Development Toward School Readiness: A Holistic Model

alan kibbe gaynor, boston university

ABSTRACT research. What is needed—although difficult to do—is bivariate

experimental research that would provide the effect sizes that are
This article describes a systemic analysis of the early childhood
needed for more precise systemic analysis.
development factors that explain the variance in readiness for
school among representative 5-year-olds in the United States. The
model expresses a theory that incorporates a broad set of causally
interactive endogenous variables that are hypothesized to be driven
by the effects of three exogenous variables: parental education, INTRODUCTION
immigrant status and racial/ethnic identity, and single-parent/
divorced/remarried vs. stable-marriage family status. While exog- School readiness is a complex concept that, overall, relates to a
child’s readiness at age 5 to learn in a school environment. Julia
enous is generally defined as “outside,” for the purposes of this anal-
Isaacs (2012) defines school readiness and the relative disadvantage
ysis, racial/ethnic identity is exogenous in that, in contrast to the
of poor children in this regard in the following terms:
endogenous variables in the model, it is not influenced—and does
not change—in interaction with other variables in the model. Poor children start school at a disadvantage. Their health,
The model was run in computer simulation mode. The results behaviors, and skills make them less prepared for kinder-
are compatible with what is known about school readiness pat- garten than children growing up under better economic
terns. While this finding does not establish validity per se, it does conditions. Fewer than half (48 percent) of poor children
suggest that the model provides a reasonable multivariate, systemic are school ready at age five, under a summary measure that
description of factors that determine readiness for school at the encompasses early math and reading skills, learning-related
age of 5 and that offer a reasonable explanation for the variance in and problem behaviors, and overall physical health. Children
school readiness among 5-year-olds. born to parents with moderate or higher incomes are much
Finally, the model was run in experimental computer simula- more likely to enter school ready to learn; three-fourths (75
tion mode to evaluate the likely effects of five interventions that percent) of these children are ready for school at age five.
relate to: (1) cognitive and academic factors, (2) health care and In other words, there is a 27-percentage point gap in school
nutrition, (3) income, (4) reducing the effect of low income on readiness between poor children and those from moderate
family stress, and (5) a combination of all these types. All the inter- or higher income families. (p. 2)
ventions were simulated by modifying the structure of the model to In a typical school, children who enter with low readiness do pro-
moderate the effects of low parental education and low income on gressively worse in the course of their schooling in comparison
other key variables in the system of early childhood development. to students who enter with average readiness—and far worse in
As expected, combining the interventions related to cognitive comparison to students who enter school with high levels of read-
and academic factors, health and nutrition, and family stress with iness. The major conclusion of my earlier work on the dynamics of
straightforward increases in low family income had a very sub- typical vs. good schooling (Gaynor, 2011/2012a) was that strong
stantial effect on the relative age of school readiness of children school leadership was critical to improving key system variables
of parents with very low educational attainment (from a relative such as teacher quality, the rigor of the curriculum for all students,
age of 2.9 years to a relative age of 4.3 years)—with very modest the professional development of teachers, and special programs
improvements in school readiness for children whose parents did to increase the academic ability and achievement of low-achieving
not graduate from college. It is acknowledged that such a set of students. Other important variables strongly influenced by school
comprehensive interventions would be very costly and, probably, leadership are careful teacher supervision, building and maintain-
politically infeasible. The more limited and less costly interven- ing community support for the school, and improving and main-
tions—e.g., cognitive and academic—vary in their likely effec- taining family support for student aspirations, self-expectations,
tiveness; the least effective interventions are those that affect only and academic skills. Leadership at the school district level was also
health and nutrition. found to be critically important.
The findings of this systemic analysis are generally consistent However, a key finding was that while these changes in school-
with the literature, at the same time emphasizing the weaknesses in ing were effective in bringing students with initially low readiness
the current knowledge base, which is largely based on correlational closer to national norms (established by scores on various high-stakes

27
standardized tests in language arts, math, and science), they were moderate-or high-income families on a range of particular readi-
not effective in closing the gap between these students and aver- ness measures—cognitive and behavioral.
age- and high-readiness students in the “good” school itself. This The connection between income and school readiness can be
was because in good schools the achievement of initially average- seen as even more compelling when placed in the context of the
and high-readiness students also improves, thus maintaining the rising proportion of poor children in school:
gap between these students and initially low-achieving students.
For the first time in at least 50 years, a majority of U.S. pub-
Although the computer simulation modeling effort that is
lic school students come from low-income families, accord-
reported in this article produced important insights into the
ing to a new analysis of 2013 federal data, a statistic that has
dynamics of schooling and their effects on the academic achieve-
profound implications for the nation. (Layton, 2015)
ment of students with different levels of school readiness, I asked
myself the same question many others have asked themselves: Why These data led me to build a computer simulation model
do 5-year-olds arrive in school with differential levels of readiness? informed by the relevant literature that would analyze the dynam-
We know that they do. ics of early childhood development. The goal was to determine
As noted earlier, poverty has been found by many researchers the implications for intervention policies that would help to
to be a significant critical causal factor. Isaacs (2012) compared lower the variance in the distribution of school readiness among
the difference between children from households with an annual 5-year-olds by increasing the school readiness of children raised
income below 100% of the poverty line ($18,000 for a family of in low-income families whose parents had attained relatively low
three in 2011 terms) and those from households with an income levels of education.
above 185% of the poverty line, a group that spans a wide spectrum
from just above 185% of poverty ($33,000 for a family of three in DEVELOPMENT OF THE MODEL
2011 terms) to much higher levels. Children who are described as
Although others have pursued this goal, it seemed that a computer
“near poor” (household income between 100% and 185% of pov-
simulation modeling approach would provide a more holistic view
erty) are also disadvantaged as they begin kindergarten, although
of early childhood development than the approaches taken by the
to a lesser extent than poor children. In this group, the 59% with
vast majority of other educational researchers. As described in
incomes just above the poverty line are ready for school at age 5.
the reports cited in this article, most of the research on the root
For children born into households with income above $100,000,
causes of low school readiness (and on its long-term effects) is cor-
school readiness increases to 86%. Recent research (Frey, 2015)
also reports that there is a significant correlation between family relational in design and focuses on individual factors. In contrast,
income and early childhood brain development, a likely further the work that is reported in this article and in my previous work
factor in explaining variances in school readiness. on school dynamics (Gaynor, 2011/2012a; 2011/2012b; 2012a;
The relationship between family income and school readiness 2012b) posits a set of causal interactions that seek to explain the cor-
is made even sharper in Figure 1. Students from poor families relations reported in the literature.
are seen to be highly more likely to score below students from The Importance of School Readiness
As noted by the authors of a themed issue of The Future of Children
Figure 1. Likelihood of Scoring Very Low (Failing to Be School- (Rouse, Brooks-Gunn, & McLanahan, 2005) that addressed the
Ready) on Measures of School Readiness by Poverty Status problem of school readiness about a decade ago, school readiness
is of great importance. The report notes that children who enter
Percentage Scoring Very Low (Failing to Be School Ready)

60 school not yet ready to learn, whether the cause is an academic,

Poor Near-Poor Moderate or High Income
50
52
social, and/or emotional deficit, will continue to have difficulties.
40
41 With regard to school success, those who score low on tests of
30
30 cognitive skills in preschool years are likely to do less well in ele-
26 25
19
21 23
20 20
mentary and high school than their higher-performing peers. Later
20
13
15
13 in their lives, they are more likely to become teen parents, engage
10 7 7
4 2 1
in criminal activities, and suffer from depression. Ultimately, they
0
Math Skills Reading Skills Learning-related Externalizing Physical Health Any of These attain less education and are more likely to be unemployed.
Behaviors Behaviors Measures
Confirming and extending the long-term effects of school read-
Source and Notes: Brookings tabulations of data from the Early Childhood Longitu- iness, Karoly, Kilburn, and Cannon (2005) reported,
dinal Study – Birth Cohort (ECLS-B). Very low is defined as more than one standard
deviation below average on the academic and behavioral measures and in poor/ Many children from disadvantaged backgrounds fail to meet
fair health on the physical health measure. grade-level expectations on core subjects. For example,
“Starting school at a disadvantage: The school readiness of poor children,” by Julia national educational assessments at grades 8 and 12 show that
B. Isaacs, 2012, p. 2, Copyright 2012 by the Brookings Institution. Permission to about 50 percent of children from at-risk backgrounds (e.g.,
reprint.
low parental education or low family income) score below

