Group Title: longitudinal follow up of preschool intervention program for Mexican-American migrant children in primary grades
Title: A longitudinal follow up of preschool intervention program for Mexican-American migrant children in primary grades
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098858/00001
 Material Information
Title: A longitudinal follow up of preschool intervention program for Mexican-American migrant children in primary grades
Physical Description: viii, 110 leaves : ; 28 cm.
Language: English
Creator: Hoffman, Mae Jewel
Copyright Date: 1977
 Subjects
Subject: Mexican American children -- Florida   ( lcsh )
Language arts (Elementary)   ( lcsh )
Foundations of Education thesis Ph. D   ( lcsh )
Dissertations, Academic -- Foundations of Education -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 99-108.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: Mae Jewel Hoffman.
 Record Information
Bibliographic ID: UF00098858
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000210060
oclc - 04168762
notis - AAX6879

Downloads

This item has the following downloads:

PDF ( 5 MBs ) ( PDF )


Full Text













A LONGITUDINAL FOLLOW UP
OF A PRESCHOOL INTERVENTION PROGRAM
FOR MEXICAN-AMERICAN MIGRANT CHILDREN
IN PRIMARY GRADES












By
MAE JEWEL HOFFMAN



















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


UNIVERSITY OF FLORIDA
1977




















To
Ann and Amy
whose encouragement and support have been
very special


















ACKNOWLEDGMENTS


The author wishes to acknowledge with appreciation the members

of her doctoral committee: Dr. Gordon Greenwood, director of the

dissertation, who gave wise guidance and assistance in the execution

of the study; Dr. Ira Gordon, who made possible the opportunity for

the writer to develop the original preschool intervention program from

which this present study evolved; Dr. William Ware and Dr. Michael

Hanes, who gave valuable direction for the research design and evalua-

tion methodology.

The author also wishes to acknowledge the school personnel of

Hardee County, Florida, and especially Mr. Jack limes, who gave time

and assistance in the data collection procedures. Without their

interest and support, this study would not have been possible.

















TABLE OF CONTENTS


Page



ACKNOWLEDGMENTS. ................ . . . ... iii

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

ABSTRACT . . ... ....................... . ...vii



CHAPTER

I. INTRODUCTION . . . . . . . ... ... .. . 1

Statement of the Problem .. . . . . . .. .... 2
Factors Influencing the Decision for the Preschool
Intervention Program .. . .... . . . 3
Summary of the Preschool Intervention Program. ..... 6
The Curriculum Component of the Preschool Intervention
Program. . .. . ... ........ .. 7
Rationale for the Curriculum Component. . . . . 9
Language Component of the Preschool Intervention
Program . . .... ......... ..... 9
Cognitive Task Component Rationale . . .... .10
Experience Component Rationale . . . ... .12
Self-concept Component Rationale . . ... .. 12
Summary of the Curriculum Component . ... . .. 13
The Home-School Interaction Component of the Preschool
Intervention Program . . .. . . . . 13
The Inservice Training Component for the Saff of the
Intervention Program . . . . . .. 14
The Evaluation Component of the Intervention Program. . 15
Results of the Intervention Program Evaluation of
Children's Gaines. .. . ............ 15
Program Continuation by Hardee County .... ...... 16
Significance of the Present Study .. .. . . . . .17
Statement of the Research Question and Hypotheses. .. .... 19
Limitations of the Proposed Study. .. ... .... 21

II. REVIEW OF THE LITERATURE ..... . . . . . 23

Federal Support for Education of the Disadvantaged Child . 24
Psychological Assumptions for Intervention Program Models. 2S












Philosophical Bases for Intervention Models . . . .31
Bilingual Early Childhood Intervention Programs of the
1960's . . . . . . . . .. . .. . . . 36
The Literature of Longitudinal Evaluation of Intervention
Programs ....... . . . . . ... . .41
Summary . . . . . . . ... . . . 50

III. DESIGN AND METHODOLOGY. ...... . . . . . 54

Subject Selection Procedures .... . . . ... 57
Instrumentation. .......... .. ........ 60
Data Selection by the Investigator . . . . ... .63
Statement of the Null Hypotheses for Statistical Analysis. 66
Summary of the Design and Methodology. . . . . ... 67

IV. RESULTS OF THE STUDY. .. . . . . . . ... 68

Consideration of Attendance During First, Second, and
Third Grade for Both Groups . . . . . . .. 77
Consideration of the Research Study. . . . . ... 78
Summary of the Results of the Study . . . . . . 7

V. SUMMARY, DISCUSSION, AND RECOMMENDATIONS. . . . . ... 1

Summary . . . . . . . . . . . . . 81
Discussion. ...... . . . . . ...... 83
Recommendations ....... . . . . . . . ..93
Recommendation One ...... . . . . . 93
Recommendation Two . . .... ............. .94
Recommendation Three . . . . . . .... . 95
Recommendation Four. .. ... . . . . . 95
Recommendation Five . . ..... . . . . . 96
Conclusion. . . . . . . . ... ...... 97

REFERENCES. .... . . . . . . . . . . 99


BIOGRAPHICAL SKETCH . . . ......


.109



















LIST OF TABLES


Table Page

1. MEANS AND STANDARD DEVIATIONS FOR THE PROGRAM GROUP
AND THE COMPARISON GROUP ON THE PRESCHOOL INVENTORY
AT KINDERGARTEN ENTRANCE .. . . . . .70

2. MEANS AND STANDARD DEVIATIONS ON RAW SCORE TOTALS FOR
THE METROPOLITAN READINESS TEST, FORM A, AT THE FIRST
GRADE FOR THE PROGRAM AND COMPARISON GROUPS. . . ... 71

3. MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD
SCALE SCORES ON THE TOTAL BATTERY, COMPREHENSIVE
TEST OF BASIC SKILLS, FOR SECOND GRADE FOR THE
PROGRAM AND COMPARISON GROUPS. .. ... . . . 72

4. MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD
SCALE SCORES ON THE LANGUAGE COMPONENT, COMPREHENSIVE
TEST OF BASIC SKILLS, FOR SECOND GRADE FOR THE PROGRAM
AND COMPARISON GROUPS. . . ... ......... 73

3. MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD
SCALE SCORES ON THE TOTAL BATTERY, COMPREHENSIVE
TEST OF BASIC SKILLS, FOR THIRD GRADE FOR THE PROGRAM
AND COMPREHENSIVE GROUPS . . . . . . .74

6. MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD
SCALE SCORES ON THE LANGUAGE COMPONENT, COMPREHENSIVE
TEST OF BASIC SKILLS, FOR THIRD GRADE FOR THE PROGRAM
AND COMPARISON GROUPS. . . .... .. . .. 75

7. MEANS AND STANDARD DEVIATIONS FOR RAW SCORES ON THE
COMMUNICATION SKILLS COMPONENT OF THE FLORIDA STATE-
WIDE ASSESSMENT TEST FOR THE PROGRAM AND COMPARISON
GROUPS AT THIRD GRADE. ....... . . . 76

8. COMPARISON OF SUBJECTS RETAINED OR PROMOTED AT THE END
OF THIRD GRADE FROM EACH OF THE TWO GROUPS . .. .. 76

















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 Philosophy



A LONGITUDINAL FOLLOW UP
OF A PRESCHOOL INTERVENTION PROGRAM
FOR MEXICAN-AMERICAN MIGRANT CHILDREN
IN PRIMARY GRADES

By

Mae Jewel Hoffman

August 1977

Chairman: Gordon E. Greenwood
Major Department: Foundations of Education

The school achievement of twenty Mexican-American migrant children

in Hardee County, Florida, who participated in a six-month pilot preschool

intervention program when they were three years old, was studied to

determine longitudinal program effects from kindergarten through third

grade. Fourteen Mexican-American migrant children, from highly similar

home environments who entered kindergarten in that county at the same

time as the program group and were in attendance in the school system

during those same four years, served as a comparison group for this study.

The six-month intervention program emphasized a child-oriented curri-

culum, focusing on growth in language, cognition,and self-esteem in

the children. To insure the acceptance of cultural and language differ-

ences, bilingual aides from the Spanish-speaking migrant community

worked with the teachers, thereby facilitating the use of Spanish,

the home language of the children, in both conversation and instruction









and bridging the cultural gap between the home and the school. English

was encouraged and used increasingly as children learned initial

concepts in Spanish, grew in their understanding of English, and became

comfortable with the school language.

The results indicated statistically significant differences in mean

scores in standardized tests in favor of the program intervention sub-

jects at the kindergarten, first, and second grade levels. The program

group appeared to evidence greater skill in language use necessary for

achievement in the Caldwell Preschool Inventory in kindergarten, the

Metropolitan Readiness Test in first grade, and the language component

of the Comprehensive Test of Basic Skills in second grade. By third

grade, the achievement scores on the Comprehensive Test of Basic Skills

and the communication component of the Florida Statewide Assessment

Test showed no statistically significant mean differences. The gains

evidenced prior to that testing period appeared to have "washed out."

Relevant to the study, however, were the number of children retained at

grade level from each of the two groups: six of the fourteen comparison

subjects had been retained with only two of the twenty program subjects

retained. This proved to be statistically significant.

The findings of this study indicated that early intervention into the

learning of Mexican-American children from the migrant population in

Hardee County did make an overall significant difference between experimental

and comparison children, even within the severe time limitations of the

intervention program. Further research is indicated to determine whether

or not a bilingual-bicultural early intervention program of longer duration,

and controlled for longitudinal program evaluation, for children of similar

home and cultural environments could increase even more significantly

the achievement gains of K-3 migrant children.


viii


















CHAPTER I

INTRODUCTION



At a national conference on Early Childhood Education and the

Chicanito in 1972, it was estimated there were 75,000 migrant children

under the age of six who traveled with their families throughout the

United States (Chavez, 1972). Chicanos, blacks, Indians, Puerto Ricans,

and Anglos made up the migrant population that moved in three broad streams

originating in Texas, California, or Florida and moving into and through-

out most of the states. Rivera (1970) reported that approximately 70

to 75 percent of the migrant labor force at that time were Mexican-

Americans. Although national concern for the education of seriously

disadvantaged children had been evidenced with the passage of the Elemen-

tary and Secondary Education Act of 1965, migrant families had not been

included in the 1960 census; therefore, they were not counted in estimating

state funding allotments. In November 1966, Title I of the Elementary

and Secondary Education Act was amended to include the children of

migratory agriculture workers to insure provisions of adequate program

consideration and funding to meet their particular needs (Rivera, 1970).

Under this amended Public Law 89-750, the Florida Department of

Education, Migrant Education Section, initiated a pilot project during

the 1969-1970 school year for four-year old migrant preschool children in

five counties in South and Central Florida. Most of the children

who attended the initial programs were "American Negro, the second group










in size was Anglo-White, and the third group Spanish, both Texas-

Mexican and Puerto Rican" (Curtis, & Klock, 1970, p. 2). The following

school year, 1970-1971, an expanded Early Childhood learning program for

both four and five-year old children was underway in counties with a

high migrant population (Combs, 1971).

It was at this time (December 1970) that the Institute for

Development of Human Resources, University of Florida, in cooperation

with the school authorities of Hardee County, Florida, entered

into an agreement with the Migrant Section, Florida Department of

Education, whereby the Institute would be responsible for the develop-

ment of a curriculum for three-year old migrant children in a pilot

program at the Zolfo Springs Elementary School site. It was the respon-

sibility of this investigator to design and develop the educational

curriculum to be used in that preschool intervention program which was

operational from January 1971 through May 1971.

Statement Of The Problem

Most of the children who entered the preschool intervention program

were from the Spanish speaking Mexican-American migrant population of

that county. It was the purpose of this study to evaluate the early

elementary school performance of those Mexican-American migrant children

who attended the intervention program, and to determine whether or not

there were longitudinal program effects on their school achievement from

kindergarten through third grade.

The preschool program was designed to effect change in the expected

school achievement of the migrant children who attended the program.

Evaluation was made using the criteria of Hardee County public schools

for determining school success through the testing procedures adopted










by that county for elementary children. Whether or not positive changes

in anticipated school achievement could be measured five years after

this intervention program was a corrernof this study.

Factors Influencing the Decision for
the Preschool Intervention Program

It is important to the problem to consider the state and local

concerns for migrant education that led to the agreement for a pilot

early childhood program in that county. Entrance into a formal school

setting was difficult for migrant children who came from restricted

educational environments and who had learned more from the indirect

modeling of peers and older siblings than from parents and other adults.

Similar to the child care patterns of most migrant families, young child-

ren were often left in the care of older children while both parents

worked in the groves or fields. Sometimes, the very young were taken

with their parents and left quite on their own during the work day from

early morning to early evening. Concern was expressed not only for

their safety and well being, but also for the lack of an enriching

learning climate (Combs, 1971).

For Hardee County migrant children, there were additional disadvan-

tages. There was no public school kindergarten program in the county

to help bridge the gap between the "free" home environment and the

traditional first grade classroom environment. Also there was the

fact that a large part of the migrant population was Mexican-American

with Spanish as their native language, and in most instances, their

only home language. Typically, Mexican-American children entered first

grade in that county with limited language facility, even in Spanish.

For most of the children, English was not at all a part of their home

world; for a few of the children, their language was neither Spanish nor

English, but a hybrid language referred to as "Spanglish" (Laosa, 1975,

p. 625).










Once in school, and with English as the only language used for

instruction, Mexican-American children experienced difficulty "catching

up" to those children who already knew the language of the school and

could gain, therefore, more easily from the instructional program.

Nedler and Sebera (1971, p. 259) wrote of the same language difficulties

and deficits found in the Mexican-American children with whom they

worked in the Southwest Educational Development Laboratory programs:

"Where the language of the home and the language of the school are

different, the problems are multiplied for the child."

The language factor, in combination with the cultural-environmental

differences of the Mexican-American migrant families, increased the

probability of these children falling behind in the formal educational

program of the public school. Reissman (1962) stressed that children

who come to school from a different culture will be at a disadvantage in

reaching the expectations of a school curriculum developed on middle-

class cultural values. This has been actualized many times in school

settings throughout the county and reported by educators developing

curricula to compensate for cultural differences in the mid to late

1960's Cazden, 1971; DiLorenzo & Salter, 1968: Kamii 1971; Weikart,

1967; 'h'i t e, 1970). Hardee County School personnel expressed particular

concern for the success of their Spanish speaking elementary school

children. Below average achievement at grade level for children from

the migrant communities, as reported by school administrators, gave

credence to their concern.

Teachers in that county were not insensitive to the problems facing

the migrant children in their classrooms. The language differences seemed

to be at the heart of the problem. Bruce Gaarder (1975, p. 4), in

reviewing the effectiveness of programs serving Spanish speaking children






5


under the Bilingual Education Act, made the following observations:

"Teachers were not prepared to work professionally through the medium

of a non-English tongue," and even those who spoke Spanish had been

"educated exclusively through English." Hardee County teachers faced

this same dilemma. There were no Spanish speaking faculty members;

teachers who might have learned Spanish at one time in their own

schooling were not able to speak it well enough for conversation with

the children let alone for instructional purposes. The use of parapro-

fessionals from the Spanish speaking migrant population was not yet con-

sidered a source of assistance in the classrooms to English speaking

teachers and Spanish speaking children.

The deficits in learning of these migrant children which created an

increase in their school failures were found in both cognitive and

affective skills which were required of them when they entered public

school. Children were unable to demonstrate cognitive potential in a

school system where standard English was the mode of communication.

Affective deficits appeared to result from feelings of nonacceptance

as a belonging member of the permanent community, which even young child-

ren learned quickly from the feedback their parents got from others

and passed on to their youngsters in many ways (Hoffman & Mottola, 1971).

In considering all of these factors, the limited academic success

for the Mexican-American child in Hardee County during these years was

undoubtedly the result of language differences in a school where no one

spoke Spanish to and with the child, of language deficits resulting from

an educationally impoverished family, of cultural differences not fully

understood and accepted by the school community as well as the larger

community, and complicated further by the very nature of the migrant

family. In the migrant family's mobility, children were often lost









to the full benefits of the educational program through natural attri-

tion caused by the families' movement in the migrant stream. Even though

migrant families in that county often returned for fall to spring field

labor, for school age children this meant continued late entrance into

and early exit from school. Mobility of families existed within the

county, too. Sometimes, the late entrance was into one of the three

elementary schools in the county and early exit was from a different

one of the schools. In Zolfo Springs, where the school population

included well over 50 percent migrant children moving in and out of the

school program during the year, and with the majority of these children

from Mexican-American homes, concern for the educational success of

these children was especially great.

