Citation
Part-time employment as a predictor of grade point average for secondary school students in an urban school district

Material Information

Title:
Part-time employment as a predictor of grade point average for secondary school students in an urban school district
Creator:
Squires, Walter G
Publication Date:
Language:
English
Physical Description:
viii, 75 leaves : ; 28 cm.

Subjects

Subjects / Keywords:
Academic achievement ( jstor )
Correlations ( jstor )
Employment ( jstor )
Grade point average ( jstor )
High school students ( jstor )
Part time employment ( jstor )
Part time students ( jstor )
Schools ( jstor )
Scope of employment ( jstor )
Students ( jstor )
Dissertations, Academic -- Educational Administration and Supervision -- UF
Educational Administration and Supervision thesis Ed. D
High school students -- Florida -- Duval County ( lcsh )
Part-time employment ( lcsh )
Prediction of scholastic success ( lcsh )
Duval County ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ed. D.)--University of Florida, 1983.
Bibliography:
Bibliography: leaves 72-74.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Walter G. Squires, Jr.

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:
029289910 ( ALEPH )
09882471 ( OCLC )

Downloads

This item has the following downloads:


Full Text












PART-TIME EMPLOYMENT
AS A PREDICTOR OF GRADE POINT AVERAGE
FOR SECONDARY SCHOOL STUDENTS
IN AN URBAN SCHOOL DISTRICT







BY









WALTER G. SQUIRES, JR.


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




UNIVERSITY OF FLORIDA


1983















Copyright 1983


by


Walter G. Squires, Jr.















To


Walter G. Squires Corine T. Squires


Parents














ACKNOWLEDGEMENTS


The author wishes to express his appreciation to those whose efforts have contributed to the completion of this dissertation. Gratitude is offered to Dr. M. Y.

Nunnery, Chairman of the Supervisory Committee, Dr. J. W. Longstreth, Dr. T. C. Healy, and Dr. D. A. Jacobsen, for their support and assistance. The author wishes to extend

a singular thank you to Dr. Nunnery for his invaluable wisdom, guidance, and patience in the actual preparation of this document.

The assistance of Mr. Ronell Poppell, Principal of N. B. Forrest High #241, given at most critical times is greatly appreciated. The author wishes to recognize those friends that helped with much of the tedious detail work so necessary for data collection: Mr. Jessie Bullard, Miss Diana Lynn Hastings, Miss Carmen Denise Blouir, Miss Janet Coffrin, Miss Tamaii L. Wilson, and Mr. William Green. A special thanks to Mr. Loel A. Cruikshank, auditor, who double checked all data entries to ensure accuracy. He is a great father-in-law.

A particular thank you is offered to Claire, the author's wife, for her prolonged effort, love, continuing

support, understanding, and encouragement during some very difficult times.









A final note of appreciation is offered to Miss Barbara Young who identified the need for study of the stated problem.
















TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS iv ABSTRACT vii CHAPTER
I INTRODUCTION, 1

The Problem 4 Delimitation and Limitations 4 Justification of the Study 5 Definition of Terms 8 Procedures 10 Organization of the Research Report 16

II CORRELATES OF ACADEMIC ACHIEVEMENT 17
AMONG HIGH SCHOOL STUDENTS: A REVIEW
OF THE RESEARCH

III PRESENTATION AND ANALYSIS OF DATA 30

Descriptive Data by Grade Level
Relative to Students Involved In the
Study 31
Correlation Matrices 37
The Results of Multiple Regression
Analysis 47
Multiple Regression Data 49

IV SUMMARY, CONCLUSIONS AND DISCUSSION 55

Summary 55 Conclusions 62 Discussion 63

APPENDICES
A DUVAL COUNTY SCHOOL SYSTEM 68
BIOGRAPHIC DATA FORM

B STUDENT INFORMATION QUESTIONNAIRE 69

C STUDENT DATA NEEDED FOR LISTINGS 71
OF SAMPLE DATA BASE

REFERENCES 72 BIOGRAPHICAL SKETCH 75














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


PART-TIME EMPLOYMENT
AS A PREDICTOR OF GRADE POINT AVERAGE
FOR SECONDARY SCHOOL STUDENTS
IN AN URBAN SCHOOL DISTRICT By

Walter G. Squires, Jr.

April, 1983


Chairman: M. Y. Nunnery Major Department: Educational Administration


Given the paucity of research and policy implications relative to the impact of part-time employment on academic performance among secondary school students, the study of the impact of part-time employment on grade point averages, after accounting for the contributions of eight other variables known to affect grades, was undertaken. From the 34,340 students enrolled in grade 8-12 of an urban school district a grade-level stratified random sample was selected. Data were collected from district-maintained student records and from a student-completed instrument relative to 470 students' employment status, the selected other presumed independent variables, and 1981-82 grade point averages.


vii







Two multiple regression analyses were run for each grade, one using whether or not the student was employed and the other the amount of employment. The form of the analysis was stepwise for all of the variables except employment status which was entered last.

Since in only three instances did the part-time employment make a significant contribution to the variations in grade point averages (accounting for 1.1% and 1.2% of 68.7% total variance and 2.2% of 52.5% total variance), it was concluded that its impact was minimal. Consistent with previous research, previous grades, academic aptitude scores, and school attendance consistently accounted for most of the variance in grade point averages.


viii















CHAPTER I
INTRODUCTION


There can be found two schools of thought on part-time employment for secondary school students. Teen-age students in junior and senior high school who work part-time jobs are thought by some educators to be adversely affected academically because of part-time work (Cole, 1980, p.44). That is, part-time work has an adverse effect on the grade point average (GPA) of working students. This school of thought also holds that the requirement for students to work part-time while attending school precludes their participation in many school related activities. It is reasoned that involvement in cocurricular activities is an integral part of the school years' development and that the "values orientation" from such participation is lost for students working while attending school (Williams, 1967, p. 36; Young, 1980, D-3).

The followers of the opposing school of thought hold concepts likely to be expressed as follows: "Part-time employment for students serves to keep them off the streets and out of trouble." "It helps them to raise their standard of living, and learn the value of a dollar."







2

"Part-time employment prepares young people to become working members of society." "Part-time employment contributes to the reduction of unemployment in general and to an increase in the gross national product in particular." A 6-city survey by the National Manpower Institute indicated that 68% of the employers surveyed were firmly in favor of youth work experience and the survey cited such benefits of working as developing maturity, self-identity, self-reliance, preparing for future career direction, establishing good working relationships, and enjoying the stimulation of working with competent professionals (Delaney, 1975, VI-14).

Relative to the stance of parents and teachers on the matter of part-time employment for high school youth, it can be argued that the concurrence of parents is expressed when they sign the appropriate forms to enable their children to secure work permits. The researcher, in preparation for the research reported herein, talked informally with a number of teachers about the issue and found a wide variation of opinion with several teachers supporting the notion of

part-time employment and other teachers equally strong in their opposition to part-time employment. The division of opinion tended to center around whether or not it impacted on their school activity both in terms of academic performance and the ability to participate in extra-curricular activities.

Some situations make it necessary for teen-age students to work part-time. Low family income, divorce, death







3

of the principal wage earner in the family, unanticipated need to include the student in the family work force, or abandonment by one or both parents may force the teenage student to fend for himself or herself financially. Socioeconomic and peer pressure to participate in activities that require a substantial outlay of funds mandate part-time work by some students. Some teen-age students work part-time jobs when it is a matter of convenience to acquire luxury items. Furthermore, government and individually-sponsored student employment programs, which have been instituted in an effort to combat students dropping from school, have encouraged students to work during part of the school day or outside classroom hours (Delaney, 1975; Dykeman, 1979, p. 365; Heyneman, 1977; Levine, 1979). The conclusion drawn by the researcher, based on available information, was that apparently administrators and counselors involved in such programs had not considered the possibility that the part-time employment could have an adverse influence on student academic performance; or, the assumption was made by these school officials that the benefits from gainful part-time employment out-weighed any possible adverse academic effects.

A basic conflict can be seen in the two points of view on part-time employment for secondary school students. Therefore, the study reported herein focused on the relationships that might exist between academic achievement as reflected in grade point averages and part-time employment by students.






4

The Problem

The problem in the study was to determine for students in each grade level, grades 8 through 12, in a single urban school district the relative impact of parttime employment upon their 1981-82 grade point average (GPA) after the contributions of sex, ethnic group, socioeconomic status (SES) , school attendance, GPA for previous school year (1980-81), self-perceptions of intelligence, extracurricular activity participation, and Stanford Achievement Test Battery (SAT) overall percentile score were taken into account.


Delimitation and Limitations

The potential participants in the study were 34,340 students who were enrolled in grades 8 through 12 of the Duval County, Florida, School System in the school year 1981-82. From this population, at each grade level a random sample of 120 students was selected for inclusion in the study.

Data in regard to 1981-82 GPA, school attendance, sex, ethnic group, SAT score, SES, and 1980-81 GPA were obtained from the Duval County School Board SIMS computer data base and student cumulative record files. All data were verified with the student permanent record files maintained at the school that the student was attending.

The data source for information pertaining to hours worked, extracurricular activity participation, self-perception of intelligence, and participation in free or






5

reduced lunch program was a researcher-designed student information questionnaire. Complete useable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students.

Multiple regression was used to determine the contributions of the independent variables to the dependent variable. In entering the variables in the multiple regression equation, the part-time employment variable was entered last in order that the impact of this variable over and beyond that of the other independent variables could be determined. The other independent variables were entered in stepwise fashion.

To the extent that these data sources were inaccurate or incomplete the study lacked validity. However, this inaccuracy was guarded against in that each data sheet was cross-checked against the cumulative folder by the researcher. Furthermore, since the study was confined to the students in a single urban school district for a specified year, the results over and beyond this population had to be regarded as suggestive rather than conclusive. That is, the data were applicable to only one school district at one point in time.


Justification for the Study

Some school administrators, parents, guidance counselors, and politicians advocate the expenditure of vast sums of money for programs that encourage students to work






6

for a variety of reasons (Crawford & Miskel, 1978; Delaney, 1975; Dykeman, 1979; Heffez, 1979; Heyneman, 1977; Watkins & Corder, 1977). Research studies dealing specifically with student employment effects on academic achievement are very limited. The results of the studies that could be found dealing with student employment were contradictory and of limited generalizability. Most were performed using very small sample size. Most were from rather limited populations (i.e., one or two schools).

School administrators have been influenced by the Department of Labor since 1970-71 to increase the participation in the Work Experience and Career Exploration Program (WECEP) and other vocational programs in order to retain dropout prone students within the school system (Dykeman, 1979, p. 363; Shively & Watts, 1980; Skallerud, 1979). The researcher had personal knowledge of the fact that classroom teachers were expected to alter or modify their lesson plans to allow cooperative training programs students to leave class early and be spared regular homework assignments. Homeroom programs, flag observances, and announcements were routinely preempted by the vocational education students being dismissed early to catch the bus to work at two district-operated skill centers.

As has been noted, there were few research studies dealing specifically with a relationship between student employment and academic achievement. Over 1400 items were found in ERIC and other sources under the subject "academic achievement"; of that number, only 120 were







7

concerned with grades 7 through 12. Of those studies

found, only two studies focused directly upon grade point average and part-time employment. These were the Gade and Peterson (1980) and Hammond (1970) studies. The results were conflicting. In the Gade and Peterson study no significant association between employment status and academic achievement was found. In the Hammond (1970) study there was evidence that the level of achievement might be significantly lower among the employed students than among the unemployed.

Millions of dollars are being spent on various vocational education programs, cooperative work programs, student unemployment surveys, and businesses' employment of students at below the minimum wage, because they work less than 40 hours per week. All these programs are operating on the basic assumption and widespread opinion that "school youth need work." They are not founded on continuing or widespread results of scholarly research.

The study reported herein does not provide a definitive answer to the question, "Should students work or not work?" It does, however, add to available knowledge of a relationship between students working part-time and their grade point averages. It was reasoned that if the findings indicated there was no meaningful relationship between students working part-time and academic achievement, it would help alleviate the concerns of some educators about the possibility of an adverse impact of working part-time on academic achievement. If there was a meaningful adverse







8

relationship between students working part-time and academic achievement, administrators should discourage student work programs. Therefore, the study was seen as relevant to administrators who make decisions on matters concerning part-time employment policies and practices for students.


Definitions of Terms

Employment. The performance of some duty, service, or obligation of time, and availability of a person for another that results in payment is employment, a job for money. Various studies have used different criteria (e.g., "over 5 hours per week," or "at least 12 hours per week"); in the study, any work for pay outside of class on a continuing basis was considered employment. Part-time jobs of an independent contractor nature such as newspaper routes were estimated to the number of hours per week performed. Babysitting jobs were approximated to the nearest hour per week on the average considering the school year.

Ethnic group. The classification system of the Duval County, Florida, school system whereby each student assigns himself or herself to one of the following groups was used: white non-Hispanic, black non-Hispanic, Hispanic, American Indian or Alaskan Native, and Asian or Pacific Islander.

Extracurricular activity. A student's response to a global question about extra curricular activity participation contained in a researcher-designed instrument was used.







9

The terms used elsewhere in referenced materials, "cocurricular activities," and "third curriculum activities" were interpreted in this study to be synonymous with extracurricular activities.

Grade point average (GPA). The GPA is the numerical average of letter grades awarded at the end of a school year. In determining the average an A = 4 points, B = 3, C = 2, D = 1, and E = 0, or failure to earn credit for the course of study. (This was consistent with district policy for the years in question in that advance placement classes did not carry additional weight.) In this study, grades for all subjects in which the students were enrolled for credit were used in determining 1980-81 GPA and the 1981-82 GPA.

School attendance. The number of days of attendance

in school for the year as recorded in the student permanent record file was used to indicate attendance.

Self-perception of intelligence (SPIN) The SPIN was the student's response to a global question about intelligence contained in a researcher-designed instrument.

Socioeconomic status (SES). A dichotomous classification based on whether or not the student participated in the free lunch or reduced price lunch specially funded program was used. Participation in free lunch or reduced price lunch program resulted in a "Yes" classification. No participation in the free lunch or reduced price lunch program was a "No" classification. A "Yes" classification






10

was considered low socioeconomic status. A "No" classification was considered a high socioeconomic status.

Stanford Achievement Test Battery (SAT). The most recent SAT score held by the Duval County School System was used. When English, Reading, and Mathematics portions were combined a single overall score was derived. The overall percentile score recorded in a student's cumulative record file was used in the study as the measure of scholastic ability.


Procedures

The procedures used to conduct the study are described in the paragraphs that follow. These are organized by study design, sample selection, instrumentation, data collection, and data treatment.

Study Design

The study was a survey type of ex post facto study. A stratified random sample of junior and senior high school (grades 8 through 12) students entered in the data base by their completion of Duval County School System Biographic Data Form (Appendix A) was selected from the listing for the 1981-82 school year by the Duval County Student Information Management System (SIMS) computer from the population of each grade 8 through 12 students for 198182. A survey, using a researcher-designed instrument (Appendix B) , was made of selected students to determine whether or not they held a part-time job during the 1981-82






11

school year. If they did work a part-time job, the number of hours worked and other relevant data were requested. The grade point averages of the selected students as well as the other required biographic information were obtained from the Cumulative Record Folders and the computer maintained data base (Appendix C).

The data were analyzed by grade level using multiple regression to determine the contribution of part-time employment and amount of part-time employment to grade point average over and beyond the contributions made by other independent variables that previous research had indicated contributed to grades. Sample Selection and Final Sample

The 34,340 students who were enrolled in grades 8 through 12 in the 1981-82 school year in the school system of Duval County, Florida, junior and senior high schools were the population from which the student participants in the study were selected. The basic procedure for selecting the participants was by a stratified random method. Specifically, for each grade, grades 8 through 12, the Duval County SIMS computer program was used to generate a 2% random sample of students which resulted in a minimum of 120 students randomly selected at each grade level. This allowed for a minimum of 10 students for each of the independent variables in the multiple regression equation and gave a safety factor of 33 1/3% to cover loss of students who moved, or in some other fashion were not







12

available for data collection. The one-third turnover approximation was based on observed rates of turnover in the classrooms during the past four years. The turnover on a county-wide basis for the year was estimated at nearly 40% by the school superintendent (Pope, 1982, p. B-l).

As a result of the selection procedure, complete useable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students. Instrumentation and Data Collection

The Duval County School System Biographical Data Form (Appendix A), which was used to gather data for the Student Information Management System (SIMS), was a basic source of information for this study. The SIMS was established for the purpose of obtaining and storing in the computer pertinent information concerning the Duval County student population. A major goal of the system was to afford quick access to student information while reducing the reports and data collection burdens on school-based personnel (Roberson, 1980, p. 20). Information provided by this source was identification of the students to be included in the study, their sex, grade level, school attended, and ethnic grouping.

The student information questionnaire (Appendix B) was the principal instrument used to obtain the other information needed, including data which was not yet stored in SIMS such as attendance, whether or not student






13

worked part-time, if so, how many hours were worked per week, self-perception of intelligence, and verification of participation in free lunch/reduced price lunch program. As previously indicated, the student information questionnaire was developed by the researcher whose concern was to ask the questions necessary to deal with the focus of the study. After its initial development, the instrument was administered to a group of approximately 30 Duval County School System students included in the population but not in the random sample. The instrument was then revised based on their responses. Thus, it was believed that the instrument was valid for the purpose used. The student permanent record file and cumulative record file were the source documents for SIMS. They were considered to take precedence should conflicting data appear in any other sources.

The consolidating instrument that was used to combine the information from all sources to establish the data base for performing the multiple regression operations is shown as Appendix C.

Data Coding and Treatment

Each bit of data was coded in a form that was usable in the multiple regression analysis. Specifically, the coding of each of the variables was as follows:

The 1981-82 grade point average (GPA) was coded in a two digit number between 0.0 and 4.0 representing y, the dependent variable.







14

The number of days of school attendance was represented by total days present out of 180 days during the 1981-82 school year as predictor variable xI.

The sex of the student was represented by (1) for male (2) for female as predicator variable x2.

Extracurricular school-sponsored activity involvement in athletic teams, band, or clubs was coded (1) to indicate the student was involved in one or more athletic teams, band or clubs, (0) to mean not involved in any school sponsored athletic teams, band, or clubs. This was predictor variable x3.

Ethnic group membership was coded with same code used by SIMS: (1) white non-Hispanic, (2) black non-Hispanic, (3) Hispanic, (4) American Indian or Alaskan Native,

(5) Asian or Pacific Islander. Ethnic group membership was predictor variable x4.

The socioeconomic status (SES) was coded as (1) for participation in the specially funded programs for socioeconomically deprived students, (0) for nonparticipation in any of the specially funded programs. The SES was predictor variable x5.

The most recent composite, or overall SAT percentile score was entered as a two digit number as predictor variable x6�

The student's self-perception of his/her intelligence (SPIN) was coded (1) to indicate a perception of intelligence as below average intelligence, (2) a self-perception of average intelligence, (3) a self-perception of above







15

average intelligence. The SPIN was predictor variable x7

Each student's grade point average for the previous school year (1980-81) was coded as a two digit number between 0.0 and 4.0 as predictor variable x8.

For the first regression analysis, whether or not the student worked at all during school year 1981-82 (employment status 1982 yes/no) was coded as (0) for not working during the school year and (1) to indicate that student worked during the school year. This is hereafter referred to as Model I analysis. In the second analysis, the amount of time worked (employment status 1982 category) was coded using a single digit number from one - eight to indicate the categories of hours worked based on the category scale shown as a part of Appendix B. This is hereafter referred to as Model II analysis. This was predicator variable x9�
X9*
The slope of correlation between the predictor variable and the dependent variable was coded "b" as is standard practice in solving the multiple regression equations.

The above coded variables were incorporated into a multiple regression equation (Roscoe, 1974, p. 362; Belanger & Boyle, 1980), that was performed for each grade level 8 through 12. As noted, multiple regression equation analyses were performed for each grade level sample. As indicated, the first analysis, Model I, was with the part-time employment treated as (1) employed or (0) not






16

employed part-time; in the second analysis, Model II, the part-time employment was coded and treated in terms of amount ranging from zero employment to 35 plus hours per week.

The form of the multiple regression equation
A
(y = b1x1 + b2x2 + b3x3 + b4x4 + b5x5 + b6x6

+ b7x7 + b8x8 + b9x9)


is such that one can determine if the mere presence of part-time employment and the amount of part-time employment can be related to grade point average. In order to do this, all the independent variables save the part-time employment (variable x9 in each instance) were entered in stepwise fashion. Then, the part-time employment variable (whether 1 or 0 to represent part-time employment, or the category of part-time employment to represent amount of employment) was entered last. This enabled a determination to be made of the amount of contribution to the dependent variable of grade point average made by the employment variable after the contribution of the other variables has been taken into account.


Organization of the Research Report

The remainder of the study is organized into three chapters. The second chapter is a review of related literature and research. The third chapter contains the presentation and analysis of data. The fourth chapter includes the summary, conclusions, and discussion.















CHAPTER II
CORRELATES OF ACADEMIC ACHIEVEMENT
AMONG HIGH SCHOOL STUDENTS:
A REVIEW OF THE RESEARCH


Two points of view about part-time work for high school youth were found in the literature along with copious reasoning for each point of view. The scholarly research to support either point of view was limited. However, there had been numerous studies of factors related to academic achievement among high school students. In order that the reader may have a framework for interpreting the data and an understanding of why it is necessary to take into account the impact of variables in addition to part-time employment on the academic achievement of students, the review presented herein, over and beyond the two studies relating to part-time employment and academic achievement, is confined to a synthesis of the research showing the relationship of variables to academic achievement. This research in regard to other variables provided the basis on which selections were made about the other independent variables that needed to be taken into account before determining the contribution of part-time employment.

Before turning directly to the other variables, it seems appropriate to deal first with the two studies which related part-time employment and academic achievement.







18

Gade and Peterson (1980) and Hammond (1970) focused on the relationship of school related behavior, and other variables affecting academic achievement, to work experience and academic achievement. Gade and Peterson (1980) found that there was no significant association between employment status and academic achievement. Their research sample consisted of 351 10th grade students enrolled in 2 urban high schools located in the upper midwestern part of the United States. The determination of achievement was based on grades received in the last two semesters in four basic subjects. Working students were designated as those employed 12 or more hours per week. The study sought to examine the relationship of student work experience to five factors: (a) gender, (b) academic achievement, (c) extracurricular involvement, (d) socioeconomic status (SES), and

(e) self-esteem. "Students who are employed seem to perform as well or better [when identified] on [the basis of] these indicators than non-working students" (Gade & Peterson, 1980, p. 69). The limitation of the sample to 10th grade students may somewhat limit generalization downward to the junior high school students and upward to the 12th grade students.