28 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
 the “basic” level of reading and math achievement, indicat- defined as those with a poverty rate exceeding 20 percent)
ing that they have less than partial mastery of the knowledge provide more limited opportunities . . . in terms of social
and skills “fundamental for proficient work” at that grade interaction, positive role models, and other resources, such
level. Other manifestations of problems in school achieve- as quality child care, health facilities, parks, and playgrounds,
ment for disadvantaged children include higher rates of spe- that are important for healthy child development. (para. 4)
cial education placement, grade repetition, and dropping
Additional benefits of access to health care are child development
out of school. Ultimately, limited skills and low educational
screening and advice to parents regarding behaviors that promote
attainment increase the likelihood of undesirable outcomes
children’s social, emotional, cognitive, and physical development.
in adulthood. Low educational attainment is associated with
Many of the model variables can also be found in the following
reduced rates of employment and with lower earnings for
graph published in a report by Isaacs (2012, p. 17):
those who are employed. Use of social welfare programs is
also higher among those with low educational attainment, as Figure 2. Overall School Readiness by Selected Child
are crime and incarceration rates. (p. 4) and Family Characteristics
Doyle, Finnegan, and McNamara (2010) also confirm and add Male
Female
Non-Hispanic White
to these findings in a review of studies that identified six lifelong Non-Hispanic Black
Hispanic
effects of school readiness across the following domains of devel- Other race
Normal birth weight
Low birth weight
opment: academic achievement (Heckman, 2006), peer relation- Maternal age: 15–19
Maternal age: 20–24
Maternal age: 25–29
ships (Ladd, 1999), psychological well-being (Alloy et al., 1999), Immigrant mother
Maternal age: 30+

teenage pregnancy (Brooks-Gunn et al., 2003), employment and Native-born mother

Married mother
Cohabiting mother
earnings (Raver, 2003), and criminal activity (Le et al., 2006). Single mother
Head Start
Preschool
No preschool
Maternal ed: <HS
Critical Factors in School Readiness Maternal ed: HS diploma
Maternal ed: some college
MA/MS
In the introduction to the previously cited report, Rouse and col- <100% FPL
Maternal ed: BA

100%–184% FPL
leagues (2005) summarize the factors that were identified as cru- HH income: ≤$25,000
185% FPL or more

cial to child readiness for school as follows: HH income: $25,001–$50,000

$50,001–$100,000
$100,001+

In essence, the message of this issue is similar: taken together, 45 50 55 60 65 70 75 80 85 90

family socioeconomic status, parenting, child health, maternal Percentage of Children Who Are School Ready (on All 5 Measures)

health and behaviors, and preschool attendance likely account “Starting school at a disadvantage: The school readiness of poor children,” by Julia B.
for most of the racial and ethnic gaps in school readiness. (p. 11) Isaacs, 2012, p. 4, Copyright 2012 by the Brookings Institution. Permission to reprint.

The Head Start Child Development and Early Learning Framework

(United States Department of Health and Human Services, Admin- ANALYSIS
istration for Children and Families, Office of Head Start, 2015) The purpose of the analysis is to depict the causal dynamics that are
includes the following major categories of readiness for school: hypothesized to explain the differences in school readiness that, as
Approaches to Learning; Social and Emotional Development; described in the Introduction, are problematic for academic achieve-
Language and Literacy; Cognition; and Perceptual, Motor, and ment, peer relationships, psychological well-being, teenage pregnancy,
Physical Development (p. 1). The model includes all but the first employment and earnings, and criminal activity (Doyle et al., 2010).
category (using sometimes different terminology). As noted ear-
The first step was to visualize the network of interacting vari-
lier, it also includes the effects of immigrant status, racial identity,
ables in what is called a “causal-loop” or a “causal-influence” dia-
and parental marital status and educational attainment, the latter
gram (see Figure 3).
known to have a strong influence on earnings and unemployment
The central thesis of this perspective on the dynamics of early
(United States Department of Labor, Bureau of Labor Statistics,
childhood development is that, unless there are effective interven-
2014) and, therefore, on such additional factors as neighborhood,
tions, these dynamics are driven in each generation by three exog-
family stress, nutrition, and health.
enous variables. These three variables, all of which are identified
With regard to the effects of home and neighborhood, the vari-
in the literature, are (1) parental education, (2) immigrant status
ables in the model mirror many of those identified in studies reported
and minority identity (White/Asian vs. Black/Hispanic/Other),
in the literature. For example, Karoly and colleagues (2005) stated,
and (3) single-parent (vs. stable-marriage) family status. However,
Although most children experience a supportive home and Isaacs (2012) describes an exception to the general pattern for
neighborhood environment with access to sufficient financial children with college-educated mothers. “[P]oor children whose
and nonfinancial resources to support healthy development, mothers have a college degree or higher are as well-prepared for
many other children do not. . . . Research has demonstrated school as other children of college-educated mothers (but this
that neighborhoods of concentrated poverty (typically small group represents only 2 percent of all poor children” (p. 4).

DEVELOPMENT TOWARD SCHOOL READINESS 29

 Figure 3. Early Childhood Development Causal-Loop Diagram

NEXT
GENERATION +
(+) + CHILD'S +
+ +
READINESS
FOR SCHOOL
+ Child's
Resilience
Single Parent
Family Status + + + +
+
+
Child's Self-

Child's Level of
(+) Reading

+
Language Child's Degree

+
Development of Positive
Behavior Child's Level of Social

+ + - + Development

+
Child's Level of
+ +
+
Conceptual
Development
(+)
Parent
Educational
+
Attainment
+
Family
Immigrant &

+ +
Minority Status Amount & Child's
Physical
+
Quality of
Health
+
Preschooling
(+)
Mother's Level of
+ - (+) +
Maternal Attachment

+ + Child's Level of
+ Emotional Security
+
-+
Family Availability of
Income Health Care
+ +
Family
+ - (+)
+
Nutrition

- + + Variety of
Food Available
+
Mean Income

+ of
Neighborhood Child's Level of

+
Maternal Attachment
Neighborhood
Unemployment,
Single Parent Crime & Drugs
Family Status

- Family Instability
& Stress

A report of the Social Genome Project, Center on Children Figure 4. Intergenerational Feedback Loop
and Families, Brookings Institution (Isaacs & Magnuson, 2011)
explains this finding when identifying a list of factors very similar +
Child’s School Experiences Child’s Life
those in the model as influencing school readiness. Controlling for and Development Achievements
parents’ level of education, marital status, and mother’s age at the
time of the child’s birth, as well as race/ethnicity, immigrant sta-
+ (+) +
tus, gender, and age in months, the report concludes: Child’s Readiness Child as Parent—
for School Effect on the
Next Generation
Parents’ education is a large factor explaining why chil- +
dren from moderate and high-income families enter school
Child’s Early Childhood
with higher reading and math skills—their parents are bet- Experiences and
Development
+
ter educated. Children’s early academic skills are higher,
on average, when parents have more years of schooling, +
and this association persists even after controlling for par- Original Parent’s
Education and
ents’ inherent abilities, according to evidence from wel- Income
fare reform evaluations and sophisticated statistical analysis
(Gennetian, Magnuson & Morris, 2008; Carneiro et al., The next step in the analysis was to transform the causal-loop
2007). In addition, the “education” effect also reflects diagram into a System Dynamics computer simulation model (a
underlying differences in parents’ skills and preferences, “stock-and-flow” diagram) in visual format (see Figure 5) and
which are often passed on to their children, by both inher- mathematized (see Appendix A for a list of the equations that com-
ited traits and upbringing. (p. 6) prise the computer simulation model).
It is further proposed (Amato, 2005), although not pursued in this The great analytic value of a computer simulation model is
analysis, that the effects are intergenerational (see Figure 4). twofold. First, the model can be analyzed as a description of “the