The preschool intervention program, implemented in Hardee County,

was an attempt to prevent the expected school failure of the disadvantaged

migrant child and especially the Mexican-American migrant child in

that county. Whether or not the program made a lasting difference became

the basis for this present study.

Summary of the Preschool Intervention Program

General objectives for the Florida Migrant Early Childhood Learning

program for four and five-year old children had been developed (Combs,

1971; Curtis & Klock, 1970). Primarily, the objectives were guidelines

for insuring a health and nutritional emphasis in an educational environ-

ment conducive to expanding language and social-personal development

to meet the needs of migrant children. Encouragement was also to be

given to parents to participate in center activities and to contribute

information as to cultural differences. These general objectives served

as the only state guidelines for planning the Hardee County program for

three-year olds.









The program was designed under the pressure of limited time and had

to be operational within a few weeks in a fieldbase some distance from

the University without the direct supervision of the curriculum designer.

Because of this, the curriculum had to be without the complexities of

laboratory design. It had to be understood easily and accepted as

valid by those responsible for its implementation in Hardee County. It

had to be planned for monitoring curriculum modification and instructional

strategies through bi-weekly on-site consultant visits. With these

restrictions, the program was developed in an attempt to bring about an

increased opportunity for future school success for the children who

would participate.

Broadly stated, the major goals of the preschool intervention program

designed by this investigator were similar to those of many early child-

hood programs: to maximize the educational potential of the participating

children and to increase the probability of their future educational

success in public schools in Ilardee County. Four components made up

the program design: an educational curriculum component, a home-school

component, a preservice-inservice training component, and an evaluation

component.

The Curriculum Component Of The Preschool Intervention Program

The curriculum resource guide (Hoffman, Mottola, 1971, p. 3-4)

presented the following considerations for the curriculum design:

1. a language-oriented program to bridge the gan between home

and school oral language;

2. a task-oriented program to increase the child's ability to

solve problems at his/her particular state of intellectual

and motor-skill development;

3. an experience-oriented program to assure opportunities for









effective relating to school and the larger community, and

expanding knowledge beyond the immediate home environment;

4. a social-emotional development program conducive to growth

in self-esteem; and

5. A health and nutritional program to enhance physical well-

being necessary for the realization of a total life program.

Restatement of the curriculum design in terms of goals for

learning for the three-year old migrant child followed the statement

of curriculum components:

1. to communicate more effectively, both physically and verbally;

2. to learn how to learn: to explore, to question, to solve

problems;

3. to supplement, augment, and compliment known home experiences

with experiences in the school community;

4. to increase awareness of self as a giving and receptive

person, worthy of recognition and acceptance, and capable

of successful experiences in the school world; and

5. to increase understandings for care of one's physical self.

As the instructional objectives of the program were developed

and presented, giving more specific direction to the teachers for

implementing the curriculum, the nature of the preschool intervention

was stressed:

1. to use the family language (Spanish) whenever appropriate

in both conversation with the children and instruction in

the program;

2. to use English simultaneously as a second language to help

children become familiar with, understand, and use the

language of the public school;

3. to provide planned concrete learning activities, geared to









each child's cognitive and experiential level; and

4. to increase the use of English in instruction when children

evidenced attainment of cognitive skills and the ability

to talk about this learning in their own language (non-

standard English or Spanish).

Rationale for the Curriculum Component

A rationale for each of the curriculum components was also

presented by Hoffman & Mottola (1971, pp. 5-23). The rationale was

stated to be easily understood by both the professionals and para-

professionals who would work in the program. Because of the time

interval between the development of the preschool curriculum and

this present study and with the influences of increased professional

knowledge and understanding about early intervention into learning,

it seems wise to depend upon condensed but direct quotes from the

original curriculum guide for the underlying considerations for the

curriculum at the time this investigator designed it. Therefore,

the following sections on language, cognitive task, experience, and

self-concept components of the curriculum design are statements of the

program design rationale taken in part from that document.

LanguageComponent Rationale. Close interaction with
adults on a verbal plane is necessary for the young child
to learn his language. If this is absent, or if the verbal
plane on which the child is communicating is one of non-
standard English [or nonstandard Spanish], language devel-
opment reflects this lack. The quantity and quality of
speech of an individual are directly related to the environ-
ment in which the person lives. Language does indeed
"mirror" life around the child.

The fact that migrant children have continued failures
in public schools indicates serious problems in communica-
tion between English speaking teachers and n nnstandard
English speaking (or Spanish speaking) children. Children
feel inadequate when no one knows how much they have to
contribute in their own language or teaches English as a
second language to supplement their own home language.









It is important that the child's native language be
accepted as his and of importance because it is his. What
is being advocated is that the migrant child needs to have
teachers who speak and understand the child's language,
build upon its strengths, and then add a school language
which will help the child make ideas and knowledge known,
secure information, learn more about the world,and grow
in ability to communicate effectively in the school world.
Children need (1) to talk and be listened to with accep-
tance of teachers and peers; and (2) to listen in order to
understand perceptually that which is being said in Spanish
and in English. Effective communication between child
and child, and child and adult, is important.

To meet the first need suggests providing experiences
that bring natural responses from the child about his own
world and environment and help him say to the school,
"My world has meaning." This happens when teachers create
opportunities for children's talking and teacher's listening.
The second language need is a prerequisite for meaningful
oral communication. If the receptive language ability--
that which involves the ability to comprehend what is being
said--and the inner language-- that which involves the
ability to think to oneself in word-- are inadequate or
restricted, then the expressive language ability--that
which involves the ability to express one's own ideas in
words--creates difficulties for school success.

Bzoch and League (1969) showed, through the develop-
ment of their Receptive-Expressive Emergent Language
Scale for measurement of language learning in infancy,
that many children evidence a lag between the two languages.
The gap between the three language abilities (stated above)
often widens as the migrant child progresses through public
school. The one obvious reason is the lack of developing
the needed oral school language.

If the child is to talk, if he is to learn new ideas
to talk about, if he is to understand others and be under-
stood by them, he must encounter language opportunities.
Frost (1968, p. 383) stated: "For the teacher, the
verbal behavior...is concrete evidence of the starting point
for the child's education. From that point education is a
matter of change, often radical change, but the beginning
of the educational encounter is brought to the school by
the child, from the environment which has shaped and educated
him until the moment of formal education begins." Communi-
cation with the child is the teacher's responsibility as
is the reconciliation of home and school.

Cognitive Task Component Rationale. In a curriculum
for three-year old migrant children, it seems appropriate
to combine the strengths of the child development-oriented
program with a compensatory educational program which,
through increased intellectual stimulation, could overcome









early learning deficiencies in these children. This is to be
done through specific teacher-directed learning episodes
geared primarily for intellectual growth -- a series of tasks
which select activities for their cognitive content.

Piaget's (1969) theory of the development of intelligence
offers suggestions for building a cognitive curriculum for
young children. It is within this framework of insight
into stages of intellectual development that there may be
answers to developing a program based on objectives necessary
as prerequisites for subsequent learning. For Piaget (Lavatelli,
1970), the child from about two to four years of age must
relearn on a conceptual level what he had learned in the
sensorimotor stage, a preverbal stage where reasoning is
accomplished by means of mental images rather than language.

The child appears to be constantly investigating his
environment and exploring the possibilities for action within
it. He must have the opportunity to manipulate things, to
ask questions, to seek answers, to compare findings. He
must have a chance to do his own learning under teacher
guidance, not teacher domination. As the child discovers
new ways of doing things, new symbols to use in communication
with himself and with others, he grows intellectually.

Ideas gleaned from the research of Kamii (1967) have been
incorporated into the design of the tasks in this curriculum.
Her understanding and experience in the practical applica-
tion of performance tasks built on Piagetian concepts are
recognized as valuable models in an attempt to build a sequence
of cognitive tasks that have meaningful order for the develop-
mental learning steps of the three-year old child.

Ideas are also based on tasks designed for the Home
Learning Center Approach to Early Stimulation project at
the University of Florida. Gordon (1970, p. 120) stated:
"The principle is that [the] child learns best, and learns
not only how to learn but also that it is fun, when you
provide him with a variety of interesting and challenging
things to do. These things teach him that he can do, he
can affect his world, he is competent. With these basic
building blocks of skills and self-confidence, additional
experiences enable him to build his ability to learn and
continue his growth in skills."

It must be remembered that tasks are for learning and
not for testing. Teachers are to provide useful experiences
so that children can grow in their knowledge and abilities.
The task-oriented phase of this curriculum includes those
cognitive tasks ranging from sorting and classification, to
ordering and seriation, patterning, sequencing, and very
early number concepts. These tasks are built on positives,
and designed without the "pressure cooker" style of some
traditional educational performance tasks.









Experience Component Rationale. The child-initiated
activities through which much of the emotional and social
growth takes place share equal importance with a culturally-
enriching program plus teacher-directed learning episodes
primarily for cognitive growth. An experience-oriented
program is one in which the environment encourages action
participation: doing, being, playing, working.

This part of the curriculum reflects the child's play as
the way the child works and learns about the environment
around him. Hymes (1968, p. 98) supported play as a time
of earnestness and intensity: "Play for young children
is not recreation activity, nor recess activity. It is not
leisure-time activity nor escape activity. Free play is thinking
time for young children. It is language time. Problem-
solving time. It is organization-for-ideas time, when the
young child uses his mind and body and his social skills and
all his powers in response to the stimuli he has met."

Free play, which is initiated and developed by the child
in individual play, in parallel play or in cooperative play,
is lacking in structure other than that given it by a child's
interests and his imagination. Teacher-promoted play, in which
materials are arranged or activities are planned by the adult
to promote learning about the larger world, is built upon the
free play of the child and furthered by materials and equipment,
field trips and visitors, the teacher may add and plan for
stimuli.

Frank (1968, p. 437) wrote: "While play may be focused
upon playthings and situations and people, it soon becomes
concerned with ideas, concepts and assumptions by which the
child carries on his many thought experiments." Spodek (1965,
p. 157) emphasized the teacher's role: 'The fact that signi-
ficant learning can develop from play need not detract from
the child's satisfaction . . The adult's selection and
engineering depends upon the learning to be gained .
The content is to be added by the teacher who will serve as
the resource person.' Providing the opportunity for rich
play-work learning experiences is a must in an educational
program for all children, and especially for disadvantaged
children.

Self-concept Component Rationale. To help children to
be and to become is an important aspect of this curriculum.
How well each child feels about himself determines his success
in doing and being. The child who sees himself as less than
others too often becomes that child who fails time after
time until at last he gives up trying and becomes the "drop
out" talked about so much.

The tasks related to building the migrant child's self-
concept are aimed at demonstrating belief in the child's
integrity as an individual, his right to be accepted as he
is, his unique abilities, his pride in himself and his
family. Goals for the child are becoming aware of self, of









others, of one's feelings, of other's feelings, of one's
own language, of a second language, of one's own way of
life, of others' life styles and of meeting success regularly.
If the early years are of the importance we believe them to
be, then feeling good about self at an early age gives greater
assurance for continued feelings has a lessened probability.

Summary of the Curriculum Component

These preceding statements from the curriculum guide (Hoffman &

Mottola, 1971, pp. 5-23) were the considerations upon which the ration-

ale for the curriculum component of the preschool intervention program

was based. The emphasis on language growth and development, both in

Spanish and in English, was a major intervention plan. If children

could be encouraged to work and play and therefore learn in their

own home language as well as the school language, it was expected that

cognitive and affective skills would be evidenced and growth promoted.

If children knew that their own cultural background was valued and

used in the learning environment, it was expected that positive feelings

of self-concept would be enhanced and achievement would be greater.

If children had an opportunity at an early age to learn more about

the larger school world, entrance into public school would be less

traumatic and adjustment made easier. If children were physically

and nutritionally cared for, learning would be potentially easier, too.

These considerations and assumptions were the bases for the curriculum

of the preschool intervention program.

The Home-School Interaction Component of the Preschool Intervention Program

The home-school interaction concept came from the philosophy of

the Florida Parent Education Model (Gordon, 1970; 1976) which supported

the premise that education is enhanced when parent and child are

involved together in the learning process. Because of the long working

hours of both parents, junior and senior high school siblings of the

preschool children were employed to work with on-site teaching personnel









in planning "home learning tasks" with the brothers and sisters in

the preschool program. Home learning activities are designed

experiences growing out of a child's interests at home or at school

and based on physical, social, emotional, and intellectual needs

(Packer, Hoffman, Bozler, 5 Bear, 1976; Shea & Hoffman, 1977). The

adolescents spent two hours a week at the center in work-play situa-

tions with the children; a second two hours each week in preparing

tasks with the teachers; and an additional four hours a week in the

home settings working with the three-year olds in specific task

direction. It was expected that the older children would help parents

understand more about the preschool program, reinforce some of the

earnings emphasized in the program, promote more use of English by the

young children at home,and generally serve as liaisons between school

and home.

Along with the use of these older children in this component of

the program, bilingual aides from the migrant community worked in

the classroom with the teachers in carrying out the daily educational

program. Their active involvement with the children facilitated the

use of Spanish with the children and also promoted a dialogue between

the Mexican-American families and the school.

The Inservice Training Component for the Staff of the Intervention Program

The Institute for Development of Human Resources provided an on-

going consultant assistance for pre and inservice training of the teaching

staff as well as program implementation monitoring. Guidance from

the curriculum designer permitted modification and expansion of the

curriculum as growth in the children was evidenced. Since the teaching

staff were not trained and experienced early childhood teachers and

the program conceptualization new and untried in that county, it

was believed that continuing on-site contact with the program









designer and sponsor would increase the chance of successful implemen-

tation. An early childhood consultant served as a liaison between

the site and the Institute for Development of Human Resources.

The Evaluation Component of the Intervention Program

The assumption was made that the children would demonstrate

increased problem solving ability as well as language growth by the

end of the intervention program, due in large part to the acceptance

and use of their own home language, the deliberate introduction of

English into their school day learning,and a planned cognitive task

component in the curriculum. To determine the effects of the program

upon the participating children, pre and posttest data were collected

by the Institute for Development of Human Resources staff on the prog-

ram group as well as a comparison group, using the following assess-

ment measures: the Arthur Adaptation of the Leiter International

Performance Scale, described as a "nonverbal Binet for young children"

(Arthur, 1952, p. 1); the Preschool Inventory (Caldwell, 1967); the

Peabody Picture Vocabulary Test (Dunn, 1965), given in English and also

in Spanish as translated for use in this program. A series of check

lists, prepared by the investigator and completed by the teachers to

sample various aspects of cognitive and language development along

with growth in positive self-concept evidenced in each child's behavior

during the program's duration, provided additional assessment information

(Hoffman & Mottola, 1971).

Results of the Intervention Program Evaluation of Children's Gains

Evaluation of the intervention efforts by members of the Institute

for Development of Human Resources indicated significant growth in

a variety of areas for the children, even within the severe time

limits of the program. The detailed evaluation results are described









in Migrant Early Childhood Education Program in Hardee County, Florida:

An Evaluation (Newell, Cage, Hoffman,& Carr, 1971). Growth in the

area of general intellectual competence for the children was reflected

by a significant increase on the Leiter International Performance Scale.

Significant improvement on the English version of the Peabody was

obtained but no significant increase was found for the Spanish version.

The Preschool Inventory indicated some growth in the children, although,

because of the difficulties in administering this measure, no statistical

analysis was done. The check lists completed on each of the children

also reflected growth in the areas of language development, cognitive

development, physical development and self-concept.

The study further reported that deficiencies in language and

cognitive skill areas of these children can be and were, in part,

overcome and, that were the program to have continued until the

children reached public school age, even greater growth in these

areas might have been noted (Newell, etal., 1971, p. 392. Although no

plans were made at that time to assess longitudinal effects of the

program, it was anticipated that the children would gain more from

their elementary school experiences because of their participation in

the intervention program.

