Citation
Community college developmental writing courses

Material Information

Title:
Community college developmental writing courses a study of student achievement and persistence
Creator:
Greenwood, Elaine A., 1945-
Publication Date:
Language:
English
Physical Description:
x, 129 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Basic writing ( jstor )
Classrooms ( jstor )
College freshmen ( jstor )
College students ( jstor )
Community colleges ( jstor )
Control groups ( jstor )
Freshman composition ( jstor )
High school students ( jstor )
Individualized instruction ( jstor )
Students ( jstor )
Academic achievement ( lcsh )
Curriculum and Instruction thesis Ed. D
Dissertations, Academic -- Curriculum and Instruction -- UF
English language -- Composition and exercises ( lcsh )
English language -- Remedial teaching ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ed. D.)--University of Florida, 1984.
Bibliography:
Bibliography: leaves 123-127.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Elaine A. Greenwood.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030595567 ( ALEPH )
11988757 ( OCLC )

Downloads

This item has the following downloads:


Full Text











COMMUNITY COLLEGE DEVELOPMENTAL WRITING COURSES: A STUDY OF STUDENT ACHIEVEMENT AND PERSISTENCE









By

ELAINE A. GREENWOOD


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF EDUCATION



UNIVERSITY OF FLORIDA


1984


















Copyright 1984

by

Elaine A. Greenwood














ACKNOWLEDGEMENTS


Special thanks go to Dr. Al Smith, my committee chairman, for his considerable insight and guidance; to Dr. James L. Wattenbarger, committee member, for his example of national leadership in higher education; to Dr. Charles D. Dziuban, committee member, for his patience in teaching me research skills; and to Donald J. Tighe, my mentor, for a decade of collaboration.

My appreciation is also due to Alban and Ellen Guenette, my parents, for their belief in me always; to Barbara Greenwood, my mother-in-law, for her example of steadfastness; and to Andrew and Ann Greenwood, my children, for their sense of humor. My deepest appreciation goes to Richard G. Greenwood, my husband, for his unwavering attention to detail, encouragement, and love.














TABLE OF CONTENTS


ACKNOWLEDGEMENTS .................................................. iii

LIST OF TABLES .................................................... vi

LIST OF FIGURES ................................................... viii

ABSTRACT .......................................................... ix

CHAPTER

THE NATURE OF THE PROBLEM ............................... 1

Problem Statement ....................................... 4
Need for the Study ...................................... 8
Delimitations ........................................... 11
Limitations ............................................. 12
Definitions ............................................. 13

2 REVIEW OF THE LITERATURE ................................ 18

Summaries of Evaluations ..............................19
Studies Related to Research Questions....................25
Research Question #1: Developmental Writing-Individualized Versus Classroom ................... 25
Research Question #2: Developmental Completers
Versus Control Groups Who Needed But Did
Not Enroll in Developmental Writing ............... 27
Research Question #3: Developmental Completers
Versus Control Groups Who Did Not Need
Developmental English ............................. 35
Research Question #4: Age as a Factor for
Developmental Students ............................ 43
Summary ................................................. 45

3 METHODOLOGY ............................................. 48

General Research Design ................................. 48
Instrumentation ......................................... 48
Sample .................................................. 49
Collection of the Data .................................. 50
Analysis of the Data .................................... 51









4 FINDINGS ................................................ 55

Background Analysis of Samples .......................... 56
Research Question #1: Classroom Versus
Individualized Developmental Course .................. 60
Research Question #2: Developmental Groups
Combined Versus Control Group of No-takes ............ 66
Research Question #3: Developmental Groups
Combined Versus High Scorers ......................... 71
Research Question #4: Younger Versus Older
Developmental Students ............................... 77
Summary ................................................. 83

5 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS ............... 87

Summary ................................................. 87
Conclusions ............................................. 90
Recommendations ......................................... 101

APPENDIX

A INITIAL ENGLISH ASSESSMENT TEST ......................... 110
B SAMPLE STUDENT TRANSCRIPT ............................... 118
C DESCRIPTION OF DEVELOPMENTAL CLASSROOM
AND INDIVIDUALIZED COURSES ............................ 120

REFERENCES ........................................................ 123

BIOGRAPHICAL SKETCH ............................................... 128














LIST OF TABLES


1 A Descriptive Analysis of Initial English Assessment
Test Scores ................................................ 57

2 Results of Analysis of Variance on the Mean Scores on
the Initial English Assessment Test by Groups .............. 58

3 A Comparison of the Mean Scores on the Initial English
Assessment Test Between the Combined Developmental
Groups and the Control Group of No-takes ................... 59

4 A Comparison of Developmental Writing Students' Mean
Scores on the Initial English Assessment Test by
Student Age at Entry ....................................... 59

5 A Comparison of the Mean Grade in Developmental English
Courses Between Students in the Classroom and
Individualized Courses ..................................... 61

6 A Comparison of Mean Grade in Freshman Composition I,
GPA, Semesters, and Credits of Developmental Completers of the Classroom and Individualized Courses ........ 62

7 A Comparison of the Completion Rate of Developmental
Courses Between the Classroom and Individualized
Groups ..................................................... 64

8 A Comparison of the Completion Rate of Freshman Composition I Between Students Who Completed the Classroom
and the Individualized Developmental Courses ............... 64

9 A Comparison of the Degrees Earned by Developmental Classroom and Individualized Course Completers .................. 65

10 A Comparison of the Mean Grade in Freshman Composition I
of Combined Developmental Groups Versus Control Group
of No-takes ................................................ 67

11 A Comparison of the Mean GPA, Semesters Enrolled, and
Credits Earned of the Combined Developmental Groups
and the Control Group of No-takes .......................... 68

12 A Comparison of the Rate of Completion of Freshman Composition I for Combined Developmental Groups Versus Control
Group of No-takes .......................................... 70









13 A Comparison of Associate Degrees Earned by Developmental
Completers Versus Control Group of No-takes ................ 71

14 A Comparison of the Mean Grades in Freshman Composition I
Earned by Developmental Completers Versus High Scorers ..... 72

15 A Comparison of Cumulative GPA, Semesters Enrolled, and
Credits Earned of the Combined Developmental Groups
Versus the Control Group of High Scorers ................... 73

16 A Comparison of the Completion Rate of Freshman Composition I for Developmental Completers Versus the Control
Group of High Scorers ...................................... 75

17 A Comparison of the Associate Degrees Earned by Developmental Completers Versus Control Group of High Scorers ..... 75

18 A Comparison of Mean Grade in Developmental Course, GPA,
Semesters Enrolled and Credits Earned by Younger and
Older Developmental Course Completers ...................... 78

19 A Comparison of the Mean Grade in Freshman Composition I
of Younger and Older Developmental Course Completers ....... 79

20 A Comparison of the Rate of Completion of a Developmental
Course by Younger and Older Students ....................... 80

21 A Comparison of the Rate of Completion of Freshman
Composition I by Younger and Older Developmental
Course Completers .......................................... 81
22 A Comparison of the Associate Degrees Earned by Younger
and Older Developmental Completers ......................... 82

23 Summary of Findings ........................................... 84

24 Summary of the Relationship of Findings to Hypotheses ......... 85














LIST OF FIGURES


1 Diagram of Groups for Research Questions 1, 2, and 3 as
Determined by Score on the Initial English Assessment
Test ....................................................... 52

2 Diagram of the Developmental Group Realignment for Research
Question 4 as Determined by Student Age at Entry ........... 53


viii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Education


COMMUNITY COLLEGE DEVELOPMENTAL WRITING COURSES:

A STUDY OF STUDENT ACHIEVEMENT AND PERSISTENCE

By

Elaine A. Greenwood

December 1984

Chairman: Dr. Al Smith

Major Department: Curriculum and Instruction

The purpose of this study was to evaluate Valencia Community College's (Florida) developmental writing program in terms of student achievement and persistence. The study contrasted the performance of students who completed a classroom developmental course, who completed an individualized developmental course, who avoided taking a developmental writing course, and who did not need a developmental writing course. The last part of the study contrasted the performance of older and younger developmental course completers.

Four student samples (developmental classroom students, developmental individualized instruction students, developmental avoiders, and high scorers) were randomly drawn from fall 1980 college entrants grouped by English placement test scores and first semester English course. Achievement was measured by grades in a developmental course, grades in Freshman Composition I, and cumulative grade point average








(GPA). Persistence was measured by completion of a developmental writing course, completion of Freshman Composition I, number of semesters enrolled, number of credits earned, and attainment of an associate degree over eleven semesters.

The study showed no significant differences on any measure between developmental classroom completers and individualized course completers. Developmental course completers did have a significantly higher cumulative GPA than did developmental avoiders. No significant differences occurred on all other measures even though the avoiders had slightly stronger entry-level writing skills than the developmental students. High scorers, who entered with strong writing skills, achieved significantly higher grades and earned more associate degrees than did developmental completers; but developmental completers equalled high scorers in Freshman Composition I completion, number of semesters enrolled, and number of credits earned. Older developmental

course completers achieved significantly higher cumulative GPAs than did younger developmental students, but age had no effect on the other measures.

Recommendations included that the Florida Department of Education should investigate the feasibility of a statewide evaluation model for developmental courses, given the current state-mandated entry-level placement and sophomore testing programs; that students should not be denied access to developmental writing courses on the basis of age; and that further research should be conducted at Valencia to determine if the recently upgraded developmental writing curriculum has improved student achievement and persistence.














CHAPTER 1
THE NATURE OF THE PROBLEM

For the past fifteen years, developmental education programs have been widespread in community colleges. Alternative instructional methods, such as individualized instruction, have been incorporated into many of these programs. A recent survey of community college presidents indicated that 97.8% of the 639 responding institutions offered developmental or remedial courses (Campbell, 1982).

The current emphasis on developmental education in community colleges grew out of the "open door" admissions policies starting in the late 1960s. "Open door" institutions began admitting nearly every adult applicant, no matter what educational credentials the applicants had. Large numbers of non-traditional students were among these new adult students. Cross (1971) defined these non-traditional students comprehensively as all adult learners who had difficulty with academic work, regardless of the cause of the difficulty. What these new students had in common was that they scored in the lowest third among national samples on standardized tests of academic ability. What they did not have in common was just about everything else. Some were representatives of minorities, some not; some were women; some were in their twenties, thirties, forties, some even older; some were married with families; many were working full-time jobs in addition to attending college. However, the lower socioeconomic classes, women, minorities, and older students had higher representation in this group than among traditional college students.









Advice on how to deal with non-traditional, academically unprepared students grew with their numbers. A great deal of theorizing occurred about alternate instructional methods. Individualized instruction was one such proposed method. The Carnegie Commission on Higher Education (1970) suggested that community colleges needed remedial education programs which were flexible and individualized. Roueche and Pitman (1973) advocated combining Bloom's mastery learning theories (Bloom, 1971) with systematic, individualized instruction in basic skills. Competency-based instruction was often part of the individualized approach (Roueche, Herrscher, & Baker, 1976). The intensity of the student-instructor relationship in individualized instruction became a hallmark of many programs (Gollattscheck, Harlacher, Roberts, & Wygal, 1976).

After fifteen years of widespread developmental programs, state legislatures in charge of funding most state higher education systems are increasingly asking pointed questions about the effectiveness of developmental and remedial programs. Many legislatures, as in Maryland, are reassessing higher education programs with an eye to budget cuts, or at least different funding approaches (Study of Remedial/Developmental, 1982). At the same time, some states have been attempting to upgrade elementary and secondary systems. Legislators, like those in Illinois, asked if developmental programs and alternative instructional methods are really helping underprepared students cope with the demands of a college education. They are also questioning if such programs will continue to be necessary as school systems produce better prepared high school graduates (Status Report, 1981). In Texas, there is even a call for the state to close the open door in the









community college system since developmental programs, according to the director of research for the Coordinating Board of the Texas College and University System, have only diluted quality in postsecondary education (Patterson-Griffith, 1983). In Florida, also, such questions are being asked--and answered with legislation. In 1982, the Florida Department of Education's Postsecondary Education Planning Commission recommended that developmental/remedial programs at the state community colleges and universities be eliminated by the year 1990 (The Master Plan, 1982). In the 1983 state legislative session, lawmakers took that advice and enacted legislation that shifted responsibility for remedial and developmental programs from higher education to district school boards (Section 232.2455, 1983). According to this law, after 1990, the only remedial courses allowed in community colleges would serve students who would be five or more years from high school graduation or who needed remediation before benefitting from vocational education (Section 240.134, 1983). Although these limitations on developmental/remedial courses were eliminated by the 1984 legislature (Section 240.117, 1984), more legislative limitations in Florida and in other states are a possibilty.

A number of developmental education theorists and researchers have expressed growing concern about the need for evaluation if developmental programs are to survive. Perry-Miller, Nolan, and Smith (1980) point out that current arguments at the local, state, and national level about whether developmental education should even be offered at the college level make it necessary that developmental educators continue to convince constituencies that tax dollars should go to developmental education. Maxwell (1979) cites the need for experimental and









quasi-experimental studies to contribute to decisions about program continuation so that programs can gain support from the populace and

from political decision makers. Akst and Hecht (1980) warn that too few remedial programs have been carefully evaluated, despite the political reality of shrinking budgets and growing demands for accountability.

The problem facing developmental educators in the mid-1980s is to provide evaluation of their programs in a manner which will have credibility with decision makers who control continued funding for postsecondary education. Without continued funding, the programs will be cut back. If legislative bills, such as the one enacted by the 1983 Florida legislature cited above, explicitly prohibit community colleges and universities from providing developmental programs, such programs could even disappear.


Problem Statement

The problem in this study was to evaluate a community college developmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an individualized format, so that the college might assess the efficacy of each course and of the entire program in terms of student persistence and achievement. The achievement and persistence of randomly selected students experiencing these two modes in the fall semester of 1980 were compared. The achievement and persistence of students from both courses combined were also compared with achievement and persistence of two randomly selected control groups: one of students who needed developmental writing but did not enroll in either course and one of





5


students who neither needed nor enrolled in a developmental writing course in the fall semester of 1980.

Three major research questions were asked about the effect of the developmental writing courses. These questions and their accompanying null hypotheses follow:

I. Does an individualized developmental writing course or a classroom based developmental writing course contribute to better achievement and longer persistence by students who need a developmental writing course when they start college?

HI: There is no significant difference in achievement (in terms of

mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental

writing course and completed the classroom course.

H 2: There is no significant difference in persistence (in terms of

completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individualized course and students who needed a developmental writing course and enrolled in

the classroom course.

II. Do students who need and complete either of the developmental writing courses achieve better and persist in college longer than students who need but do not enroll in a developmental writing course?









H3: There is no significant difference in achievement (in terms of

mean grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course and students who needed and completed a

developmental writing course.

H4: There is no significant difference in persistence (in terms of

completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who

needed and completed a developmental writing course.

III. Do the developmental writing courses help students with weak writing skills at entry catch up, in terms of achievement and persistence, with students who do not need a developmental writing course when they start college?

H 5: There is no significant difference in achievement (in terms of

mean grade in Freshman Composition I and cumulative grade point average) between students who needed and completed a developmental

writing course and students who neither needed nor enrolled in a

developmental writing course.

H 6: There is no significant difference in persistence (in terms of

completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed and completed a developmental writing course and students who neither

needed nor enrolled in a developmental writing course.









The effect of developmental writing courses on older adult students as well as on recent high school graduates was also considered in this study for two reasons. First, the developmental courses were designed for non-traditional students who comprised the majority of students enrolling in these courses, and non-traditional students as a group varied greatly in age. Second, one argument for discontinuing developmental level courses made by some state legislators was that the

upgrading of primary and secondary schools would result in stronger high school graduates, which could diminish or eliminate the need for college level developmental courses. However, it is important for legislators to consider that many developmental students are much older than eighteen and, therefore, have not graduated from recently upgraded school systems. Legislators need to know how the developmental courses

have affected the achievement and persistence of older adults as well as younger students. In Florida, legislation enacted in 1983 limited

remedial and developmental education after 1990 to two classes of students, one of which was the adult who had been "five years or more out of sequence from high school graduation". (Section 240.134 (3) (a),

Florida Statutes, 1983). It is important to ascertain if this arbitrary age demarcation point is a valid one.

Thus, the major research question asked about student age as a factor in the effect of developmental writing courses in this study was as follows:

IV. After completing a developmental writing course, do older adults achieve and persist any differently from recent high school graduates? H7: There is no significant difference in achievement (in terms of

mean grade in the developmental course, mean grade in Freshman









Composition I, and cumulative grade point average) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course

at age 17 to 21.

H8: There is no significant difference in persistence (in terms of

completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled

in a developmental writing course at age 17 to 21.



Need for the Study

All instructional programs should be evaluated to ascertain if they are effective in attaining their goals. The need is perhaps more critical for developmental programs. At most community colleges, a

substantial proportion of entrants are non-traditional students, and often the majority have weaknesses in the basic skills. At Miami-Dade Community College, for example, about two-thirds of the entering students need course work in one of the college's developmental areas; nearly one-half of the entering high school graduates need course work in all developmental areas, and with a mean student age of twentyeight, large numbers of these entering students will not benefit from upgraded secondary programs (McCabe, 1983). At Valencia Community College, the comparable figures range from 57% to 79% of the students needing course work in one or more of the college's developmental courses (Riles, 1983). These figures are typical of the large numbers









of students entering other Florida community colleges and needing developmental programs. Besides having to acquire skills they lack at entry and continue through the college curriculum, Florida students must also achieve the approved level of proficiency on the College Level Academic Skills (Rule 6A-10.31, 1982), an official statewide list of communications and computations competencies, and pass a test on these skills (Rule 6A-10.312, 1984). Community college students who

fail this test neither receive an Associate in Arts degree (Rule 6A10.313, 1982) nor are they allowed to matriculate into the upper division of a Florida public university (Rule 6A-10.314, 1984). To this end, according to a recent memo by the Florida State Commissioner of Education to the presidents of the Florida community colleges and state universities,

Institutions will be required to assure that students who enter their college credit programs have effective opportunity to
achieve the skills to the level required by the passing scores.
Admissions counseling, developmental and remedial programs, standards for course completion, and standards of student progress that are required for retention as well as enrollment and funding will be impacted by the passing scores on CLAST (College Level
Academic Skills Test). (Turlington, 1983, p.3)

The continuing need for developmental programs in Florida community colleges is clear.

The need for evaluating these developmental and remedial programs is equally clear, especially given the current political climate and recent legislation. According to 1983 statutes, which have since been rewritten, starting in 1990, primary responsibility for remedial and

developmental programs would have been given to Florida county public school districts. Remedial and developmental programs at community colleges would only have been permitted to serve students needing basic skills in order to be successful in vocational programs or students who









completed high school five or more years before starting college (Section 240.134, 1983). Although these limitations have since been written out of the statutes (Section 240.117, 1984) substantive studies

might play a part in shifting political opinion toward future possible legislation concerning developmental education in Florida and other states.

But even ignoring the current political climate, the time has come for more community colleges to find out systematically if their developmental programs have been effective in reaching their goals. According to Darrel Clowes (1984), developmental programs are unique in their respective colleges in that the output of the developmental program is the input for almost every other program on campus. How many students make a successful transition into the mainstream curriculum and how well they do there are important and available data.

Such data have not been tapped at Valencia Community College, the site of the developmental courses under study in this research. The developmental writing courses in existence since 1976 have not been evaluated in a thorough and systematic way. Data have been available about the percent of students who completed the courses and the mean grade received in those courses, but no breakdown by age and only one

comparison with students who needed but did not enroll in the courses has been made. No comparison between outcomes of the classroom course

and the individualized course has been made. There has also been no study of student persistence and achievement for more than one semester after completion of these courses. Therefore, although informal formative evaluation of the courses has contributed to their evolution, it is indeed time for a systematic, summative evaluation of this program.









Delimitations

The delimitations of the study were as follows. Only the developmental writing courses at Valencia Community College, West Campus, were evaluated. The developmental reading, study skills, and mathematics courses were not involved. Valencia does not have a developmental education department or division per se. The developmental mathematics courses are a part of the Mathematics Department; the developmental writing, reading, and study skills courses are a part of the Communications Department. Instructors who teach in all these developmental courses are part of and report to their respective academic departments. There is no separate funding for a developmental program; funds come through the respective department budgets. There is no comprehensive developmental program and no counseling component automatically attached to the developmental courses, although many students in these courses also belong to the Special Services program or the Handicapped Student Services program, both of which involve students in extensive counseling. There are frequent contacts and referrals back and forth between instructors teaching the developmental courses and counselors in these programs.

Therefore, the scope of this study was confined to student outcomes from two courses, the classroom developmental writing course and the individualized developmental writing course at one Florida community college campus. The data were drawn from students who entered the college in the fall semester of 1980.









Limitations

The findings in this study could be generalized to the same type of populations and courses described herein. The overall population was the group of students who chose to enter an urban community college with an open door admissions policy. The subpopulations were those students who needed and enrolled in a developmental writing course, those who needed but did not enroll in a developmental writing course, and those who did not need and did not enroll in a developmental writing course. A randomly selected sample was drawn from each of these subpopul ations.

The study probably should not be applied to developmental programs

broader in scope than this one, that is, to programs which may include courses and/or supplemental lab work for students needing a variety of academic course work and which may or may not include a counseling component.

Findings pertaining to the classroom developmental writing course may be generalized to other classroom developmental writing courses. Findings pertaining to the individualized developmental writing course may be generalized to other individualized developmental writing courses with the following cautions. This individualized developmental writing course took place in a laboratory setting and carried elective credit; the instructors worked with a maximum of twelve and usually only ten students per section, and there were no student tutors. The instructors conferred with each student each time the student came to his or her regularly scheduled lab meeting. Many individualized developmental courses have sections with much higher numbers of students, with student tutors or paraprofessionals who carry on much of the








direct instruction, with instructors who do not have the opportunity to monitor each student's progress at each lab meeting, and even with variable scheduling. Findings from this study should only be generalized to other courses with one or a combination of these differences with extreme caution.

Furthermore, this study focused on only quantifiable data of

achievement and persistence. It did not attempt to measure affective outcomes of the student-instructor interaction involved in these courses. It did not measure value added characteristics these courses might have had, that is, what value the course might have had even for

students who withdrew during the course or later before completing a degree. It did not attempt to ascertain if students who eventually

withdrew from college stayed in college longer than they might have if they had not enrolled in a developmental writing course.

Finally, this study did not attempt to delineate why a course was successful or unsuccessful in increasing student persistence and achievement, just that it did or did not.



Definitions

Achievement is the level of academic attainment as measured by grade

earned in the developmental writing course, subsequent grade in

Freshman Composition I, and grade point average in all courses.

A classroom course is any course which meets in a regular college

classroom and is taught by traditional instructional methods, primarily lecture and discussion.

A classroom developmental writing course is a developmental writing

course taught in a classroom using traditional instructional methods.








Developmental completers are students who have completed either the

classroom or the individualized developmental writing course.

Developmental courses are courses in basic skills such as reading,

writing, mathematics, and study skills in which the instructor desires to nurture students' self-concepts, ability to cope with academic strategies, and self-confidence as they learn new skills.

Developmental programs are programs which include a group of basic

skills courses such as reading, writing, mathematics, and study skills courses along with a counseling component and possible supplemental lab work. Developmental programs often compose their own department or division within the college's administrative

structure.

A developmental writing course is a course in which the instructor desires to nurture students' self-concepts, ability to cope with academic strategies, and self-confidence when they are learning the

skills involved in the process of basic composition. The site could be either classroom or lab; the instructional method could

be either traditional or individualized.

An entry level test is a test or series of tests administered to students after they complete admission but before they register for their first semester of classes. The purpose of the test(s) is to

help place students into courses appropriate to their level of ability and skill achievement. English entry tests usually designate whether students have sufficient writing skills for success

in Freshman Composition I or whether they should enroll in a developmental writing course.









Freshman Composition I is the standard course in rhetoric taught to

first year college students. In Florida, the common course numbering system designation usually used for Freshman Composition I

is ENC 1101.

Grade Point Average (GPA) is the total of a student's quality points

earned in each course divided by the total number of semester

hours in which the student enrolled and did not withdraw.

High scorers are students who score above the cutoff on the Initial

English Assessment Test and enroll directly into Freshman Composition I.

An individualized course is a course in which a one-to-one learning

relationship exists between the instructor and student; in which the instructor assigns materials appropriate to each student's needs, monitors student progress constantly, and provides constant

feedback to the student on progress made; and which takes place in

a learning laboratory.

An individualized developmental writing course is a course in writing

skills with the characteristics of both an individualized course

and a developmental writing course.

The Initial English Assessment Test is an objective test of basic English skills written by English faculty at Valencia Community College and used as the entry level English test to recommend placement into a developmental writing course or into Freshman Composition I during the year when the students measured in this

study enrolled.

A learning laboratory is an environment designed especially for one-toone instruction, usually including individual study carrels,








tables, shelves, file cabinets, and storage units appropriate to a variety of books, audiovisual materials and equipment, and possibly computer software and hardware for computer assisted instruction.

Long-term measures are measurements of student achievement and persistence that occur a semester or longer after completion of a developmental course.

"No-takes" are students who score below the cutoff on the Initial

English Assessment test, who are therefore advised to enroll in a

developmental writing course, but do not.

Non-traditional students are college students of any age or race and of

either sex who, for whatever reason, have difficulty with academic work in college, who usually score in the lower third on nationally normed tests, and who would benefit from developmental courses to improve their skills and their chance of success in

mainstream courses.

Older students are students who are twenty-two years old or older when

they enroll in a developmental course.

Persistence is longevity of matriculation as measured by the rate at

which students complete a course, complete another course subsequent to the first, complete a series of semesters within the four years of this study, earn a cumulative number of credits, and earn

an associate degree.

Recent high school graduates are students who have graduated from high

school and who are age seventeen to twenty-one when they enroll in

a developmental course.






17


Short-range measures are measurements of student achievement and persistence that occur during or at the end of the semester in which

the student is enrolled in a developmental course.

Younger students are students who have graduated from high school and

who are age seventeen to twenty-one when they enroll in a developmental course (same as "recent high school graduates").















CHAPTER 2
REVIEW OF THE LITERATURE

More than one evaluation specialist has noted that current evaluation of remedial and developmental programs presents great diversity and little effort to synthesize that diversity (Akst & Hecht, 1980; Dumont, Bekus, & Tallon, 1981; Grant & Hoeber, 1978; Kendrick & Thomas, 1970; Kulik, Kulik, & Shwalb, 1983; Richardson, Martens, & Fisk, 1981; Tinto & Sherman, 1974). Comparisons between studies are difficult because of the differences in methodology and the lack of agreement on appropriate measurement criteria. In addition, there is no clear pattern to the results. Positive, nonsignificant, and even negative results have all been reported for entire developmental programs and for developmental writing courses in both classroom and individualized instructional modes.

The results of the better studies extant in the literature follow here. First come the summaries of evaluation studies. The rest appear in the same order as the major research questions being asked in this study: (1) what does the research tell us about the achievement and persistence of students in classroom versus individualized developmental writing courses; (2) what does the research tell us about the achievement and persistence of students who need and complete a developmental writing course as opposed to students who need but do not enroll in such a course; (3) what does the research tell us about the achievement and persistence of students who need and complete a






19
developmental writing course as opposed to students who do not need and do not enroll in such a course; and (4) what does the research tell us about the achievement and persistence of older adult students who need and complete a developmental writing course.



Summaries of Evaluations

The following summaries and a meta-analysis concern developmental programs, not developmental writing courses specifically. Metaanalysis is sometimes criticized as having such a sheer amount of data grouped and analyzed together that regression toward the mean causes misleading results. However, many researchers and decision makers look to summaries and meta-analyses as pointing out the major trends in the area being studied.

Kendrick and Thomas (1970) reviewed studies on disadvantaged college students, remarking that it seemed premature in 1970 to be reviewing these studies because the emphasis nationwide was still on expanding educational opportunity. The high priority nationwide on admitting such students to college had only existed for a few years. Kendrick and Thomas focused their report more on attempts to discern which traits these disadvantaged students had which would be most predictive of success. They noted that it was hard to ascertain the impact of remedial programs on helping disadvantaged students attain a bachelor's degree because so few community college students were going on to four year schools. They also noted that when studies showed good retention rates for these students, the studies too often had not been designed to determine if innovation in the program or other factors such as less demanding courses, a lighter course load, or atypical persistence patterns were responsible for the positive results.








Four years later, Tinto and Sherman (1974) analyzed evaluations of special services programs for disadvantaged students in secondary and postsecondary institutions. Like Kendrick and Thomas, they also bemoaned the difficulty of synthesizing reports that focused on different outcomes, different populations, different points in time, different educational settings, different measures, and different research designs, most of which they found seriously flawed. Tinto and Sherman concluded that special services programs, with their heavy emphasis on counseling, tutoring, personal and cultural enrichment in addition to remedial course work, did seem to decrease dropout rates and increase retention of disadvantaged students. However, the increase did not reach the retention level for non-remedial students. In addition, special services programs seemed to have little positive impact on academic achievement. The special services students attained about the same GPA in college as they had attained in high school. Even more worrisome was the fact that disadvantaged students who were not part of special services programs did about as well as the special services students. However, although tutoring and counseling seemed to make no difference in academic achievement, some studies showed that special service students did seem to make important changes in attitudes, values, and motivation as evidenced by their higher persistence to graduation rate than non-special services and disadvantaged students had. Tinto and Sherman cautioned that these results may have been affected by self-selection into the programs. Since most studies were flawed by not having random assignment of students to groups, the higher rate of persistence (about 50%) might have been due to the more motivated disadvantaged students deciding to take








advantage of the special services program while less motivated students decided to ignore it. Overall, the special services programs with better results were usually integrated into the regular college program; they were not separate. The students in separate programs were often stigmatized; they usually did not fare as well in retention and persistence patterns as the students in integrated programs.

Roueche and Snow (1977) disagreed with Tinto and Sherman on this last point. Roueche and Snow's study was a survey of the developmental education programs of 150 public two-year colleges and 150 public four-year colleges. They concluded that in four-year colleges, at least, having a separate department or division of basic studies was significantly related to the respondents also rating their program as successful. A significant relationship did not exist between this factor and self-reported successful programs for two-year colleges, though. Other factors with statistically significant relationships to self-reported success were written course objectives given to students, tutors trained to develop positive self-concepts in their tutees, peer counselors selected on the basis of their effectiveness and trained to work on improving their students' self-concepts, faculty who compared their students favorably with other students at their college, full-time counselors who were attached to the developmental courses and who consulted in curriculum development, and successful retention of students until their third semester in college. However, the main problem with the Roueche and Snow study was that it was a self-report survey. When respondents indicated that their program was successful, they were asked for their opinion and not required to comment based on any substantive research at their institution.









Grant and Hoeber (1978) reiterated the problem of first finding studies which were empirically based and second comparing those studies given the broad range of designs and outcome measures employed. When concluding whether or not basic studies programs were working, Grant and Hoeber wrote:

The question can be answered in two ways--yes, those involved in and committed to those programs are 'working' very hard; no, the programs themselves are not 'working' very well, or, more accurately, a dearth of empirically based evaluation research on
these programs makes the question moot. (p. 1)

They did decide that some developmental programs were doomed--those with curriculum and instructional methods similar to what developmental students had experienced in elementary and secondary school. Such curriculum and teaching methods had not worked in these students' previous twelve years of schooling; they probably would not work in college either. Grant and Hoeber recommended that programs be modified for each student and that recognition of different learning styles and rates of growth should lead to greater use of mastery learning. They also concluded that the literature clearly indicated that one type of developmental program could not possibly be successful as the universal model for all postsecondary programs. Just as programs needed to be modified for individual students, programs needed to be individually tailored for each campus in order to fit each institution's needs.

Richardson, Martens, and Fisk (1981) compiled results of developmental programs for their research into functional literacy in colleges and universities. Retention or persistence was usually measured in terms of continual enrollment or decline in the dropout rate.








Community colleges frequently reported positive results in retention, universities less often. Richardson et al. speculated that many institutions could not supply such data because they did not have it or would not supply such data because they wished to suppress it. As for achievement measurements, most studies used GPA as the criterion for success. Often, students' GPAs during remediation dropped significantly as soon as they took regular college classes. In some studies, completing a developmental writing course often contributed to an improved GPA, but completing, failing, or withdrawing from a developmental reading course made no difference in GPA at all. As for studies comparing methods of instruction, the most common comparison was between individualized approaches contrasted to traditional classroom instruction. The results were sometimes positive, sometimes not significant, sometimes negative, leaving confusion as to the effectiveness of individualized programs.

The most current meta-analysis is Kulik, Kulik, and Shwalb's empirical study of research on college programs for high-risk and disadvantaged students (1983). Following systematic selection of quantitative, experimental studies, the authors chose 60 reports out of 504 documents to apply multivariate statistical analysis to in order to integrate findings. Each of the chosen studies had to involve highrisk college students in special programs, had to report on measures made on these students plus on one or more control groups with similar aptitude, and had to have no methodological flaws. The measures had to involve achievement in terms of GPA and persistence in terms of the proportion of students who remained enrolled during the course of each study.