30 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
 Figure 5. The Computer Simulation Model

Emotional
Security Child's Academic
~ Language Development
Effect of Emotional Security
on Readiness for School

~
Effect of the Level of Aspir & Self
Expect on School Readiness
~ Child's Level or Readiness
for School
Effect of Maternal Attachment ~ Maternal
~ on Child Behavior Effect of Resilience on Attachment
Child's Level of
Effect on Child Behavior Readiness for School
Social Development
of Family Instab & Stress

Child's Physical ~
Rate of Increase Health Effects of Physical Healthon
or Decrease in Conceptual Development Child's Resilience
Social Development
Child's Level of
~ Conceptual Development
Effect of Parent Education on
the Child's Social Development ~
Effect of the Child's PreSchooling
on Social & Conceptual Develop Child's Aspiration and
Parent Rate of Increase in Self Expectation
Educational Attainment Conceptual Development

Child's Self Reading

~
Effect of Self Reading on Child's Age
Lang & Voc Development Child's Academic Quality and Intensity of Child's
Language Development Formal and Informal PreSchooling

Increase in
Child's Age Child's Social Child's Level of ~
~ Development Divided Social Development
Rate of Increase in Academic by Age Effect of Parent Education on
Effect of Level of Standard English Child's Aspiration and Self Expect
Vocabulary and Language Development
& Acad Cult on Voc & Lang Dev

~ ~ ~
Effect of the Neighborhood Effect of Standard English & Effect of Maternal Attachment on
on Vocabulary and Language Level of Standard
~ English & Acad Cult Academic Culture in the Home Aspiration & Self Expectation
Development
Effect of Parent Education in the Home Parent
on PreSchooling Educational Attainment

~
Effect of Immigr & Min Status on ~
Effect of Parent Education on Child
Engl & Acad Culture in the Home Effect of Parent Education Child's Degree of
Vocab & Language Development
~ on Conceptual Development Positive Behavior ~
Effect of Parent Education on Effect of Maternal Attachment
English & Acad Cult in the Home on Child Behavior

Parent
Educational Attainment
~
Family Immigrant Effect of Child Behavior Goal of Maternal
~ and Minority Status ~ on Maternal Attachment Attachment
Effect of Family Income on Effect on Child Behavior Maternal
on Quality of PreSchooling of Family Instab & Stress Attachment

~
Effect of Family Instability &
Stress on Materna Attachment Flow 1
~
Mean Income of
Income Based Family Income Affordable Neighborhood
on Education ~ Natural Maternal
~ Neighborhood Gang Attachment
Neighborhood Rate Crime & DrugTrafficking
~
of Unemployment Effect of Crime & Drug
Effect of Single Parent Family Instability
Status on Family Income Trafficking on Emotional Security and Stress

~
~ Effect of Immigrant or Effect of Neighborhood Income Child's Physical Emotional
Effect of PEA on Minority Status on Income Level on Variety of Food Health Security
Use of Health Care Effect of Single Parent
Single Parent
Effect of Nutrition Status on Maternal Attachment
Family Status Family Nutrition
on Physical Health
~ ~
Family Immigrant Availability of Health Care Effect of Family Instability
and Minority Status & Stress on Emotional
Security
~
Effect of Family Instability &
Stress on Physical Health Single Parent
~ Family Status
Effect of Availability on Degree
of Utilization of Health Care Effect of Single Parent
Effect of Health Care Satus on Family Stress
Utilization on Physical Health Family Instability
and Stress

Degree of Utilization
of Health Care Effect of Income
on Family Stress Family Income

way things are,” and while based generally on understandings from As noted earlier, the numbers are only illustrative because prior
the literature, the model constitutes a “theory of the problem” and empirical research has not produced specific effect sizes among
allows one to generate or construct interventions, some of which the variables that collectively, and interactively, generate differ-
have been tried, and for which there are, therefore, some effects ent levels of school readiness.
data. Secondly, and very importantly, the structure of the model
can be altered “experimentally” and “run” to simulate different A THEORY OF EARLY CHILDHOOD DEVELOPMENT
interventions without the risks and costs of empirical experimen- THAT GENERATES SIGNIFICANT DIFFERENCES IN
tation. Both procedures were used, and the remainder of this arti- SCHOOL READINESS
cle is rooted in these two related but different kinds of analyses.
Figure 5 presents the model. The report of the Center on Children and Families (Figure 1)
provides empirical data to indicate a 27% gap in school readiness
The Basic Model: Illustrative Simulation Results between children of poor and moderate- or high-income families
Having formulated a theoretical computer simulation model of (48% vs. 75%).The theory presented in this analysis is that this dis-
the current “system” of early childhood development that pro- crepancy is the result of interactions among a set of variables that
duces a range of students’ readiness for school, it was then pos- are caused by the effects of exogenous differences in parental edu-
sible to run the model to see the extent to which it generated cation, ethnic and racial identity, and being a child of a stable mar-
results consistent with what is known about the distribution of riage vs. being a child of a single parent (unmarried or divorced
school readiness among students with different backgrounds. It is mother) or of a second marriage. These interactions are shown in
important to note, however, that the numbers from the computer a causal-loop diagram (Figure 4), a stock-and-flow diagram (Fig-
simulation are illustrative only. That is, they show the effects, ure 5), and a set of equations (see Appendix A) that facilitate the
when run through the theoretical system of variables which com- simulation of the basic theoretical model. An appropriate subset of
prises the model, that reflect the idea that different backgrounds these equations can later be modified to test structural changes in
tend, in general, to develop different levels of school readiness. the model that represent various policy initiatives.

DEVELOPMENT TOWARD SCHOOL READINESS 31

 As can be seen by examining these various model formats, the MODELING SOFT VARIABLES
theory represented by the model projects the same variables for:
Ideally, science is about variables that can be measured as well as about
(1) children of parents with higher and lower educational levels,
empirically evidenced effect sizes. However, as is well known, this is
(2) Black, Hispanic, and non-Asian immigrant children, (3) White
not the world we live in. Many researchers do experimental work on
and Asian children, and (4) children of single parents and parents in
specific variables and the effects that so-called “independent” variables
a stable marriage. However, it is the influence of the fundamental have on these so-called “dependent” variables. But these experiments
differences in the effects of these driver variables working through are not generally holistic in nature. In addition to or instead of exper-
the various interacting endogenous variables in the model that, on imental work, many of us are committed to theorizing holistically
average, produce the differences in school readiness among chil- based on wide-ranging historical research in our fields, as evident in
dren at age 5. my earlier work on various achievements attained by students in dif-
The theory posits a set of secondary variables that include: ferent schools and in the work reported here.
income and employment, neighborhood, family stress, maternal In the course of this work, I have dealt with variables such as
attachment, health, nutrition, aspiration and self-expectation, age, family income, test scores, number of out-of-school suspen-
perseverance, language and vocabulary development, concept sions, and neighborhood crime rate that are measurable, one of the
development, and social development, among others. It is the dif- characteristics of so-called “hard” variables, but I have also had to
ferences in the values of these secondary variables in interaction include more variables—such as the child’s motivation and level
with one another that ultimately produce the differences among of perseverance, family stress, teacher quality, strength of school
students in school readiness as reported by the Center on Children leadership, and community support for the school—for which
and Families (see Figures 1, 2, and 3). only estimated values are available.
The same is true of effect sizes. As noted, most of the research
RUNNING THE BASIC MODEL reported in the literature is correlational, not experimental. Cor-
relations are, by their nature, relational, not causal. When thinking
The model, which is displayed in the formats noted, represents
systemically, there are several major problems with correlational
this “theory of the problem” of the uneven distribution of readi-
research in contrast to experimental research. One of these prob-
ness among 5-year-olds entering school. It is important to observe
lems is that in correlational findings, the Beta weights are depen-
that the causal effects in the model, with rare exceptions (e.g., the dent upon the inclusion or exclusion of variables in the regression
effect of education on income, which is projected in precise num- analysis, and in a lack of co-variability among these variables. As
bers from census data; see Figure 6) are projected as general in a consequence, the results can change with the set of variables
nature—e.g., high (3), low (1), or medium (2)—or as fractional included by the researcher.
values of 1 by using general rather than exact effect sizes. This is Another problem is implicit in this first problem: correlational
because my reading of the literature suggests that, as noted, the findings are basically systemic in nature. That is, a regression anal-
research is essentially correlational, not experimental, in nature, ysis does not provide direct information about, for example, the
and thus exact effect sizes are generally unknown. effect of Variable A on Variable B. Beta weights are not effect sizes;
they do not show the causal effects of changes in one variable on
Figure 6. United States Department of Labor, Bureau of changes in another. This essential point is consistent, for example,
Labor Statistics (2013). Earnings and Unemployment Rates
with a similar statement by Cass Sunstein and Reid Hastie (2015):
by Educational Attainment
“Unfortunately, it is always difficult to define the true causal factor
Unemployment rate in 2013 (%) Median weekly earnings in 2013 (%)
in correlations of this type [high correlation between Factor C and
2.2 Doctoral degree 1,623
performance on problem-solving tasks]” (p. 209).
2.3 Professional degree 1,714
Also, regression analysis tests for the linear relationships
3.4 Master’s degree 1,329