Program Continuation by Hardee County

Short-term gains, as measured by the tests used for achievement

data collection, were evident in the children's performance in the

intervention program. Whether there would be long-term effects

from the program was only a hoped-for expectation. However, the

program attracted attention in its possibilities for Mexican-American

children to enter public school with an educational "head start." The

Hardee County Board of Public Instruction decided to continue such a

program for migrant children. With the addition of a new kindergarten









program, some of the original program children had a opportunity for

continued preschool education.

There is evidence of a self-report nature from the teachers in

the program that the objectives of the pilot project curriculum

continued to be stressed during the following year. The teachers'

observation logs of children's behaviors indicated similar teaching

strategies for children's learning as those recommended in the initial

program. The bilingual aides remained as assistants in the program

and it can be expected that because of their presence Spanish was

used with children needing the familiar home language for communication

purposes. There was no formal assessment made of children's growth;

and direct influence from the initial investigator and sponsorship

by the Institute for Development of Human Resources terminated along

with the pilot program, June 1971.

Significance of the Present Study

The significance of this study lies in the potential for answering

questions related to the elementary school success of the Mexican-

American children who participated in the pilot preschool intervention

program at age three and who have been enrolled in elementary schools

in that county from kindergarten through third grade. The questions

raised are (1) What do the scores on standardized achievement

measures indicate for these children as a group when considered as

evidence of school performance success? (2) Can the evaluation

procedures used in this study indicate longitudinal effects of an inter-

vention program upon school performance of these children? In other

words, did the opportunity for a "head start" into kindergarten make

any difference for the children in measureable ways?









At the time of this present study, twenty of the twenty-six

Mexican-American children from the preschool intervention program

were enrolled in one of the three elementary schools in Hardee County.

It was because of the present enrollment of these children in this

school system, the on-going support given preschool education in the

county, the desire to have an objective evaluation made to indicate

the value of such a program, as well as the professional interest of

the investigator that the study of these children's school success was

proposed and questions asked as to the effects of their participation

in the intervention program upon the school achievement record.

School achievement is relative and school success is admittedly

a vague term. Most elementary schools use traditional standardized

achievement test scores to determine in part how well each of the

students does comparedto national, state, or local norms for those

assessment measures. Cumulative records are kept for each child to

plot this school progress. Also included in most school records are

evidencesof learning behaviors which indicate need for special

compensatory programs within the school system. For example, special

education, learning disabilities, speech correction, or social-

emotional problems are noted with recommendations for follow-up and

assignment to classes designed to meet these needs are indications of

problems children are having in maintaining success in school.

Attendance is also noted; the number of days absent during the enrolled

months is considered to have an effect on how well a child is expected

to keep up with the educational program. Some demographic data pertaining

to the home environment is recorded with the expectancy that educational

background of the parents, occupation employment, number of children

in the family, and mobility of the family give further indications of






19

predictive school achievement. These factors are all considered by

most teachers as contributing to the school performance of children.

If the intervention program children show significant gains in

school achievement test scores and these gains can be attributed to

longitudinal effects of the preschool program, there may be implications

for program development and implementation in the regular school

program in Hardee County or in other counties with similar school

populations including many Mexican-American migrant children. With

the continuation of migrant education funding of early childhood

education programs for three, four, and five-year olds, there may also

be implications suggested by the results of this study for curriculum

determination and emphasis in such programs. Perhaps the real signi-

ficance of this study lay in the opportunity to examine the school

achievement for these particular children under real life conditions

as opposed to tightly controlled, experimental conditions.

State of the Research Question and Hypotheses

The research question addressed in this study is: Did the preschool

intervention program contribute to the school achievement of the Mexican-

American children who participated in that program as determined by

Hardee County public school criteria for primary grade success? If the

children under study were affected positively by their attendance in the

preschool program, then they would evidence greater achievement during

the primary grades than a comparison group who did not attend the

intervention program.

It was anticipated that the program group would have higher scores

on predictive measurements for school readiness and for reading readi-

ness ad kindergarten and first grade than the comparison group who did

not attend the preschool program. It was expected that the program

group would evidence higher achievement scores on standardized test






20

batteries, especially those tests measuring language achievement,

than the comparison group. It was further expected that the number

of children retained at grade level by the end of third grade would

be less for the program group than the comparison group. It was

also anticipated that school attendance would be as regular for one

group as the other, even though all of the children might not have

enrolled at the beginning of the fall term and some might have left

before school was officially over.

The following research hypotheses were formulated:

1. Group means on a standardized test for school readiness

at the beginning of kindergarten will indicate significant

difference in favor of the program intervention group.

2. Group means on a standardized test for reading readiness at

the beginning of first grade will indicate significant

difference in favor of the program intervention group.

3. Group means on a standardized achievement test at second

grade will indicate significant difference in favor of the

program intervention group.

4. Group means on a standardized achievement test at third

grade will indicate significant difference in favor of the

program intervention group.

5. The number of children retained at grade level by the end

of third grade will be less for the program intervention group

than for the comparison group.

6. Attendance records will indicate similar absenteeism for each

group due to the migratory nature of all of the families.










Limitations of the Proposed Study

Whatever limitations are ever present in quasi-experimental post

hoc studies are certainly potentially present in this study. The

difficulties of evaluating longitudinal effects of early intervention

programs on children's elementary school achievement are reported

with consistency in the literature (Abt Associates, 1974, 1975, 1976;

Campbell, & Erlebacher, 1970; Cicirelli, 1969; Datta, 1975; DiLorenzo,

Salter, & Brady, 1969; Payne, Mercer, Payne, Davison, 1973; Rivlin, &

Timpane, 1975; Stodolsky, 1971; White, 1970).

Since a longitudinal study was not proposed and planned a priori

the factor of experimental-control group design was weakened. Random

assignment to the preschool intervention program was not made. Children

were enrolled from self-selection procedures of the parents, with the

opportunity for participation made available through notices sent

home to parents of school-age children and through home solicitation

by the bilingual aides employed to work in the program. Only two of

the original control group were still in the county; therefore, the

comparison group was selected from those Mexican-American migrant

children who entered kindergarten at the same time as the program

group and were still in school through third grade.

The assessment measures used in the evaluation of children's

school performance were those achievement tests used by the school

system to assess school achievement and not selected by the investigator

to measure defined criterion objectives from the original preschool

intervention program.

The nature of the migrant family in its mobility was reflected

in the assessment data available for use in this study: tests varied,

depending upon the child's absence from school at the time a regularly









scheduled test was given; some test scores are missing completely

because of late entrance and early exit of the migrant children; some

tests differed because of children changing schools within the district

and therefore could not be used.

There was no controlling for classroom-teacher assignment. The

variables found in teacher-student interactions were recognized as

variables for which there were no controls. It was also considered

that Hardee County teachers might well have had expectations for

increased school performance from the program intervention children.

Their comments during the days spent by the investigator in data

collection indicated their regard for the value of the program. This

teacher-expectation may have had a self-fulfilling prophecy effect.

Within these considered limitations, it was hoped that by following

procedures deemed viable and recommended for this kind of evaluative

study, assessment could be made as to the longitudinal effects of the

preschool intervention program upon the school performance of the

participating children. Under the existing circumstances, it would

be unwise to anticipate generalization beyond thi p'-pulation in this

particular study.


















CHAPTER II

REVIEW OF THE LITERATURE



In the past decade, writing in the field of early childhood

education has been abundant. Authors have traced the history of early

education and child development as well as presented new understandings

in the cognitive and affective development of young children in

numerous books and articles. Reports of innovative programs have

appeared in many monographs. Curriculum guides, emphasizing different

approaches to instructional procedures, have been developed. Entire

journals have been published to provide updated information on research

underway or to promote dialogue between differing philosophical

positions regarding early intervention into learning. In the last few

years researchers and educators have brought together the literature

on early intervention programs.

For the purpose of this study, it was important to review rele-

vant literature available at the time of the investigator's designing

of the preschool intervention program. What had been learned in the

following seven years might well have influenced differently a

design of an intervention program today. The writer chose to review

the literature from a chronological approach, looking at the advent of

the federal government into education for the disadvantaged child,

at the psychological underpinnings for the initial early intervention

programs of the 1960's, at the philosophical bases underlying the









rationale for model intervention programs from which the Hardee

County preschool program design drew heavily, and at the programs

designed especially to meet the needs of non-English speaking preschool

children in 1970. Finally, a review of the literature related to

the longitudinal effects of early intervention into learning as well

as the continuing problems inherent in evaluating early intervention

programs from a longitudinal consideration was made. This chapter

is addressed to these areas.

Federal Support for Education of the Disadvantaged Child

Programs for systematic intervention into early learning of

young children came into national prominence in the 1960's with the

advent of President Johnson's War on Poverty. Under the guidelines

of the Office of Economic Opportunity (OEO) in 1964, the children

for whom these compensatory preschool programs were to be carried

out were determined in the first Head Start projects. Children

whose families were economically, educationally, and socially disad-

vantaged were to be the target population. Within the first year of

Head Start, over a half-million children had been in attendance

(Evans, 1975).

Having defined the population, OEO called for a program emphasizing

medical, dental, and nutritional attention for the children, social

services for the families, active parent involvement,and an educational

curriculum that would provide disadvantaged children a "head start"

into kindergarten or first grade. There was no precedence for such

a large scale intervention program; nor had early childhood research

efforts clarified any single approach to instruction as the best

one. The traditional nursery schools and existing public school

kindergartens, designed primarily to enrich and enhance social-emotional






25

experiences for middle class children, had to serve as models for

the instructional program of initial Head Start efforts (Hunt, 1972)

under the pressure of too little planning time, lack of trained

staff and material resources, and yet thousands of children in hundreds

of varied settings. Existing health, welfare,and social service

agencies were the sources for the other mandated services.

It is important to note that parent involvement was encouraged

and promoted in two ways: through volunteer participation or in

paid aide positions in Head Start classrooms, and as members of Commu-

nity Action Program agency (CAP) governing boards (Economic Opportunity

Act, 1964). These CAP boards were responsible for decision making

at the local level concerning the approval of plans for Head Start

programs as well as other programs to be funded to overcome problems

in the poverty community.

Poverty parents found it financially difficult, if not impossible,

to serve in volunteer capacities. Many parents, even though they had

experienced limited success in schools themselves, were employed

as classroom aides. This was a new experience for both the parents

of this group and the professionals in schools. In some communities,

where CAP agencies gave the responsibility of Head Start programs to

groups other than the public schools, members of the poverty community

to be served were the lead teachers of the Head Start children.

Local communities, especially their governing agencies, were not

accustomed to making educational decisions with this new group of

citizens having an active and shared voice; nor were the spokesmen

for the poverty community experienced in the decision making process

of which they now found themselves such an important part. However,

the traditional policy-makers of a community and the almost politically-

disinfranchized citizenry of the poverty community joined together and






26

learned to work effectively to begin to meet the expressed needs of

Head Start children and their families. The impact of such a War on

Poverty and the intervention program, Head Start, was to be felt for

many years.

The second major thrust taken by the federal government to assist

financially in the education of disadvantaged children occurred with

the passage of the historic Elementary and Secondary Education Act of

1965. This legislation gave state education agencies the responsibility

for developing a comprehensive educational program that would provide

opportunities for local school systems to finance and implement

programs to insure increased educational success for the educationally

deprived child (Rivera, 1970). It had been assumed that migrant

children would be among those who would most benefit from the legislation.

However, as stated in the introduction to this study, migrants were

not included in the 1960 census and, therefore, were not counted in

estimating the monetary allocations needed for the eligible states.

In November of the following year, 1966, Title I of the Elementary and

Secondary Education Act was amended to include the children of migrant

laborers. Migrant children were defined by the Office of Education as

children of migrant agricultural workers who, with in the past year,

have moved across county or state boundaries for the purpose of obtaining

agricultural work or related food processing activities. An amendment

was made to that definition (Public Law 90-247) to include those

migrant children whose parents have established a permanent home

within the past five years (Rivera, 1970, p. 43).

Much had been written by late 1970 on the social, economic,

health,and political plight of the migrant adult and his family.

However, Cheyney (1972, p. ix), in compiling literature for a book

on the education of migrant children, expressed concern with the









"dearth of information available that would help classroom teachers

improve their understanding of migrant children and deal with the problems

peculiar to them as individuals." Elizabeth Sutton's early migrant re-

search and experiences, and subsequent chapter in Cheyney (1972, pp. 110-

111) pointed out the need for modification of curriculum content to meet

the migrant school-age child's particular needs, but stressed there was

no stereotype of the migrant child, that the theory of individual differ-

ences among children held true for migrant children as well as all child-

ren, and "finally, the basic principles of good teaching are the same

of the migrant child as they are for all children."

With the enactment of the amendment to the Elementary and Secondary

Act to include children of migrant laborers, thirty states had initiated

migrant programs and fourteen had expanded programs already in existence

in 1967. These programs varied from the development of instructional

units that emphasized oral language development of school-age children

to summer programs for migrant families traveling in the migrant stream.

Some states developed state-wide surveys to locate migrants eligible

for specialized programs. In some of zhe southwestern states, public

schools were considering the problems of late entrance into and early

exits from schools of the many migrant children in the school population.

It was under funding from the passage of this federal legislation that

Florida initiated a pilot preschool program for four-year old children

during the 1969-1970 school year and made possible the funding for the

pilot three-year old program in Hardee County. The 1969-1970 evaluation

of Florida Migrant child compensatory programs summarized these programs

developed and implemented in the state under the Elementary and Secondary

Education Act of 1965 (State Department of Education, 1970).









The literature most relevant to this study was found in research

undertaken with young children in limited bilingual education programs

and in other early childhood education programs for disadvantaged

children. Since these programs had psychological underpinnings in

agreement with what was known about early intervention programs in the

late 1960's, it was important to review the rationale underlying consi-

derations for most early childhood intervention programs at the time

of federal legislation for education of the disadvantaged. The view-

points discussed in this part of the literature review were very

influential in the professional thinking of this investigator at the

time for curriculum decision-making in the preschool intervention

program design.

Psychological Assumptions for Intervention Program Models

Certainly the psychological underpinnings in the conceptualization

of Head Start advantages for young children were strongly affected by

the research and writings of child development psychologists whose

position reflected belief in the value and necessity of enriched

early educational experiences for positive, and even accelerated,

cognitive development of many children. Havighurst (1948) had

refined the concept of developmental tasks, promoted by Daniel Prescott

at the University of Chicago, to include most of learning from

infancy through later maturity. In stressing the importance of

"teachable moments when the individual is ripe to learn," Havinghurst

(1948, pp. 27-28) stated that "formal education may be defined as a

procedure setup bya society to help children achieve certain of their

developmental tasks." The educational implications of this statement

suggested that "teachable moments" for certain developmental tasks

to be learned by young children could be recognized and acted upon









more systematically by teachers than by leaving such teaching to

chance by parents in educationally disadvantaged homes.

Bruner's views (1960) that selective and careful training would

accelerate learning, and that there was nothing that could not be

taught any child if taught at the child's stage of development were

important in early childhood education considerations at that time.

His interpretations of the structure of knowledge, along with the

relationships made of early learning to later learning, were undoubtedly

among the philosophical and psychological viewpoints that brought atten-

tion to young children's educational needs during this time of intervention

consideration (Evans, 1975, p. 1).

The writings of Benjamin Bloom (1964) gave further indications

of the importance of the early years for intellectual development as

well as the significance of the home and cultural environment to this

development. Children from low socio-economic status and minority

groups were showing consistently lower scores on measures of intelli-

gence than children from high socio-economic groups. Stodolsky

and Lesser (1967) pointed out that as children increased in age, there

tended to be even larger mean differences in intelligence between

the two groups. If Bloom's arguments that the greatest amount of

intellectual growth occurs during the first five or six years of life

and that the child's environment during that period of growth is of

utmost importance, then the planned and systematic learning opportunities

through educational intervention programs for disadvantaged children

were due important consideration.

The publication of Hunt's (1961) work on the relationship of

intelligence to experience contributed new and profound insights into

the question of fixed intelligence held quite firmly until then. His






30

position that intelligence was neither fixed nor predetermined but

was closely related to the early learning experiences of the child

suggested clearly that environmental changes for learning might

greatly enhance the cognitive abilities of disadvantaged children.