In terms of achievement, Kulik et al. concluded that the developmental programs as a whole had basically positive results. In 44 of 57 studies including GPA as a measure, GPA was higher for students in special programs than for similar students who were not in special programs. Significantly higher GPAs for students in special programs were reported in 16 of 17 studies where the difference reached a significant level. When meta-analysis techniques were applied, the typical report would have cited a GPA of 2.03 for special program students but 1.82 for control groups, which the authors admitted was a small difference, but considered statistically reliable. Regression analysis showed that three features of the programs were associated with higher GPA: newer programs (perhaps due to novelty effect), programs characterized as remedial or developmental studies (as opposed to isolated academic skills instruction, guidance sessions, and comprehensive support systems), and programs reported in documents by the Educational Resources Information Center (ERIC), rather than in journal articles and dissertations.

Kulik et al. continued with meta-analysis of persistence. They defined persistence as "the proportion of students initially admitted who remained enrolled at the college during the period examined in the study" (p. 400). Of the 30 studies measuring persistence, 21 showed higher persistence for special program students, 11 of which were statistically significant. Further analysis indicated that the typical persistence rate across the studies was 60% for special program students versus 52% for control students. The authors judged the special programs as having a statistically reliable effect on persistence, but it was smaller and harder to see than the effect on GPA.








Using regression analysis, the authors concluded that how early students started in a special program (high school was better than freshman year in college, which was better than second year in college, etc.) was the only consistently important factor on persistence.

The authors were disappointed in the results for community college programs, which, by and large, produced no significant difference in GPA and little effect on persistence. Program effects were stronger at four-year colleges and strongest at doctoral universities.



Studies Related to Research Questions

Research Questions #1: Developmental Writing--Individualized Versus
Classroom

Whether an individualized approach to developmental writing instruction or the traditional classroom works better is a question many researchers have asked. Their answers have produced mixed results and, more often than not, statistically nonsignificant results.

Sutton and Arnold (1974) were concerned that classroom instruction denied academically disadvantaged students the personal attention they required to make significant gains. They theorized that individualized instruction or instruction in a writing laboratory might be more effective than classroom remedial English courses. Their writing lab approach incorporated tutoring with programmed instruction. The student population was 244 entering freshmen at a southern regional university in the fall of 1970. Sutton and Arnold found during the two years of the study that there was no apparent difference in the number of students who persisted in college nor in the dropout rate between students who experienced the individualized course in the









writing lab and students who experienced a remedial English course in the traditional lecture-discussion mode. However, the achievement of the individualized course students was higher than for the classroom course students; the individualized course students earned higher mean grades in subsequent English courses as well as higher cumulative GPAs than did the classroom course students.

Epes, Kirkpatrick, and Southwell (1980) designed and reported on an individualized, audio-tape enhanced course as contrasted with traditional, classroom-based developmental writing courses at three colleges of the City University of New York. Their new course did involve some whole-group instruction, but students worked primarily in a writing lab at their own pace on individually assigned audio-tapes, exercises, and writing assignments created by the authors. Most measures used in this study pertained to attitude toward English courses or toward writing in general (no significant difference) or writing improvement scales (error-count reduction was significantly better on two of six errors for the individualized students; holistic evaluation of student writing at both the beginning and end of the semester showed no significant difference). In terms of persistence, there was also no significant difference between the individualized and the classroom students. Of the 243 lab students, 140 (57.6%) passed while of the 199 traditional classroom students, 129 (64.8%) passed. The difference was not significant.

On the other hand, Brown (1984) found that a teacher-paced, classroom developmental English course seemed to result in higher grades in developmental English and in higher term GPAs than did a self-paced developmental English course. Using a matched subject








design, Brown compared student results in a fall semester 1977, teacher-paced course with student results in a fall semester 1980, self-paced course at Quinsigamond Community College. Students were matched on the basis of age, sex, and score on a standardized, reading pre-test. No significant difference occurred on three of the five measures: post-test standardized reading scores, completion rate of the developmental English course, and number of students persisting in college for four consecutive semesters. The students in the teacherpaced, classroom course did significantly outperform the self-paced student group on the remaining two measures: final grade in developmental English and GPA during the semesters under study.


Research Question #2: Developmental Completers Versus Control Group

Who Needed But Did Not Enroll in Developmental Writing

Those studies which set up a control group most often concern a group of students who need and enroll in the developmental writing course and another group of students who need to, but for whatever reason, do not enroll in a developmental writing course. These studies are often judged to be flawed because students are not assigned randomly to the experimental and the control groups. Choosing students randomly for the developmental course group and the control group may be a strong experimental design, but many researchers have eschewed that option. They or their institutions felt a moral dilemma faced them; for the sake of a design, they could not deny students with weak skills the chance to be helped by a developmental course. Consequently, researchers have more often chosen to form a control group from students who were accidentally left out of developmental courses, who could not get into a developmental course because it had








closed out, who could not fit the developmental course into their schedule, or who chose to bypass a developmental program or course and enroll directly in regular freshmen courses.

Unfortunately, there are design flaws inherent in accidental control groups. For example, all types of accidental control groups could easily be composed of non-representative samples of students. The population of students who could not get into a developmental course because it had closed out might easily have a higher percentage of the weakest students. Students who could not fit the developmental course into their schedule might primarily be older adults working full-time jobs during the day and trying to register at a college with a limited night schedule. This population might be stronger as a group than primarily younger students who could fit a developmental course into their day-time schedule more easily. Students whose entry scores indicated the need for a developmental course, yet who chose to bypass it and instead enroll in regular freshman courses might be either so self-confident and highly motivated that they do well in the regular freshman courses or so unwise about their own abilities that they fail in large numbers.

Some randomly assigned control group studies do follow below, but most researchers cited here opted for the more accidental type of control group because of their reluctance to deny the possible benefits of a developmental course to any student.


Randomly assigned control groups

Losak and Burns' early study is the most often quoted study in the literature (1971). Students registering full-time at Miami-Dade








Community College in the fall of 1969 and scoring too low on the entry level English test for normal placement into freshman English, were randomly assigned into three groups. The first were registered for an experimental remedial program called the Community College Studies Program, an interdisciplinary, team taught program including counselors and student tutors in addition to instructors in English, the natural sciences, and the social sciences. The second group was enrolled in a traditional remedial reading course. The third group was enrolled in regular freshman English. Although the entry test indicated that students in all three groups needed a remedial English course, there was no significant difference in the number of students in each group who passed regular freshman English. There was no significant difference in the GPA of students in the three groups. Nor was there a difference in the results of a reading post-test. Finally, there was no difference in the attrition rate. Losak and Burns wondered if the newness of the Community College Studies Program or the relatively short period of time (three months) during which this course or the traditional remedial course tried to effect major changes in skills may have contributed to the discouraging results.

In an often praised, early study, Sharon (1972) randomly selected some students who would have been required to take remedial English and placed them instead in the regular freshman English course. The only students excluded from this random control group were students who scored extremely low on the entry assessment, and they comprised only 4% of the students who needed remedial English. The rest of the students judged as needing remedial English were enrolled in that course. There was no significant difference in the persistence of the









two groups to the end of their respective courses. The students who received remediation did not pass in significantly greater numbers or withdraw in significantly fewer numbers than those in the control group who had been placed in freshman English. On the other hand, there was a significant difference between the mean grade of the two groups in freshman English. Students in the control group placed directly into freshman English earned a mean grade of 1.85 whereas the students who were placed in remedial English, completed it, and went on to freshman English earned a mean grade in freshman English of 2.27. Nevertheless, Sharon concluded that "the effectiveness of remedial courses in college should not be taken for granted . . . remediation in English has only limited value" (p. 62).

Mowatt (1978) found positive results in both achievement and persistence of developmental English students in a five year, randomized control group study. Non-traditional students entering West Valley College in 1970 and scoring below the college's cutoff on two entry English assessments either enrolled in developmental English or did not. Over the five years, the mean GPA was significantly higher for former developmental English students at 2.41 than for non-traditional students who did not enroll in developmental English at 2.16. Over the same time period, the former developmental English students also persisted better in terms of mean number of cumulated units at 40.66 versus 26.24 for those who needed but did not take developmental English. Mowatt concluded that the developmental English course did help non-traditional students improve their GPAs and to accumulate almost twice as many units as similar students who did not enroll in the developmental course.









Non-random control groups

Lovejoy (1974) found no significant difference in achievement measures, but partially positive results in persistence of academically disadvantaged students entering Western Christian College from 1961 to 1971. Two of his three groups consisted of students in a traditional, formal course in remedial English which heavily emphasized grammar. One group took this course concurrently with regular freshman English; the second took it before enrolling in freshman English. A third group, also academically disadvantaged, did not enroll in the remedial English course at all. In terms of final grade in regular freshman English, neither remedial English group showed a significant improvement over the non-remedial group. In terms of GPA in the six areas of study Lovejoy selected to include, again neither remedial English group showed improvement. The only significant gain came in persistence into a second year of college. The group which completed the remedial grammar course before enrolling in freshman English did continue in significantly larger numbers than the other two groups into a second year. However, the effect did not continue into the junior or senior year.

Lunsford's (1978) study of the remedial English course at an Ohio college used a variety of English skills measures in addition to checking persistence and achievement patterns of two groups scoring low on an entry English test. Although low scorers were randomly selected and invited to enroll in the remedial English course, only students who chose to did. From the others was drawn a control group matched with the remedial English students on ACT scores, sex, and high school rank. Some of the students in the control group enrolled









in freshman composition while some enrolled in no English course at all during their first semester. In terms of English skills, remedial English students had a significantly higher score on their holistically graded ending essay than they did on their first writing sample. The control group did not; in fact, their ending essay mean score was very similar to the remedial students' first writing sample score. This suggested that the control group's writing skills had not improved while the remedial group's had.

In terms of persistence and achievement, Lunsford also found that remedial English was beneficial to students. Of the 94 students who enrolled in remedial English, 86 completed, but only 61 of them passed (70%). In other words, about 65% of those enrolled passed the course. Of these completers, 75% went on to complete freshman English in the following term. On the other hand, only 35% of control group students enrolled in freshman composition the first semester, and of these 72% completed. The mean grade in freshman composition for the former remedial students was 1.8 with 80% having a C or above whereas the mean grade for the control students was 1.2 with only 13% having a grade of C or above. Two and a half years after the beginning of the study, 55% of the remedial students were still enrolled at the university. There were no long term persistence figures cited for the control group, however.
Presley (1981) reported on one of the annual evaluation surveys conducted throughout the Georgia postsecondary system. Georgia, since 1974, had had a statewide developmental education program under the university system's Board of Regents. Each college and university had a department of developmental education, and students who scored low









on the statewide entry test had to enroll in the course or courses which addressed their weaknesses. These students could not enroll in a freshman course in that area unless they had successfully completed the assigned developmental course. When Presley wrote, the evaluation surveys focused on persistence, but no longer required achievement figures because the consistently higher showings by students who did not need developmental courses convinced state administrators that developmental and non-developmental students were clearly not equal groups and that judging developmental students by the norm of the nondevelopmental students was setting an impossibly high criterion by which to judge the developmental programs.

Presley found that system-wide about 8% of the developmental English students were dropping out of the developmental courses, while the dropout rate for all courses was 10%. Between 50% and 60% of developmental students were passing developmental English and moving into freshman composition, although only 21% managed this in one semester. At Presley's own institution, raising the entry test cutoff score had resulted in two groups with comparable SAT scores; the new group was required to take the developmental English course while the previous year's students with the same scores had not. Of the students required to take developmental English and who passed, 72.6% subsequently completed freshman English on their first try while only 52.7% of the comparably scoring students who had not taken developmental English completed freshman English on their first try. These findings suggested that the majority of Georgia developmental English students had trouble passing the developmental English course the first semester they enrolled, but that eventually more than half did








pass developmental English. Those that did pass then had a major advantage; they had a substantially better chance of passing freshman composition than they would have had if they had not taken developmental English.

Bers conducted two studies in the fall semesters of 1980 and 1981 on entering students at Oakton Community College. Of the fall 1980 entrants, Bers (1982a) noted that 32% of all entering students scored below the cutoff on the entry-level English test, but only 53% of these voluntarily enrolled in the developmental English course. Of those, only 53% completed with a passing grade whereas 83% of students enrolled in regular freshman English passed. The remedial course was harder to complete successfully than was freshman English. Bers (1982b) found similar persistence problems with fall 1981 developmental English students; 55% passed the developmental English course with 21% withdrawing, a withdrawal rate twice as high as in the overall student body. Of those who passed, 68% enrolled in freshman English. However, having passed developmental English did not give these students an edge over students who needed but had not completed developmental English. The rate at which developmental completers passed freshman English was 89%, but 74% of non-completers passed, not a significant difference.

Baker (1982) had serious reservations about the efficacy of remedial English at two-year Snow College. His only favorable finds were that remedial English students were more likely to complete freshman English than the control group who needed remedial but instead enrolled in freshman English. However, Baker's other findings suggested that remedial English was not clearly a more effective way to








improve student writing skills. For example, scores on an objective English test and on pre- and post-essays graded independently, were not significantly different for the remedial English students and for the control group. In other words, the study suggested that students improved their writing skills as much by avoiding remedial English as by taking it. In terms of retention, the students in regular English fared better: only 14% of them withdrew before the end of the term, whereas 24% of the remedial English students withdrew.


Research Question #3: Developmental Completers Versus Control Groups

Who Did Not Need Developmental English

Reap and Covington (1980) compared the attrition and achievement on a total of 4,900 students in three developmental English courses and the regular freshman English course from the fall of 1973 to the spring of 1980. During this seven year period, the proportion of entering students registered in developmental English steadily increased from 15% to 33% with the rest enrolling in regular freshman English. With three levels of developmental courses, the dropout rate was the worst at the lowest course level, averaging 40%. The middle level did best with a withdrawal rate average of 26%, but the rate again rose to 30% in the highest level of developmental English. This was about the same attrition rate as freshman English experienced at 29%. Combining the withdrawal rates of all developmental courses for 1977-1980, Reap and Covington found that the attrition rate was about equal for them and the freshman English course, 29.4% and 28.5% respectively. Moreover, in terms of achievement, the average grades in each of the courses were about the same, with the exception of the middle level developmental English course, whose average was slightly








higher: 1.7 in lowest level, 2.3 in middle level, 1.9 in highest level developmental, and 1.9 in freshman English. Reap and Covington here reported positive results, but these were short-term measurements, taken during and at the end of one semester. The long-term measures reported were not as positive. Former developmental English students were more apt to complete freshman English successfully if they had been in a developmental course only one semester. Moreover, although an average of 30% of entering students enrolled in a developmental course, only about 16% of the students in freshman English had completed a developmental English course or courses. Obviously, substantial numbers of students who started in developmental English did not persist to enrolling in Freshman Composition.

Braxton et al. (1980) studied the persistence and achievement patterns of developmental studies students at Thomas Nelson Community College from 1975 to 1977. Their program had English, reading, and math courses run by a division of developmental studies. Of the students mandatorily placed in the developmental English course, 62% completed the course. Of the students who completed the course, 71% passed. Of those who did not pass the first time, 48% were successful in passing the second quarter they were enrolled in the course and 49% were successful the third semester. When students had successfully passed the developmental English course, they did almost as well in regular freshman English as students who did not need developmental English. Of the developmental English completers, 68% earned a C or better grade in freshman English compared with 73% of the nondevelopmental students. Braxton et al. also supplied long term data on the persistence of all developmental studies students, including









English students. Of all the students enrolled at the college in the fall quarter of 1975, 62% of non-developmental students reenrolled in the winter quarter, but the higher percentage of 73% of developmental students reenrolled. Exactly two years after their admission, 9% of non-developmental students had graduated, while somewhat fewer, 6%, of the developmental students had graduated. The authors noted that two calendar years were probably not enough time to take accurate measures of graduation rates because many adult part-time students would have needed a longer period to complete their programs, because a large number of students who enrolled there were not degree-seeking students, and because students placed in developmental studies courses had a delayed entry into the regular curriculum which normally prolonged their program beyond two years. As for overall achievement patterns, developmental students' average GPA did lag; it was significantly lower than the GPA for non-developmental students. In all, the developmental English students at this community college did about as well in freshman English as non-developmental students, once the developmental students passed their developmental course. Developmental students persisted to the second quarter at a higher rate than non-developmental students; however, they had a significantly lower GPA, and they graduated within two years at a somewhat lesser rate than non-developmental students.

Smittle (1982) in an overall evaluation of Santa Fe Community College's developmental studies program, found the persistence and achievement of developmental writing students encouraging. Of the entering students in the fall 1981 semester, 76% of those required to take Basic Writing Skills completed the course. This would mean that









24% dropped out of the course before the end of the semester. Of course completers, 75% passed. That meant that of the students required to enroll in developmental writing, 57% passed it on their first try. The following semester, 71% of the students who had enrolled in Basic Writing Skills returned for more college courses. For a long term measure, Smittle compared the passing rate in the winter 1982 College Composition course of successful fall Basic Writing Skills students with that of students who scored above the entry cutoff and had not taken the developmental course. She concluded that the former developmental students, of whom 44% passed College Composition, did as well as the students who had not needed developmental English, of whom 46% passed College Composition. Therefore, Smittle suggested that the developmental English students who passed Basic Writing Skills had about an equal chance for success in the freshman English course as did the stronger students.

Barton (1984) cited mixed results from a study on students entering a developmental education program at a multicampus, urban community college during the fall of 1980. Approximately one-quarter of all entering students enrolled in the developmental program. When compared with non-developmental students, developmental English course completers did not fare well in their subsequent college English course, having a significantly lower mean grade than the nondevelopmental students. However, Barton was encouraged that 68% of the developmental students completed the college level English course with a grade of C or higher. In terms of long range achievement, the developmental program completers had a significantly lower cumulative GPA. The more developmental courses the students took, the lower








their cumulative GPA was apt to be. However, when Barton analyzed the relationship between number of credits earned and cumulative GPA, he found that the cumulative GPA of students who had taken only one developmental course increased as the number of quarters they were enrolled increased, eventually surpassing the cumulative GPA for students who had taken no developmental courses. In long term persistence measures, Barton found that although there was no significant difference in the number of quarters enrolled between full-time developmental and non-developmental students, part-time developmental students did have a significantly higher number of quarters enrolled.


Studies with two control groups

Many studies used two control groups, one of students who fell below the cutoff score on entry testing but did not enroll in developmental English, and one of students who scored above the cutoff, indicating they did not need the developmental course. Both these groups were contrasted with the experimental group of students who needed and enrolled in developmental English.

Dudley (1978) reported that an individualized English remedial skills course at the State University College in Brockport, New York, seemed to have produced positive results when the remedial skills students' achievement and persistence were compared to two control groups, one of students who needed but did not enroll in the remedial English course and one of students who did not need the remedial course based on the entry level English test battery. Dudley used only long term measures: grade in regular freshman English; grade in a freshman speech course; GPA in first, second, third, and fourth









semester of enrollment; and retention rate over the first, second, third, and fourth semesters. There were no differences in speech course grades between students who completed the individualized remedial English course and students in the control groups. But in freshman English, the former remedial English students earned a statistically significantly higher mean grade than did the control group of students who scored above the cutoff on the entry English test and therefore were exempted from remedial English. However, there was no significant difference in the former remedial students' grades in freshman English compared with the grades for students who scored below the cutoff but did not enroll in remedial English. The former remedial English students did, on the other hand, earn a statistically significantly higher GPA than this control group did in their second and third semesters of enrollment. There was, even more encouragingly, no difference between the GPA of the former remedial students and the other control group, those who did not need remedial English. In addition, 35% of the remedial students who persisted to the fourth semester showed an upward trend in GPA while only 9% of students in the control groups did. As for retention, the dropout rate for the controls was higher than for the remedial students: higher for the high scoring control group in the second semester, higher for both control groups in the third semester, and higher in the fourth semester for the low scoring control group of students who should have taken remedial English but did not. Dudley concluded that the individualized remedial English class, a course seemingly quite similar to the one studied in this research, helped some students survive in college who probably would not have without this course.









Linthicum (1979) reported on a statewide evaluation of developmental/remedial programs in Maryland community colleges and ended up questioning the effectiveness of these programs but at the same time admitting that the range of some pieces of data might have been a significant limitation of this research. For example, there was no significant difference in the GPA, retention, and persistence rates between students who had scored low on entry testing and enrolled in developmental English and students who had scored low but did not enroll in it. But Linthicum noted that since different institutions used different cutoff scores on the entry level test, the student populations may have been too disparate to have been analyzed as a group. In the comparison of developmental English students with the other control group, the students who did not need developmental English, the developmental students did not fare well. In achievement, developmental English students achieved a cumulative GPA ranging from 1.66 to 2.91 at various institutions while the stronger students achieved GPAs between 2.58 and 3.21. Developmental English students earned a significantly lower grade in freshman English than non-developmental students did. On the brighter side, there was no significant difference between the length of time developmental English students and stronger English students continued taking courses in college. Although the ratio of courses completed to those attempted statewide was 76% for the developmental English students but 90% for the stronger students, Linthicum cautions that this reflects a large disparity between institutions (59% to 94% for developmental students, 75% to 96% for non-developmental students), which was due to variations in withdrawal policies.









In New Jersey, statewide evaluation of what are called remedial programs has had more positive results (New Jersey Basic Skills Council, 1983). As in Maryland, the results for individual community colleges, colleges, and universities were uneven with substantial ranges of positive to negative results, but the overall picture was reported as pointing to some success for the state's postsecondary remedial programs. For example, of the full-time students in community colleges, 65% passed their remedial writing course statewide. The range among the community colleges was wide, however, running from a low of 46% passing at one institution to a high of 82% at another. For parttime students, the results were not quite as favorable with an average passing rate of 59% and a range of 28% to 90%. Moreover, community college students who completed remedial writing dropped out of college after their first semester at a much lower rate (14%) than did either the students who needed but did not complete remediation (33%) or the high scoring students who did not need remedial writing at entry (20%). In terms of completion of regular freshman English courses, community college students as a whole who completed remedial English passed regular freshman English at about the same rate as those who did not need remedial writing and better than those who needed but did not complete remedial writing. In terms of GPA, across all community colleges, remedial writing completers earned a mean GPA of 1.93, lower than the high scoring entry students' mean of 2.31 but higher than the mean of 1.32 for low scoring students who did not complete remedial writing. Instead of the mean number of credits earned, this report also listed the ratio of credits earned to credits attempted. Again, the remedial writing completers did not fare as well as the stronger








students at .78 to .81 respectively, but the completers did do better than the students who needed but did not complete a community college remedial writing course at .63.

The authors of the New Jersey study also included data for reading, computation, and elementary algebra courses and for all public institutions of higher education in the state. They noted that the wide variations in statistics reported from various institutions must not be overlooked in drawing conclusions about remedial programs at specific colleges. They urged that attention be paid to the characteristics of the more successful programs in an attempt to upgrade the weaker ones. Finally, they cautioned that it may be too early to tell about the overall effectiveness of the state's remedial programs and that further one and two year follow-up studies will help clarify how effective the programs have been.


Research Questions #4: Age as a Factor for Developmental Students

Linthicum (1979), in a study noted earlier, also found that when results were factored by age, older students who completed developmental English did just as well in freshman English as the control group of students who did not need a developmental course. The developmental students over 30 were significantly more successful than the 19-21 year olds, who had the lowest entry test scores in the age breakdown. In addition, developmental course completers age 22 or older achieved the same mean grade in freshman English, had the same retention, persistence, and cumulative GPA patterns as students who never needed developmental English. These older students comprised only 35% of the developmental English students, however, and the








majority of the developmental students were age 19-21, in the lowest scoring of all Linthicum's age categories. Therefore, although the 22 and older students' performance was equal to the performance of the non-developmental English students, the performance of the younger students brought the overall performance of developmental students below that of the control group.

Whittle (1980) found that at Piedmont Virginia Community College, there was no relation between age and grade in the developmental English course. The students who were most successful completed only one quarter of developmental English. The chance for success diminished with each successive quarter a student spent in developmental English. In freshman composition, however, the highest grades were received by those former developmental students who were enrolled part-time, took fewer quarters of developmental English, were older and were female, in that order, according to Whittle's weighted linear combination of predictor variables correlating success in developmental English and in freshman composition. This result also suggested that age was a factor in former developmental students' success in regular courses and that older students later achieved at a higher level than recent high school graduates.

Dumont and Jones (1983) also did a predictive analysis study involving age as a factor in developmental course success. As part of a statewide study on selected remedial education programs in a Southeastern state, Dumont and Jones found at one regional university that age was the most important predictor in students' success in the remedial composition course. Older students did better than younger students in earning grades of C or higher. The other predictors were, in








descending importance, the student's confidence in his or her ability to master composition, sex (females did better), score on the locally developed composition pre-test, the student's perception of the course's usefulness, and race.



Summary
Although some studies cited encouraging results of developmental English programs, others did not. For example, the rate of successful completion of developmental English courses seemed to range between 50% to 70%. About one-fourth to one-third of successful completers did not go on to enroll in freshman English.

When contrasting results for individualized versus classroom modes of developmental courses, these studies showed mixed results. Students in both the classroom and individualized courses completed their developmental English at about the same rate in the three studies measuring this. On the other hand, one study cited higher grades in freshman English and higher GPA for individualized course completers while another reported the opposite, that classroom course completers had higher freshman English grades and GPA. None of the studies showed significant difference in persistence when using longterm measures.

More mixed results occurred among the studies with a control group of students who needed but did not complete developmental English. In performance in regular freshman English, developmental completers did better either in passing or in attaining a mean higher grade than the controls in four studies but showed no significant difference in three reports. Of the studies measuring GPA, three








cited significantly higher cumulative GPA for the developmental completers while three reported no significant differences. Persistence over two or more semesters was higher for former developmental students in three studies, about equal to the controls in one study, but lower than the controls in two studies. The only two reports measuring credits or units earned both cited significantly higher accumulation for the former developmental students. Several authors noted, however, that the high attrition rate among students who originally enrolled in the developmental course, a fate which the control students had not suffered, could have contributed to this positive result. Of course, these studies were conducted at a mixture of community colleges, four-year colleges, and universities. As Kulik et al. cautioned in their meta-analysis on cumulative GPA and persistence, community college studies by and large resulted in no significant differences between developmental program completers and control groups of similar aptitude who did not complete developmental programs.

Developmental completers were more likely to do worse than the control groups of stronger students who did not need developmental English, but occasionally held about equal with them. Developmental completers earned a higher grade in regular freshman English than stronger students in only one study. In two studies, the no significant difference in freshman English grade may also be viewed as encouraging since the developmental English completers originally had started college with a measurable disadvantage in English skills. Two studies did, however, find the developmental completers earning significantly lower grades. Also encouraging were the two studies








measuring completion rate in freshman English; both reported no significant difference between former developmental and stronger students. On longer term measures, developmental students did not fare as well with four out of five studies having significantly higher GPAs for the stronger students (one with no significant difference). The only study citing credits earned showed stronger students as higher, and the only study citing degrees earned showed stronger students as higher also. Finally, the number of semesters of persistence was more encouraging with three studies reporting no significant difference and one reporting partially better persistence for developmental completers.

In the three studies involving older versus younger developmental students, older students fared better than younger on the few measures used. Grade in developmental course was higher for older students in one study but showed no significant difference in another. A higher proportion of older students completed their developmental English course in the one study while being older was one of four factors related to higher completion rates in another. For long term measures, one study cited older students as having higher cumulative GPAs and more semesters completed than younger students.

With mixed results like these, in some cases involving relatively few studies applicable to the research questions in this study, there emerges a need for more research on the effects of developmental English programs, especially in Florida and other states where such programs in all disciplines are coming under increasing legislative scrutiny and mandates.














CHAPTER 3
METHODOLOGY

General Research Design

This ex post facto study employed a quasi-experimental, nonequivalent control group design. Archival data were the source for pretest scores partially responsible, along with self-selection, for placement of fall, 1980, Valencia Community College students into the four groups in this study described below. Student transcripts were the source for the records of student persistence and achievement.



Instrumentation

Data were collected from computer tapes of entry-level testing results from fall 1980, and from the college's student data base. The records included the Initial English Assessment (IEA) test scores; demographic data on age, sex, and race; and student transcripts.

The Initial English Assessment Test at Valencia Community College (see Appendix A) was developed and validated by a team of the college's English instructors under the direction of a consultant expert in English objective test writing. Reliability was established and cutoff scores were set with the help of the Vice-President of Institutional Research. The cutoff scores were as follows:

22-35 High Scores -- students should enroll in Freshman Composition I.

19-21 Inconclusive Scores -- students may enroll in either a developmental English course or in Freshman Composition I.









0-18 Low Scores -- students should enroll in a developmental English course.
Academic advising occurred after a student had submitted an application to the college and completed the entry level placement tests in English, reading, mathematics, and study skills. During the academic advising session, a member of the counseling staff reviewed the student's Initial English Assessment test score, along with the other initial assessment test scores in reading, mathematics, and study skills. The counselor also reviewed the student's transcript from secondary school, if available, and discussed which courses the initial assessments indicated the student should take.

The data were organized by the English course status of students during their first semester, that is, whether their entry test scores indicated a need for developmental English course work and they enrolled in the classroom, individualized, or no developmental course or whether their entry test scores indicated no need for developmental English course work and they enrolled in regular Freshman Composition I. The data were further organized by student age at entry to the college, as reported by the student.



Sample
The total population involved in this study consisted of 1,431 Valencia Community College students with the following characteristics:

(1) enrolled at the college for the first time in the fall semester of the 1980-1981 school year,
(2) were first-time-in-college students, not transfers, (3) completed the Initial English Assessment test, and
(4) registered for classes on the West Campus.

The 1,431 students were divided into four groups on the basis of their Initial English Assessment score and their first semester English









course status. Then, 100 students in each group were randomly selected. Included were both full-time and part-time students and both day and evening students.

The 400 students in the four random samples were enrolled in the following courses in Session I, 1980-1981:

100 in the classroom course who needed a developmental course (IEA
Score of 0-18),

100 in the individualized course who needed a developmental course
(IEA Score of 0-18),

100 not in an English developmental course even though their entry
assessment scores indicated they needed one (IEA score of 018),
100 in regular Freshman Composition I because their scores were
over the cutoff (IEA score of 22-35).

Students with inconclusive scores of 19-21 were excluded from the study.

Of the 100 not in an English developmental course even though their entry assessment scores indicated they needed one, about fourfifths enrolled directly in Freshman Composition I while about onefifth enrolled in no English course at all in the fall, 1980, semester.



Collection of the Data
Data were collected from June 1980, when the first round of Initial English Assessment testing and academic advisement sessions were conducted for students entering in the fall session of 1980-81, through April, 1984, a span of eleven semesters, nearly four academic years.

Two reasons existed for conducting a nearly four year study on students in a community college. First was to compensate as much as possible for the extended time period many part-time students need to complete their associate degree. Second was to compensate for the









possible semester or more that students originally enrolling in developmental courses often need to add to their academic careers since they must delay taking one or more of the standard freshman courses while they strengthen their skills in developmental courses.

The following data were collected: student Social Security number; age at entry, sex, and race as reported by student; score on Initial English Assessment; which developmental English course enrolled in during Session I, 1980-81, if any; completion of that course; grade in that course; enrollment in Freshman Composition I; grade in Freshman Composition I; cumulative GPA; cumulative number of semesters completed; cumulative number of credits earned; and associate degree earned, if any.



Analysis of the Data

Four groups were studied in order to answer the first three research questions concerning achievement and persistence of developmental students in each course and of the developmental students versus students in the control groups. Two were experimental groups; two were control groups. The two experimental groups consisted of the following:

DC - developmental classroom students who needed a developmental
course and enrolled in the classroom course (IEA score of 018), and

DI - developmental individualized students who needed a developmental course and enrolled in the individualized course (IEA
score of 0-18).

The two control groups consisted of the following:

CNO - control group of no developmental course students who needed
a developmental course but did not enroll in either the individualized or the classroom course (IEA score of 0-18), and









CHI - control group of high scorers on the IEA who neither needed
nor enrolled in a developmental course because their scores
were above the high cutoff score (IEA score of 22-35).

IEA scores of 19-21 were judged to be inconclusive. Counselors recommended each student to take either Freshman Composition I or a developmental English course depending on other factors like high school

English grades. Students with inconclusive scores were, therefore, not

included in this study. Figure 1 represents the relationship between

the four groups by IEA score.



EXPERIMENTAL IEA CONTROL
GROUPS SCORE GROUPS


DC
Devel opmental
classroom
group 0-1
0-13


DI CNO
Developmental Control with
individualized no developgroup mental course




CHI
Control with
22-35 _ high scores & no need for
developmental
course


Figure 1. Diagram of Groups for Research Questions 1, 2, and 3 as Determined by Score on Initial English Assessment Test


The two developmental groups were combined and then redivided according to age in order to answer the fourth research question









concerning achievement and persistence of students four or more years out of high school versus that of students who have recently graduated from high school. Figure 2 depicts this realignment.


DC
Developmental
Classroom
Students


+



DC
Developmental Individualized
Students


DC + DI
Age 17-21








DC + DI
Age 22 or Older


Figure 2. Diagram of the Control Group Realignment for Research Question 4 as Determined by Student Age at Entry


A frequency table was constructed, and the descriptive statistics of mean, standard deviation, median, mode, range, skewness, and kurtosis were computed for the IEA scores over all four groups. Descriptive statistics of sex, race, and age were noted over all four groups.