4.0 Bachelor’s degree 1,108

between the independent variables and the dependent variable:
5.4 Associate’s degree 777
Some college,
Regression analysis also has an assumption of linearity.
7.0 727

7.5
no degree
High school diploma 651
Linearity means that there is a straight-line relationship
11.0 Less than a
high school diploma 472
between the IVs and the DV. This assumption is important
All workers: 6.1% All workers: $827
because regression analysis only tests for a linear relation-
Note: Data are for persons age 25 and over. Earnings are for full-time wage and salary workers.
Source: Current Population Survey, U.S. Bureau of Labor Statistics, U.S. Department of Labor
ship between the IVs and the DV. Any nonlinear relation-
Data Table
These education categories reflect only the highest level of education attained. They do not take into account completion of training
ship between the IV and DV is ignored. You can test for
programs in the form of apprenticeships and other on-the-job training, which may also influence earnings and unemployment rates. For more
information on training, see http://www.bls.gov/emp/ep_table_education_summary.htm and http://www.bls.gov/emp/ep_table_education_by_train.htm. linearity between an IV and the DV by looking at a bivariate
BLS has some data on the employment status of the civilian noninstitutional population 25 years and over by educational attainment, sex, race, and Hispanic origin online.
The Census Bureau also has some data on education attainment online. scatterplot (i.e., a graph with the IV on one axis and the
“America’s youth at 25: School enrollment, number of jobs held and labor market DV on the other). If the two variables are linearly related,
activity: Results from a longitudinal study,” a report of the United States Depart- the scatterplot will be oval. (Princeton University Library,
ment of Labor, Bureau of Labor Statistics (2013).
Data and Statistical Services, 2007, para. 13)

32 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
Thus, correlational analysis, on which most educational research 6 (parent doctorate or equivalent). When parental education is
related to the factors that influence school readiness is based, held constant, racial and ethnic effects are not as strong. Given the
stands in contrast to the systemic analysis that is represented in this data that show that children of a White or Asian stable marriage
article. The theory described here explains the interactions among have attained at age 5, the school readiness that is known to affect
a set of individual causal relationships that collectively account for their subsequent achievement, for the purposes of this analysis,
the variability in school readiness among a representative sample White or Asian stable marriage status was considered to be the
of 5-year-olds. “standard.” Then the gain or loss in readiness (shown as the child’s
Unfortunately, as noted earlier, inter-variable causal effect sizes relative “readiness age”) varies under different family conditions.
are generally unknown. Thus, in this model, effect sizes are pos- Children of non-White or non-Asian parents with minimal educa-
ited only in highly generalized forms as high, medium, or low or tion in stable marriages had a readiness age just 1.04% below their
as hypothesized fractional values between 0 and 1. Thus, the vast White or Asian counterparts while children of single or remar-
preponderance of the effect sizes in my models are—of necessity— ried non-White or non-Asian parents who had graduated from
estimates, not known experimental effect sizes, estimated effect high school had a readiness age 24% below their White or Asian
sizes that are shown either as fractions of one or, more frequently, counterparts from stable marriages. In fact, it appears that given
as non-linear table values. the relationships posited in this model, other than parental educa-
This kind of theoretical imprecision has been a common obser- tion, the largest negative exogenous factor is single parenthood or
vation throughout the history of science. Siddhartha Mukherjee, divorce and remarriage.
the author of The Emperor of All Maladies: A Biography of Cancer With regard to parent education, it is important to keep in
(2010), writes that, on the one hand, “Science begins with count- mind that this variable is laden with the effects of race and ethnic-
ing. To understand a phenomenon, a scientist must first describe it; ity. Therefore, effects that are attributed in the model (in which
to describe it objectively, he must first measure it” (p. 19). But a educational attainment is an independent exogenous variable)
later quote, this one attributed to Howard Skipper, a noted cancer incorporate some of the effects of race and ethnicity. In a graph
researcher, presents the view that, “A model is a lie that helps you published by the United States Department of Labor, Bureau of
see the truth” (p. 491). Labor Statistics (2013), the percentage of 25–29 year olds who
My sense is that holistic, interactive, computer simulation mod- completed a bachelor’s degree or higher between 1990 and 2013
els, involving many “soft variables” and “soft relationships,” with all varied between 43% and 60% for Asians/Pacific Islanders, between
their problems of imprecision, help to provide a holistic sense of 25% and 40% for Whites, between 13% and 22% for Blacks, and
interaction, to identify areas where further research is needed, and between 9% and 15% for Hispanics (p. 3). Clearly, this makes the
to inform policymakers about the complexity of the problems they numbers in Table 1 more consistent with these data.
are attempting to resolve. With regard to marital status, data were gathered on a nationally
representative sample of men and women who were ages 14 to 22
Effects of Initial Conditions on Children’s Readiness when they were first interviewed in 1979 (United States Depart-
for School ment of Labor, Bureau of Labor Statistics, 2013). Respondents
The following table shows, in general, the ultimate effects in the were interviewed annually until 1994 and since then have contin-
model of parental education, minority identity and immigrant status, ued to be interviewed on a biennial basis. The detailed information
and stable or non-stable marriage on a child’s readiness for school: on fertility, marital transitions, and employment is collected in a
format that allows the dating of the specific events.
Table 1. Effects of Minority Identity and Immigrant Status on
School Readiness Marriage and divorce patterns differ according to race and
ethnicity. Whites are about twice as likely as Blacks and His-
Remarriage

Remarriage
Other and

Other and
Education

panics to have earned a bachelor’s degree. The difference in

Parent or

Parent or
Marriage

Marriage
White or

White or
Parental

Asian &

Asian &
Stable

Stable
Single

Single

divorce rates between college graduates and those with less

education translates into different divorce rates for Blacks
10 2.88 (Standard) 2.84 (–1.4%) 2.85 (–1.04%) 2.84 (–1.4%)
and Hispanics compared to Whites. (p. 4)
12 4.28 (Standard) 3.34 (–22.0%) 3.42 (–20.1%) 3.23 (–24.5%)
14 4.62 (Standard) 3.93 (–14.9%) 4.53 (–1.9%) 3.68 (–20.3%) The report continues,
16 5.56 (Standard) 4.59 (–17.4%) 5.13 (–7.7%) 4.47 (–19.6%)
[T]he divorce rate for first marriages is nearly 20 percent-
18 6.04 (Standard) 5.86 (–3.0%) 5.91 (–2.2%) 5.16 (–15.0%)
age points lower for those who have completed their bache-
20 6.48 (Standard) 6.29 (–2.9%) 6.22 (–4.0%) 6.02 (–7.1%)
lor’s degree compared with those who have completed high
Regardless of other family conditions, the effects of parental edu- school, regardless of whether they have some college or not.
cation are very strong in the model. Depending on parental edu- The gap is even greater, approaching 30 percentage points,
cation, the average child’s readiness for school at age 5 varies from when comparing those with a college degree to those with
a relative age of under 3 (10th-grade parental education) to over less than a high school diploma. (p. 6)