William Fowler's research and writings (1962a; 1962b) presented

justification for those who believed that learning in the early years

had not been maximized for children; that children could be taught

much more than they had been at home or in preschool programs; that,

under the existing nursery school programs for young children, there

was only a marking of time in learning. He proposed a systematic

approach to devising sequential learning tasks that would increase in

complexity and stimulate learning potential. Fowler's position was

somewhat reinforced by Irving Sigel (1965, p. 3) in his contention

that early education "can make a greater contribution to the child's

cognitive development than is usually realized." However, Sigel

cautioned against de-emphasizing the positive emotional climate of

most middle class nursery school programs, and encouraged preschool

education to address itself to the disadvantaged child who could

benefit most from intellectual stimulation in such a program because

of its virtually nonexistence in most poverty homes.

It was also in the 1960's that Jean Piaget, labeled "giant in

the nursery" by David Elkind (1972) in a New York Times article, was

"discovered" by American educators and psychologists. Translations

and interpretations of his complex writings were more available and

accessible for study. Flavell's (1963) translation of Piagetian

psychology provided a through and scholarly work for those wanting

to understand Piaget's theory of developmental stages in the cognitive

growth of children and to make application of this understanding in their

own research. The conclusions reached by Hunt and Bloomsuggeted their










studied analysis and consideration of Piaget's work. It was also evident

in later research designs for early childhood programs that the theories

of this eminent psychologist (Piaget, 1952; 1973; Piaget & Inhelder, 1969)

had tremendous effect in the thinking of the authors of many intervention

programs rFlkind, & Flavell, 1969; 'accoby, 5 Tellner, 1970).

The rationale for early intervention programs was summarized in

Evans (1975, p. 6), and was attributed by that author to assumptions found

in Blank (1970) and Sigel (1970). This rationale is quoted and paraphrased

here because it represents the thinking of many concerned with intervention

programs at that time: "Children are, by nature, malleable and their

growth and development can be modified extensively in a variety of

directions. The earlier one can effect a plausible intervention, the

better"; the provision of qualitatively sound experience can compensate

for basic lacks in children's environments. "Since the school's scholas-

tic emphasis demands certain basic learning capabilities, these capabili-

ties must become the focus for early intervention." Cumulative development

is involved, whether the influence of early experience is salutary

or hindering. Resources must be marshalled to prevent or remediate such

disorders in cognitive performance that children fail to achieve their

human potential.

Philosophical Bases for Intervention Models

It was important to stress once more that research efforts had not

singled out any one best approach to instructional programming when

Head Start was initiated. As communities struggled with the realities

of a new educational program underway, educational psychologists

and early childhood specialists were carefully designing and beginning

to implement research-oriented intervention programs which had far-

reaching effects on the future of Head Start and Follow Through, the










planned variation approach for each of these programs, and the continued

early childhood thrust [Revlin & Timpane (Eds.) 1975]. Both politically

and educationally speaking, it had been hoped that Head Start as

originally carried out in the early years of the program would provide

fast and long lasting results for poverty children's success in school.

However, there were many public school and nursery school approaches

being used with little attention given to what specific curriculum would

be most effective. Goals and objectives were ill-defined and usually

not stated in measurable ways. The Westinghouse Report (Cicirelli,

et al., 1969], to be discussed more fully later in this chapter,

emphasized in its recommendations the need for measurable results.

Head Start Planned Variation and Follow Through were responses to

this criticism.

Philosophical differences were evident among the educators who

were designing and implementing intervention programs (Butler, 1970;

Evans, 1975; Maccoby & Zellner, 1970; Weber, 1970). Programs varied

from highly structured language-math orientation (Bereiter 6 Engelmann,

1966) to traditionally formal instructional programs (Karnes, 1969;

Klause & Gray, 1968); from behavior analysis and modification emphasis

(Bushnell, Worbel,& Michaelis, 1968) to cognitively-oriented Piagetian

programs (Weikart, Rodgers, Adcock,& McClelland, 1971); from responsive

care approaches (Nimnicht, McAfee,& Meier, 1969) to planned enrichment

programs such as that of Bank Street (Biber, Wickens, Shapiro,&

Gilkeson, 1971), the parent education and involvement programs of Gordon

(1969, 1970a, 1970b), Levenstein (1970), and Schaefer (1970, 1972) to

the programs building upon the strengths of the child's family language

(Hughes. Wcetzel, 5 Henderson, 1969; Nedler, 1966, 1967,1971). From

some of these research efforts, and others underway at various










colleges, universities or early childhood laboratories, came the

models selected for Planned Variation Head Start programs, reflecting

the philosophical differences in assumptions underlying the program

designs.

Certain of these programs had philosophical bases that were, in

part, the considerations for this investigator's preschool interven-

tion program design. It seems important to focus on these while

recognizing that the philosophical orientation of the other programs

were just as valid and viable.

David Weikart and his associates (1971) had spent several years

working with preschool children from disadvantaged home environments

in the early Perry Preschool Project. From their understanding of

the value of Piaget's contribution to knowledge about the developmental

stages of young children's cognitive growth came the Cognitively Oriented

Curriculum also known as the High/Scope Planned Variation Head Start

(and later, Follow Through) Model. Psychologically rooted in Piagetian

theory, the concepts of the program were philosophically based in the

need for children to be active participants in their own learning and

that a variety of experiences must be made available for learning to

become all that is possible for the young child (Weikart, et al., 1971;

Weikart, 1972). Classification, seriation, spatial relations, and

temporal relations were cognitive understandings presented so that

the child experienced these concepts on the motoric level, then the

verbal level, but always in physical manipulation of the environment

by the child (Kanii, 1972; Weikart, et al., 1971).

The Responsive Educational Program, directed by Ninnicht

and growing out of the philosophy of the New Nursery School (Nimnicht,

McAfee, & Meier, 1967, 1969),directed its attention to an autotelic










responsive environment: activities were done for their own sake; choices

by the child as learner were promoted and accepted; and adults

responded to children's choices with few teacher-initiated activities

insisted upon for child participation. As the program has developed

in Follow Through, children continue to work at their own pace with

materials and equipment designed to promote self-confidence in

problem-solving skills, sensory discrimination, and expressing ideas

through language and art media. It would seem that the philosophy

suggests the potential for increased learning when children select

their own learning experiences and are assisted by adults responding

to children's self-initiated activities.

Bank Street had for some fifty years developed an outstanding

child-oriented program for preschool children in the belief that the

interaction of the child with the social and physical environment of

learning was of equal importance with the recognition that the child

passes through developmental stages of cognitive and affective growth

(Shapiro, & Biber, 1972). Educational goals for children in this program

were the same for those who came from either advantaged or disadvantaged

home environments. Biber, Shapiro, Wickens, and Gilkeson (1971, p. 7)

stated that: "A program of education is derived from a system of

values -- an image of man and man's relation to man ... .Bank

Street College programs are based on a consistent philosophy of

education comprising values, goals, and strategies congruent with a

humanistic approach. Central to this philosophy are concepts of

competence, interperson relatedness, individuality and creativity."

This approach for Planned Variation Head Start stressed children

building positive feelings about themselves with teachers guiding the

development of plans, choices, and activities with language conducive









to the formation of ideas and expression of feelings (Rivlin, & Timpane,

1975).

Ira Gordon's (1969; 1970) early research in the Parent Educator

Program evidence a philosophy in support of the competence of parents in

the education of their children. His premise that education is enhanced

when parent and child are involved together in the learning process was

the basis for the use of trained parent educators to visit in the homes

of participating project children to work with mother-baby and mother-

child activities that would promote both cognitive and affective growth

in the children. The parent educators were women from the same poverty

community as the families whose homes they visited. The Florida Parent

Education Planned Variation Head Start and Follow Through Models stressed

this same belief that as parents increased in their competency and feelings

of self-esteem children would evidence enhanced intellectual growth and

self-worth. This model was unique in the planned variation efforts in

that no curriculum changes were set forth for classroom programs; the

school curriculum was accepted with the addition of the parent educator

as classroom-home liaison and parent involvement in the overall program

planning through the Policy Advisory Committee (Gordon, 1970; 1972a; 1972b).

From these varying philosophical emphases, the design of the

Hardee County preschool intervention program took shape. The "inter-

person relatedness" of Bank Street's program, promoting concepts of

competence, individuality, and feelings of worth about self, supported

the rationale for the social-emotional development component of this

investigator's preschool curriculum. The New Nursery School philosophy

of an autotelic responsive environment in which choices are made by

the child as learner to the experience-oriented component of the preschool

curriculum. The task-oriented component drew quite directly from the










Cognitively Oriented Curriculum of Weikart and his associates--a

Piagetian approach to learning based in the belief that children must

be active participants in their own learning. The use of adolescents

as home-school liaisons and the employment of paraprofessionals from

the Mexican-American community as assisting teachers in the classrooms

came from the Florida Parent Education Model of Head Start Planned

Variation and Follow Through, designed by Ira Gordon.

The language-oriented component of the preschool intervention

program was an attempt to emphasize the worth and need for the child's

own native language to be accepted and used in the early childhood

classroom setting. Support for this approach was found in some of the

bilingual early childhood programs for preschool children in the 1960's.

A review of literature at that time follows.

Bilingual Early Childhood Intervention Programs of the 1960's

Even as researchers turned their attention to various strategies

of intervention in the early education of young disadvantaged children,

few programs focused directly on the problems of children who spoke

little if any English and how best to facilitate their entrance into

the expectations of English speaking schools. It is true that the New

Nursery School (Nimnicht, Meier, & McAfee, 1967) initial program was

primarily for Spanish surnamed children who, in addition to environmental

deprivation, had a different culture and often a different language.

The Tuscon Early Education Model (Hughes, Wetzel, & Henderson, 1969)

also emphasized starting with the child's family language; however,

the program was not specifically designed for non-English speaking children.

School districts, primarily in the southwestern United States,

were beginning to establish elementary school programs for bilingual






37

education. Texas and New Mexico led in this effort at that time. One

exemplary public school program, initiated in 1963, was found in

Coral Way Elementary School, Dade County, Miami, Florida, where a first

"two-way bilingual" education program was conceptualized (Logan, 1970;

Inclan, 1972). John and Horner (1971, p. 28) reported that the student

body was mainly middle-income, half English speaking and half Spanish

speaking Cuban refugees. They further reported that the program goal was

to produce "literate, educated bilinguals" in both groups of children.

With fairly limited bilingual education in public schools, preschool

bilingual programs were even fewer. Those for which program descriptions

were available (John and Horner, 1971) and planned for Spanish speaking

children between the ages of three and five were the Community Play

Center Preschool in Redwood City, California, Escuela Hispana Montessori,

in New York City (one of the first of the preschools to include bilingual

emphasis in its program), St. Paul's Episcopal School in Brownsville, Texas,

and Dos Mundos School in Corpus Christi, Texas.

Probably the most well-known research program for bilingual-bicultu-

ral children in 1970,developed in response to cultural differences rather

than emphasis on cultural deprivation, was the Southwest Educational

Development Laboratory's planned Bilingual Early Childhood Education

Program (Nedler, 1966; 1967; 1971; 1972; 1975). In program development,

the primary objective was to develop new methods for teaching English as

a second language to Spanish-speaking children between the ages of three

and six. Dr. Nedler (1972, p. 69) stated: "As new research pertaining

to developmental theory has appeared, and as data related to the instruc-

tional program has accumulated, the conceptual design has been revised."

These revisions were reflected in the initial program as contrasted with

the program four or five years later.







38

Quoting from the proposal for the initial 1965 program at the

Good Samaritan Center in San Antonio, Texas, Nedler (1975) wrote that

the program would emphasize increasing the children's understanding and

use of both Spanish and English and provide experiences to enrich the

children's lives in preparing them for adaption to the American culture.

In an earlier article (1966, p. 3) Nedler stated: "We accept the idea

that a child's language is a part of him, and as we accept the child,

we accept, recognize and give status to his native tongue." With this

program emphasis and a philosophy of acceptance of Spanish at a time

when Texas had just reversed state laws forbidding Spanish in schools,

the Good Samaritan Center Program began.

A rather traditional activity-centered approach with the adults

speaking either Spanish or English to the children produced a "naturalis-

tic" learning environment. Observations of the children's behavior

indicated that they felt comfortable and secure in the center, appeared

to feel good about themselves, listened while being addressed in

English, but conversed mostly in Spanish. They really only used English

in a "practicing" way. A more structured approach to teaching English

was then selected (Nedler, 1971; 1972). Strategies were changed to

present concepts first in Spanish, and when the concepts were mastered,

they were systematically introduced in English. This more structured

approach was adopted as a "developmentally appropriate" and comprehensive

learning system for young Spanish speaking children (Evans, 1975, p. 173;

Nedler, 1975).

Nedler (1972, pp. 39-90) attributed the conceptualization of the

Southwest Educational Development Laboratory's early childhood program

for bilingual children to research and writings of other early childhood

educators and psychologists such as Bruner, Cazden. Hunt, Bernstein,

Bloom, Blank,and Ausubel. Some of the references made especially for









the approach used in language development came from the literature on

bilingual education. The Lambert and Tucker (1972) report on the

St. Lambert experiment, where English-Canadian children were taught

French and English alternately and then simultaneously, demonstrated

that children learned and developed thinking skills in a second

language with no adverse effects from that particular learning process.

Hake's (1965) study on psychological aspects of bilingualism pointed

to three specific problems related to second language learning: (1) dif-

ferences in word order must be learned; interference and negative transfer

inevitably results for the bilingual child; (2) sounds that do not

appear in the person's native language causes discrimination difficulties

because of the differences signaled in meaning in the new language;

(3) the selection of the appropriate response in the second language

causes interference with the new learning because of the previously

learned response in the native language.

H. T. Manuel (1965) wrote that Mexican-American children, and

especially the disadvantaged Spanish speaking children, face all of the

problems of the disadvantaged child of any race but compounded by the

lack of even good Spanish. When they enter school, growth in their

own native language is arrested, if not stopped completely, because

English is the language for all school communication purposes; and

their lack of out-of-school contact with English slows down the

development of skill in the second language.

Language development training has been an important part of all

preschool intervention programs. The questions raised seem to rest

with what, when,and how best to help disadvantaged children attain

necessary language skills for future academic success in school.









Differences in language among social classes were stressed in the

research and writings of Bernstein (1961). Labov (1970) suggested

there was an equality of functional linguistic ability in children with

nonstandard language. Vygotsky (1962) seemed to be in agreement with

Piaget in that the development of intellectual structures are built

through experience with language following the "knowingness." Sinclair-

De-Zwart (1969, p. 316) presented a Piagetian conception that

"the sourcesof intellectual operations are not found in language, but

in the preverbal, sensorimotor period where a system of schemes is

elaborated that prefigures certain aspects of the structures of classes

and relations, and elementary forms of conservation and operative reversi-

bility." Lavatelli (1970, p. 65), in writing about Piaget's language

theory, referred to Sinclair's work and her interpretation of Piaget

in making this conclusion: "Language is not the source of logic- but,

rather, is structured by logic."

These studies seemed to indicate an agreement that a young

child must experience the knowing of objects, experiences,and ideas

before language is added by the child to express this knowingness.

Varied strategies for language training have been found in research from

the Blank and Solomon (1968) and Cazden (1966) tutorial and "modeling"

treatment to Berieter-Englemann's (1966) structured language drill.

Programmed materials have been and still are used in many early childhood

centers. Dunn, Horton,and Smith's (1965) Peabody Language Development

Kit is one example of a comprehensive program for language stimulation

and development. Gotkin's (1968) Matrix Games, the Language Master

used in the New Nursery School, and the Talking Typewriter, The Talking

Page and the Voice Mirror from the Responsive Environments Corporation

are other examples of approaches to early language development for

teachers' use with young children.










Certainly, children must become competent in the language of the

formal school. In the light of the Coleman Report (1966) that school

achievement is enhanced when children have a sense of control over their

environment and feel good about themselves, any approach to language

training that reflects understanding of the language and culture of the

child may well contribute to the child's feelings of self-esteem and

personal worth. For the non-English speaking child, "bilingual education

may be the means of eliminating or at least minimizing the confusion

and shame created in the non-English speaking child by the common

educational practices of the day" (John and Horner, 1970, p. 149).

