Post on 17-Jun-2020
ED 100 492
AUTHORTITLE
INSTITUTION
SPONS AGENCY
PUB DATENOT!
AVAILABLE FROM
FOPS ?ETC?')ESCRTPTORS
IDENTIFIERS
ABSTRACT
DOCUMENT RESUME
24 PS 007 573
Gordon, Ira J.The Florida Parent Education Early InterventionProjects: A Longitudinal Look.ERIC Clearinghouse on Early Childhood Education,Urbana, Ill.; Florida Univ., Gainesville. Coll. ofEducation.Children's Bureau (DREW) , Washington, D.C. Div. ofResearch and Evaluation.; National Inst. of Education(DREW), Washington, D.C.Jan 7536p.; Adapted from paper presented at theMerrill-Palmer Conference on Infant Education(Detroit, Michigan, February 1973)Publications Office /IPEC, College of Education,University of Illinois, 805 W. Pennsylvania Ave.,Urbana, Illinois 618(7,1 (Catalog No. 125, $1.60)
MF-$0.75 HC-$1.85 PLUS POSTAGEEmotional Development; Group Experience; *HomeVisits; Infants; *Intervention; *LongitudinalStudies; Mother Attitudes; Paraprofessional SchoolPersonnel; Parent Child Relationship; *Parenteducation; Preschool Children; *Program Evaluation;slf Concept; Sex Differences; Socialization;Standardized Tests*Parent Education Project; PEP
This article presents a longitudinal analysis of theFlorida Parent Education Programs which were designed to provideintervontion services to socioeconomically disadvantaged parents ofyoung children through home visits by paraprofessional personnel. Theprograms include the: (1) Parent Education Project (PEP); (2) EarlyChile Stimulation through Parent Education Project; (3) Home LearningCenter Project; (4) Instructional Strategies in Infant Stimulationnroject; and (5) Project Follow Through. Data were collected fromchildren, ages 1-6, who had entered the programs since 1966 and hadparticipated from one to three years. Skills and attitudes of themothers were also assessed. Descriptions of the projects' effects onthe rarticipants are given. The measurement design includedexoerimental and control groups, assessed by standardized tests,Interviews and Parent Educator Weekly Report Forms. The resultsindicate *hat the parent education approach has had lasting effects
children and on some aspects of family life. (SDH)
THE FLORIDA PARENT EDUCATION EARLY INTERVENTION PROJECTS:A LONGITUDINAL LOOK
Ira J. Cordon
University of Florida
Available from:Publications Office/IRECColle,;e of Education/University of Illinois603 West Pennsylvania AvenueUruana, Illinois 61801
January 1975
Price: Catalog ftrj
The projects reported herein were partially supported by grants fromtne Fund for the Advancement of Education, Children's Bureau (PHS-R-306),and tne National Institute of Mental Health, Department of Health, Education,ard Welfare (5-R0.1 -Mh 16037, NIH-MH-17347). This paper is adapted from onepresented at the Merrill-Palmer Conference on Infant Education in Detroit,February, 1973.
CO' material in this publication WIS prepared pursuant to a contract with:Iti,nal Institute of Education, U.S. Department of Health, Educations'eiefare, partially supported by a contract with the Office of Child
Devlopment. Contractors undertaking such projects under governmentsponorship are encouraged to express freely their judgement inprofessional and technical matters. Prior to publication, the manuscript14 submitted to the Area Committee for Early Childhood Education attie :ni7ersity nf Illinois for critical review and determination ofprofessional competence. Points of view or opinions, however, do notne,.;essarily represent of government position or policy, or theofficial views or opinions of the Area Committee.
Since 1966 we have conducted a series of intervention research efforts,
in wnich paraprofessionals serwd as nome visitor parent educators who
demonstrated specially designed home learning activities to the parent
(usually the mother) so that she, in turn, would engage in broadly-defined
instructional interaction with her child. Table 1 shows the chronology of
these projects along the top line, witn their spin-offs on the other lines.
Here we are concerned with only the top line.
The PEP project (Gordon, 1967) was a basic engineering effort to answer
practical questions as whether we could develop and install a delivery system
and develop a set of materials to deliver. Obviously, the existence of Table 1
indicates we were successful. In the PEP project, we had 150 experimental
families and two control groups of about 30 families each. In one control
group, graduate nurses visited the Families on a systematic basis, but
conducted no parent education (to explore the Hawthorne effect). The other
control group was the standard kind. Families were randomly assigned to treat-
ment and to those two control groups.
The first effort was followed by the Early Child Stimulation through
DareW, Education, Project (Gordon, 169c) which was a little more sophisticated,
a little m;re organized, and also a little more complicated. The original
exb,?rimental group was divided, and half the families were randomly assigned
to a new control group. Since we found no significant differences on the scores
of the control groups when the children were age one, we treated them as a
common pool and randomly assigned half to the experimental group in the second
year. Tnis gave us four groups (see Table 2 for design).
A group of new parent educators and one professional supervisor were
instructed to develop their own curriculum so we could explore the question
Research
Development
'readier
Education
Technical Assistance:
Training, Development
and Evaluation
Dissemination
TABLE 1
The Florida Parent Education Programs
Longitudinal Study 1966
P.E.P. ---4
E.C.S.P.E.P.
H.L.C. --,I.S.I.S.
(1966-7)
(1967-9)
(1968-70)
(1970-2) 1974
Social Roots (1972-3)
Reinforce (1972-3)
Follow Through, Florida Model 1968
--
Headstart 1969-72
Planned Variation
Teacher Corps 1970
--
TDDI, TDDS 1971-73
Alachua EPDA 1971-73
PCC Chattanooga 1969-71
F
Appalachian Reg. Comm. 1971-73
Anderson, S. C., 1972
Hplinc,s Co., Ohio, 1972
Publications:
Two Boo:ss, Ref,
ar:Ji Reports, Dissertations, Articles
Conferences:
Headstart, AERA, APA, SRCD, NAEYC, Merrill-Palmer
Film:
Playing for Keeps !ARC)
Video Tapes
Workshops:
Local, Regional, National
of whether Piaget-based, language-oriented curriculum was any better or
any worse than a curriculum put together by people who had a lot of experience
with infants, but not much theoretical background. A comparison of the two
curriculums test scores at age one) indict ed that it did not really make
any difference which one we used.
This investigation led to the Home Learning Center Project (Gordon, 1972).