Hammond (1970) found that the level of academic achievement was significantly lower among employed 12th grade students than among the unemployed. The rapidity of changes in behavior and personality in the teen years may produce considerably different results in lower and higher







19

grades. A predictor variable of the grade level of the student would help identify this effect.

A need for categorization of the origins of influences on student achievement has been indicated by research.

In attainment models where individual student achievement is at issue, individual student attributes, other nonschool inputs to schooling, and school inputs must be measured at the individual level to avoid misleading results. (Bidwell & Kasarda, 1980, p.
426)

Variables Related to the Background of the Students

There has been considerable research undertaken that focused on selected variables related to the background of students in relation to academic achievement. Sex, socioeconomic status (SES), standardized measures of academic ability (i.e., intelligence tests), and/or standardized measures of academic achievement (e.g., SAT) have been taken into account as background variables.

Finn, Dulberg, and Reis (1979) found in cross-national studies of educational attainment that the influence or interrelationship of the sex of the student, the teacher, and/or their interactions may be dependent on the national culture. More specifically the need to keep cultural sex discrimination in mind in drawing conclusions from sex based data was expressed as follows: "The explicit relationship of sex-segregated schooling to the attainments of boys and girls makes this issue especially worthy of exploration" (Finn, Dulberg, & Reis, 1979, p. 494). They concluded that any study of educational attainment in terms of school attendance, literacy, academic performance, or







20

preparation for gainful employment would reveal that the patterns for men and women differ markedly according to the nationality under consideration. These differences cannot be divorced from the process of schooling itself.

Opportunity to learn is one of the significant correlates of pupil achievement. Opportunity, as defined in these studies, refers to the degree of exposure to the course material and the required skills. . . . Many of the data reported in this summary suggest that inferior performance in either sex is due in part to the inadequacy of time and experience provided by instructional practices. (Finn, Dulberg, & Reis, 1979, p.497) Because of its frequent consideration, inclusion of the gender variable in the study reported herein seemed reasonable. However, based on the results of Finn, Dulberg, and Reis (1979) , the sex variable could have minimal influence in a population of one school district in the United States.

The socioeconomic status (SES) of the student is also typically considered to be a major influence on academic achievement. One of the studies that contributed a different point of view was done in Puerto Rico by Ronald L. Nuttall in 1967-68. The model involved two levels of achievement, two levels of sex, and five levels of socioeconomic status. The sample size was reduced by screening criteria from 2,500 to 800 seniors and some juniors in the largest public high school. In terms of family background there were no statistically significant interactions. This study indicated that the same factors were important for academic achievement for both boys and girls and for students from all socioeconomic status levels. Students who







21

were doing well academically desired, and were working to attain higher status occupations than those who were not doing well academically. The lower achieving students expected and desired a lower level occupation. These effects held across socioeconomic status levels. The importance of these background factors to academic achievement was explained as follows:

Personality traits associated with academic success did differ somewhat by sex and socioeconomic background.
Low achieving boys as contrasted to high achieving boys and to both high and low achieving girls, were markedly less intelligent. A similar pattern existed for the traits of conscientiousness where again the low achieving boys were markedly deficient in conscientiousness.
These low scores for intelligence and conscientiousness held at all socioeconomic levels. (Nuttall, 1972,
p. 83)

The positive relationship between the educational expectations of the adolescent and the socioeconomic status (SES) of his/her family has been consistently replicated. The students from the "better family" expected to achieve more and go farther than "poor folks." However, family expectations and adoption of those values by the student are inconsistent. Rehberg, Schafer, and Sinclair (1970) found that ambition, personal educational expectations, and measured intelligence so overshadow socioeconomic stratification that achievement predictions based predominately on SES alone would not be valid. Mobility attitudes were more strongly associated with stratification of destination than with stratification of origin (p. 46). Poor youth may work part-time and rich youth may work part-time. But neither of these categories has been shown to be an exclusively significant determinant of academic achievement.







22

In light of the studies that have been done in regard to the variables of sex, socioeconomic class, and measures of ability, it is obvious that there are some conflicts. However, since there seemed to be some indication that these may be influential, it was decided to include all three of these variables in the study reported herein. For convenience, given the high relationship between "intelligence" and "achievement," instead of using a measure labelled intelligence to portray scholastic ability, the Stanford Achievement Test was used. (It was available in the students' records.)


Variables Related to Students' Self-perceptions

Huge quantities of resources are being invested in work-study programs from eighth grade through high school (Dykeman, 1979). How helpful is the work-study program for the 14 and 15 year old junior high school student who is

already undergoing "identity confusion," a reflection of a general redefinition of self, of others, and of the world which takes place in the early adolescent (Bailey & Bailey, 1974, p. 210)? The self-perceptions held by the student of how intelligent the student felt himself/herself to be were found to be very important by Nuttall (1972).

The single most important trait related to academic performance was found to be intelligence both actual and self-perceived. . . . The school should routinely monitor or how intelligent the child thinks he/she is.
(Nuttall, 1972, p. 87)

Self-perceptions of scholastic ability were found related to academic achievement in studies by Bailey and







23

Bailey (1974) who confirmed Nuttall's (1972) finding. Using 341 students from different grade levels, Bailey and Bailey (1974) found a shift in self-perceptions of scholastic ability--from low at eighth grade to high at 12th grade. At the 8th grade level, the males overrated their actual ability, while females tended to underrate their actual ability. The 12th grade students seemed to have the most realistic view of their ability. High school students not only revealed the most stable self-perceptions of the groups studied, but the most realistic self-views (p.211).

The relationship among home background, school achievement, and adolescent values (defined more broadly than self-perceptions of intelligence and including values related to work, satisfaction, self-worth, and morality) was studied by Kobett (1979). Here again, overall grade point average was used as a measure of school success. One of the strongest conclusions resulting was that "there is no evident relationship between adolescent values and school achievement" (p. 162).

The interests of students as they relate to both working part-time and academic achievement were central to the work of Gill (1980) who said, "What students are most and least interested in is certainly a significant variable in the educational process" (p. 160). More specifically,

the interest of the student in school, peace, love, life, and the opposite sex may be very significant (Gill, 1980, p. 162).







24

Research such as that cited provides support for the notion that students' self-perceptions of intelligence versus measured intelligence are legitimate points to consider in a study of the relationship of part-time work to academic achievement. Measures of adolescent values orientation seemed to have no relation to academic achievement. Even though interests and related value system may be related to academic achievement, since the study reported herein was based primarily on records, it was assumed that the attributes were randomly distributed in the population.


Other Relevant Variables

Student attitudes toward school and work, teacher sex and student sex interactions, and extent of participation in extracurricular school activities have been examined as relevant variables.

In a survey of 1248 high school students in North and Central Georgia, their attitudes and their future in the world of work were probed. Their attitudes about the need for freedom and independence in their future work and their feelings concerning certain conditions under which they would want to work, as well as their philosophy in general, were sampled and analyzed. Their attitudes, as shown by the correlation of their perception of the need for achievement in the classroom and success on the job, were not deemed to be very positive. "These students were,







25

perhaps, recalling some historical examples of this discrepancy: U.S. Grant, Wagner, Edison, Churchhill, von Braun, etc." (Schab, 1979, p. 603).

The influence of the teacher and the interactions of some teacher characteristics and some student characteristics have been researched. Is the gender of the teacher, that is interacting with boys or girls, a significant factor to include in the influence of work on academic achievement? Teacher's sex and student's sex interactions were studied by the Thomas L. Good Center for Research in Social Behavior. The differing levels of achievement in the classroom of boys and girls in junior high school classes were investigated in two male and two female teacher's classes. Three-way analysis of variance was made using teacher sex, student sex, and subject matter. "Inspection of the variables involved confirmed the conclusion that sex bias [of the teacher] was not a factor affecting teacher-student interactions in these classrooms" (Good et al. 1973, p. 78). Sex differences were found in classroom interaction patterns to be mostly due to student sex and not the sex of the teacher. Teachers were found to be primarily reactive in their teacher role to the different interactive practices that boys and girls present to the teacher. The teacher's role, as the adult in charge of the class, and not the teacher's sex was found to be the most significant factor. The sex of the student was significant. The quality of classroom life, as reflected in achievement, for high-achieving boys was found to be vastly different






26

from the experience that low-achieving boys had. The difference between high and low girls was found to be much less than the difference between high and low boys (Good et al. 1973, p. 85-86).

The relevance of students' participation in school activities as an influence on working and academic achievement has been viewed by some as significant in determining the quality of high school student life (Yarworth & Gauthier, 1978, p. 336). The related argument holds that high levels of participation in the "third curriculum" (i.e., cocurricular activities) are associated with high academic achievement. The alternative, inability to participate in the "third curriculum" because of working, has been assumed by some authors to result in lowered academic achievement (Williams, 1967; Young, 1980).

The "normal" participation of students in school activities was tentatively established and measured by Joseph S. Yarworth and William J. Gauthier, Jr. (1978) in a study of 459 high school students drawn from five Pennsylvania high schools. The relationship between various aspects of student self-concept and student participation in the extra and cocurricular activity programs of several Pennsylvania high schools was explored. It combined psychological with personal variables in its examination of student participation in the extra and cocurricular activity programs, both athletic and non-athletic. Some findings of the study were as follows:







27

Not only was there a difference in the self-concept scores of high-and-low-frequency participators, but this difference is significant for all categories of student activities. Large numbers of students do not participate in the extra and cocurricular activities of the school. The study established the importance of selfconcept in the relationship between academic achievement and participation not only in athletics but in nonathletic activities and the school activity program as a whole. The study dispelled the myth that school activities appeal equally to every student and that school activities are used by large numbers of students to complete the high school experience. (Yarworth &
Gauthier, 1978, p. 342)

Greater participation in school activities might relate to academic high achievers in some ways (Yarworth & Gauthier, 1978, p. 335). Furthermore, "the lack of participation in school activities is a significant characteristic of the dropout" (Bell, 1967, p. 251). Participation in cocurricular activities cannot be inferred to be a causative or a regulatory agent for high academic achievement (Yarworth & Gauthier, 1978).

The attitudes of students toward work or academic achievement are recognized as being influential. A survey of students' attitudes was beyond the scope of the study reported herein. However, in addition to the previously noted sex variable a variable relative to extracurricular school activities was used.

The possibility of the confounding effects of chronological sequence, interval, span of consideration, and year group (or generation) comparison was researched by Ferguson and Maxey (1978), who studied the trends in the academic performance of high school and college students. They found a substantial increase in the grades awarded by high






28

school and college faculty over a 10-year period, 19661976. The ACT test scores for college freshmen were concurrently on the decline. They concluded that academic achievement improvement, as measured by grade point average differential over time, may show gains based on relaxing of harsh grading standards rather than factors that would be relevant if a shorter interval of comparison was used (Ferguson & Maxey, 1978, p. 509).

Other factors affecting academic achievement of students have been grouped by Charles E. Bidwell and John D. Kasarda (1980). They noted:

In short, in attainment models where individual student achievement is at issue, individual student attributes, other nonschool inputs to schooling, and school inputs must be measured at the individual level to avoid
misleading results. (p. 426)

The influence of both "schooling," the process through which instruction occurs, a structure of actions by students and teachers, and "school," the organization that conducts instruction and distributes school resources to individual students, are recognized as having recently been shown to have significant influence on academic achievement of individuals (Bidwell & Kasarda, 1980, p. 404).

To control the possible source of error of shifting grade point averages over time owing to a possible relaxing of grading standards, the study reported herein was kept to the minimum possible span of years. For example, data from all included grade levels was taken for the same consecutive calendar year time periods, as compared to stretching the







29

study over a 10-year period. The relationship of the rather

global concepts of "school" and "schooling" to academic achievement was considered too complex to introduce at this early stage of research relative to academic achievement and part-time employment.














CHAPTER III
PRESENTATION AND ANALYSIS OF DATA


As repeatedly noted, concern as to the influence of students' part-time work on schooling has been expressed in many newspapers, magazines, and professional journals. Little research has been directed toward this concern. The 34,340 students who were enrolled in grades 8 through 12 (junior and senior high school) in the Duval County School System for the 1981-82 school year constituted the target population for the study of the relationship between employment status and academic achievement among students in an urban school district. A grade level stratified random sample of students was selected from this population. Using a researcher-designed instrument and the permanent record files of the school district, the needed data regarding the independent variables and dependent variable were collected. Complete usable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students. Presented in the present chapter are descriptive data by grade relative to students involved in the study, correlation matrices, and the results of the multiple regression analysis of the data.







31

Descriptive Data by Grade Level
Relative to Students Involved In the Study

Table 1 contains descriptive data in regard to the

dependent variable and presumed independent variables about the students by grade level. More specifically, contained in Table 1 are data about the 1981-82 grade point average (GPA-82) i.e., the dependent variable: days present 1981-82, sex, participation in extracurricular activities, ethnic group, participation in a program for the socioeconomically deprived student (SES), Stanford Achievement

Test (SAT) , self-perception of intelligence (SPIN) , grade point average 1980-81 (GPA-81), employment status (Model

I-worked yes/no 1981-82), and employment status (Model II-category of hours worked) which were the presumed independent variables.

As can be seen from the table, the GPA-82 of students ranged from a low of 1.9728 for grade 10 students to a high of 2.6967 for grade 12 students, within a range of 0 to 4.0

for grade 8 students to a range of .7 to 4.0 for grade 12 students. Days present in 1981-82 ranged from 91 to 180 days for grade 10 students to 148 to 180 days for grade 9 students. The range of mean attendances for all grade levels was 164.6222 to 169.2222 days. The sample according to sex varied from 60% female and 40% male for grade 11 students to 50% male and 50% female in grade 12.















Table 1. Descriptive Data by Grade Level
for Students Involved In The Study.

Variable Grade Level
8 9 10 11 12 n=106 n=90 n=92 n=90 n=92


GPA-82 (y) Range
Mean
Standard Deviation
Days Present 1981-82,(x 1)Range
Mean
Standard Deviation Sex (x2
Male
Female
Extracurricular Activities (x3)
Yes No
Ethnic Group (x4)
white, Non-Hispanic black, Non-Hispanic
Hispanic
American Indian/Alaskan Native
Asian/Pacific Islander
Participation in Program for Socioeconomically Deprived
(SES) (Xs)
Yes No
SAT Scores (x 6), Range
Mean
Standard Deviation Self-perception of
intelligence (x7), Range


Mean
Standard Deviation GPA-81 (x8) Mean
Standard Deviation Employment Status (x9
Model I Yes Model I No
Model II, category of Not Employed
1-4 hours per week


5-9
10-14 15-19
20-24 25-29
30-34


hours hours hours hours hours hours


week week
week week
week week


35 + hours per week


hrs. worked
(0) 79 (5) 5 (6) 11
(7) 5 (8) 3 (9) 1 (10) 1 (11) 1 (12) 0


0-4.0 3-4.0 .2-4.0 .5-4.0 .7-4.


2.1722 .8465 148-180 169.2222 12.8726


1.9728 .8581 91-180 166.8587
13.4406


2.1911 .7255 106-180 164.6222
14.5892


2.6967 .7152 139-180 166.2609 9.7607


2.0311 .9425 99-180
165.9716 20.4657


50 (46%) 46 (49%) 44 (48%) 36 (40%) 46 (50%) 56 (54%) 54 (51%) 48 (52%) 54 (60%) 46 (50%)

67 (63%) 63 (70%) 48 (52%) 46 (51%) 63 (69%) 39 (37%) 27 (30%) 44 (48%) 44 (49%) 29 (31%)


67 (63%) 38 (36%)
0
1 (1%)
0



27 (25%) 79 (75%)
0-98 53.5110
25.6256

1-3*
2.2547
.5175
0-4.0
2.2287
1.0260


55 (61%) 32 (36%) 1 (1%)
0
2 (2%)



25 (28%) 65 (72%)
0-94 46.5444 27.3763

1-3*
2.2444
.5041
.1-4.0
2.0911
.8069


(64%) (34%)
(1%) 1 (1%)


16 (17%) 76 (83%)
4-99
51.8913 23.9598


1-3*
2.1630
.4753 0-3.9
2.0586 .9260


66 (74%) 21 (23%)
1 (1%)
0
2 (2%)



10 (11%) 80 (89%)
9-97 55.9111
21.0418


1-3*
2.2111
.4369 .6-4.0
2.1711
.7020


(67%) (32%)
(1%)


12 (13%) 70 (87%)
6-98 57.5000
24.5041


1-3*
2.3152
.4901
.7-4.0
2.4826 .7129


27 (25%) 19 (21%) 25 (27%) 33 (37%) 59 (64%) 79 (75%) 71 (79%) 67 (73%) 56 (63%) 33 (36%)


(75%)
(5%) (10%)
(5%)
(2%)
(1%)
(1%)
(1%)


(79%)
(8%)
(6%)
(1%)
(2%)
(1%)
(1%)

(2%)


(73%)
(6%)
(3%) (10%)
(1%)
(4%)
(1%)
(1%)
(1%)


(63%) (4.5%)
(1%)
(3%)
(8%)
(8%)
(4%)
(4%)
(2%)


(36%)
(3%)
(4%)
(1%) (20%) (12%)
(7%)
(5%) 12%)


2 means average, 3 means above average


0


*1 means below average,






33

The highest participation in extracurricular activities was found to be in grade 9 students with 70% participation in one or more extracurricular activities. The lowest level of participation was for grade 11 students with 51% participation in extracurricular activities.

All ethnic groups were represented, although not in each grade level. Hispanics were not included in the grade 8 group. However, a single American Indian/Alaskan Native was included in grade 8.

The indicator of lower socioeconomic status, participation in reduced price or free lunch program, showed highest participation in grades 8 and 9 with 25% and 28% respectively. Grades 10, 11, and 12 had participation rates of 17%, 11%, and 13% respectively.

The average SAT percentile scores for the stratified samples were 53.5000 for grade 8, 46.5444 for grade 9, 51.8913 for grade 10, 55.9111 for grade 11, and 57.5000 for grade 12 students. Only grade 9 students fell below the national norm.

Students self-perception of intelligence was scaled as 1 of below average intelligence, 2 of average intelligence, and 3 of above average intelligence. The grade level mean scores of 2.2547 for grade 8 students, 2.2444 for grade 9 students, 2.1630 for grade 10 students, 2.2111 for grade 11 students, and 2.3152 for grade 12 students show that the grade level groups of students perceived themselves generally to be slightly above average intelligence. When the SAT scores are compared to self-perception of intelligence,







34

except for grade 9, this opinion cannot be contested.

Grade point average 1980-81 (GPA-81) had a range from 0.0 to 4.0. The means were 2.2287 for grade 8 students, 2.0911 for grade 9 students, 2.0586 for grade 10 students, 2.1711 for grade 11 students, and 2.4826 for grade 12 students. These means show the academic achievement level for their previous school year to be slightly above for all grade groups.

The data about Model I employment status (a dichotomous yes/no response) indicated that 25% of grade 8 students worked more than one-half of school year 1981-82, 21% of grade 9 students worked more than one-half of school year 1981-82, 27% of grade 10 students worked more than one-half of school year 1981-82, 37% of grade 11 students worked more than one-half of school year 1981-82, and 64% of grade 12 students worked more than one-half of school year 1981-82. Model II (the number of hours worked each week by students) showed that of those students who worked in grade 8, 10% worked 5 - 9 hours per week and only 1% worked as much as 30 - 34 hours per week. The greatest percent (8%) of grade 9 students who worked, worked, only 1 - 4 hours per week; 5% worked 5 - 9 hours per week, 2% worked 15 - 19 hours per week, 1% worked 20 - 24 hours per week, 1% worked 30 - 34 hours per week, and 2% worked 35 or more hours per week. The proportion of grade 10 students who worked 1 - 4 hours per week was 6%, 5 - 9 hours per week was 3%, 10 - 14 hours per week was 10%, 15 - 19 hours per week was 1%, 20 24 hours per week was 4%, 25 - 29 hours per week was 1%,







35

30 - 34 hours per week was 1%, and 35 or more hours per week was 1%. Four grade 11 students (4%) worked 1 - 4 hours per week, 1 student worked 5 - 9 hours per week, 3 students

(3%) worked 10 - 14 hours per week, 8 students (8%) were found in each category of 15 - 19 hours per week and 20 - 24 hours per week, 4 students, (4%) worked in each category 25 - 29 hours per week and 30 - 34 hours per week, and 2 students (2%) worked more than 35 hours per week. Grade 12 had the highest overall percentage of students who worked part-time --64%. Thus, 36% of the grade 12 students did not work part-time. Three percent of the grade 12 students worked part-time for 1 - 4 hours per week for more than one-half of the school year, 4% worked 5 - 9 hours per week, 1% worked 10 - 14 hours per week, 20% worked 15 - 19 hours per week, 12% worked 20 - 24 hours per week, 7% worked 25 29 hours per week, 5% worked 30 - 34 hours per week, and 12% worked part-time 35 or more hours per week.

Summarizing Model II (categories of hours worked per week), grade 8 students ranged from 1 - 4 hours per week through 30 - 34 hours per week, with a mean of less than 4 hours per week. Grade 9 students had the same range with a mean of less than 4 hours per week. Grade 10 students ranged from 1 - 4 hours per week through 35 hours or more with a mean of less than 4 hours per week. Grade 11 had the same range and a group mean below 4 hours per week. Grade 12 students also had the same range with the group mean being between 4 and 5 hours per week of part-time employment. The percentage of students who did work were working







36

sufficiently longer hours per week to raise the mean for the entire group to more than 4 hours per week.


Correlation Matrices

In order for the reader to have a better understanding of the impact of the several presumed independent variables on the presumed dependent variable, in this section the correlation matrices among the independent variables and between each of the independent variables and the presumed dependent variable are presented. By analyzing these correlations between the several variables the reader can understand the amount of "overlap" among the variables. If the correlations are high the different variables are measuring somewhat the same thing; if they are low then different things are being measured. Ideally, when dealing with predictor variables in regression analysis low correlations are desired.

In Tables 2 through 11 the correlations within each grade level are identical except as they involve the employment status (x9) , of Model I or Model II and the dependent variable (y). The negative values of some correlations are readily seen herein but are not so apparent when observing the multiple regression data. The greatest negative correlation (-.3113) in grade 8 independent variables occurred between ethnic group and Stanford Achievement Test (SAT) percentile scores. The greatest positive independent, variable correlation (.5907) occurred between SAT percentile scores and previous grade point average (GPA-81).