The language component of the preschool intervention program

designed for the Hardee County project was based on the early bilingual

preschool studies of Nedler,and Piaget's concept that young children

need cognitive and affective experiences paralleling the growth of

language.

The Literature on Longitudinal Evaluation of Intervention Programs

At the time of the preschool intervention program described in

Chapter I, the effects of early intervention into learning had not yet

been substantiated. There had not been time to evaluate longitudinal

effects of programs; there were not enough children in the middle

elementary school grades who had experienced well-implemented programs

prior tc their entrance into public school systems. Those evaluations

that were available in 1970 and 1971 provided conflicting results.

The reasons are many: clearly defined program goals and objectives

were not always stated, data collection procedures were often inadequate,

measurements did not always measure precisely-stated objectives, equi-

valent comparison or control groups were often missing, teacher variables

were not usually considered, and research efforts for those programs







42

of the 1960's had not been adequately funded (Butler, 1970; Evans,

1975).

While many individual Head Start programs, as well as research

programs designed with evaluation procedures built into the program,

reported immediate and short term gains in children's scores on

intelligence and achievement tests, these gains on a "variety of measures

varied in statistical and educational significance from marginal to

quite substantial"(Datta, 1975, p. 81).

Some of the conflicting results reported on longitudinal effect

studies before Head Start Planned Variation was well under way are

summarized here. Fuller (1960) found that out of seventeen studies

dealing with the relationship of early childhood education to later

school achievement, fourteen indicated positive results; Van Der Reit

(1967) reported that children in preschool programs performed signifi-

cantly better than children who had not been in such programs by the

end of first grade. Pitts (1968) found that the length of preschool

attendance was related to affective development but not to academic

readiness. The review of Head Start research (Grotberg, 1969) was

contradictory concerning the long range impact of Head Start; children

did not seem to lose what they gained from the experience but leveled

off to a point which allowed other children to catch up.

Weikart (1969, p. 3), in reviewing evaluation studies through 1969,

wrote: "The basic conclusion is that the more structured or task-oriented

the program, the greater the gains in immediate intellectual competence,

and where follow-up data are available, academic achievement." Also in

that year DiLorenzo, Salter, and Brady (1969) reported to the New York

State Education Department that cognitively-oriented and more structured

programs produced, in general, language and intellectual gains in









disadvantaged children greater than child-oriented programs. Structured

programs, emphasizing language and cognitive development, showed more

lasting gains in studies conducted by Karnes, Teska, and Hodgins (1970),

Karnes (1968), Miller and Dyer (1970) and Bissell (1971). Stodolsky

(1971, p. 16) suggested that gains of children in structured programs

may be the result of a "well-defined treatment and (one) measuring the

effects of a something that is specificable and largely the same for the

participating children."

The most critical study evaluating the cognitive and affective

gains made during the primary grades of selected children who had

participated in either the first summer or full year Head Start program

was the Westinghouse Report (Cicirelli, et al., 1969). The children were

in the first, second, and third grades when testing was done in the

fall of 1968. Standardized achievement tests plus personal-social

development and school motivation measures were used. No differences

on any measure at any grade between summer participants and nonparticipants

were found; and although differences were found between Head Start

children who attended a full year's program and nonparticipating children

at the first grade level, no effects were found after the first grade.

One very positive outcome of the Westinghouse Study showed up in the

parents' tremendous approval of Head Start programs for their children.

Obviously, they believed that both children and parents had benefitted

greatly from all and whatever Head Start meant to them.

The critical evaluation of early Head Start efforts coupled with the

1966 Coleman Report (Smith, 1975, p. 156), which had concluded that "with

the exception of the social class composition of the student body, varia-

tions in educational inputs bore little relation to variations in outcomes

once the effect of family and background influences had been removed,"

encouraged skepticism regarding the worth or many educational programs,









including intervention programs. But there was another side of the

evaluation results.

A review of the Westinghouse Study by Kean (1970, p. 449) suggested

that the "statistics used in the analysis overpowered the data available."

The major findings of the study were based on data collected from six

instruments plus survey and interview data on nearly 2,000 children

from 104 Head Start Centers across the country out of a population of

almost 13,000 centers. Kean's (1970, p. 449) reactions were expressed

as follows: "What was the value of asking how well children did in

third grade when Head Start Centers in some cases were working with child-

ren who had never before held a crayonor a fork? How, indeed, could

anyone have the audacity to suggest that he was really going to measure

cognitive and affective development at the third-grade level?"

Kean (1970, pp. 450-451) gave scholarly credit to a well-conceived

and well-executed study within the restrictions of an ex post facto

design. However, he raised the following questions: (1) Was the study

designed to answer political questions rather than educational questions?

(2) Was there a hidden curriculum in many centers that resulted in

objectives not measured by the report? (3) Were the "right" children

included in Head Start programs? (4) What was the relationship of the

nutritional component of Head Start to the outcome of intellectual and

emotional development measured by the statistical analysis? (5) Could

the interaction of Head Start with ESEA Title One Programs in some of

the same areas have been responsible for someof the washout effects

found in third grade children? Kean (1970, p. 151) further suggested

that the most important questions to be asked might well be "how might

the schools be different or whether the Head Start program should be

considered successful and the school unsuccessful."







45

In an analysis of the Cicirelli et al., report. Smith and Bissell

(1970) presented a comprehensive history of Head Start and of the

national evaluation. They, too, raised serious questions about the sampling

procedures used in the study and presented a reanalysis of the data which

suggested that some of the full-year Head Start centers were effective.

In these authors' judgements, stratified random sampling should have

been used instead of simple random sampling. The varying characteristics

of centers which Smith and Bissell believed related to their teaching

effectiveness would have been considered through stratified random sampling

procedures. They also expressed concern that more than half the centers

refused to participate in the evaluation or were unqualified for partici-

pation. Equally important to the selection of the centers for evaluation

was the selection of individual children. Smith and Bissell (1970, p. 74)

point out the Westinghouse Report shows "large discrepencies between the

Head Start and control intercorrelations" among socio-economic variables

and achievement measures.

Their conclusions (Smith and Bissell, 1970, p. 97), based on statis-

tical reanalysis of the Westinghouse Study data for first grade results

of children who had been in Head Start for a full year, stated: "The

intercorrelations among socio-economic and school readiness variables were

very different for Head Start and control samples . (and) the Head

Start group scored higher than the control group on the Metropolitan

Readiness Test by a large enough margin for us to consider the differences

'educationally significant.'" Their final argument was for continued

intervention into the elementary school years which, in effect, supported

the recommendations of the Westinghouse Report for planned program

variations and follow-up programs,with clearly defined goals and objec-

tives that could be measured and evaluated, and characteristics of

programs that could be replicated by other centers and schools.








Evaluation of longitudinal effects of early intervention programs

on children's achievement in elementary school continued to present

difficulties to evaluators of Head Start and Follow Through Planned Variation

programs, as reported by Elmore, McDaniels, Datta, Smith,and Cohen in Rivlin

and Timpane (1975). The Brookings Panel on Social Experimentation sponsored

a conference of experts in the field of early education effectiveness

for disadvantaged children in the spring of 1973. At that time, the

fact remained that "definitive answers to questions about how best to

improve the education of young children from deprived homes have not emerged

from these programs . no model has come forth as robust in all or most

situations (nor) does the analysis promise to yield definitive prescrip-

tions" (Rivlin & Timpane, 1975, pp. 11-12).

Garry McDaniels (1975, pp. 52-53) summarized the problems in proving

an educational model successful. The answers to how much difference should

be expected at any given year between experimental program groups and

nonexperimental groups depends upon how the measurements taken are inter-

preted and on the comparisons made with "clearly no best comparison and

thus no best metric." He stressed the importance of a set of criteria

for success but questioned the criteria; and he noted that "any field

experiment is subject to a variety of influences that the experimenter

cannot control," with the implication that evaluation of field experiments

has been made most difficult under these givens.

Evaluation continued. Follow Through evaluations gave indication of

persistent problems in carrying out meaningful, objective critiques of

programs where even well-planned data collection procedures did not always

"mesh" with the program objectives and implementation, therefore causing

statistical analytic difficulties (Abt Associates, 1974-1976; SRI,

1974).










The Southwest Education Development Laboratory, from which Nedler

initiated research in intervention strategies for Spanish speaking

preschool age children, continued to study the longitudinal effects

of its bilingual early childhood educational programs. Early findings

indicated that the "educational disadvantage which the Mexican-American

child exhibits is, in large, a language deficit and not a total intellec-

tual deficit" and suggested that the findings supported "the need for

additional study of the effects of alternate strategies designed to

meet the needs of disadvantaged preschool children who speak a language

that differs from that of the wider community" (Nedler, 1971, p. 266).

As those research projects from Austin were field-tested in public

school classrooms during the 1971-1972 school year, the major findings

of the evaluation study included the following gains (Southwest

Educational Development Laboratory):

1. Children made significant gains between the pre and post

tests on both the English and Spanish versions of the

Auditory Test of Language Comprehension;

2. Children made significantly higher gains on a nonverbal

measure of general intellectual development, the Raven Pro-

gressive Matrices, giving indications of superior perceptual

and cognitive growth;

3. Seventy-five percent of the children successfully mastered

the objectives of the visual, auditory, prewriting, and

ideas and concepts elements of criterion-referenced tests;

Nedler, (1975, p. 482) reviewed the eight years of testing and

development that had gone into a model program to teach English as a

second language. From a naturalistic approach to a phonetic approach,

from an adaptation of a programmed approach to a fourth approach










designed to develop the child's basic English vocabulary (after concepts

included had been taught in Spanish) and to reach "certain related basic

structures of the English language and deal only with English syntas,"

the research has continually studied improved ways of helping Spanish

speaking children develop intellectually in an English-oriented school

environment.

In another report from a planned bilingual program in the El Paso

Public School District, where the intention was for all kindergartners

and first graders to be bilingual by the 1973-1974 school year, the initial

results of a study of second grade Mexican-American students showed that

those children tested in Spanish performed better in English studies

than their peers who had been instructed in English (Cross, & Bridgewater,

1973). The longitudinal effects of early bilingual education upon later

elementary school achievement are not known as yet; however, there has

been early indication that performance has increased when Spanish speaking

children are taught in their own language first with English taught as a

second language.

One of the most interesting and potentially valuable longitudinal

effects studies underway at the present time is that funded by the

Office of Child Development through the Education Commission of the

States. A consortium of twelve principal investigators is engaged in

"two interrelated efforts to provide better information on the actual

effects of preschools on low income children: (I) pooling the original

longitudinal data collected by the individual projects, and (2) collecting

current year follow-up data from all projects in identical formal for

use in both pooled and separate analyses" (Murray, 1977, p. 3).









Chaired by Irving Lazar, the consortium members are: Kuno Beller,

Martin Deutsch, Cynthia Deutsch, Ira Gordon, Susan Gray, Merle Karnes,

Phyllis Levenstein, Louise Miller, Francis Palmer, David Weikart,

Myron Woolman and Edward Zigler. The Murray study dealt primarily

with the results of studying the effects of attending a preschool

program or not attending one and emphasized the way in which family

variables affect evaluations.

Of particular interest to the researcher or evaluation specialist

is the consideration of home environment as it contributes to the school

achievement of young children. For example, in the research efforts

of Ira Gordon, significant influence from the home environment is indicated

upon children's school achievement. Parents who participated with

their children in a Home Learning Center program indicated a "change in

the mother's perception of herself and her child, in her role as teacher

of her child, and some change in the motivational as well as cognitive

system of the child" (Guinagh & Gordon. 1976, p. 48). Participation in

this program did make a difference in the school achievement of the

children several years later. Although this research emphasized the

involvement of parent and child in the home and did not stress a school

program, it is clear that the attitude of the participating parents as

compared with control group parents affected positively the academic

attainments of the children when they entered public school.

In a study at the University of Florida, Hanes and Shea (1977)

have shown direct relationship between the home environment and the

predicted school achievement, especially in reading, of children in

two of the Florida Parent Education Follow Through Model communities.

Walberg and Marjoribanks (1976, p. 548) concluded from their studies

of home environment contributions to cognitive gains in children in










school settings that . family environment is an obvious correlate

of the usual criterion tests" and suggested that . family environ-

ment measures should be considered for inclusion in experimental and

correlational studies of educational effects."

In the Murray paper (1977, p. 57), the conclusions of the analysis

of the data from the study indicates "that for lower class populations,

significant correlations exist between mother education, SES, family

size, and birth order and IQ but these correlations appear to be weaker

than they are for the general population." Murray (1977, p. 57)

further concludes that there is "little evidence that preschools in

general effect the correlations between these variables and IQ, although

there are a few instances in which such effects appear for a particular

preschool project."

What these recent studies of longitudinal effects of early

intervention programs seem to suggest is that there may well be other

variables to consider than just school achievement scores.

Summary

The literature indicates that legislation passed for financial

support from the federal government gave impetus to the early inter-

vention programs of the 1960's. The Economic Opportunity Act of 1964

provided opportunities for communities to direct attention to the

educational concerns for disadvantaged preschool children through the

initiation of Project Head Start. The Elementary and Secondary Act of

1965, and its amendment in 1966 to include monetary support for programs

aimed at migrant workers' children, placed responsibility on individual

state departments of education to plan and develop school programs

that would promote success for the educationally disadvantaged. With-

out these two landmark pieces of legislation, it seems unlikely that










the implementation of intervention programs could have been undertaken

in such large scale fashion.

Undoubtedly, the research of early childhood educators, psychologists,

and child development specialists contributed greatly to the decision

to begin such a large scale intervention into learning program. Hunt's

work indicating that intelligence is directly related to experience and

is not fixed nor predetermined along with Bloom's argument in favor of

the importance of the early years for intellectual development supported

other viewpoints suggesting that intervention into the disadvantaged

child's early learning could increase cognitive and affective development

in a child who might otherwise fail in the traditional educational process.

There were other research efforts that resulted in conflicting beliefs

as how best to intervene in the learning of young children. Bruner

interpreted the structure of knowledge in such a way that he adopted a

strong belief that there was nothing that could not be taught any child

if taught at the child's particular developmental level. On the other

hand, translations of Piaget's complex research writings evidenced his

equally strong conviction that children must progress quite systematically

through defined developmental states, that one stage led to the next,

and that learning was best not pushed when a child was not yet ready

for particular learning stages.

Such literature reflected the differing philosophical assumptions

of educators as early intervention programs were designed and implemented.

These differences varied from traditional child-oriented programs to

highly structured cognitive and behavioral programs to nonclassroom home

involvement programs. Such varying positions served as underpinnings for

the interventions that were designed. One fact was clear: the provision









of quality learning experiences for disadvantaged children was intended

to compensate for basic lacks in children's environments and attempts

were made to develop programs to prevent the cumulative deficits that

helped create the failure of many children to achieve in public schools.

The research efforts of Nedler at the Southwest Educational

Development Laboratory in bilingual education for young children was

also underway at the time of intervention program direction. The

acceptance of the Spanish speaking child's home language and the teaching

of basic concepts in Spanish before teaching in English was demonstrating

initial achievement gains for the children in these programs. However,

like all of the intervention programs regardless of philosophical bent

and implementation procedures, immediate gains were evidenced but

longitudinal effects are still cautiously interpreted at the time of

this present study.

Individual research projects, such as those of Weikart, Karnes,

Bereiter-Englemann, Gordon, Nimnicht and others demonstrated gains in

children beyond the short-term program. However, the Westinghouse Report

on the longitudinal evaluation of the field-based, nonexperimental

Head Start program caused considerable debate. Short-term gains were

reported, but those gains appeared to washout completely after first

grade--a disappointing result for those who had perhaps expected more

than was possible under the early implementation of that program. The

most promising results of the Cicirelli evaluation were the recommendations

for Planned Variation Head Start, the opportunity to compare varying

early intervention program implementations to attempt to determine the

best means of intervention, and Planned Variation Follow Through, the

opportunity to follow through early elementary grades with the same

intervention models as Head Start.