The CHI group was nonequivalent to the other three groups since its members had to have scored above the cutoff on the Initial English Assessment in order to be members of this group, whereas membership in the other three groups was partially determined by scores below the cutoff. Mean scores on the IEA by groups were analyzed using analysis of variance followed by the Scheffe Procedure to determine if any other








groups had means with significant differences. One-way analysis of

variance was also applied to means of the developmental groups divided by age.

The independent variable was the score on the Initial English Assessment. The dependent variables were divided into three sets: those to which one-way analysis of variance was applied, those to which multivariate analysis of variance was applied, and those to which chisquare followed by the Yates Correction was applied. One-way analysis of variance (ANOVA) was used to establish if any statistically significant differences at the .05 level existed for the following dependent variables:

grade in developmental course
grade in Freshman Composition I.

Multivariate analysis of variance (MANOVA) followed by multivariate F tests was used to establish if any statistically significant differences at the .05 level existed between the values of the long-term variables measured in each of the groups. MANOVA allows possible interactions among the dependent variables to be evaluated. When the multivariate F value following MANOVA was significant, univariate F tests were used to determine which variables were significantly different in and of themselves. The set of dependent variables analyzed with MANOVA included the following:

cumulative grade point average (GPA),
number of semesters enrolled in nearly four years,
number of credits earned in nearly four years.

Chi-square analysis followed by the Yates Correction was applied to the third set of variables:

number and percent of students completing a developmental course,
number and percent of students completing Freshman Composition I
number and percent of students completing an associate degree.















CHAPTER 4
FINDINGS

This study was designed to evaluate a community college developmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an individualized format, so that the college might assess the efficacy of each course and of the entire program in terms of student persistence and achievement. Four main research questions were asked about achievement and persistence patterns of the developmental English students. Eight null hypotheses, two relating to each research question, were tested.
First, comparisons were made between students enrolling in the classroom developmental English course and the individualized developmental English course to see if an individualized developmental writing course or a classroom based developmental writing course contributed to better achievement and longer persistence by students who needed a developmental writing course when they started college. Second, comparisons were made between students completing a developmental course and those who needed but did not enroll in either developmental course to see if students who needed and completed either of the developmental writing courses achieved better and persisted in college longer than students who needed but did not enroll in a developmental writing course. Third, comparisons were made








between students completing a developmental course and students with stronger writing skills who did not need a developmental writing course to see if the developmental writing courses helped students with weak writing skills at entry catch up, in terms of achievement and persistence, with students who did not need a developmental writing course when they started college. Finally, comparisons were made of developmental course completers divided into two age groups, those age 17 to 21 and those age 22 or older, to see if older adults achieved and persisted any differently from recent high school graduates after completing a developmental writing course.

This chapter presents the data collected in this study. There are five sections. The first describes background data on the four groups and correlations among independent variables across the four groups. The remaining four sections address each of the four main research questions and present statistical evidence that either proved or disproved each of the study's eight null hypotheses.



Background Analysis of Samples

When all four groups were analyzed, the following characteristics emerged. Females made up 55% and males 45% of the samples. Whites composed 75.7%, Blacks 17.5%, American Indians 2.5%, Asians 0.7%, Hispanics 3.0%, and others 0.5% of these students. Students aged 17 to 21 comprised 82.5% of the samples while students 22 or older made up the remaining 17.5%. Their scores on the Initial English Assessment test had a mean of 17.0. The test scores for the combined groups did not form a normal distribution, as can be seen in Table 1, because three of the four groups were, as planned, chosen from students








scoring at or below the cutoff score of 18 on the IEA. The fourth group consisted only of students scoring at or above the high cutoff of 22 on the IEA. Consequently, the students with inconclusive scores of 19 to 21 were excluded from the study. TABLE 1

A Descriptive Analysis of Initial English Assessment Test Scores


Mean 17.02 Median 16.00 Standard Deviation 5.23 Mode 17.00 Range 32.00 Skewness .53 Minimum 3.00 Kurtosis .10 Maximum 35.00



In order to test the assumption that the groups under study were equivalent, except for the control group of students scoring high on the entry test, one-way analysis of variance (ANOVA) was performed on the mean score on the Initial English Assessment test for each of the four groups. The resultant F value was significant, so the Scheffe Procedure was applied to indicate where the differences occurred. As shown in Table 2, there was no significant difference in the mean score of the classroom course students at 14.16 and the individualized developmental course students at 13.67. The individualized developmental group did have a significantly lower mean at 13.67 than did the control group of "No-takes," (students who needed but did not enroll in a developmental course) at 15.19. There was, as expected, a significant difference between the mean of the control group of high scoring students at 24.60 and the means of the other groups.









TABLE 2

Results of Analysis of Variance on the Mean Scores on the Initial English Assessment Test by Groups

Source df SS MS F

Between Groups 3 7783.58 2594.53 328.33* Within Groups 396 3129.29 7.90 Total 399 10912.88 Scheffe Procedure

Group Mean DC DI CNO CHI DC (developmental classroom) 14.16 DI (developmental individualized) 13.67 CNO (control, No-takes) 15.19 * CHI (control, high scorers) 24.60 * * *
* p < .05


In addition to the significant difference between the individualized developmental group mean on the IEA and that of the control group of No-takes, there was also a significant difference in the means on the IEA when the individualized developmental group and the classroom developmental group were combined. This combined experimental group was important in answering the second research question. As seen in Table 3, the combined developmental groups' mean of 14.14 was significantly lower than the No-takes' mean of 15.19.

As for the developmental students divided according to age, there was also no significant difference at p < .05 between the mean score on the entry test between younger and older students as recorded in Table 4.









TABLE 3

A Comparison of the Mean Scores on the Initial English Assessment Test Between the Combined Developmental Groups and the Control Group of Notakes
Standard
Group Mean Deviation N DC + DI (Combined 14.14 2.91 200 Developmental
Groups)

CNO (No-takes) 15.19 2.86 100 Source df SS MS F Between Groups 73.50 1 73.50 8.78* Within Groups 2494.47 289 8.37 p < .05


TABLE 4

A Comparison of Developmental Writing Students' Mean Scores on the Initial English Assessment Test by Student Age at Entry
Standard
Age Mean Deviation N

17-21 14.17 2.89 168

22+ 14.00 3.06 32

Source df SS MS F Between Groups 1 .75 .74 .09 Within Groups 198 1683.33 8.50 p < .77


These findings were important to all of the following comparisons pertaining to the major research questions and their null hypotheses. Since there were no significant differences between the IEA means of the two experimental developmental course groups and between the younger developmental course students and the older developmental









course students, the independent variable of IEA score for these groups was essentially equal. The inequality of the mean IEA score between the combined developmental groups and the control group of Notakes implies that the No-takes had somewhat stronger writing skills at entry than students electing to enroll in a developmental writing course, which has implications in the interpretation of later analyses. The inequality of the mean IEA score between the combined developmental groups and the control group of high scoring students underscored the continued inequality in some of the dependent variables measured between these two groups.


Research Question #1:

Classroom Versus Individualized Developmental Course

The first research question was concerned with whether or not an individualized developmental writing course or a classroom based developmental writing course contributed to better achievement and longer persistence by students who needed a developmental writing course when they started college. The first hypothesis stated that there was no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course.

Table 5 presents the results of analysis of variance applied to grades in the two different developmental courses. Of the 100 students enrolled in the classroom course, 86 completed it that semester.









Of the 100 students enrolled in the individualized course, 75 completed. Of the completers, the mean grade in the classroom course was 2.85, while the mean grade in the individualized course was 3.05. Analysis of variance resulted in an F value of 2.08, which was not significant at p < .05.


TABLE 5

A Comparison of the Mean Grade in Developmental English Courses Between Students in the Classroom and Individualized Courses
Standard
Group Mean Deviation N DC Classroom 2.85 .93 86 DI Individualized 3.05 .86 75


Sourc Betwe

Within p < .]


e df SS MS F en Groups 1 1.68 1.67 2.08 n Groups 159 128.82 .80 15


The mean grade which developmental course completers earned in Freshman Composition I also did not show any significant difference between students who completed the classroom or the individualized developmental course (see Table 6). Of the 86 classroom course completers, 65 enrolled in Freshman Composition I and 49 completed it with a mean grade of 2.51. Of the 75 individualized course completers, 57 enrolled in Freshman Composition I and 44 completed it with a mean grade of 2.73. Multivariate analysis of variance was applied to the variables of grade in Freshman Composition I, GPA, number of semesters enrolled, and number of credits earned. MANOVA resulted in a multivariate F of .48, which meant there were no









significant differences at the .05 significance level on any of these measures between the two developmental groups. TABLE 6

A Comparison of Mean Grade in Freshman Composition I, GPA, Semesters, and Credits of Developmental Completers of the Classroom and Individualized Courses


Variable

GRADE IN FRESHMAN COMP. I GPA


SEMESTERS



CREDITS


Multivariate F = .48; di


Group

DC DI DC DI DC DI DC DI f = 5,


Mean 2.51 2.73 2.46 2.54 5.16 5.29

39.94 41.61

89; p < .79


Standard Deviation

1.05

.84 .47 .50

2.25 2.52

22.00 21.71


N


The cumulative GPA of these developmental course and Freshman Composition I completers also showed no significant difference, as seen in Table 6. Former classroom developmental completers amassed a GPA mean of 2.46 while former individualized developmental completers amassed a 2.54. MANOVA showed a non-significant multivariate F of .48.

Consequently, in terms of achievement, of those students who completed their respective developmental writing course, the mean grade they earned in the developmental course, the mean grade in Freshman Composition I, and the mean cumulative GPA showed no significant difference between groups. Therefore, the first hypothesis--that there








was no significant difference in achievement (in terms of mean grade in developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between the students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course--was not rejected. The students in both courses achieved about equally as well.

The second hypothesis concerned persistence, that there was no significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individualized course and students who needed a developmental writing course and enrolled in the classroom course.

As seen previously in Table 6, the mean number of semesters completers stayed enrolled at the college and the number of credits they earned were not significantly different between the two courses. The classroom completers stayed enrolled a mean of 5.16 semesters and earned a mean of 39.94 credits while the individualized completers stayed enrolled a mean of 5.29 semesters and earned a mean of 41.61 credits. The MANOVA analysis yielded a non-significant multivariate F of .48 at the .05 level.

Table 7 shows that of the 100 students enrolled in the classroom developmental course, 86 (86%) completed it while individualized course completers numbered 75 (75%). This resulted in a chi-square









value of 3.18 after the Yates Correction; this was non-significant at the p < .05 value with 1 degree of freedom.


TABLE 7

A Comparison of the Completion Rate of Developmental Courses Between the Classroom and Individualized Groups
Original
Group Number Completers %

DC 100 86 86.0% DI 100 75 75.0% Chi-square after Yates Correction = 3.18; df = 1; p < .07

Also non-significant was the difference between the number of developmental course completers who went on to enroll in and complete Freshman Composition I, as seen in Table 8. Of the 86 classroom completers, 65 enrolled in Freshman Composition I and 49 (75.4% of those completers who enrolled) finished Freshman Composition I with a grade of A, B, C, or D. Of the 75 individualized completers, 57 enrolled in Freshman Composition I and 44 (77.2% of completers who enrolled) completed it successfully. This resulted in a chi-square value of .00 after the Yates Correction, which does not equal a significant p < .05 value at 1 degree of freedom.


TABLE 8

A Comparison of the Completion Rate of Freshman Composition I Between Students Who Completed the Classroom and the Individualized Developmental Courses
Original Developmental Enrolled in Completed
Group Number Completers Fresh. Comp. I Fresh. Comp. I %

DC 100 86 65 49 75.4% DI 100 75 57 44 77.2% Chi-square after Yates Correction = .00; df = 1; p < .98








Finally, there was also no significant difference between the rate at which developmental course completers earned an associate degree (see Table 9). At the end of 11 semesters, 14 of the 86 classroom course completers (16.3%) had earned a degree while 12 of the 75 individualized course completers (16.0%) had earned a degree. That equalled 14% of the original 100 classroom developmental students and 12% of the original individualized developmental students. The chisquare value of .00 after the Yates Correction meant no significance at the p < .05 level.


TABLE 9

A Comparison of the Degrees Earned by Developmental Classroom and Individualized Course Completers

Earned % of
Group Completers Degree Completers

DC Classroom 86 14 16.3% DI Individualized 75 12 16.0% Chi-square after Yates Correction = .00; df = 1; p < 1.00


Therefore, the second hypothesis--that there was no significant difference between the persistence of students completing the classroom developmental writing course and the individualized developmental writing course--was not rejected. The students completing both courses seem to have persisted about equally as well.

In sum, no significant differences appeared on any of the measures of achievement and persistence between students who enrolled in the classroom developmental writing course and those who enrolled in the individualized developmental writing course. The answer to the first question is that the individualized course and the classroom








course contributed about equally to the achievement and persistence of students who needed a developmental writing course when they started college.


Research Question #2:

Developmental Groups Combined Versus Control Group of No-takes

The second research question was concerned with whether or not students who needed and completed either of the developmental writing courses achieved better and persisted in college longer than students who needed but did not enroll in a developmental writing course. The third hypothesis stated that there was no significant difference in achievement (in terms of grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course (the "No-takes") and students who needed and completed a developmental writing course.

With regard to achievement measures, there was no significant difference between the Freshman Composition I grade of developmental completers and that of the No-takes who had enrolled in Freshman Composition I (see Table 10). Of the classroom developmental completers, 49 also completed Freshman Composition I. Of the individualized developmental completers, 44 also completed Freshman Composition I. Together, 93 students completed both a developmental course and Freshman Composition I. Of the original 100 control group of No-takes, 78 enrolled in Freshman Composition I (22 enrolled in no English course during the fall 1980 semester). Of those 78, 59 completed Freshman Composition I with a grade of D, C, B, or A. The mean grade of the developmental completers was 2.61 while the mean grade of the control









group of No-takes was 2.63. The F value of .01 after analysis of variance was not significant at the .05 level. TABLE 10

A Comparison of the Mean Grade in Freshman Composition I of Combined Developmental Groups Versus Control Group of No-takes
Standard
Group Mean Deviation N DC + DI 2.61 .96 93 CNO 2.63 .81 59

Source df SS MS F Between Groups 1 .01 .01 .01 Within Groups 150 124.39 .81 p < .91


The one significant difference which did occur was between the cumulative GPA of the combined developmental completers and the control group of No-takes as seen in Table 11. In order to control for multiple interactions between the dependent variables of cumulative GPA, number of semesters of persistence, and number of credits earned, all of which were long-term measures, multiple analysis of variance was applied. The developmental completers totalled 161 (86 who completed the classroom course and 75 who completed the individualized course). They were compared to the 100 students in the control group of No-takes. The mean GPA for the developmental completers was 2.39 while the mean GPA for the control No-takes was 2.00, which proved to be a significant difference. MANOVA analysis yielded a significant multivariate F of 4.01. Further analysis with univariate F tests showed significance at the .05 level only for the GPA and not for the








number of semesters enrolled or for the number of credits earned. Therefore, the developmental completers did show a significantly higher cumulative GPA than did those students who should have enrolled in a developmental writing course but did not.


TABLE 11

A Comparison of the Mean GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups and the Control Group of No-takes
Standard
Variable Group Mean Deviation N GPA DC + DI 2.39 .63 161 CNO 2.00 1.21 100 SEMESTERS DC + DI 4.31 2.57 161 CNO 3.95 2.69 100 CREDITS DC + DI 30.37 22.94 161 CNO 25.25 24.36 100 Multivariate F = 4.01; df = 3, 257; p < .01* Univariate F Tests
Variable Univariate F df p

GPA 11.34 * 1, 259 .01 *

SEMESTERS 1.17 1, 259 .28 CREDITS 2.93 1, 259 .09
* p < .05


As a result, the third hypothesis--that there is no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course and students who









needed and completed a developmental writing course--was partially rejected because of the analysis related to GPA. The developmental completers did achieve a higher cumulative GPA than the control group of No-takes. In addition, although the developmental groups started with significantly lower writing skills as measured by the IEA, developmental completers did earn about the same grade in Freshman Composition I as the No-takes.

The fourth hypothesis was that there was no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course.

In terms of persistence, there were no significant differences. Not all students who completed a developmental course enrolled in Freshman Composition I. Table 12 shows that of the original 100 in each developmental group, 86 and 75 respectively completed the classroom and individualized courses. Of these, 65 classroom course completers and 57 individualized course completers totalled 122 developmental completers who enrolled in Freshman Composition I. Of this total, 93 (76.2%) finished Freshman Composition I with a D, C, B, or A grade. On the other hand, of the 100 control group of No-takes, 78 enrolled in Freshman Composition I and 59 (75.6% of those enrolled) finished it that semester. The difference was not significant at the p < .05 level according to chi-square analysis which resulted in a value of .00 after the Yates Correction.









TABLE 12

A Comparison of the Rate of Completion of Freshman Composition I for Combined Developmental Groups Versus Control Group of No-takes
Original Completed Enrolled in Completed
Group Number Developmental Fresh. Comp. I Fresh. Comp. I %

DC 100 86 65 49 75.4% DI 100 75 57 44 77.2% DC + DI 200 161 122 93 76.2% CNO 100 - 78 59 76.6% Chi-square after Yates Correction = .00; df = 1; p < 1.00


As noted in Table 11, no significant difference existed between the mean number of semesters enrolled and the credits earned in 11 semesters between the combined developmental groups and the control No-takes. Table 11 lists the mean number of semesters enrolled for the developmental completers as 4.31 while the value for the No-takes was 3.95. Likewise, the mean number of credits earned was 30.37 for the developmental completers while the value for the No-takes was 25.25. While these numbers are slightly higher for the developmental completers than for the No-takes, the differences are not significant at the p < .05 level.

In numbers of students earning an associate degree, no significant difference existed either between the combined developmental groups and the control No-takes. Table 13 shows that of the 161 combined developmental completers, 26 (16.1%) earned an associate degree while of the 100 control group of No-takes, 17 (17%) earned a degree within the 11 semesters of this study. With a chi-square value after the Yates Correction of .00, there was no significant difference between these rates at the .05 level.









TABLE 13

A Comparison of Associate Degrees Earned by Developmental Completers Versus Control Group of No-takes

Original Developmental Earned Group Number Completers Degree DC + DI 200 161 26 16.1% CNO 100 - 17 17.0% Chi-square after Yates Correction = .00; df = 1; p < .99


Therefore, the fourth hypothesis--that there was no significant difference in persistence between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course--was not rejected. Although the developmental group started with lower mean entry test scores than the No-takes, the developmental completers equalled the No-takes in terms of both short and long term persistence.

Therefore, one significant difference did occur in the measures of achievement (cumulative GPA for developmental completers was higher) but none in the measures of persistence, between the performance of the combined developmental groups (DC + DI) and the control group of No-takes (CNO), students whose scores on the IEA indicated that they needed a developmental writing course, but they did not enroll in one.


Research Question #3:
Developmental Groups Combined Versus High Scorers

The third research question was concerned with whether or not the developmental writing courses helped students with weak writing skills








at entry catch up, in terms of achievement and persistence, with students who did not need a developmental writing course when they started college. The fifth hypothesis stated that there was no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course.

On these achievement measures, the high scorers (CHI) continued scoring high (see Table 14). Their mean grade in Freshman Composition I and cumulative GPA were both higher than for the developmental completers. The developmental completers' mean Freshman Composition I grade was 2.61 whereas the high scorers' mean was 3.28, a significant difference when analysis of variance yielded an F value of 24.12.


TABLE 14

A Comparison of the Mean Grades in Freshman Composition I Earned by Developmental Completers Versus High Scorers
Standard
Group Mean Deviation N

DC + DI 2.61 .96 93 CHI 3.28 .84 80

Source df SS MS F Between Groups 1 19.83 19.83 24.12* Within Groups 171 143.06 .82
* p < .05


The mean GPA of 2.66 earned by the high scorers also was significantly higher than the GPA of 2.39 for the developmental completers









(see Table 15). When the long term measures of GPA, number of semesters enrolled, and number of credits earned were analyzed with multivariate analysis of variance, the multivariate F of 3.75 was significant at the p < .05 level, indicating that one or more of the pair of means were significantly different. When univariate F tests were applied, the value for cumulative GPA of 6.02 was significant at the .05 level.


TABLE 15

A Comparison of Cumulative GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups Versus the Control Group of High Scorers


Variable

GPA



SEMESTERS



CREDITS



Multivariate F = .01;


SI


Group Mean D DC + DI 2.39 CHI 2.66 DC + DI 4.31 CHI 4.48 DC + DI 30.37 CHI 36.43 df = 3, 257; p < .01*

Univariate F Tests


standard aviation

.63
1.11 2.57 2.59 22.94 27.55


N

161 100 161 100 161 100


Variable Univariate F df p

GPA 6.02 * 1, 259 .01 *

SEMESTERS .27 1, 259 .61 CREDITS 3.68 1, 259 .06
* p < .05








Consequently, the fifth hypothesis--that there was no significant difference in achievement between developmental completers and students who neither needed nor enrolled in a developmental writing course--was rejected. Developmental writing course completers, as a group, earned a significantly lower grade in Freshman Composition I and a significantly lower cumulative GPA than students judged as not needing a developmental writing course at entry. Completing a developmental course did not help them "catch up" to students who started college with significantly higher abilities in English as measured by the IEA.

The sixth hypothesis stated that there was no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course.

In terms of persistence, three of the four measures showed no significant differences between the developmental completers and the high scorers. First, the rate at which each type of student completed Freshman Composition I was statistically similar. Table 16 shows that of the 122 developmental completers who enrolled in Freshman Composition I, 93 (76.2%) completed with a grade of D, C, B, or A while 80 (80.0%) of the high scorers did. This yielded a chi-square value of .26 after the Yates Correction, which was not significant at the p < .05 level.








TABLE 16

A Comparison of the Completion Rate of Freshman Composition I for Developmental Completers Versus the Control Group of High Scorers

Original Completed Enrolled in Completed
Group Number Developmental Fresh. Comp. I Fresh. Comp. I %

DC + DI 200 161 122 93 76.2% CHI 100 - 100 80 80.0% Chi-square after Yates Correction = .26; df = 1; p < .61


Likewise, as seen above in the reporting of MANOVA applied to long term measures, found in Table 15, the mean number of semesters enrolled by the two groups showed no significant difference at the .05 level. The developmental completers enrolled for an average of 4.31 semesters while the high scorers enrolled for an average of 4.48 semesters. Similarly, the number of credits earned by each group, at 30.37 for the developmental completers and 36.43 for the high scorers, was not significantly different either.

However, significantly more high scorers received an associate degree within the 11 semesters of this study. Table 17 illustrates that of the 161 developmental completers, 26 (16.1%) earned a degree while of the 100 high scorers, 32 (32.0%) earned one. The chi-square value of 8.07 after the Yates Correction was a significant difference.


TABLE 17

A Comparison of the Associate Degrees Earned by Developmental Completers Versus Control Group of High Scorers
Original Developmental Earned
Group Number Completers Degree % DC + DI 200 161 26 16.1% CHI 100 - 32 32.0% Chi-square after Yates Correction = 8.07; df = 1; p < .05*









As a result, the sixth hypothesis--that there was no significant difference in persistence between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course--was rejected only in terms of the number of degrees earned. Developmental completers received fewer associate degrees than students who had stronger writing skills at entry. However, developmental completers, even though they scored significantly lower on the entry English test, appear to have completed Freshman Composition I at about the same rate, to have stayed enrolled at the college about the same number of semesters, and to have earned a similar number of credits as students who scored significantly higher on the Initial English Assessment Test.

In summary, significant differences occurred in both measures of achievement but in only one of the four measures of persistence when the combined developmental groups (DC + DI) were compared with the group of students who scored above the cutoff score, on the Initial English Assessment Test. These high scoring students were judged as not needing a developmental writing course and enrolled directly into Freshman Composition I. That high scorers had a higher mean in the achievement measures and in half of the persistence measures is not surprising given their high performance on the entry English test. However, on three of the persistence measures, there was no significant difference between the developmental completers, who were low scorers, and the high scorers.








Research Question #4:
Younger Versus Older Developmental Students

The fourth research question was concerned with whether or not, after completing a developmental writing course, older adults achieved and persisted any differently from recent high school graduates. The seventh hypothesis stated that there was no significant difference (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21.

Table 18 lists the results of the multiple analysis of variance used to control for interaction among dependent variables of grade in developmental course, GPA, number of semesters enrolled, and number of credits earned in the 11 semesters of the study. The results of the MANOVA analysis yielded a multivariate F of 3.22, which indicated that a significant difference at the .05 level was at work across the variables. To pinpoint where the difference lay, univariate F tests were applied.

As seen in Table 18, younger and older students earned about the same mean grade in a developmental course since the univariate F value of 2.50 means there was no significant difference at the .05 level between the younger students' mean of 2.89 and the older students' mean of 3.19. The cumulative grade point average for the two groups, however, was significantly different. Cumulative GPA for younger students averaged 2.33, while the cumulative GPA for the older students was significantly higher at 2.70, yielding a significant F value of 7.47.









TABLE 18

A Comparison of Mean Grade in Developmental Course, GPA, Semesters Enrolled and Credits Earned by Younger and Older Developmental Course Completers


Si


Variable GRADE IN
DEVELOPMENTAL
COURSE

GPA



SEMESTERS



CREDITS



Multivariate F = 3.22;


Group Mean D 17-21 2.89 22+ 3.19 17-21 2.33 22+ 2.70 17-21 4.41 22+ 3.77 17-21 31.85 22+ 22.69

df = 5, 155; p < .01*

Univariate F Tests


tandard eviation

.92 .95 .62 .64 2.60 2.40 23.42 18.80


N

135

26

135

26 135

26 135

26


Variable Univariate F df p

GRADE IN
DEVELOPMENTAL 2.50 1, 159 .11
COURSE

GPA 7.47 * 1, 159 .01 *

SEMESTERS 1.37 1, 159 .24 CREDITS 3.53 1, 159 .06
* p < .05


As seen in Table 19, mean grade in Freshman Composition I was not significantly different between the younger and older groups of developmental course completers. The mean grade of 2.56 for 17-21 year olds and 2.92 for the age 22 and older group yielded a non-significant F value of 1.60 at the .05 level when analysis of variance was applied.








TABLE 19

A Comparison of the Mean Grade in Freshman Composition I of Younger and Older Developmental Course Completers

Standard
Group Mean Deviation N 17-21 2.56 .94 80 22+ 2.92 1.04 13

Source df SS MS F Between Groups 1 1.47 1.47 1.60 Within Groups 91 85.12 .91 p < .20


Therefore, the seventh hypothesis--that there was no significant difference between achievement of students age 22 and older and students age 17-21 in developmental courses--was rejected only in terms of cumulative GPA. Older students did have a significantly higher cumulative GPA than recent high school graduates. However, on the other two achievement measures of grade in developmental course and grade of developmental completers in Freshman Composition I, the hypothesis failed to be rejected. Older and younger students seemed to achieve about the same in these courses on two of the three measures of achievement.

The eighth and last hypothesis stated that there was no significant difference in persistence (in terms of completing the developmental course, completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who








enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21.

No significant difference occurred between the number of younger and older students completing their developmental writing courses, as seen in Table 20. Of the total of 200 students enrolled in either the classroom or the individualized course, 168 were age 17-21 and 32 were age 22 or older. Of the younger students, 135 or 80.4% completed their course; of the older students, 26 or 81.3% completed their course. That resulted in a non-significant chi-square value of .00 after the Yates Correction.


TABLE 20

A Comparison of the Rate of Completion of a Developmental Course by Younger and Older Students

Original Completed

Group Number Developmental %

17-21 168 135 80.4% 22+ 32 26 81.3% Chi-square after Yates Correction = .00; df = 1; p < 1.00


Similarly, the number of developmental completers subsequently completing Freshman Composition I showed no significant difference by age (see Table 21). Of the 135 developmental completers in the 17-21 age group, 107 enrolled in Freshman Composition I and 80, or 74.8% of those enrolled, completed with a grade of D, C, B, or A. Of the 26 older developmental completers, 15 enrolled in Freshman Composition I and 13, or 86.7%, completed it. The chi-square value of .48 after the Yates Correction was not significant.








TABLE 21

A Comparison of the Rate of Completion of Freshman Composition I by Younger and Older Developmental Course Completers

Completed Enrolled in Completed
Group Developmental Fresh. Comp. I Fresh. Comp. I %

17-21 135 107 80 74.8% 22+ 26 15 13 86.7% Chi-square after Yates Correction = .48; df = 1; p < .49


The MANOVA analysis of possible interacting short and long-term measures cited in Table 18 included data on long-term persistence measures of semesters and credits completed. There was no significant difference in number of semesters enrolled between the younger and older groups. The 17-21 group had a mean number of semesters enrolled of 4.41 while the 22 and older students were enrolled a mean of 3.77 semesters. The univariate F value of 1.37 was not significant at the p < .05 level. The number of credits earned by the two groups also showed no significant difference, as also seen on Table 18. The 17-21 group had a mean cumulative credits earned of 31.85 while the 22 and older students had earned 22.69. The univariate F of 3.53 was not significant. As a result, it appeared that the recent high school graduates persisted better than students who had been out of high school four or more years.

Finally, the number of students earning an associate degree from each age group was not significantly different. Table 22 shows that of the 135 younger developmental completers, 24, or 17.8%, earned an associate degree. Of the 26 older developmental completers, 2, or








7.7%, earned an associate degree. While this may appear to be a significant difference, chi-square with the Yates Correction yielded a non-significant value of .98. TABLE 22

A Comparison of the Associate Degrees Earned by Younger and Older Developmental Completers


Developmental Earned Group Completers Degree

17-21 135 24 22+ 26 2

Chi-square after Yates Correction = .98; df = 1; p < .32


17.8% 7.7%


Consequently, the eighth and last hypothesis--that there was no significant difference in persistence between developmental students aged 22 and older and more recent high school graduates--was not rejected. Older students persisted about the same as younger students in completing their developmental course, in completing Freshman Composition I, in staying enrolled by semesters, in earning credits, and in earning an associate degree about as well as did younger students.

In conclusion, only one significant difference, the long-range achievement measure of cumulative GPA, occurred between the younger developmental students age 17-21 and the older developmental students age 22 or more. Older students did earn a higher cumulative GPA than younger students. On all the short-range measures and on the other long-range measures, there were no significant differences.









Summary

In summary, Table 23 provides a quick recap of the findings detailed in this chapter.

There were no significant differences between the achievement and persistence measures of students in the classroom developmental and the individualized developmental groups. Only one significant difference occurred when the combined developmental groups were compared with the students who should have taken a developmental course, according to the cutoff scores on the Initial English Assessment, but did not. Students who completed a developmental course did end up with a higher mean GPA than did the No-takes. The other achievement and persistence measures showed no significant difference, even though the No-takes had a slight, but significantly higher IEA mean score.

When the developmental completers were compared to the control group of students who had scored high on the Initial English Assessment and were therefore judged as not needing a developmental writing course, the high scorers did have significantly higher results on both achievement measures, but not on three of the four persistence measures. In terms of the rate of completing Freshman Composition I, the mean number of semesters enrolled in 11 semesters, and in the number of credits earned, the developmental completers showed no significant difference from the students judged stronger in writing skills at entry.

Finally, there were no significant differences on all but one of the achievement and persistence measures between recent high school graduates aged 17-21 and older students aged 22 or more. The exception came in the long-term achievement measure of cumulative GPA. The older students had a significantly higher cumulative GPA.









TABLE 23

Summary of Findings

Group Measure Significance


Classroom vs. Individualized Developmental Course





Combined Developmental Completers vs. No-takes



Combined Developmental Completers vs. High Scorers


17-21 Year Olds vs. 22 and Older Developmental Students


Grade in Developmental Course Grade in Freshman Composition I Cumulative GPA

Completing Developmental Course Completing Freshman Composition I Semesters Enrolled Credits Earned Associate Degrees Earned

Grade in Freshman Composition I Cumulative GPA

Completing Freshman Composition I Semesters Enrolled Credits Earned Associate Degrees Earned

Grade in Freshman Composition I Cumulative GPA

Completing Freshman Composition I Semesters Enrolled Credits Earned Associate Degrees Earned

Grade in Developmental Course Grade in Freshman Composition I Cumulative GPA

Completing Developmental Course Completing Freshman Composition I Semesters Enrolled Credits Earned Associate Degrees Earned


not significant not significant not significant


significant significant significant significant significant


not significant +


not not not not


significant significant significant significant


not significant not significant not significant


not significant not significant


significant significant significant significant significant


* Developmental completers' means were significantly higher at p < .05.

** High scorers' means were significantly higher at p < .05.

*** Older students' means were significantly higher at p < .05.

+ Developmental completers' means were similar at p < .05 to the Notakes' means although No-takes had higher entry skills.

++ Developmental completers' means were similar at p < .05 to the high scorers' means.








Table 24 summarizes the relationship of the findings to the hypotheses.


TABLE 24

Summary of the Relationship of Findings to Hypotheses


Hypotheses


Status


Hl: There is no significant difference in
achievement (in terms of mean grade
in the developmental course, mean grade
in Freshman Composition I, and cumulative GPA) between students who needed
a developmental writing course and completed the individualized course and
students who needed a developmental course and completed the classroom
course.

H2: There is no significant difference in
persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four
years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individualized course and students who needed a
developmental writing course and enrolled
in the classroom course.