We followed the cnildren tnrouf;n tne tilird year o'T life. However, we made one
significant cnange. Up to tnis point all of the intervention had been of the
home visit nature, on a once-a-Neck schonulo. We felt that a group experience
for 2-year-olds would be an important addition. The children were placed in what
we called, °home learning centers" or backyard centers, five children at a time,
for four hours a week in two, 2-hour periods. These centers were in homes
of mothers already in the project (urban homes in the Gainesville area, and
rural homes around the 12-county area). Some of the Gainesville homes were
loc,ted in needy-opened housing projects and in turnkey housing in the east
Gainesville section. "he mother who lived in a home center was employed as
ar aide to tne bacKyard center director. a parent educator converted into a
norre learning center director as well as a home visitor. Each parent educator
s`.111 carried 13 children, so she met groups in the center and continued to
rreet with the mothers on a once-a-week basis. We also added new 2-year-olds in
tne program so we could loo;', specifically at the effects of people coming in
at i'je 2 and flaying one year of the combined program versus thos" who had a
continuing program. We are still engaged in the longitudinal study of ti'ese
Tables 2 and 3 contain the treatment design And the basic measurement
tools
'able 3 shows a language measure for ch,ldren at 24 and 36 months and their
mother;. This measure was not part of the origihal project, but was the work
U1
caa
02
I
TABLE 2
Longitudinal Study - Treatment Design
Child's Age by Months
Group
3-12
12-24
24-36
48, 60, 72
1)
EHome Visit
HV
HJme Learning Center/HV
Test
2)
EE/C
HV
HV
Control
Test
3)
C/EE
Control
HV
HLC/HV
Test
4)
E/C/E
HV
Control
HLC/HV
Test
5)
E/CC
HV
Control
Control
Test
6)
C/E/C
Control
HV
Control
Test
7)
C/C/E
Control
Control
HLC/HV
Test
8)
CControl
Control
Control
Test
-5-
of Resnick (1972). It was a measure of the language in a free play situation
with the mother present during the 5-minute period before the child moved
into the actual ,estirg situation.
We were interested in how children hchaved in the Home Learning Center.
The Weld, shown in Table 3, is a situational event sampling pruedure of
child behavior in various hone learning center settings, such as free play,
one-to-one adult-child interaction, small roup instruction. The SEMS is
the Scott Effectiveness Motivation Scale used by Kronstadt (1973) as a
measure of achievemnt motivation.
Since the project vil'; a parent education project, we felt that it was
important to gather a variety of information about the mothers (not because
we weren't interested in fathers, but because in half the families there was
no father present consistently in the home, and therefore it was far easier
to measure the essential caretaker, the mother). The Social Reaction Inventory
is a measure of internal-external control of reinforcements based on the
'flit -r, toe Now I See Myself is a self-report scale of feelings of inter-
personal adequacy, home-school relationships, and feelings of competence.
We gathered a good deal of demograp6ic information about the size and
oomposi:.ion of tne family, the not.ier's education, her age, number of children,
housing conditions, and so forth. The longitudinal work is still in process.
The )1('reh are now reacnin 6 years ref age.
1,e,, on 'nose tnree drojects, there were a number of questions and
r.flr:orn tnat Arose over tne 'oars, so Dr. Jester and I designed the Instruc-
'frltri.e; in intan-:. Stic,uldtion project (Gordon ind Jester, 1972).
We did not know at tne time that tne acronym, ISIS, was the Egyptian goddess
of fertility, which was a sLarre because we were really after the other end of
the 1 ine!
TABLE 3
Longitudinal Study - Measurement Design
(Infant's Age in Months)
3
=.
12
24
36
48
60
72
Experimental
and Control
Experimental
Only
Experimental
and Control
Social
Reaction
Inventory
How I See
Myself
Griffith
Series
SRI
H1SM
Language
Demographic
Children
Bayley
Series
Language
TOB
Weld SEMS
Mothers
SRI
HISM
Language
Demos
Experimental
PEWR
PEWR
PEWR
Only
(Parent Educator Weekly Report on Visit)
Stanford-
S-B
Binet
TOB
TOB
Leiter
Leiter
PPVT
PPVT
Language
Interview
INT
Language
PEWR
S- B
TDB
S-B
TOB
Preschool
Preschool
Inventory
Inventory
()PVT
Home
Interview
-7-
With ISIS, we examined professiooal versus paraprofessional home visiting.
We had used only paraprofesionals, but other projects, notably DARCEE, Levenstein,
and Weikart, had used drofessionals. We were interested in the fact that in
tne earlier projects we seemed to be getting a bv-treatment effect: riirls
seemed to be benefiting more from the intervention than boys (Lally, 1969).
Vet some of our otner data seemed to indicate that maternal attitudes were more
influential in affecting boys' performance than in affecting girls' performance
(merman, 1971; Etheridge, 1972). We wanted to look more closely at the social-
ization process. We also wanted to see if it makes any difference whether you
work directly witn the cnild, or whether the focus is on the mother on the
assurption that she, in turn, will work with the child. The families in this
sample were divided into the various treatment groups necessary to look at
trese guestAns.
in tne :SiS project, every six weeks the home visit took place in an off-
rArP.J1 (lartrent. We video-taped the home visit beginning at the 3-month point,
"AO a ve array of raw data on tne 128 families showing the growth in
t-e ,lotrer, in -e and tne cnanges in behavior of the parent educator
(;ier tie. 7-nine ta:)es are being reanalyzed in our two current projects
")72; riordon, 1972c) .
A row,nw of 7arlei snows now a programmatic research effort can lead
av-listance, cnanges in University programs, and
0 Tmr!ral ;)uHic.
r.
Ir r.,,omp 4
"Al invo:p'd in Ylo 0110w Through project as a
,Ar r-odl!; (Inarerizel by the V,P of paraprofessionals
lit pri)gram, rut ako includes the work of paraprofessionals in
r.lalsroom and a much greater involverPnt of the parents themselves in the
BEST COPY AVAILABLE
decision-making process including curriculum development as well as other
aspects of the program. These activities are related to the reseach effort.
Another outgrowth of the basic research has been the development of teacher
education programs: the Teacher Corps Project and the Teacher Training in
Developing Institutions Programs in the Department of Childhood Education.
Both use tne parent education and involvement philosophy and model.
Analysis of the Program
The procedure for analysis was developed by the Association for Supervision
and Curriculum Development Comission on Instructional Theory (Gordon, 1968),
which leaned heavily on ideas from Robert Travers. The premise is that any
instructional program or cur6cu1um program can be analyzed in terms of the way_.
it handles the interaction among three major sets of variables: the pupil
:naricteristics (or the assumed pupil characteristics), program goals, and
tne instructional setting characteristics (Figure 1). in order for this scheme
to rrd,ce snse, nowever, we must go back a step furthen and seek the derivation
of the goals. nnat basic assumptions or postulates or hypotheses did we have?
/4 (4,PV911Ped a series of postulates and assumptions about the child and about
tre Anther. For each of tnese postulates and assumptions, it is possible to
wna*_ we assumed the pupil characteristics to be as children entered the
loo at whd t (J Irwr fdctors in the environment might be playing a
r,-)10 :n rf.suit or in cootriuuting to the operation. We have
"(1,;-ci'Jr.1,-)nic factors.
t s-,ible to state a goal and to describe
tr1,- -01,-; by which wn would attempt to get to that goal.