53


A scarcity of studies evaluating the longitudinal effects of early

intervention programs upon migrant children was discovered by the inves-

tigator. Perhaps this is partly explained by the problems inherent in

conducting such studies: the mobility of the children's families,

the bilingual-bicultural nature of the Mexican-American family and its

children, the lack of controlled evaluation procedures planned into the

program design and implementation. Thus it would appear that there is

a definite need for conducting further studies to help determine the

impact of field-based intervention programs for Spanish speaking migrant

children.


















CHAPTER III

DESIGN AND METHODOLOGY



The procedures followed in this study were those of "controlled

inquiry" (Kerlinger, 1973, p. 392), rather than true experimentation.

A quasi-experimental design, combining a multiple time-series design and

a pretest-posttest comparison group procedure, was used.

Riecken and Boruch (1974, p. 116) have described the problems in-

herent in nontrue experimental designs, but suggested that the multiple

time-series and pretest-posttest comparison group design in combination

makes possible "hardheaded evaluation" when experimental control is

not available. Popham (1975, p. 213) stated that the advantage of the

time-series design is "that it provides a post hoc tool to evaluate

educational interventions for which no comparison-group contract was

preplanned". Messick and Barrows (1972, p. 276) wrote that the "power

of this design nonequivalentt control-group pretest-posttest procedure]

is increased substantially if it is extended to a multiple time-series

having repeated measurement of the two groups over time". Kerlinger

(1973, pp. 341-346) further supported this compromise design: a long-

itudinal time design and a compromise experimental group-control design.

The results from combining these two designs can be interpreted with

caution when a true experimental approach to a study is not possible.

In selecting this combination of designs, the investigator did so

with the following considerations: (1) protests, given in kindergarten











to both the program and comparison group, would initiate the evaluation

study of school achievement--recognizing that a pretest after an

original treatment (the preschool intervention program) covaries out

some of the treatment effect; (2) treatment would be understood to be

those learning experiences planned through the regular school curriculum

and provided each group of children during each school year from kinder-

garten through third grade; (3) achievement tests, given the children

at each grade level would be used for evaluation of school performance

each year; (4) the final posttest would be those tests given at the third

grade level; and (5) the number of children retained at grade level by

the end of third grade would give further indication of school performance.

Although this evaluation study did not provide for a controlled and

predetermined experimental treatment at intermittent time periods, the

consideration of on-going school learning experiences and opportunities

along with periodic measurement processes seemed to satisfy the require-

ments for a modification of the multiple-time series design suggested

in Campbell and Stanley (1966, pp. 37-45). The use of pretest-posttest

measurements with a comparison group satisfied the conditions required

for the nonequivalent control group design (Campbell & Stanley, 1966,

pp. 47-50).

Campbell and Stanley (1966. p. 40) forewarned of potential sources

of invalidity for quasi-experimental designs. In combining these two

designs, there is some indication that the various sources of experi-

mental crrur, especially threats to internal validity, can be reduced

to a minimum. History, as a source of internal validity in the multiple

time-series design, is controlled for by the comparison group being

added. The possibility of instrumentation being a threat to internal











validity in the time-series design is controlled for through the

com''ining of the comparison group design, and the interaction of

selection and maturation of subjects in the comparison group design

is more adequately controlled for in combination with the time-series

design.

The effects of history and maturation are considered as occurring

within each of the groups over the same period of time, the four years

including kindergarten through third grade years. Although the subjects

from the intervention program group were given the Preschool Inventory

(Caldwell, 1967) in May of 1971, it was doubtful that the effects of

such protesting at that time would present a threat to test-retest

validity potential a year and a half later when the same test was

administered at the beginning of kindergarten to both groups.

The effect of instrumentation on internal validity was controlled

for with tests given to subjects during the same period of time and in

the same testing environment. The exception to this was in the case of

make-up tests given to a single subject; this occurred only in two or

three instances, according to the school records. Neither the program

group nor the comparison group was selected on the basis of test

criteria.

Mortality was not controlled for in as much as only twenty of the

twenty-six Mexican-American children from the preschool intervention

group were enrolled in Hardee County schools from kindergarten through

third grade. There was no way of knowing how the absence of those six

subjects contributed to the differences in group means on tests.

The variables of teacher selection and teacher-student interaction

that might have had direct effect on the subjects' school achievement

were among those variables in this evaluation for which there were no











controls. Selection of tests to measure specific program criteria was

another uncontrolled factor. These are all recognized as problems

inherent in non-true experimental designs and will be dealt with

accordingly in a cautious interpretation of the results of this

evaluation.

Subject Selection Procedures

Two groups of subjects were used in this study: the preschool

intervention program group and a comparison group. Twenty-two of the

twenty-eight children who were enrolled for the entire original pilot

program were still attending school in Hardee County through the 1975-

1976 school year. Because it was the intent of this study to evaluate

the program's effects on the school achievement of the Mexican-American

children from the program, two of the twenty-two were excluded from the

study on the basis of race: one black, and one white

The comparison group was selected from enrollment registration

lists for the kindergarten year, 1972-1973, when the program group

entered kindergarten. The forty-eight Spanish surnamed children who

entered kindergarten that fall were then checked for present enrollment

in that county school system. Excluding the intervention program

children, fourteen children were available for the comparison group

composite. Families were contacted for permission for the children

to be included in the study. Permission was granted by all fourteen

families for the investigator to have access to school records necessary

for data collection.

When the two groups were studied for demographic comparisons, the

similarity between the two groups was much greater than might have been

expected. School records did not provide some of the information needed










for statistical analysis of the data; nor was the information contained

in cumulative records about the families precise enough for true analysis

of demographic data. Moreover, some of the information obtained came

from self-report of parents visited by the investigator, from direct

observation of the investigator in visiting the migrant communities, and

from the verbal reports of those schools personnel who know the families

well. These observations, verbal reports, and what recorded data were

available served as the basis for the following demographic comparisons

to be made.

All children fell within the Florida kindergarten entrance age, from

four years, nine months (the youngest) to five years, eight months (the

oldest) with the majority between four years, ten months and five years,

four months. There were a total of seventeen males and seventeen females,

with the program group having a ratio of twelve males to eight females

and the comparison group five males to nine females.

Spanish is the native home language for all of the families. It is

the language used for family communication as well as the social language

between friends. Although some of the parents understand and are able

to use English when necessary, they choose to speak Spanish in their

home environment. Most of the children find little reinforcement at

home for the English used in school. In those homes where limited

English is spoken along with the Spanish, children use the hybrid

language, "Spanglish."

The migrant population, as defined by the United States Office of

Education, are those agricultural workers who move across county or

state lines for the purpose of seeking work or who have so moved for

employment reasons within the last five years. Many of the families of











children in both groups are either presently in the interstate or

intrastate migrant stream. Some of the families in each group are

becoming "resident" migrant--they now live more permanently in the

area although their work continues to be that of the migrant laborer.

Few of the families have been out of the migrant population as defined

by state and federal criteria for more than two or three years; this

number may be three or four such families.

The employment level of the fathers of each group of subjects is,

for the most part, that of unskilled labor in the citrus groves,

vegetable fields, or in packing plants. Three fathers, two of program

children and one of a child in the comparison group, are skilled laborers.

A relatively equal number of mothers work from each group; about the

same proportion of mothers are unemployed and are listed in school

records as housewives. Twelve mothers, seven from the program group

and five from the comparison group, did not have employment status listed

in school records. Information provided by school paraprofessionals

knowing these families suggested that some of these mothers work at

times, but not necessarily all of the time.

As might be expected from the employment level of the adult family

members, the educational level of the parents is quite low. This infor-

mation is not contained in school records. However, from self-report and

from information given by school personnel, it is safe to assume that

not more than two or three parents have high school education. Most of

them have less than eighth grade schooling with some of the parents

having attended only sporadically for three or four years.

Families live in either rented trailers or small houses available

for migrant workers near the groves and in migrant worker communities.











Some families live in or very near the three small towns in the county

in similar rental housing or, in a few instances, in trailers or

houses of their own.

The number of children in the families was also noted. In the

program group, ten families have between one and three children and

ten families have between four and eight children. In the comparison

group, four families have between one and three children with ten families

having between four and eight children. Within the program group, five

children were the oldest, four were the youngest, and two were only

children; in the comparison group, three were the oldest, four were the

youngest, and there were no only children.

From the demographic data, it was evident that the comparison

group shared very similar home environments with the program group.

This finding was important since there were no other children from

which to select a comparison group who entered kindergarten in the fall

of 1972, who were of Mexican-American background, who had Spanish as

the home language, and who were in school in Hardee County, Florida,at

the time of this study.

Instrumentation

The instruments used in the study were those tests given to child-

ren in Hardee County's elementary schools. They were the measures used

to determine, in part, school success from an achievement point of

view for all children.

During the 1972-73 through 1975-76 school years, with which this

study is concerned, the following tests were administered in the fall

and in the spring and within two week intervals: the Preschool Inventory

(Caldwell, 1967) upon entrance into kindergarten, followed by the










Metropolitan Readiness Test, Form B (MRT-B) (Hildreth, Griffiths, &

McGauvran, 1969) in the spring of that year; Form A of the MRT in the

fall of first grade and, in some instances, the Metropolitan Achievement

Test (MAT) (1971) or the Comprehensive Test of Basic Skills (CTBS)

(1973) at the end of first grade. In second and third grades, children

were given the Florida Statewide Assessment Test (Testing Division,

State Department of Education, Florida) for math and communication

skills achievement in the fall of the year.

The Caldwell Preschool Inventory is an instrument designed to be

administered to individual children between three and six years old.

This instrument was developed by Bettye Caldwell during the early

summer weeks of the 1965 Head Start programs to provide an assessment

of achievement in those areas regarded as important and necessary

for school success. Although the Preschool Inventory was not developed

as a test of basic intelligence, it contains items for assessment

that are correlated with performance on intelligence tests. The

instrument's goals are to assess the degree of disadvantagement

which children from educationally and socially-economic disadvantaged

home environments have when entering school. Expectations are that

these observed deficits can be eliminated or reduced during the school

experience and therefore show changes associated with educational

intervention in a pretest-posttest measurement (Preschool Inventory

Handbook, 1970).

The screening instrument was field-tested nationally during the

first Head Start summer and preliminary standardization determined.

On measures of internal consistency, the reliability coefficients for

the total sample used for standardization were .91 on the Kuder-Richardson









(KR20) coefficient and .92 on the corrected split-half coefficient. The

test is a relatively difficult test for all age groups with the mean

for the percent of number of items ranging from 25.6 at the youngest

age group to 42.4 for the oldest age group. The standard error of

measurement is from 3.1 to 3.9 in the standardized sample (Preschool

Inventory Handbook, 1970, pp. 20-22).

The MRT was designed to measure the extent to which beginning

school children have developed skills and abilities that contribute

to first-grade instructional readiness. It serves to assist teachers

in classifying children as to the particular help and attention they

may need in further development of skills necessary for language

attainment, visual and auditory perception, number knowledge, motor

skills and muscular coordination, and the ability to attend and follow

directions. This test has been standardized with the predictive validity

of the MRT across groups estimated at .60, a value considered very

good for test results for five and six-year old children taking their

first group-administered test lUildreth etaL, 1969, pp. 27-28). The

Manual further indicated that the split-half value ranged from .90

to .95 with the measurement error of an individual total score ranging

from three to five points.

The CTBS is intended to be given students who have been taught

according to differing instructional approaches with the aim of measuring

those academic skills common to all curricula (CTBS Examiner's Manual,

1974). A total battery score is produced on subtests measuring reading,

language, and math concepts with science and social studies added to

the higher levels of the test batteries. All of the items on the

battery met the content criteria for the test; an appropriate level

of difficulty for the grades for which they were intended was determined;










and they had point-biserial correlation coefficients greater than .20.

CTBS was standardized on a large national sample of students from

kindergarten through grade twelve, randomly selected from all parts

of the United States in 1972-1973 (CTBS Examiner's Manual, 1974, p. 2).

The Florida Statewide Assessment Test for third grade students

was designed to "measure specific educational objectives which identify

the skills Florida students should achieve from their educational

experience" (Department of Education, State of Florida, 1973, p. iii).

Following the procedure established for a Statewide Assessment Program

under the Educational Accountability Act of 1971, objectives, chosen

by teachers and educators throughout the state, identified a number of

reading-related and math-related skills. Achievement of the objectives

is measured through objective-referenced tests with each objective

measured by one or more items. Furthermore, the Florida Assessment

Test is weighted for school districts and individual schools within

each district on five nonschool factors related to student test perfor-

mance: income of families, minority group membership, occupation of

parent, college education attainment of parent, and Spanish as the

family native language. A predicted range for scores is assigned

to the school with observed school scores and individual scores used

in comparison. This test is in the process of validation.

Data Selection by the Investigator

In studying the school records, it was evident that children were

absent sometimes at the administration of various tests; and, although

make-up tests were given in a few cases, this apparently was not always

possible or practical as evidenced by the absence of many scores for

data analysis. The three schools did not use the same tests each

year and in instances where children changed schools during the year,

the test data reflected this.









It was necessary for the investigator to determine which test scores

to use in the data analysis so that the largest number of subjects

would be represented by the achievement results and at the same testing

period each year. Although the Bianchini and Loret Anchor Test Study

Linn, 1975) has provided means for converting eight different achieve-

ment test scores to a similar score scale for fourth, fifth and sixth

grade levels, there is no procedure for converting test scores to a

single score scale for lower elementary grade levels. It was the

decision of the investigator to analyze the data from the fall testing

periods, the most consistent testing time for the most subjects in this

study.

Data from the kindergarten entrance scores on the Caldwell Preschool

Inventory were analyzed using the total raw scores. The Handbook for

the Preschool Inventory (1970) does not recommend that subtest scores

be routinely obtained for all children given this test. On the basis

of this recommendation, it was decided to analyze the total raw score

data.

The fall scores on the MRT-A given to first graders were analyzed,

also using total raw scores. The decision for using total raw scores

was made on basis of the "total score (numbers and copying also included)

providing a slightly better prediction of reading and spelling than do

the four Readiness tests which might appear to be more directly

related to such learning" (Hildreth, Griffiths & McGauvran, 1969, p. 17).

The CTBS total battery was analyzed for both second and third grade

subjects. The subtest scores for language were analyzed separately

because of the emphasis placed on language development in the preschool

program. It was of interest to the investigator to determine whether

or not the language component of the intervention program would appear









to have a relationship to second and third grade language achievement.

The raw scores were converted to standard scale scores to enable analysis

of data when different levels of the test had been given to different

children in the same grade. This is a procedure recommended in the CTBS

Examiner's Manual, Expanded Edition (1974).

The communication skills subtest of the Florida Statewide Assessment

Test for third grade students was also included in the analytic processes.

Again, the language area was of particular interest to this study

because of the preschool program emphasis on language development. These

five tests were those given to the largest number of subjects from each

of the two groups at the same testing period. The dependent variables

were analyzed in relationship to the independent variable, group

membership.

The computing services available to the University of Florida through

the Northeast Regional Data Center were used for a one-way analysis of

variance procedure for each of the dependent variables analyzed in rela-

tionship to the independent variable group means for each of the tests

at each grade level analyzed in relationship to group membership. The

statistical package program, SPSS (Nie, Hull, Jenkins, Steinbrenner,

Bent, 1975) was used. F ratios were converted to t scores at the a=.05

level for a one-tailed test (Roscoe. 1975, p. 288).

The z ratio for independent proportions was used to analyze the

data pertaining to the number of subjects from each group retained at

grade level by the end of third grade as compared to those from each

group who were promoted. This is a procedure recommended for determining

a difference between uncorrelated proportions (Guilford & Fruchter,

1973, p. 162).

Attendance was studied, but no statistical analysis of the data was

made. Information available from cululative records of the subjects










was not detailed enough to enable statistical analysis. The information

available will be reported in narrative fashion.

Statement of the Null Hypothesis for Statistical Analysis

To answer the research question as to whether or not the intervention

program evidenced longitudinal effects on the school achievement of

participating children, the predictive hypotheses were restated in null

hypothesis form for the puprose of statistical analysis of data.

1. The group mean for the preschool intervention group on the

Preschool Inventory given at entrance into kindergarten will

be equal or less than the group mean for the comparison

group.

2. The group mean for the preschool intervention group on the

MRT-A given in first grade will be equal to or less than

the group mean for the comparison group.