H3: There is no significant difference in
achievement (in terms of mean grade in Freshman Composition I and cumulative
GPA) between students who needed but
did not enroll in a developmental writing course and students who needed and
completed a developmental writing course.

H4: There is no significant difference in
persistence (in terms of completing Freshman Composition I, enrolling a
mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between
students who needed but did not enroll in a developmental writing course and
students who needed and completed a
developmental writing course.


Was not rejected.








Was not rejected.










Rejected for cumulative GPA.

Was not rejected for grade in Fresh. Comp. I.

Was not rejected.








Table 24--Continued

Hypotheses


H5: There is no significant difference in
achievement (in terms of mean grade in Freshman Composition I and cumulative
GPA) between students who needed and
completed a developmental writing course
and students who neither needed nor enrolled in a developmental writing course.

H6: There is no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of
semesters in nearly four years, earning a mean number of credits, and earning an
associate degree) between students who
needed and completed a developmental writing course and students who neither needed
nor enrolled in a developmental writing
course.

H7: There is no significant difference in
achievement (in terms of mean grade in
the developmental course, mean grade in Freshman Composition I, and cumulative
GPA) between students who enrolled in a
developmental writing course at age 22 or
older and students who enrolled in a developmental writing course at age 17 to 21.

H8: There is no significant difference in
persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly
four years, earning a mean number of
credits, and earning an associate degree) between students who enrolled in a developmental writing course at age 22 or older
and students who enrolled in a developmental writing course at age 17 to 21.


Rejected.






Rejected for associate degree.

Was not rejected for completing Fresh. Comp. I, semesters, and credits.


Rejected for cumulative GPA.

Was not rejected for grades in developmental course and Fresh. Comp. I.

Was not rejected.


Four hypotheses were not rejected because there were no significant differences on all variables between the groups being measured. Three hypotheses were rejected only for one applicable measure, but were not rejected on other measures. Only one hypothesis was rejected


on all measures.


Status














CHAPTER 5
SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS

The problem in this study was to evaluate a community college developmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an individualized format, so that the college might assess the efficacy of each course and of the entire program in terms of student persistence and achievement. Three major questions were asked about the effect of the developmental writing courses. The fourth question asked if age was a factor in developmental students' performance.

The first research question asked whether an individualized developmental writing course or a classroom based developmental writing course contributed to better achievement and longer persistence by students who needed a developmental writing course when they started college. The first accompanying null hypothesis stated that there was no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course. The second null hypothesis stated that there was no significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of









credits, and earning an associate degree) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course.

The second research question was concerned with whether or not students who needed and completed either of the developmental writing courses achieved better and persisted in college longer than students who needed but did not enroll in a developmental writing course. The third null hypothesis accompanied this research question and stated that there was no significant difference in achievement (in terms of

mean grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. The fourth null hypothesis stated that there was no

significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course.

The third research question was concerned with whether or not the developmental writing courses helped students with weak writing skills

at entry catch up, in terms of achievement and persistence, with students who did not need a developmental writing course when they started college. The fifth null hypothesis stated that there was no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who









needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. The sixth null hypothesis stated that there was no significant difference in persistence (in terms of subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course.

The fourth research question was concerned with whether or not, after completing a developmental writing course, older adults achieved and persisted any differently from recent high school graduates. The seventh null hypothesis stated that there was no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. The eighth null hypothesis stated that there was no significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21.

The 1,431 students who entered Valencia Community College in the 1980 fall semester, who were first-time-in-college students, who completed the Initial English Assessment (IEA), and who enrolled in West









Campus courses were divided into four groups: those scoring below the IEA's cutoff and enrolling in a classroom-based developmental writing course, those scoring below the cutoff and enrolling in an individualized developmental writing course, those scoring below the cutoff but not enrolling in either developmental course, and those scoring above the cutoff and enrolling in regular freshman English. Random samples of 100 students were drawn from each of these four groups.

Achievement was measured as grade in developmental course, grade in regular freshman composition, and cumulative GPA. Persistence was measured as completion of a developmental English course, completion of a regular freshman composition course, number of semesters enrolled, number of credits earned, and attainment of an associate degree.

Achievement and persistence data came from student transcripts. IEA scores came from computer archives.

Chi-square followed by the Yates Correction, analysis of variance followed by the Scheffe test or an F test, and multivariate analysis of variance followed by the multivariate F test and univariate F tests were applied as appropriate.



Conclusions

As is typical for community college developmental programs (Kulik et al., 1983), these findings produced primarily no significant differences. However, once achievement and twice persistence measures seemed to indicate positive impact from developmental writing course work.

In answer to the first research question, the individualized developmental writing course compared favorably with the classroom developmental course on all measures. At entry, the individualized




Full Text

PAGE 1

COMMUNITY COLLEGE DEVELOPMENTAL WRITING COURSES: A STUDY OF STUDENT ACHIEVEMENT AND PERSISTENCE By ELAINE A. GREENWOOD A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 1984

PAGE 2

Copyright 1984 by Elaine A. Greenwood

PAGE 3

ACKNOWLEDGEMENTS Special thanks go to Dr. Al Smith, my committee chairman, for his considerable insight and guidance; to Dr. James L. Wattenbarger, committee member, for his example of national leadership in higher education; to Dr. Charles D. Dziuban, committee member, for his patience in teaching me research skills; and to Donald J. Tighe, my mentor, for a decade of collaboration. My appreciation is also due to Alban and Ellen Guenette, my parents, for their belief in me always; to Barbara Greenwood, my mother-in-law, for her example of steadfastness; and to Andrew and Ann Greenwood, my children, for their sense of humor. My deepest appreciation goes to Richard G. Greenwood, my husband, for his unwavering attention to detail, encouragement, and love. iii

PAGE 4

TABLE OF CONTENTS ACKNOWLEDGEMENTS ........•.....••••••...•..•..•.•••••••.•.•••..•..• iii LIST OF TABLES.................................................... vi LI ST OF FIGURES ....•..•••....•.•......••..•.•...•.•••.....••••..•. vi i i ABSTRACT CHAPTER 1 2 .......................................................... ix THE NATURE OF THE PROBLEM ..••••.•.......•.••.•.•..•.••.• 1 Problem Statement. ...................................... 4 Need for the Study ...................................... 8 Delimitations... ...... ... .... ...... ........... .......... 11 Limitations......... ... ..... ... ..... ... ..... ......... ... 12 Definitions............................................. 13 REV I EW 0 F TH E LI T E R AT U R E ..•••••••.•••••••....•....•....• 18 Summaries of Evaluations .•••..•.•..•••••...•.•..•••.• 19 Studies Related to Research Questions................... 25 Research Question #1: Developmental WritingIndividualized Versus Classroom •••••••••....•.••.• 25 Research Quest ion #2: Developmental Completers Versus Control Groups Who But Did Not Enroll in Developmental Writing •••••.•.•.....• 27 Research Question #3: Developmental Completers Versus Control Groups Who Did Not Need Developmental English ....•......•......•.•••.•...• 35 Research Question #4: Age as a Factor for Developmental Students ......•..••.•••..•••.••••.•. 43 Summary................................................. 45 3 METHODOLOGY ..•......•.....•.........•.•.•••............. 48 General Research Design •.....•...•.••...•••....•.....•.. 48 Instrumentation .........•.•..•.•••..•.•.•...•••.••..••.. 48 Sample.................................................. 49 Collection of the Data ..•.•.••.•••..•••.•.•••••••••..•.• 50 Analysis of the Data......... .......... ... ... ...... ..... 51 iv

PAGE 5

4 FINDINGS................................................ 55 Background Analysis of Samples ...•.....................• 56 Research Question #1: Classroom Versus Individualized Developmental Course .................. 60 Research Question #2: Developmental Groups Combined Versus Control Group of No-takes ............ 66 Research Question #3: Developmental Groups Combined Versus High Scorers .........•............... 71 Research Question #4: Younger Versus Older Developmental Students ..................•.•.......... 77 Summary................................................. 83 5 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS .........•..... 87 Summa ry ................................................. 87 Conclusions.. ... ........ ......... ... .... ............. ... 90 Recommendations ......•........................•......... 101 APPENDIX A INITIAL ENGLISH ASSESSMENT TEST .....................•... 1 1 0 B SAMPLE STUDENT TRANSCRIPT ............................... 118 C DESCRIPTION OF DEVELOPMENTAL CLASSROOM AND INDIVIDUALIZED COURSES ..................•.......•. 120 REFERENCES ••.••.•..•..•••••.••.•.•.••.•••.•••••••••.••••••••••.••• 123 BIOGRAPHICAL SKETCH............................................... 128 v

PAGE 6

LI ST OF TABLES 1 A Descriptive Analysi s of Initial English Assessment Test Scores ................................................ 57 2 Results of Analysis of Variance on the Mean Scores on the Initial English Assessment Test by Groups ....•.••••.•.. 58 3 A Comparison of the Mean Scores on the Initial English Assessment Test Between the Combined Developmental Groups and the Control Group of No-takes................... 59 4 A Comparison of Developmental Writing Students' Mean Scores on the Initial English Assessment Test by Student Age at Entry ••.•••.•••.....••..•••••.••••.•..•••••• 59 5 A Comparison of the Mean Grade in Developmental English Courses Between Students in the Classroom and Individualized Courses •.••••.•.••••.•.•.••••..••...•.••..•• 61 6 A Comparison of Mean Grade in Freshman Composition I, GPA, Semesters, and Credits of Developmental Com pleters of the Classroom and Individualized Courses 7 A Comparison of the Completion Rate of Developmental Courses Between the Classroom and Individualized 62 Groups ......••.••••..•..•.••..•..•••........•.••••••••..... 64 8 A Comparison of the Completion Rate of Freshman Composi tion I Between Students Who Completed the Classroom and the Individualized Developmental Courses .•••••.••.••... 64 9 A Comparison of the Degrees Earned by Developmental Classroom and Individualized Course Completers •••....•.••.••.••• 65 10 A Comparison of the Mean Grade in Freshman Composition I of Combined Developmental Groups Versus Control Group of No-takes ................................................ 67 11 A Comparison of the Mean GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups and the Control Group of No-takes •.....•.••••...•..••••.••• 68 12 A Comparison of the Rate of Completion of Freshman Composi tion I for Combined Developmental Groups Versus Control Group of No-ta kes ..••.••.............•.•••...........•..•.. 70 vi

PAGE 7

13 A Comparison of Associate Degrees Earned by Developmental Completers Versus Control Group of No-takes •••.•••••....••. 71 14 A Comparison of the Mean Grades in Freshman Composition I Earned by Developmental Completers Versus High Scorers 72 15 A Comparison of Cumulative GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups Versus the Control Group of High Scorers .•.••.•......•••.•. 73 16 A Comparison of the Completion Rate of Freshman Composition I for Developmental Completers Versus the Control Group of Hi gh Scorers ...................................... 75 17 A Comparison of the Associate Degrees Earned by Devel opmental Completers Versus Control Group of High Scorers 75 18 A Comparison of Mean Grade in Developmental Course, GPA, Semesters Enrolled and Credits Earned by Younger and O l der Developmental Course Completers .••••••.•.•••.......•• 78 19 A Comparison of the Mean Grade in Freshman Composition I of Younger and Older Developmental Course Completers .•••..• 79 20 A Comparison of the Rate of Completion of a Developmental Course by Younger and 01 der Students ..••.••.•••...••..•••.. 80 21 A Comparison of the Rate of Completion of Freshman Composition I by Younger and Older Developmental Course Campl eters .......................................... 81 22 A Comparison of the Associate Degrees Earned by Younger and Older Developmental Completers ...•.•.••.•...••.••..•.•• 82 23 Summary of Fi ndings ........................................... 84 24 Summary of the Relationship of Findings to Hypotheses •.•••..•• 85 vii

PAGE 8

LIST OF FIGURES 1 Diagram of Groups for Research Questions 1, 2, and 3 as Determined by Score on the Initial English Assess m ent Test....................................................... 52 2 Diagram of the Developmental Group Realignment for Research Question 4 as Determined by Student Age at Entry ••.•.•.•••. 53 viii

PAGE 9

Abstrac t of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education COMMUNITY COLLEGE DEVELOPMENTAL WRITING COURSES: A STUDY OF STUDENT ACHIEVEMENT AND PERSISTENCE By Elaine A. Greenwood December 1984 Chairman: Dr. Al Smith Major Department: Curriculum and Instruction The purpose of this study was to evaluate Valencia Community College's (Florida) developmental writing program in terms of student achi evement and pers i stence. The study contrasted the performance 0 f students who completed a classroom developmental course, who completed an individualized developmental course, who avoided takin g a developmental writing course, and who did not need a developmental writing course. The last part of the study contrasted the performance of older and younger developmental course completers. Four student samples (developmental classroom students, developmental individualized instruction students, developmental avoiders, and high scorers) were randomly drawn from fall 1980 college entrants grouped by Engl ish pl acement test scores and fi rst semester Engl i sh course. Achievement was measured by grades in a developmental course, grades in Freshman Composition I, and cumulative grade point average ix

PAGE 10

(GPA) • Persistence was measured by completion of a developmental writing course, completion of Freshman Composition I, number of semesters enrolled, number of credits earned, and attainment of an associate degree over eleven semesters. The study showed no significant differences on any measure between developmental classroom completers and individualized course completers. Developmental course completers did have a significantly h i gher cumulative GPA than did developmental avoiders. No significant dif-ferences occurred on all other measures even though the avoiders had slightly stronger entry-level writing skills than the developmental students. High scorers, who entered with strong writing skills, achieved significantly higher grades and earned more associate degrees than did developmental completers; but developmental completers equalled high scorers in Freshman Composition I completion, number of semesters enrolled, and number of credits earned. Older developmental course completers achieved significantly higher cumulative GPAs than did younger developmental students, but age had no effect on the other measures. Recommendations included that the Florida Department of Education should investigate the feasibility of a statewide evaluation model for developmental courses, given the current state-mandated entry-level pl acement and sophomore testi ng programs; that students shoul d not be denied access to developmental writing courses on the basis of age; and that further research should be conducted at Valencia to determine if the recently upgraded developmental writing curriculum has improved student achievement and persistence. x

PAGE 11

CHAPTER 1 THE NATURE OF THE PROBLEM For the past fifteen years, developmental education programs have been widespread in community colleges. Alternative instruc-tional methods, such as individualized instruction, have been incor-porated into many of these programs. A recent survey of community college presidents indicated that 97.8% of the 639 responding institutions offered developmental or remedial courses (Campbell, 1982). The current emphasis on developmental education in community colleges grew out of the "open door" admissions policies starting in the late 1960s. "Open door" institutions began admitting nearly every adult applicant, no matter what educational credentials the ap-plicants had. Large numbers of non-traditional students were among these new adult students. Cross (1971) defined these non-traditional students comprehensively as all adult learners who had difficulty with academic work, regardless of the cause of the difficulty. What these new students had in common was that they scored in the lowest third among national samples on standardized tests of academic abil-ity. What they di d not have in common was just about everythi ng else. Some were representatives of minorities, some not; some were women; some were in their twenties, thirties, forties, some even older; some were married with families; many were working full-time jobs in addition to attending college . However, the lower socioeco-nomic women, minorities, and older students had higher rep-resentation in this group than among traditional college students. 1

PAGE 12

2 Advice on how to deal with non-traditional, academically unprepared students grew with their numbers. A great deal of theorizing occurred about alternate instructional methods. Individualized in-struction was one such proposed method. The Carnegie Commission on Hi gher Educati on (1970) suggested that community colleges needed remedial education programs which were flexible and individualized. Roueche and Pitman (1973) advocated combining Bloom's mastery learning theories (Bloom, 1971) with systematic, individualized instruction in basic skills. Competency-based instruction was often part of the individualized approach (Roueche, Herrscher, & Baker, 1976). The intensity of the student-instructor relationship in individual ized instruction became a hallmark of many programs (Gollattscheck, Harlacher, Roberts, & Wygal, 1976). After fifteen years of widespread developmental programs, state legislatures in charge of funding most state higher education systems are increasingly asking pointed questions about the effectiveness of developmental and remedial programs. Many legislatures, as in Maryland, are reassessing higher education programs with an eye to budget cuts, or at least different funding approaches (Study of Remedial/Developmental, 1982). At the same time, some states have been attempting to upgrade elementary and secondary systems. Legislators, like those in Illinois, asked if developmental programs and alternative instructional methods are really helping underprepared students cope with the demands of a college education. They are also questioning if such programs will conti nue to be necessary as school systems produce better prepared high school graduates (Status Report, 1981). In Texas, there is e ven a call for the state to close the open door in the

PAGE 13

3 community college system since developmental programs, according to the director of research for the Coordinating Board of the Texas College and University System, have only diluted qual ity in postsecondary edu cation (Patterson-Griffith, 1983). In Florida, also, such ques t ions are being aSked--and answered with legislation. In 1982, the Florida Department of Education's Postsecondary Education Planning Commission recommended that devel opmenta 1 /remedi a 1 programs at the sta te community colleges and universities be eliminated by the year 1990 (The Master Plan, -1982). In the 1983 state legislative session, lawmakers too k that advice and enacted legislation that shifted responsibility for reI medial and developmental programs from higher education to district school boards (Section 232.2455, 1983). According to this law, after 1990, the only remedial courses allowed in community colleges woul d serve students who would be five or more years from high school graduation or who needed remediation before benefitting from vocational edu cation (Section 240.134, 1983). Although these limitations on developmental/remedial courses were eliminated by the 1984 legislature (Se c tion 240.117, 1984), more legislative limitations in Florida and in other states are a possibilty. A number of developmental education theorists and researchers have expressed growi ng concern about the need for eva 1 uati on if develop mental programs are to survive. Perry-Miller, Nolan, and Smith (1980) point out that current arguments at the local, state, and national level about whether developmental education should even be offered at the college level make it necessary that developmental educators continue to convince constituencies that tax dollars should go to develop mental education. Maxwell (1979) cites the need for experimental and

PAGE 14

4 quasi-experimental studies to contribute to decisions about program continuation so that programs can gain support from the populace and from political decision makers. Akst and Hecht (1980) warn that too few remedi al programs have been carefully evaluated, despite the pol itical real ity of shrinking budgets and growing demands for accountabil ity. The problem facing developmental educators in the mid-1980s is to provide evaluation of their programs in a manner which will have credibil ity with deci sion makers who control continued funding for postsecondary education. Without continued funding, the programs will be cut back. If legislative bills, such as the one enacted by the 1983 Florida legislature cited above, explicitly prohibit community colleges and universities from providing developmental programs, such programs could even disappear. Problem Statement The problem in this study was to evaluate a community college de velopmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an indi vi dua 1 i zed forma t, so that the college mi ght assess the effi cacy of each course and of the entire program in terms of student persistence and achievement. The achievement and persistence of randomly selected students experiencing these two modes in the fall semester of 1980 were compared. The achievement and persistence of students from both courses combined were also compared with achievement and persistence of two randoml y sel ected contro 1 groups: one of students who needed de velopmental writing but did not enroll in either course and one of

PAGE 15

5 students who neither needed nor enrolled in a developmental writing course in the fall semester of 1980. Three major research questions were asked about the effect of the developmental writing courses. These questions and their accompanying null hypotheses follow: 1. Does an individualized developmental writing course or a classroom based developmental writing course contribute to better achievement and longer persistence by students who need a developmental writing course when they start college? HI: There is no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course. H 2 : There is no Significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enroll i ng a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individualized course and students who needed a devel opmenta 1 writ i ng course and enrolled in the classroom course. II. Do students who need and complete either of the developmental writing courses achieve better and persist in college longer than students who need but do not enroll in a developmental writing course?

PAGE 16

6 H3: There is no s i gnifi cant di fference in achi evement (i n terms of mean grade in Freshman Composition I and cumulative ' grade point average) between students who needed but di d not enro 11 ina de velopmental writing course and students who needed and completed a developmental writing course. H 4 : There is no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. III. Do the development al writing courses help students with weak writing skills at entry catch up, in terms of achievement and persistence, with students who do not need a developmental writing course when they start college? H5: There is no s i gni fi cant difference in achi evement (i n terms of mean grade in Freshman Composition I and cumulative grade point average) between students who needed and completed a developmental writi ng course and students who nei ther needed nor enro 11 ed ina developmental writing course. H6: There is no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed and comp 1 eted a deve 1 opmenta 1 writ i ng cou rse and students who neither needed nor enrolled in a developmental writing course.

PAGE 17

7 The effect of developmental writing courses on older adult students as well as on recent high school graduates was also considered in this study for two reasons. First, the developmental courses were de signed for non-traditional students who comprised the majority of students enroll ing in these courses, and non-traditional students as a group varied greatly in age. Second, one argument for di sconti nui ng developmental level courses made by some state legislators was that the upgrading of primary and secondary schools would result in stronger high school graduates, which could diminish or eliminate the need for college level developmental courses. However, it is important for legislators to consider that many developmental students are much older than eighteen and, therefore, have not graduated from recently upgraded school systems. Legislators need to know how the developmental courses have a ffec ted the achi evement and pers i s tence of older adults as well as younger students. In Florida, legislation enacted in 1983 limited remedial and developmental education after 1990 to two classes of students, one of which was the adult who had been "five years or more out of sequence from high school graduation". (Section 240.134 (3) (a), Florida Statutes, 1983). It is important to ascertain if this arbitrary age demarcation point is a valid one. Thus, the major research question asked about student age as a factor in the effect of developmental writing courses in this study was as follows: IV. After completing a developmental writing course, do older adults achieve and persist any differently from recent high school graduates? H7: There is no s i gnifi cant difference in achi evement (i n terms of mean grade in the developmental course, mean grade in Freshman

PAGE 18

8 Composition I, and cumulative grade point average) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course a tag e 17 to 21. H8: There is no significant difference in persistence (in terms of completin g the developmental course, subsequently completing Freshman Composition I, enroll ing a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who enrolled in a develop mental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. Need for the Study All instructional programs should be evaluated to ascertain if they are effective in attaining their goals. The need is perhaps more critical for developmental programs. At most community colleges, a substantial proportion of entrants are non-traditional often the majority have weaknesses in the basic skills. students, and At Miami-Dade Community College, for example, about two-thirds of the entering students need course work in one of the cOllege1s developmental areas; nearly one-hal f of the entering high school graduates need course work in all developmental areas, and with a mean student age of twentyeight, large numbers of these entering students will not benefit from upgraded secondary programs (McCabe, 1983). At Valencia Community Col lege, the comparable figures range from 57% to 79% of the students needing course work in one or more of the college1s developmental courses (Riles, 1983). These figures are typical of the large numbers

PAGE 19

9 of students entering other Florida community colleges and needing developmental programs. Besides having to acquire skills they lack at entry and continue through the college curriculum, Florida students must also achieve the approved level of proficiency on the College Level Academic Skills (Rule 6A-10.31, 1982), an official statewide list of communications and computations competencies, and pass a test on these skills (Rule 6A-10.312, 1984). Community college students who fail this test neither receive an Associate in Arts degree (Rule 6A10.313, 1982) nor are they allowed to matriculate into the upper division of a Florida public university (Rule 6A-10.314, 1984). To this end, according to a recent memo by the Florida State Commissioner of Education to the presidents of the Florida community colleges and state universities, Institutions will be required to assure that students who enter their college credit programs have effective opportunity to achieve the skills to the level required by the passing scores. Admissions counseling, developmental and remedial programs, standards for course completion, and standards of student progress that are required for retention as well as enrollment and funding will be impacted by the passing scores on CLAST (College Level Academic Skills Test). (Turlington, 1983, p.3) The continuing need for developmental programs in Florida community colleges is clear. The need for evaluating these developmental and remedial programs is equally clear, especially given the current political climate and recent legislation. According to 1983 statutes, which have since been rewritten, starting in 1990, primary responsibility for remedial and developmental programs would have been given to Florida county publ ic schoo 1 di stri cts. Remedial and developmental programs at community colleges would only have been permitted to serve students needing basic skills in order to be successful in vocational programs or students who

PAGE 20

10 completed high school five or more yea r s before starting college (Section 240.134, 1983). Although these limitations have s ince been written out of the statutes (Se ction 240. 117, 1984) subs tantive studies might playa part in shifting political opi nion toward futu re possible legislation concern ing developmental education in Florida and other states. But even ignoring the current political climate. the time h as come for mor e community colleges to find out systematically if their d e v el opmental programs have been effective in reaching their goals. A cc o rd ing to Darrel Clowes (1984). developmental programs are uniqu e in t heir respective coll eges in that the output of the developmental program i s the input for almost every other program on campus. How many students make a successful transition into the mainstream curri culum and h o w well they do there are important and available data. Suc h data have not been tapped at Valencia Community Colle g e . the site of the developmental courses under study in thi s research. The developmental writing courses in e xistence since 1976 have not been evaluated in a thorough and systema tic way. Data have been available about the percent of students who completed the courses and the m ea n grade received in those courses, but no breakdown by age and only one comparison with students who needed but did not enroll in the courses has been made. No compari son between outcomes of the cl assroom course and the individualized course has been made. There has also bee n no study of student persistence and achievement for more than one semester after completion of these courses. Therefore. although i nforma l formative evaluation of the courses has contributed to their evolu tion, it is indeed time for a systematic, summative evaluation of t h i s p r o gram.

PAGE 21

11 Del imitations The delimitations of the study were as follows. Only the develop mental writing courses at Valencia Community College, West Campus, were evaluated. The developmental reading, study skills, and mathematics courses were not involved. Valencia does not have a developmental education department or division per se. The developmental mathematics courses are a part of the Mathematics Department; the developmental writing, reading, and study skills courses are a part of the Communica tions Department. Instructors who teach in all these development al courses are part of and report to thei r respective academi c departments. There is no separate funding for a developmental program; funds come through the respective department budgets. There is no comprehensive developmental program and no counseling component automatically attached to the devel opmenta 1 courses, although many students in these courses also belong to the Special Services program or the Handicapped Student Services program, both of which involve students in extensive counseling. There are frequent contacts and referrals back and forth between instructors teaching the developmental courses and counselors in these programs. Therefore, the scope of thi s study was confi ned to student outcomes from two courses, the classroom developmental writing course and the individualized developmental writing course at one Florida community coll ege campus. The data were drawn from students who entered the college in the fall semester of 1980.

PAGE 22

12 Limitations The findings in this study could be generalized to the same type of populations and courses described herein. The overall population was the group of students who chose to enter an urban community college with an open door admissions policy. The subpopulations were those students who needed and enrolled in a developmental writing course, those who needed but did not enroll in a developmental writing course, and those who did not need and did not enroll in a developmental writing course. A randomly selected sample was drawn from each of these sUbpopulations. The study probably should not be applied to developmental programs broader in scope than this one, that is, to programs which may include courses and/or supplemental lab work for students needing a variety of academic course work and which mayor may not include a counseling component. Findings pertaining to the classroom developmental writing course may be generalized to other classroom developmental writing courses. Findings pertaining to the individualized developmental writing course may be generalized to other individualized developmental writing courses with the following cautions. This individualized developmental writing course took place in a laboratory setting and carried elective credit; the instructors worked with a maximum of twelve and usually only ten students per section, and there were no student tutors. The instructors conferred with each student each time the student came to his or her regularly scheduled lab meeting. Many individualized developmental courses have sections with much higher numbers of students, with student tutors or paraprofessionals who carryon much of the

PAGE 23

13 direct instruction, with instructors who do not have the opportunity to monitor each student's progress at each lab meeting, and even with variable schedul ing. Findings from this study should only be general ized to other courses wi th one or a combi nati on of these di fferences w i th extreme caution. Furthermore, this study focused on only quantifiable data of achievement and persistence. It did not attempt to measure affective outcomes of the student-instructor interaction involved in these courses. It did not measure value added characteristics these courses might have had, that is, what value the course might have had even for students who withdrew duri ng the course or 1 a ter before compl eti ng a degree. It did not attempt to ascertain if students who eventually withdrew from college stayed in college longer than they might have if they had not enrolled in a developmental writing course. Finally, this study did not attempt to delineate why a course was successful or unsuccessful in increasing student persistence and achievement, just that it did or did not. Definitions Achievement is the level of academic attainment as measured by grade earned in the developmental writing course, subsequent grade in Freshman Composition I, and grade point average in all courses. A classroom course is any course which meets in a regular college classroom and is taught by traditional instructional methods, primarily lecture and discussion. A classroom developmental writing course is a developmental writing course taught in a classroom using traditional instructional meth ods.

PAGE 24

14 Developmental completers are students who have completed either the classroom or the individualized developmental writing course. Developmental courses are courses in basic skills such as reading, writing, mathematics, and study skills in which the instructor desires to nurture students' self-concepts, ability to cope with ac ademic strategies, and self-confidence as they learn new skills. Developmental programs are programs which include a group of basic skills courses such as reading, writing, mathematics, and study skills courses along with a counseling component and possible sup plemental lab work. Developmental programs often compose their own department or division within the college's administrative structure. A developmental writing course is a course in which the instructor desires to nurture students' self-concepts, ability to cope with ac ademic strategies, and self-confidence when they are learning the skills involved in the process of basic composition. The site could be either classroom or lab; the instructional method could be either traditional or individualized. An entry level test is a test or series of tests administered to students after they complete admission but before they register for their first semester of classes. The purpose of the test(s) is to help place students into courses appropriate to their level of ability and skill achievement. English entry tests usually designate whether students have sufficient writing skills for success in Freshman Composition I or whether they should enroll in a de velopmental writing course.

PAGE 25

15 Freshman Composition I is the standard course in rhetoric taught to first year college students. In Florida, the common course numbering system designation usually used for Freshman Composition I is ENC 110l. Grade Point Average (GPA) is the total of a student's quality points earned in each course divided by the total number of semester hours in which the student enrolled and did not withdraw. High scorers are students who score above the cutoff on the Initial English Assessment Test and enroll directly into Freshman Composi tion I. An individualized course is a course in which a one-to-one learning relationship exists between the instructor and student; in which the instructor assigns materials appropriate to each student's needs, monitors student progress constantly, and provides constant feedback to the student on progress made; and which takes place in a learning laboratory. An individualized developmental writing course is a course in writing skills with the characteristics of both an individualized course and a developmental writing course. The Initial English Assessment Test is an objective test of basic English skills written by English faculty at Valencia Community College and used as the entry level English test to recommend placement into a developmental writing course or into Freshman Composition I duri ng the year when the students measured in thi s study enrolled. A learning laboratory is an environment designed especially for one-to-one instruction, usually including individual study carrels,

PAGE 26

16 tables, shelves, file cabinets, and storage units appropriate to a variety of books, audiovisual materials and equipment, and possi bly computer software and hardware for computer assisted instruction. Long-term measures are measurements of student achievement and persistence that occur a semester or longer after completion of a de velopmental course. "No-takes" are students who score below the cutoff on the Initial English Assessment test, who are therefore advised to enroll in a developmental writing course, but do not. Non-traditional students are college students of any age or race and of either sex who, for whateve r reason, have difficulty with academic work in college, who usually score in the lower third on nationally normed tests, and who would benefit from developmental courses to improve their skills and their chance of success in mainstream courses. Older students are students who are twenty-two years old or older when they enroll in a developmental course. Persistence is longevity of matriculation as measured by the rate at whi ch students compl ete a course, compl ete another cou rse subse quent to the first, complete a series of semesters within the four years of this study, earn a cumulative number of credits, and earn an associate degree. Recent high school graduates are students who have graduated from high school and who are age seventeen to twenty-one when they enroll in a developmental course.

PAGE 27

17 Short-range measures are measurements of student achievement and per sistence that occur during or at the end of the semester in which the student is enrolled in a developmental course. Younger students are students who have graduated from high school and who are age seventeen to twenty-one when they enroll in a develop mental course (same as "recent high school graduates").

PAGE 28

CHAPTER 2 REVIEW OF THE LITERATURE More than one evaluation specialist has noted that current evaluation of remedial and developmental programs presents great diversity and little effort to synthesize that diversity (Akst & Hecht, 1980; Dumont, Bekus, & Tallon, 1981; Grant & Hoeber, 1978; Kendrick & Thomas, 1970; Kulik, Kulik, & Shwalb, 1983; Richardson, Martens, & Fisk, 1981; Ti nto & Sherman, 1974). Comparisons between studies are difficult because of the d i fferences in methodology and the lack of agreement on appropriate measurement criteria. In addition, there is no clear pat-tern to the results. Positive, nonsignificant, and even negative results have all been reported for entire developmental programs and for developmental writing courses in both classroom and individualized instructional modes. The results of the better studies extftnt in the literature follow here. First come the summaries of evaluation studies. The rest appear in the same order as the major research questions being asked in this study: (1) what does the research tell us about the achieveme nt and persistence of students in classroom versus individualized developmental writing courses; (2) what does the research tell us about the achievement and persistence of students who need and complete a devel-opmental writing course as opposed to students who need but do not enroll in such a course; (3) what does the research tell us about the achievement and persistence of students who need and complete a 18

PAGE 29

19 developmental writing course as opposed to students who do not need and do not enroll in such a course; and (4) what does the research tell us about the achievement and persistence of older adult students who need and complete a developmental writing course. Summaries of Evaluations The following summaries and a meta-analysis concern developmental programs, not developmental writing courses specifically. Metaanalys i s is sometimes criticized as having such a sheer amount of data grouped and analyzed together that regress i on toward the mean causes misleading results. However, many researchers and decision makers look to summaries and meta-analyses as pointing out the major trends in the area being stud ied. Kendrick and Thomas (1970) reviewed studies on disadvantaged co 11 ege students, remark i ng that it seemed premature in 1970 to be reviewing these studies because the emphasis nationwide was still on expanding educational opportunity. The high priority nationwide on admitting such students to college had only existed for a few years. Kendri ck and Thomas focused thei r report more on attempts to discern which traits these disadvantaged students had which would be most predictive of success. They noted that it was hard to ascertain the impact of remedial programs on helping disadvantaged students attain a bachelor's degree because so few community college students were going on to four year schools. They also noted that when studies showed good retention rates for these students, the studies too often had not been des i gned to determi ne if i nnov at i on in the program or other factors such as less demanding courses, a lighter course load, or atypical persistence patterns were responsible for the positive results.