Anaijsis of r,nild (",',4ractr!ristics
There are several sets of basic assumptions. The first set is: (1) the
wnlld erter tne situation with some level of intellectual performance
PupilCharacter letka
Ins IructIonal SI NotionChemist laki
GooiChotoct.
4"
ASSUMPTIONS/POSTULATES
BLe,1!Li)31
vv p.1
Fig. 1 Adapted from the transactional network between pupil goal,
and instructional situation characteristics. From I. J. Gordon, ed.,
Criteria for Thc:)rics of Instruction (Associatioi. for Supervision and
Curriculum Development, 1968), p. 17.
:; 4) 0 1 `)
-1d-
(although we trid no way of measuring that level at the entrance age of 3 months),
and (2) intellectual performance is a function of experience and not a given.
Our instructional situation involved a series of sequenced tasks. However,
the activities were not sequenced in such a way that you had to follow task 4
after 3 after 2 after 1, they were rather sets of sequences. The choice of which
task to use at any particular time was up to the parent educator and the mother;
but, generally, choices were made in terms of expectation about the growth of
the child. Language input was built into the materials for the mother and in
the parent educator's instructions and demonstrations.
Tables 4 and 5 contain the results shown when the children reached age 5.
Pemember that Group 1, the first experimental group, which had been in the program
for three years, had been out of the program for two years at the time of
testing. Children in Group 2 had been out of the program for three years,
and children who were in for the first years only (Group 5) have been out of
the program for four years. As the children enter kindergarten are there any
lasting effects over a period from at least two years to four years after the
pronram? At age 5 the children in the experimental group were superior to
ILnose in the control group on the Stanford-Binet and on the Caldwell Preschool
Inventory, tne only two measures thlt we have so far used.
When the children were age 3 and 4, we factor analyzed the Stanford-Binet
(7a.Jie 6). Children who were in tne program for three years scored significantly
6.iver tnan the con trols across all tnree factors (Table 7, Gordon, 1971).(Age
4 Factors are snown on Tables 2 and 9.) Those children who were in the program
for triree years, th9 first two, and the first year only scored significantly
rhej r than the controls on all tnree factors (riordon, 1972). The differences
i! age 4 are clearer than Irip 3, and, interestingly, even a bit clearer at
age 5.
Jo It,
TABLE 6
Stanford-Binet Factors
Used in Group Comparisons
At Age Three
lits1COPY
411.111.4..
Factor 1 Language
,111.
S-D LevelDescription
11-6Identifying Objects by Use
11-6Picture Vocabulary
111-6Comparison of Balls
111-6Discrimination of Animal Pictures
111-6Response to Pictures
IVPictorial Identification
IVDiscrimination of Forms
Factor II Memory
S-B LevelDescription
11-6Obeying Simple Commands
IIIPicture Memories
111-6Sorting Buttons
IVNaming Objects From Memory
IVPictorial Identification
Factor III Perceptual Motor
S-B LevelDescription
Stringing BeadsBlocking: BridgeCopying a Circle.Comparison of 11a115
Patience: Pictures
Sorting Buttons.
") 0 1
Means
TA
BL
E 7
and Standard Deviations for Three
Standard-Einet Factors at Age 3 by
Number of Years and Timing of Participation
in the
Stimulation Program
ryeFactor
=11,1,
Language
Memory
PerccDt..:al-rotor
Croup
Years
IVS.D
.S.D
.S.D
.
Iall 3
27
3.30a
1.98
2.26b
1.48
3.00c
1.82
2first 2
17
2.64
1.97
2.12
1.69
2.82b
1.63
3last 2
83.00
1.55
2.50
1.31
4.12
1.25
41 and 3
11
2.91
1.70
1.73
1.27
2.82
1.17
S1 only
10
2.10
1.60
1.40
1.17
2.30
1.49
62 only
10
2.00
1.41
1.60
1.08
1.90
1.45
73 only
56
2.78
1.75
1.75
1.38
3.39e
1.41
8Cc:ntrols
51
2.33
1.81
1.61
1.23
2.98E1.64
a.
Hilt - -r than groups 5, 6,
and 8
p<.05
b.
th3r, L: ups S and 8
p<.05
C.
Eig.:-.er than group
6p
<.05
d.
Higher than groups 1, 2, 4,
3, 6,
e.
Higher than gro,;p
5 and 6
and 8
p<.05
p<.05
f.
Higher than group
6p
<.0S
-13-
TABLE 8
Stanford -Dinet Factors at Age 4
IIMINI
BEST COPY V,711.A31E
Factor 1 Cognitive Processes: Symbolic
s-n lxvo) 6 Item Lo:+lia. Description
(4) -I .53 Picture Vocabulary
(4)-3 .R2 Opposite analogies
(4)-4 .4A Pictorial identification
(4-6)-2 .S1 Comprehension II
(4-6)-4 .R4 Opposite analoLies
(4-6)-4 .57 Materials
(5)-3 .42 Definitions
(6)-6 .46 Maze tracing
Factor II Visual Discrimination
:;;-11 r, Item Lonil.n7
.SO
Descrition
(3-6) -2 Patience: Pictures
(3-6)-2 .55 Discrinination of animal pictures
(3-6) 5 .52 Sorting but t-ons
(1)-1 .43 Picture vocal.lulary
(4)-5 .65 Discrinination of, forms
(4)-2 .74 raming objects from memory
(4-6)-1 .73 Aesthetic conparison
(4-6)-3 .63 Pictorial similarities and differences
(4-6) 5 .48 Three commissions
(5) 5 .60 Pictorial similarities and differences II
Factor III Cognitive Processes: Iconic
S-11 rr,%.-1 6 'tell LoH:n^ Descri,tion
(c-6)-4ri
' 1I'
) )
I
1
1
Pr2v(7,11:c to pictures
err-Irchcnion 1
Pictorial identification
( 1 ) (.1 .43 Comprehrnsion 11
- ) 4 .44 Matcririls
(1-6)-5 .49 Three comsiens
(S)-1 .70 Picture completion; Man
(,)-3 .67 Definitions
(6)-6 .55 Haze tracing
IIM441.
1111444.416.