3. The group mean for the preschool intervention group on the

CTBS total battery given at second grade will be equal to

or less than the group mean for the comparison group.

4. The group mean for the preschool intervention group on the

language component of the CTBS given at second grade will

be equal to or less than the group mean for the comparison

group.

5. The group mean for the preschool intervention group on

the CTBS total battery at third grade will be equal to or

less than the group mean for the comparison group.

6. The group mean for the preschool intervention group on the

language component of the CTBS given at third grade will

be equal to or less than the group mean for the comparison

group.









7. The group mean for the preschool intervention group on

the communication skills subtest of the Florida Statewide

Assessment Test at third grade will be equal to or less

than the group mean for the comparison group.

8. The number of children in the preschool intervention group

retained at a grade level by the end of third grade will

be equal to or more than the number of children retained

from the comparison group.

Summary of the Design and Methodology

To make possible the evaluation of the longitudinal effects of a

preschool intervention program for three-year old Mexican-American migrant

children upon their school achievement from kindergarten through third

grade, a quasi-experimental post hoc procedure was used. The study design

combined a multiple time-series design and a non equivalent control-group

pretest-posttest procedure, with the comparison group selected to attain

the greatest similarity possible.

Conventional analytic techniques were used in studying the relation-

ship of the independent variables, scores on each of the tests at each

grade level, to the dependent variable, group membership as represented

by the program group and the comparison group. The conversion of F ratios to

t scores was used for the one-tailed test at the .05 level of signi-

ficance in a one-way analysis of variance procedure. The z ratio for

independent proportions was used to determine the difference between

the number of subjects in each group retained at grade level by the end

of third grade.


















CHAPTER IV

RESULTS OF THE STUDY



The purpose of this study was to evaluate the longitudinal effects

of a preschool intervention program in Hardee County, Florida, on the

school performance of Mexican-American children in the primary grades.

The twenty Mexican-American children from the original intervention

program who were still attending school in that county through third

grade were considered members of the program group. Fourteen Mexican-

American migrant children who entered kindergarten in the fall of 1972

at the same time asthe program subjects comprised the comparison group.

In ex post facto quasi-experimental studies, where a control group

is not planned a prior, there is a risk of selecting a nonequivalent

comparison group. The comparison group for this study was scrutenized

carefully to determine similarity to the original program group. When

comparisons were made on the basis of the employment and educational levels

of parents, the number of children in the family and birth order, the

migrant status of families, age and entrance into public school,and

Spanish as the home language, the two groups were found to be homogeneous

in terms of home environment conditions that might contribute to any

differences in achievement scores found in the assessments made at each

grade level. It was assumed that differences in the two groups would

result from attending or not attending the preschool intervention program.









Data were collected from public school records on assessment measures

used in that county school system to determine school success of children

from kindergarten entrance through third grade. Although there were

thirty-four subjects in the combined groups, the number of subjects for

whom scores were recorded in school records varied due to late enrollment,

absence from school at the time of the testing program,or because sub-

jects had not been administered the same tests. Each test result

reflected these differences in total number of scores recorded and is

reported in each data analysis.

The results of the data analyses for testing each of the null

hypotheses related to the research question are reported in the following

sections.

Test of the First Hypothesis: The group mean for the preschool inter-

vention group on the Preschool Inventory

given at entrance into kindergarten will

be equal to or less than the group mean

for the comparison group.

A one-way analysis of variance procedure was used to test this null

hypothesis. In a two-sample situation, this statistical procedure is

mathematically equivalent to the t test (Roscoe, 1975, p. 292). Conver-

ting the F ratio to a t score for a one-tailed test (Roscoe, 1975, p. 288),

the computed t of 2.1897 was found to be significant at a = .05 level.

Group means and standard deviations, using raw score totals, are shown

in Table 1 along with the computed t score.









Table 1

MEANS AND STANDARD DEVIATIONS FOR THE PROGRAM GROUP AND THE
COMPARISON GROUP ON THE PRESCHOOL INVENTORY AT KINDERGARTEN ENTRANCE

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 12 37.0000 9.8719 2.1897*

Comparison 5 25.0000 11.3798


*Significant at p < .05

On the basis of this statistical analysis, the null hypothesis was

rejected; the program intervention subjects scored higher on the

Preschool Inventory than the comparison group subjects. The Preschool

Inventory Handbook (1970, p. 4) stated that the Inventory was developed

as "an instrument that was sensitive to experiences and could thus be

used to demonstrate changes associated with education intervention."

The educational significance of the difference in the test means between

the two groups lies in the assumption that the preschool intervention

program provided educational experiences for that group of children

which evidenced increased readiness for kindergarten.

Test of the Second Hypothesis: The group mean for the preschool inter-

vention group on the MRT-A given in

first grade will be equal to or less

than the group mean for the comparison

group.

A one-way analysis of variance procedure was used to test this null

hypothesis as in the previous data analysis. Again, the F ratio provided

with this analytic technique was converted to a t score for a one-tailed

test of significance. Means and standard deviations for raw score totals

along with the t score at the a = .05 level are shown in Table 2.









TABLE 2

MEANS AND STANDARD DEVIATIONS ON RAW SCORE TOTALS FOR
THE METROPOLITAN READINESS TEST, FORM A, AT THE FIRST
GRADE FOR THE PROGRAM AND COMPARISON GROUPS

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 17 53.2941 11.7035 1.7306*

Comparison 11 44.0000 16.7809


*Significant at p < .05

The null hypothesis was rejected; the program intervention group

scored statistically significantly higher on the MRT-A in the fall of

first grade.

The MRT provides for an interpretation of scores by converting the

total scores to letter ratings on a five-point scale, from A (high) to

E (low). The significance of this "readiness status" by letter rating

is to enable teachers "to group pupils into five readiness levels for

the provision of instruction best suited to the present status of each

group" (Hildreth, Griffiths, McGauvran, 1969, p. 11). These predictive

characteristics of the MRT indicated that thirteen of the seventeen

program group subjects fell in the B and C letter rating, suggesting

average to high normal readiness status. Four program subjects were

viewed as being likely to have difficulty in first grade work as

evidenced from a letter rating of D, low normal readiness status. In

the comparison group, six of the eleven subjects were in the B and C

rating for readiness with five subjects in the D and E letter rating,

low to low normal readiness status. These letter ratings for readiness

status indicated support for the statistical analysis which rejected

the null hypothesis.









Test of the Third Hypothesis: The group mean for the preschool inter-

vention group on CTBS total battery given

at second grade will be equal to or less

than the group mean for the comparison

group.

Raw scores on this test battery were converted to expanded standard

scale scores so that different levels of the CTBS could be equated (CTBS

Examiner's Manual, Expanded Edition, 1974, p. 74). The total battery was

comprised of scores from math, reading, and language subtests. Analysis

was made with a one-way analysis of variance procedure as in the previous

analytic procedures.

Although there was a mean difference of 19.1190 between the total

battery scores of two groups, it was not statistically significant at the

a = .05 level. The null hypothesis that the program group would score

equal to or less than the comparison group was retained. The means and

standard deviations for test scores in this analysis are shown in

Table 3.

TABLE 3

MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD SCALE
SCORES ON THE TOTAL BATTERY, COMPREHENSIVE TEST OF BASIC SKILLS,
FOR SECOND GRADE FOR THE PROGRAM AND COMPARISON GROUPS

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 18 168.8333 41.4165 1.1392 (n.s.)

Comparison 7 149.7143 24.0743


It should be pointed out that four of the comparison subjects for

whom scores were analyzed at the first grade level were retained at that

grade and were not included in the second grade test analysis.









Test of the Fourth Hypothesis: The group mean for the preschool inter-

vention group on the language component

of the CTBS given at second grade will

be equal to or less thanthe group mean

for the comparison group.

The mean difference of 34.6587 on the language component of the CTBS

battery was found statistically significant at the a = .05 level for a

one-tailed test using the one-way analysis of variance procedure with

the F ratio converted to a t score. Means and standard deviations for

these data are shown in Table 4.

TABLE 4

MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD SCALE
SCORES ON THE LANGUAGE COMPONENT, COMPREHENSIVE TEST OF BASIC
SKILLS, FOR SECOND GRADE FOR THE PROGRAM AND COMPARISON GROUPS

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 18 247.94444 42.8671 1.8365*

Comparison 7 213.2857 40.9094


*Significant at p < .05

The null hypothesis was rejected; the program subjects scored higher

on the language subtest of the CTBS than the comparison group. Even though

the null hypothesis that the program group would do equally well or less

well on the total CTBS battery than the comparison group was retained,

the rejection of this fourth hypothesis seemed to have educational signi-

ficance. It appeared that the program subjects were better able to use the

language necessitated for this grade level achievement test than the

comparison group.

Test of the Fifth Hypothesis: The group mean for the preschool intervention

group on the CTBS total battery at third grade









Test of the Fifth Hypothesis (cont.): will be equal to or less than the

group mean for the comparison group.

The same statistical procedure was followed for analysing these data

as for previous data. Expanded standard scale scores were used so that

different levels of the CTBS could be equated. The null hypothesis was

retained when the analysis showed no statistically significant difference

between the mean scores of the two groups on the total test battery.

The means and standard deviations are shown in Table 5.

TABLE 5

MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD SCALE
SCORES ON THE TOTAL BATTERY, COMPREHENSIVE TEST OF BASIC SKILLS,
FOR THIRD GRADE FOR THE PROGRAM AND COMPREHENSIVE GROUPS

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 14 215.1429 52.3110 0.4037 (n.s.)

Comparison 7 224.2857 40.7870


Test of Sixth Hypothesis: The group mean for the preschool intervention

group on the language component of the

CTBS given at third grade will be equal

or less than the group mean for the compar-

ison group.

Means and standard deviations for the language component scores on

the CTBS are shown in Table 6. A one-way analysis of variance procedure

was used for these data and, with the results of the analysis, the null

hypothesis was retained at the .05 level of significance.






75

TABLE 6

MEANS AND STANDARD DEVIATIONS FOR EXPANDED STANDARD SCALE
SCORES ON THE LANGUAGE COMPONENT, COMPREHENSIVE TEST OF BASIC
SKILLS, FOR THIRD GRADE FOR THE PROGRAM AND COMPARISON GROUPS

GROUP NUMBER MEAN STANDARD DEVIATION t SCORE

Program 14 276.2856 62.9517 0.1483 (n.s.)

Comparison 7 280.2856 43.8665


It is at this analytic step that the two group means show no statis-

tically significant differences on any measure at a grade level--the

total battery scores and the language subtest scores on the third grade

CTBS. Although it should be pointed out once more that the four compari-

son children retained at first grade were not included in these test

results, the same children taking the second grade test from the compari-

son group were those taking the third grade test. With this in mind,

it appears that any effect of the preschool intervention program on

school performance at third grade had washed out so that the comparison

group had now caught up with the program group.

Test of the Seventh Hypothesis: The group mean for the preschool inter-

vention group on the communication skills

subtest of the Florida Statewide Assess-

ment Test at third grade will be equal to

or less than the group mean for the

comparison group.

The communication skills subtest of the Florida Statewide Assessment

Test at third grade was the final test to be analyzed. Again, the one-

way analysis of variance procedure was used for data analysis, converting

the F ratio to a t score for a one-tailed test. The equality of school

achievement in this area for both groups in third grade was confirmed with

the results shown in Table 7. The null hypothesis was retained at the

a = .05 level.
















GR(

Pr

Cor


76

TABLE 7

MEANS AND STANDARD DEVIATIONS FOR RAW SCORES ON THE COMMUNICATION
SKILLS COMPONENT OF THE FLORIDA STATEWIDE ASSESSMENT TEST FOR
THE PROGRAM AND COMPARISON GROUPS AT THIRD GRADE

OUP NUMBER MEAN STANDARD DEVIATION t SCORE

ogram 11 71.0000 16.1555 0.1303

mparison 8 72.0000 17.5255


(n.s.)


Test of the Eighth Hypothesis: The number of children in the preschool

intervention group retained at a

grade level by the end of third grade

will be equal to or more than the

number of children retained from the

comparison group.

To test this final hypothesis, the z ratio for independent proportions

was used to analyze the data. Of the twenty program subjects, one was

retained in first grade and one at the end of third grade. Of the four-

teen comparison subjects in the study, three were retained at first grade,

one was assigned to a special education classroom and two were retained

at the end of third grade. The results of the analytic procedure are

shown in Table 3, with the rejection of the null hypothesis at the a = .05

level.

TABLE 8

COMPARISON OF SUBJECTS RETAINED OR PROMOTED
AT THE END OF THIRD GRADE FROM EACH OF THE
TWO GROUPS

PROMOTED RETAINED


Program Group*

Comparison Group


6 (including one subject
assigned to special
education)


*Significant: z = -2.231, p > .05









Consideration of Attendance During First, Second, and Third Grade for
Both Groups

No statistical analysis was made on the attendance factor as related

to school performance. The data available and collected from school

records, indicating the number of weeks children were enrolled and the

number of days absent during the year, was not sufficient for statistical

analysis. However, there were interesting observations to be seen from

some of the data recorded in cumulative records.

Days absent for program subjects ranged from 0 to 41 days in first

grade; from 0 to 31 days in second; from 0 to 21 days in third. A total

of nine different children in this group entered school late or exited

early during those three years, thereby being enrolled for less than

the 36 weeks of the school year. The number of weeks enrolled for program

group subjects were 12, 22, 24, 30, 31, 33, 35, and the full 36 weeks.

Days absent for comparison group subjects ranged from 1 to 22 days

in first grade; from 1 to 22 days in second; from 0 to 32 days in third

grade. A total of seven different children in this group entered school

late or existed early during the three year period. The number of weeks

enrolled for the comparison group subjects were 9, 18, 24, 26, 27, 30,

34, 23, and the full 36 weeks. The subjects retained at first grade

were not included in the second and third grade information on attendance

since their records were not used for test data collection after first

grade.

Certainly the mobility of the migrant family is substantiated in

these data. The pattern of absenteeism reinforced the concern expressed

by county school personnel for the success of migrant children in that

school system. The expectation that both groups of children would evidence

the same enrollment and absenteeism patterns was substantiated with the

attendance information presented here.






78

Consideration of the Research Question

It is evident from the statistical analyses of the data used in this

study that the subjects from the preschool intervention program scored

higher on standardized tests at the entrance into kindergarten, and in

the fall of first grade, as well as on the language component of the CTBS

in second grade. Although there were no statistically significant differ-

ences between the program group and the comparison group on the tests

given at the third grade level, the number of program children promoted

by the end of third grade as contrasted to the number of comparison

children promoted by the same time was statistically significant in

favor of the program group. With these considerations, the judgement

of the investigator is that the preschool intervention program did

contribute positively to the school achievement of the Mexican-American

children who participated in that program. The contributions of the pro-

gramappeared to be in the area of language development, evidenced not

only in the language component of the second grade CTBS but also in the

first grade MRT which has a strong emphasis on language skills for

reading readiness. A full discussion of the educational significance of

the statistical findings of this study is found in the following chapter.

Summary of the Results of the Study

The statistical analyses of data for each grade level, comparing the

program intervention group and the comparison group, provided the investi-

gator with evidence for the rejection or retention of each null hypothesis.

This information is summarized as follows:

1. The null hypothesis that the group mean for the preschool inter-

vention group on the Preschool Inventory given at entrance into

kindergarten will be equal to or less than the group mean for

the comparison group was rejected.






79

2. The null hypothesis that the group mean for the preschool inter-

vention group on the MRT-A given in first grade will be equal to

or less than the group mean for the comparison group was rejected.

3. The null hypothesis that the group mean for the preschool inter-

vention group on the CTBS total battery given at second grade

will be equal to or less than the group mean for the comparison

group was retained.

4. The null hypothesis that the group mean for the preschool inter-

vention group on the language component of the CTBS given at

second grade will be equal to or less than the group mean for

the comparison group was rejected.

5. The null hypothesis that the group mean for the preschool inter-

vention group on the CTBS total battery given at third grade

will be equal to or less than the group mean for the comparison

group was retained.

6. The null hypothesis that the group mean for the preschool inter-

vention group on the language component of the CTBS given at

third grade will be equal to or less than the group mean for the

comparison group was retained.