Table

Variables


Days Present x1
1982
x2
Sex

Extracurricular x3
Activities
x4
Ethnic Group Socioeconomic x5
Status (SES)

SAT Percentile x6 Score

Self-perceived x7
Intelliqence

GPA 1981 x8 Employment Status 1982 x9 Model I yes/no


GPA 1982 y


2. Correlation Matrix for Eighth Grade Group, Model I

xl x2 x3 �4 x5 x6 x7 x8


1.0000 -.0972 -.0193 -.1765 -.0279 .2313 .1266 .2717


1.0000 .0011


1.0000


.1398


.0855 1.0000


-.0643


-.1826 .2443 1.0000


-.0530 .0633


-.3113


-.1889 1.0000


9


-.1416


-.0544 .0888 -.1512 .0735 .1211 .0419 .1044 -.1215 -.1584 .0892 -.2698 .0558 .3515 .5907 .0098 1.0000 .2694 -.1629

1.0000 -.1882


7


.4007


.0197 .1442 .0545


-.2202 .5504 .4014 7813


1.0000 -.2248


1.0000










Table 3. Correlation Matrix for Eighth Grade Group, Model II

Variables x1 x2 x3 x4 x5 x6 x7 x8 9


Days Present xI 1.0000 -.0972 .0193 -.1765 -.0279 .2313 .1266 .2717 -.1690 .4007
1982


Sex

Extracurricular x3 Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 x9 Model II Category


GPA 1982 y


1.0000 .0011 .1397 -.0643


1.0000 .0855 .1826


1.0000 .2443 1.0000


-.0530 -.0544 .0633 .0735 .3112 .1044


-.1888 .0892 1.0000 .3515


1.0000


.0888 -.1148 .1211 .0340


-.1215 -.1535


-.2698 .0444 .5907 .0207 .2694 -.1700 1.0000 -.1871


1.0000


.0197 .1442
CO

-.0545


-.2202 .5504 .4014


.7813


-.2142 1.0000










Table 4. Correlation Matrix

Variables x1 x2 x3

Days Present x 1.0000 .0500 .1061
1982


for Ninth Grade Group, Model I x4 x5 x6 x7 x8 .1416 .0675 .0872 .1958 .4813


Sex

Extracurricular
Activities


Ethnic Group Socioeconomic
Status

SAT Percentile Score
Self-perceived
Intelligence GPA 1981 Employment Status 1982 Model I yes/no


GPA 1982


1.0000 .0388


1.0000


9

-.1280


.4156


.1071 .1852 1.0000


.0839 .0963 .3633 1.0000


-.1339


-.1116


-.1257


-.1784 1.0000


-.0108 .3675


-.0080 .0804 .1758 1.0000


.0806


.1256


- .0003


-.1487


.5500 .3202 1.0000


-.0931 .0416


-.1807 .0307 .0647 .0193 .0498 1.0000


.1103 .1483 .0299


-.0808 .4944 .3188 .7801 .0235 1.0000










Table

Variables


Days Present x1 1982


Sex

Extracurricular x3
Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 x9 Model II Category


GPA 1982 y


5. Correlation Matrix for Ninth Grade Group, Model II x 1 x2 x3 x4 x5 x6 x7 x8


1.0000 .0500


1.0000


.1060 .0388 1.0000


.1416 .1071 .1851 1.0000


.0675 .0839 .0963 .3632 1.0000


.0872


-.1339


-.1116


-.1257


-.1784 1.0000


.1958


-.0108 .3675


-.0080 .0804 .1758 1.0000


.4813 .0806 .1256 .0391


-.1487 .5500 .3202 1.0000


-.1553


-.0960


-.0137


-.1825


-.0396 .0931


-.0444 .0307 1.0000


.4155 .1103 .1483 .0299


-.0808 .4944 .3188 .7801 .0358 1.0000










Table 6. Correlation Matrix for Variables x1 x2 x3 x4


Tenth Grade Group, Model I x5 x6 x7 x8 9


Days Present x1
1982


Sex 2 Extracurricular x3
Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x
Intelligence

GPA 1981 x Employment Status 1982 Model I x9 yes/no


GPA 1982 V


1.0000 .0819 .0534 .1706


1.0000


-.1514 1.0000


-.0560 .0076 1.0000


.0928


-.1048 .0374 .5388 1.0000


.1132 .0756 .1554


-.4856


-.4588


1.0000


.0793


-.1075 .1921


-.0688


-.1582 .2602 1.0000


.1576 .0722 .1673


-.2156


-.2720 .5996 .3200 1.0000


.0210


-.3044 .1935


-.0306 .0420


-.0495 .1511


-.0628 1.0000


.3143 .0413 .1709


-.0982


-.2172 .6197 .4312 .73131


-.1179 1.0000










Table

Variables Days Present x1
1982


Sex

Extracurricular x3
Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 Model II x9 Category


GPA 1982 y


7. Correlation Matrix xI 1 x2 x 3


l.'0000 .0819 .0534


1.0000 -.1514


1.0000


for Tenth Grade Group, Model II x 4 x5 x6 x7 x8


.1706 .0559 .0075 .0000


.0928


-.1047 .0374 .5388 1.0000


.1132 .0756 .1554


-.4856


-.4588 1.0000


.0793


-.1075 .1921


-.0688


-.1583 .2602 1.0000


.1576 .0722 .1673


-.2156


-.2721 .5996 .3200 1.0000


.0157


-.2889 .1888


.0372 .1185


-.1216


.1088


-.1599 1.0000


.3143 .0413 .1709


-.0982


-.2172 .6197 .4313 .7313


-.1930 1.0000










Table

Variables


Days Present x1
1982


Sex

Extracurricular x3
Activities
x4
Ethnic Group Socioeconomic x5
Status

SAT Percentile 6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 Model I x9 yes/no


GPA 1982 y


8. Correlation Matrix for Eleventh Grade Group, Model I

xI x2 x3 x4 x5 x6 x7 x8


1.0000 .0525


1.0000


.1553


-.0635 1.0000


.2260


-.0759 .0668 1.0000


.1700


-.0722


-.0786 .3723 1.0000


.0030 .0642 .1764


-.0283


-.2198 1.0000


.0896


-.0313 .1683 .1588


-.0904 .5777 1.0000


.1350 .1377 .1410 .0443


-.0816 .5990 .3497 1.0000


-.2088 .0187 .0285


-.2463


-.1296 .1304 .0094 .0224 1.0000


.2973 .0383 .1481 .0617 .0485 .4696 .2151 .6553


-.1492 1.0000










Table

Variables


Days Present x1
1982


Sex

Extracurricular x3
Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status Model II x9 Category


GPA 1982 y


9. Correlation Matrix for Eleventh Grade x1 x2 x3 x4 x5 x6


1.0000 .0525


1.0000


.1553


-.0635 1.0000


.2260


-.0759 .0668 1.0000


.1700


-.0721


-.0786 .3723 1.0000


.0030 .0642


.1764


-.0283


-.2198 1.0000


Group, x7


-.0896


-.0313 .1683 .1588


-.0904 .5777 1.0000


Model II x8


.1350 .1377 .1410 .0443


-.0815 .5990 .3498 1.0000


-.2415 .0452 .0323


-.2522


-.1338


.0636


-.0915


-.0328 1.0000


.2973 .0383 .1481 .0617 .0484 .4696 .2151 .6553


-.1632 1.0000










Table

Variables Days Present x1
1982


Sex 2 Extracurricular x3
Activities
x4
Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score

Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 Model I x9 yes/no


GPA 1982 y


10. Correlation Matrix for Twelfth xI x2 x3 x4 x5


1.0000 -.1097 .3098 .04729 -.0636


1.0000 .0945 -.0220 -.1291


1.0000 .0685 -.0945 1.0000 .3883


1.0000


Grade Group, Model I x 6 x 7 x 8


.2639 .1848 .2603


-.0232 -.1115 .1318 .4080 .3308 .3536


-.4563 .0103 -.2186


-.3086 .0144 -.0224 1.0000 .4269 .4497


1.0000 .4216


1.0000


9


-.1153 .0680 .0471


-.3607


-.2487 .1520


-.0743


-.0120 1.0000


9


.3492 .0688 .3157


-.1975


-.0663 .5583 .4419 .6365


-.0735 1.0000









Table

Variables Days Present x1
1982
x2
Sex

Extracurricular x3
Activities


Ethnic Group Socioeconomic x5
Status

SAT Percentile x6 Score
Self-perceived x7
Intelligence

GPA 1981 x8 Employment Status 1982 Model II x9 yes/no


GPA 1982 y


11. Corre

x1


1.0000


lation Matrix for Twelfth Grade Group, Model II x2 x3 x4 x5 x6 x7 x8 9


-.1097 .3097 .0473 -.0636 .2639 .1848 .2603 -.0823 .3493


1.0000 .0945 -.0220 -.1291 -.0232 -.1115 .1318 .0094 .0688


1.0000 .0685 -.0945 .4080 .3308 .3536 .0873 .3158


1.0000 .3883 -.4563 .0103 -.2186 -.3249 -.1975


1.0000 -.3086 -.0144 -.0224 -.2917 -.0662


1.0000 .4268 .4496 .1745 .5583 1.0000 .4216 -.0190 .4419 1.0000 .0093 .6365


1.0000 -.0289


1.0000







47

These data hold true for both the Model I and Model II tables. The greatest negative correlation for grade 8 students between an independent variable and the dependent variable (GPA-82) occurred between Model I employment status and GPA-82 with a correlation of -.2248. The greatest positive correlation for grade 8 students between independent variables and the dependent variable (GPA-82) occurred between GPA-81 and GPA-82 with a positive correlation of .7813. As shown in Table 3, when categories of employment status (Model II) were incorporated into the matrix, relationships between ethnic group and SAT scores and between SAT scores and GPA-81 held constant; SES replaced employment status as the most negative relationship with the dependent variable. Whether or not these relationships are significant at the .05 level in the multiple regression data is determined by the F ratios established from consideration of all variables interacting among themselves and the sample size.

It must be noted that 1982 employment status, whether Model I or Model II, had a negative correlation with GPA-82, except for the grade 9 students. The correlation being most negative (-.2248) was for Model I employment status for the grade 8 students and least negative (-.0289) was Model II employment status for the grade 12 students.

The Results of Multiple
Regression Analysis

The reader will recall the basic focus of the study was to determine the relative impact of employment status








on students' grade point average when the other variables which are normally considered to contribute to academic achievement have been taken into account. In order to accomplish this purpose a regression analysis was used. In the regression analysis used the variables, with the exception of employment status, were permitted by the computer program to enter into the regression analysis in the order in which they made a contribution to the variation in the presumed dependent variable, i.e., grade point average. Once all these had been entered, employment status was entered to determine the extent to which it made a contribution over and above the contributions made by the other independent variables utilized in this study. The results of the multiple regression analysis for each of the grade levels using both whether the student worked or not and the number of hours worked, i.e., Model I and Model II, are presented in Tables 12 through 16.

In Tables 12 through 16 the variables are listed in order of their contribution to GPA-82 (the dependent variable - ^). The contributory relationship of employment in the equation is portrayed as an individual line of data. That is, Model I and Model II employment status should be interpreted from the tables as separate and not as cumulative variables.

Only in grades 10 and 11 did employment status make a significant contribution at the .05 level to the variation in grades (GPA-82). Model I and Model II had a significant relationship for grade 10 students. Only Model I had a










Multiple Regression Data for Eighth Grade Students


R2 proportion of
variance in
Multiple dependent variable Variables x R Correlation accounted for by variables
Grade Point x8
Average 1981 .7813 .6104 * Self-perceived x7
Intelligence .8006 .6498 * Days Present x1
1982 .8270 .6839 * Socioeconomic x5
Status 8286 .6866

Ethnic Group .8304 .6896 SAT Percentile x6
Score .8330 .6940 Extracurric- x3
ular Activity .8338 .6953

Sex .8340 .6956 Employment Sta- x9
tus 1982 Mod. I .8351 .6975 Employment Sta- x
tus 1982 Mod.II .8344 .6963

* Variables remaining when those contributing at less


Cumulative Stand. R2 2Error of Change R Estimate


.6104 .0614 .0341 .0025 .0030 .0026 .0013 .0013 .0019 .0007 than the


.6104 .5910 .6498 .5631 .6839 .5376 .6866 .5379 .6896 .5380 .6940 .5369 .6953 .5385 .6956 .5410 .6975 .5420 .6963 " .5413

.05 level have been removed.


Table 12








Table 13


Variables

Grade Point Average 1981 SAT Percentile Score
Self-perceived
Intelligence


Sex
Days Present x1
1982
Extracurric- x3
ular Activity
Socioeconomic x5
Status

Ethnic Group Employment Sta- x9 tus 1982 Mod. I Employment Sta- x9 tus 1982 Mod.II

*Variable remaining


Multiple Regression Data for Ninth 6rade Students

R2 proportion of
variance in Cumulative Multiple dependent variable R C a R Correlation accounted for by Change R variables

.7801 .6086 * .6086 .6086 .7840 .6147 .0061 .6147 .7874 .6200 .0053 .6200 .7903 .6247 .0047 .6247 .7930 .6290 .0043 .6390 7945 .6313 .0023 .6313 .7947 .6317 .0004 .6317 .7948 .6318 .0001 .6318 .7949 .6319 .0001 .6319 .7952 .6324 .0006 .6324 when those contributing less than the .05 level have been


Stand. Error of Estimate


.5326

.5314 .5309 .5306 .5307 .5322 .5352 .5384

.5416 .5413 removed.








Table 14 Multiple Regression Data for Tenth Grade Students


Variables Grade Point Average 1981 SAT Percentile Scores
Days Present
1982
Self-perceived
Intelligence

Ethnic Group Extracurricular Activity

Sex
Socioeconomic
Status
Employment Status 1982 Mod. I Employment Status 1982 Mod.II


Multiple R Correlation


.7312 .7655 .7903 .8111

.8218 .8219 .8219 .8219 .8287 .8290


R2 proportion of
variance in
dependent variable
accounted for by
variables

.5347 * .5860 * .6246 * .6580 * .6755 *

.6756 ** .6756 ** .6756 **

.6868 *

.6884 *


Cumulative
R 2


R 2
Change .5347 .0513 .0386 .0254 .0175 .0000 .0000 .0000 .0112 .0128


.5347 .5860 .6246 .6580 .6755 .6756 .6756 .6756 .6868 .6884


Stand. Error of Estimate


.5886 .5583 .5347 .5133 .5029 .5057 .5087 .5118 .5059 .5047


* Variables remaining after those contributing at less than the .05 level have been removed.
** Carried to the fifth place Extracurricular Activity, 2
Sex and Socioeconomic Status accounted for only .00001 R change.







Table 15 Multiple Regression Data for Eleventh Grade Students

R2 proportion of
variance in 2 Cumulative Stand. Multiple dependent variable R 2 Error of Variables x R Correlation accounted for by Change R Estimate variables
Grade P1oint x 8
Average 1981 .6552 .4294 * .4294 .4294 .5518

Days Present x1
1982 .6883 .4738 * .0444 .4738 .5523

SAT Percentile x6
Scores 6983 .4877 * .0139 .4877 .5283

Socioeconomic x 5
Status .7041 .4958 .0081 .4958 .5272

Self-perceived x7
Intelligence .7061 .4986 .0028 .4986 .5289

Sex2 .7082 .5016 .0030 .5016 .5304

Ethnic Group .7090 .5028 .0002 .5028 .5330

Extracurric- x3
ular Activity .7093 .5032 .0004 .5032 .5360

Employment Sta- x9
tus 1982 Mod. I .7248 .5253 * .0221 .5253 .5272

Employment Sta- x9 .5321 tus 1982 Mod.II .7187 .5165 .0133 .5165
* Variables remaining after those contributing at less than the .05 levelhave been removed








Table 16 Multiple Regression Data for Twelfth Grade Students

R2 proportion of
variance in Cumulative Stand. Multiple dependent variable R2 Error of Variables xn R Correlation accounted for by Change R Estimate variable
Grade Point x8
Average.6364 .4051 * .4051 .4051 .5547 SAT Percentile x6
Scores .7056 .4980 * .0849 .4980 .5124 Days Present x1
1982 .7195 .5177 * .0197 .5177 .5050 Self-perceived x7
Intelligence .7266 .5279 .0123 .5279 .5025

Sex .7283 .5304 .0025 .5304 .5041 Socioeconomic x5
Status .7300 .5328 .0003 .5328 .5058 Extracurric- x3
ular Activity .7312 .5347 .0019 .5347 .5078
x4
Ethnic Group .7314 .5349 .0002 .5349 .5106 Employment Sta- x9
tus 1982 Mod. I .7359 .5415 .0076 .5415 .5101 Employment Sta- x9
tus 1982 Mod.II .7333 .5378 .0029 .5378 .5122

*Variables remaining after those contributing at less than the .05 level have been removed.






54

significant relationship for grade 11 students. In grades, 8, 9 and 12 neither Models I or II employment status made a significant contribution to the variation in GPA-82. Grade Point Average 1981 (GPA-81) was significant at all grade levels accounting for .4051 of the GPA-82 variance at twelfth grade to .6104 at the eighth grade.

SAT percentile scores were the next most influential variable for grade 9, 10, and 12 students. Self-perceived intelligence was second for grade 8 students. Days present 1982 was second for grade 11 students. Days present was statistically significant at the .05 level for grade 8, 10, 11 and 12 students. Only in grade 10 was ethnic group of significant statistical influence.














CHAPTER IV
SUMMARY, CONCLUSIONS, AND DISCUSSION


Summary

The problem of the study was to determine for students in each grade level, grades 8 through 12, in a single urban school district the relative impact of part-time employment to their 1981-82 grade point average (GPA) after the contributions of sex, ethnic group, socioeconomic status, school attendance, GPA for previous school year (1980-81), self-perceptions of intelligence, extracurricular activity participation, and SAT overall percentile score were taken into account. Specifically, the following steps were taken:

1. A determination was made of the relationship

between the independent variables of GPA 1980-81, sex, ethnic group, socioeconomic status, school attendance, self-perceptions of intelligence, extracurricular activity participation, and SAT percentile scores and the dependent variable GPA

1981-82.

2. After determination of the relationship of the

above listed independent variables and the dependent variable, the independent dichotomous variable









of student part-time employment (yes or no-Model I) was introduced and its contribution to the relationship, if any, was determined.

3. After determination of the relationship of the

independent variables listed in #1 to the dependent variable, except part-time employment, the extent of part-time employment (number of hours worked-Model II) was introduced in place of Model I and its contribution to the relationship, if any, was

determined.

The study was confined to those students in a single urban school district for which complete usable data could be obtained. More specifically, complete usable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students for a total of 470 students. Data in regard to 1981-82 GPA, school attendance, sex, ethnic group, SAT scores, SES, and 1980-81 GPA were obtained from the Duval County School Board SIMS computer data base and student cumulative record files. All data were verified with the student permanent record files maintained at the school that the student was attending.

The data source for information pertaining to hours worked, extracurricular activity participation, self-perception of intelligence, and participation in free or reduced price lunch program was a researcher-designed student information questionnaire.







57

To determine the contribution of the several presumed independent variables to grades multiple regression analysis was used. The nature of the multiple regression was stepwise for all the variables except the employment variable. For each grade level two multiple regression equations were run, one using employed or not (yes or no-Model I), the other one employment status by category of hours worked (Model II).

The output of the analysis included means and standard deviations for each variable, correlation coefficients among the variables, the proportion of the variance accounted for by each of the entire models, the proportion accounted for by each variable in each of the models, and the standard error of estimate.

The major findings emerging from the data analysis were as follows:

1. In regard to the descriptive data collected

from the students, it was found that the 1981-82 grade point average of students ranged throughout the entire scale of 0.0000 to 4.0000. The means were slightly above the theoretical average (2.0000) for all grades except grade 9 where it was slightly below the theoretical average with a mean of 1.9728. Days present ranged from a minimum 91 for grade 10 to a minimum of 147 days for grade 9 students. The maximum attendance of 180 days was achieved by all grade groups. The sample according







58

to sex varied from 60% female and 40% male for grade 11 students to 50% male and 50% female in grade 12. Participation in extracurricular activities was found to range from 70% participation in one or more activities in grade 9 to a low of 51% participation in one or more activities in grade 11. All ethnic groups were represented in the sample. Socioeconomic status (SES) was indicated by whether or not the student participated in the reduced price or free lunch program. This participation ranged from 11% in grade 11 to 28% in grade 9. The mean Stanford Achievement Test (SAT) overall percentile scores ranged from 46.5444 for grade 9 to 57.5000 for grade 12; only grade 9 students fell below the national norm. Selfperception of intelligence (SPIN) was scaled, 1 of below average intelligence, 2 of average intelligence, and 3 of above average intelligence. The grade level groups of students perceived themselves generally to be slightly above average intelligence. The 1980-81 grade point average ranged from 0.0000 to 4.0000. The range of grade group means was from 2.0511 for grade 10 to 2.4826 for grade 12. The proportion of students employed (Model I employment status) ranged from 21% in grade 9 to 64% in grade 12. In regard to the extent of employment (Model II employment status), it was found that 10% of grade 8 students worked 5-9 hours







59

per week, 8% of grade 9 students worked 1-4 hours per week, 10% of grade 10 students worked 10-14 hours per week, 8% of grade 11 students worked 15-19 hours per week, 8% of grade 11 students worked 20-24 hours per week, and 20% of grade 12 students worked 15-19 hours per week. (These percentages were the largest by category for each

grade level group.)

2. In regard to the correlations among the variables,

it was found that the correlations for grade 8 students ranged from a negative correlation of -.3113 between ethnic group and SAT overall percentile scores, to a correlation of .7813 between 1980-81 grade point average (GPA-81) and 1981-82 grade point average (GPA-82). Correlations for grade 9 students ranged from a negative correlation of -.1825 between ethnic group and Model II employment status to a correlation of .7801 between GPA-81 and GPA-82. Grade 10 correlations ranged from a negative correlation of -.4856 between

ethnic group and SAT percentile scores to a correlation of .7313 between GPA-81 and GPA-82. Grade 11 correlations ranged from a negative correlation

of -.2198 between SES and SAT percentile scores to a correlation of .6553 between GPA-81 and GPA-82.

Correlations for grade 12 ranged from a negative of -.4563 between ethnic group and SAT percentile







60

scores to a correlation of .6365 between GPA-81 and GPA-82. The correlations between employment status and 1981-82 grade point average were negative in value for all grades except grade 9, even though only in two instances were these significant at the

.05 level.

3. In regard to the findings emerging from the multiple regression data, the predictive value of independent variables, less part-time employment, varied greatly. The 1980-81 grade point average (GPA-81) was significant at all grade levels accounting for from about 41% of the variance in grades at the twelfth grade to about 61% at eighth grade. SAT overall percentile score was the second most influential variable for grades 10 and 12 students. Self-perception of intelligence (SPIN) was the second most influential variable for grade 8 students. Days present during 1982 was the second most influential variable for grade 11 students and it was statistically significant at the .05 level for grade 8, 10, 11, and 12 students.

Only in grade 10 was ethnic group of statistical significant influence. In regard to the predictive value of independent variables including Model I employment, only in grades 10 and 11 was part-time employment found to be statistically significant at the .05 level. Its contribution was negative and






61

was by far overshadowed by GPA-81, days present 1982, and SAT percentile scores. In regard to the predictive value of independent variables including

Model II employment (categories of hours worked) , only in grade 10 was it found to be statistically significant. Again, its contribution was negative and was by far overshadowed by GPA-81, SAT percentile scores, days present 1982, and SPIN.

4. In terms of the specific contributions of employment status to an increase in the percent of variance in the dependent variable (GPA-82), it was found that in grade 8 Model I employment contributed .0019% of a total of .6975% and Model II employment status contributed .0007% of a total of .6975%. In grade 9 Model I employment contributed .0001% of a total of .6319% and Model II employment status contributed .0006% of a total of .6324%. In grade 10 Model I employment contributed .0112% of a total of .6868%; Model II employment status contributed .0128% of a total of .6884%. In grade 11 Model I employment contributed .0221% of a total of .5253% and Model II employment status contributed .0133% of a total of .5165%. In grade 12 Model I employment contributed .0076% of a total of .5415% and Model II employment status contributed .0029%

of a total of .5378%.






62

Conclusions

In regard to the basic focus of the investigation, the extent to which part-time employment has an impact on the grade point average of high school students, the conclusion is that the impact of part-time employment on the academic achievement of the studied students is minimal. This conclusion seems to be justified, because in only very few instances was part-time employment found to make a significant contribution to the variation in grade point average. More specifically, in grade 10 it was found that it made a significant contribution in the sense that the students that worked made poorer grades, but this contribution followed the contributions of GPA-1981, SAT percentile scores, days present 1982, SPIN, and ethnic group and contributed less than 2% to the approximately 69% of the variance in grades accounted for by all of the presumed independent variables combined. In grade 11 it was found that part-time employment made a contribution after the contributions of GPA-81, days present 1982, and SAT percentile scores were determined. The contribution was about 2% of the approximately 53% of the variance in 1981-82 grades accounted for by all of the presumed independent variables. In the other instances studied there was no significant impact either positively or negatively. Thus, the conclusion is reached that it has minimal impact.

Even though not central to the basic focus of the investigation, the other major conclusion is that the best predictors of students' grade point averages are previous







63

grades, SAT percentile scores, and attendance. If one reviews the studies reported in Chapter II, it shows

that in numerous studies, e.g., Gade and Peterson (1980), Hammond (1970) , Bidwell and Kasarda (1980), Finn, Dulberg, and Reis (1979), Nuttall (1972), Rehberg, Schafer, and Sinclair (1970) , Bailey and Bailey (1974) , Kobett (1979) , Schab (1979), Good (1973), and Yarworth and Gauthier (1978), the importance of these factors has been noted. Thus, this conclusion, which is peripheral to the central purpose of this investigation, reconfirms the previous studies which lead to the generalization that these variables impact most heavily on grade point averages.


Discussion

The reader will recall that a justification offered for the study was that there was a great deal of controversy about whether, as a policy matter, students should be

encouraged to seek part-time employment during their secondary school years. Proponents of one view contended that grade point averages of teenage students in junior and senior high school who worked part-time were lower than those of similar students who did not work. That is, part-time work hindered academic achievement as measured by grade point average. Central to this point of view also was

the opinion that students required to work while attending school were denied active involvement in school-related activities. Justification for this belief rested on reports which suggested that involvement in cocurricular activities







64

enhanced essential social development and that the "values orientation" from such activities was lacking in students who worked while attending school. (Williams, 1967, p. 36; Young, 1980, D-3).

Advocates of the opposing view argued that part-time

employment prepared students to become self-sufficient wage-earners in society's work force. These educators held that students' work experiences helped them raise their

living standard and aided society by reducing unemployment and increasing the gross national product. A survey of

employers revealed that 68% favored youth work experience. Resultant benefits included development of maturity, selfidentity, self-reliance, career direction, work relationships, and work enjoyment. (Delaney, 1975, VI-14).

Family financial situations sometimes required students to work. Variables dictating the need to work have ranged from low family income to divorce, death of the family wage earner, or abandonment by one or both parents. In addition, students often worked to relieve peer pressure to participate in school-related activities that involved substantial financial commitments. Others sought employment as a convenient means to obtain luxury items. Student employment programs sponsored by government and private agencies to prevent the termination of education have had a further effect of encouraging students to work part-time during or after school hours (Delaney, 1975; Dykeman, 1979, p. 365; Heyneman, 1977; Levine, 1979).








If the major conclusion of this investigation is accepted as reasonable and valid, the implication for policy makers is that if the concern is with academic achievement there should be no policy discouraging students from working. If working is deemed necessary or desirable for other reasons, i.e., financial hardship, students might even be encouraged to seek a reasonable amount of part-time employment.

It is recognized that the conclusions of the investigation reported herein obviously must be regarded as suggestive, not conclusive, because only one urban school district was involved at only one point in time. However, since the sampling procedure ensured a random sample, one can feel fairly comfortable that the data were in fact applicable for that school district at that point in time. Furthermore, if the investigation was replicated and similar results were found, the implication for policymakers relative to the lack of clear linkage between part-time employment and grade point average would be strengthened.

Given the situation, when policy makers are faced with a decision of what to do about part-time employment, if action is taken to curb employment or strongly encourage employment, it has to be made on some other basis, other than impact in academic areas.

Another argument not germaine to the central focus of this study was, did part-time employment impact on the students participation in extracurricular activities? From








a review of the data collected for study, it can be determined that the correlations between employment status and extracurricular activities participation for each of the grade levels were as follows:
Grade 8 Model I .0419 Grade 8 Model II .0340 Grade 9 Model I .0416 Grade 9 Model II - .0137 Grade 10 Model I .1935 Grade 10 Model II .1888 Grade 11 Model I .0285 Grade 11 Model II .0323 Grade 12 Model I .0471 Grade 12 Model II .0873

The range of correlations from -.0137 in grade 9 Model II to .1935 in grade 10 Model I does not show that part-time employment had a drastic impact on extracurricular activities. Also, the extracurricular activities variable was not found to make a statistically significant contribution to the dependent variable (GPA-82) at any grade level. Further inspection of the correlation matrices will yield similar relationships for the other variables that were not found among the variables classified as significant.

In summary, consistent with the view expressed by Nuttal (1972), Rehberg, Schafer, and Sinclair (1970), the argument that poor students are adversely affected academically because of simply being poor and having to work was not confirmed by these data. If a secondary student is employed part-time, the other contributions to grade point average such as, previous years grades, attendance, SAT percentile scores, and self-perception of intelligence will,







67

percentile scores, and self-perception of intelligence will, according to the data of the study, far overshadow the impact of students working, or not working, part-time while attending secondary schools.










Add Cl,,,,.,. I). . lA5t NAMI


DUVAL COUNTY SCHOOL SYSTEM BIOGRAPHIC DATA FORM


N" .I1h


If~ III


COMPUTERUSE ONlY SECTION C FREE LUNCH STATUS
tAICH SEOUINC(


o o a a a a a B ~ . .. r,. ...h N..
i i I i i i i I Y... Ii...Ir. i , I ii .


SECTION 0 OTHER PROGRAMS
* 4 4 4 4 4 * a DIHFC IIONS S!,! i. mlt l i ...ll. . I... ili-iiiiCuiiii
I* I 7 % Y,* N, Yi:s No Yes Nu
a 4 6 6 4 6 6 4 1 % 9 1 7 7 1 7 7 7 1 2 6 10
* a a~ a 3 7 1
;t- M 2 t . t 4 It1FI


I I I I I I I I I I II I I I I
----------------------------


NlIllil NI IlIIIMll It


I I I I I I


N ,, ! I


AREA ENCLOSED BY A SOLID RED LINE IS TO BE COMPLETED BY A CLASSROOM TEACHER
OR GUIDANCE COUNSELOR
SECTION A DIRECTIONS. Answer Ouestions


AI OI0 . .TH


W I Ii,. i4 u1t J *' t I * * IS C

0Ii, t I..,, Mil. a a * It At,,ll" I |I l

J..
Hlap,ieIt gM.,v s 4 4


A ;., t,, 1I,s,, Jul 1 * 6
An.y 7 7
N~in

As,a,, or Oct up * P-~ 1,c Nuy I, Islam t I


Then go to Section B.
YES


1 The stuinlit is classified its .n eceptiotial education studemiI)


NO


2 The stuidlel is tiew to Duval Cotitily pi.blic schools?


3. The slu(et was relained in his her grade last school year?

4 (hilh schools only) The student is classified as a vocational sl~idewt7


SECTION B SPECIALLY FUNDED PROJECTS DIRECTIONS: Darken as many bubbles as apply for this student,


1st Grade Teachers Indicate il this student was enrolled in One ol these pireschool piogranis.

HII..I .l IStrtI
Pit, Klrmilhi.iit.miI til I



All Teachers Indicate any of the following specially Iiiniled pto ldilrms in which this student ctrrently is enrolled


c illil I.Ilo y Edticiall

ESSA Bisi
ESSA Pilu Folluw 1lliimnlh C mi,.tt


Title I Realing Title I Mailh
A
B
C
Nonte u Ilth'se


TRANSACTION DATE


MON lii



M-l, I





Oi,d,m. 4I A,..im. a Om.1-.r Is


YEAR



19
79 80 8I
82 83
84 85 86
87


I I I I I I I I 1 1 1 1 1 1 1 I F

- - - - - - - - - - - - - - -


C



o


0












0K

. -2
t-I m.1


I J


I lli'.l


N,.*.iN-.










APPENDIX B
STUDENT INFORMATION QUESTIONNAIRE
STUDENT PERCEPTIONS, ACTIVITIES, AND EMPLOYMENT




Name School
last first middle

Address (Number and St.)

City State Zipcode Student Identification Number (if known) Please answer the following questions by darkening in the appropriate bubble "Yes" or "No" after the question. Please read all the questions (both sides) before starting to answer the first one.

1. School year, 1981-82, I participated in at least one, or more, of the school activities of athletic teams, band, chorus, orchestra, or other school sponsored clubs: YES 0, NO 0. 2. In comparison to my fellow students, I think that my intelligence is: Below Average 0, Average 0, Above Average 0. 3. During the more than one-half of school year 1981-82, I held a part-time job: YES 0, NO 0. 4. During the more than one-half of school year 1980-81, I held a part-time job: YES 0, NO 0. If your answers to questions Number Three and Four were "NO" that completes the information needed. Please turn in this sheet. If your answer to question Number Three or Four was "YES", please select an hours worked category that is closest to the amount of time you were usually employed each week for more than one-half of that year.











5. During the 1980-81, 1981-82 school years, I was paid for doing part-time work (including odd jobs, such as mowing lawns, or babysitting, and so forth) for about 1 to 4 hours per week:


1980-81 YES 0,


NO 0. 1981-82


1980-81


6. I worked 7. I worked 8. I worked 9. I worked 10. I worked 11. I worked 12. I worked Thank you for


5 to 9 hours per week: 10 to 14 hours per week: 15 to 19 hours per week: 20 to 24 hours per week: 25 to 29 hours per week: 30 to 34 hours per week: 35 or more hours per week: you help in performing this


YES YES YES YES YES YES YES


YES, NO 0.

1981-82

YES 0 NO 0 YES 0 NO 0 YES 0 NO 0 YES 0 NO 0 YES 0 NO 0 YES 0 NO 0 YES 0 NO 0


bit of educational research.


Individual identified responses will not be released without your advance permission.


On free lunch/reduced lunch program. Yes No









APPENDIX C
STUDENT DATA NEEDED FOR LISTINGS
OF SAMPLE DATA BASE


1. Name (last, first, initials). 2. Student Identification Number.

3. Sex (1) male (2) female.

4. Ethnic group: (1) White non-Hispanic, (2) Black non-Hispanic,

(3) Hispanic, (4) American Indian or Alaskan Native (5) Asian or

Pacific Islander.

5. Year grade level in school during 1981-82 6. School attended 1981-82 school year # 7. School attended 1980-81 school year # 8. School attended 1979-80 school year # 9. Grade point average school year 1981-82 10. Grade point average school year 1980-81 11. Participation in extracurricular activities (yes/no). 12. Self-perception of intelligence: (1) below average, (2) average,

(3) above average.

13. Number of hours worked per week for most of the school year

sorted by categories: (1) none, (2) 1-4 hours per week, (3) 5-9 hours per week, (4) 10-14 hours per week, (5) 15-19 hours per week, (6) 20-24 hours per week, (7) 25-29 hours per week, (8)

30-34 hours per week, (9) 35 or more hours per week.

14. Student participation in any of the specially funded projects in

Section B of SIMS Biographical Data Form, (yes/no). 15. The number of days present during 1981-82













REFERENCES


Bailey, R. C. & Bailey, K. G. Self-perceptions of scholastic
ability at four grade levels. Journal of Genetic
Psychology, 1974, 124, 197-212.

Belanger, R. R. & Boyle, R. D. Stepwise multiple regression,
Cupertino, California: Apple Computer Incorporated,
1980.

Bell, 'J. W. Comparison of dropouts and non-dropouts
on participation in school activities. Journal of
Educational Research, 1967, 60(6), 248-251.

Bidwell, C. E. & Kasarda, J. D. Conceptualizing and measuring the effects of school and schooling. American
Journal of Education, August 1980, 88, 401-430.

Cole, Sheila, Send our children to work? Psychology Today,
July 1980, pp.44-56; 50-58; 60-68.

Crawford, G. & Miskel, C. Experience based career education
at wichita east high school: A third party evaluation. Kansas: Wichita Public Schools, July 1978.
(ERIC Document Reproduction Service No. ED 150 285)

Delaney, W. The transition from school to work: A study
of laws, regulations, and practices restructuring work experience and employment opportunities. Ann Arbor, Michigan: University of Michigan, 1975. (ERIC Document Reproduction Service No. ED 124 762)

Dykeman, B. WECEP: initiating a career education program
for the 14 and 15 year old student. Education, 1979,
99(4), 363-367.

Ferguson, R. & Maxey, E. 'J. Trends in the academic performance of high school and college students. Journal of
College Personnel, November 1978, 19, 505-511.

Finn, 'J. D., Dulberg, L. & Reis, 'J. Sex differences in
educational attainment: A cross-national perspective,
Harvard Educational Review, November 1979, 49(4),
477-503.

Gade, E. & Peterson, L. Comparison of working and nonworking high school students on school performance, socioeconomic status, and self-esteem. Vocational
Guidance Quarterly, September 1980, 29, 65-69.









Gill, N. T. Comparison of high school students interests
across three grade and ability levels. High School
Journal, 'January 1980, 63, 160-166.

Good, T. L. , Sikes, 'J. N. & Brophy, 'J. E. Effects of
teacher sex and student sex on classroom interaction.
Journal of Educational Psychology, 1973, 65(1), 74-87.

Hammond, W. A. Scholastic achievement and part-time employment. The Clearinj House, 1970, 44 (7), 465-467.

Heffez, 'J. The effects of part-time employment on high
school students' arade point averages and rate of school attendance. New York: EDRS Research Report, 1979. (ERIC Document Reproduction Service No. ED
173 592)

Heyneman, S. P. Six views on three issues related on
education and work, report of a symposium in debate form held june 28, 1977. Washington, D. C.: National Institute of Education (DHEW), 1977. (ERIC Document
Reproduction Service No. ED 179 687)

Kobett, L. A. Relationship between home background, school
achievement, and adolescent values. Education, 1979,
100(2), 158-164.

Levine, S. V. Psychological and social effects of youth
unemployment. Children Today, November-December 1979,
8,6-9.

Nuttall, R. L. Do the factors affecting academic achievement differ by the socioeconomic status or sex of the students? Ann Arbor, Michigan: University of Michigan, 1972. (ERIC Document Reproduction Service No. ED 064
465)

Pope, M. C. Students in Duval County have high mobility
rate. The Florida Times Union, May 17, 1982, pp.B-l.

Rehberg, R. A., Schafer, W. & Sinclair, J. Toward a temporal sequence of adolescent achievement variables.
American Sociological Review, 1970, 35, 34-48.

Roberson, D. R. Division of research, planning and evaluation policies and procedures. Jacksonville, Florida:
Duval County Schools Printing Office, 1980.

Roscoe, J. T. Fundamental research statistics for the
behavioral sciences (2nd ed). New York: Holt, Rinehart and Winston, Inc., 1974.









Schab, F. Adolescence in the south: A comparison of
blacks and whites not prepared for the world of work.
Adolescence, 1979, 14(55), 599-605.

Shively, 'J. & Watts, R. Final evaluation report for the
conecuh county part D experience-based career education program. Charleston, West Virginia: Appalachia Educational Laboratory, 1980. (ERIC Document Reproduction
Service No. ED 181 221)

Skallerud, R. D. Career Education: Planning, learning,
understanding, succeeding--ce+, final report, september 1, 1976 to august 31, 1979. Bismark, North Dakota: North Dakota State Board of Vocational Education, 1979.(ERIC Document Reproduction Service No. ED
179 726)

Watkins R. W. & Corder, R. Student outcomes and participant
opinions in experienced-based career education schools.
external evaluators final report on the experiencedbased career education program, vol. VI . Berkley, California: Educational Testing Service, 1977. (ERIC
Document Reproduction Service No. ED 155 181)

Williams, N. Part-time jobs advantages and disadvantages.
School and Community, 'January 1967, pp.36;54.

Yarworth, 'J. S. & Gauthier, W. �J. Relationship of student
self-concept and selected personal variables of participation in school activities. Journal of Educational
Psychology, 1978, 70, 335-344.

Young B. Better education most critical need. The Florida
Times-Union, 'January 27, 1980, pp.D-3.













BIOGRAPHICAL SKETCH


Walter G. Squires, Jr., an earth-space science and biology teacher at N. B. Forrest High School, Jacksonville, Florida, has taught science subjects for 5 years spanning a period of 30 years. Mr. Squires received his bachelor's degree in psychology from the University of Florida in 1951, the Master of Education from the University of North Florida in 1976, and the Doctor of Education in 1983 from the University of Florida. He is a Commander in the United States Navy (retired) . He was awarded U. S. Patent No. 3,035,285 for an explosively anchored navigational buoy. He was recognized in 1972, as one of the foremost authorities in the U. S. Navy in the field of mine warfare. Mr. Squires is the father of three children and has four grandchildren. He is a member of Kappa Delta Pi, the

National Society for the Study of Education, the American Legion, National Sojourners, a 32" Mason, and is a disabled veteran.















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 d9gree of Doctor of Education.

"M. Y/Wunnery .a, Ed/ Chairman

Prof 'ssor of Educati a
Administration a d 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.



./W. Longstr th, Ed D. )
(Associate Professor, 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.


T. C. Healy, Ph.

Associate Professor 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.



I'. A . a c s e n, E d .D Profesor of Instructional Lead e/r s h i p and S u ppor t



This dissertation was submitted to the Graduate Faculty of the Department of Administration in the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Education.

April, 1983
Dean for Graduate Studies and Research




Full Text

PAGE 1

PART-TIME EMPLOYMENT AS A PREDICTOR OF GRADE POINT AVERAGE FOR SECONDARY SCHOOL STUDENTS IN AN URBAN SCHOOL DISTRICT BY WALTER G. SQUIRES, JR. A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF EDUCATION UNIVERSITY OF FLORIDA 1983

PAGE 2

Copyright 1983 by Walter G. Squires, Jr.

PAGE 3

To Walter G. Squires Corine T. Squires Parents

PAGE 4

ACKNOWLEDGEMENTS The autlior wishes to express his appreciation to those whose efforts have contributed to the completion of this dissertation. Gratitude is offered to Dr. M. Y. Nunnery, Chairman of the Supervisory Committee, Dr. ‘ J . W. Longstreth, Dr. T. C. Healy, and Dr. D. A. Jacobsen, for their support and assistance. The author wishes to extend a singular thank you to Dr. Nunnery for his invaluable wisdons, guidance, and patience in the actual preparation of this document. The assistance of Mr. Ronell Poppell, Principal of N. B. Forrest High #241, given at most critical times is greatly appreciated. The author wishes to recognize those friends that helped with much of the tedious detail work so necessary for data collection: Mr. Jessie Bullard, Miss Diana Lynn Hastings, Kiss Carmen Denise Blouir, Miss Janet Coffrin, Miss Tammi L. Wilson, and Mr. William Green. A special thanks to Mr. Loel A. Cruikshank, auditor, who double checked all data entries to ensure accuracy. He is a great father-in-law. A particular thank you is offered to Claire, the author's wife, for her prolonged effort, love, continuing support, understanding, and encouragement during some very difficult times. IV

PAGE 5

A final note of appreciation is offered to Miss Barbara Young who identified the need for study of the stated problem. V

PAGE 6

TABLE OF CONTENTS PAGE ACKNOWLEDGEMENTS iv ABSTRACT vii CHAPTER I INTRODUCTION, 1 The Problem 4 Delimitation and Limitations 4 Justification of the Study 5 Definition of Terms 8 Procedures 10 Organization of the Research Report 16 II CORRELATES OF ACADEMIC ACHIEVEMENT 17 AMONG HIGH SCHOOL STUDENTS: A REVIEW OF THE RESEARCH III PRESENTATION AND ANALYSIS OF DATA 30 Descriptive Data by Grade Level Relative to Students Involved In the Study 31 Correlation Matrices 37 The Results of Multiple Regression Analysis 47 Multiple Regression Data 49 IV SUMMARY, CONCLUSIONS AND DISCUSSION 55 Summary 55 Conclusions 62 Discussion 63 APPENDICES A DUVAL COUNTY SCHOOL SYSTEM 68 BIOGRAPHIC DATA FORM B STUDENT INFORMATION QUESTIONNAIRE 69 C STUDENT DATA NEEDED FOR LISTINGS 71 OF SAMPLE DATA BASE REFERENCES 72 BIOGRAPHICAL SKETCH 75 VI

PAGE 7

Abstract of Dissertation Presented to the Graduate Council of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Education PART-TIME EMPLOYMENT AS A PREDICTOR OF GRADE POINT AVERAGE FOR SECONDARY SCHOOL STUDENTS IN AN URBAN SCHOOL DISTRICT By Walter G. Squires, Jr. April, 1983 Chairman; M. Y. Nunnery Major Department; Educational Administration Given the paucity of research and policy implications relative to the impact of part-time employment on academic performance among secondary school students, the study of the impact of part-time employment on grade point averages, after accounting for the contributions of eight other variables known to affect grades, was undertaken. From the 34,340 students enrolled in grade 8-12 of an urban school district a grade-level stratified random sample was selected. Data were collected from district-maintained student records and from a student-completed instrument relative to 470 students' employment status, the selected other presumed independent variables, and 1981-82 grade point averages. vii

PAGE 8

Two multiple regression analyses were run for each grade, one using whether or not the student was employed and the other the amount of employment. The form of the analysis was stepwise for all of the variables except employment status which was entered last. Since in only three instances did the part-time employment make a significant contribution to the variations in grade point averages (accounting for 1.1% and 1.2% of 68.7% total variance and 2.2% of 52.5% total variance), it was concluded that its impact was minimal. Consistent with previous research, previous grades, academic aptitude scores, and school attendance consistently accounted for most of the variance in grade point averages. viii

PAGE 9

CHAPTER I INTRODUCTION There can be found two schools of thought on part-time employment for secondary school students. Teen-age students in junior and senior high school who work part-time jobs are thought by some educators to be adversely affected academically because of part-time work (Cole, 1980, p.44). That is, part-time work has an adverse effect on the grade point average (GPA) of working students. This school of thought also holds that the requirement for students to work part-time while attending school precludes their participation in many school related activities. It is reasoned that involvement in cocurricular activities is an integral part of the school years' development and that the "values orientation" from such participation is lost for students working while attending school (Williams, 1967, p. 36; Young , 1980 , D-3) . The followers of the opposing school of thought hold concepts likely to be expressed as follows: "Part-time employment for students serves to keep them off the streets and out of trouble." "It helps them to raise their standard of living, and learn the value of a dollar." 1

PAGE 10

2 "Part-time employment prepares young people to become working members of society." "Part-time employment contributes to the reduction of unemployment in general and to an increase in the gross national product in particular." A 6-city survey by the National Manpower Institute indicated that 68% of the employers surveyed were firmly in favor of youth work experience and the survey cited such benefits of working as developing maturity, self-identity, self-reliance , preparing for future career direction, establishing good working relationships, and enjoying the stimulation of working with competent professionals (Delaney, 1975, VI-14) . Relative to the stance of parents and teachers on the matter of part-time employment for high school youth, it can be argued that the concurrence of parents is expressed when they sign the appropriate forms to enable their children to secure work permits. The researcher, in preparation for the research reported herein, talked informally with a number of teachers about the issue and found a wide variation of opinion with several teachers supporting the notion of part-time employment and other teachers equally strong in their opposition to part-time employment. The division of opinion tended to center around whether or not it impacted on their school activity both in terms of academic performance and the ability to participate in extra-curricular activities . Some situations make it necessary for teen-age students to work part-time. Low family income, divorce, death

PAGE 11

3 of the principal wage earner in the family, unanticipated need to include the student in the family work force, or abandonment by one or both parents may force the teenage student to fend for himself or herself financially. Socioeconomic and peer pressure to participate in activities that require a substantial outlay of funds mandate part-time work by some students. Some teen-age students work part-time jobs when it is a matter of convenience to acquire luxury items. Furthermore, government and individually-sponsored student employment programs, which have been instituted in an effort to combat students dropping from school, have encouraged students to work during part of the school day or outside classroom hours (Delaney, 1975; Dykeman, 1979, p. 365; Heyneman, 1977; Levine, 1979). The conclusion drawn by the researcher, based on available information, was that apparently administrators and counselors involved in such programs had not considered the possibility that the part-time employment could have an adverse influence on student academic performance; or, the assum.ption was made by these school officials that the benefits from gainful part-time employment out-weighed any possible adverse academic effects. A basic conflict can be seen in the two points of view on part-time employment for secondary school students. Therefore, the study reported herein focused on the relationships that might exist between academic achievement as reflected in grade point averages and part-time employment by students.

PAGE 12

4 The Problem The problem in the study was to determine for students in each grade level, grades 8 through 12, in a single urban school district the relative impact of parttime employment upon their 1981-82 grade point average (GPA) after the contributions of sex, ethnic group, socioeconomic status (SES), school attendance, GPA for previous school year (1980-81), self-perceptions of intelligence, extracurricular activity participation, and Stanford Achievement Test Battery (SAT) overall percentile score were taken into account. Delimitation and Limitations The potential participant s in th e study were 34,340 s t udents who were enrolled in grades 8 through 12 of the Du V a 1 Co unty , . Florida, School System in the school year 1 9 81-82 . Fr om this populate .on , at each grade level a r a ndom sample of 120 students was sel ected for inclusion in the study. Data in regard to 1981-82 GPA, school attendance, sex, ethnic group, SAT score, SES, and 1980-81 GPA were obtained from the Duval County School Board SIMS computer data base and student cumulative record files. All data were verified with the student permanent record files maintained at the school that the student was attending. The data source for information pertaining to hours worked, extracurricular activity participation, self-perception of intelligence. and participation in free or

PAGE 13

5 reduced lunch program was a researcher-designed student information questionnaire. Complete useable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students. Multiple regression was used to determine the contributions of the independent variables to the dependent variable. In entering the variables in the multiple regression equation, the part-time employment variable was entered last in order that the impact of this variable over and beyond that of the other independent variables could be determined. The other independent variables were entered in stepwise fashion. To the extent that these data sources were inaccurate or incomplete the study lacked validity. However, this inaccuracy was guarded against in that each data sheet was cross-checked against the cumulative folder by the researcher. Furthermore, since the study was confined to the students in a single urban school district for a specified year, the results over and beyond this population had to be regarded as suggestive rather than conclusive. That is, the data were applicable to only one school district at one point in time. Justification for the Study Some school administrators, parents, guidance counselors, and politicians advocate the expenditure of vast sums of money for programs that encourage students to work

PAGE 14

6 for a variety of reasons (Crawford & Miskel, 1978; Delaney, 1975; Dykeman, 1979; Heffez, 1979; Heyneman, 1977; Watkins Si Corder, 1977). Research studies dealing specifically with student employment effects on academic achievement are very limited. The results of the studies that could be found dealing with student employment were contradictory and of limited gener al i zabi 1 i ty . Most were performed using very small sample size. Most were from rather limited populations (i.e., one or two schools). School administrators have been influenced by the Department of Labor since 1970-71 to increase the participation in the Work Experience and Career Exploration Program (WECEP) and other vocational programs in order to retain dropout prone students within the school system (Dykeman, 1979, p. 363; Shively & Watts, 1980; Skallerud, 1979). The researcher had personal knowledge of the fact that classroom teachers were expected to alter or modify their lesson plans to allow cooperative training programs students to leave class early and be spared regular homework assignments. Homeroom programs, flag observances, and announcements were routinely preempted by the vocational education students being dismissed early to catch the bus to work at two district-operated skill centers. As has been noted, there were few research studies dealing specifically with a relationship between student employment and academic achievement. Over 1400 items were found in ERIC and other sources under the subject "academic achievement"; of that number, only 120 were

PAGE 15

7 concerned with grades 7 through 12. Of those studies found, only two studies focused directly upon grade point average and part-time employment. These were the Gade and Peterson (1980) and Hammond (1970) studies. The results were conflicting. In the Gade and Peterson study no significant association between employment status and academic achievement was found. In the Hammond (1970) study there was evidence that the level of achievement might be significantly lower among the employed students than among the unemployed . Millions of dollars are being spent on various vocational education programs, cooperative work programs, student unemployment surveys, and businesses' employment of students at below the minimum wage, because they work less than 40 hours per week. All these programs are operating on the basic assumption and widespread opinion that "school youth need work." They are not founded on continuing or widespread results of scholarly research. The study reported herein does not provide a definitive answer to the question, "Should students work or not work?" It does, however, add to available knowledge of a relationship between students working part-time and their grade point averages. It was reasoned that if the findings indicated there was no meaningful relationship between students working part-time and academic achievement, it would help alleviate the concerns of some educators about the possibility of an adverse impact of working part-time on academic achievement. If there was a meaningful adverse

PAGE 16

8 relationship between students working part-time and academic achievement, administrators should discourage student work programs. Therefore, the study was seen as relevant to administrators who make decisions on matters concerning part-time employment policies and practices for students. Definitions of Terms Employment . The performance of some duty, service, or obligation of time, and availability of a person for another that results in payment is employment, a job for money. Various studies have used different criteria (e.g., "over 5 hours per week," or "at least 12 hours per week"); in the study, any work for pay outside of class on a continuing basis was considered employment. Part-time jobs of an independent contractor nature such as newspaper routes were estimated to the number of hours per week performed. Babysitting jobs were approximated to the nearest hour per week on the average considering the school year. Ethnic group . The classification system of the Duval County, Florida, school system whereby each student assigns himself or herself to one of the following groups was used: white non-H i spani c , black non-H i span i c , Hispanic, American Indian or Alaskan Native, and Asian or Pacific Islander . Extrac ur ric ular activity . A student's response to a global question about extra curricular activity participation contained in a researcher-designed instrument was used.

PAGE 17

9 The terms used elsewhere in referenced materials, "cocurricular activities," and "third curriculum activities" were interpreted in this study to be synonymous with extracurricular activities. Grade point average (GPA) . The GPA is the numerical average of letter grades awarded at the end of a school year. In determining the average an A = 4 points, B = 3, C = 2, D = 1, and E = 0 , or failure to earn credit for the course of study. (This was consistent with district policy for the years in question in that advance placement classes did not carry additional weight.) In this study, grades for all subjects in which the students were enrolled for credit were used in determining 1980-81 GPA and the 1981-82 GPA. School attendance . The number of days of attendance in school for the year as recorded in the student permanent record file was used to indicate attendance* Self-perception of intelligence (SPIN) The SPIN was the student's response to a global question about intelligence contained in a researcher-designed instrument. Socioeconomic status (SES) . A dichotomous classification based on whether or not the student participated in the free lunch or reduced price lunch specially funded program was used. Participation in free lunch or reduced price lunch program resulted in a "Yes" classification. No participation in the free lunch or reduced price lunch program was a "No" classification. A "Yes" classification

PAGE 18

10 was considered low socioeconomic status. A "No" classification was considered a high socioeconomic status. Stan ford Achievement Test Battery (SAT) . The most recent SAT score held by the Duval County School System was used. When English, Reading, and Mathematics portions were combined a single overall score was derived. The overall percentile score recorded in a student^s cumulative record file was used in the study as the measure of scholastic ability. Procedures The procedures used to conduct the study are described in the paragraphs that follow. These are organized by study design, sample selection, instrumentation, data collection, and data treatment. Study Design The study was a survey type of ex post facto study. A stratified random sample of junior and senior high school (grades 8 through 12) students entered in the data base by their completion of Duval County School System Biographic Data Form (Appendix A) was selected from the listing for the 1981-82 school year by the Duval County Student Information Management System (SIMS) computer from the population of each grade 8 through 12 students for 198182. A survey, using a researcher-designed instrument (Appendix B) , was made of selected students to determine whether or not they held a part-time job during the 1981-82

PAGE 19

11 school year. If they did work a part-time job, the number of hours worked and other relevant data were requested. The grade point averages of the selected students as well as the other required biographic information were obtained from the Cumulative Record Folders and the computer maintained data base (Appendix C) . The data were analyzed by grade level using multiple regression to determine the contribution of part-time employment and amount of part-time employment to grade point average over and beyond the contributions made by other independent variables that previous research had indicated contributed to grades. Sample Selection and Final Sample The 34,340 students who were enrolled in grades 8 through 12 in the 1981-82 school year in the school system of Duval County, Florida, junior and senior high schools were the population from which the student participants in the study were selected. The basic procedure for selecting the participants was by a stratified random method. Specifically, for each grade, grades 8 through 12, the Duval County SIMS computer program was used to generate a 2% random sample of students which resulted in a minimum of 120 students randomly selected at each grade level. This allowed for a minimum of 10 students for each of the independent variables in the multiple regression equation and gave a safety factor of 33 1/3% to cover loss of students who moved, or in some other fashion were not

PAGE 20

12 available for data collection. The one-third turnover approximation was based on observed rates of turnover in the classrooms during the past four years. The turnover on a county-wide basis for the year was estimated at nearly 40% by the school superintendent (Pope, 1982 , p. B-1 ) . As a result of the selection procedure, complete useable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students. Instrumentation and Data Collection The Duval County School System Biographical Data Form (Appendix A), which was used to gather data for the Student Information Management System (SIMS), was a basic source of information for this study. The SIMS was established for the purpose of obtaining and storing in the computer pertinent information concerning the Duval County student population. A major goal of the system was to afford quick access to student information while reducing the reports and data collection burdens on school-based personnel (Roberson, 1980, p. 20). Information provided by this source was identification of the students to be included in the study, their sex, grade level, school attended, and ethnic grouping. The student information questionnaire (Appendix B) was the principal instrument used to obtain the other information needed, including data which was not yet stored in SIMS such as attendance, whether or not student

PAGE 21

13 worked part-time, if so, how many hours were worked per week, self-perception of intelligence, and verification of participation in free lunch/reduced price lunch program. As previously indicated, the student information questionnaire was developed by the researcher whose concern was to ask the questions necessary to deal with the focus of the study. After its initial development, the instrument was administered to a group of approximately 30 Duval County School System students included in the population but not in the random sample. The instrument was then revised based on their responses. Thus, it was believed that the instrument was valid for the purpose used. The student permanent record file and cumulative record file were the source documents for SIMS. They were considered to take precedence should conflicting data appear in any other sources. The consolidating instrument that was used to combine the information from all sources to establish the data base for performing the multiple regression operations is shown as Appendix C. Data Coding and Treatment Each bit of data was coded in a form that was usable in the multiple regression analysis. Specifically, the coding of each of the variables was as follows: The 1981-82 grade point average (GPA) was coded in a two digit number between 0.0 and 4.0 representing y, the dependent variable.

PAGE 22

14 The number of days of school attendance was represented by total days present out of 180 days during the 1981-82 school year as predictor variable x^. The sex of the student was represented by (1) for male (2) for female as predicator variable Extracurricular school-sponsored activity involvement in athletic teams, band, or clubs was coded (1) to indicate the student was involved in one or more athletic teams, band or clubs, (0) to mean not involved in any school sponsored athletic teams, band, or clubs. This was predictor variable x^. Ethnic group membership was coded with same code used by SIMS: (1) white non-Hispanic, (2) black non-Hispanic, (3) Hispanic, (4) American Indian or Alaskan Native, (5) Asian or Pacific Islander. Ethnic group membership was predictor variable . The socioeconomic status (SES) was coded as (1) for participation in the specially funded programs for socioeconomically deprived students, (0) for nonparticipation in any of the specially funded programs. The SES was predictor variable x^. The most recent composite, or overall SAT percentile score was entered as a two digit number as predictor variable X . 6 The student's self-perception of his/her intelligence (SPIN) was coded (1) to indicate a perception of intelligence as below average intelligence, (2) a self-perception of average intelligence, (3) a self-perception of above

PAGE 23

15 average intelligence. The SPIN was predictor variable Each student's grade point average' for the previous school year (1980-81) was coded as a two digit number between 0.0 and 4.0 as predictor variable x_ . O For the first regression analysis, whether or not the student worked at all during school year 1981-82 (employment status 1982 yes/no) was coded as (0) for not working during the school year and (1) to indicate that student worked during the school year. This is hereafter referred to as Model I analysis. In the second analysis, the amount of time worked (employment status 1982 category) was coded using a single digit number from one eight to indicate the categories of hours worked based on the category scale shown as a part of Appendix B. This is hereafter referred to as Model II analysis. This was predicator variable X9. The slope of correlation between the predictor variable and the dependent variable was coded "b" as is standard practice in solving the multiple regression equations . The above coded variables were incorporated into a multiple regression equation (Roscoe, 1974, p. 362; Belanger & Boyle, 1980) , that was performed for each grade level 8 through 12. As noted, multiple regression equation analyses were performed for each grade level sample. As indicated, the first analysis. Model I, was with the part-time employment treated as (1) employed or (0) not

PAGE 24

16 employed part-time; in the second analysis. Model II, the part-time employment was coded and treated in terms of amount ranging from zero employment to 35 plus hours per week . The form of the multiple regression equation (y = + b^x^ + b,x^ + b^x^ + b^x 2 2 3 3 4 4 5 5 6 6 is such that one can determine if the mere presence of part-time employment and the amount of part-time employment can be related to grade point average. In order to do this, all the independent variables save the part-time employment (variable x^ in each instance) were entered in stepwise fashion. Then, the part-time employment variable (whether 1 or 0 to represent part-time employment, or the category of part-time employment to represent amount of employment) was entered last. This enabled a determination to be made of the amount of contribution to the dependent variable of grade point average made by the employment variable after the contribution of the other variables has been taken into account. Organization of the Research Report The remainder of the study is organized into three chapters. The second chapter is a review of related literature and research. The third chapter contains the presentation and analysis of data. The fourth chapter includes the summary, conclusions, and discussion.

PAGE 25

CHAPTER II CORRELATES OF ACADEMIC ACHIEVEMENT AMONG HIGH SCHOOL STUDENTS: A REVIEW OF THE RESEARCH Two points of view about part-time work for high school youth were found in the literature along with copious reasoning for each point of view. The scholarly research to support either point of view was limited. However, there had been numerous studies of factors related to academic achievement among high school students. In order that the reader may have a framework for interpreting the data and an understanding of why it is necessary to take into account the impact of variables in addition to part-time employment on the academic achievement of students, the review presented herein, over and beyond the two studies relating to part-time employment and academic achievement, is confined to a synthesis of the research showing the relationship of variables to academic achievement. This research in regard to other variables provided the basis on which selections were made about the other independent variables that needed to be taken into account before determining the contribution of part-time employment. Before turning directly to the other variables, it seems appropriate to deal first with the two studies which related part-time employment and academic achievement. 17

PAGE 26

18 Gade and Peterson (1980) and Hammond (1970) focused on the relationship of school related behavior, and other variables affecting academic achievement, to work experience and academic achievement. Gade and Peterson (1980) found that there was no significant association between employment status and academic achievement. Their research sample consisted of 351 10th grade students enrolled in 2 urban high schools located in the upper midwestern part of the United States. The determination of achievement was based on grades received in the last two semesters in four basic subjects. Working students were designated as those employed 12 or more hours per week. The study sought to examine the relationship of student work experience to five factors: (a) gender, (b) academic achievement, (c) extracurricular involvement, (d) socioeconomic status (SES) , and (e) self-esteem. "Students who are employed seem to perform as well or better [when identified] on [the basis of] these indicators than non-working students" (Gade & Peterson, 1980, p. 69). The limitation of the sample to 10th grade students may somewhat limit generalization downward to the junior high school students and upward to the 12th grade students. Hammond (1970) found that the level of academic achievement was significantly lower among employed 12th grade students than among the unemployed. The rapidity of changes in behavior and personality in the teen years may produce considerably different results in lower and higher

PAGE 27

19 grades. A predictor variable of the grade level of the student would help identify this effect. A need for categorization of the origins of influences on student achievement has been indicated by research. In attainment models where individual student achievement is at issue, individual student attributes, other nonschool inputs to schooling, and school inputs must be measured at the individual level to avoid misleading results. (Bidwell & Kasarda, 1980, p. 426) Variables Related to the Background of the Students There has been considerable research undertaken that focused on selected variables related to the background of students in relation to academic achievement. Sex, socioeconomic status (SES), standardized measures of academic ability (i.e., intelligence tests), and/or standardized measures of academic achievement (e.g., SAT) have been taken into account as background variables. Finn, Dulberg, and Reis (1979) found in cross-national studies of educational attainment that the influence or interrelationship of the sex of the student, the teacher, and/or their interactions may be dependent on the national culture. More specifically the need to keep cultural sex discrimination in mind in drawing conclusions from sex based data was expressed as follows: "The explicit relationship of sex-segregated schooling to the attainments of boys and girls makes this issue especially worthy of exploration" (Finn, Dulberg, & Reis, 1979, p. 494). They concluded that any study of educational attainment in terms of school attendance, literacy, academic performance, or

PAGE 28

20 preparation for gainful employment would reveal that the patterns for men and women differ markedly according to the nationality under consideration. These differences cannot be divorced from the process of schooling itself. Opportunity to learn is one of the significant correlates of pupil achievement. Opportunity, as defined in these studies, refers to the degree of exposure to the course material and the required skills. . . . Many of the data reported in this summary suggest that inferior performance in either sex is due in part to the inadequacy of time and experience provided by instructional practices. (Finn, Dulberg, & Reis, 1979, p.497) Because of its frequent consideration, inclusion of the gender variable in the study reported herein seemed reasonable. However, based on the results of Finn, Dulberg, and Reis (1979), the sex variable could have minimal influence in a population of one school district in the United States. The socioeconomic status (SES) of the student is also typically considered to be a major influence on academic achievement. One of the studies that contributed a different point of view was done in Puerto Rico by Ronald L. Nuttall in 1967-68. The model involved two levels of achievement, two levels of sex, and five levels of socioeconomic status. The sample size was reduced by screening criteria from 2,500 to 800 seniors and some juniors in the largest public high school. In terms of family background there were no statistically significant interactions. This study indicated that the same factors were important for academic achievement for both boys and girls and for students from all socioeconomic status levels. Students who

PAGE 29

21 v/ere doing well academically desired, attain higher status occupations than doing well academically. The lower expected and desired a lower level effects held across socioeconomic s and were working to those who were not achieving students occupation. These tatus levels. The importance of these background factors to academic achievement was explained as follows: Personality traits associated with academic success did differ somewhat by sex and socioeconomic background. Low achieving boys as contrasted to high achieving boys and to both high and low achieving girls, were markedly less intelligent. A similar pattern existed for the traits of conscientiousness where again the low achieving boys were markedly deficient in conscientiousness. These low scores for intelligence and conscientiousness held at all socioeconomic levels. (Nuttall, 1972, p. 83) The positive relationship between the educational expectations of the adolescent and the socioeconomic status (SES) of his/her family has been consistently replicated. The students from the "better family" expected to achieve more and go farther than "poor folks." However, family expectations and adoption of those values by the student are inconsistent. Rehberg, Schafer, and Sinclair (1970) found that ambition, personal educational expectations, and measured intelligence so overshadow socioeconomic stratification that achievement predictions based predominately on SES alone would not be valid. Mobility attitudes were more strongly associated with stratification of destination than with stratification of origin (p. 46). Poor youth may work part-time and rich youth may work part-time. But neither of these categories has been shown to be an exclusively significant determinant of academic achievement.

PAGE 30

22 In light of the studies that have been done in regard to the variables of sex, socioeconomic class, and measures of ability, it is obvious that there are some conflicts. However, since there seemed to be some indication that these may be influential, it was decided to include all three of these variables in the study reported herein. For convenience, given the high relationship between "intelligence" and "achievement," instead of using a measure labelled intelligence to portray scholastic ability, the Stanford Achievement Test was used. (It was available in the students' records.) Variables Related to Students' Self-perceptions Huge quantities of resources are being invested in work-study programs from eighth grade through high school (Dykeman, 1979). How helpful is the work-study program for the 14 and 15 year old junior high school student who is already undergoing "identity confusion," a reflection of a general redefinition of self, of others, and of the world which takes place in the early adolescent (Bailey & Bailey, 1974, p. 210)? The self-perceptions held by the student of how intelligent the student felt himself/herself to be were found to be very important by Nuttall (1972). The single most important trait related to academic performance was found to be intelligence both actual and self-perceived. . . . The school should routinely monitor or how intelligent the child thinks he/she is. (Nuttall, 1972, p. 87) Self-perceptions of scholastic ability were found related to academic achievement in studies by Bailey and

PAGE 31

23 Bailey (1974) who confirmed Nuttall's (1972) finding. Using 341 students from different grade levels, Bailey and Bailey (1974) found a shift in self-perceptions of scholastic ability — from low at eighth grade to high at 12th grade. At the 8th grade level, the males overrated their actual ability, while females tended to underrate their actual ability. The 12th grade students seemed to have the most realistic view of their ability. High school students not only revealed the most stable self-perceptions of the groups studied, but the most realistic self-views (p.211). The relationship among home background, school achievement, and adolescent values (defined more broadly than self-perceptions of intelligence and including values related to work, satisfaction, self-worth, and morality) was studied by Kobett (1979). Here again, overall grade point average was used as a measure of school success. One of the strongest conclusions resulting was that "there is no evident relationship between adolescent values and school achievement" (p. 162) . The interests of students as they relate to both working part-time and academic achievement were central to the work of Gill (1980) who said, "What students are most and least interested in is certainly a significant variable in the educational process" (p. 160). More specifically, the interest of the student in school, peace, love, life, and the opposite sex may be very significant (Gill, 1980, p. 162) .

PAGE 32

24 Research such as that cited provides support for the notion that students' self-perceptions of intelligence versus measured intelligence are legitimate points to consider in a study of the relationship of part-time work to academic achievement. Measures of adolescent values orientation seemed to have no relation to academic achievement. Even though interests and related value system may be related to academic achievement, since the study reported herein was based primarily on records, it was assumed that the attributes were randomly distributed in the population. Other Relevant Variables Student attitudes toward school and work, teacher sex and student sex interactions, and extent of participation in extracurricular school activities have been examined as relevant variables. In a survey of 1248 high school students in North and Central Georgia, their attitudes and their future in the world of work were probed. Their attitudes about the need for freedom and independence in their future work and their feelings concerning certain conditions under which they would want to work, as well as their philosophy in general, were sampled and analyzed. Their attitudes, as shown by the correlation of their perception of the need for achievement in the classroom and success on the job, were not deemed to be very positive. These students were.

PAGE 33

25 perhaps, recalling some historical examples of this discrepancy: U.S. Grant, Wagner, Edison, Churchhill, von Braun, etc." (Schab, 1979, p. 603). The influence of the teacher and the interactions of some teacher characteristics and some student characteristics have been researched. Is the gender of the teacher, that is interacting with boys or girls, a significant factor to include in the influence of work on academic achievement? Teacher's sex and student's sex interactions were studied by the Thomas L. Good Center for Research in Social Behavior. The differing levels of achievement in the classroom of boys and girls in junior high school classes were investigated in two male and two female teacher's classes. Three-way analysis of variance was made using teacher sex, student sex, and subject matter. "Inspection of the variables involved confirmed the conclusion that sex bias [of the teacher] was not a factor affecting teacher-student interactions in these classrooms" (Good et al. 1973, p. 78). Sex differences were found in classroom interaction patterns to be m.ostly due to student sex and not the sex of the teacher. Teachers were found to be primarily reactive in their teacher role to the different interactive practices that boys and girls present to the teacher. The teacher's role, as the adult in charge of the class, and not the teacher's sex was found to be the most significant factor. The sex of the student was significant. The quality of classroom life, as reflected in achievement, for high-achieving boys was found to be vastly different

PAGE 34

26 from the experience that low-achieving boys had. The difference between high and low girls was found to be much less than the difference between high and low boys (Good et al. 1973, p. 85-86) . The relevance of students' participation in school activities as an influence on working and academic achievement has been viewed by some as significant in determining the quality of high school student life (Yarworth & Gauthier, 1978, p. 336). The related argument holds that high levels of participation in the "third curriculum" (i.e., cocurricular activities) are associated with high academic achievement. The alternative, inability to participate in the "third curriculum" because of working, has been assumed by some authors to result in lowered academic achievement (Williams, 1967; Young, 1980). The "normal" participation of students in school activities was tentatively established and measured by Joseph S. Yarworth and William J. Gauthier, Jr. (1978) in a study of 459 high school students drawn from five Pennsylvania high schools. The relationship between various aspects of student self-concept and student participation in the extra and cocurricular activity programs of several Pennsylvania high schools was explored. It combined psychological with personal variables in its examination of student participation in the extra and cocurricular activity programs, both athletic and non-athletic. Some findings of the study were as follows;

PAGE 35

27 Not only was there a difference in the self-concept scores of high-and-low-f requency participators, but this difference is significant for all categories of student activities. Large numbers of students do not participate in the extra and cocurricular activities of the school. The study established the importance of selfconcept in the relationship between academic achievement and participation not only in athletics but in nonathletic activities and the school activity program as a whole. The study dispelled the myth that school activities appeal equally to every student and that school activities are used by large numbers of students to complete the high school experience. (Yarworth & Gauthier, 1978, p. 342) Greater participation in school activities might relate to academic high achievers in some ways (Yarworth & Gauthier, 1978, p. 335). Furthermore, "the lack of participation in school activities is a significant characteristic of the dropout" (Bell, 1967, p. 251). Participation in cocurricular activities cannot be inferred to be a causative or a regulatory agent for high academic achievement (Yarworth & Gauthier, 1978) . The attitudes of students toward work or academic achievement are recognized as being influential. A survey of students' attitudes was beyond the scope of the study reported herein. However, in addition to the previously noted sex variable a variable relative to extracurricular school activities was used. The possibility of the confounding effects of chronological sequence, interval, span of consideration, and year group (or generation) comparison was researched by Ferguson and Maxey (1978) , who studied the trends in the academic performance of high school and college students. They found a substantial increase in the grades awarded by high

PAGE 36

28 school and college faculty over a 10-year period, 19661976. The ACT test scores for college freshmen were concurrently on the decline. They concluded that academic achievement improvement, as measured by grade point average differential over time, may show gains based on relaxing of harsh grading standards rather than factors that would be relevant if a shorter interval of comparison was used (Ferguson & Maxey, 1978, p. 509). Other factors affecting academic achievement of students have been grouped by Charles E. Bidwell and John D. Kasarda (1980) . They noted: In short, in attainment models where individual student achievement is at issue, individual student attributes, other nonschool inputs to schooling, and school inputs must be measured at the individual level to avoid misleading results, (p. 426) The influence of both "schooling," the process through which instruction occurs, a structure of actions by students and teachers, and "school," the organization that conducts instruction and distributes school resources to individual students, are recognized as having recently been shown to have significant influence on academic achievement of individuals (Bidwell & Kasarda, 1980, p. 404). To control the possible source of error of shifting grade point averages over time owing to a possible relaxing of grading standards, the study reported herein was kept to the minimum possible span of years. For example, data from all included grade levels was taken for the same consecutive calendar year time periods, as compared to stretching the

PAGE 37

29 study over a 10-year period. The relationship of the rather global concepts of "school" and "schooling" to academic achievement was considered too complex to introduce at this early stage of research relative to academic achievement and part-time employment.

PAGE 38

CHAPTER III PRESENTATION AND ANALYSIS OF DATA As repeatedly noted, concern as to the influence of students ' part-time work on schooling has been expressed in many newspapers, magazines, and professional journals. Little research has been directed toward this concern. The 34,340 students who were enrolled in grades 8 through 12 (junior and senior high school) in the Duval County School System for the 1981-82 school year constituted the target population for the study of the relationship between employment status and academic achievement among students in an urban school district. A grade level stratified random sample of students was selected from this population. Using a researcher-designed instrument and the permanent record files of the school district, the needed data regarding the independent variables and dependent variable were collected. Complete usable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students. Presented in the present chapter are descriptive data by grade relative to students involved in the study, correlation matrices, and the results of the multiple regression analysis of the data. 30

PAGE 39

31 Descriptive Data by Grade Level Relative to Students Involved In the Study Table 1 contains descriptive data in regard to the dependent variable and presumed independent variables about the students by grade level. More specifically, contained in Table 1 are data about the 1981-82 grade point average (GPA-82) i.e., the dependent variable: days present 1981-82, sex, participation in extracurricular activities, ethnic group, participation in a program for the socioeconomically deprived student (SES), Stanford Achievement Test (SAT), self-perception of intelligence (SPIN), grade point average 1980-81 (GPA-81), employment status (Model Iworked yes/no 1981-82), and employment status (Model IIcategory of hours worked) which were the presumed independent variables. As can be seen from, the table, the GPA-82 of students ranged from a low of 1.9728 for grade 10 students to a high of 2.6967 for grade 12 students, within a range of 0 to 4.0 for grade 8 students to a range of .7 to 4.0 for grade 12 students. Days present in 1981-82 ranged from 91 to 180 days for grade 10 students to 148 to 180 days for grade 9 students. The range of mean attendances for all grade levels was 164.6222 to 169.2222 days. The sample according to sex varied from 60% female and 40% male for grade 11 students to 50% male and 50% female in grade 12.

PAGE 40

32 Table 1. Descriptive Data by Grade Level for Students Involved In The Study Variable 8 n=106 9 n = 90 Grade Level 10 n n=92 n=90 12 n=92 GPA-82 (y) Range 0-4.0 .3-4.0 .2-4.0 .5-4.0 .7-4.0 Mean 2.0311 2.1722 1 .9728 2.1911 2.6967 Standard Deviation .9425 .8465 .8581 .7255 .7152 Days Present 1981-82, (x)Range 99-180 148-180 91-180 106-180 139-180 Mean ' 165.9716 169.2222 166.8587 164.6222 166.2609 Standard Deviation 20.4657 12.8726 13.4406 14.5892 ( 1.7607 Sex (x ) Male 50 (46%) 46 (49%) 44 (48%) 36 (40%) 46 (50%) Female 56 (54%) 54 (51%) 48 (52%) 54 (60%) 46 (50%) Extracurricular Activities (x ) Yes ^ 67 (63%) 63 (70%) 48 (52%) 46 (51%) 63 (69%) No 39 (37%) 27 (30%) 44 (48%) 44 (49%) 29 (31%) Ethnic Group (x^) white, Non-Hispanic 67 (63%) 55 (61%) 59 (64%) 66 (74%) 62 (67%) black, Non-Hispanic 38 (36%) 32 (36%) 31 (34%) 21 (23%) 29 (32%) Hispani c 0 1 (1%) 1 (1%) 1 (1%) 1 (1%) American Indian/Alaskan Native 1 (1%) 0 0 0 Asian/Pacific Islander 0 2 (2%) 1 (U) 2 (2%) Participation in Program for Socioeconomically Deprived (SES) (X.) Yes ^ 27 (25%) 25 (28%) 16 (17%) 10 (11%) 12 (13%) No 79 (75%) 65 (72%) 76 (83%) 80 (89%) 70 (87%) SAT Scores (x ) , Range 0-98 0-94 4-99 9-97 6-98 Mean 53.5110 46.5444 51 .8913 55.9111 57 .5000 Standard Deviation 25.6256 27.3763 23.9598 21.0418 24 .5041 Self-perception of intelligence (x ) , Range 1-3* 1-3* 1-3* 1-3* 1 -3* Mean ' 2.2547 2.2444 2.1630 2.2111 2 .3152 Standard Deviation .5175 .5041 .4753 .4369 .4901 GPA-81 (X ) 0-4.0 .1-4.0 0-3.9 .6-4.0 7-4.0 Mean 2.2287 2.091 1 2.0586 2.1711 2 .4826 Standard Deviation 1.0260 .8069 .9260 .7020 .7129 Employment Status (x ) Model I Yes ® 27 (25%) 19 (21%) 25 (27%) 33 (37%) 59 (64%) Model I No 79 (75%) 71 (79%) 67 (73%) 56 (63%) 33 (36%) Model II, category of hrs. worked Not Employed (0) 79 (75%) 71 (79%) 67 (73%) 56 (63%) 33 (36%) 1-4 hours per week (5) 5 (5%) 7 (8%) 5 (6%) 4 (4.5%) 3 (3%) 5-9 hours per week (6) 11 (10%) 5 (6%) 3 (3%) 1 (1%) 4 (4%) 10-14 hours per week (7) 5 (5%) 1 (1%) 9 (10%) 3 (3%) 1 (1%) 15-19 hours per week (8) 3 (2%) 2 (2%) 1 (1%) 8 (8%) 18 (20%) 20-24 hours per week (9) 1 (1%) 1 (1%) 4 (4%) 8 (8%) 1 1 (12%) 25-29 hours per week (10) 1 (1%) 1 (1%) 1 (1%) 4 (4%) 6 (7%) 30-34 hours per week (11) 1 (1%) 1 (1%) 4 (4%) 5 (5%) 35 + hours per week (12) 0 2 (2%) 1 (1%) 2 (2%) 1 1 12%) *1 means below average, 2 means average, 3 means above average

PAGE 41

33 The highest participation in extracurricular activities was found to be in grade 9 students with 70% participation in one or more extracurricular activities. The lowest level of participation was for grade 11 students with 51% participation in extracurricular activities. All ethnic groups were represented, although not in each grade level. Hispanics were not included in the grade 8 group. However, a single American Indian/Alaskan Native was included in grade 8. The indicator of lower socioeconomic status, participation in reduced price or free lunch program, showed highest participation in grades 8 and 9 with 25% and 28% respectively. Grades 10, 11, and 12 had participation rates of 17%, 11%, and 13% respectively. The average SAT percentile scores for the stratified samples were 53.5000 for grade 8, 46.5444 for grade 9, 51.8913 for grade 10, 55.9111 for grade 11, and 57.5000 for grade 12 students. Only grade 9 students fell below the national norm. Students self-perception of intelligence was scaled as 1 of below average intelligence, 2 of average intelligence, and 3 of above average intelligence. The grade level mean scores of 2.2547 for grade 8 students, 2.2444 for grade 9 students, 2.1630 for grade 10 students, 2.2111 for grade 11 students, and 2.3152 for grade 12 students show that the grade level groups of students perceived themselves generally to be slightly above average intelligence. When the SAT scores are compared to self-perception of intelligence.

PAGE 42

34 except for grade 9, this opinion cannot be contested. Grade point average 1980-81 (GPA-81) had a range from 0.0 to 4.0. The means were 2.2287 for grade 8 students, 2.0911 for grade 9 students, 2.0586 for grade 10 students, 2.1711 for grade 11 students, and 2.4826 for grade 12 students. These means show the academic achievement level for their previous school year to be slightly above for all grade groups. The data about Model I employment status (a dichotomous yes/no response) indicated that 25% of grade 8 students worked more than one-half of school year 1981-82, 21% of grade 9 students worked more than one-half of school year 1981-82, 27% of grade 10 students worked more than one-half of school year 1981-82, 37% of grade 11 students worked more than one-half of school year 1981-82, and 64% of grade 12 students worked more than one-half of school year 1981-82. Model II (the number of hours worked each week by students) showed that of those students who worked in grade 8, 10% worked 5-9 hours per week and only 1% worked as much as 30 34 hours per week. The greatest percent (8%) of grade 9 students who worked, worked, only 1-4 hours per week; 5% worked 5-9 hours per week, 2% worked 15 19 hours per week, 1% worked 20 24 hours per week, 1% worked 30 34 hours per week, and 2% worked 35 or more hours per week. The proportion of grade 10 students who worked 1-4 hours per week was 6%, 5-9 hours per week was 3%, 10 14 hours per week was 10%, 15 19 hours per week was 1%, 20 24 hours per week was 4%, 25 29 hours per week was 1% 9

PAGE 43

35 30 34 hours per week was 1%, and 35 or more hours per week was 1%. Four grade 11 students (4%) worked 1-4 hours per week, 1 student worked 5-9 hours per week, 3 students (3%) worked 10 14 hours per week, 8 students (8%) were found in each category of 15 19 hours per week and 20 24 hours per week, 4 students, (4%) worked in each category 25 29 hours per week and 30 34 hours per week, and 2 students (2%) worked more than 35 hours per week. Grade 12 had the highest overall percentage of students who worked part-time — 64%. Thus, 36% of the grade 12 students did not work part-time. Three percent of the grade 12 students worked part-time for 1-4 hours per week for more than one-half of the school year, 4% worked 5-9 hours per week, 1% worked 10 14 hours per week, 20% worked 15 19 hours per week, 12% worked 20 24 hours per week, 7% worked 25 29 hours per week, 5% worked 30 34 hours per week, and 12% worked part-time 35 or more hours per week. Summarizing Model II (categories of hours worked per week) , grade 8 students ranged from 1-4 hours per week through 30 34 hours per week, with a mean of less than 4 hours per week. Grade 9 students had the same range with a mean of less than 4 hours per week. Grade 10 students ranged from 1-4 hours per week through 35 hours or more with a mean of less than 4 hours per week. Grade 11 had the same range and a group mean below 4 hours per week. Grade 12 students also had the same range with the group mean being between 4 and 5 hours per week of part-time employment. The percentage of students who did work were working

PAGE 44

36 sufficiently longer hours per week to raise the mean for the entire group to more than 4 hours per week. Correlation Matrices In order for the reader to have a better understanding of the impact of the several presumed independent variables on the presumed dependent variable, in this section the correlation matrices among the independent variables and between each of the independent variables and the presumed dependent variable are presented. By analyzing these correlations between the several variables the reader can understand the amount of "overlap" among the variables. If the correlations are high the different variables are measuring somewhat the same thing; if they are low then different things are being measured. Ideally, when dealing with predictor variables in regression analysis low correlations are desired. In Tables 2 through 11 the correlations within each grade level are identical except as they involve the employment status (x ) , of Model I or Model II and the dependent 9 variable (y) . The negative values of some correlations are readily seen herein but are not so apparent when observing the multiple regression data. The greatest negative correlation (-.3113) in grade 8 independent variables occurred between ethnic group and Stanford Achievement Test (SAT) percentile scores. The greatest positive independent, variable correlation (.5907) occurred between SAT percentile scores and previous grade point average (GPA-81).