PAGE 30

20 Four years later, Tinto and Sherman (1974) analyzed evaluations of spec i a 1 servi ces programs for di sadvantaged students in secondary and postsecondary institutions. Like Kendrick and Thomas, they also bemoaned the difficulty of synthesizing reports that focused on different outcomes, different populations, different points in time, different educat i ona 1 settings, different measures, and different research designs, most of which they found seriously flawed. Tinto and Sherman concluded that special services programs, with their heavy emphasis on counseling, tutoring, personal and cultural enrichment in addition to remedial course work, did seem to decrease dropout rates and increase retention of disadvantaged students. However, the increase did not reach the retention level for non-remedial students. In addition, special services programs seemed to have little positive impact on academic achievement. The special services students attained about the same GPA in college as they had attained in high school. Even more worrisome was the fact that disadvantaged students who were not part of special services programs did about as well as the special services students. However, although tutoring and counseling seemed to make no difference in academic achievement, some studies showed that special service students did seem to make important changes in attitudes, values, and motivation as evidenced by their higher persistence to graduation rate than non-special services and disadvantaged students had. Ti nto and Sherman caut i oned that these results may have been affected by self-selection into the programs. Since most studies were flawed by not having random assignment of students to groups, the higher rate of persistence (about 50%) might have been due to the more motivated disadvantaged students deciding to take

PAGE 31

21 advantage of the special services program while less motivated students decided to ignore it. Overall, the special services programs with better results were usually integrated into the regular college program; they were not separate. The students in separate programs were often stigmatized; they usually did not fare as well in retention and persistence patterns as the students in integrated programs. Roueche and Snow (1977) disagreed with Tinto and Sherman on this 1 ast poi nt. Roueche and Snow I s study was a survey of the deve 1 op mental education programs of 150 public two-year colleges and 150 public four-year colleges. They concluded that in four-year colleges, at least, having a separate department or division of basic studies was significantly related to the respondents also rating their program as successful. A significant relationship did not exist between this factor and self-reported successful programs for two-year colleges, though. Other factors with statistically significant relationships to self-reported success were written course objectives given to students, tutors trained to develop positive self-concepts in their tutees, peer counselors selected on the bq.sis of their effectiveness and trained to work on improving their students' self-concepts, faculty who compared their students favorably with other students at their college, full-time counselors who were attached to the develop mental courses and who consulted in curriculum development, and suc cessful retention of students until their third semester in college. However, the main problem with the Roueche and Snow study was that it was a self-report survey. When respondents indicated that their program was successful, they were asked for their opinion and not required to comment based on any substantive research at their institution.

PAGE 32

22 Grant and Hoeber (19 7 8 ) reiterated the p roblem of f irst finding studies which were empirically base d and second comparing those studies g iven the broad range of designs and outcome measures employed. When concluding whether or not basic studies programs wer e working , Grant and Hoeber wrote: The question can be answered in two ways--yes, those involved in and committed to those programs are 'wor k i ng' very hard; no, the programs themselves are not 'working' very well, or, more accurately, a dearth of empirically based evaluation research o n these programs m akes the que stion moot. (p. 1) They di d deci de that some deve 1 opmenta 1 programs were doomed--those with curriculum and instructional methods similar to what developmenta 1 students had experi enced in elementary and secondary school. Such curriculum and teach in g methods had not worked in these students' pre vious twelve years of schooling; they probably would not work in college either. Grant and Hoeber recommended that programs be modified for each student and that recognition of differen t learning styles and rates of growth should lead to greater use of mastery learning. They also concluded that the literatur e clearly indicated that one type of developmental prog ram could not possibly be success-ful as the universal model for all postsecondary programs. Just as programs needed to be modified for individual students, pro grams needed to be individually tailored for each campus in o r der to fit each institution's needs. Richar dson, Martens , and Fisk (1981) compiled results of develop-mental programs for their research into functional literacy in col-leges and universities. Retention or persistence was usually measured in terms of continual enrollment or decline in the dropout rate.

PAGE 33

23 Community colleges frequently reported pos it i ve results in retent i on, universities less often. Richardson et al. speculated that many institutions could not supply such data because they d id not have it or would not supply such data because they wished to suppress it. As for achievement measurements, most studies used GPA as the criterion for success. Often, students' GPAs during remediation dropped significantly as soon as they took regular college classes. In some studies, completing a developmental writing course often contributed to an improved GPA, but completing, failing, or withdrawing f rom a develop mental reading course made no difference in GPA at all. As for studies comparing methods of instruction, the most corrrnon comparison was between individualized approaches contrasted to traditional classroom instruction. The results were sometimes positive, sometimes not sign ifi cant, sometimes negative, 1 eavi ng confus i on as to the effect i ve ness of individualized programs. The most current meta-analysis is Kulik, Kulik, and Shwalb's empirical study of research on college program s for high-risk and disadvantaged students (1983). Following systematic selection of quantitative, experimental studies, the authors chose 60 reports out of 504 documents to apply multivariate statistical analysis to in orde r to integrate findings. Each of the chosen studies had to involve highrisk college students in special programs, had to report on measures made on these students plus on one or more control groups with similar aptitude, and had to have no methodological flaws. The measures had to involve achievement in terms of GPA and persistence in terms of the proportion of students who remained enrolled during the course of each study.

PAGE 34

24 In terms of achievement, Kulik et al. concluded that the devel op mental programs as a whole had basically positive results; In 4 4 of 57 studies including GPA as a measure, GPA was higher for students in special programs than for similar students who were not in special pro g r ams. S ignificantly h igher GPAs for students in special pro grams were reported in 16 of 17 studies where the difference reached a significant level. When meta-analysis techniques were appl i e d, the typical report would have cited a GPA of 2.03 for special p rogram students but 1.82 for control groups, which the authors admitted was a small difference, but considered statistically reliable. Regression ana lys i s showed that three features of the programs were assoc i ated with higher GPA: newer programs ( perhaps due to novelty effect), progra m s characterized as remedial or developmental studies (a s opposed to isolated academic skills instruction, gui dance sessi ons, and comprehensive support systems), and p ro g r a m s reported in documents by the Educational Resources Info r mation Center (ERIC), rather than in journal articles and dissertations. Kulik et al. continued with meta-analys i s of pers istence. They defined persistence as lithe proportion of students initially admitted who rem ained enrolled at the college dur ing t h e period exa m ined in the study" (p. 400). Of the 30 studies measuring persistence, 21 s h owed highe r persistence for special program students, 11 of which were statistically significant. Further analysis ind icate d that the t ypical persistence rate across the studies was 60% for special p rogram students versus 52% for control s tudents. The author s judge d the special programs as having a statistically reliable effect on persistence, but it was smaller and harder to se e than the effec t on GPA.

PAGE 35

25 Using regression analysis, the authors concluded that how early students started in a special program (high school was better than fresh-man year in college, which was better than second year in college, etc.) was the only consistently important factor on persistence. The authors were disappointed in the results for community col-lege programs, which, by and large, produced no significant difference in GPA and little effect on persistence. Program effects were stronger at four-year colleges and strongest at doctoral universities. Studies Related to Research Questions Research Quest i on s #1: Developmental Writing--Individualized Versus Classroom Whether an individualized approach to developmental writing instruction or the traditional classroom works better is a question many researchers have a sked . Their answers have produced mixed results and, more often than not, statistically nonsignificant results. Sutton and Arnold (1974) were concerned that classroom i nstruc-tion denied academically disadvantaged students the personal attention they required to make significant gains. They theorized that individ-ual ized instruction or instruction in a writing laboratory might be more effective than classroom remedial English courses. Their writing 1 ab approach incorporated tutori ng with programmed instruction. The student population was 244 entering freshmen at a southern regional university in the fall of 1970. Sutton and Arnold found durin g the two years of the study that there was no apparent difference in the number of students who persisted in college nor in the dropout rate between students who experienced the individualized course in the

PAGE 36

26 writing lab and students who experienced a remedial English course in the traditional lecture-discussion mode. However, the achievement of the individualized course students was higher than for the classroom course studen ts; the i n d iv i dual ized course students earned higher mean grades in subsequent English courses as well as higher cumulative GPAs than did the classroom course students. Epes, K i rkpatri ck, and Southwell (1980) des igned and reported on an individualized, audio-tape enhanced course as contrasted with traditional, classroom-based developmental writing courses at three col -1 eges of the City Un i vers i ty of New York. Thei r new course did involve some whole-group instruction, but students worked primarily in a writing lab at their own pace on individually assigned audio-tapes, exercises, and writing assignments created by the authors. Most measures used in this study pertained to attitude toward English courses or toward writing in general (no significant difference) or writing improvement scales (error-count reduction was significantly better on two of six errors for the individualized students; holistic evaluation of student writing at both the beginning and end of the semester showed no significant difference). In terms of persistence, there was also no significant difference between the individualized and the classroom students. Of the 243 lab students, 140 (57.6%) passed while of the 199 traditional classroom students, 129 (64.8%) passed. The difference was not significant. On the other hand, Brown (1984) found that a teacher-paced, classroom developmental English course seemed to result in higher grades in developmental English and in higher term GPAs than d i d a self-paced developmental English course. Using a matched subject

PAGE 37

27 design, Brown compared student results in a fall semester 1977, teacher-paced course with student results in a fall semester 1980 , self-paced course at Quinsigamond Community College. Students were matched on the basis of age, sex, and score on a standardized, read ing pre-test. No significant difference occurred on three of the five measures: post-test standardized reading scores, completion rate of the developmental English course, and number of students persisting in college for four consecutive semesters. The students in the teacher-paced, classroom course did significantly outperform the self-paced student group on the remaining two measures: final grade in developmental English and GPA during the semesters under study. Research Question #2: Developmental Completers Versus Control Group Who Needed But Did Not Enroll in Developmental Writing Those studies which set up a control group most often concern a group of students who need and enroll in the developmental writing course and another group of students who need to, but for whatever reason, do not enroll in a developmental writing course. These stud-i es are often judged to be flawed because students are not ass i gned randomly to the experimental and the control groups. Choosing stu-dents randomly for the deve 1 opmenta 1 course group and the control group may be a strong experimental des i gn, but many researchers have eschewed that option. They or their institutions felt a moral dilemma faced them; for the sake of a design, they could not deny students with weak sk ills the chance to be helped by a deve 1 opmenta 1 course. Consequent ly, researchers have more often chosen to form a control group from students who were accidentally left out of developmental courses, who could not get into a developmental course because it had

PAGE 38

28 closed out, who could not fit the developmental course into their schedule, or who chose to bypass a developmental program or course and enroll directly in regular freshmen courses. Unfortunately, there are design flaws inherent in accidental control groups. For example, all types of accidental control groups cou 1 d eas ily be composed of non-representat i ve samples of students. The population of students who could not get into a developmental course because it had closed out might easily have a higher percentage of the weakest students. Students who could not fit the developmental course into their schedule might primarily be older adults working full-time jobs during the day and trying to register at a college with a limited night schedule. This population might be stronger as a group than primarily younger students who could fit a developmental course into their day-time schedule more easily. Students whose entry scores indicated the need for a developmental course, yet who chose to bypass it and instead enroll in regular freshman courses might be either so self-confident and highly motivated that they do well in the regular freshman courses or so unwise about their own abilities that they fail in large numbers. Some randomly assigned control group studies do follow below, but most researchers cited here opted for the more accidental type of control group because of their reluctance to deny the possible benefits of a developmental course to any student. Randomly assigned control groups Losak and Burns' early study is the most often quoted study in the literature (1971). Students registering full-time at Miami-Dade

PAGE 39

29 Community College in the fall of 1969 and scoring too low on the entry level English test for normal placement into freshman English, were randomly assigned into three groups. The first were registered for an experimental remedial program called the Community College Studies Program, an interdisciplinary, team taught program including counselors and student tutors in addition to instructors in English, the natural sciences, and the social sciences. The second group was enrolled in a traditional remedial reading course. The third group was enrolled in regular freshman English. Although the entry test indicated that students in all three groups needed a remedial English course, there was no significant difference in the number of students in each group who passed regular freshman English. There was no significant difference in the GPA of students in the three groups. Nor was there a difference in the results of a reading post-test. Finally, there was no difference in the attrition rate. Losa k and Burns wondered if the newness of the Community College Studies Progra m or the relatively short period of time (three months) during which this course or the traditional remedial course tried to effect major changes in skills may have contributed to the discouraging results. In an often praised, early study, Sharon (1972) randomly selected some students who would have been required to take remedial English and placed them instead in the regular freshman English course. The only students excluded from this random control group were students who scored extremely low on the entry assessment, and they compri sed only 4% of the students who needed remedial English. The rest of the students judged as needing remedial English were enrolled in that course. There was no significant difference in the persistence of the

PAGE 40

30 two groups to the end of thei r respect i ve courses. The students who received remediation did not pass in significantly greater numbers o r withdraw in significantly fewer numbers than those in the control group who had been placed in freshman English. On the other hand, there was a significant difference between the mean grade of the two groups in freshman Engl ish. Students in the control group placed directly into freshman English earned a mean grade of 1.85 whereas the students who were placed in remedial English, completed it, and went on to freshman Engl ish earned a mean grade in freshman Engl ish of 2.27. Nevertheless, Sharon concluded that "the effectiveness of remedial courses in college should not be taken for granted remediation in English has only limited value" (p. 62). Mowatt (1978) found positive results in both achievement and persistence of developmental English students in a five year, randomized control group study. Non-traditional students entering West Valley College in 1970 and scoring below the college's cutoff on two entry English assessments either enrolled in developmental English or did not. Over the five years, the mean GPA was significantly higher fo r former developmental English students at 2.41 than for non-traditional students who did not enroll in developmental English at 2.16. Over the same time period, the former developmental English students also persisted better in terms of mean number of cumulated units at 40.66 versus 26.24 for those who needed but did not take developmental English. Mowatt concluded that the developmental English course did help non-traditional students improve their GPAs and to accumulate almost twice as many units as similar students who did not enroll in the developmental course.

PAGE 41

31 Non-random control groups Lovejoy (1974) found no significant difference in achievemen t measures, but partially positive results in persistence of academically disadvantaged students entering Western Christian College from 1961 to 1971. Two of his three groups consisted of students in a traditional, formal course in remedial English which heavily emphasized grammar. One group took this course concurrently with regular freshman English; the second took it before enrolling in freshman English. A third group, also academically disadvantaged, did not enroll in the remedial English course at all. In terms of final grade in regular freshman English, neither remedial English group showed a significant improvement over the non-remedial group. In terms of GPA in the six areas of study Lovejoy selected to include, again neither remedi a 1 Engl ish group showed improvement. The on ly sign ifi cant gain came in persistence into a second year of college. The group which completed the remedial grammar course before enrolling in freshman English did continue in significantly larger numbers than the other two groups into a second year. However, the effect did not continue into the junior or senior year. Lunsford's (1978) study of the remedial English course at an Ohio college used a variety of English skills measures in addition to checking persistence and achievement patterns of two groups scoring low on an entry English test. Although low scorers were randomly selected and invited to enroll in the remedial English course, only students who chose to did. From the others was drawn a control group matched with the remed i a 1 Eng 1 ish students on ACT scores, sex, and high school rank. Some of the students in the control group enrolled

PAGE 42

32 in freshman composition while some enrolled in no English course at all during their first semester. In terms of English skills, remedial English students had a significantly h i gher score on their holistically graded ending essay than they did on their first writing sample. The control group did not; in fact, their ending essay mean score was very similar to the remedial students' first writing sample score. This suggested that the control group's writing skills had not improved while the remedial group's had. In terms of persistence and achievement, Lunsford also found that remedial English was beneficial to students. Of the 94 students who enrolled in remedial English, 86 completed, but only 61 of them passed (70%). In other words, about 65% of those enrolled passed the course. Of these completers, 75% went on to complete freshman English in the following term. On the other hand, only 35% of control group students enrolled in freshman composition the first semester, and of these 72% completed. The mean grade in freshman composition for the former remedi a 1 students was 1. 8 with 80% havi ng a C or above whereas the mean grade for the control students was 1.2 with only 13% having a grade of C or above. Two and a half years after the beg i nn i ng of the study, 55% of the remedial students were still enrolled at the university. There were no long term persistence figures cited for the control group, however. Presley (1981) reported on one of the annual evaluation surveys conducted throughout the Georgia postsecondary system. Georgia, since 1974, had had a statewide developmental education program under the university system's Board of Regents. Each college and university had a department of deve 1 opmenta 1 educat i on, and students who scored low

PAGE 43

33 on the statewide entry test had to enroll i n the c o urs e or c ourses which addressed their weaknesses. These students cou ld not enroll in a freshman course in that area unless they had successf ully com p le ted the assigned developmental course. When Presley wrote, the eval uation surveys focused on persistence, but no longe r requ i red a c hieveme nt figures because the consistently higher showings by students who d id not need developmental courses convinced state administr ato r s t hat deve 1 opmenta 1 and non-deve 1 opmenta 1 students were clea r ly not equal groups and that judging developmental students by the nor m of th e nondevelopmental stude nts was setting an impossibly hig h criterion by which to judge the developmental programs . Presley found that system-wide about 8% of the d e v elopmental English students were dropping out of the developmental courses, while the dropout rate for all cour ses was 10%. B etween 50% and 60% of developmental stu dents wer e passing developmental English a n d moving i nto freshman composit i on, although only 21% managed this in one semester. At Presley's own institution, raisin g t h e ent r y test cutoff score had resu lted in two groups with comparab 1 e SAT score s ; t h e new group was required to take the developme ntal English cou rse whil e the previ ous year's students with the same scores had not. Of the s tudents required to take developmental Engl ish and who passed, 72.6 % subseque n t l y completed freshman English on their f i r s t t r y whi l e only 52.7% of the comparably scori ng students who had not taken deve 1 opmental English completed freshman English on their first try. The se findings suggested that the majority of Georgi a developmental E nglish students had trouble passing the developmental English cour s e the first semester they enrolled, but that event ually more than half did

PAGE 44

34 pass developmental English. Those that did pass then had a major advantage; they had a substantially better chance of passing freshman composition than they would have had if they had not taken developmental English. Bers conducted two studies in the fall semesters of 1980 and 1981 on entering students at Oakton Community College. Of the fall 1980 entrants, Bers noted that 32% of all entering students scored below the cutoff on the entry-level English test, but only 53% of these voluntarily enrolled in the developmental English course. Of those, only 53% completed with a passing grade whereas 83% of students enrolled in regular freshman English passed. The remedial course was harder to complete successfully than was freshman English. Bers (1982Q) found similar persistence problems with fall 1981 developmental English students; 55% passed the developmental English course with 21% withdrawing, a withdrawal rate twice as high as in the overall student body. Of those who passed, 68% enrolled in freshman English. However, having passed developmental English did not give these students an edge over students who needed but had not completed develop mental English. The rate at which developmental completers passed freshman English was 89%, but 74% of non-completers passed, not a significant difference. Baker (1982) had serious reservations about the efficacy of remedial English at two-year Snow College. His only favorable finds were that remedial English students were more likely to complete freshman English than the control group who needed remedial but instead enrolled in freshman English. However, Baker's other findings sug gested that remedial English was not clearly a more effective way to

PAGE 45

35 improve student writing skills. For example, scores on an objective English test and on pre-and post-essays graded independently, were not significantly different for the remedial English students and for the control g roup. In other words, the study suggested that students improved their writing skills as much b y avoiding remedi al English as by taking it. In terms of retention, the students in regular English fared better: only 14% of them withdrew before the end of the term, whereas 24% of the remedial English students withdrew. Research Quest i on #3: Deve 1 opmenta 1 Compl eters Versus Control Groups Who Did Not Need Developmental English Reap and Covington (1980) compared the attrition and achiev ement on a total of 4,900 students in three developmental English courses and the regular freshman English course from the fall of 1973 to the spring of 1980. During this seven year period, the proportion of entering students registered in developmental Engl ish steadi ly increased from 15% to 33% with the rest enrolling in regular freshman English. With three levels of developmental courses, the dropout rate was the worst at the lowest course level,' averaging 40%. The middle level did best with a withdrawal rate average of 26%, but the rate again rose to 30% in the highest level of developmental English. This was about the same attrition rate as freshman Engl ish experienced at 29%. Combining the withdrawal rates of all developmental courses for 1977-1980, Reap and Covington found that the attrition rate was about equal for them and the freshman English course, 29.4% and 28.5% respectively. Moreover, in terms of achievement, the average grades in each of the courses were about the same, with the exception of the middle level developmental English course, whose average was slightly

PAGE 46

36 higher: 1.7 in lowest level, 2.3 in middle level, 1.9 in highest level developmental, and 1.9 in freshman English. Reap and Covington here reported positive results, but these were short-term measurements, taken during and at the end of one semester. The long-term measures reported were not as positive. Former developmental English students were more apt to complete freshman Engl ish successfully if they had been in a developmental course only one semester. Moreover, although an average of 30% of entering students enrolled in a developmental course, on ly about 16% of the students in freshman Eng 1 ish had completed a developmental English course or courses. Obviously, substantial numbers of students who started in developmental English did not persist to enrolling in Freshman Composition. Braxton et al. (1980) studied the persistence and achievement patterns of developmental studies students at Thomas Nelson Community College from 1975 to 1977. Their program had English, reading, and math courses run by a division of developmental studies. Of the students mandatorily placed in the developmental English course, 62% completed the course. Of the students who completed the course, 71% passed. Of those who did not pass the first time, 48% were successful in passing the second quarter they were enrolled in the course and 49% were successful the third semester. When students had successfully passe d the developmental English course, they did almost as well in regular freshman English as students who did not need developmental English. Of the developmental English completers, 68% earned a C or better grade in freshman English compared with 73% of the non developmental students. Braxton et al. also supplied long term data on the persistence of all developmental studies students, including

PAGE 47

37 English students. Of all the students enrolled at the college in the fall quarter of 1975, 62% of non-developmental students reenrolled in the winter quarter, but the higher percentage of 73% of developmental students reenrolled. Exactly two years after their admission, 9% of non-developmental students had graduated, while somewhat fewer, 6%, of the deve 1 opmenta 1 students had graduated. The authors noted that two calendar years were probably not enough time to take accurate measures of graduation rates because many adult part-time students would have needed a longer period to complete their programs, because a large number of students who enrolled there were not degree-seek i ng students, and because students placed in developmental studies courses had a delayed entry into the regular curriculum which normally prolonged thei r program beyond two years. As for overall achi evement patterns, developmental students' average GPA did lag; it was significantly lower than the GPA for non-developmental students. In all, the developmental English students at this community college did about as well in freshman English as non-developmental students, once the developmental students passed their developmental course. Develop mental students persisted to the second quarter at a higher rate than non-developmental students; however, they had a significantly lower GPA, and they graduated withi n two years at a somewhat 1 esser rate than non-developmental students. Smittle (1982) in an overall evaluation of Santa Fe Community College's developmental studies program, found the persistence and achievement of developmental writing students encouraging. Of the enteri ng students in the fall 1981 semester, 76% of those requi red to take Basic Writing Skills completed the course. This would mean that

PAGE 48

38 24% dropped out of the course before the end of the semester. Of course completers, 75% passed. That meant that of the students required to enroll in developmental writing, 57% passed it on their first try. The following semester, 71% of the students who had enrolled in Basic Writing Skills returned for more college courses. For a long term measure, Smittle compared the passing rate in the winter 1982 College Composition course of successful fall Basic Writing Skills students with that of students who scored above the entry cutoff and had not taken the deve 1 opmenta 1 course. She cone 1 uded that the former developmental students, of whom 44% passed College Composi tion, did as well as the students who had not needed developmental English, of whom 46% passed College Composition. Therefore, Smittle suggested that the deve 1 opmenta 1 Engl ish students who passed Bas i c Writ i ng Sk ill s had about an equal chance for success in the freshman English course as did the stronger students. Barton (1984) cited mixed results from a study on students entering a developmental education program at a multicampus, urban community college during the fall of 1980. Approximately one-quarter of all entering students enrolled in the developmental program. When compared with non-developmental students, developmental English course completers did not fare well in their subsequent college English course, having a significantly lower mean grade than the non deve 1 opmenta 1 students. However, Barton was encouraged that 68% of the developmental students completed the college level English course with a grade of C or higher. In terms of long range achievement, the developmental program completers had a significantly lower cumulative GPA. The more developmental courses the students took, the lower

PAGE 49

39 their cumulative GPA was apt to be. However, when Barton analyzed the relationship between number of credits earned and cumulative GPA, he found that the cumu 1 at i ve GPA of students who had taken on ly one deve 1 opmenta 1 course increased as the number of quarters they were enrolled increased, eventually surpassing the cumulative GPA for students who had taken no developmental courses. In long term persistence measures, Barton found that although there was no significant difference in the number of quarters enrolled between full-time developmental and non-developmental students, part-time developmental students did have a significantly higher number of quarters enrolled. Studies with two control groups Many studies used two control groups, one of students who fell below the cutoff score on entry testing but did not enroll in develop mental English, and one of students who scored above the cutoff, indicating they did not need the developmental course. Both these groups were contrasted with the experimental group of students who needed and enrolled in developmental English. Dudley (1978) reported that an individualized English remedial skills course at the St ate University College in Brockport, New York, seemed to have produced positive results when the remedial skills students' achievement and persistence were compared t o two control groups, one of students who needed but did not enroll in the remedial English course and one of students who did not need the remedial course based on the entry level Engl ish test battery. Dudley used only long term measures: grade in regular freshman Engli sh; grad e in a freshman speech course; GPA in first, second, third, and fourth

PAGE 50

40 semester of enrollment; and retention rate over the first, second, third, and fourth semesters. There were no differences in speech course grades between students who completed the individualized remedial English course and students in the control groups. But in freshman English, the former remedial English students earned a statistically significantly higher mean grade than did the control group of students who scored above the cutoff on the entry Engl i sh test and therefore were exempted from remedial English. However, there was no significant difference in the former remedial students' grades in freshman Engl ish compared with the grades for students who scored below the cutoff but did not enroll in remedial English. The former remedial English students did, on the other hand, earn a statistically significantly higher GPA than this control group d id in their second and third semesters of enrollment. There was, even more encouragingly, no difference between the GPA of the former remedial students and the other control group, those who did not need remedial English. In addition, 35% of the remedial students who persisted to the fourth semester showed an upward trend in GPA whi le only 9% of students in the control groups did. As for retention, the dropout rate for the controls was higher than for the remedial students: higher for the high scoring control group in the second semester, higher for both control groups in the third semester, and highe r in the fourth semester for the low scoring control group of students who should have taken remedial English but did not. Dudley concluded that the individualized remedial English class, a course seemingly quite similar to the one studied in this research, helped some students survive in college who probably would not have without this course.

PAGE 51

41 Linthicum (1979) reported on a statewide evaluation of develop mental/remedial programs in Maryland community colleges and ended up questioning the effectiveness of these programs but at the same time admitting that the range of some pieces of data might have been a significant limitation of this research. For example, there was no significant difference in the GPA, retention, and persistence rates between students who had scored low on entry testing and enroll ed in developmental English and students who had scored low but did not enroll in it. But Linthicum noted that since different institutions used different cutoff scores on the entry level test, the student populations may have been too disparate to have been analyzed as a group. In the comparison of developmental English students with the other control group, the students who did not need developmental English, the developmental students did not fare well. In achievement, developmental English students achieved a cumulative GPA ranging from 1.66 to 2.91 at various institutions while the stronger students achieved GPAs between 2.58 and 3.21. Developmental English students earned a significantly lower grade in freshman English than non-developmental students did. On the brighter side, there was no significant difference between the length of time developmental English students and stronger English students continued taking courses in college. Although the ratio of courses completed to those attempted statewide was 76% for the developmental English students but 90% for the stronger students, Linthicum cautions that this reflects a large disparity between institutions (59% to 94% for developmental students, 75% to 96% for non-developmental students), which was due to variations in withdrawal policies.

PAGE 52

42 In New Jersey, statewide evaluation of what are called remedial programs has had more positive results (New Jersey Basic Skills Coun cil, 1983). As in Maryland, the results for individual community colleges, colleges, and universities were uneven with substantial ranges of positive to negative results, but the overall picture was reported as pointing to some success for the state's postsecondary remedial programs. For example, of the full-time students in community colleges, 65% passed their remedial writing course statewide. The range among the community colleges was wide, however, running from a low of 46% passing at one institution to a high of 82% at another. For parttime students, the results were not quite as favorable with an average passing rate of 59% and a range of 28% to 90%. Moreover, community college students who completed remedial writing dropped out of college after their first semester at a much lower rate (14%) than did either the students who needed but did not complete remediation (33%) or the high scoring students who did not need remedial writing at entry (20%). In terms of completion of regular freshman English courses , community college students as a whole who completed remedial English passed regul ar freshman Engl ish at about the same rate as those who did not need remedial writing and better than those who needed but did not complete remedial writing. In terms of GPA, across all community colleges, remedial writing completers earned a mean GPA of 1.93, lower than the high scoring entry students' mean of 2.31 but higher than the mean of 1.32 for low scoring students who did not complete remedial writing. Instead of the mean number of credits earned, this report also listed the ratio of credits earned to credits attempted. Again, the remedial writing completers did not fare as well as the stronger

PAGE 53

43 students at .78 to .81 respectively, but the completers did do better than the students who needed but did not complete a community college remedial writing course at .63. The authors of the New Jersey study also included data for reading, computation, and elementary algebra courses and for all public institutions of higher education in the state. They noted that the wide variations in statistics reported from various institutions must not be overlooked in drawing conclusions about remedial programs at specific colleges. They urged that attention be paid to the characteristics of the more successful programs in an attempt to upgrade the weaker ones. Finally, they cautioned that it may be too early to tell about the overall effectiveness of the state's remedial programs and that further one and two year follow-up studies will help clarify how effective the programs have been. Research Questions #4: Age as a Factor for Developmental Students Linthicum (1979), in a study noted earl ier, also found that when results were factored by age, older students who completed develop mental English did just as well in freshman English as the control group of students who did not need a developmental course. The developmental students over 30 were significantly more successful than the 19-21 year 01 ds, who had the lowest entry test scores in the age breakdown. I n add it ion, deve 1 opmenta 1 course comp 1 eters age 22 or older achieved the same mean grade in freshman English, had the same retention, persistence, and cumulative GPA patterns as students who never needed developmental English. These older students comprised only 35% of the developmental English students, however, and the

PAGE 54

44 majority of the developmental stude nts were age 19-21, in the lowest scoring of all Linthicum's age categories. Therefore, although the 22 and older students ' performance was equal to the performance of the non-developmental English students, the performance of the younger students brought the overall performance of developmental students below that of the control group. Whittle (1980) found that at Piedmont Virginia Community Coll ege, there was no relation between age and grade in the developmental English course. The students who were most successful completed only one quarter of developmental English. The chance for success dimin ished with each successive quarter a student spent in develop mental English. In freshman composition, however, the highest grades wer e received by those former developmental students who were enrolled part-time, took fewer quarters of developmental English, were older and were female, in that order, according to Whittle's weighted linear combination of predictor variables correlating success in develop mental English and in freshman composition. This result also sug gested that age was a factor in former developmental students' success in regular courses and that older students later achieved at a higher level than recent high school graduates. Dumont and Jones (1983) also did a predictive analysis study involving age as a factor in developmental course success. As part of a statewide study on selected remedial education programs in a South eastern state, Dumont and Jones found at one regional university that age was the most important predictor in students' success in the remedial compositjon course. Older students did better than younger students in earning grades of C or higher. The other predictors were, in

PAGE 55

45 descending importance, the student's confidence in his or her ability to master composition, sex (females did better), score on the locally developed composit ion pre-test, the student's perception of the course's usefulness, and race. Summary Although some studies cited encouraging results of developmental English programs, others did not. For example, the rate of successful completion of developmental Engl ish courses seemed to range between 50% to 70%. About one-fourth to one-third of successful completers did not go on to enroll in freshman English. When contrasting results for individualized versus classroom modes of developmental courses, these studies showed mixed results. Students in both the classroom and individualized courses completed their developmental English at about the same rate in the three studies measuring this. On the other hand, one study cited higher grades in freshman English and higher GPA for individualized course completers while another reported the opposite, that classroom course comp 1 eters had higher freshman Eng 1 ish grade s and GPA. None of the studies showed significant difference in persistence when using long term measures. More mixed results occurred among the studies with a control group of students who needed but did not complete developmental English. In performance in regular freshman English, developmental completers did better either in passing or in attaining a mean higher grade than the controls in four studies but showed no significant difference in three reports. Of the stud i es measuri ng GPA, three

PAGE 56

46 cited significantly higher cumulative GPA for the developmental completers while three reported no significant differences. Persistence over two or more semesters was higher for former developmental students in three studies, about equal to the controls in one study, but lower than the controls in two studies. The only two reports measur ing credits or units earned both cited significantly higher accu mula tion for the former developmental students. Several authors noted, however, that the high attrition rate among students who originally enrolled in the developmental course, a fate which the control students had not suffered, could have contributed to this positive result. Of course, these studies were conducted at a mixture of community colleges, four-year colleges, and universities. As Kulik et al. cautioned in their meta-analysis on cumulative GPA and persistence, community college studies by and large resulted in no significant differences between deve 1 opmenta 1 program comp 1 eters and control groups of simi 1 ar aptitude who did not complete deve 1 opmenta 1 programs. Developmental completers were more likely to do worse than the control groups of stronger students who did not need developmental English, but occasionally held about equal with them. Developmental completers earned a higher grade in regular freshman English than stronger students in only one study. In two studies, the no significant difference in freshman English grade may also be viewed as encouraging since the developmental English completers originally had started college with a measurable disadvantage in English skills. Two studies did, however, find the developmental completers earning significantly lower grades. Also encouraging were the two studies

PAGE 57

47 measuring completion rate in freshman English; both reported no significant difference between former developmental and stronger students. On longer term measures, development al students did not fare as well with four out of five studies having significantly higher GPAs for the stronger students (one with no significant difference). The only study citing credits earned showed stronger students as higher, and the only study citing degrees earned showed stronger students as higher also. Finally, the number of semesters of persistence was more encouraging with three studies reporting no significant difference and one reporting partially better persistence for developmental completers. In the three studies involving older versus younger developmental students, older students fared better than younger on the few measures used. Grade in developmental course was higher for older students in one study but showed no sign i f i cant d i ff erence in another. A higher proportion of older students completed their developmental Englis h course in the one study while being older was one of four factors related to highe r completion rates in another. For long term measures, one study cited older students as having higher cumulative GPAs and more semesters completed than younger students. With mixed results like these, in some cases involving relatively few studies applicable to the research questions in this study, there emerges a need for more research on the effects of developmental English programs, especially in Florida and other states where such programs ina 11 disc i P 1 i nes are comi ng under i ncreas i ng 1 eg is 1 at i ve scrutiny and mandates.

PAGE 58

CHAPTER 3 METHODOLOGY General Resear ch Design This ex post facto s tudy employed a quasi-experimental, non-equivalent control group design. Archival data were the source for pretest scores partially responsible, along with self-selection, for placement of fall, 1980, Valencia Community College students into the four groups in thi s study descri bed below. Student transcri pts were the source for the records of student per s istence and achievement. Instrumentation Data were collected from computer tapes of entry-level testing results from fall 1980, and from the college's student data base. The records inc luded the Initial English Assessment (lEA) test sc ores ; demographic data on age, sex, and race; and student transcripts. The Initial English Assessment Test at Valencia Community College (see Appendix A) was developed and validated by a team of the college's English instructors under the direction of a consultant expert in Eng-lish objective test writing. Reliability was established and cutoff scores were set with the help of the Vice-President of Institutional Research. The cutoff scores were as follows: 22-35 High Scores -students should enroll in Freshman Composi tion I. 19-21 Inconclusive Scores 48 students may enroll in either a de velopmental English course o r i n Fre shman Composition I .

PAGE 59

49 0-18 Low Scores --students should enroll in a develop mental English course. Academic advising occurred after a student had submitted an appli-cation to the college and completed the entry level placement tests in English, reading, mathematics, and study skills. During the academic advising session, a member of the counseling staff reviewed the student's Initial English Assessment test score, along with the other ini-tial assessment test scores in reading, mathematics, and study skills. The counselor also reviewed the student's transcript from seconda ry school, if available, and discussed which courses the initial assess-ments indicated the student should take. The data were organized by the Engl ish course status of students during their first semester, that is, whether their entry test scores indicated a need for developmental English course work and they enrolled in the classroom, individualized, or no developmental course or whether their entry test scores indicated no need for developmental English course work and they enrolled in regular Freshman Composition I. The data were further organized by student age at entry to the college, as reported by the student. Sample The total population involved in this study consisted of 1,431 Valencia Community College students with the following characteristics: (1) enrolled at the college for the first time in the fall semes-ter of the 1980-1981 school year, (2) were first-time-in-college students, not transfers, (3) completed the Initial English Assessment test, and (4) registered for classes on the West Campus. The 1,431 students were di vi ded into four groups on the bas i s of their Initial English Asses sment score and their first semester English

PAGE 60

50 course status. Then, 100 students in each grou p wer e randomly selec-ted. Included were both full-time and part-time students and both day and evening students. T h e 400 students in the four random samples were e nrolled in the following courses in Session I, 1980-1981: 100 in the classroom c ourse who needed a deve lopmental course (lE A Score of 0-18), 100 in the individualized course who needed a deve l o pmental course (lEA Score of 0-18), 100 not in an English developmental cours e even tho ugh thei r entry assessment scores i ndi cate d they needed one (l E A score of 018) , 100 in regular Freshman Composition I because thei r scores were over the cutoff (lEA score of 22-35). Students with inconclusive scores of 19-21 wer e exc lude d f r o m the study. Of the 100 not in a n English developmental cou r se eve n though their entry assessment sc o r es indicated they need e d one, abo u t f o ur f ifth s enrolled directly in Freshm a n C o m position I while abou t one-fifth enrolled in no Engli sh course at all in the fall, 1980 , se m ester. C o llection of the Dat a Data were collected from June 1980, when the first round of Ini -tial English Assessment testing and academic advisement sessio n s were conducted for students entering in the fall session of 1 980-8 1 , t hrough April, 1984, a span of eleven semesters, nearly four academic years. Two reasons exi sted for conducti ng a nearly four year study on students ina community coll ege. Fi rst was to compensate as much as possible for the extended time period many part-time students n eed to complete their associate degree. Second w as to compensate for the

PAGE 61

51 possible semester or more that students originally enrolling in devel-opmental courses often need to add to their academi c careers since they must dela y tak ing one or more of the standa rd freshm a n courses whil e they strengthen their skills in developmental courses. The following data were collected: studen t Social Security number; age at entry, sex, and race as reported by student; score on Initial English Assessment; which developmental English course enrolled in during Session I, 1980-81, if any; completion of that course; grade in that course; enrollment in Freshman Composition I; grade in Freshman Composition I; cumulative GPA; cumulative number of semesters com-pleted; cumulative number of credits earned; and associate degree earned, if any. Analysis of the Data Four groups were studied in order to answer the first three re-search questions concerning achieve ment and persistence of developmental students in each course and of the developmental students versus students in the control groups. Two were experimental groups; two were control groups. The two experimental g r oups consisted of the follow-ing: DC developmental classroom students who needed a d e velop mental course and enrolled in the classroom course (lEA score of D IS), and 01 developmental individualized students who needed a develop mental course and enrolled in the individualized course (lEA score of 0-18). The two control groups consisted of the following: CNO -control group of no devel opmenta 1 course students who needed a developmental course but did not enroll in either the individualized or the classroom course (lEA score o f 0-18), and

PAGE 62

52 CHI -control group of high scorers on the lEA who neither needed nor enrolled in a developmental course because their scores were above the high cutoff score (lEA score of 22-35). lEA scores of 19-21 were judged to be inconclusive. Counselors recom-mended each student to take either Freshman Composition I or a developmental English course depending on other factors like high school English grades. Students with inconclusive scores were, therefore, not i ncluded in this study. Figure 1 represents the relationship betwee n the four groups by lEA score. EXPERIMENTAL GROUPS DC Developmental classroom group 01 Developmental individualized group lEA SCORE 0-1 8 22 -35 ___ CONTROL GROU PS CNO Control with no develop mental course CHI Control with high scores & no need for developmental course Figure 1. Dia ram of Grou s for Research uestions 1, 2, and 3 as Determined by Score on Initia English Assessment Test The two devel opmenta 1 groups were combi ned and then redi vi ded ac-cording to age in order to answer the fourth research question

PAGE 63

53 concerning achievement and persistence of s tudents four o r more years out of high school versus that of students who have recently g raduated from high school. DC Developmental Classroom Students + DC Developmental Individuali zed Students Figure 2. tion 4 as Figure 2 depicts this realignment. DC + 01 Age 17-2 1 DC + 0 1 Age 22 or Older nment for Researc h u es A frequency table was constructed, and the d escript ive statistics of mean, standard deviation, median, mode, range, skewness, and kurto-sis were computed for the lEA scores over all four groups. Descriptive statistics of se x , race, and age were noted over all four groups. The CHI group was nonequivalent to the other three groups since its members had to have scored above the cutoff on the I nitial E nglish Assessment in order to be members of this gro u p, whereas membership i n the other three groups was partially determined by scores belo w the cutoff. Mean scores on the I EA by groups were analyzed us i ng ana 1 ys is of variance followed by the Scheffe Procedure to determine if any other

PAGE 64

54 groups had means with s i gnifi cant di fferences. One-way analys i s of variance was also applied to means of the developmental groups divided by age. The independent variable was the score on the Initial English As-sessment. The dependent variables were divided into three sets: those to which one-way analysis of variance was applied, those to which m ul-tivariate analysis of variance was applied, and those to which chi-square followed by the Yates Correction was applied. One-way analysis of variance (ANOVA) was used to establish if any statistically signifi-cant differences at the .05 level existed for the following dependent variables: grade in developmental course grade in Freshman Composition I. Multivariate analysis of variance (MANOVA) followed by multivariate F tests was used to establish if any statistically significant d if-ferences at the .05 level existed between the values of the long-term variables measured in each of the groups. MANOVA allows possible in-teractions among the dependent variables to be evaluated. When the multivariate F value following MANOVA was significant, univariate F tests were used to determine which variables were significantly differ-ent in and of themse l ves. The set of dependent variables analyzed w ith MANOVA included the following: cumulative grade point average (GPA), number of semesters enrolled in nearly four years, number of credits earned in nearly four years. Chi-square analysis followed by the Yates Correction was applied to the third set of variables: number and percent of students completing a developmental course, number and percent of students completing Freshman Composition I number and percent of students completing an associate degree.

PAGE 65

CHAPTER 4 FINDINGS This study was designed to evaluate a community college developmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an individualized format, so that the college might assess the efficacy of each course and of the ent i re program in terms of student pers i s-tence and achievement. Four main research questions were asked about achievement and persistence patterns of the developmental English stu-dents. Eight null hypotheses, two relating to each research question, were tested. First, comparisons were made between students enrolling in the classroom developmental English course and the individualized develop mental English course to see if an individualized developmental writing course or a classroom based developmental writing course con-tributed to better achievement and longer persistence by students who needed a developmental writing course when they started college. Second, comparisons were made between students completing a developmental course and those who needed but did not enroll in either developmental course to see if students who needed and completed either of the developmental writing courses achieved better and per-sisted in college longer than students who needed but did not enroll in a developmental writing course. Third, comparisons were made 55

PAGE 66

56 between students completing a developmental course and students with stronger writing skills who did not need a developmental writing course to see if the developmental writing courses helped students with weak writing skills at entry catch up, in terms of achievement and persistence, with students who did not need a developmental writing course when they started college. Finally, comparisons were made of developmental course completers divided into two age groups, those age 17 to 21 and those age 22 or older, to see if older adults achieved and persisted any differently from recent high school graduates after completing a developmental writing course. This chapter presents the data collected in this study. There are five sections. The first describes background data on the four groups and correlations among independent variables across the four groups. The remaining four sections address each of the four main research questions and present statistical evidence that either prove d or disproved each of the study's eight null hypotheses. Background Analysis of Samples When all four groups were analyzed, the following characteristics emerged. Females made up 55% and males 45% of the samples. Whites composed 75.7%, Blacks 17.5%, American Indians 2.5%, Asians 0.7%, Hispanics 3.0%, and others 0.5% of these students. Students aged 17 to 21 comprised 82.5% of the samples while students 22 or older made up the remaining 17.5%. Their scores on the Initial English Assessment test had a mean of 17.0. The test scores for the combined groups did not form a normal distribution, as can be seen in Table 1, because three of the four groups were, as planned, chosen from students

PAGE 67

57 scori ng at or below the cutoff score of 18 on the lEA. The fourth group cons i sted on ly of students scori ng at or above the hi gh cutoff of 22 on the lEA. Consequently, the students with inconclusive scores of 19 to 21 were excluded from the study. TABLE 1 A Descriptive Analysis of Initial English Assessment Test Scores Mean 17.02 Median 16.00 Standard Deviation 5.23 Mode 17.00 Range 32.00 Skewness .53 Minimum 3.00 Kurtosis .10 Maximum 35.00 In order to test the assumption that the groups under study were equivalent, except for the control group of students scoring high on the entry test, one-way analysis of variance (ANOVA) was performed on the mean score on the Initial English Assessment test for each of the four groups . The resultant F value was significant, so the Scheffe Procedure was applied to indicate where the differences occurred. As shown in Table 2, there was no significant difference in the mean score of the classroom course students at 14.16 and the individualized developmental course students at 13.67. The individual ized developmental group did have a significantly lower mean at 13.67 than did the control group of IINo-takes, II (students who needed but did not enroll in a developmental course) at 15.19. There was, as expected, a significant difference between the mean of the control group of high scoring students at 24.60 and the means of the other groups.

PAGE 68

58 TABLE 2 Results of Analysis of Variance on the Mean Scores on the Initial English A s sessment Test by Groups Source df SS MS F Between Groups 3 7783.58 3129.29 10912.88 2594.53 7.90 328.33* With in Groups Total Group DC (developmental 396 399 classroom) Scheffe 01 (developmental individualized) CNO (control, No-takes) CHI (control, high scorers) * p < .05 Procedure Mean DC 01 CNO CHI 14.16 13.67 15.19 * 24.60 * * * In addition to the significant difference between the individualized developmental group mean on the lEA and that of the control group of No-takes, there was also a significant difference in the means on the lEA when the individualized developmental group and the classroom developmental group were combined. This combined experimental group was important in answering the second research question. As seen in Table 3, the combined developmental groups' mean of 14.14 was significantly lower than the No-takes' mean of 15.19. As for the developmental students divided according to age, there was also no significant difference at p < .05 between the mean score on the entry test between younger and older students as recorded in Table 4.

PAGE 69

59 TABLE 3 A Comparison of the Mean Scores on the Initial English Assessment Test Between th e Combined Developmen t a l Groups and the Control Group o f N o takes Group DC + 01 (Combined Developmental Groups) CNO (No-takes) Source Between Groups Within Groups p < .05 TABLE 4 Mean 14.14 15.19 df 73.50 2494.47 SS 1 289 Standard Deviation 2.91 2.86 MS 73.50 8 .37 N 200 100 F 8.78* A Comparison of Developm ental Writing Students ' Mean Scores on t h e Initial English Assessment Test by S tu dent Age at Entr y Standard Age Mean Deviation N 17-21 14.17 2.89 168 22+ 14.00 3 .06 32 Source df SS MS F Between Groups 1 .75 .74 . 0 9 Withi n Groups 198 1683.33 8.50 p < .77 These findings were important to all of the following comparisons pertaining to the major research questions and their null hypoth eses . Since there were no s ignificant differences between the l E A m eans of the two experimental developmental course groups and between the younger developmental course stud e nt s and the older developmental

PAGE 70

60 course students, the independent variable of lEA score for these groups was essentially equal. The inequality of the mean lEA score between the combined developmental groups and the control group of Notakes implies that the No-takes had somewhat stronger writing skills at entry than students electing to enroll in a developmental writing course, which has implications in the interpretation of later analy-ses. The inequality of the mean lEA score between the combined developmental groups and the control group of high scoring students underscored the continued inequality in some of the dependent vari-ables measured between these two groups. Research Question #1: Classroom Versus Individualized Developmental Course The first research question was concerned with whether or not an individualized developmental writing course or a classroom based developmental writing course contributed to better achievement and longer persistence by stu dents who needed a developmental writing course when they started college. The first hypothesis stated that there was no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course. Table 5 presents the results of analysis of variance applied to grades in the two different developmental courses. Of the 100 stu-dents enrolled in the classroom course, 86 completed it that semester.

PAGE 71

61 Of the 100 students enrolled in the individualized course, 75 com-pleted. Of the completers, the mean grade in the classroom course was 2.85, while the mean grade in the individualized course was 3.05. Analysis of variance resulted in an F value of 2.08, which was not significant at p < .05. TABLE 5 A Comparison of the Mean Grade in Developmental English Courses Between Students in the Classroom and Individualized Courses Standard Group Mean Deviat ion N DC Classroom 2.85 .93 86 01 Individualized 3.05 .86 75 Source df SS MS F Between Groups 1 1.68 1.67 2.08 Within Groups 159 128.82 .80 p < .15 The mean grade wh i ch deve 1 opmenta 1 course comp 1 eters earned in Freshman Composition I also did not show any significant difference between students who completed the classroom or the individualized developmental course (see Table 6). Of the 86 classroom course completers, 65 enrolled in Freshman Composition I and 49 completed it wi th a mean grade of 2.51. Of the 75 individual ized course com-p 1 eters, 57 enrolled in Freshman Compos it i on I and 44 completed it with a mean grade of 2.73. Multivariate analysis of variance was applied to the variables of grade in Freshman Composition I, GPA, number of semes ters enro 11 ed, and number of cred i ts earned. MANOVA resulted in a multivariate F of .48, which meant there were no

PAGE 72

62 significant differences at the .05 significance level on any of these measures between the two developmental groups. TABLE 6 A Comearison of Mean Grade in Freshman Comeosition I! GPA2 Semesters! and Credits of Develoemental Comeleters of the Classroom and Individualized Courses Standard Variable Groue Mean Deviation N GRADE IN DC 2.51 1.05 49 FRESHMAN COMP. I 01 2.73 .84 44 GPA DC 2.46 .47 49 01 2.54 .50 44 SEMESTERS DC 5.16 2.25 49 DI 5.29 2.52 44 CREDITS DC 39.94 22.00 49 01 41.61 21. 71 44 Multivariate F = .48; df = 5, 89; P < .79 The cumu 1 at i ve GPA of these deve 1 opmenta 1 course and Freshman Composition I completers also showed no significant difference, as seen in Table 6. Former classroom developmental completers amassed a GPA mean of 2.46 while former individualized developmental completers amassed a 2.54. MANOVA showed a non-significant multivariate F of .48. Consequently, in terms of achievement, of those students who com-pleted their respective developmental writing course, the mean grade they earned in the deve 1 opmenta 1 course, the mean grade in Freshman Composition I, and the mean cumulative GPA showed no significant dif-ference between groups. Therefore, the first hypothesis--that there

PAGE 73

63 was no significant difference in achievement (in terms of mean grade in developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between the students who needed a developmental writing course and completed the individual ized course and students who needed a developmental writing course and completed the classroom course--was not rejected. The students in both courses achieved about equally as well. The second hypothes is concerned pers is tence, that there was no significant difference in persistence (in terms of completing the deve 1 opmenta 1 course, subsequent ly comp 1 et i ng Freshman Compos it i on I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individualized course and students who needed a developmental writing course and enrolled in the classroom course. As seen previously in Table 6, the mean number of semesters completers stayed enrolled at the college and the number of credits they earned were not significantly different between the two courses. The classroom completers stayed enrolled a mean of 5.16 semesters and earned a mean of 39.94 credits while the individualized completers stayed enrolled a mean of 5.29 semesters and earned a mean of 41.61 credits. The MANOVA analysis yielded a non-significant multivariate F of .48 at the .05 level. Table 7 shows that of the 100 students enrolled in the classroom developmental course, 86 (86%) completed it while individualized course completers numbered 75 (75%). This resulted in a chi-square

PAGE 74

64 value of 3.18 after the Yates Correction; this was non-significant at the p < .05 value with 1 degree of freedom. TABLE 7 A Comparison of the Completion Rate of Developmental Courses Between the Classroom and Individualized Groups Original Group Number Completers % DC 100 86 86.0% 01 100 75 75.0% Chi-square after Yates Correction = 3.18; df = 1; p < .07 Also non-significant was the difference between the number of deve 1 opmenta 1 course camp 1 eters who went on to enroll in and complete Freshman Composition I, as seen in Table 8. Of the 86 classroom completers, 65 enrolled in Freshman Composition I and 49 (75.4% of those completers who enrolled) finished Freshman Composition I with a grade of A, B, C, or D. Of the 75 individualized completers, 57 enrolled in Freshman Composition I and 44 (77.2% of completers who enrolled) com-pleted it successfully. This resulted in a chi-square value of .00 after the Yates Correction, which does not equal a significant p < .05 value at 1 degree of freedom. TABLE 8 A Comparison of the Completion Rate of Freshman Composition I Between Students Who Completed the Classroom and the Individualized Develop mental Courses Original Developmental Enrolled in Completed Group Number Completers Fresh. Camp. I Fresh. Camp. I % DC 100 86 65 49 75.4% 01 100 75 57 44 77 .2% Chi-square after Yates Correction .00; df = 1; p < .9 8

PAGE 75

65 Finally, there was also no significant difference between the rate at which developmental course completers earned an associate degree (see Table 9). At the end of 11 semesters, 14 of the 86 classroom course completers (16.3%) had earned a degree while 12 of the 75 individualized course completers (16.0%) had earned a degree. That equalled 14% of the original 100 classroom developmental students and 12% of the original individualized developmental students. The chi-square value of .00 after the Yates Correction meant no significance at the p < .05 level. Chi-square after Yates Correction = .00; df = 1; p < 1.00 Therefore, the second hypothesis--that there was no significant difference between the persistence of students completing the class-room developmental writing course and the individualized developmental writing course--was not rejected. The students completing both courses seem to have persisted about equally as well. In sum, no significant differences appeared on any of the mea-sures of achievement and persistence between students who enrolled in the classroom developmental writing course and those who enrolled in the individualized developmental writing course. The answer to the first question is that the individualized course and the classroom

PAGE 76

66 course contributed about equally to the achievement and persistence of students who needed a developmental writing course when they started co 11 ege. Research Question #2: Developmental Groups Combined Versus Control Group of No-takes The second research quest i on was concerned wi th whether or not students who needed and completed either of the developmental writing courses achieved better and persisted in college longer than students who needed but did not enroll in a developmental writing course. The third hypothesis stated that there was no significant difference in achievement (in terms of grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course (the "No-takes") and students who needed and completed a developmental writing course. With regard to achievement measures, there was no significant difference between the Freshman Composition I grade of developmental completers and that of the No-takes who had enrolled in Freshman Com position I (see Table 10). Of the classroom developmental completers, 49 also completed Freshman Composition I. Of the individualized developmental completers, 44 also completed Freshman Composition I. Together, 93 students completed both a developmental course and Freshman Composition 1. Of the original 100 control group of No-takes, 78 enrolled in Freshman Composition I (22 enrolled in no English course during the fall 1980 semester). Of those 78, 59 completed Freshman Composition I with a grade of D, C, B, or A. The mean grade of the developmental completers was 2.61 while the mean grade of the control

PAGE 77

67 group of No-takes was 2.63. The F value of .01 after analysis of variance was not significant at the .05 level. TABLE 10 A Comparison of the Mean Grade in Freshman Composition 1 of Combined Developmental Groups Versus Control Group of No-takes Standard Group Mean Deviation N DC + 01 2.61 .96 93 CNO 2.63 .81 59 Source df SS MS F Between Groups 1 .01 .01 .01 Withi n Groups 150 124.39 .81 p < .91 The one significant difference which did occur was between the cumulative GPA of the combined developmental completers and the control group of No-takes as seen in Table 11. In order to control for multiple interactions between the dependent variables of cumulative GPA, number of semesters of persistence, and number of credits earned, all of which were long-term measures, multiple analysis of variance was app 1 i ed. The deve 1 opmenta 1 comp 1 eters totalled 161 (86 who com-pleted the classroom course and 75 who completed the individualized course). They were compared to the 100 students in the control group of No-takes. The mean GPA for the developmental completers was 2.39 while the mean GPA for the control No-takes was 2.00, which proved to be a significant difference. MANOVA analysis yielded a significant multivariate F of 4.01. Further analysis with univariate F tests showed significance at the .05 level only for the GPA and not for the

PAGE 78

68 number of semesters enrolled or for the number of credits earned. Therefore, the developmental completers did show a significantly higher cumulative GPA than did those students who should have enrolled in a developmental writing course but did not. TABLE 11 A Comparison of the Mean GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups and the Control Group of No-takes Standard Variable Group Mean Deviation N GPA DC + 01 2.39 .6 3 161 CNO 2.00 1.21 100 SEMESTERS DC + 01 4.31 2.57 161 CNO 3.95 2.69 100 CREDITS DC + 01 30.37 22.94 161 CNO 25.25 24.36 100 Multivariate F = 4.01; df = 3, 257; P < .01* Univariate F Tests Variable Univariate F df P GPA 11.34 * 1, 259 .01 * SEMESTERS 1.17 1, 259 .28 CREDITS 2.93 1, 259 .09 * p < .05 As a result, the third hypothesis--that there is no significant difference in achievement (in terms of mean grade in Freshman Composition 1 and cumulative grade point average) between students who needed but did not enroll in a developmental writing course and students who

PAGE 79

69 needed and completed a developmental writing course--was partially rejected because of the analysis related to GPA. The developmental completers did achieve a higher cumulative GPA than the control group of No-takes. In addition, although the developmental groups started with significantly lower writing skills as measured by the lEA, development al completers did earn about the same grade in Freshman Composition I as the No-takes. The fourth hypothesis was that there was no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earn ing a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. In terms of persistence, there were no significant differences. Not all students who completed a developmental course enrolled in Freshman Composition I. Table 12 shows that of the original 100 in each developmental group, 86 and 75 respectively completed the classroom and individualized courses. Of these, 65 classroom course completers and 57 individualized course completers totalled 122 developmenta 1 comp 1 eters who enro 11 ed in Freshman Compos it i on 1. Of th is total, 93 (76.2%) finished Freshman Composition I with a 0, C, B, or A grade. On the other hand, of the 100 control group of No-takes, 78 enrolled in Freshman Composition I and 59 (75.6% of those enrolled) finished it that semester. The difference was not significant at the p < .05 level according to chi-square analysis which resulted in a value of .00 after the Yates Correction.

PAGE 80

70 TABLE 12 A of the Rate of of Freshman it i on I for Combined Versus Control of No-takes On gi na 1 Completed Enrolled in Completed Number Fresh. I Fresh. I % DC 100 86 65 49 75.4% 01 100 75 57 44 77 .2% DC + 01 200 161 122 93 76.2% CNO 100 78 59 76.6% Chi-square after Yates Correction = .00; df = 1; p < 1.00 As noted in Table 11, no significant difference existed between the mean number of semestp.rs enrolled and the credits earned in 11 semesters between the comb; ned deve 1 opmenta 1 groups and the control No-takes. Table 11 lists the mean number of semesters enrolled for the developmental completers as 4.31 while the value for the No-takes was 3.95. Likewise, the mean number of credits earned was 30.37 for the developmental completers while the value for the No-takes was 25.25. While these numbers are slightly higher for the developmental completers than for the No-takes, the differences are not significant at the p < .05 level. In numbers of students earning an associate degree, no signifi-cant difference existed either between the combined developmental groups and the control No-takes. Table 13 shows that of the 161 combined developmental completers, 26 (16.1%) earned an associate degree while of the 100 control group of No-takes, 17 (17%) earned a degree within the 11 semesters of this study. With a chi-square value after the Yates Correction of . 00, there was no significant difference between these rates at the .05 level.

PAGE 81

71 TABLE 13 A Comparison of Associate Degrees Earned by Developmental Completers Versus Control Group of No-takes Group DC + 01 CNO Original Number 200 100 Developmental Completers 161 Earned Degree 26 17 Chi-square after Yates Correction = .00; df = 1; p < .99 % 16.1% 17.0 % Therefore, the fourth hypothesis--that there was no significant difference in persistence between students who needed but did not enro 11 ina deve 1 opmenta 1 writ i ng course and students who needed and comp leted a deve 1 opmenta 1 writ ing course--was not rejected . Although the developmental group started with lower mean entry test scores than the No-takes, the developmental completers equalled the No-takes in terms of both short and long term persistence. Therefore, one significant difference did occur in the measures of achievement (cumulative GPA for developmental completers was higher) but none in the measures of pers i s tence, between the performance of the combi ned deve 1 opmenta 1 groups (DC + 01) and the control group of No-takes (CNO), students whose scores on the lEA indicated that they needed a developmental writing course, but they did not enroll in one. Research Question #3: Developmental Groups Combined Versus High Scorers The third research question was concerned with whether or not the developmental writing courses helped students with weak writing skills

PAGE 82

72 at entry catch up, in terms of achievement and persistence, with stu-dents who did not need a developmental writing course when they started college. The fifth hypothesis stated that there was no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. On these ach i evement meas ures, the high scorers (CH I) cont i nued scoring high (see Table 14). Their mean grade in Freshman Composition I and cumulative GPA were both higher than for the developmental com-p 1 eters. The deve 1 opmenta 1 comp 1 eters' mean Freshman Compos it i on I grade was 2.61 whereas the high scorers' mean was 3.28, a significant difference when analysis of variance yielded an F value of 24.12. TABLE 14 A Comparison of the Mean Grades in Freshman Composition I Earned by Developmental Completers Versus High Scorers Standard Group Mean Deviation N DC + 01 2.61 .96 93 CHI 3.28 .84 80 Source df SS MS F Between Groups 1 19.83 19.83 24.12* Within Groups 171 143.06 .82 * P < .05 The mean GPA of 2.66 earned by the high scorers also was significantly higher than the GPA of 2.39 for the developmental completers

PAGE 83

73 (see Table 15). When the long term measures of GPA, number of semes-ters enrolled, and number of credits earned were analyzed with multi-variate analysis of variance, the multivariate F of 3.75 was significant at the p < .05 level, indicating that one or more of the pair of means were significantly different. When univariate F tests were applied, the value for cumulative GPA of 6.02 was significant at the .05 level. TABLE 15 A Comparison of Cumulative GPA, Semesters Enrolled, and Credits Earned of the Combined Developmental Groups Versus the Control Group of High Scorers Standard Variable Group Mean Deviation N GPA DC + 01 2.39 .63 161 CHI 2.66 1.11 100 SEMESTERS DC + 01 4.31 2.57 161 CHI 4.48 2.59 100 CREDITS DC + 01 30.37 22.94 161 CHI 36.43 27.55 100 Multivariate F = .01; df = 3, 257; P < .01* Univariate F Tests Var iable Univariate F df P GPA 6.02 * 1, 259 .01 * SEMESTERS .27 1, 259 .61 CREDITS 3.68 1, 259 .06 * P < .05

PAGE 84

74 Consequently, the fifth hypothesis--that there was n o significant difference in achievement between developmental completers and stu dents who neither needed nor enrolled in a developmental writing cour se--was rejected. Developmental writing course completers, as a group, earned a significantly lower grade in Freshman Composit io n I and a significantly lower cumulative GPA than students judged a s not needing a developmental writing course at entry. Completing a developmental course did not help them "catch up" to studen t s who started college with significantly higher abilities in English as measured by the lEA. The sixth hypothesis stated that there was no sign ificant differ ence in pers i stence (i n terms of comp 1 et i ng Freshman Compos it i o n I , enrolling a mean number of semesters in nearly four years, ea rni n g a mean number of credits, and earning an associate degree ) between stu dents who needed and completed a developmental w riting cours e and students who neither needed nor enrolled in a develop m ental writing course. I n terms of pers i stence, three of the fou r measures s howed no significant differences between the developmental com p leters and the high scorers. First, the rate at which each type of stu dent compl eted Freshman Composition I was statistically similar. Table 16 shows that of the 122 developmental completers who enr olled i n Fresh man C o mposition I, 93 (76.2%) completed with a grade of D, C, B, or A while 80 (80.0%) of the high scorers did. This yielded a chi-square valu e o f .26 after the Yates Correction, which was not significant at the p < .05 level.

PAGE 85

75 TABLE 16 A Comparison of the Completion Rate of Freshman Composition I for Developmental Completers Versus the Control Group of High Scorers Group Original Completed Number Developmental Enrolled in Fresh. Compo I Completed Fresh. Compo I % DC + 01 CHI 200 100 161 122 100 93 80 76.2% 80.0% Chi-square after Yates Correction = .26; df = 1; P < .61 Likewise, as seen above in the reporting of MANOVA applied to long term measures, found in Table 15, the mean number of semesters enrolled by the two groups showed no significant difference at the .05 level. The developmental completers enrolled for an average of 4.31 semesters while the high scorers enrolled for an average of 4.48 semesters. Simil arly, the number of credits earned by each group, at 30.37 for the developmental completers and 36.43 for the high scorers, was not significantly different either. However, significantly more high scorers received an associate degree within the 11 semesters of this study. Table 17 illustrates that of the 161 developmental completers, 26 (16.1%) earned a degree while of the 100 high scorers, 32 (32.0%) earned one. The chi-square value of 8.07 after the Yates Correction was a significant difference. TABLE 17 A Comparison of the Associate Degrees Earned Developmental Com-pleters Versus Control Group of High Scorers Original Developmental Earned Group Number Completers Degree % DC + 01 200 161 26 16.1% CHI 100 32 32.0% Chi-square after Yates Correction 8.07; df -1; P < .05*

PAGE 86

76 As a result, the sixth hypothesis--that there was no significant difference in persistence between students who needed and ' completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course--was rejected only in terms of the number of degrees earned. Deve 1 opmenta 1 comp 1 eters recei ved fewer associate degrees than students who had stronger writing skills at entry. However, developmental completers, even though they scored significantly lower on the entry Engl ish test, appear to have completed Freshman Composition I at about the same rate, to have stayed enrolled at the college about the same number of semesters, and to have earned a similar number of credits as students who scored significantly higher on the Initial English Assessment Test. In summary, significant differences occurred in both measures of achievement but in only one of the four measures of persistence when the combined developmental groups (DC + Dr) were compared with the group of students who scored above the cutoff score, on the Initial Engl ish Assessment Test. These hi gh scori ng students were judged as not needing a developmental writing course and enrolled directly into Freshman Composition I. That h igh scorers had a higher mean in the achievement measures and in half of the persistence measures is not surpri sing gi ven thei r hi gh performance on the entry Engl ish test. However, on three of the persistence measures, there was no significant d iff erence between the deve 1 opmenta 1 comp 1 eters , who were low scorers, and the high scorers.

PAGE 87

77 Research Question #4: Younger Versus Older Developmental Students The fourth research question was concerned with whether or not, after completing a developmental writing course, older adults achieved and persisted any differently from recent high school graduates. The seventh hypothesis stated that there was no significant difference (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative grade point average) between stu-dents who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. Table 18 lists the results of the multiple analysis of variance used to control for interaction among dependent variables of grade in developmental course, GPA, number of semesters enrolled, and number of credits earned in the 11 semesters of the study. The results of the MANOVA analysis yielded a multivariate F of 3.22, which indicated that a significant difference at the .05 level was at work across the vari-ables. To pinpoint where the difference lay, univariate F tests were applied. As seen in Table 18, younger and older students earned about the same mean grade in a developmental course since the univariate F value of 2.50 means there was no significant difference at the .05 level between the younger students I mean of 2.89 and the older students I mean of 3.19. The cumulative grade point average for the two groups, however, was significantly different. Cumulative GPA for younger students averaged 2.33, while the cumulative GPA for the older students was significantly higher at 2.70, yielding a significant F value of 7.47.

PAGE 88

78 TABLE 18 A of Mean Grade in Course! GPAz Semesters Enro 11 ed and Credits Earned Younger and Older Course Standard Variable Mean Deviation N GRADE IN 17-21 2.89 .92 135 DEVELOPMENTAL COURSE 22+ 3.19 .95 26 GPA 17-21 2.33 .62 135 22+ 2.70 .64 26 SEMESTERS 17-21 4.41 2.60 135 22+ 3.77 2.40 26 CREDITS 17-21 31.85 23.42 135 22+ 22.69 18.80 26 Multivariate F = 3.22; df = 5, 155; p < .01* Univariate F Tests Variable Univariate F df GRADE IN DEVELOPMENTAL 2.50 1, 159 .11 COURSE GPA 7.47 * 1, 159 .01 * SEMESTERS 1. 37 1, 159 .24 CREDITS 3.53 1, 159 .06 * p < .05 As seen in Table 19, mean grade in Freshman Composition I was not significantly different between the younger and older groups of devel opmental course completers. The mean grade of 2.56 for 17-21 year olds and 2.92 for the age 22 and older group yielded a non-signif i cant F value of 1.60 at the .05 level when analysis of variance was applied.

PAGE 89

79 TABLE 19 A Comparison of the Mean Grade in Freshman Composition I of Younger and Older Developmental Course Completers Standard Group Mean Deviation N 17-21 2.56 .94 80 22+ 2.92 1.04 13 Source df SS MS F Between Groups 1 1.47 1.47 1.60 Within Groups 91 85.12 .91 p < .20 Therefore, the seventh hypothes i s--that there was no sign ifi cant difference between achi evement of students age 22 and older and students age 17-21 in development al courses--was rejected only in terms of cumulative GPA. Older students did have a significantly higher cumulative GPA than recent high school graduates. However, on the other two ach i evement measures of grade in deve 1 opmenta 1 course and grade of developmental completers in Freshman Composition I, the hypothesis failed to be rejected. Older and younger students seemed to ach i eve about the same in these courses on two of the three mea-sures of achievement. The eighth and last hypothesis stated that there was no signifi-cant difference in persistence (in terms of completing the developmental course, completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earn ing an associate degree) between students who

PAGE 90

80 enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. No significant difference occurred between the number of younger and older students completing their developmental writing courses, as seen in Table 20. Of the total of 200 students enrolled in either the classroom or the individualized course, 168 were age 17-21 and 32 were age 22 or older. Of the younger students, 135 or 80.4% completed their course; of the older students, 26 or 81.3% completed their course. That resulted in a non-significant chi-square value of .00 after the Yates Correction. TABLE 20 A Comparison of the Rate of Completion of a Developmental Course by Younger and Older Students Group 17-21 22+ Original Number 168 32 Completed Developmental 135 26 Chi-sq uare after Yates Correction = .00; df = 1; P < 1.00 % 80.4% 81.3% Similarly, the number of developmental completers subsequently completing Freshman Composition I showed no significant difference by age (see Table 21). Of the 135 developmental completers in the 17-21 age group, 107 enrolled in Freshman Composition I and 80, or 74.8% of those enrolled, completed with a grade of D, C, B, or A. Of the 26 older developmental completers, 15 enrolled in Freshman Composition I and 13, or 86.7%, completed it. The chi-square value of .48 after the Yates Correction was not significant.

PAGE 91

81 TABLE 21 A Comparison of the Rate of Completion of Freshman Composition I by Younger and Older Developmental Course Completers Group 17-21 22+ Completed Developmental 135 26 Enrolled in Fresh. Compo I 107 15 Completed Fresh. Compo I % 80 74.8% 13 86.7% Chi-square after Yates Correction = .48; df = 1; p < .49 The MANOVA analysis of possible interacting short and long-term measures cited in Table 18 included data on long-term persistence mea-sures of semesters and credits completed. There was no significant difference in number of semesters enrolled between the younger and older groups. The 17-21 group had a mean number of semesters enrolled of 4.41 while the 22 and older students were enrolled a mean of 3.77 semesters. The univariate F value of 1.37 was not significant at the p < .05 1 eve 1 . The number of cred i ts ea rne d by the two groups also showed no significant difference, as also seen on Table 18. The 17-21 group had a mean cumulative credits earned of 31.85 while the 22 and older students had earned 22.69. The un i v ar i ate F of 3.53 was not significant. As a result, it appeared that the recent high school graduates persisted better than students who had been out of high school four or more years. Finally, the number of students earning an associate degree from each age group was not significantly different. Table 22 shows that of the 135 younger developmental completers, 24, or 17.8%, earned an assoc i ate degree. Of the 26 older deve 1 opmenta 1 comp 1 eters, 2, or

PAGE 92

82 7.7%, earned an associate degree. While this may appear to be a sig-nificant difference, chi-square with the Yates Correction yielded a non-significant value of .98. TABLE 22 A of the Associate Degrees Earned Younger and Older Developmental Earned Group Completers Degree % 17-2 1 l35 24 17.8% 22+ 26 2 7.7% Chi-square after Yates Correction = .98; df = 1; p < .32 Consequent ly, the ei ghth and 1 ast hypothes i s--that there was no significant difference in persistence between developmental students aged 22 and older and more recent high school graduates--was not rejected. Older students persisted about the same as younger students in completing their developmental course, in completing Freshman Composition I, in staying enrolled by semesters, in earning c redits, and in earning an associate degree about as well as did younger stu-dents. In conclusion, only one significant difference, the long-range ach i evement measure of cumu 1 at i ve GPA, occurred between the younger devel opmenta 1 students age 17-21 and the older devel opmenta 1 students age 22 or more. Older students did earn a higher cumulative GPA than younger students. On all the short-range measures and on the other long-range measures, there were no significant differences.

PAGE 93

83 Summary In summary, Table 23 provides a quick recap of the findings detailed in this chapter. There were no significant differences between the achievement and persistence measures of students in the classroom developmental and the individualized developmental groups. Only one significant difference occurred when the comb i ned deve 1 opmenta 1 groups were compared with the students who should have taken a developmental course, according to the cutoff scores on the Initial English Assessment, but did not. Students who completed a developmental course did end up with a higher mean GPA than did the No-takes. The other ach i evement and persistence measures showed no significant difference, even though the No-takes had a slight, but significantly higher lEA mean score. When the deve 1 opmenta 1 comp 1 eters were compared to the contro 1 group of students who had scored high on the Initial English Assessment and were therefore judged as not needing a developmental writing course, the high scorers did have significantly higher results on both achievement measures, but not on three of the four pers i stence measures. In terms of the rate of completing Freshman Composition I, the mean number of semesters enroll ed in 11 semesters, and in the number of credits earned, the developmental completers showed no significant difference from the students judged stronger in writing skills at entry. Finally, there were no significant differences on all but one of the achievement and persistence measures between recent high school graduates aged 17-21 and older students aged 22 or more. The exception came in the long-term achievement measure of cumulative GPA. The older students had a significantly higher cumulative GPA.

PAGE 94

84 TABLE 23 Summary of Findings Group Classroom vs. Individualized Developmental Course Combined Developmental Completers vs. No-takes Combined Developmental Completers vs. High Scorers 17-21 Year Olds vs. 22 and Older Developmental Students Measure Grade in Developmental Course Grade in Freshman Composition I Cumulative GPA Completing Developmental Course Completing Freshman Composition Semesters Enrolled Credits Earned Associate Degrees Earned Grade in Freshman Composition I Cumulative GPA Completing Freshman Compos it ion Semesters Enrolled Cred its Earned Associate Degrees Earned Grade in Freshman Composition I Cumulative GPA Completing Freshman Composition Semesters Enrolled Credits Earned Associate Degrees Earned Grade in Developmental Course Grade in Freshman Composition I Cumulative GPA Completing Developmental Course Completing Freshman Composition Semesters Enrolled Credits Earned Associate Degrees Earned Significanc e not significant not significant not significant not significant I not significant not significant not significant not significant not significant + * I not significant + not significant + not sign ificant + not significant + ** ** I not significant ++ not significant ++ not significant ++ ** not significant not significant *** not significant I not significant not significant not significant not significant * Developmental completers' means were significantly h i gher at p < .05. ** High scorers' means were significantly higher at p < .05. *** Older students' means were significantly higher at p < .05. + Developmental completers' means were similar at p < .05 to the No takes' means although No-takes had higher entry skills. ++ Developmental completers' means were similar at p < .05 to the h igh scorers' means.

PAGE 95

85 Table 24 summarizes the relationship of the findings to the hypotheses. TABLE 24 Summary of the Relationship of Findings to Hypotheses Hypotheses There is no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative GPA) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental course and completed the classroom course. There is no significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed a developmental writing course and enrolled in the individu alized course and students who needed a developmental writing course and enrolled in the classroom course. There is no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative GPA) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. There is no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. Status Was not rejected. Was not rejected. Rejected for cumulative GPA. Was not rejected for grade in Fresh . Compo I. Was not rejected.

PAGE 96

86 Table 24--Continued Hypotheses There is no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative GPA) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. There is no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. There is no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and cumulative GPA) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. There is no significant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a develop mental writing course at age 17 to 21. Status Rejected. Rejected for associate degree. Was not rejected for completing Fresh. Compo I, semesters, and credits. Rejected for cumulative GPA. Was not rejected for grades in developmental course and Fresh. Comp. 1. Was not rejected. Four hypotheses were not rejected because there were no signifi-cant differences on all variables between the groups being measured. Three hypotheses were rejected only for one applicable measure, but were not rejected on other measures. Only one hypothesis was rejected on all measures.

PAGE 97

CHAPTER 5 SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The problem in this study was to evaluate a community colleg e developmental writing program that included two courses with different instructional modes, one a classroom presentation and the other an i n-dividualized format, so that the college might assess the efficacy of each course and of the ent ire program in terms of student pers is ten c e and achievement. Thre " e major questions were asked about the effec t o f the developmental writing courses. The fo urth question a sked if age was a factor in developmental students' performance. The first research question asked whether an individualized devel-opmental writing course or a classroom based developmental writing course contributed to better achievement and longer persistence by stu-dents who needed a developmental writing course when they started c o l -lege. The first accompanying null hypothesis stated that there was no significant difference in achievement (in terms of mean grade in the developmental course, mean grade in Freshman Composition I, and c u mula-tive grade point average) between students who needed a develop mental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course. The second null hypothesis stated that there was no s ignifi-cant difference in persistence (in terms of completing the developmental course, subsequently completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of 87

PAGE 98

88 credits, and earning an associate degree) between students who needed a developmental writing course and completed the individualized course and students who needed a developmental writing course and completed the classroom course. The second research question was concerned with whether or not students who needed and completed either of the developmental writing courses achieved better and persisted in coll ege longer than students who needed but did not enroll in a developmental writing course. The third null hypothesis accompanied this research question and stated that there was no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who needed but did not enroll in a developmental writing course and stu dents who needed and completed a developmental writi ng course. The fourth null hypothesis stated that there was no significant difference in persistence (in terms of completing Freshman Composition I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate de gree) between students who needed but did not enroll in a developmental writing course and students who needed and completed a developmental writing course. The third research question was concerned with whether or not the developmental writi ng courses hel ped students with weak writi ng skill s at entry catch up, in terms of achievement and persistence, with students who did not need a developmental writing course when they started college. The fifth null hypothesis stated that there was no significant difference in achievement (in terms of mean grade in Freshman Composition I and cumulative grade point average) between students who

PAGE 99

89 needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. The sixth null hypothesis stated that there was no significant difference in persistence (in terms of subsequently completing Freshman Composi tion I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) be tween students' who needed and completed a developmental writing course and students who neither needed nor enrolled in a developmental writing course. The fourth research question was concerned with whether or not, after completing a developmental writing course, older adults achieved and persis ted any differentl y from recent hi gh school graduates. The seventh null hypothesis stated that there was no significant difference in achi evement (i n terms of mean grade in the devel opmenta 1 course, mean grade in Freshman Composition I, and cumulative grade point average) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. The eighth null hypothesis stated that there was no significant difference in persistence (in terms of completing the devel opmenta 1 course, subsequentl y compl eti ng Freshman Composit i on I, enrolling a mean number of semesters in nearly four years, earning a mean number of credits, and earning an associate degree) between students who enrolled in a developmental writing course at age 22 or older and students who enrolled in a developmental writing course at age 17 to 21. The 1,431 students who entered Valencia Community College in the 1980 fall semester, who were first-time-in-college students, who completed the Initial English Assessment (lEA), and who enrolled in West

PAGE 100

90 Campus courses were divided into four groups: those scoring below the lEA's cutoff and enrolling in a classroom-based developmental writing course, those scoring below the cutoff and enrolling in an individualized developmental writing course, those scoring below the cutoff but not enroll ing in either developmental course, and those scoring above the cutoff and enrolling in regular freshman English. Random samples of 100 students were drawn from each of these four groups. Achievement was measured as grade in developmental course, grade in regular freshman composition, and cumulative GPA. Persistence was measured as completion of a developmental English course, completion of a regular freshman composition course, number of semesters enrolled, number of credits earned, and attainment of an associate degree. Achievement and persistence data came from student transcripts. lEA scores came from computer archives. Chi-square followed by the Yates Correction, analysis of variance followed by the Scheffe test or an F test, and multivariate analysis of variance followed by the multivariate F test and univariate F tests were applied as appropriate. Conclusions As is typical for community college developmental programs (Kuli k et al., 1983), these findings produced primarily no significant differences. However, once achievement and twice persistence measures seemed to indicate positive impact from developmental writing course work. In answer to the first research question, the individualized de velopmental writing course compared favorably with the classroom developmental course on all measures. At entry, the individualized

PAGE 101

91 students' mean score on the lEA had been sl ightly lower, although not significantly so, than the classroom students' mean. More classroom students finished their developmental writing course than individualized students did, but the individualized students achieved a higher mean grade. In neither case was the difference significant. This is consistent with lack of significance in completion found by all three individualized versus classroom course studies discussed earlier (Brown, 1984; Epes et al., 1980; Sutton & Arnold, 1974). It is worth noting, however, that the completion rate out of both of Valencia Community College's developmental writing courses was higher at 80.5% than the developmental/remedial English course completion rates measured in nearly all other studies cited in this investigation. Sutton and Arnold (1974) reported drop-out rates in the 40% range, leaving persistence in the 60% range for both individualized and classroom courses. Epes et al. (1980) reported passing rates of individualized students at 64.8% and of classroom based students at 57.6%. Lunsford (1978) reported 65%, Bers (1982!, 1982E) 53% and 55% in successive years, Braxton et al. (1980) at 62%, and Barton (1984) at 68%. The New Jersey Basic Skills Council (1983) cited a statewide average of 65% with five community colleges having a higher passing rate than Valencia (the top being 82%), but twenty-four having a lower rate (the low being 46%). Finally, Smittle (1982), at another Florida community college, reported 76% of the developmental students passing their de velopmental writing course. The Valencia rate of 86% for the classroom course and 75% for the individualized course ranked favorably against these other studi es. More students were compl eti ng developmental writing course work and having the opportunity to continue into the regular

PAGE 102

92 freshman Engl ish curri cul urn wi th the background these courses afforded than at most other institutions cited in these studies. Developmental completers also finished regular freshman composition at about the same rate: 75.4% of former classroom students and 77.2% of former individualized students. The grades these students received showed no significant difference, too, with former classroom course students earning a mean. of 2.51 and former individualized students earning a mean of 2.73. This grade result falls between the other two studies measuring grade in freshman English of former developmental English students. Sutton and Arnold (1974) found individualized completers had higher freshman English grades, but Brown (1984) found classroom course completers had higher grades. No significant differences in cumulative GPAs between classroom and individualized developmental course completers was a finding that was at odds with both the Sutton and Arnold (1974) and the Brown (1984) studies. Valencia's classroom developmental completers averaged a cumulative GPA of 2.46 while the individualized developmental completers averaged a 2.54 GPA; no significant difference was found here. But Sutton and Arnold (1974) found higher cumulative GPAs for individualized completers, while Brown (1984) found higher GPAs for former class-room developmental students. Neither of these two earl ier studies found significant differences in persistence. As measured by number of semesters enrolled, credits earned, or degrees received, this Valencia study revealed no significant difference in persistence either. In answer to the second research question, there was one measure on which the developmental completers significantly outperformed students who also scored below the entry English test cutoff but who did

PAGE 103

93 not enroll in a developmental writing course (No-takes). On the rest of the achievement and on all persistence measures, the differences were not significant. However, since the No-takes had a small, but significantly higher mean entry English test score, this nonsignificance is encouraging. Both developmental completers and the No-takes achieved mean grades in freshman composition and completed this course at about the same rate. The developmental completers earned a mean grade of 2.61 in Freshman Composition I with 76.2% of them completing the course while the No-takes earned a grade of 2.63 with 75.6% of them completing. This is comparable to three of six studies cited earlier in which there were no si gni fi cant differences in freshman Engl ish grades. I n the other three studies, however, the developmental course completers achieved a higher mean grade. In these earlier studies, however, the developmental completers and the No-takes had about the same entry level ability, whereas in the present study, the developmental completers had somewhat lower writing skills at entry, as measured by the lEA, yet were able to achieve about the same grade in Freshman Composi tion I as the No-takes did. Two other studies measuring completion rate of developmental students in freshman English found that more developmental course completers finished freshman composition than No-takes. This Valencia study did not find similar results. One of the major reasons for open admission colleges to add developmental or remedial writing courses to their curriculum is to enhance weak students' skills so that they can become more successful in the regular freshman English sequence. At Valencia, the developmental completers finished Freshman Composition I

PAGE 104

94 at the same rate as the No-takes. Again, however, this is a somewhat encouraging result. Valencia's No-takes were not equal in ability to the developmental groups at entry. Although the No-takes did score below the lEA cutoff score, their mean score was still significantly higher than the mean lEA score for students who needed and enrolled in the developmental courses. Therefore, although the developmental course students were at a disadvantage in terms of their entry level writing skills, they were able to complete Freshman Composition I, after completing a developmental writing course, at the same rate as the somewhat stronger No-takes. There is certainly precedent in the literature for more discourag ing results than these in the Valencia study. The meta-analysis of Tinto and Sherman (1974) concluded the developmental courses usually did not result in improved freshman English completion rates over control groups with similar aptitudes. Losak and Burns (1971) concluded the same. Sharon (1972) concluded that "the effectiveness of remedial courses in college should not be taken for granted" (p. 62). Finally, even though Baker (1982) found that former remedial English students were more likely to complete freshman English than the control group, he concluded that on no other measures were the remedial students superior and that, therefore, the effectiveness of Snow College's remedial English course was doubtful. Kulik et al., (1983) wondered if the fault lay in the developmental courses themselves, that they were not challenging enough to prepare students for freshman Engl i sh and that corrvnunity coll ege educators woul d do well to upgrade these courses so that developmental completers would have a better chance to outperform No-takes by achieving better and persisting longer

PAGE 105

95 in college. Although at Valencia, the developmental completers d i d not outperform the No-takes in completing Freshman Composition I, the developmental course completers did equal the completion rate of the somewhat stronger No-takes. There was also no signi ficant difference in the long-term persistence measures of semesters enrolled, credits earned, and degree attained between developmental completers and No-takes. This is consistent with one study, not as good as four studi es whi ch reported that developmental/remedial completers did better than No-takes on one or two of these measures, and better than two studies where the decimation of the ranks of developmental students while they were in the develop mental course led to sharp reduction in retention rates while the Notakes were staying more semesters and/or earning more credits. Since Valencia's developmental students completed their writing courses at a higher rate than at most other institutions noted earlier, develop mental completers did not suffer this decimation effect and went on to reenroll and earn credits at least as favorably as the No-takes, even though the No-takes started with a slight advantge in writing skills. GPA is where developmental completers outstripped the control group of No-takes by 2.34 to 2.00, a statistically significant difference. Dudl ey (1978), Mowatt (1978), and the New Jersey Bas i c Ski 11 s Council (1983) found the same positive results, although Linthicum (1979), Losak and Burns (1971), and Lovejoy (1974) found no Significant differences. What this finding may suggest is a long-ter m positive effect on the grades of developmental completers as they take other courses at the college. Developmental writing course completers seemed to earn better grades than the sl ightly stronger No-takes in other

PAGE 106

96 college courses across the curriculum during the semesters they stayed enrolled. This long-term achievement of earning a significantly highe r cumulative GPA than the No-takes is an encouraging finding. In answer to the third research question, developmental completers "caught up" to the stronger students, as measured by the lEA, on three of the six achievement and persistence measures. Although their English scores were significantly lower at entry testing, developmental completers persisted in completing freshman composition, in staying enrolled at the college, and in earning credits at about the same rate as the hi gh scori ng group. On the other measures, developmental completers were lower. Overall, these findings were consistent with most of the studies cited in Chapter 2. A semester-long course in developmental writing was not enough to raise completers' grades in freshman composition to the level of the stronger students. Developmental compl eters received a mean grade of 2.61, whereas the students scoring over the cutoff on the entry English test received a mean grade of 3.28. The other studies reported mixed results on this measure with two drawing the same conclusion (Barton, 1984, Linthicum, 1979), two finding no significant difference (Braxton et al., 1980 and New Jersey, 1983) and one finding that developmental completers actually outdid the stronger students in freshman composition courses (Dudley, 1978). That the same proportion of developmental completers finished freshman composition successfully as did stronger students was consistent with the two other previous studies measuring completion rates (Reap & Covington, 1980 and Smittle, 1982). One might be tempted to consider that the developmental English course had successfully helped

PAGE 107

97 developmental completers prepare for success in freshman composition, at least to the extent that just as many of the former developmental students as the stronger students were abl e to conti nue to the end of the course and pass it. However, this encouraging finding must be tem pered with the results on grades earned in Freshman Composition I. Al though developmental completers persisted to the end and passed at about the same rate, it is not encouraging that they earned a significantly lower mean grade than the high scorers did. In addition, the No-takes, with a somewhat stronger wri ti ng abil i ty at entry than the devel opmenta 1 students but without the background of the deve 1 opmenta 1 course, completed freshman composition at the same rate as the developmental completers, and, therefore, at about the same rate as the stronger students, also. There is a distinct possibility that factors other than English preparation may be at work equally on the developmental completers, No-takes, and high scorers, affecting the completion rate in freshman composition. On two of the four long-term measures of achievement and persistence, the stronger English students outperformed the developmental completers. Consistent with four out of five studies cited in Chapter 2, Va1encia 's developmental completers attained a significantly lower cumulative GPA than the stronger students, 2.39 to 2.66 respectively. Barton (1984), Braxton et al. (1980), Linthicum (1979), and the New Jersey Basic Skills Council (1983) all reported similar negative results. Only Dudley (1978) reported no significant difference on cumulative GPA between developmental/remedial completers and the initially higher scoring students. The encouraging point to remember, though, is that the developmental completers did gain ground over the No-takes on

PAGE 108

98 this measure, even though they did not attain as high a GPA as the stronger students. In terms of long-term persistence, developmental completers enrolled in college about as long as the stronger students did. The mean numbers of semesters enrolled were 4.31 to 4.48 respectively. This is consistent with the three studies marking this measure (Barton, 1984; Dudley, 1978; Linthicum, 1979). Again, the performance of the No-takes must also be considered in interpreting this finding. The No-takes, with a slight but significant advantage over the developmental students in terms of entry level writing skills, did enroll semester after semester at about the same rate as the developmental completers. One might consider that the developmental completer s "caught up" with both the No-takes and the high scorers on this measure. However, as with the completion of freshman English rates discussed earlier, there is a distinct possibility that factors other than English preparation may be at work equally on the developmental completers, No-takes, and high scorers, affecting the number of semesters of enrollment. In addition, on the long-term persistence measure of credits earned, the developmental completers again held about equal to the hi gher scori ng students. 30.37 credits earned was The former developmental students' mean of not significantly different from the high scorers' mean of 36.43. This is not consistent with the only other study measuring credits earned. The New Jersey Basic Skills Council (1983) found developmental/remedial English completers earned significantly fewer credits than stronger students. For a thi rd time, the performance of the No-takes affects interpretation of the present study's results. The No-takes' mean credits earned was not

PAGE 109

99 significantly different from the developmental completers' mean credits. However, the No-takes started with a slight. advantage over the developmental students, so perhaps the developmental courses did hel p the completers to "catch up" to the No-takes and to the high scorers in terms of number of credits earned. Again, however, factors other than English preparation might be at work equally over all three groups. Finally, the proportion of degrees earned was significantly higher for the stronger students as compared to the developmental course completers. This finding suggests that developmental writing course completers were less successful in completing the requirements for a degree. In fact, the stronger students received an associate degree at about twice the rate as the developmental completers (32% to 16%). This finding is consistent with the only other study found using this persistence measure . Braxton et al. (1980) also found that develop mental completers earned significantly fewer degrees than students with strong writing skills at entry. In answer to the last research question, the older developmental students, those aged 22 or older, achieved a higher cumulative GPA than the younger developmental students, those more recent high school graduates aged 17 to 21. All other achi evement and pers i stence measures showed no Significant differences between these two groups, although most short term measures showed better resul ts for the other students whil e more long-term measures showed better results for the younger ones. The question about older versus younger developmental student achievement and persistence was posed for two reasons: first, because community college non-traditional students vary greatly in age, and age

PAGE 110

100 is sometimes cited as significantly affecting performance in college and secondly, because the 1983 Florida legislature passed a statute limiting developmental and remedial programs in community colleges after 1990 to older students who were not recent high school graduates (Section 240.134 (3) (a), Florida Statutes, 1983). The finding that older students earned a Significantly higher cumulative GPA is consistent with the findings of both Linthicum (1979) and Whittle (1980). The older Valencia Community College students' mean cumulative GPA was 2.70; the younger students' was 2.33. Whether or not the developmental course work contributed to this effect could not be determined since there was no significant difference in either grade in or proportion of students completing a developmental writing course between the older and younger groups. This finding agrees with Whittle (1980) that no significant difference in developmental English grade occurs between the age groups, but di sagrees wi th Dumont and Jones (1983) who found that older students attained a significantly higher developmental grade. In this study, Valencia's older students did achieve a better developmental grade (3.19 to 2.89 for younger students) and did persist slightly better in completing the developmental course (81.3% to 80.4%); however, the differences were not statistically significant. Since no significant differences existed for grade in or completion rate of freshman composition, no effect of the developmental courses could be determined here, also. This finding contrasted with all three studies cited above, all of which found some indication that older students achieved higher grades in regular freshman Engl ish. At Valencia, the mean grades were higher (older students' 2.92 to younger

PAGE 111

101 students' 2.56) and the completion rates were better (older students 86.7% to 74.8%) but given the smaller number of older developmental completers than younger students, these differences were not statistically significant. As for long-term persistence by age, none of the research cited in this study provided data on the measures used in this Valencia analysis. The younger students did earn more credits as a group than did the older students (31.85 to 22.69), but the difference was not significant. Younger students also reenrolled in more semesters than the older students (4.41 to 3.77), but this was not a significant difference either. Younger students al so graduated at a hi gher rate than older students (17.8% to 7.7%), but high mortality and the disparity in group sizes may have contributed to the nonsignificance of these differences, also. One might conclude that students aged 22 and older may do better in their courses than more recent high school graduates, but they may also find it harder to reach graduation than recent high school graduates do. In either case, no solid indication existed that older students' success in developmental English courses contributed to their possible success in regular freshman English or to their higher cumulative GPA. No solid indication existed, either, that younger students' success in developmental English courses contributed to their higher persistence to graduation. Recommendations Recommendations based on the findings above must be tempered by recent changes in Florida postsecondary evaluation procedures

PAGE 112

102 concerning entry level and sophomore level testing. These statewide changes have made subsequent collegewide changes in curriculum and testing procedures at Valencia. First of all, further research should be conducted to ascertain if the individualized course by itself might have produced significant results when compared to a control group of No-takes. The Initial English Assessment Test mean score for the individualized developmental course students was significantly lower than that of the No-takes, and lower even than the mean for classroom developmental students. There is a possibility that students who completed the individualized course, since they started significantly weaker than the No-takes, even weaker than the classroom developmental group, might have made progress in the individualized course sufficient to bring them up to the level of the No-takes on some of the measures. If so, then the individualized course by itself might then have been producing significant results. Unfortunately, in the present study, the effect of the individualized course in relation to the No-takes can not be separated from the effect of the classroom based developmental course. A coroll ary may be that the cl assroom course students may have progressed beyond the 1 evel of the No-takes on more measures than just cumulative GPA, but that students who completed the individualized course, with a lower entry score than the No-takes, did not do as well on the measures, and the net effect was that no significant differences showed when the individualized course completers and the classroom course completers were combined and contrasted with the No-takes. Further research along these lines should help to clarify this issue.

PAGE 113

103 Considerable curriculum changes have occurred across the state of Flori da since the fall of 1980 when these students enrolled in college for the first time. The Florida legislature mandated that two programs be put into effect at the public postsecondary institutions: the College Level Academic Skills Project (CLASP) and Rule 6A-10.30, the so-called Gordon Rule (Rule 6A-10.30, 1982). This rule has changed college-level English courses by requiring that students write a minimum of 6,000 words in each of four courses in the communications or humanities areas. By and large, state postsecondary institutions have opted not to include developmental and remedial English courses under this mandate, but Freshman Composition I and II have been restructured at many Florida colleges to meet this requirement. On the other hand, the College Level Academic Skills Project has heavily impacted develop-mental, remedial, and regular college Engl ish as well as reading and mathematics programs. The CLASP involves entry level testing and testing at the sopho-more level on a specific set of communications and computations compe-tencies. Each institution has been required to use a nationally nor med entry level test from a list of approved tests judged as measuring those competencies. Community colleges individually chose which entry tests to use and what cutoff scores to set. By state law, students scoring below the cutoffs were mandated into developmental or remedi al courses in reading, writing, and mathematics. Consequently, at Valencia the Initial English Assessment, whic h had been locally writ-ten, was abandoned. In its place, faculty chose the Missouri College Engl ish Test, and new cutoffs were determi ned. Another resul tis that the pool of No-takes dri ed up. Instead of about half of the low

PAGE 114

104 scoring students electing to enroll in a developmental English course and the other half skipping it, all low scorers are now mandated to enroll in the developmental Engl ish program. If further research were to be done on Valencia English students, consideration of the change in test and the change in control group make-up would have to be made. A control group could still be formed with students who circumvent the mandate or who enrolled in a developmental course but dropped out before the end of the semester. studies have used such a control group. As a result of the second part of CLASP, the sophomore test, several curriculum changes have occurred. For example, the CLASP writing competencies already addressed in Freshman Composition I and in the de velopmental courses were identified; those not already addressed were added to these two areas. Textbooks were changed, and curriculum materials were added to ensure coverage of the CLASP competencies. The classroom developmental course, in particular, was upgraded to include a 11 the grammar, punctuat ion, sentence structure, usage, and mechani cs CLASP writing competencies in addition to paragraph writing skills. On the other hand, the individualized course kept its multi-level instructional character. Because of this, the faculty recommended that counselors enroll the lower half of the students scoring below the Missouri College English Test cutoff to be placed in the individualized course and the upper half to be placed in the classroom course. Thus, the student populations of the two courses now have an important difference in abi 1 i ty 1 evel at entry. Furthermore, the nature of the two courses has also changed. Consequently, comparing the two instructional modes

PAGE 115

105 to determine which might have been more effective, the way this present study has, would not be valid now. The major change in Freshman Compos it i on I, bes ides inc rea sed emphasis on the CLASP competencies, has entailed holistic cross-grading of the final exam essay. Faculty were trained in holistic grading techniques according to practices recommended by the Educational Test-ing Service. Instructors now do not grade their own students' final exam essays. Instead, two other instructors holistically decide what level of competency each essay exhibits and assigns it a score from 1 to 4. Each grader does not know how the other has graded the paper. A third grader evaluates papers receiving divergent scores. This proce dure is del iberately structured to match that used to grade the essay portion of the College Level Academic Skills Test, the sophomore level test which makes up part two of the CLASP project. The result has been that Valencia students with weak writing skills find it harder to pass Freshman Composition I. This also might have an effect on how well former developmental students do in their regular freshman English course. Since Freshman Composition I is now more difficult to pass, former developmental students, with a firmer grounding in some of the CLAST competencies, might do better than the No-takes who somehow circumvented the mandate that they take develop mental English or the drop-outs of the developmental English courses. It would be especially interesting to contrast how the classroom developmental completers now do in comparison to the non-completers since the classroom course focuses so extensively on a large portion of the CLASP competencies.

PAGE 116

106 One 1 ast aspect of CLASP mi ght have an impact on the results of any further studies in Florida similar to this one. As of August 1984, again by state law, all students must take and pass the College Level Academic Skills Test (CLAST) in order to receive an Associate in Arts degree and continue their education with upper division courses at a public Florida university. Since students who do not pass the CLAST may not receive an Associate in Arts degree until they do, this might also impact the proportion of developmental completers, non-completers, and high entry scoring students who receive a degree. If research were conducted on a more recent cohort of entering Valencia students, the first research question of this study would have to be abandoned si nce the two devel opmenta 1 courses at VCC no longer have enough in common to compare the results of the two instructional modes. However, the second research question concerning developmental completers versus No-takes might be worthwhile even though the character of the No-takes would have changed from developmental course avoiders to non-completers. If the completers had significantly higher achievement and persistence than the drop-outs, college policy should perhaps be changed to mandate completion of the developmental course in addition to mandating enrollment in it. In the present study, no significant differences, except in cumulative GPA, occurred between the developmental completers and the No-takes. If this finding were repeated with more recent cohorts of compl eters and non-compl eters, then serious questions about the effectiveness of the upgraded classroom developmental course would have to follow. Si nce thi sis a mandated course, educators might feel a moral imperative to overhaul the course unt il the results were more pos it i ve.

PAGE 117

107 In fact, if the state continues to mandate developmental courses, it probably has a responsibility to evaluate them. At twenty-eight Florida community and junior colleges, students scoring below a variety of cutoffs on a variety of entry level tests are being required to enroll in a variety of developmental reading, writing, and mathematics courses. It would be naive to assume that these courses all have the desired result of increasing students' abilities to cope with regular freshman courses and to become proficient in the CLASP skill s. Those programs which are successful need to be identified and then emulated by programs which are not. Without a fairly coherent and widely used evaluation process, the relative merits of programs within this multiplicity can not be determined. The emphasis on short-term and long-term achievement and persistence measures used in the present study comprises one research structure which might be applied on a broader basis. The New Jersey Basic Skills Council's study (1983) offers a similar model following reading, writing, and mathematics basic skills courses with measures of achievement and persistence in subsequent courses in English, mathematics, social studies, humanities, natural and physical science courses. Although the current New Jersey report does not contain long term measures, the Council plans to require all state institutions of higher education to extend their follow-up studies to include second and third year results. Whatever model is used, some statewide research ought to be conducted. As Clowes (1984) pointed out in his stage model of evaluation for remedial and developmental programs, the output of these programs becomes the input for the mainstream curriculum, and, therefore:

PAGE 118

108 the proof of the quality of a remedial program exists not in the abi 1 i ty of students to survi ve wi thi n that program but rather in the ability of students to complete the remedial program and make a successful transition into the mainstream curricula of the institution. (p. 15) Within Florida, the model would need to take two steps beyond: student mastery of the CLASP skills and successful passing of the CLAST test. As for the last research question concerning performance of older versus younger students, this study shows that on most measures, there is no significant difference in achievement and persistence, other than the fact that older students earn an appreciably higher cumulative GPA. On the basis of these findings, one would be hard pressed to argue that developmental English course work had a greater positive effect on older students than on recent high school graduates. If there are any benefits inherent in developmental programs, then there seems little reason, on the basis of these results, to restrict admission to them to students four or more years out of high school. In 1984, the Florida legislature seemed to agree because it changed its 1983 position on limiting what it now referred to as "college preparatory courses" to older students after 1990, striking language references to students by age and to 1990 as a developmental/remedial course elimination date (Section 24, Senate Bill 923, 1984). In conclusion, the following specific recommendations seem war-ranted by this research: 1. At Valencia Community College, further research should be conducted on the 1980 fall entrants to ascertain if individ-ualized course completers by themselves and classroom course compl eters by themselves showed more pos it i ve outcomes than students of similar aptitude at entry who avoided enrolling in a developmental writing course.

PAGE 119

109 2. At Valencia Community College, further research should be conducted on 1984 fall entrants, for whom the upgraded developmental courses and more stringent Freshman Composit ion I exit criteria exist, and for whom passing the CLAST test is mandatory for an Associate in Arts degree. Developmental compl eters shoul d be compared wi th non-compl eters. Adjust ments on the developmental writing curriculum should follow unless clear positive results are obtained. 3. The Florida Department of Education should investigate the feasibility of a statewide evaluation model for developmental and remedial courses. This model should be devised by a panel of developmental education experts working in the state's postsecondary institutions, primarily at the community colleges. The panel should seek broad based, statewide consensus on the model's features and implementation proce dures before its formal adoption. Evaluation should be con ducted primarily by a team or teams of developmental education experts, both faculty and administrators, preferably working within Florida public postsecondary institutions. 4. Nowhere should students be denied access to developmental or remedial writing courses on the basis of age.

PAGE 120

APPENDIX A INITIAL ENGLISH ASSESSMENT TEST Valencia Community College There are 35 questions. Each question has only one correct answer. Read the question. Read the four choices. Mark your choice on the answer sheet for English. 1. Which sentence follows the same pattern as this example? Example: I called him president of the class. A. I found Mark and returned him to his parents. B. I appointed Jim leader of the scout troop. C. I attended the karate class regularly . D. I bought vegetables and fruit when I went shopping. 2. Which sentence is NOT punctuated correctly? A. We spend Labor Day on my father-in-laws houseboat. B. Louise asked me why she should borrow my dictionary? C. The first runner who crosses the finish line will be declared the winner. D. The pretty little American girl had her picture taken in her Easter dress. 3. Which underlined pair of words in the following sentence should be separated by a comma? Some employees at this plant are data processors working the third shift and others are officers guarding classified areas. A. plant, are B. processors, working C. shift, and D. officers, guarding 4. Which underlined word is correct? A. Either the workers or their boss will share their lunch. B. Neither of the women has her independence. C. Either Dan or Ted will sign-their name. D. Neit her the general nor his officers have revealed his secrets. 110

PAGE 121

l11 5. In which sentence are all the verbs correct? A. He stopped moving after he had swum ten mi 1 es in the Gulf of Mexico. B. Have they spoke to you about their problems since they arrived? C. Yesterday morning he had took his car to the garage before he went to work. D. Before their Coldspot freezer broke, they had froze half a cow. 6. Which sentence needs punctuation? A. The man who is wearing the ten gallon hat is the manager of the circus. B. Terry asked me if I would help him with his math. C. Terry said that I should show him the right page. D. Inside the fireplace was centered on the wall. 7. Which sentence punctuated correctly? A. Did he say, II A 11 pol iticians are corrupt! II B. Did he say, II A 11 po 1 it i c i an s are corrupt?1I C. Did he say, IIAll pol iticians are corruptll? D. Did he say, II All politicians are corrupt!lI? 8. Which of the following is a complete sentence? A. I loved her almost at first sight. B. Although we never spoke after that day. C. The first and only time I saw her. D. If she only knew that I loved her so. 9. Which of the underlined words is the subject of the sentence? A. The flowers, wilting in the sun, need rain. B. Where is the tall in the yellow dress? C. Waiting for the decision of the judge, Billie Jean glared at Chris. D. Will the life raft keep the survivors safe in rough seas? 10. Which sentence follows the same pattern as this example? Example: Since he was chairman of the committee, John handed President Carter the trophy. A. Since he was the doctor on call, he gave his patient a prescription. B. Since she always wanted to be alone, we called Marie Greta Garbo. C. Since she had two class periods to cover, the teacher chose a rather long poem. D. Since a storm was developing, the swimming instructor gave orders to clear the beach.

PAGE 122

112 11. Which sentence contains a misspelled word? A. No amount of money could pay for the misery you have caused. B. He is generous if you happen to be one of his friends. C. If the string is to loose, the package will fall apart. D. If you work at it, your thought can become a clear para graph. 12. Which sentence is NOT punctuated correctly? A. Before the janitor came to clean the room, the dirty coffee cups and crump 1 ed papers 1 ittered the tab 1 e and the tile floor. B. The used coffee cups, marked with lipstick and filled with cigarette butts, sat on the table and the crumpled papers were strewn on the tile floor. C. Although there were lipstick-stained coffee cups on the table and crumpled papers everywhere on the tile floor, the janitor skipped cleaning the conference room. D. The janitor swore three times while refusing to clean the tables covered with used coffee cups and the floor strewn with crumpled papers. 13. Which sentence contains a misspelled word? A. Can the Red Carpet Motel accommodate so many quiet guests? B. They had already given the room a thorough cleaning. C. The Weavers occasionally have trouble paying their rent by the first of the month. D. Even though she stepped on the brakes, she accidently backed into the post. 14. Which sentence needs commas? A. The President aged and infirm died in office. B. The infirm twenty-ninth President died in office. C. The aged and infirm President died in office. D. The aged President died in office at the end of June. 15. Which of the sentences below offers the least effective support for the following beginning sentence in a paragraph? The manager of a local convenience store decided to do something about the young boys who had been stealing beer. A. She called the police immediately. B. She chased two boys as they left the store. C. She tripped a boy hiding a can of beer in his socks. D. She paced impatiently while waiting for the police.

PAGE 123

113 16. Choose the correct words to fill in the blanks. The are in the closet. A. childrens coats l B. childrenls coats C. childrens l coats D. childrens coatis 17. Which sentence contains a misspelled word? A. Studing his algebra every night, Tom managed to pass his course with a C. B. Pliers and scissors are based on the same principle. C. The demonstration proceeded through town to the airport. D. It is already nine olclock, and we were supposed to be there by ei ght. 18. Which sentence follows the same pattern as the example below? Example: The spy gave his partner the stolen documents. A. The judge made him apologize to the officer. B. The bank teller handed in his letter of resignation. C. The minister gave an interesting sermon on Sunday. D. The minister told the congregation a joke. 19. Worried by the red ink, the frightened students stared at the grade on the second test. In the sentence above, the underlined words describe A. frightened B. student C. grade D. test 20. Which sentence in the following paragraph expresses the main idea? (1) A snakels skin is dry and smooth, not wet and slimy as many people think. (2) A snake does not stare at his prey in order to hypnotize it; a snake simply has no eyelids. (3) Even the feared rattlesnake would much rather flee from a human, given time, than exercise his rattle. (4) It seems that our legends and old wives tales have given the snake a bad shake. A. (1) B. (2) C. (3) D. (4)

PAGE 124

114 21. Which of the following sentences is punctuated correctly? A. Bill borrowed his neighbor's 1awnmower; his neighbor stole Bill's rake. B. When Bill borrowed his neighbor's 1awnmower; his neighbor stole Bill's rake. C. Bill borrowed his neighbor's 1awnmower; and his neighbor stole Bill's rake. D. Having borrowed his neighbor's 1awnmower; Bill gave the neighbor a new rake. 22. Unless we stop polluting our environment, we are headed for disaster. In the sentence above, the underlined group of words is A. as important as the rest of the sentence. B. the main idea of the sentence. C. less important than the rest of the sentence. D. able to stand alone as a sentence. 23. Which sentence is correct? A. To get well, the patient requires an operation. B. After snooping around the attic, a cowboy suit was discov ered. C. In old age, my father's impatience with new ideas became intolerable. D. Walking in the snow, my nose felt frozen. 24. Which comma in the following sentence is NOT correct? At the Third Century America exhibit at Cape Canaveral, a Saturn rocket, like those used to launch astronauts to the moon, lies on its side, each stage has its own concrete stand. A. Canaveral, a B. rocket, like C. moon, 1 i es D. side, each 25. She could not remember his face, but she cou1dn' t forget his warm eyes. In the sentence above, the underlined word means that A. the two parts are related and of equal importance. B. the second part is less important than the first part. C. the second part is more important than the first part. D. the two parts are not related, but they are equal.

PAGE 125

ll5 26. Choose the words that complete the following sentence correctly. H e won the race because -------------------A. he drove his car faster than the others. B. he drove his car smoother than the others. C. he drove his car more careful than the others. D. he drove his car less dangerous than the others. 27. Choose the correct word to fi 11 in the blank. The Mi 11 ers I telephone is disconnected because _________ moving. A. there B. their C. they're D. thier 28. Which of the following is NOT a complete sentence? A. A good man is hard to find. B. Without a doubt, you1re right. C. Raccoons are active mainly at night. D. Until they came upon a man who led the way. 29. Choose the words which best complete this sentence. On the first day of our vacation, we went first to the driving range, then to the bowling alley, next to the swimming pool, and fi na 11y _________ _ A. to the racing cars. B. to the cars at the racing track. C. to the race track. D. we went to drive the race cars. 30. Which underlined word tells you what the duck does? A mi grat i ng duck seldom fl i es away from the safety of a honk i ng flock. A. migrating B. flies C. away D. honking 31. Fill in the blank with the most appropriate word. Diane's poise _________ her flawless figure. A. complimented B. comp 1 i cated C. compromised D. complemented

PAGE 126

116 32. Which of the underlined groups of words does NOT contain the subject? A. Lying in the sun can be dangerous. B. There are several ways to skin a cat. C . The redheaded boy is the class president. D. The warm afternoon sun felt pleasant. 33. Which of the following words best completes this sentence? John is ready to fight at the drop of a hat; he is very A. arrogant B. overbearing C. supercilious d. pugnacious 34. Which sentence does NOT belong with the others? (1) The turtle's protective shell has allowed this animal form to survive for over 200 million years. (2) The protection from severe weather and rough terrain afforded by the shell has allowed the turtle to migrate readily from dying lakes to pros pering ones. (3) A dying lake is one whose ecosystem can no longer support the feeding and mating habits of turtles and other species. (4) Also, the large bony shell protects the turtle from being eaten by enemies; only man and the larger alligators can penetrate the defense of this enduring reptile. A . (1) B. (2) C . (3) D. (4) 35. The following sentence states the main idea of a paragraph: Dreaming is not time wasted; it promotes good mental health. Select the topic that best tells what the rest of the paragraph will cover. A. symbo 1 ism of dreams B. 1 ength of dreams C. effects of dreams D. interpretation of dreams cValencia Community College 1978 Developed 1977 by Elaine Greenwood, William Reicherts, Evelyn Shirkey, Don Tighe, Audrey Williams (Valencia Community College) and Abby Spero (intern teacher, University of Iowa) consulting with Nancy Johnson (Northern Virginia Community College).

PAGE 127

117 Revised 1978 by Celia Cullom, Elaine Greenwood, Anne Hammond, William Reicherts, Evelyn Shirkey, Mary Steffancin, Don Tighe (Language Usage Instrument Committee), Valencia Community College. Revised 1979 by the Language Usage Instrument Committee.

PAGE 128

APPENDIX B SAMPLE STUDENT TRANSCRIPT OS/23/84 01 NOT AN OFFICIAL tRANSCRIPT FOR VALENCIA INTERNAL USE ONLY VALENCIA COMMUNITY COLLEGE SESSION I 1980-81 ART2110 CERAMICS I A 3.0 3.0 3.0 12.0 ART2600 PHOTOGRAPHY I W 3.0 EC01000 BASIC ECONOMICS W 3.0 ENC1103 FRESH COMP I C 3.0 3.0 3.0 6.0 SESSION TOTAL 6.0 6.0 6.0 3.00 SESSION 2 1980-81 ART2110 CERAMICS I W 3.0 ENCl136 FRESH COMP II W 3.0 LIT2220 WRLD LT ENLIT-PR C 3.0 3.0 3.0 6.0 TPP1110 BEG. ACTING B 3.0 3.0 3.0 9.0 SESSION TOTAL 6.0 6.0 6.0 2.50 SESS ION 4 1980-81 ARTl301 DRAW ING I C 3.0 3.0 3.0 6.0 POS1041 AMER GOVERN I W 3.0 SESSION TOTAL 3.0 3.0 3.0 2.00 SESS ION 1 1981-82 ANT2000 I NTRO ANTHROPOLO D 3.0 3.0 3.0 3.0 DEP2003 DEVELOPMENTAL PSY W 3.0 SESSION TOTAL 3.0 3.0 3.0 1.00 SESSION 4 1981-92 ENC1102 FRESHMAN COMP II C 3.0 3.0 3.0 6.0 PSYlO12 GEN PSYCHOLOGY W 3.0 SESSION TOTAL 3.0 3.0 3.0 2.00 118

PAGE 129

119 SESSION 1 1982-83 DEP2003 DEVELOPMENTL PSY F 3.0 0.0 3.0 POS1041 AMER GOVERN I B 3.0 3.0 3.0 9.0 SPN1000 BASIC SPANISH W 3.0 SESSION TOTAL 6.0 3.0 6.0 1. 5 0 SESSION 2 1982-83 HUM2213 HUM GREEK-GOTHIC C 3.0 3.0 3.0 6.0 SPNllOO ELEM SPANISH I W 3.0 SESSION TOTAL 3.0 3.0 3.0 2.00 TOTAL VALENCIA 30.0 27 .0 30.0 2.10 ALL COLLEGE CUMULATIVE 30.0 27 .0 30.0 2.10 TOTAL W COURSES 0 TOTAL R COURSES 0 END OF TRANSCRIPT

PAGE 130

APPENDIX C DESCRIPTION OF DEVELOPMENTAL CLASSROOM AND INDIVIDUALIZED COURSES In the fall semester of 1980, the developmental writing program at Valencia Community College consisted of two courses, one a traditional classroom course and the other an individualized course. Members of the West Campus Communications Department had originally designed both courses five years earlier based on an analysis of five hundred student papers which faculty judged as showing deficiencies in basic writing skills. Faculty had then developed course objectives for the classroom course based on the most frequent errors found during this analysis. Objectives for the individualized course were to be flexible depending on the needs of the students who enrolled in it. For example, the objectives for a student with spelling problems might consist primarily of specific spelling improvement outcomes while t h e objectives for a student whose main problems were writing sentence fragments and run-on sentences mi ght cons i s t primarily of i mprovi ng sentence structure. In 1980, all faculty teaching the classroom course used the common set of course objectives, which were distributed to students during the first week of class. The text used by all classroom course students was Meyers' Writing with Confidence (1979). The common set of topics presented during the course consisted of parts of speech; parts of sentences; subject-verb agreement; pas t and perfect tenses for regular and irregular verbs; verb tense consistency; regular and 120

PAGE 131

121 irregular noun plurals and possessives; pronoun-antecedent agreement; pronoun case and consistency; correct forms of adjectives and adverbs; avoi dance of sentence fragments, comma sp 1 ices, and run-on sentences through coordination and subordination; correct use of commas and semicolons; and paragraph construction. In addition to the regular three hours a week of classroom instruction, some faculty teaching the classroom course assigned one or more audio-taped programs housed in the Writing Lab as supplemental work. Students also wrote and rewrote at 1 east one paragraph a week during the fifteen weeks of the course. The common final exam consisted of two parts: an objective grammar, usage, and punctuation test and three paragraphs written on different topics. The objective test was scored by computer, and grades were assigned using a common grading scale. Instructors graded their own students' paragraph section of the final exam and assigned final grades. Students in the individualized course were required to meet with their instructor in the Writing Lab a minimum of three hours per week on a regular, assigned schedule. Students were encouraged to use the lab as many additional hours as they wished. During the first week of class, faculty analyzed student writing samples and results of the reading test to determine which of several textbooks (or no textbook) would best meet the needs of each student. The most frequently assigned texts were Fawcett and Sandberg's Grassroots (1976); Daiker, Kerek, and Morenberg's The Writer's Options (1979); Feinstein's Programmed Spelling Demons (1973); Langan's English Skills (1979); and McClelland and Hale's English Grammar Through Guided Writing: Verbs (1976).

PAGE 132

122 Instructors were assigned only ten students per class section . Sitting at tables or carrels, students worked in their text or on one of the audio-visual programs which supplemented their program. Instructors conferred with each student, one-to-one, at nearly each class meeting. During these conferences, instructor and stu dent reviewed what the student had accomplished since the last meeting, discussed concepts and skills the student was then studying, discussed how successful the student had been in completing practice exercises and written paragraphs, and planned what assignments were to come next. Instructors graded students on number of concepts and skills mastered during the semester in addition to degree of mastery. The final exam consisted of writing three paragraphs which faculty g raded on the basis of how well the student was able to use the concepts and skills studied during the semester. The Coordinator of the Writing Lab, a full-time faculty mem ber, taught both courses and coordinated the efforts of all other full-tim e and part-time faculty members teaching both courses.

PAGE 133

REFERENCES Akst, G., & Hecht, M. (1980). Program evaluation. In A.S. Trillin & Associates (Eds.), Teaching basic skills in college (pp. 261-296). San Francisco: Jossey-Bass. Baker, R. G. (1982, March). A comparison of college f reshmen achievement in remedial English courses and in freshman composition courses at a two-year college. Paper presented at the annual meet ing of the American Educational Research Association, New York. (ERIC Document Reproduction Service No. ED 214 615) Barton, M. A. (1984). A study of the effectiveness of selected aspects of the developmental education program at an urban multicampus community college. Dissertation Abstracts International, 44, 3256A. --Bers, T. H. Assessment of mandatory placement in communications: Fall 1981. Des Plaines, IL: Oakton Community College, Office of Institutional Research. (ERIC Document Reproduction Service No. ED 214 599) Bers, T. H. Follow-up study of students enrolled in develop mental communications, Fall 1981. Des Plaines, IL: Oakton Com munity College, Office of Institutional Research. (ERIC Document Reproduction Service No. ED 223 289) Bloom, B. S. (1971). Mastery learning and its implications for cur riculum development. In E. W. Eisner (Ed.), Confronting curriculum reform (pp. 17-49). Boston: Little, Brown and Company. Braxton, P. C., Douthat, S. G., Hale, D. A., Lawson, H. C., Safko, G.M., & Sims, D. C. (1980). A program evaluation and student follow-up of developmental studies students at Thomas Nelson Com munity College, 1975-1977. Hampton, VA: Thomas Nelson Community College. (ERIC Document Reproduction Service No. ED 203 893) Brown, J. W. (1984). The effects of teacher-pacing versus studentpacing on high-risk students at the community college level. Dissertation Abstracts International, 44, 2962A. Campbell, R. (1982). Developmental and remedial education: A survey of AACJC members, 1981. Monroe, MI: Monroe County Community College. (ERIC Document Reproduction Service No. ED 221 247) Carnegie Commission on Higher Education. (1970). The open-door college. New York: McGraw-Hill. 123

PAGE 134

124 Clowes, D. A. (1984). The evaluation of remedial/developmental pro grams: A stage model of program evaluation. Journal of Developmental Education, Q(1), 14-15, 27-30. Cross, K. P. (1971). Beyond the open door: New students to higher education. San Francisco: Jossey-Bass. Daiker, D., Kerek, A., & Morenberg, M. (1979). The writer's options. New York: Harper & Row. Dudley, J. R. (1978). A remedial skills course for underprepared college students. Journal of Educational Research, 21, 143-148. Dumont, R. G., Bekus, A. J., & Tallon, W. A. (1981, May). Evaluating the guality of basic skills programs. Paper presented at the annual forum of the Association for Institutional Research, Minneapolis, MN. Dumont, R. G., & Jones, J. T. (1983). Discriminant analysis applied to basic skills programs. Community College Journal for Research and Planning. 1(2), 14-33. Epes, M., Kirkpatrick, C., & Southwell, M. (1980). An evaluation of the com -lab roject, final re ort. Jamaica, NY: York College, City Unlverslty 0 New York. ERIC Document Reproduction Service No. ED 194 909) Fawcett, S., & Sandberg, A. (1976). Grassroots. Boston: Houghton Mifflin. Feinstein, G. W. (1973). Programmed spelling demons. Englewood Cliffs, NJ: Prentice-Hall. Gollattscheck, J. F., Harlacher, E. L., Roberts, E., & Wygal, R. (1976). College leadership for community renewal. San Francisco: Jossey-Bass. Grant, M. K., & Hoeber, D. R. (1978). Basic skills programs: Are they working? (AAHE-ERIC/Higher Education Research Report No.1). Washington, DC: ERIC Clearinghouse on Higher Education, The George Washington University. Kendrick, S. A., & Thomas, C. L. (1970). Transition from school to college. Review of Educational Research, 40, 151-179. Kulik, C-L. C., Kulik, J. A., & Shwalb, B. J. (1983). College programs for high-risk and disadvantaged students: A meta-analysis of findings. Review of Educational Research, 53, 397-414. Langan, J. (1979). English skills. New York: McGraw Hill.

PAGE 135

125 Linthicum, D. S. (1979). Statewide assessment of developmental/remedial education at Mar land communit colle es. Annapolis, MD: Maryland State Boar for Community Co ERIC Document Reproduction Service No. ED 175 514) Losak, J., & Burns, N. (1971). An evaluation of the community college studies program for the tear 1969-1970. Miami, FL: Miami-Dade Junior College. ERIC Document Reproduction Service No. ED 056 683) Lovejoy, D. F. (1974). Assessment of a remedial English program for academically disadvantaged young adults at Western Christian Col lege. Dissertation Abstracts International, 34, 5662-A. Lunsford, A. (1978). Measurable improvements in the writing of remedial college students. (ERIC Document Reproduction Service No. ED 155 725 ) The master plan for Florida postsecondary education: Report and recommendations of the Postsecondar Education Plannin Commission. 1982 • Tal ahassee, FL: State Board of Educatlon. Maxwell, M. (1979). Improving student learning skills. San Francisco: Jossey-Bass. McCabe, R. H. (1983, February). Miami-Dade Community College should provide whatever services are needed, including developmental education, to assist as many students as possible to h ig h levels of achievement. Miami, FL: Miami-Dade Community College. McClelland, L., & Hale, P. (1976). English grammar throurh guided writing: Verbs. Englewood Cliffs, NJ: Prentice-Hal. Meyers, A. (1979). Writing with confidence. Glenview, IL: Scott Foresman. Mowatt, W. D. (1978). The impact of non-traditional students in selection or non-selection of developmental English within a community college setting. Dissertation Abstracts International, 39, 153-A. --New Jersey 8asic Skills Council. (1983). Report on the Effectiveness of Remedial Programs in New Jersey Public Colleges and Universities, Fall, 1982. Trenton, NJ: New Jersey Department of Higher Education, Basic Skills Council. Patterson-Griffith, M. C. (1983, September 21). Insur ing academic quality: Regrettably it's time for the state to step in. The Chronicle of H i gher Education, p. 64.

PAGE 136

126 Perry-Miller, M., Nolan, J., & Smith, J. (1980, April). Developmental education evaluation model. Paper presented at the meeting of the American Educational Research Association, Boston, MA. (ERIC Document Reproduction Service No. ED 193 299) Presley, J. W. (1981, November). Evaluating developmental English programs in Georgia. Paper presented at the annual meeting of the National Council of Teachers of English, Boston, MA. (ERIC Document Reproduction Service No. ED 214 166) Reap, M. C., & Covington, H. C. (1980). Evaluation of the effectiveness of the develo mental studies ro ram. Houston, TX: North Harris County College. ERIC Document Reproduction Service No. ED 197 798) Richardson, R. C., Jr., Martens, K. J., & Fisk, E. c. (1981). Functional literacy in the cOlle,e setting (AAHE-ERIC/Higher Education Research Report No.3., 1981 . Washington, DC: American Association for Higher Education. (ERIC Document Reproduction Ser vice No. ED 211 032) Riles, C. (1983, July). PAGSS follow-up study. Unpublished manuscript, Valencia Community College, Orlando, FL. Roueche, J. E., Herrscher, B. R., & Baker, G. A. (1976). Time as the variable, achievement as the constant: Competency-based instruction in the community cOllege. Washington, DC: American Association of Community and Junior Colleges. Roueche, J. E., & Pitman, J. C. (1973). A modest proposal: Students can learn. San Francisco: Jossey-Bass. Roueche, J. E., & Snow, J. J. (1977). Overcoming learning problems. San Francisco: Jossey-Bass. Rule 6A-IO.30 Florida Administrative Code. (1982). Rule 6A-10.31 Flori da Administrative Code. (1982). Rule 6A-IO.312 Florida Administrative Code. Rule 6A-10.313 Florida Administrative Code. Rule 6A-IO.314 Florida Administrative Code. Section 24, Florida Senate Bill 923. (1984). Section 232.2455, Florida Statutes. (1983). Section 240.117, Florida Statutes. (1984). Section 240.134, Florida Statutes. (1983). (1984) • (1982) . (1984).

PAGE 137

127 Sharon, A. T. (1972). Assessing the effectiveness of remedial college courses. Journal of Experimental Education, 41(2), 60-62. Smittle, P. (1982). Evaluation of Santa Fe Community College develop mental studies prorram. Unpublished manuscript, Santa Fe Community College, Gainesvil e, FL. Status re ort on remediation in hi her education. (1981). Springfield: I inois State Board of Higher Education. ERIC Document Reproduction Service No. ED 210 079) Sutton, D., & Arnold, D. S. (1974). The effects of two methods of compensatory freshman English. Research in the Teaching of English, ., 241-249. Tinto, V., & Sherman, R. H. (1974). The effectiveness of secondary and hi her education intervention ro rams: A critical review of re search. New York: Teachers Col ege, Co umbia University. ERIC Document Reproduction Services No. ED 101 042) Turlington, R. D. (1983, May 2). Performance standards for the College-Level Academic Skills Test (CLAST) (Memorandum to the presidents of Florida community colleges and state universities). Tallahassee, FL: State of Florida Department of Education. Whittle, D. H. P. (1980). Success in freshman English preceded by de velopmental English: Predicting success through persistence and selected demographic variables. Dissertation Abstracts International, 41, 1363A-1364A.

PAGE 138

BIOGRAPHICAL SKETCH Elaine A. Greenwood was born Elaine A. Guenette in Salem, Massachusetts, December 23, 1945. After graduating from Salem Classical and High School in 1963, she attended Boston College, graduating magna cum laude with a Bachelor of Science degree in English and education in 1967. Having taught English for two years at Wakefield Public High School, Wakefield, Massachusetts, she returned to Boston College and Harvard University for graduate work. She completed a Master of Arts in English at Boston College in 1971. After moving to Orlando, Florida, Greenwood taught one year as a substitute teacher at Robinswood Junior High School and as a part-time instructor at Mid-Florida Technical Institute, both in the Orange County Public School system. In 1974, she joined the adjunct faculty of Valencia Community Col lege. In 1976, she became a full-time faculty member as Coordinator of the Writing Lab. This position combined individual ized instruction of developmental writing students with training and supervision of adjunct faculty teaching in the Writing Lab. She has received in-house grants to study microcomputer applications in English instruction, videotapes for staff development in individualized instruction, and materials development for the Writing Lab. In 1984, she was chosen for an internal administrative internship as Assistant to the President of the College. 128

PAGE 139

129 At the state and national level, Greenwood has served in numerous capacities with the College Level Academic Skills Project and the Florida Developmental Education Association. She has made presentations at the following conferences: National Master Teachers' Seminar, Southeast Region Conference on English in the Two Year College, Florida Developmental Education Association, Florida Association of Community Colleges, Florida Association of Media Educators, South Atlantic Modern Language Association, and College Board Assessment and Placement Workshop. She is currently serving as Coordinator of CLAST/Gordon Activities at Valencia Community College.

PAGE 140

I certify that I have read this study and that in my oplnlon it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. ' Professor of Instructional Leadership and Support I certify that I have read this study and that in my oplnlon it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. es L. Wattenbarger rofessor of Educational Administration and Supervision I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Education. Charl es D. Dziuban Professor of Instructional Leadership and Support This dissertation was submitted to the Graduate Faculty of the Co llege of Educat i on and to the Graduate Schoo 1 and was accepted as partial fulfillment of the requirements for the degree of Doctor of Education. December 1984 Dean, College of Dean for Graduate Studies and Research