4) 1
TABLE 9
Means and Standard Deviations for Throe
Stanford-Binct Factors at Asc 4 by
Number of Years and Timing of Participation
in Program
03
Gr1:1:n
Years
Cognitive rroccsscs:
Symbolic
SD
1:11 3
24
3.67a,c,c,g
3.3S
2first 2
14
2.794
2.FG
3second 2
92.78
2.S2
1:Ind 3
11
1.36
1.69
S1 only
10
3.70a,d,f,h
3.47
2 only
IS
1.60
2.20
73 only
54
2.11
2.1(3
3control
1.50
Factor
Visual
Discrimination
so
Cognitivc Processes:
iconic
5.SS
e
b f
5.79
3.65
2.23
4.21b f
4.21
1,
3.27
2.53
6.44b c
1.SS
3.73
2.59
4.64
3.SS
1.66
5.q0b,f
3.23
4..sja,d,c
3.22
3.10
2.67
2.40
2.23
S.713c
2.37
3.72b
2.11
4.2q
2.66
2.SS
2.15
than
than
greup
group
F = 2.30, p4.01
cnc-tnilcd.
ono-tailed.
cnc-tilcd.
onc-tai lcd.
t1lan
f111::lier thin
rItiy.;Icr
hIlii;hcr than
F = 2.5S, p<.05
r = 2.10, 11<:.CS
group 6,6,
};rot;) 7,
7,
p<.01,
.01,
p p
<-6S,
c- c- toiled.
c7c-tailed,
enc-tailed.
-15-
The long range effects are evident; they are statistically significant.
Tne question is whether these effects are practically significant. They are
not magnificent gains of 24 IQ points; they are more in the neighborhood of 8
and 9 IQ points. On toe factors toey look a little better. There are eight
items listed in Table 8 under the Cognitive Processes: Symbolic factors. The
mean differences were as much as two items, which is a pretty worthwhile
difference. But it would be foolish to say that this difference now showing
up at ages toree, four, and five was the effect of a single activity or a set
of specific activities offered at age one. Instead, it may simply be that we
did something to encourage that mother to work with her children during the
intervening years, and this "something" is still paying off. When we look
at the maternal factors we will see a little more specifically what the
motners' actions were.
The second set of assumptions were related to time. The first of this
set was that the child's age at entry would affect the holding power. Our
assurtion was, the earlier the wetter . As indicated on Table 2, we are able
to examine each 1-year program and each 2-year program by the starting age of
cnild. Binet scores snow tnat those cnildren who were in the first year
only are significantly nigher than toose in the second year only and tend to
t,e nigher than those in the tnird year only. There may also be a recency
effect for The tnird year, since the scores of these children surpass those
of tr,e second year. Tnose children who were in the program for the first
yoar orly "ave Leer, rut nnw four ears and stfli score significantly higher
tfldr flc? cortrol ;. 'nosh crii;lren w,1-) were in for tne third year only, also still
scrp-0 r17,;ner t:rian tri 7ne,-e are no differences in scores for thP
consisf.ent 2-year qrouDs. They are nvp'er than the controls, but equal to each other.
7,1P ;cores on tne Caldwel; PrrInrini Inventory show that the first-year
oni, group scored higher t.nan tne controls, but the other single year
groups did not. t3oth of toe tv,o-consecutive-year groups scored higher than
the controls, but not differently from each other. The second-year-only
group has consistently scored low at ages 3, 4, and 5, although there was no
difference on the (lriffitls measure at entry point into the group.
I have an idea that there are at least two factors involved. First, we
knew pretty well what we wantec to do in the sequence of materials from 3
months to 12 months, but we were a lot more vanue about materials to use
from 12 months to 24 months. 4e did a lot of experimenting, on a trial and
error basis, all during the year. Second, I believe that you have something
goinn for you in the first year that you don't in the second. The child is
(rowing very rapidly in the first year of life, and I am perfectly willing to
take advantage of magic: (If the mothers assumed that this magic growth was
partly the result of wnat we were .!,cling, that served as a positive reinforcement
for the mothers.) However, growth slows down in the second year of life, and
tne cnildren become a good deal more mobile. They do not want to sit still
and pay attention in the same fasnion.
Of critical i7portance is the fact that we are able to demonstrate, four
/ears later, the effects of a ninimal interve ntion program in the first year
:t wi;1 De imnortant, however, to see if these effects hold up at
1,e 5 and what happens as the children move on into school.
7,e .)' t17e assumption concerned length of time the children were in
'1; It a crisp of "tne loncer the program, the better?" The
An1 scores shows that there is not any nice,
r. ,:hat eernes is that the combination
oar ne-, and cOn-,11,7/ of treatment.
Trw t i r l set of ass.,r1Dtions had to glo with sex differences. We have
eilminel oAr data in a variety of wa/s: within treatment Groups, across
" i) I (1
- 7-
treati,ent r3ups, and reiiardless of treatment grouts. The Stanford-Binet scores
slow no significant difi'erence at uck 5 by oroup, but there are some differences
on tre Pr'scnool Inventery for tnese children who were in the program for two
consecutive years. nerwise, tne pictu-e does not seem completely clear.
The TOO is a !iieasire of task-oriented behavior which Earl Schaefer derived
tne opservat.iul foin used du-ing Bayley testing. e have been
with our testing at eacn year. e find no sex differences. We find
ne differences on our rating, 5ased on observation in the Home Learning Center,
of effectiveness motivation (Kronstadt, 1973). Lally (1968) found that there
wis A sex-by-treatrent effect at age one; the significant differences between
experi-ental; and controls on the Griffiths seemed to tie due to the _girls. In
7).e-snick's (1372) study of the language of trio -year -olds in a free play situation,
here were no significant differences by sex in expressive lanquace; but
Di 31e 3, according to the 27 different ways in which he scored his tanes, he
fnd tnat -.,ensures favored ti expressive language of girls, and 10 of these
s!rr;ficant, includinh tne rurber of words used, the number of different
Norris used, and the mean length of remark.
,e helieve tnat tne sex differences are not so much quantitative (although
4'9r tney tne/ favor the girls), but qualitative, in terms of the inter-
r9iti;nsnips a-long variables. In ry view, this organizing concept, how the
gut togetner, is 5otn r.4-)re intriguing and more useful than score-
e ') r.!:ItO o aff ct. Although we have used
'r"Y; .c1; and intellectual performance measurements
*P ;ricfith;, ;Ay , and were Also concerned with affective
1eiel7nmpr!t of ton yii Id. Tnere are', a; you well know, a number of severe
pronlems in this area. triore no satisfactory way to measure self-concept
-18-
in the 3-month-old, 1-year-old, 2-year-old, and even the 3-year-old. llhat
,vr haft to do was use some otIer reasures and theoretically link them as reflect-
oig aspects .Jf self-conce;It. je used the (1E1S, an event-sampling procedure
still not fu;lv analyzed (Neld), and the Task ')riented Behavior (lob) scores.
At age 3, the SE'tS scores are elated to Stanford-3inet, although the trend
is for this to be -lore true for girls. 3oys' SPIS scores are significantly
related to TOB, girls' are nit. SES scores are related to Resnick's major
child language variables: vocalizations, number of words used, number of
different words used, mean length of remark, number of nouns used and number
of verbs used (Kronstadt, 1973). Earl Schaeferii takes the position that
expressive language is not only an intellectual variable. He has found a
number of relationsnips between it and other ways in which he measures affect.