PAGE 45

Table 2. Correlation Matrix for Eighth Grade Group, Model 37 CN in CN 00 o a^ O o rH o rH in CN in o ro CN O rH o 0) i-H X X X X X X X X X 0 (T3 Vj m OJ 'O CJ c \ o in r-H QJ U in u JJ a in CU •rH > c o> c V 0) D E in -u •rH 0) fN >1 > CL) •H *H 0 0 c QJ O' 4-1 CO W U -P Vj c 01 O -rH C CT^ 0) CN Vj ---I ID 0 in u VJ i-H rH OJ I-H 00 D > O 3 yi OJ I-H 00 E M CXi CT» U -H O 0) 4J (V a OJ (Ti >H in rH »C -U •»H O <0 nn 0) 1 -P rH O 3 I-H to u o c -rH 4-1 Vj OJ C I-H 4-1 OJ >1 X 4J < s: o in tH 0 I-H I-H < CL <0 TJ 03 0) X 4J 0 < o QJ Ph E -P O D in u w in in in in o u CO e: o o o o < CM 00
PAGE 46

38 rCN in CN ro CN O o cr* O o fH fH H* o ”>1 o H in CN m O 00 fH o O iH o CN in CN o • • • « • • • • • • 1 1 I fH o 00 O in ro fH O ro o o ro VO iH m in CN 00 o fH o fH O O fH fH o • • • • • • • • • 1 1 1 1 1 iH 00 fH m 00 n' o fH 00 fH fH (T» o CT> o CO CN CN VO cr» VO o M 00 CN o fH fH CN in CN o M X • • • • • • t • I 1 fH d) VO in CN m O VO m CT^ X o 0 CN LO rO 00 in o s: rH o o fH o ro o X • • • • • • • 1 rH a D m o c^ CN 00 o 0 r-H ro m fH 00 o U ro m VO fH 00 o o >x> CN o o ro fH o X • • • • « • , fX V4 n3 X >4 o Eh to to (D Ti c X to C CN o a> D e X •H dj (N to > 0) CD •H -H 0 0 c 0) C7I X 00 CJ W OV i4 X u c d> O -H C CTi 0^ fH >4 -H KD O w o V4 iH fH (1) X M CN u o > O D J4 (1) X CO E tH CO Oi V -H O 0) X (U a 1 in (Ti (0 X •H 0 to Oi a; 1 X fH O D fH fH w V4 O c •H X »4 X c fH X o Q CO u u w CO CO CO u cd CO s: o

PAGE 47

Table 4. Correlation Matrix for Ninth Grade Group, Model 39 VO ro CT^ 00 00 rH in m o CO cn o 00 o ro (>1 f— 1 rH CM CO I — 1 oo CM rH rH O o ro ro • • « * • 1 • • • • o rH VO ro 00 o CD m rH o o a\ cyv o C \ (D c a 0) 3 E •H 0) CN W > 1 (V (N VJ -rl o o w o >J fH rH QJ H 1-1 00 D > O 3 i-l (1) H 00 E H &i 1 W ro 4J •H 0 <0 Of 01 1 4J rH 0 3 fH w U O c •H u 44 C fH -U
PAGE 48

Table 5. Correlation Matrix for Ninth Grade Group, Model II 40 f>i CTi CD X X VD X in X X n X CM X X m ro ro IT) o 00 i-H •H rH fH • • • ro o m VO ro iD rH r~~i o O • 1 • I • 1 ro VO VO o m 00 oo CN o rH • « • 00 00 m m o rOs nH VO »— 1 O ro • • 1 • CN VO rro rH 00 ro rH o rH fH • • 1 • 1 m o\ ro rro VO 00 cr> o o o • • • VO rH iH rH rin o 00 iH rH iH f • • O 00 O VO 00 o o ro o fH o o o o m o o o o o o o o o CTN 00 00 rH CO o CT* o 00 O in o CN 00 . (V X X X X X X X X X * — 1 Vl XI Vi o m ro 0) TD 0) cn •iH 1 — 1 V 1—4 0) o 0) Vi 4-) D to a •rH •fH > c 4_) m c O •H 0) CN ro > 0) •H *H 0 0 c Vl c 0) O -r4 c a^ 0> Vj -i-i u 0 to o V 1-4 rH (D rH M CN Vi 3 > O 3 V , to ro 4-) •rl 0 ro cv (U 1 4J rH 0 3 iH iH to V U4 C fH 4-1 01 >iCD X 4J < o to E-I o fH M < a ro T! < ro cTi
PAGE 49

41 ro ro On CM CM rCM ro o 0 O \£> no U-l rin o o 'T rH o o o X X • • • • •rH 1 fH Vj 4J o <0 ro rH o s: m in o m o rH o c X • • • o 1 rH fH 4J o (C rH o 1 — I 00 o X • 1— ( d) r-^ XI »0 E-i w I-H CM ro in lO r-* 00 o^ CD X X X X X X X X CO X <>1 rH D XI u XJ (0 fO c CO M (0 C o a; D e. X) •H (U > OJ •H ’rl 0 0 C (V cr> X) I-H w U 4J u c 03 O -H C 03 0) U -H u 0 tn o rH rH 03 CN V-i o > O D IH (D rH 00 e 0 00 (X is: o (Ti CO ro JJ •H 0 10 Oi 0) 1 4J #H o c I-H m CT\ U O c •H X> u iH C iH CN \ '-H X . 4J < x: O CO Eh 0 rH M < Cu w < 03 1 CJ3

PAGE 50

Table 7. Correlation Matrix for Tenth Grade Group, Model II 42 m m o O in rrH rcr» CM O 00 r— H o rH CM (N in ro o X • • t • • • • • 1 1 rH m in i-H CO ro CM o r-* CM CO 00 O o ro o^ V£> in VD o o r— 1 rH O rH CM o X • • • • • t • 1 1 1 rH CM VD 00 O cn in in in 00 o rH in 00 in o fH o rH o X • • • • • . 1 1 rH CO 00 o (N r00 o o m ro o in O o m o X • • • • . 1 rH c^ in o o in ro rin o o r~^ o o o X • • • . o m o m in o m o rH o X • • . 1 rH O o GO o CM O o X « • rH o o o r— ( p X • »— f in CM m in OO o X X X X X X X X 1— 1 )-l (15 (0 <15 'O <15 rH o rH <15 o 4J D in a *iH •P > c (0 c O ID D e 4-1 •H d5 > 05 •rH "rH o 0 c 05 cn W Vj -P u c <15 C5 *H 0) Vj -h o o in u P rH rH Vj D > O D p <15 ( — 1 CO CM O -H o <15 -P <15 a <15 CJV 00 (13 4J •H 0 (0 Cli 05 1 -P rH in 1 >— 1 X .p < x: o ca Eh 0 rH M < (0
PAGE 51

43 m CO rH m VD rH ro CN r00 00 rH CO rH 00 00 o ro CD ro VD in 0) O CTi CO o r— I in VD o u CN r'D o (N o O o V4 X • • • • 0 1 1 — 1 U-l ro in o X LTJ CO o •H LT) ID o u n rH O o 4J X • • t nj 1 rH S in o c (N o 0 in o •H c CO IH c O dJ D E 4J •H o > (D *iH "rH o 0 c 0) cn P rH W JH 4J p c H -H o 0 tn O >H rH rH 0) t3 Z3 > O 3 >H CJ rH 00 E O 04 CN O -H o 0) jj 0) a H o c •r< 4J p IP c rH (N \ X -p < x: o to Eh 0 rH M < OrCO W CD OJ X 4-1 0 < O 0) 04 E CTi ID Q cn u u CO CO CO CO o ta rH >1 o o o o <>1 CN CD < 04 o

PAGE 52

Table 9. Correlation Matrix for Eleventh Grade Group, Model II 44 ro ro rH VO rH ro (N o 00 CO rH 00 (TV in in n o ro VO VO rH in VD o CvJ o rH O o CN VO r — i o • • t • • • • • • • 1 rH in CM ro CM 00 VO in 00 o rH in CM CM ro fO rH CM o a^ ro in ro VO OV ro o ro ro 00 ro cn rH CO 00 o r' o 00 ro VO in cr* ro ro o rH rH o m o X « • • • • • • 1 1 1 rH o CM ro 00 o ro VO 00 ov o o VO CN rH o KD o o rH O c cn fD c O 0) D e 4J •H 0> > o •H *H O o c 0 ) cn 4-1 w >J 4-> c (U O -H c 0) -H o 0 w V ^4 rH rH 01 IH >, CM D > O D Vj (U rH 00 E M W <30 CXi fN O -H o dJ 4J 1 o (TV 00 <0 4J •iH 0 (0 a 0) 1 4J rH 0 r-i cr rH tn u o c •H 4J >4 V4H C rH 0) 01 1— 1 X .4-> < x: O CO Eh 0 iH M
PAGE 53

45 (N CO in ro ro o^ tn 00 m rVO 00 rH VO ro C>, VD rH VO in ro rm o ro rH o uo VO o • • • t 1 • 1 • • • • 1 m o rH o ro o o m 00 ro 00 CM CM o -l X (C (C 0) TD 01 (C3 t— 1 o rH 0) u X jj 3 W a H •fH > c cn w IT3 c O 0) 3 e 4J •H 0) > 01 •H *H o o C 0) X »H w VJ 4J xi c 0) CJ -H C OJ 01 U -H o 0 W CJ Xl r—t rH 01 XI Vj 3 > U 3 u OJ I-H 00 E 0 (X CN O -H o 0) X 0) Cu 0) cn >i5; 0 CO (0 4J •iH O fO 0^ 0) 1 X rH o c W CTv n u c •H X u X c rH CN \ X 4-1 < XI CJ w E-> O X M < DjOO cn (C 01 X X 0 < O 01 Oh E CTv 0) Q U3 u u cn CO CO cn o M X >1 o o o o <>. (N 00 o^ < ClO

PAGE 54

Table 11. Correlation Matrix for Twelfth Grade Group, Model II 46 ro 00 00 m CN ro ro m rH CN CN o CO CM rH ro CN O o X • « • 1 • t • • rH 00 in 00 ro 00 O I-H o O VO o CO rH ro rH rH CN o i-H rH ro O O o X • • • • • I * • rH c CO fl3 c O 0) 3 e x> •H CD •H -H O o c ro cr> 4J W Vj X) u c ro O -H c Vj -h a 0 w u U -H rH CD V-J D > CJ D Vj ro iH 00 Pu CN O •'"1 o ro a ro CJ^ >1 CO fO -U •H o ro di ro 1 -tJ rH 0 w u o c •H X> >4 IH C rH X 4J < s: o w Eh 0 fH H < CD CD 0) X 4J o < u ro PL4 E Q CO u W cn CO CO O 1982 Model II Xg 1.0000 -.0289

PAGE 55

47 These data hold true for both the Model I and Model II tables. The greatest negative correlation for grade 8 students between an independent variable and the dependent variable (GPA-82) occurred between Model I employment status and GPA-82 with a correlation of -.2248. The greatest positive correlation for grade 8 students between independent variables and the dependent variable (GPA-82) occurred between GPA-81 and GPA-82 with a positive correlation of .7813. As shown in Table 3, when categories of employment status (Model II) were incorporated into the matrix, relationships between ethnic group and SAT scores and between SAT scores and GPA-81 held constant; SES replaced employment status as the most negative relationship with the dependent variable. Whether or not these relationships are significant at the .05 level in the multiple regression data is determined by the F ratios established from consideration of all variables interacting among themselves and the sample size . It must be noted that 1982 employment status, whether Model I or Model II, had a negative correlation with GPA-82, except for the grade 9 students. The correlation being most negative (-.2248) was for Model I employment status for the grade 8 students and least negative (-.0289) was Model II employment status for the grade 12 students. The Results of Multiple Regression Analysis The reader will recall the basic focus of the study was to determine the relative impact of employment status

PAGE 56

48 on students ' grade point average when the other variables which are normally considered to contribute to academic achievement have been taken into account. In order to accomplish this purpose a regression analysis was used. In the regression analysis used the variables, with the exception of employment status, were permitted by the computer program to enter into the regression analysis in the order in which they made a contribution to the variation in the presumed dependent variable, i.e., grade point average. Once all these had been entered, employment status was entered to determine the extent to which it made a contribution over and above the contributions made by the other independent variables utilized in this study. The results of the multiple regression analysis for each of the grade levels using both whether the student worked or not and the number of hours worked, i.e.. Model I and Model II, are presented in Tables 12 through 16. In Tables 12 through 16 the variables are listed in order of their contribution to GPA-82 (the dependent variaA ble y) . The contributory relationship of employment in the equation is portrayed as an individual line of data. That is. Model I and Model II employment status should be interpreted from the tables as separate and not as cumulative variables. Only in grades 10 and 11 did employment status make a significant contribution at the .05 level to the variation in grades (GPA-82) . Model I and Model II had a significant relationship for grade 10 students. Only Model I had a

PAGE 57

Table 12 Multiple Regression Data for Eighth Grade Students 49 'D > 0 o 0 • JJ O »H VO Ov 0 cr» in 0 0 ro u "D 0 H3 O • > rH •H •sr CO n rH CD CO CO os cv o> Cv 0 0 VO VO VO VO VO VO VO VO VO VO D CM • • • • • • « • • • e « in 3 0 u • 0 0) Cr» rH in 0 VO m ro ON XJ c o »H CM cn CM iH rH iH 0 rj rH VO m 0 0 0 0 0 0 0 C VO o o 0 0 0 0 0 0 0 rg K O • • x: XJ w 0) >i to *— < 0 ^ JD 0 Vj CO C -H 0 o * XJ C -H ^ «4_! f-H rg 0 ^0 XI •H (U > -D XJ Oi-H 0) 3 > VO VO VO VO VO VO VO VO VO VO 3 0 Vj X) O JQ U tc c o •H a > 41 o a, <0 XJ S 0) C OS X! 0 U c 0 0 •H to 0) Jj 0 Cl. XJ •H QJ XJ U r) VO o VO 0 CO 0 rH ro c t-i Vj o CO 0 m ro in CO 0 3 0 00 o CN CM ro m m ro ro * x: z u 00 00 CO CO 00 CO CO CO s • • • • • • • • OS n> c •H c c 00 r-i in VO ro CM •H X X X X X X X X X X X fT3 E > 1 M 1 M 0 XI 0) 0 XJ m rt3 M Vj CO 0 O 0 rH •*H XJ • 4J • cr» > c XJ •H a •fH I > CO 'D C/3 T? CO 4J rH • H 0 c e 3 XJ 0 -r-f 0 0 0 w c 0 cn 0 0 0 c •H JJ XJ X XJ S rH o •r^ 0> O -r^ w c ij 0 u 0 C C 0 O' U .H 0 0 to 0 U .u c a» CM 0 (M iT3 Ch rg 0 w u 3 U 3 E CO E CD •H Vj a 0 O4 . a^ >iCTi u O 0 1 XJ 00 0 •H a 0 0 ro 0 f—i ITJ Vj 'tJ > U-1 c to o> •r~< XJ c U Vj rH r—i iH > H3 .*0 < *-J M >n»~< 0 W j:: H 0 4 J 3 X a, to Oi w > J-i 0 r3 0 4J < 0 X 0 E Zi E S 3 o CO a C/3 w t/3 CO Da cn DO XJ U XJ

PAGE 58

Table 13 Multiple Regression Data for Ninth (5rade Students 50 0> • tp p o\ VD O' cn ro ro Tf 'O ro Vj -H lO m m vn in in in in in in W Vj W U U Da a; > •p o ro cn r** 00 o 0 H o u c o o iQ p CO o c^ (-H rH rH OJ «T3 CN o CN OJ m ro ro ro ro ro CJ iC \D VO VO VO VO VO VO VO VO VO > 3 • • • • • • • • • • ra e 3 CJ VD fH m ro ro rH VO level h o CO VO m TT 04 o o O o 04 CT» o o o c o O o o o o in d: c VO o o o o O o o o o o • • • • • • • • • • s: CJ the o >1 *— I X) U-I 0 C3 W to * c •HOC) C -H Jj vp i-H 0 C3 jO VO o •H T: O Vj -U 0 U <0 c o a > (1) o a TO N orcs: • • • • c CO VO CN X X X X X rH TO 'D TO CC rH TO TO a^ *P > c p rH P ‘H TO w c c TO CJ» TO •rH TO TO TO -P rH 0 Cn O U rH x> 0^ TO u OJ r-< TO P TO a TO •p TO TO a TO 1 p > U Uh C TO TO < O rH M X > P < TO TO TO CJ 10 io w W o ro o00 Ov Ov rH rH rH rH 04 OJ ro ro ro ro ro VO VO VO VO VO VO • • • • • • o in CO o> 04 ro rr in . 1 tp 1 M p TO TO M •p TO P • p • p 1 > •P a to 'O to c TO •p E 3 0 0 TO •P p O 0 P s P s to Vh TO c P c c TO P < o to u TO 04 Q» 04 U 3 u 3 ' B 00 E 00 Cl, CN TO P 01 P TO >1 a^ >i0> CO TO TO o TO •H 0 rH C rH (0 o U rH •p P C rH rH >1^ P 3 TO to x: a, w a w TO X o p 6 3 E 3 Q u to u to P U P C fO s: -p w w <1> C •H 4J D 'H u 4J c o o *

PAGE 59

51 w +j c 0 ) 'V D 4-1 w ( 1 ) 'V 03 >4 U £1 4J c (W Eh >4 O 44 (0 4J (0 O c o •H w w 0) >4 cn i-H D S 0) rH jO fO Eh • O’ • 4J to oo rn ON 00 •H o o o in VO VO VO 00 4J to 00 in in in in VO 00 m m CO . rH U-t J3 (n CP o ns U O * c c •H 0 « * * * * •K •«( •H C --I W 44 X! x> o m ro c VO o m VO VO VO CO 3 •H -H to CO m m in m VO £ OJ O o u m 00 CM in 00 •rH u c Jj XJ fO in m VO VO VO VO VO VO VO CO Vh 0 c c > CO x> a-H C 3 VO C o vj t; 0 • o U (0 c u o Cm 01 u a ^0 OJ CM 0) w u; o JZ c X) 0 (N in m rH CO Ov Ov o •H rH in o rH rH rH rH rH cc CL »H VM •H to 4J U t-i U rp 3 0 c e: u •rH c c: •rH to e OJ C CO to rH m CM m OV o> Vh X X X X X X X X X X X to >. 1 tH 1 t-H 0» rH 0) rz (V x> fO C3 M fH 00 rH o o •r^ a x> • x> • X! CTS •H 4J > c c. 1 > •H to t: to ’V to 4J tH 4J c • H <>) 3 a -H e 0 o •rH to c c o 0) cn o •H 4J o X) S Jj s u -l rH o >4 < 0 to a; CM 0) CM > Xi Qu r? u u 0) rH 3 U 3 B CO E CO fO u >0'. >, On •H c u w a^ VW c c U rH •H X) rH rH fO < H O rH M x: X) 3 X o to a to D. W 4c > Vj < V c (D X) X 0> o B 3 E 3 in in c to to to to to CO x» CO X> >1 oJ •H > o < u (C • 'C D -H O Vj e O IT3 «3 £ rH a (D x: > 4J C» u-l O O; • JZ 4J
PAGE 60

52 in m Eh V. lu 0 aJ CO CO fH m CN fO CO CM rOn 00 o o ro o in in in in CN in (N in (N in m in m in ro in (N in c/5 u (0 W 4J C a; n 4-) W dJ T) (0 i-i U , C 0) > (V Cd i-i O 4-1 (0 4J (0 Q C o *H w w a; Di 0) « (U a H 4J f-H D S u (U > •H 4J ro rj E 3 U o: CT» C >, <0 Vj CO C *H 0 O H Vj UJ f-H <0 0) > 'D 4J fH «3 Qj rH #-l o ^ U rH Vj 3 O s u ce; o ro On O CO (N CO c CO iH VO in X X X X X rH a> CO u OS •H 4J iH c 4J B to c o c c o H a’ 10 CJ c fH 0 Ds 0> o 0 to XJ Oj IT3 u u O 3 U Oh 10 ON u •H JJ Q fZ < *H Eo U CO > u ra < o O O Q CO oo CO TJ O > c •H d; 0» Ds 0 •*-< Vj fH cj ^ Cl. U-J c »H M (U w . (0 M a 4-1 4J t XJ • 1 c/5 'D CO TJ 3 o > 0 C o »H *H XJ 2: XJ s: Vj Vj xJ c c o Vj O flj CN a; CN :3 < E CO E CO o o >1 CN >tOS H CC Vj 0 H 0 »H c Vj ru fH *H x: 4J fH a 10 a CO 4J X 3 E 3 E 3 £l} U U 4J CO XJ T? O E fO 0^ > o in o <1; -C 4J c CO ji: CO to 0) cn c 4J 3 4J C o Q) (0 O JC u
PAGE 61

Table 16 Multiple Regression Data for Twelfth Grade Students 53 'XD o > • IW JJ o in rH 00 00 VO rH CM o C CM m CM in O O CM fc c e m fH o o o o o rH rH rH q; Vj -h in in m m in in in in in in u 0 4-> • W Vj 0) c V4 U o u fO o s: > •r~i rH o as oo Oi in 00 rH CN in 00 o CM rH p^ O cs 05 o r^ m in in m in in in in o D • rH B D m u o a; 0) O' rH as r*^ cn in m as CM VO as -C c in a^ CM CM o rH O CM JJ OJ o 00 rH iH o o O o o O (N JZ o o O o o o o o o c 05 U 03 x: JJ (1) >1 w iH (A n (V o *H m 00 ro CN TT rH p^ AJ V o a\ rH (N m m ro fO ro cri >-i c 4J 4J (U in m in in in in in m c 0 • • •H a O D JJ O Vj T3 O D U r3 c o a, > 0) o H 04 <0 Vj N (1) JJ o: 'D C o c o o V£) in V£> m o CM as CO •r^ \£> in as VO 00 o rH rH m CO 03 1 M M a CJ T3 C •rH 1 > a CO T? c/: T3 Vj 4J 4J c •H to Pu CM a oj 0) 4J U JH o >1 as >4 Os 03 •H O (1) a. Q) 00 1 4J o ca «cj •»H o rH 0 rH •rH Vj nj > to o> iw C •rH OJ Vh rH c rH rH 5h O n < &H O >--1 rH M X O CO 4-> D x: a to a w ca > Vj < u (D 0) 0 X B D fc D > o w cn Q w CO CO u u hi Jj u JJ K

PAGE 62

54 significant relationship for grade 11 students. In grades, 8, 9 and 12 neither Models I or II employment status made a significant contribution to the variation in GPA-82. Grade Point Average 1981 (GPA-81) was significant at all grade levels accounting for .4051 of the GPA-82 variance at twelfth grade to .6104 at the eighth grade. SAT percentile scores were the next most influential variable for grade 9, 10, and 12 students. Self-perceived intelligence was second for grade 8 students. Days present 1982 was second for grade 11 students. Days present was statistically significant at the .05 level for grade 8, 10, 11 and 12 students. Only in grade 10 was ethnic group of significant statistical influence.

PAGE 63

CHAPTER IV SUMMARY, CONCLUSIONS, AND DISCUSSION Summary The problem of the study was to determine for students in each grade level, grades 8 through 12, in a single urban school district the relative impact of part-time employment to their 1981-82 grade point average (GPA) after the contributions of sex, ethnic group, socioeconomic status, school attendance, GPA for previous school year (1980-81), self-perceptions of intelligence, extracurricular activity participation, and SAT overall percentile score were taken into account. Specifically, the following steps were taken: 1. A determination was made of the relationship between the independent variables of GPA 1980-81, sex, ethnic group, socioeconomic status, school attendance, se 1 f -per cep t i ons of intelligence, extracurricular activity participation, and SAT percentile scores and the dependent variable GPA 1981-82. 2. After determination of the relationship of the above listed independent variables and the dependent variable, the independent dichotomous variable 55

PAGE 64

56 of student part-time employment (yes or no-Model I) was introduced and its contribution to the relationship, if any, was determined. 3. After determination of the relationship of the independent variables listed in #1 to the dependent variable, except part-time employment, the extent of part-time employment (number of hours worked — Model II) was introduced in place of Model I and its contribution to the relationship, if any, was determined . The study was confined to those students in a single urban school district for which complete usable data could be obtained. More specifically, complete usable data were obtained from 106 eighth grade students, 90 ninth grade students, 92 tenth grade students, 90 eleventh grade students, and 92 twelfth grade students for a total of 470 students. Data in regard to 1981-82 GPA, school attendance, sex, ethnic group, SAT scores, SES, and 1980-81 GPA were obtained from the Duval County School Board SIMS computer data base and student cumulative record files. All data were verified with the student permanent record files maintained at the school that the student was attending. The data source for information pertaining to hours worked, extracurricular activity participation, self-perception of intelligence, and participation in free or reduced price lunch program was a researcher-designed student information questionnaire.