DEVELOPMENT TOWARD SCHOOL READINESS 33

It is understandable that the opportunity for parents to be well on nutrition and health care. This involved making the effect on
educated has a strong effect on a child’s school readiness (see Table the availability of health care the same up to a mean neighborhood
1). In addition to other characteristics that distinguish college income of $48,000 a year and making the effect on the variety of
graduates from those with less education, parental educational food the same up to a mean neighborhood income of $72,000 a
attainment is affected by race and ethnicity and is consistent with year, a figure chosen arbitrarily to essentially eliminate the effect of
the theory of intergenerational effects illustrated in Figure 4. The neighborhood and income on the variety of food available.
point is that parental education in the present generation is an out- To simulate the third proposed intervention, changes were
come of the effects of early childhood development in the prior made to reduce the effect of low parental educational attainment
generation, which is why reducing school readiness gaps and sub- on income, raising minimum family income from $27,000 to
sequent academic achievement and educational attainment gaps is $50,000 a year regardless of parental educational attainment.
vitally important and a matter of social justice. To simulate the fourth proposed intervention, changes were
made to make the effect of family income on family stress the
EXPERIMENTAL RESULTS same up to $90,000 a year, a figure chosen arbitrarily to essen-
Structural changes were made in the model to mimic the effects of tially eliminate the effect of family income on family stress. The
a set of experimental interventions, each designed to accomplish fifth experiment looked at the effects of combining all the first
one of the policy goals to attain equality of opportunity for 5-year- four interventions. The effects of these simulated interventions are
olds as they enter school. shown in Table 2.
As expected, combining cognitive and academic interven-
1. Decreasing the negative effects of low parental education on the
tions, health and nutrition interventions, and family stress inter-
quality of the child’s early childhood education and vocabulary
ventions with straightforward increases in low family income had
and concept development,
a very significant effect on the relative age of school readiness of
2. Diminishing the effects of poverty and low parental education
children of parents with very low educational attainment (from
on child nutrition and health care,
3. Diminishing the effects of parental education on low family a relative age of 2.9 years to a relative age of 4.3 years)—with
income, very modest improvements in school readiness for children with
4. Diminishing the effects of low income on family stress, and parents who enrolled in but did not graduate from college. Such
5. Multiple interventions (1–4 combined). a set of comprehensive interventions would require substantial
time, effort, and money and, therefore, they probably have lim-
Changes in the Model ited political feasibility.
To simulate the first proposed intervention, changes were made However, theoretically in the model, the more limited, and less
to reduce the effect of low parental education on each of a set of costly, interventions vary in their likely effectiveness. Those deal-
cognitive effects, including preschooling. This involved making ing with income and family stress seem theoretically to be poten-
the effects of parent education on each of these child develop- tially the most effective, with cognitive and academic interventions
ment variables the same for parent education up to Grade 15. These following. The least effective interventions are those that are
effects included: child vocabulary and language development, directed solely at health and nutrition. It is important to note again
English and home academic culture, concept development, and that the interventions tested have focused on the lowest parental
preschooling. educational achievement and the lowest family income group—
To simulate the second proposed intervention, changes were those with parent educational attainment of 10th grade and family
made to reduce the effect of low parental educational attainment income of $27,000 a year.

Table 2. Effects of Test Interventions in the Model

Reduce Effect Reduce Effect
Parent of Low Parent of Low Parent Reduce Effect
Educational Standard Education Education of Low Parent Reduce Effect of Combination
Attainment Asian-White on Cognitive on Health & Education on Low Income on of All Four
(Grade Level) Readiness Development Nutrition Income Family Stress Interventions
10 2.9 3.5 3.1 3.9 3.8 4.3
12 4.3 4.5 4.3 4.3 4.3 4.5
14 4.6 4.7 4.6 4.6 4.6 4.7
16 5.6 5.6 5.6 5.6 5.6 5.6
18 6.0 6.0 6.0 6.0 6.0 6.0
20 6.5 6.5 6.5 6.5 6.5 6.5

34 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
CONCLUSIONS we’re serious, we have to do things that overcome the dam-
ages of poverty. We have to meet their health needs, their
What has been reported in this article has been an exercise in sys-
mental health needs, after-school programs, summer pro-
tems analysis. The problem addressed, which has an extensive lit-
grams, parent engagement, early-childhood services. These
erature, is the variability in the readiness of 5-year-olds for school.
are the so-called wraparound services. Some people think of
While the analysis is theoretical in nature, a kind of thought exper-
them as add-ons. They’re not. They’re imperative.”
iment, it is generally consistent with the literature on school read-
iness and early childhood development and yet, at the same time, No, it’s not easy and there’s more to the problem than closing the
emphasizes the limitations of the current knowledge base. Most of school readiness gap and the academic achievement gap, per se.
the research available on the development of readiness for school There is also the larger task of closing the “life gap,” the dramatic
is correlational in nature. What is needed, although it is difficult to difference in standards of living that characterize families with dif-
do, is bivariate experimental research that would provide the effect ferent educational backgrounds and incomes—which goes well
sizes that are needed for more precise systemic analysis. beyond scores on a limited range of high-stakes tests.
The overarching conclusion from studying the “achievement gap”
over the past several years, and looking systematically at the dynam- References
ics of both schooling and early childhood development, is not opti- Amato, P. R. (2005). The impact of family formation change on the
mistic.The policies that work in schools to improve the achievement cognitive, social, and emotional well-being of the next generation. The
of initially low-readiness children work also to improve the achieve- Future of Children, 15(2), 75–86.
ment of initially average- and high-readiness students. Doyle, O., Finnegan, S., & McNamara, K. A. (2010). Differential teacher
This is not to say that we should not invest in “good schools”: and parent ratings of school readiness in a disadvantaged community. Dublin,
strong school leadership, high-quality teachers, rigorous curricula, IE: Geary Institute, University College.
and all the other factors that characterize “good schools.” Bring- Frey, S. (2015, August). Study of brain scan shows impact of poverty on academic
ing initially low-readiness students closer to national academic achievement. Oakland, CA: EdSource.
achievement norms even if other students do better is a virtuous Gaynor, A. K. (2011/2012a). Different students: How typical schools
and worthy effort. are built to fail and need to change: A structural analysis. Journal of
At the level of early childhood development, as the analytic Education, 192(2/3), 13–27.
Gaynor, A. K. (2011/2012b). A reflection on the structural analysis and
work reported in this article suggests, implementing policies to
the case study. Journal of Education, 192(2/3), 31–32.
close the “readiness gap” is costly, imperfect, and probably politi-
Gaynor, A. K. (2012a). The racial, ethnic, and social class achievement
cally difficult. However, the reports of a teacher and an advocate gaps: A systems analysis. International Education Studies, 5(1), 29–49.
offer evidence of the critical need to address this gap that, as has Gaynor, A. K. (2012b, Summer).Typical schools need change. The Connector,
been noted, is a matter of social justice. The teacher is James Bou- isee Systems. Retrieved from http://www.iseesystems.com/community/
tin, who told the story of one of his former students in a blog that connector/Zine/2012_Summer/TypicalSchoolsNeedChange.aspx
was quoted by Valerie Strauss in The Washington Post on December Isaacs, J. B. (2012, March). The Social Genome Project: Starting school at
24, 2014: a disadvantage: The school readiness of poor children. Washington, D.C.:
Center on Children and Families, Brookings Institution.
Getting to and from school was not the only challenge Guill- Isaacs, J., & Magnuson, K. (2011). The Social Genome Project: Income and
ermo faced, though. His father abandoned his mother and education as predictors of children’s school readiness. Washington, D.C.:
siblings when he was 4 years old after some years of ver- Center on Children and Families, Brookings Institution.
bal and physical abuse, and his mom could not get a regular Karoly, L. A., Kilburn, M. R., & Cannon, J. S. (2005). Research brief:
housing situation on her own. Although I didn’t learn about Children at risk: Consequences for school readiness and beyond. Santa Monica,
these facts until after he’d left my classroom, it made a lot of CA: The RAND Corporation.
sense. Guillermo was a student who had suffered the loss and Layton, L. (2015, January 16). Majority of U.S. public school students are
abuse of his father, and the financial instability of his mother. in poverty. The Washington Post. Retrieved from http://wbhsenglish
On top of that, he struggled with the same challenges that .net/nic hols11/hhar r is11/tag/httpwww-washingtonpost
teenagers who don’t face such tremendous trauma deal with -comlocaleducationmajority-of-us-public-school-students-are-in
on a daily basis: hormonal changes, fitting in at school, and -poverty20150115df7171d0-9ce9-11e4-a7ee-526210d665b4_story
-html/
finding an identity.
Mukherjee, S. (2010). The emperor of all maladies: A biography of cancer. New
The advocate is Michael Rebell, Campaign for Educational Equity, York, NY: Scribner.
Teachers College, Columbia University, quoted by Linda Layton in Princeton University Library, Data and Statistical Services. (2007).
an article in TheWashington Post on January 16, 2015. Introduction to regression. Retrieved from http://www.princeton.edu
Rouse, C., Brooks-Gunn, J., & McLanahan, S. (Eds.). (2005). School
“We have to think about how to give these kids a meaningful readiness: Closing racial and ethnic gaps. The Future of Children, 15(1).
education,” he said. “We have to give them quality teachers, Strauss, V. (2014, December 24). We are trying to close the achievement
small class sizes, up-to-date equipment. But in addition, if gap all wrong. The Washington Post. Retrieved from http://www

DEVELOPMENT TOWARD SCHOOL READINESS 35

 .washingtonpost.com/blogs/answer-sheet/wp/2014/12/24/we United States Department of Labor, Bureau of Labor Statistics.
-are-trying-to-close-the-achievement-gap-all-wrong-teacher/ (2014). Employment projections. Retrieved from http://nces.ed.gov/
Sunstein, C., & Hastie, R. (2015). Wiser: Getting beyond groupthink to make programs/coe/indicator_caa.asp
groups smarter. Cambridge, MA: Harvard Business School Publishing.
United States Department of Health and Human Services, Administration
for Children and Families, Office of Head Start. (2015). The Head Start Alan Kibbe Gaynor is associate professor emeritus, Boston University School
child development and early learning framework: Promoting positive outcomes of Education. Dr. Gaynor can be reached at agaynor@bu.edu.
in early childhood programs serving children 3–5 years old. Washington,
D.C.: Author.
United States Department of Labor, Bureau of Labor Statistics. (2013).
America’s youth at 25: School enrollment, number of jobs held and labor market
activity: Results from a longitudinal study. Washington, D.C.: Author.

36 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
Appendix A. Model Equations

Child’s_Academic__Language_Development(t) = Child’s_Academic__Language_Development(t - dt) + (Rate_of_Increase_in_

Academic__Vocabulary_and_Language_Development) * dt
INIT Child’s_Academic__Language_Development = 0

INFLOWS:
Rate_of_Increase_in_Academic__Vocabulary_and_Language_Development = 1*((Quality_and_Intensity_of_Child’s_Formal_
and_Informal_PreSchooling+Effect_of_the_Neighborhood_on_Vocabulary_and_Language_Development+2*Effect_of_Parent_
Education_on_Child_Vocab_&_Language_Development+2*Effect_of_Level_of_Standard_English_&_Acad_Cult_on_Voc_&_Lang_
Dev+Effect_of_Self_Reading_on_Lang_&_Voc_Development)/7)
Child’s_Age(t) = Child’s_Age(t - dt) + (Increase_in_Child’s_Age) * dt
INIT Child’s_Age = 0.01

INFLOWS:
Increase_in_Child’s_Age = 1
Child’s_Level_of_Conceptual_Development(t) = Child’s_Level_of_Conceptual_Development(t - dt) + (Rate_of_Increase_in_
Conceptual_Development) * dt
INIT Child’s_Level_of_Conceptual_Development = 0

INFLOWS:
Rate_of_Increase_in_Conceptual_Development = (1*((3*Effect_of_Parent_Education_on_Conceptual_Development+2*Effect_
of_the_Child’s_PreSchooling_on_Social_&_Conceptual_Develop+2*Effect_of_Standard_English_&_Academic_Culture_in_the_
Home+Effects_of_Physical_Healthon_Conceptual_Development)/8))
Child’s_Level_of_Social_Development(t) = Child’s_Level_of_Social_Development(t - dt) + (Rate_of_Increase_or_Decrease_in_
Social_Development) * dt
INIT Child’s_Level_of_Social_Development = 0

INFLOWS:
Rate_of_Increase_or_Decrease_in_Social_Development = ((Effect_of_Maternal_Attachment_on__Child_Behavior+Effect_on_
Child_Behavior_of_Family_Instab_&_Stress+Effect_of_the_Child’s_PreSchooling_on_Social_&_Conceptual_Develop+Effect_of_
Parent_Education_on_the_Child’s_Social_Development)/4)
Maternal_Attachment(t) = Maternal_Attachment(t - dt) + (Flow_1) * dt
INIT Maternal_Attachment = 2

INFLOWS:
Flow_1 = Goal_of_Maternal_Attachment-Maternal_Attachment
Availability_of_Health_Care = GRAPH(Mean_Income_of__Affordable_Neighborhood)
(20000, 1.00), (27000, 1.00), (34000, 1.20), (41000, 1.75), (48000, 2.00), (55000, 2.20), (62000, 2.50), (69000, 2.50), (76000,
2.80), (83000, 3.00), (90000, 3.00)
Child’s_Aspiration_and_Self_Expectation = 2*((Effect_of_Parent_Education_on_Child’s_Aspiration_and_Self_Expect+Effect_of_
Maternal_Attachment_on_Aspiration__&_Self_Expectation)/2)
Child’s_Degree_of_Positive_Behavior = IF (2*(Child’s_Social_Development__Divided_by_Age_Age*((Effect_of_Maternal_
Attachment_on__Child_Behavior+Effect_on_Child_Behavior_of_Family_Instab_&_Stress))))<3 THEN((Child’s_Social_
Development__Divided_by_Age_Age*((Effect_of_Maternal_Attachment_on__Child_Behavior+Effect_on_Child_Behavior_of_
Family_Instab_&_Stress)/2))) ELSE (3)

DEVELOPMENT TOWARD SCHOOL READINESS 37

Child’s_Level_or_Readiness_for_School = (Child’s_Age*(Child’s_Level_of_Social_Development/Child’s_Age+Child’s_Level_
of_Conceptual_Development/Child’s_Age+Child’s_Academic__Language_Development/Child’s_Age+Effect_of_the_Level_of_
Aspir_&_Self_Expect_on_School_Readiness+Effect_of_Resilience_on_Readiness_for_School+Effect_of_Emotional_Security_on_
Readiness_for_School)/6)
Child’s_Physical_Health = IF (((Effect_of_Health_Care_Utilization_on_Physical_Health+Effect_of_Nutrition_on_Physical_
Health)/2)*Effect_of_Family_Instability_&_Stress_on_Physical_Health)<3 THEN (Effect_of_Health_Care_Utilization_on_Physical_
Health+Effect_of_Nutrition_on_Physical_Health)/2*Effect_of_Family_Instability_&_Stress_on_Physical_Health ELSE 3
Child’s_Resilience = (Child’s_Level_of_Social_Development/Child’s_Age+Child’s_Aspiration_and_Self_Expectation+Quality_and_
Intensity_of_Child’s_Formal_and_Informal_PreSchooling+2*Maternal_Attachment)/5
Child’s_Self_Reading = Child’s_Academic__Language_Development/Child’s_Age
Child’s_Social_Development__Divided_by_Age_Age = Child’s_Level_of_Social_Development/Child’s_Age
Degree_of_Utilization_of_Health_Care = IF (2*((3*Effect_of_Availability_on_Degree_of__Utilization_of_Health_Care+2*Effect_
of_PEA_on_Use_of_Health_Care)/5))<3 THEN ((3*Effect_of_Availability_on_Degree_of__Utilization_of_Health_Care+2*Effect_
of_PEA_on_Use_of_Health_Care)/5) ELSE (3)
Effects_of_Physical_Health on_Conceptual_Development = GRAPH(Child’s_Physical_Health)
(1.00, 0.7), (1.20, 0.75), (1.40, 0.8), (1.60, 0.85), (1.80, 0.9), (2.00, 1.00), (2.20, 1.20), (2.40, 1.30), (2.60, 1.35), (2.80, 1.40),
(3.00, 1.50)
Effect_of_Availability_on_Degree_of__Utilization_of_Health_Care = GRAPH(Availability_of_Health_Care)
(1.00, 0.5), (1.20, 0.5), (1.40, 0.5), (1.60, 0.6), (1.80, 0.7), (2.00, 0.8), (2.20, 0.85), (2.40, 0.9), (2.60, 1.00), (2.80, 1.00), (3.00, 1.00)
Effect_of_Child_Behavior_on_Maternal_Attachment = GRAPH(Child’s_Degree_of_Positive_Behavior)
(1.00, 0.67), (1.20, 0.7), (1.40, 0.7), (1.60, 0.75), (1.80, 0.8), (2.00, 0.9), (2.20, 0.95), (2.40, 1.00), (2.60, 1.00), (2.80, 1.10),
(3.00, 1.20)
Effect_of_Crime_&_Drug_Trafficking_on_Emotional_Security = GRAPH(Neighborhood_Gang_Crime_&_DrugTrafficking)
(1.00, 1.00), (1.20, 1.00), (1.40, 0.95), (1.60, 0.9), (1.80, 0.85), (2.00, 0.8), (2.20, 0.75), (2.40, 0.7), (2.60, 0.65), (2.80, 0.6),
(3.00, 0.5)
Effect_of_Emotional_Security_on_Readiness_for_School = GRAPH(Emotional_Security)
(0.00, 0.7), (0.3, 0.00), (0.6, 0.7), (0.9, 0.75), (1.20, 0.75), (1.50, 0.75), (1.80, 0.85), (2.10, 1.00), (2.40, 1.00), (2.70, 1.10), (3.00, 1.20)
Effect_of_Family_Income_on_on_Quality_of_PreSchooling = GRAPH(Family_Income)
(20000, 0.7), (33000, 0.8), (46000, 0.9), (59000, 0.95), (72000, 1.00), (85000, 1.00), (98000, 1.20), (111000, 1.30), (124000,
1.35), (137000, 1.40), (150000, 1.50)
Effect_of_Family_Instability_&_Stress_on_Emotional_Security = GRAPH(Family_Instability_and_Stress)
(1.00, 1.20), (1.20, 1.10), (1.40, 1.00), (1.60, 0.9), (1.80, 0.85), (2.00, 0.8), (2.20, 0.75), (2.40, 0.7), (2.60, 0.65), (2.80, 0.6),
(3.00, 0.5)
Effect_of_Family_Instability_&_Stress_on_Materna_Attachment =
GRAPH(Family_Instability_and_Stress)
(1.00, 1.20), (1.20, 1.10), (1.40, 1.00), (1.60, 1.00), (1.80, 0.95), (2.00, 0.9), (2.20, 0.8), (2.40, 0.75), (2.60, 0.7), (2.80, 0.7),
(3.00, 0.67)
Effect_of_Family_Instability_&_Stress_on_Physical_Health = GRAPH(Family_Instability_and_Stress)
(1.00, 1.20), (1.20, 1.10), (1.40, 1.00), (1.60, 0.9), (1.80, 0.85), (2.00, 0.8), (2.20, 0.75), (2.40, 0.7), (2.60, 0.65), (2.80, 0.6),
(3.00, 0.5)
Effect_of_Health_Care_Utilization_on_Physical_Health = Degree_of_Utilization_of_Health_Care

38 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5
Effect_of_Immigrant_or__Minority_Status_on_Income = IF (Family_Immigrant__and_Minority_Status = 1) THEN (.8) ELSE (1)
Effect_of_Immigr_&_Min_Status_on_Engl_&_Acad_Culture_in_the_Home = IF (Family_Immigrant__and_Minority_Status=1)
THEN (.8) ELSE (1)
Effect_of_Income_on_Family_Stress = IF (Family_Income<50000) THEN (3) ELSE IF (Family_Income<90000) THEN (2) ELSE (1)
Effect_of_Level_of_Standard_English_&_Acad_Cult_on_Voc_&_Lang_Dev =
GRAPH(Level_of_Standard_English_&_Acad_Cult_in_the_Home)
(1.00, 0.5), (1.20, 0.6), (1.40, 0.7), (1.60, 0.8), (1.80, 0.9), (2.00, 1.00), (2.20, 1.00), (2.40, 1.00), (2.60, 1.00), (2.80, 1.10), (3.00, 1.20)
Effect_of_Maternal_Attachment_on_Aspiration__&_Self_Expectation =
GRAPH(Maternal_Attachment)
(1.00, 0.5), (1.20, 0.5), (1.40, 0.5), (1.60, 0.6), (1.80, 0.7), (2.00, 0.8), (2.20, 0.85), (2.40, 0.9), (2.60, 1.00), (2.80, 1.00), (3.00, 1.20)
Effect_of_Maternal_Attachment_on__Child_Behavior = GRAPH(Maternal_Attachment)
(1.00, 0.7), (1.20, 0.75), (1.40, 0.8), (1.60, 0.85), (1.80, 0.9), (2.00, 1.00), (2.20, 1.10), (2.40, 1.20), (2.60, 1.25), (2.80, 1.30),
(3.00, 1.35)
Effect_of_Neighborhood_Income_Level_on_Variety_of_Food = GRAPH(Mean_Income_of__Affordable_Neighborhood)
(20000, 1.00), (33000, 1.00), (46000, 1.20), (59000, 1.75), (72000, 2.00), (85000, 2.20), (98000, 2.50), (111000, 2.50), (124000,
2.80), (137000, 3.00), (150000, 3.00)
Effect_of_Nutrition_on_Physical_Health = Family__Nutrition
Effect_of_Parent_Education_on_Child’s_Aspiration_and_Self_Expect = GRAPH(Parent___Educational_Attainment)
(10.0, 1.00), (11.0, 1.30), (12.0, 1.50), (13.0, 1.75), (14.0, 2.00), (15.0, 2.10), (16.0, 2.70), (17.0, 2.80), (18.0, 3.00), (19.0, 3.00),
(20.0, 3.00)
Effect_of_Parent_Education_on_Child_Vocab_&_Language_Development = GRAPH(Parent___Educational_Attainment) (10.0, 0.7),
(11.0, 0.8), (12.0, 0.9), (13.0, 0.95), (14.0, 1.00), (15.0, 1.00), (16.0, 1.20), (17.0, 1.30), (18.0, 1.35), (19.0, 1.40), (20.0, 1.50)
Effect_of_Parent_Education_on_Conceptual_Development =
GRAPH(Parent___Educational_Attainment) (10.0, 0.7), (11.0, 0.75), (12.0, 0.8), (13.0, 0.85), (14.0, 0.9), (15.0, 1.00), (16.0,
1.20), (17.0, 1.30), (18.0, 1.35), (19.0, 1.40), (20.0, 1.50)
Effect_of_Parent_Education_on_English_&_Acad_Cult_in_the_Home = GRAPH(Parent___Educational_Attainment) (10.0, 0.5),
(11.0, 0.55), (12.0, 0.6), (13.0, 0.7), (14.0, 0.9), (15.0, 1.00), (16.0, 1.20), (17.0, 1.30), (18.0, 1.35), (19.0, 1.40), (20.0, 1.50)
Effect_of_Parent_Education_on_PreSchooling = GRAPH(Parent___Educational_Attainment) (10.0, 0.7), (11.0, 0.8), (12.0, 0.9),
(13.0, 0.95), (14.0, 1.00), (15.0, 1.00), (16.0, 1.20), (17.0, 1.30), (18.0, 1.35), (19.0, 1.40), (20.0, 1.50)
Effect_of_Parent_Education_on_the_Child’s_Social_Development = GRAPH(Parent___Educational_Attainment) (10.0, 0.7), (11.0,
0.8), (12.0, 0.9), (13.0, 0.95), (14.0, 1.00), (15.0, 1.00), (16.0, 1.20), (17.0, 1.30), (18.0, 1.35), (19.0, 1.40), (20.0, 1.50)
Effect_of_PEA_on_Use_of_Health_Care = GRAPH(Parent___Educational_Attainment) (10.0, 1.00), (11.0, 1.00), (12.0, 1.00),
(13.0, 2.00), (14.0, 2.00), (15.0, 2.00), (16.0, 3.00), (17.0, 3.00), (18.0, 3.00), (19.0, 3.00), (20.0, 3.00)
Effect_of_Resilience_on_Readiness_for_School = GRAPH(Child’s_Resilience) (1.00, 0.5), (1.20, 0.6), (1.40, 0.7), (1.60, 0.8),
(1.80, 0.9), (2.00, 1.00), (2.20, 1.00), (2.40, 1.00), (2.60, 1.00), (2.80, 1.10), (3.00, 1.20)
Effect_of_Self_Reading_on_Lang_&_Voc_Development = GRAPH(Child’s_Self_Reading) (1.00, 0.5), (1.20, 0.6), (1.40, 0.7), (1.60,
0.8), (1.80, 0.9), (2.00, 1.00), (2.20, 1.00), (2.40, 1.00), (2.60, 1.00), (2.80, 1.10), (3.00, 1.20)
Effect_of_Single_Parent_Satus_on_Family_Stress = IF (Single_Parent_Family_Status=1) THEN (1.2) ELSE (1)
Effect_of_Single_Parent_Status_on_Family_Income = IF (Single_Parent_Family_Status=1) THEN (.75) ELSE (1)
Effect_of_Single_Parent_Status_on_Maternal_Attachment = IF (Single_Parent_Family_Status=1) THEN (.9) ELSE (1)

DEVELOPMENT TOWARD SCHOOL READINESS 39

Effect_of_Standard_English_&_Academic_Culture_in_the_Home = GRAPH(Level_of_Standard_English_&_Acad_Cult_in_the_
Home) (1.00, 0.7), (1.20, 0.75), (1.40, 0.8), (1.60, 0.85), (1.80, 0.9), (2.00, 1.00), (2.20, 1.20), (2.40, 1.30), (2.60, 1.35), (2.80,
1.40), (3.00, 1.50)
Effect_of_the_Child’s_PreSchooling_on_Social_&_Conceptual_Develop = GRAPH(Quality_and_Intensity_of_Child’s_Formal_and_
Informal_ PreSchooling) (1.00, 0.7), (1.20, 0.7), (1.40, 0.7), (1.60, 0.8), (1.80, 0.85), (2.00, 0.9), (2.20, 1.00), (2.40, 1.00), (2.60,
1.10), (2.80, 1.30), (3.00, 1.50)
Effect_of_the_Level_of_Aspir_&_Self_Expect_on_School_Readiness = GRAPH(Child’s_Aspiration_and_Self_Expectation) (1.00,
0.5), (1.20, 0.5), (1.40, 0.5), (1.60, 0.6), (1.80, 0.7), (2.00, 0.8), (2.20, 0.85), (2.40, 0.9), (2.60, 1.00), (2.80, 1.00), (3.00, 1.20)
Effect_of_the_Neighborhood_on_Vocabulary_and_Language_Development = GRAPH(Mean_Income_of__Affordable_
Neighborhood) (20000, 0.5), (33000, 0.6), (46000, 0.7), (59000, 0.8), (72000, 0.9), (85000, 1.00), (98000, 1.00), (111000, 1.00),
(124000, 1.00), (137000, 1.20), (150000, 1.50)
Effect_on_Child_Behavior_of_Family_Instab_&_Stress = GRAPH(Family_Instability_and_Stress) (1.00, 1.00), (1.20, 1.00), (1.40,
1.00), (1.60, 1.00), (1.80, 1.00), (2.00, 1.00), (2.20, 0.9), (2.40, 0.7), (2.60, 0.6), (2.80, 0.55), (3.00, 0.5)
Emotional_Security = (((Quality_and_Intensity_of_Child’s_Formal_and_Informal_PreSchooling+Child’s_Physical_
Health)/2)*(Effect_of_Crime_&_Drug_Trafficking_on_Emotional_Security+Effect_of_Family_Instability_&_Stress_on_Emotional_
Security)/2)
Family_Immigrant__and_Minority_Status = 2
Family_Income = (Income_Based_on_Education*Effect_of_Immigrant_or__Minority_Status_on_Income)*Effect_of_Single_Parent_
Status_on_Family_Income
Family_Instability_and_Stress = IF (Effect_of_Income_on_Family_Stress*Effect_of_Single_Parent_Satus_on_Family_Stress)<3 THEN
(Effect_of_Income_on_Family_Stress*Effect_of_Single_Parent_Satus_on_Family_Stress) ELSE 3
Family__Nutrition = Effect_of_Neighborhood_Income_Level_on_Variety_of_Food
Goal_of_Maternal_Attachment = IF (Natural_Maternal_Attachment*Effect_of_Child_Behavior_on_Maternal_Attachment*Effect_
of_Single_Parent_Status_on_Maternal_Attachment*Effect_of_Family_Instability_&_Stress_on_Materna_Attachment)<3 THEN
(Natural_Maternal_Attachment*Effect_of_Child_Behavior_on_Maternal_Attachment*Effect_of_Single_Parent_Status_on_Maternal_
Attachment*Effect_of_Family_Instability_&_Stress_on_Materna_Attachment) ELSE 3
Income_Based_on_Education = GRAPH(Parent___Educational_Attainment) (10.0, 25000), (11.0, 35000), (12.0, 50000), (13.0,
55000), (14.0, 70000), (15.0, 75000), (16.0, 90000), (17.0, 100000), (18.0, 120000), (19.0, 130000), (20.0, 150000)
Level_of_Standard_English_&_Acad_Cult_in_the_Home = 2*Effect_of_Parent_Education_on_English_&_Acad_Cult_in_the_
Home*Effect_of_Immigr_&_Min_Status_on_Engl_&_Acad_Culture_in_the_Home
Mean_Income_of__Affordable_Neighborhood = Family_Income
Natural_Maternal_Attachment = 3
Neighborhood_Gang_Crime_&_DrugTrafficking = GRAPH(Neighborhood_Rate_of_Unemployment) (2.25, 1.00), (3.12, 1.00),
(4.00, 1.00), (4.88, 1.50), (5.75, 1.80), (6.62, 2.00), (7.50, 2.50), (8.38, 2.75), (9.25, 3.00), (10.1, 3.00), (11.0, 3.00)
Neighborhood_Rate_of_Unemployment = GRAPH(Mean_Income_of__Affordable_Neighborhood) (20000, 11.0), (33000, 10.0),
(46000, 7.50), (59000, 7.00), (72000, 5.40), (85000, 5.00), (98000, 4.00), (111000, 3.40), (124000, 3.20), (137000, 2.25), (150000,
2.25)
Parent___Educational_Attainment = 16
Quality_and_Intensity_of_Child’s_Formal_and_Informal_PreSchooling = 2*((Effect_of_Family_Income_on_on_Quality_of_
PreSchooling+Effect_of_Parent_Education_on_PreSchooling)/2)
Single_Parent_Family_Status = 2

40 J O U R N A L O F E D U C AT I O N • V O L U M E 1 9 5 • N U M B E R 3 • 2 0 1 5