:t Is interesting gnat tne effectiveness flotivation, as measured by observing
penavior of youngsters in tne backyard enter, was significantly related to
tne expressive language output of children in the 5-minute free play situation
set up by ?esnir'K.
703 nas been nirp.ly related to tne intellectual test scores of children
3, 1, and 5. If Tf); is a measure of self-concept, then we can see the
r971',:On;r1i .,ere Petween t.rlis aspect of affect and cognition. Further, Resnick
j7?) *.qlt the exprey;i4e language measures of children at age 2 were
;tAnfori-31r et -,cores of children age 3. We have a com-
t-e 'I -)f t-e -Ancet -self-concept," but we are still
'r;
"aterri-
'Jr" .)r)1! .
/--tr1or7Y11
no';An ;-1:iraYor, bro-; im because, in 1956, people were making
-,r.nAefer, Der;onai corriunication, December 13, 1972.
an assumotion thot one of tne r:asons that children do not do well in School
Docau'',e soretning is wrong in tne hove OaDeled the deficit model). So our
first .issumbtion was that the lower Class nother, as defined by income, does
not ne....ss?rily se erself as a teacher of ner child. She lacks effective
,lotivational and instructional techniques. Long range gains depend on changing
tne none so tilt the motner sees nerself as a teacher of her child, and possesses
sill in teacnind ner child. ;ie assumed (and let me stress the word "assumed")
t-it tie ,-otner entered tie pogrvi with a lack of skill in teaching the child
ir: lice of orientation toward self as teacher. Since our early programs we
nO ti3Y of a'isPssind "the skill of the mother," this was only an assumption,
an, qe proceeded on our way using tnis assumption. In 1973 we would explore this
w- d; area in TAite a diffrent fashion, but our goal at that time was to
. 3Dilit/ of tne plrent to teach her child specific activities. The
lo;trJctional situation cnaracteristic was therefore to send somebody into the
-oe to i7Ow tne r-otner now to do a specific activity, encourage her to try it
w.tn tne chi i1, and to follow ud 4itn weekly home visits, introducing a continuous
;et oc Tne riore visitors would talk with the mother about how well
*.ne cirTh anr', involye ner in the operation.
;0? P'.14f) been able to assess -iaternal teaching skill through use of the
ii.!eotape; in tne pro;ect '(;orlon and Jester, 1972). We used a variation
-0 ee,]i.r-),:al -it.egorie,. scale wnic- we.s based on interaction process
in! "11-):4°"! *_fl.! 0:7 i9r41YOnS of parent educator, mother, and infant.
1 -eg ca -_e.1 r/ 9-V! tnree seconds and these behaviors
lrl: Z.? Di tne pattern of interaction between
age l, ;t14 rj' !qnntal Development Index scores
nd J!rr! ;(?r- very clearly some teacning patterns that
jrd °trier,. -egatiyely, to tne child performance at age 1.
a
n was a .mttern:" I do somethini, you do
; do soetnind, yon, lo :t is a very rapid transact.iona1
tno o!ner hind, staino,: behavior, talking to the chiid
wItqpit nirn to resr:ond or necossarily paying any attention to his
re,Jorses, (typical profe;sorial behavior) nas a negative correlation to
fjszinatin(i anoat tneSe two findings is how closely
t!'e. !le in Altr :oar's work in Foilow Througn (Soar, 1972) and with so much
of t'e classroorl ooservdtion researLn. others within this popu-
ation vary conilerably in now roach of this :)attorn they use. Therefore, it
wte possible noy that 'we nave, if vie wanted to start over again, a way
ei,,, enterini behavior of :.Narents that we did not have ba6 in 1966
Irtrorc!, indirate that one of the basic assumptions
0 "tiny tn1t of r1 horodeneous deficient population,
:t not tr,e. Sr) e of the rotners were extreely skillful in "ping-pow)"
t°e e,1 of tne pror;ect, and nad probably been extremely skillful
were ever Involved with tie-. :',ut other Withers were unskillful at
esiecially tnose in the control group. Jestee and Guinagh (1972),
,,.; i; rnid:!rl, and ford qrea: range in the ability of parents to
r;!il O -eir youn, I.' assumption that mothers lack skill
11
'.Or'.iOn of the :;o:141ationl but for that portion, I think
irrfI ' ,' rior.
to 1;fct cni Hren lern from their
at. liyinq in a nor-e
1,; zines, etc., miTit
There were indeedivl!les lr° rot m;)ortwt,
1 111era r f n 0; 4. n rAteriak aere lacking. In our
dro;A mo mett.eril;') tnoor of were needed by building some
1701,F 10
ncs Per,: i%f: AP. Cvitrr,
Catel;,IVIC
P I
21/1i.
2A/I5
27/15
15/21
15/27
801 COPY AMILARE
for Pefinini: icachin4
--1wItrIr:!io.17:
Pchvior Pencription
rev,vidc, rotiler initiates
Pmher elicits, bal,y rer.ponds
rther initiates, baby responds
P-Iter directs, baby revonds
11,,v responds, rothcr elicits
resi,onds, rpthor directs..AffPctiv,_. inter77tir-1
11-17/21,22,21 baby b.-..!.aos, ilothcr warms, accepts,anplifics
F.1Y,117-1...N... MINN
15/13 Baby responds, baby arplifics
1 :. J. (or.:. dri; .10;!, :ri .,t.-.7.:ztic!:al !'.1-r-l^c,:ie; In Infanttr.---i: ...,:.on. OO,'.), t.. -. , :-;:....rnt,. Pr'VZ:r11107:/# 1972,2, P. 77.
of our none tiviics around tnese and introducing them is- a meaningful fashion into
:fit! .e -.1rovIleJ and Li:e cadazines and other printed materials. We
doviAed e,koerionces oy teacninq rothrs to use simple objects and
tie nouie for i..;aiking ifoailos, dolls, and games. Mothers were
1-terv1ewed when tneir children were a(le 1 try a team consisting of an anthropologist
nArce, noitner of ww.);; .;ere heavily involved in the project.
''o' r; in tie e(:,-.?rimental grow) indicated were significantly more involved in
witn tieir cnildren, in prinu materials into the home, and in buying
r',) rn educitIonal tor, than were the control group mothers. On the other :sand,
t-e tester 4n0 asKed tne '.1others sore questions while she was doing the testing,
eou,d tnat snout 43 0; *.nt? control 7others (unidentified to her) indicated that
"ii ;earn-?:: of tne value of soe of these thir.is simply by watching the
7)enailor in tne testing situation. The question has been raised about the
e"e':ts fv: testing is an intervention. If you recall our child effects, testing
!n;Affi:ient, necause ooth tne experi:iental and the control groups had continued
Drolri effects overshadow testing effects.
7-. tnird anal irry Pasic set of assumptions has to do with language. There is
1,:ur,sion in the current literature on the deficit versus difference approach
t) ;Age. 'serertneless, our 1966 assumption was that the child structure of
m linwstic leveloDTent, is determined by the linguistic pattern in
to inter:)retations of Basil Bernstein (1961), a child expoc'ed
;
;. ;-,att,r,1 in tne no-le would nave more difficulty developing and
7 gi-,r,;_-). tne'iretical area is controversial; Bernstein him-
Linguists now feel that all
1;0; -.(-)rirfn tnat any lanquaie can e:=press 6)5tract
';(4 4r1,411 tare 4-! Asitior yn.it now a language is used in the home r-IJ,es
!rerrorflov, dI f f ,rnnci. in tne rie40100r4Int. Of a cnild. WP assumed that in a "lower ''lass'
' I;
3-
home tne frequency of adult interaction with the child would be less than in
"favored' nomes, and tnat the kind or language used with the child might be
-ore of tne ordering, forbidding, and comoanding type than the reasoning,
questioning type of language. Therefore, we tried to build the use of verbs
and adjectives and total sentences and questions into our materials.
7esr,ic',( f),^A tnat several of his variables were significantly related to
tne Stanford-3inet score at age 3. Two major variables were: (1) the number
-)f l'"erent orris tnat riotners used, and (2) the mothers' interrogatory
ntences to tneir 2-year-old children in a 5-minute free play time. The
:orrelations between tnese two maternal variables at child age 2 and i.he
:itanora-iiinet at child age 5 is approximately .50. Considering the 3-vear-
;1:, end tnat tne language sample had been very small (5 minutes), the finding
sJiborts tne notion of the strong relationship between home use of language and
:mild intellectual performance. Again, we find wide variability in the group.
'e;nick's ,!ata and our follow-up study indicate a clear relationship between
-dtprri; ;anguie Denavior and child performance within this population.
lan uage assumption is that the important time in language devel-
tne receve language period, j) to 2 years of age. The pattern of
0' 71^'Hiaqe at that tire, before the child is doing much talking
rv Dui
cri'1:1. 4e issu-ed tnat in "lower class" homes there was a low
.),-1/cr;ltion directed at tne child. This does not mean that the
r ;ang'41e, but ratner, tnat sne does not have a verbal interaction
`ne/ nO
rild -a/ ;Je -,urrounded with a lot of people talking,
n.g7AJer of homes, coil iren are getting
'Ar;,Ajo cr,-;fl teeilsion and radio, but it is mostly background noise. It
); not be,ng ;pecifical;y at tnese children, nor does it require any
.1,11 of rar,On-,P from tnpm. -)ince we were concerned about increasing the frequency
-24-
of verbal interaction, we stressed the importance of language in the home
visit, providing a variety of words in relation to each activity.
Our data come primarily from the Parent Educator Weekly Reports (PEWR)
completed by the parent educators after each home visit. Included was a very
short checklist abo.,' the presence of certain verbal activities, such as: did
the mother use words with the child? Did she speak directly to the child,
face to face? This checklist of 13 items is very primitive. There were no
frequency counts (verbal interaction simply occurred or did not occur), and
no allowance was made for the length of time parent educators were in the home.
evertheless, Jester and 3ailey (1969), who studied the first experimental group,
Herran 0)71) and Etheridge (1971), who analyzed language as part of larger
investigations of maternal effects on the performance of 2-year-olds, all found
significant low out reliable correlations, (in the 30's) between the language
beh /ior as gathered on that primitive scale and the child's performance on the
Gr)ffith's measure at age 1, and on the Bayley measure at age 2.
The reciprocal categories tapes were examined to see if the program resulted in
increased amount of interaction between mother and infant and the amount of
instr.J,tional interaction. There is a rising line from 3 months to 12 mcnths in
botn the amount of general mother-infant interaction and mother-infant instructional
interaction. This is an indication that we achieved the goal of increasing
the frequen,:y of verbal interaction. Further, we demonstrated a positive rela-
tionship between adult language and child performance (Gordon and Jester, 1972).
The fourth naternal set of variables was affect. The literatul'e in the
early suggested that disadvantaged mothers saw themselves as inadequate,
that tney had lower self-esteem than middle class parents, thlt they had less
feeling of control over the environment and more fer :Hng of being victims of
fate and circumstances. We thought that a program such as ours mig!Cc change
that picture. Our assumption was that when the mother realized that what she
Wii doing witn tne cnild was paying off, she would feel more adequate as a
teacher and mother. 4e also assumed that she had low self-esteem when she
entered the program and our goal was to raise her self-concept. Because of the
tine it took to develop the scales, we were not able to use them on the original
1266 population. We used them only on the second cohort involved in the cur-
riculum comparison study (Gordo o, 1969c). We found that the Now I See Myself
scale dii not snow and significant change from the mother's score when the
child WAS 12 months of age. Since then we have refactored that instrument on
tne Follow Through population but we have not yet reanalyzed these data. The
Follow Through data suggest that this program does lead to the improvement of
self-esteem as measured by the Now I See Myself scale (Greenwood et al., 1972).
Bilker (1970) modified the language of the Rotter I-E Scale with the
nelp of the parent educators. We labeled it the Social Reaction Inventory. We
felt tnat tne mothers entering the program would have a high belief in external
control of reinforcement. Our hypothesis was that a; a mother saw that what
S^c! WAS doing with her child was really having an effect, she might begin to feel
; more control over wat happened to herself as well as to her child.
the 'Arne group of ITiothers was studied on the Now I See Myself scale, mothers
-0 c!r! to a more internal view. 4hen we compared our mothers with other samples,
tner were -.lore external than at least a comparative national high school sample
is,rg tnp standard I-E scale, and were more external than parent educators
e-porel in tne Floridl p;Irent education projects in the 11 communities in
whi,:n 4c? 4r-)ror Piront; in t- Florida Follow Through program have moved toward
1 mon int(!rnil new (1-Ireenwood et al., 1972).
otn of the schedules are self-reports; frt.(?), are weak measures of what
we are after. One can question their validity. We could say that if the
" 0
-26-
parents have moved toward more sense of control and more self-esteem this
snould show up in what they do. We have studied the housing patterns, birth
rates, and the involvement of the children in other programs after leaving
this program. The experimentals nave improved their housing Patterns by
moving to better housing as it opened in the Gainesville area, and by moving
less often than previously. Inc ay.:.rane experimental mother had about three
children when sne started the program and the average control mother had
tnree-plus children. The experimental mothers have given birth to fewer
children, even after leaving the project (Gordon, 1971, 1972). The program
supplied no information. When parents were interviewed more of the experimental
nothers reported that they had tried to put their children into other kinds of
cnild development programs than did the control. These are side effects, but
tnink worthwhile side effects. 1 believe they relate to control over the
environment.
The fifth maternal assn, ption was that the mother's different expectations
for Days and girls would influence ner behavior and child performance. We were
"at only interested in tne sex differences in child performance, but also in
41/. trlt .otners miqnt relate differ.mtly to boy and girl babies. We have
-any Iniings on tnis particular dirension. We assumed that there would be
»cialization patterns and different verbal behavior and different
itttples toward toys and toward girls. Two dissertations, Herman (1971) and
H37l), examined tne relationship between maternal variables and
C children on tne 3ayley scales at age 2. They found that (1)
atYtide index and amount of positive verbal behavior on the PEWR
liff.rentilted rnotnerl of nign from low scoring boys and girls on MDI and PDI
'Herman), (2) ')R.1 and HI'IM factor scales differentiated mothers of high and
low boys (Herman), (3) extent of talk on PEWR differentiated high and low boys
" 0 .2 41
-27-
and girls on F131 and POI (Etheridge), (4) HISM factors score differentiated
high and low buys on 1DI (Etheridge), and (5) HIS1 factor scores differen-
tiated high and low boys on TO[ (Etheridge). In general, maternal self-
attitudes, maternal verbal oehavior, and maternal attitudes toward the project
seemed to be more influential in determining (within the experimental group
of boys) those boys tnat scored high and those that scored low. There is an
effect for the girls, but the predominant effect is for the boys. We have
the interesting situation tnat treatment effects seem to be more related to
the girls, maternal personality and behavior within the experimental group
seems to be more influential on boys.
The ISIS project was designed to shed further light on the above
prohlem. In tne ISIS project we compared professional with paraprofessional
teaching of boy and girl babies Poth when babies were taught directly and
when rnotners were taught and later taught their babies. We found fascinating
sex-by-treatment differences. For example, girl babies taught (either directly
or through tne mother) oy professionals, averaged 12 points higher on the
3ayley than did the bo1 rabies they taught. No such differences appear for
piriprofessionals, and tne overall test score averages are equal for the
two ,arent educator groups. The sa,i* pattern held on our other measures.
;:f,v,i,,,itIon of tr)e intractIon data indicate tnat professionals seemed to
ittehl more to the r,otriers, particularly the mothers of girls: paraprofessionals
-,or-ed to attend more to th,7? baby. Pecan that the poor (negatively related
teaulinh ;)attern wls sustained adult talk. The professionals
u;e-: :)-ittPrh 4itn fr,_!;u,nc/ when the children were from 3 to 12
mortn1 of Aq0 (especially with tne no/s), fAistaine6 baby activity was a nattern
relate'"; to POI. Again, professionals seemed to encourage less of this when
they taught the child directly, than did the paraprofessionals; and again, the
boys taught oy professionals snow up at the bottom (Gordon and Jester, 1972).
-2d-
ate do not know why this picture emerged, or whether it is re;Aicable. There
is still a great deal to be learned, and ISIS raised more questions than
answers about early sociolization as a function of sex.
We had several demographic assumptions. We found a significant but low
correlation between awount of motner's education and Stanford-Binet scores at
a4e 5. There were sore indications that these homes were crowded, but no clear
picture of effects. We found no solid support for the general assumption that
marital status affects a child's test performance.
The Griffiths measure and our series items have been factor analyzed
(Maurelli, 1969, 1971), aim we have begun studies across age on the various
tests. We nave, as I mentioned earlier, considerable data yet to be analyzed,
Particularly of the type in which early information is related to later
performance.
Implications
At tne silolest level , we have demonstrated the efficacy of one approach
to parents of young cnildren which seems to have a lasting effect not only
on tne cnildren but also on some aspects of family life. We have also demon-
strated that such a program should out confine itself to only the traditional
e.(erimental-control design, but that much which needs to be learned comes
from a program of r,,ultivariate studies of factors vithin the experimental
;roar. ''.ei-cyjri:mic and personal , 'oonicn influence the course of events.
1.r.1 -,rnw It twin ;o -Cal led disadvantaged group is by no means homo-
Irl the nol r.tr i; tic? :'j(111? Cils; group with whom we've worked
ir Tnrouln and °tr?'" spin-off programs. Programs, tnerefore, must
move awl/ from oversimolified and erroneous concepts built around such terms
.0tner deficit or difference models. We need to move toward better subject-
by-treatment designs. This will require the development of effective measures
-29-
of parents' entry skills. Here, I believe we have made a contribution
through the ISIS project.
As we examine our procedures, we know with much confidence that the
basic model works. It can be done and it can be transnlanted. We have
also learned that the model is acciotable to parents across social class,
income and ethnic lines. In spite of earlier criticisms by some naive socio-
logists, the program, both in the infant and Follow Through projects, is
perceived by parents from all groups as helpful and desirable, as strength-
ening tne role of the family and supporting parents' desires fnr their
children.
Improvements in tne delivery system need to be made, in keeping with what
we hake learned apout parent educator-mother-baby interaction. Examination of
the videotapes, for example, shows that a major problem is convincing parent
educators to be more open and flexible and less ordering and autocratic. It
is a - latter of training. '4ith a videotape system we can engage in a more
careful prescription of the training operation than we were able to do earlier.
4e would want to capitalize on encouraging more of the ping-pong behavior that
lid originally. 1e have developed a preliminary list of what we call
desirable parenting behaviors wnich we did not have before. There is a task
orientation problem wnich occurs when the parent educator demonstrates an
activit: with the child. The Parent often focuses on the activity and forgets
the child, failing to attend to the cues the chili is giving and urging the
-1111 to r;et tne activity dory ?. would now suggest: let the child stay
with an activity witnout interruption, if possible; attend more to the comfort
of the Chl:d than to the activity; respond to what the baby is doing; play
and talk with him. if the cnild is doing something, do not interrupt; let the
activity you want flow from tne child's behavior. These are different suggestions
than we could nave stated in 1966.
-3J-
We have also developed a preliminary list of parent educator "do's" and
"don'ts." Parent educators interrupt mothers, null the attention of the child
from the mother, and do not play ping-pong with the mother. There is a good
deal more precision we could introduce in training.
Another area which needs improvement is the measurement of the affective
domain- -both for adults and children. Our results show that even with limited
measurement tools there is a relationship between maternal affective factors
and child cognitive development; there is a relationship between cognition
and affect in the children; that the program has an impact beyond intellectual
Performance scores. However, we need far more knowledge, which requires
better reasurement, so that we can help parents in the affective domain to
view themselves better and to relate to the child in mentally healthy ways
which enhance his ',ense of v:lf-esteem.
Although we have made progress, we have a way to go yet. Continued
Progress will require the cooperative work of many longitudinal studies, and,
of course, hard cash. I believe we have shown that such money is well spent,
in terms of our goals for sound and effective family life.
'' 4+ I'll
References
Bernstein, B. Social class and linguistic development: A theory of sociallearning. In A. Halsey, 3. Floud, and C. Anderson (Eds.), Education,economy_and society, New York: The Free Press of Glencoe, 1961, 288-
Bilker, L. Loc:_s of J-E.control evectancLand exPectanci changes inJisadvantaledlio_tie_rs.--DOCCOraloTssertatTon,517ersity ofMrida,
Bradsaaw, C.E. between paternal_ and infant performancein environmentally disadvantaged homes. DoctordrdissertaCon, University
Etneri7(.1e, F. Tnefelltions)i,) of marital status of the mother and sex ofthe infant to naternai
u
varicOle scores and infant performance. Doctoral_
lissertation, niversity or lorida,
Gordon, I.J. 1,_parent education approach to provision of early stimulationcon tne cUltur-qy disadvantaged. rThal report to Che Filia7157-TheAdvancement of Education established by the Ford Foundation, 1967.?iTeographed.
(Ed.). Criteria for theories of instruction. 1ashington,D.C.:Association for Supervision and Curriculum Development, NEA, 1968.
(Ed.). Reaching the chld through parent education: TheFlorida_a2proach. Gaines0-11477)a.: flhi-titute frOTTeVelopment ofihman Resources, 1969a. (Available through ERIC-ECE, Urbana, Ill.4ED-057-330).
Florida parent education projects: A schematic representa-tion. Gainesville,
Institute for Develovent of Human Resources, 1969b.
7.1rj; Chill station through parent education. Final reportP,yre,p1, DeDartrent of Heal-th,Taucation and' Welfare.
'ii7eor.apnel. F11.: Institute for Development of HunanPesources, 1)69c. (Available through ERIC -ECE, Urbana, 0E0-038-166).
7Jerelod;H ;went :)o-vr?r. In E. Grotberq (Ed.), Critical issuesin researr, related to disadvantal0 children.. Princeton, N.J.: Educa-tional Testing Service, September 1969d, Seminar 05, 1-24.
stimulation via parent education. Children, March/April-1169e7 6(2) , 57-59.
1970Baby iearniliatririluyi baby play. New York: St. Mertin's Press,
"
iucItioReaching the young chi 1d tnrough parent education. Childhood
February 1)70 b, 46(5), 247-249.
A noire lean center a2proach to early spriulation. A----progress report on Gr4ritill-MH-16-0P-6.1, 'Gainesville, Fla.:
Institute for Development of Hunan Resources, July 1971.
An instructional theory approach to the analysis of selectedearly childhood programs. In I.J. Gordon (Ed.) Early childhoodeducation, seventy-first yearbook of the National Society for the Studyof Education. Chicago: University of Chicago Press, 1972 a, 203-228.
A nome_llarning center as roach to early stimulation. A;1T-ogress report on Gran: 45-r.,1-!Iii-16/T17-1-4,
Institute for Development of Human Resources, June 1972 b.
Inc) social roots of conpetence. National Institute for MentalHeirtn project statement. Mimeographed. Gainesville, Fla.: Institutetor Development of Human Resources, 1972 c.
Gordon, I. J., Guinagh, B., Jester, R. E., & Associates. Child learningthrough child_plat. New York: St. Martin's Press, 1972.
Gordon, :.J., Jester, R. E. Instructional stratenies in infant stimulatio%JSAS, Catalog of_ Selected Documents in Psychology, 1972, 2, 122.
Gordon, I. j., & Lally, J. R. Intellectual stimulation for infants andtpldlers. (lairesvi.le, Fla.: Institute for Development ofTuman?e;ourcei, 1 )67.
rireenool, 76;!Snnr...? to local Follow Through Programs, Final='ep(:.rt, Der.e,ber 1972. r1linesville, Fla.: Institute for Developmentof 4-lan )e7io,Jrcir-, ,and Tiorida Fducational Research and DevelopmentMA^cii, 1)72. !.*.iiry:raln(id. 53 Pp. [Our, aPoendix.
r:rf-dn -aternal variable scores and infant'-!rclr"linr: In 3 %ooro stirailation_populatior.i7/To-ctoral
'.:nimrSlty of Florida, 1971.
Jester, a. E. Tne effect of r,:binforcerent infant performance. lational:,11HtLiti? for !lental th project statement. ,'Mimeographed.
11.: i rr, t l t.,A to for Development of Human Resources, 1972.
Jeste-, ) . . ,1,..',--..:,e-?-.., scores on the nriffithS lental/-'";, 0.. 1; A-, A ..rCi)(-1 ' 1drv:uarl usage in the hore. In I.J.
1 CW '. :,;;Oc:1,11, t:-I'fi; th ifilil throuq'l parent education: The
" '..- '' jte for Development-WC--, , '-i,j7 n ., e, / ; 0- : i: : .fl : : Tns-tIt
-,4-4n )e;f7), :J65, 21-31.
.;ester, R. E . , rJuiraqn, B. 1 rihw Hrents read to children. Thn17_11!.1Practice, J6ne 1972, 11( ), 171-177.
Kron:*- t. D. The relatIonstip of motivat, iiono 4chiPvi and intellectual, and 1 enavi oraLfunc_ti onina Chreei.
vivantaj oi c ni 11.ren. iissertat:,,n, University of
Lally, t. P. A s'./ ;cores of tra i nod and untrained twelve month oldenvironmental iv to )riv+'i infants on th" "ilri ffi ths .".ental DeveTontient,,(11.!". Doctoral dissertation, iiniversi ty of -Florida,
,
laure'I., ). A for Inale.ic sud/ of a series of intellectual stimulationtr,is, for 1 nint tod,:lers. In J. 'Inrdon ! Associates, Reachingtfle thriJIn educarlon: The Florida ciporoacii. Aainec01 e,F:a.: InstitJte fOr )evelOX:ent of Hliman Resources, 1909, V.44,
!IAA re 1. 1 1.1 ./ Y. Lid v of t,ie r; rif :!en ta 1 Development:11;-,ert.ation, Jniver;ity of Florida, 19/I
:4esn , . "). an(4, n",:ellect_Jua 1 and b"iavioral functioning; van ta-octoral dissertai ton,
)7
Soar, 4, analysis of selected Follow-Through.)le of a .)rocess a-.)proach to evalJation. In I. J.,.)r ;n E)(11..:/cni1,:noo: education, seventy-first yearbook of the
q7/ or tne '.4tudy or nuCation. Chicago: University ofCnicaqo Press, 1972, 229-259.
L. 3!iiiora: correlates of intellectual erformance among disad-vanta-..ed three /ear old Negro cni dren. on,-F5i7sityof riorida, unfinisned.