PAGE 65

57 To determine the contribution of the several presumed independent variables to grades multiple regression analysis was used. The nature of the multiple regression was stepwise for all the variables except the employment variable. For each grade level two multiple regression equations were run, one using employed or not (yes or no-Model I), the other one employment status by category of hours worked (Model II) . The output of the analysis included means and standard deviations for each variable, correlation coefficients among the variables, the proportion of the variance accounted for by each of the entire models, the proportion accounted for by each variable in each of the models, and the standard error of estimate. The major findings emerging from the data analysis were as follows: 1. In regard to the descriptive data collected from the students, it was found that the 1981-82 grade point average of students ranged throughout the entire scale of 0.0000 to 4.0000. The means were slightly above the theoretical average (2.0000) for all grades except grade 9 where it was slightly below the theoretical average with a mean of 1.9728 . Days present ranged from a minimum 91 for grade 10 to a minimum of 147 days for grade 9 students. The maximum attendance of 180 days was achieved by all grade groups. The sample according

PAGE 66

58 to sex varied from 60% female and 40% male for grade 11 students to 50% male and 50% female in grade 12. Participation in extracurricular activities was found to range from 70% participation in one or more activities in grade 9 to a low of 51% participation in one or more activities in grade 11. All ethnic groups were represented in the sample. Socioeconomic status (SES) was indicated by whether or not the student participated in the reduced price or free lunch program. This participation ranged from 11% in grade 11 to 28% in grade 9. The mean Stanford Achievement Test (SAT) overall percentile scores ranged from 46.5444 for grade 9 to 57.5000 for grade 12; only grade 9 students fell below the national norm. Selfperception of intelligence (SPIN) was scaled, 1 of below average intelligence, 2 of average intelligence, and 3 of above average intelligence. The grade level groups of students perceived themselves generally to be slightly above average intelligence. The 1980-81 grade point average ranged from 0.0000 to 4.0000. The range of grade group means was from 2.0511 for grade 10 to 2.4826 for grade 12. The proportion of students employed (Model I employment status) ranged from 21% in grade 9 to 64% in grade 12. In regard to the extent of employment (Model II employment status) , it was found that 10% of grade 8 students worked 5-9 hours

PAGE 67

59 per week, 8% of grade 9 students worked 1-4 hours per week, 10% of grade 10 students worked 10-14 hours per week, 8% of grade 11 students worked 15-19 hours per week, 8% of grade 11 students worked 20-24 hours per week, and 20% of grade 12 students worked 15-19 hours per week. (These percentages were the largest by category for each grade level group.) 2. In regard to the correlations among the variables, it was found that the correlations for grade 8 students ranged from a negative correlation of -.3113 between ethnic group and SAT overall percentile scores, to a correlation of .7813 between 1980-81 grade point average (GPA-81) and 1981-82 grade point average (GPA-82) . Correlations for grade 9 students ranged from a negative correlation of -.1825 between ethnic group and Model II employment status to a correlation of .7801 between GPA-81 and GPA-82. Grade 10 correlations ranged from a negative correlation of -.4856 between ethnic group and SAT percentile scores to a correlation of .7313 between GPA-81 and GPA-82. Grade 11 correlations ranged from a negative correlation of -.2198 between SES and SAT percentile scores to a correlation of .6553 between GPA-81 and GPA-82. Correlations for grade 12 ranged from a negative of -.4563 between ethnic group and SAT percentile

PAGE 68

60 scores to a correlation of .6365 between GPA-81 and GPA-82. The correlations between employment status and 1981-82 grade point average were negative in value for all grades except grade 9, even though only in two instances were these significant at the .05 level. 3. In regard to the findings emerging from the multiple regression data, the predictive value of independent variables, less part-time employment, varied greatly. The 1980-81 grade point average (GPA-81) was significant at all grade levels accounting for from about 41% of the variance in grades at the twelfth grade to about 61% at eighth grade. SAT overall percentile score was the second most influential variable for grades 10 and 12 students. Self-perception of intelligence (SPIN) was the second most influential variable for grade 8 students. Days present during 1982 was the second most influential variable for grade 11 students and it was statistically significant at the .05 level for grade 8, 10, 11, and 12 students. Only in grade 10 was ethnic group of statistical significant influence. In regard to the predictive value of independent variables including Model I employment, only in grades 10 and 11 was part-time employment found to be statistically significant at the .05 level. Its contribution was negative and

PAGE 69

61 was by far overshadowed by GPA-81, days present 1982, and SAT percentile scores. In regard to the predictive value of independent variables including Model II employment (categories of hours worked), only in grade 10 was it found to be statistically significant. Again, its contribution was negative and was by far overshadowed by GPA-81, SAT percentile scores, days present 1982, and SPIN. 4. In terms of the specific contributions of employment status to an increase in the percent of variance in the dependent variable (GPA-82) , it was found that in grade 8 Model I employment contributed .0019% of a total of .6975% and Model II employment status contributed .0007% of a total of .6975%. In grade 9 Model I employment contributed .0001% of a total of .6319% and Model II employment status contributed .0006% of a total of .6324%. In grade 10 Model I employment contributed .0112% of a total of .6868%; Model II employment status contributed .0128% of a total of .6884%. In grade 11 Model I employment contributed .0221% of a total of .5253% and Model II employment status contributed .0133% of a total of .5165%. In grade 12 Model I employment contributed .0076% of a total of .5415% and Model II employment status contributed .0029% of a total of .5378%.

PAGE 70

62 Conclusions In regard to the basic focus of the investigation, the extent to which part-time employment has an impact on the grade point average of high school students, the conclusion is that the impact of part-time employment on the academic achievement of the studied students is minimal. This conclusion seems to be justified, because in only very few instances was part-time employment found to make a significant contribution to the variation in grade point average. More specifically, in grade 10 it was found that it made a significant contribution in the sense that the students that worked made poorer grades, but this contribution followed the contributions of GPA-1981, SAT percentile scores, days present 1982, SPIN, and ethnic group and contributed less than 2% to the approximately 69% of the variance in grades accounted for by all of the presumed independent variables combined. In grade 11 it was found that part-time employment made a contribution after the contributions of GPA-81, days present 1982, and SAT percentile scores were determined. The contribution was about 2% of the approximately 53% of the variance in 1981-82 grades accounted for by all of the presumed independent variables. In the other instances studied there was no significant impact either positively or negatively. Thus, the conclusion is reached that it has minimal impact. Even though not central to the basic focus of the investigation, the other major conclusion is that the best predictors of students ' grade point averages are previous

PAGE 71

63 grades, SAT percentile scores, and attendance. If one reviews the studies reported in Chapter II, it shows that in numerous studies, e.g., Gade and Peterson (1980), Hammond (1970), Bidwell and Kasarda (1980), Finn, Dulberg, and Reis (1979), Nuttall (1972), Rehberg, Schafer, and Sinclair (1970), Bailey and Bailey (1974), Kobett (1979), Schab (1979) , Good (1973) , and Yarworth and Gauthier (1978) , the importance of these factors has been noted. Thus, this conclusion, which is peripheral to the central purpose of this investigation, reconfirms the previous studies which lead to the generalization that these variables impact most heavily on grade point averages. Discussion The reader will recall that a justification offered for the study was that there was a great deal of controversy about whether, as a policy matter, students should be encouraged to seek part-time employment during their secondary school years. Proponents of one view contended that grade point averages of teenage students in junior and senior high school who worked part-time were lower than those of similar students who did not work. That is, part-time work hindered academic achievement as measured by grade point average. Central to this point of view also was the opinion that students required to work while attending school were denied active involvement in school-related activities. Justification for this belief rested on reports which suggested that involvement in cocurricular activities

PAGE 72

64 enhanced essential social development and that the "values orientation" from such activities was lacking in students who worked while attending school. (Williams, 1967, p. 36; Young , 1980 , D-3 ) . Advocates of the opposing view argued that part-time employment prepared students to become self-sufficient wage-earners in society's work force. These educators held that students' work experiences helped them raise their living standard and aided society by reducing unemployment and increasing the gross national product. A survey of employers revealed that 68% favored youth work experience. Resultant benefits included development of maturity, selfidentity, self-reliance, career direction, work relationships, and work enjoyment. (Delaney, 1975, VI-14). Family financial situations sometimes required students to work. Variables dictating the need to work have ranged from low family income to divorce, death of the family wage earner, or abandonment by one or both parents. In addition, students often worked to relieve peer pressure to participate in school-related activities that involved substantial financial commitments. Others sought employment as a convenient means to obtain luxury items. Student employment programs sponsored by government and private agencies to prevent the termination of education have had a further effect of encouraging students to work part-time during or after school hours (Delaney, 1975; Dykeman, 1979, p. 365; Heyneman, 1977; Levine, 1979).

PAGE 73

65 If the major conclusion of this investigation is accepted as reasonable and valid, the implication for policy makers is that if the concern is with academic achievement there should be no policy discouraging students from working. If working is deemed necessary or desirable for other reasons, i.e., financial hardship, students might even be encouraged to seek a reasonable amount of part-time employment . It is recognized that the conclusions of the investigation reported herein obviously must be regarded as suggestive, not conclusive, because only one urban school district was involved at only one point in time. However, since the sampling procedure ensured a random sample, one can feel fairly comfortable that the data were in fact applicable for that school district at that point in time. Furthermore, if the investigation was replicated and similar results were found, the implication for policymakers relative to the lack of clear linkage between part-time employment and grade point average would be strengthened. Given the situation, when policy makers are faced with a decision of what to do about part-time employment, if action is taken to curb employment or strongly encourage employment, it has to be made on some other basis, other than impact in academic areas. Another argument not germaine to the central focus of this study was, did part-time employment impact on the students participation in extracurricular activities? From

PAGE 74

66 a review of the data collected for study, it can be determined that the correlations between employment status and extracurricular activities participation for each of the grade levels were as follows: Grade 8 Model I .0419 Grade 8 Model II .0340 Grade 9 Model I .0416 Grade 9 Model II .0137 Grade 10 Model I .1935 Grade 10 Model II .1888 Grade 11 Model I .0285 Grade 11 Model II .0323 Grade 12 Model I .0471 Grade 12 Model II .0873 The range of correlations from -.0137 in grade 9 Model II to .1935 in grade 10 Model I does not show that part-time employment had a drastic impact on extracurricular activities. Also, the extracurricular activities variable was not found to make a statistically significant contribution to the dependent variable (GPA-82) at any grade level. Further inspection of the correlation matrices will yield similar relationships for the other variables that were not found among the variables classified as significant. In summary, consistent with the view expressed by Nuttal (1972), Rehberg, Schafer, and Sinclair (1970), the argument that poor students are adversely affected academically because of simply being poor and having to work was not confirmed by these data. If a secondary student is employed part-time, the other contributions to grade point average such as, previous years grades, attendance, SAT percentile scores, and self-perception of intelligence will.

PAGE 75

67 percentile accord ing impact of attending i scores, and self-perception of intelligence will, to the data of the study, far overshadow the students working, or not working, part-time while econdary schools.

PAGE 76

APPENDIX A DUVAL COUNTY SCHOOL SYSTEM BIOGRAPHIC DATA FORM’ cr o 'LL < t< Q 1 Cl < IT O o C3 S LU tLO > LO c o X u w > t2 X o u .1 5 < 5 i •i < 2 w 5 ± < * S Z s ^ 5 ^ i og O V) v> < < s C3 O c UJ uj {/) O UJ g| UJ o < “ UJ UJ CC ® < O o s S ^ w C 2 < D O u CO lu 2 qS ^ UJ D CC ^ 5 c O < o c 5 u UJ — < O 2 O u < CO 2 < c 1*1 •! O • 8 : 5 J J V J 1 r ^ .2 > u “ CJ a t) 5 c < CD U Z M M C C -• CJ ^ 2 'c i O 5 w I ^ ^ r i £ o s ^ ^ c y -• ^ *« c z 2 o i a. u ^ ' 5— — 3 ? d CL *“ !“•=<< I L Z = CO CO S •= z ui i/i 0 2 CD (J UJ uJ w C < > d! 1 j j Z i I ^ 1 0«w0Z-'«XjXZ0 0K»^3 >)K> • V • u • u • u t u s u 9 w 0 w 0 w 0 w 0 w 3 w 0 w 3 u k 9 Z U 9 Z h. 9 Z W 9 Z W u z 9 Z c. 9 X w 9 Z t J X J X J X J X J X J X J X J J Z 0 X 1 Z 0 ^ X Z 0 iX z 0 *. X z 0 kX z 0 »• X z 0 X z 0 0 c 0 c 9 K 0 K 0 c 0 s 0 c 0 s I" •I K 3 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > > X > J X > > X » X t X >• J X I X > i X > • u a V a u a J a u a u a u a w a u a w 3 u 9 w 9 w 0 W 0 U 0 w 0 w 0 w 0 y 0 W 0 y 0 w 0 y y 9 Z u 9 Z y 9 Z w 9 Z k o Z w O Z w 9 Z w 9 X k 0 Z k 9 X k 9 Z k 9 Z k 9 X k 9 Z X J X J X J X J X J X J X J X J X J X J X J X J X z 0 k X Z C k X z 0 k X Z 0 k 3 Z 0 k X z 0 k Z Z 0 k X z 0 k X z 0 k X Z 0 k X Z O k X z 0 k X z 0 k X Z 0 k 0 z 0 X 0 t 0 t 9 X 9 X 9 X 9 X 9 X 9 X 0 t 9 X 9 X 9 X «* h u h M K a 3 > 3 > 5 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > 3 > ) « t X y J X * X > i X y t X y i X > t X y i X y t X y t X y t X > t X y t X y £ 2 VI r X J u 2 o r z c C 2 " 2 g (J u 9 u CO 01 o ^ : 2 «n o f** » 68 JULl

PAGE 77

APPENDIX B STUDENT INFORMATION QUESTIONNAIRE STUDENT PERCEPTIONS, ACTIVITIES, AND EMPLOYMENT Name School_ last first middle Address (Number and St.) ^ City State Zipcode Student Identification Number (if known)_ Please answer the following questions by darkening in the appropriate bubble "Yes" or "No" after the question. Please read all the questions (both sides) before starting to answer the first one. 1. School year, 1981-82, I participated in at least one, or more , of the school activities of athletic teams, band, chorus, orchestra, or other school sponsored clubs; YES 0, NO 0 . 2. In comparison to ray fellow students, I think that my intelligence is; Below Average 0, Average 0, Above Average 0. 3. During the more than one-half of school year 1981-82, I held a part-time job; YES 0, NO 0. 4. During the more than one-half of school year 1980-81 , I held a part-time job; YES 0, NO 0. If your answers to questions Number Three and Four were "NO" that completes the information needed. Please turn in this sheet. If your answer to question Number Three or Four was "YES", please select an hours worked category that is closest to the amount of time you were usually employed each week for more than one-half of that year. 69

PAGE 78

70 5. During the 1980-81, 1981-82 school years, I was paid for doing part-time work (including odd jobs, such as mowing lawns, or babysitting, and so forth) for about 1 to 4 hours per week: 1980-81 YES 0, NO 0. 1981-82 YES, NO 0. 1980-81 1981-82 6. I worked 5 1 to 9 hours per week: YES 0 NO 0 YES 0 NO 0 7. I worked 10 to 14 hours per week: YES 0 NO 0 YES 0 NO 0 8. I worked 15 to 19 hours per week: YES 0 NO 0 YES 0 NO 0 9. I worked 20 to 24 hours per week : YES 0 NO 0 YES 0 NO 0 10. I worked 25 to 29 hours per week : YES 0 NO 0 YES 0 NO 0 11. I worked 30 to 34 hours per week: YES 0 NO 0 YES 0 NO 0 12. I worked 35 or more hours per week: YES 0 NO 0 YES 0 NO 0 Thank you for you help in performing this bit of educational research. Individual identified responses will not be released without your advance permission. On free lunch/reduced lunch program. Yes No

PAGE 79

APPENDIX C STUDENT DATA NEEDED FOR LISTINGS OF SAMPLE DATA BASE 1. Name (last, first, initials). 2. Student Identification Number. 3. Sex (1) male (2) female. 4. Ethnic group: (1) White non-Hispanic, (2) Black non-Hispanic, (3) Hispanic, (4) American Indian or Alaskan Native (5) Asian or Pacific Islander. 5. Year grade level in school during 1981-82 . 6. School attended 1981-82 school year # 7. School attended 1980—81 school year # 8. School attended 1979-80 school year # 9. Grade point average school year 1981-82 10. Grade point average school year 1980-81 11. Participation in extracurricular activities (yes/no). 12. Self-perception of intelligence: (1) below average, (2) average, (3) above average. 13. Number of hours worked per week for most of the school year sorted by categories: (1) none, (2) 1-4 hours per week, (3) 5-9 hours per week, (4) 10-14 hours per week, (5) 15-19 hours per week, (6) 20-24 hours per week, (7) 25-29 hours per week, (8) 30-34 hours per week, (9) 35 or more hours per week. 14. Student participation in any of the specially funded projects in Section B of SIMS Biographical Data Form, (yes/no). 15. The number of days present during 1981-82 71

PAGE 80

REFERENCES Bailey, R. C. & Bailey, K. G. Self-perceptions of scholastic ability at four grade levels. Journal of Genetic Psychology , 1974, 124 , 197-212. Belanger, R. R. & Boyle, R. D. Stepwise multiple regression , Cupertino, California: Apple Computer Incorporated, 1980. Bell, 'J . W. Comparison of dropouts and non-dropouts on participation in school activities. Journal of Educational Research , 1967, ^(6), 248-251. Bidwell, C. E. & Kasarda, J. D. Conceptualizing and measuring the effects of school and schooling. American Journal of Education , August 1980, 401-430. Cole, Sheila, Send our children to work? Psychology Today, July 1980, pp. 44-56; 50-58; 60-68. Crawford, G. & Miskel, C. Experience based career education at Wich ita east high school: A third party evaluat io n. Kansas: Wichita Public Schools, July 1978. (ERIC Document Reproduction Service No. ED 150 285) Delaney, W. The transition from school to work: A study of laws, regulations, and practices restructuring work experience and employment opportunities . Ann Arbor, Michigan: University of Michigan, 1975. (ERIC Document Reproduction Service No. ED 124 762) Dykeman, B. WECEP: initiating a career education program for the 14 and 15 year old student. Education, 1979, ^(4) , 363-367. Ferguson, R. & Maxey, E. J. Trends in the academic performance of high school and college students. Journal of College Personnel , November 1978, I^, 505-511. Finn, J . D., Dulberg, L. & Reis, ' J . Sex differences in educational attainment: A cross-national perspective. Harvard Educational Review, November 1979, 49(4), 477-503. Gade, E. & Peterson, L. Comparison of working and nonworking high school students on school performance, socioeconomic status, and self-esteem. Vocational Guidance Quarterly , September 1980, 65-69. 72

PAGE 81

73 Gill, N. T. Comparison of high school students interests across three grade and ability levels. High School Journal , January 1980 , 160-166. Good, T. L. , Sikes, 'J . N. & Brophy, ' J . E. Effects of teacher sex and student sex on classroom interaction. Journal of Educational Psychology , 1973, ^(1), 74-87. Hammond, W. A. Scholastic achievement and part-time employment. Th e Cl e ar ing House , 19 7 0 , (7), 465-467. H e f f e z , ' J . The effects of part-time employment on high school stu de nt s^ g rade point averages and rate of school attendance . New York: EDRS Research Report, 1979. (ERIC Document Reproduction Service No. ED 173 592) Heyneman, S. P. Six views on three issues related on education and work, report of a symposium in debate form held june 28 , 1977 . Washington, D. C.: National Institute of Education (DHEW) , 1977. (ERIC Document Reproduction Service No. ED 179 687) Kobett, L. A. Relationship between home background, school achievement, and adolescent values. Education, 1979, 100 (2) , 158-164. Levine, S. V. Psychological and social effects of youth unemployment. Children Today, November-December 1979, 8,6-9. Nuttall, R. L. Do the factors affe cting academic achievement differ by the socioeconomic status or sex of the students ? Ann Arbor, Michigan: University of Michigan, 1972. (ERIC Document Reproduction Service No. ED 064 465) Pope, M. C. Students in Duval County have high mobility rate. The Florida Times Union , May 17, 1982, pp.B-1. Rehberg, R. A., Schafer, W. & Sinclair, J. Toward a temporal sequence of adolescent achievement variables. American Sociological Review , 1970, 34-48. Roberson, D. R. Division of research, planning and evalua tion policies and procedures . Jacksonville, Florida: Duval County Schools Printing Office, 1980. Roscoe, J . T. Fundamental research statistics for the behavioral sciences (2nd ed) . New York: Holt, Rinehart and Winston, Inc., 1974.

PAGE 82

74 Schab, F. Adolescence in the south: A comparison of blacks and whites not prepared for the world of work. Adolescence , 1979, ]J^(55) , 599-605. Shively, 'J. & Watts, R. Final evaluation report for the conecuh county part D experience-based career education program . Charleston, West Virginia: Appalachia Educational Laboratory, 1980. (ERIC Document Reproduction Service No. ED 181 221) Skallerud, R. D. Career Education: Planning, learning, understanding, succeeding — ce+, final report, September 1, 1976 to august 31, 1979 . Bismark, North Dakota: North Dakota State Board of Vocational Education, 1979. (ERIC Document Reproduction Service No. ED 179 726) Watkins R. W. & Corder, R. Student outcomes and participant opinions in experienced-based career education schools . external evaluators final report on the experiencedbased career education program, vol. VI . Berkley, California: Educational Testing Service, 1977. (ERIC Document Reproduction Service No. ED 155 181) Williams, N. Part-time jobs advantages and disadvantages. School and Community , January 1967, pp.36;54. Yarworth, 'J . S. & Gauthier, W. J. Relationship of student self-concept and selected personal variables of participation in school activities. Journal of Educational Psychology , 1978, 1 ^, 335-344. Young B. Better education most critical need. The Florida Times-Union , 'January 27, 1980, pp.D-3.

PAGE 83

BIOGRAPHICAL SKETCH Walter G. Squires, Jr., an earth-space science and biology teacher at N. B. Forrest High School, Jacksonville, Florida, has taught science subjects for 5 years spanning a period of 30 years. Mr. Squires received his bachelor's degree in psychology from the University of Florida in 1951, the Master of Education from the University of North Florida in 1976 , and the Doctor of Education in 1983 from the University of Florida. He is a Commander in the United States Navy (retired). He was awarded U. S. Patent No. 3,035,285 for an explosively anchored navigational buoy. He was recognized in 1972 , as one of the foremost authorities in the U. S. Navy in the field of mine warfare. Mr. Squires is the father of three children and has four grandchildren. He is a member of Kappa Delta Pi, the National Society for the Study of Education, the American Legion, National Sojourners, a 32' Mason, and is a disabled veteran. 75

PAGE 84

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. 'M Y/ Wl:nnery, Ed.D., Chairman Profyessor of Educational Administration a/d 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. jJ./W. Longst^'^th, Ed/. D. ) (Associate Professor, 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. iT C. Healy, Ph,^. Associate Professor of Educational Administration and Supervision

PAGE 85

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. V/ yt A . a c o b s' e n , E d . D . Professor of Instructional Lead e^r ship and Support This dissertation was submitted to the Graduate Faculty of the Department of Administration in the College of Education and to the Graduate Council, and was accepted as partial fulfillment of the requirements for the degree of Doctor of Education. April, 1983 Dean for Graduate Studies and Research