7. The null hypothesis that the group mean for the preschool inter-

vention group on the communication skills subtest of the Florida

Statewide Assessment Test at third grade will be equal to or less

than the group mean for the comparison group was retained.

8. The null hypothesis that the number of children in the preschool

intervention group retained at a grade level by the end of third

grade will be equal to or more than the number of children retained

from the comparison group was rejected.






80

The findings supported the conclusion that the preschool intervention

program contributed in positive ways to the school achievement of the

participating Mexican-American children even though the statistical evidence

is not powerful in its entirety.


















CHAPTER V

SUMMARY, DISCUSSION, AND RECOMMENDATIONS



Summary

In the late fall of 1970, this investigator was asked to design an

educational curriculum for a pilot program for three-year old migrant

children in Hardee County, Florida. The preschool intervention program,

operational from January 1971 through May 1971, was sponsored by the

Institite for Development of Human Resources, University of Florida, in

cooperation with the public schools of that county and funded by the

Migrant Section of the State Department of Education.

The purpose of the present study was to evaluate the longitudinal

effects of the preschool intervention program on the school achievement

from kindergarten through third grade of the Mexican-American migrant

children who participated in the early childhood program. Twenty children

of Mexican-American backgrounds from the original program were enrolled

in school in Hardee County at the time of the study. Fourteen other

Mexican-American children who entered kindergarten at the same time as

the program group, in the fall of 1972, were selected for an ex post facto

comparison group.

School records for the total thirty-four subjects were studied as

sources of data from which the evaluation of school achievement was made.

Scores from standardized tests,adopted by the school system and given in

the fall of each grade level year, were used as measures of school






82

achievement. Each of the scores for the two groups of subjects were

analyzed with a one-way analysis of variance procedure, converting F

ratios to t scores at the .05 level of significance for a one-tailed

test. The z ratio for independent proportions was used to analyze the

data comparing the number of subjects from each group retained at grade

level by the end of third grade. Attendance and enrollment data were

studied but with no statistical analysis made of the data in relationship

to school achievement. From the analysis of all data, the following

results were obtained:

1. The program intervention group scored significantly higher on

the Preschool Inventory at entrance into kindergarten than the

comparison group.

2. The program intervention group scored significantly higher on

the Metropolitan Readiness Test, form A, in the fall of first

grade than the comparison group.

3. There was no statistically significant difference between the

two groups in mean scores for the total battery, Comprehensive

Test of Basic Skills, given in the fall of second grade.

4. The program intervention group scored significantly higher on

the language component of the Comprehensive Test of Basic Skills

in the fall of second grade than the comparison group.

5. There was no statistically significant difference between the

two groups in mean scores for the total battery, Comprehensive

Test of Basic Skills, given in the fall of third grade.

6. There was no statistically significant difference between the

two groups in mean scores for the language component, Compre-

hensive Test of Basic Skills, given in the fall of third grade.








7. There was no statistically significant difference between the

two groups in mean scores for the communication skills component

of the Florida Statewide Assessment Test given in the fall of

third grade.

8. The program intervention group had significantly fewer subjects

retained at grade level by the end of third grade than the

comparison group.

Although the longitudinal gains in achievement evidenced "wash-out"

by the fall of third grade for the program group, it was apparent that

the preschool intervention program did have a positive effect into second

grade. The number of subjects retained at grade level by the end of

third grade for each group was significantly different in favor of the

program group. The educational significance of these statistical findings

are discussed in the following section to provide a basis for the professional

judgement of the investigator that the preschool intervention program

contributed in positive ways to the school achievement of the Mexican-

American migrant children who attended the program and who were studied in

this evaluation.

Discussion

This study has been one of controlled inquiry rather than true

experimentation. Post hoc studies can only be made with clear understanding

of the lack of those controls that limit investigation to a quasi-

experimental design and therefore to systematic scrutiny of the available

data for conceptualization of findings rather than just statistical analyses

considered. Nonrandomization of subjects, an ex post facto comparison

group, lack of control for instrumentation, noncontrol for teacher-student-

classroom interaction all contributed to the evaluation difficulties

inherent in this particular nontrue experimental study. Within the









limitations imposed by the available data, the findings must be interpreted

with a degree of caution; and, yet, meaningful trends emerged.

Like Head Start as well as other early childhood intervention programs

discussed in Chapter II, there were immediate gains evidenced in the pre

and posttest scores at the beginning and ending of the preschool inter-

vention program itself (Newell, et al., 1971). The severe time limitations

of that five-month program prompted questioning as to whether or not

such gains could last in measurable ways through another year before

entrance into kindergarten. The Preschool Inventory, however, indicated

statistically significant differences between the scores of the program

group and the comparison group, indicating that the experiences provided

by the preschool program contributed to learning changes associated with

educational intervention. Although the number of comparison subjects

taking that test at the beginning of kindergarten was relatively small,

the program subjects did show a "head start" into public school. This

suggested that the experiences for cognitive, social,and language develop-

ment that were a part of the planned preschool program prepared the

program group more than just the home readiness factors contributed to

the comparison group's readiness for kindergarten.

It might have been expected that Mexican-American migrant children,

coming from culturally different and educationally disadvantaged homes

with Spanish rather than English as the home language for communication

and learning, would score considerably lower than a cross-section of

English speaking disadvantaged children the same age. However, it was

interesting to note that in determining norms for the Preschool Inventory,

based on Head Start data throughout the United States in the fall of 1968,

the mean and standard deviation for norms considered as mid-percentile

ranks for age five years to five years, five months were 38.4 and 9.7









respectively for scores obtained from children in the South East region

(Preschool Inventory Handbook, 1970, p. 18). This indicated only a slight

difference in mean for the program group as compared to the regional norms

for the Inventory. It appeared that the preschool intervention program

compensated for their potentially increased deficits in school readiness

and contributed to raising the Mexican-American migrant children's level

of school preparedness.

A larger number of subjects from both groups had recorded scores for

the Metropolitan Readiness Test in the fall of first grade. These scores,

too, were statistically significant with the program group evidencing

increased skills deemed necessary for first grade success. It would seem

that the opportunity to be in a school setting,and to experience at an

earlier age than kindergarten an educational curriculum intended to

promote-language and basic cognitive skills, tended to enhance school

readiness in both kindergarten and first grade for the program group.

By the end of first grade, only one of the program subjects was

retained at grade level; three of the comparison subjects were retained

in first grade and one other child was assigned to a special education

classroom. The statistical evidence, along with the promotion and retention

factors, point to the fact that three years after the pilot preschool

program ended in May of 1971, the children under study were maintaining

longitudinal effects from the experience through the end of first grade,

May 1974.

In the fall of second grade, the total battery scores on the Comprehen-

sive Test of Basic Skills showed a mathematical difference in means of

over nineteen points in favor of the program group. This difference was

not statistically significant, however, even though there was a trend

towards greater school achievement for the program group. The significant

difference was found in the language subtest scores of the CTBS at this









grade level. It seemed apparent that the language emphasis in the preschool

intervention program had a great enough effect to maintain language gains

for the children from that program. The early introduction to English

and the instructional language of the school evidenced positive contribu-

tions to achievement in language needs at this grade level.

The "wash-out" of gains in third grade was not surprising. Head

Start evaluations led to the recognition of a need for Follow Through

programs to promote continuous achievement gains for school-age children.

Without such a Follow Through program emphasis, five years would have been

a long time to expect a five-month intervention program to reflect

measurable effects in school achievement.

The fact that the analysis of the CTBS at both second and third grades

did not include scores for four comparison group subjects who were retained

in first grade (or assigned to a special education classroom, in one case)

and one program group subject retained in first grade, provides opportunity

for outcome speculation. One might infer that only the "better" of the

comparison students were included in the computation of means for those

tests; or, that had the four retained students taken these test batteries

at the time and their scores computed with the rest of the comparison group,

the differences in means might have been significant in favor of the prog-

ram group.

This study did not attempt to compare statistically the findings on

the various measures to national norms, nor did the study compare findings

to the district norms for that county's test scores. However, it was

obvious when studying the test data that both groups of children were in

most instances below national norms and district norms on the CTBS in

both second and third grade. Educationally, this does not bode well for






87

either group. Certainly, school systems and their educators are not

satisfied to have any group of students preforming below average on

achievement tests.

From the test records, the following information supported the obser-

vations just stated. The district grade equivalent mean for the second

grade CTBS battery was 1.6 at the time these subjects were assessed on

that test. For the program group four of the eighteen subjects were at

district level or above, seven were between 1.0 and 1.5 grade equivalent,

and seven were below 1.0. For the seven comparison subjects, none were

at district level grade equivalent mean, three were between 1.0 and 1.5

and four were below 1.0. At the third grade level and of the fourteen

program group subjects taking the CTBS, one was at the district grade

equivalent mean of 2.4, three were between 2.0 and 2.3, four were between

1.6 and 2.0 and six were below 1.6. The comparison group of seven showed

from that data that one subject was at district mean, three were between

2.0 and 2.3, two were between 1.6 and 2.0 and one was below 1.6.

When looking at the raw scores for the communication skills tests

for the Florida Statewide Assessment Test, the expected range for the

district was between the 34th percentile and the 60th percentile. Five

of the eleven program subjects for whom scores were recorded fell within

this range; six of the eight comparison group subjects were within the

expected range.

These results substantiate Carter's (1970, p. 17-18) writings:

"Mexican Americans as a group fail to achieve well on standard tests

of academic achievement; . the gradepoint average of the Mexican-

American group is generally lower than that of the Anglo group within

the same school or district; . Mexican-Americans achieve at a rate

substantially lower than national norms or their local Anglo counterparts









[in the area of language arts]; . [they do] relatively well in the

'fundamentals,' [of math] approaching local norms, but far below local

norms in areas dependent on language ability."

Much has been written about the low academic achievement and IQ

of Mexican-American children (Carter, 1970; Zirkel, 1972). In an article

in The Urban Review, Zirkel reviewed many of the studies and research

efforts that have looked at the results of standardized tests taken

by Mexican-American children in both English and in Spanish. He wrote

that as early as 1935, Manuel, then at the University of Texas, reported

that Mexican-American students demonstrated higher scores on the

Stanford-Binet Intelligence Test when given the Spanish version than

when given the English version. Zirkel (1972, p. 34) looked at the

Coleman Report, for example, and pointed out that the "scores of verbal

ability were consistently lower than nonverbal ability scores for

Mexican-American and Puerto Rican pupils in grades 1 to 12." He also

referred to Mahakian's study, reported ina1939 Elementary School Journal,

that showed 87% of a total number of over 200 Spanish speaking children

in elementary grades achieved higher scores on standardized reading

test when given that test in Spanish. And in referring to a study by

Davis and Bersonke, Zirkel (1972, p. 35) stated that although the

"differences between Spanish and English administrations of the MRT were

mostly nonsignificant for a group of Mexican-American first graders,

.the mean scores on the subtest that most appropriately reflected

their language background (Word Meaning) revealed a significant difference

favoring the Spanish version."

This discussion is perhaps more relevant to the present study than

might seem. It raised the question of how much difference is enough

for a program to have had effect if children are still below "norms"






89

and if tests are unable to show just what children might actually know

if the assessment procedures were in their favor. The question must

be answered in studies yet to be made whether or not children who speak

a home language different from the school language should be assessed

on achievement tests developed for and written in English.

It was also apparent from the data that there were other consider-

ations than just achievement test scores on which teachers made judgements

as to the relative success of each child at each grade level. This

was evident in the number of children from each group promoted or

retained at grade level by the end of third grade. One is tempted to

suggest that if different criteria were established for promotion other

than the scores on standardized tests given to students, then perhaps

the test results were of less importance in the school's general

appraisal of school performance and success.

Teachers in Hardee County had expressed concern for the number of

days migrant children were absent each year and equated this absence

factor with school achievement. It seems appropriate to consider

what some of the research has said about attendance and school perfor-

mance in this discussion. Jencks (1972, pp. 93-94), in his contribution

to the published papers from the Harvard University Faculty Seminar

on the Coleman Report, stated that "exposure to schooling showed no

appreciable relationship to achievement." He further stated that

"findings do not suggest that efforts to develop 'year around schooling'

or 'after school' programs are likely to boost achievement on tests of

the type used in EEOS [referring to the Coleman Report]."

Soar (1973, p. 111) reported that "more frequent absences

from school were associated with a decrease in motivation and a decrease






90

in creativity measures," but that absence measures failed to show a

relationship to school achievement.

The Stanford Research Institute (1974, pp. 255-256) reported that,

in analyzing Follow Through data for classroom observation evaluation

in 1972-1973, first and third grade children tended to be absent less

in classrooms where there was a high degree of "child independence,

child questioning, adults responding, individualized instruction, and

open-ended questioning . and where children and adults show more

positive affect [smiling and laughing)." The report also stated that

children were absent more from classrooms where they worked in large

groups more, where there was corrective feedback from teachers, and

where children were punished more frequently. The report on attendance

concluded with the observation that "although the data are correlational

and causal effects cannot be attributed to the instructional processes,

the correlations are high enough, and the sample large enough, to suggest

some directions for further research in absenteeism."

Whether or not attendance in school is significantly related to

school performance is subject for another study, not this one. However,

it seems to this investigator that, for children who come from education-

ally and socioeconomically disadvantaged home environments as do the

Mexican-American migrant children, and who have the added disadvantage

of a native home language that is different than the formal instructional

language of the school, consistent absence from school may preclude

these children from benefiting from the total educational program.

It is possible that absenteeism may well contribute to poorer school

performance in areas that the school system believes are important for

testing purposes. Again, it is recognized that this is speculation and

beyong the scope of this study. The facts are that many of the children









in this study were not in school for the full academic year; school

authorities were and are concerned with their number of absences and

nonenrollment weeks, and teachers believe that these absences contri-

bute negatively to school achievement.

There is also the question of whether or not preschool education

has a positive effect on school achievement for children. This has

been discussed and argued pro and con by professionals and lay persons.

Jencks' writings on the Coleman Report (1972, p. 92) questioned

seriously preschool contributions to later school success: "If the

nursery and kindergarten programs prevalent around 1960 had an effect

on achievement, it was extremely small [and that there] is not very

strong evidence for the theory that nursery school attendance boosts

achievement." However, this observation was made about nursery schools

for primarily middle class children. In 1960, few disadvantaged

children from poverty homes had nursery schools to attend. The programs

of the late 1960's and early 1970's grew out of the support for early

childhood education for less advantaged children with educational

emphasis on a curriculum to compensate for deficits in learning.

Today, there seems to be little doubt that "good" preschool programs

can contribute positively to children's learning.

In the Murray (1977, pp. 56-57) study referred to in Chapter II, the

consortium conducting longitudinal research on early childhood education

projects for disadvantaged children has found that "preschool effects

on IQ tend to last for at least three years." The results of the analysis

of the data from the twelve projects being studied indicated that "even

when some of the environmental variables most common associated with IQ

are controlled for, well run preschool programs using a great variety

of approaches produce significant, immediate improvements in IQ which

can last for several years."









In the data collection trips made to Hardee County by the investigator,

one thing was very evident. Teachers and administrators were quick to

give unsolicited praise to the intervention program. They reported

that the program group seemed more confident in their school relationships,

felt "good" about themselves, and did better in school than other children

from the same backgrounds. The preschool program had been a part of

the school system since its inception until the fall of the present

school year, 1976, when the responsibility for implementation was given

to the Spanish Mission in Zolfo Springs, Florida. Several members of

the school community expressed the hope that the program might come

under the auspices of the public schools again. There seemed to be,

from these kinds of statements, the belief that the preschool program

had value and merit for its participating children and that it would

be worth having within the confines of the school system for greater

continuity of educational programming.

The statistical findings and the over-all interpretation of these

results indicated a substantial trend towards positive contribution

to the longitudinal effects of the preschool intervention program on

school achievement. It was apparent that the program did make a difference

into second grade for the Mexican-American children who participated

in that pilot program and who were enrolled in Hardee County schools

at the time of this study. It was also evident that even though the

gains faded in third grade with the comparison group "catching up"

to the program group, there was a determination of over-all grade level

success made by teachers that led to fewer children from the program

group being retained by the end of the third grade than from the

comparison group. If these kinds of significant trends could be realized




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs