Search Type Title Author/Investigator Search Advanced Search

Table of Contents | Previous | Next

PREDICTORS AND OUTCOMES OF PROGRAM PARTICIPATION, AND CORRELATES OF CHILDREN’S COGNITIVE DEVELOPMENT AT THE EDUCATIONAL ALLIANCE’S EARLY HEAD START

Mark Spellmann, Catherine Tamis-Lemonda, Maria Yarolin, Lisa Baumwell,
Joanne Roberts, and the NYU Early Childhood Research Team
New York University

Background And Research Questions

This is a brief summary of predictors and outcomes of participation in the Early Head Start (EHS) program of The Educational Alliance in New York City. Correlates of children’s cognitive development are also explored. The Educational Alliance, a Settlement House serving families on Manhattan’s Lower East Side for over 100 years, provided center-based EHS services at two sites:

• The EHS center at the Educational Alliance.

• Teen-Aid High School—a New York City Board of Education program for pregnant and parenting teens.

Many families randomly assigned to EHS did not actually participate in the program—42 percent of children assigned to EHS were rated by program staff as having “poor attendance” at the EHS childcare centers. Two research questions thus emerged from this context:

• What family characteristics predicted EHS program participation?

• What child and family characteristics were affected by EHS program participation?

In addition to exploring outcomes of participation, we were also interested in exploring correlates of children’s cognitive development, as a first stage toward modeling pathways to gains in cognitive development. This inquiry was guided by the final research question to be addressed in this study:

• What child and family characteristics were associated with children’s cognitive development?

Method

Research Participants

Participants included all those families who were randomly assigned at either Teen Aid or the Educational Alliance (N = 141). Demographic characteristics of the sample collected at baseline are presented below in Table 1.

TABLE 1
SAMPLE FREQUENCIES

	Frequency	Percent
Ethnicity
African-American	59	48%
Hispanic	51	41%
Asian	8	6%
Caucasian	2	2%
Mixed	3	2%
Caribbean	1	1%
Did not identify	17
Family's country of origin
Puerto Rico	22	16%
Dominican Republic	3	2%
Mexico	4	3%
Caribbean Island	7	5%
USA	89	61%
China	6	4%
Indian	1	1%
Guatemala	1	1%
Ecuador	3	2%
Panama	1	1%
Central American Country	2	2%
Puerto Rico/Dominican	1	1%
West Indian	1	1%
Mother's age at child's birth
14-15	10	7%
16-17	42	31%
18-19	24	18%
20-29	31	30%
30-39	16	12%
40(+)	2	2%
Primary language in the home
English	109	77%
Spanish	25	18%
Chinese	6	4%
Other	1	1%
Marital Status
Single	67	64%
Live with partner or Married	31	29%
Separated or Divorced	5	5%
Widowed	1	1%
Educational Status
Some Junior High School	2	2%
Graduated Junior High School	7	7%
Some High School	64	68%
Graduated Junior High School	6	6%
GED	2	2%
Some College	7	7%
College Graduate	6	6%
Status of Baby's Father
Residential	28	32%
Not residential, but involved with child and mother	33	37%
Not residential, but involved with child only	4	5%
No contact/not involved	23	26%
Social Services
AFDC	29	36%
Medicaid	77	82%
Food stamps	25	31%
WIC	63	78%
SSI/SSD	9	11%

Note: Due to missing data, not all categories total to 141

Program Participation Ratings

EHS staff rated families on two dimensions of program participation—child attendance at the EHS childcare centers and parent involvement with EHS social service staff. Ratings were based upon a four-point scale. Consistency of a children’s attendance was rated as “Poor,” “Fair” “Good” or “Excellent.” Ratings of how often parents worked with EHS social services staff ranged from “Not at all” to “Occasionally” to “Fairly Often” to “Regularly.” Ratings were generated at the end of the completion of the program. For this analysis, children’s attendance was collapsed into one of two categories—“Fair to Excellent Attendance” or “Poor Attendance.”

For the analysis of predictors of program participation, only those randomly assigned to EHS were included (N = 67). For the analysis of outcomes of participation, those randomly assigned to EHS, but rated “Poor Attendance,” were excluded. Thus in the analysis of outcomes of participation, the “Fair(+) Attendance” variable contrasts EHS families whose children had fair to excellent attendance (N = 39) with the control group (N = 71). The rationale for this grouping was to compare children who participated in the EHS program with those who did not.

Table 2 summarizes children’s attendance and parent involvement by site. As can be seen in the final row of Table II, parent involvement was almost twice as great at the Teen Aid site, compared to the Educational Alliance site (51 percent versus 28 percent). This difference was likely due to the program setting. Teen Aid mothers were attending high school on-site, thus they were available to work with the EHS social worker most school days.

TABLE 2
CHILD ATTENDANCE AND PARENT INVOLVEMENT BY PROGRAM SITE

	Frequency	Percent
Total Randomly Assigned	141	100%
Randomly assigned through the Educational Alliance	77	55%
Randomly assigned through Teen Aid	64	45%
Total assigned to EHS	70	50%
Total assigned to control group	71	50%

TABLE 2A
CHILD ATTENDANCE AND PARENT INVOLVEMENT BY PROGRAM SITE

	Combined Sites	Ed. Alliance	Teen Aid	Combined Sites	Ed. Alliance	Teen Aid
Child Attendance	671	36	31
Poor	28	14	14	42%	39%	48%
Fair	11	3	8	16%	8%	28%
Good-Excellent	28	19	9	42%	53%	32%
Parent Involvement	671	36	31
Not at all	22	14	8	33%	39%	26%
Occasionally	19	12	7	28%	33%	23%
Fairly Often-Regularly	26	10	16	39%	28%	51%

1Staff unable to rate 3 families

Instruments

Dyad Ratings

Children and their caregivers were videotaped playing together in their homes. Observations were conducted during home visits when children were six, fourteen, twenty-four, and thirty-six months old. Dimensions mothers were rated on included positive affect, positive touch, positive verbal reinforcement, responsiveness, emotional attunement, participation with child, structuring, overall consistency, language use, caregiver, quality of language, use of teaching loop, achievement orientation, inventiveness with toys, and sophistication of play. Factor analysis demonstrated that these dimensions loaded on a single factor, termed “Maternal Didactic Responsiveness”. Dyadic interaction was rated on three dimensions—mutual enjoyment, mutual communication, and reciprocal interaction. Factor analysis demonstrated these dimensions loaded on a single factor, termed “Dyad Mutuality”. Ratings of children’s language use and communication abilities formed the “Child Quality of Communication factor”. Children’s positive affect and positive touch formed the “Child Positive Affect factor”.

Scale Scores

Measures from the national evaluation collected when children were 14 and 24 months were included in the analyses. Additionally, Bayley Mental Development Index (MDI) scores from 36 months were also included. Scales and observational measures from our local research study were administered when children were six, 14, 24, and 36 months old. Table 3 reports on the psychometric properties of scales in our local survey that were included in this report.

Results

Predictors of Participation

All scale scores and dyadic observation measures included in our six-month assessment were tested against the two measures of EHS program participation—child attendance and parent involvement (“Fair to Excellent Attendance versus Poor Attendance” and “How Much Parents Worked with EHS Social Service Staff”). These data may not have constituted true baseline measures for all families, as a few children began attending the EHS center-based care from the age of 4 months; but they represent our earliest data from our families.

As can be seen in Table 4, exposure to violence significantly lowered the degree to which families participated in the program. Cultural values also affected participation. More traditional cultural values were associated with lower program participation. These findings give rise to the question of whether cultural values might have been associated with exposure to violence. Follow-up analyses showed no significant correlation between cultural values and exposure to community or domestic violence.

Fathers—both in mothers’ family of origin, and in babies’ current family—played a significant role in affecting program participation. Current father involvement predicted higher attendance and involvement with social service staff. Harsh fathering that mothers experienced in their families when they were growing up was associated with lower attendance. Warm, accepting fathering was associated with higher attendance. Measures of quality of parenting, quality of mother-infant interaction, maternal mental health, social support (except from baby’s father), and the quality of the relationship between EHS mothers and their own mothers when they were growing up (family of origin) were not significantly associated with participation. Only correlations that reached at least a trend toward significance are reported in Table 4.

TABLE 3
INTERNAL CONSISTENCY OF LOCAL MEASURES1

Psychosocial Variables - Support, Psychological Well-Being	hronbach's Alpha
Practical Support - Vaux Social Support Record	0.74
Emotional Support- Vaux Social Support Record	0.62
Advice and Guidance Support- Vaux Social Support Record	0.74
Support from EHS: All staff	0.88
What I Got from EHS: Growth as a Parent	0.95
What I Got from EHS: Personal Growth	0.92
What I Got from EHS: Family Program Bond	0.96
What I Got from EHS: Child Development	0.93
Working Alliance Inventory Total (WAI)	0.97
Conflict in Approach - WAI	0.78
Goal - WAI	0.97
Emotional Bond - WAI	0.96
Merhabian Empathy Scale	0.83
Parenting Stress Inventory (PSI): General	0.88
Parenting Stress Inventory (PSI): Child	0.86
Maternal Efficacy Scale	0.92
PTSD (Impact of Events Scales)	0.91
Center for Epidemiological Studies-Depression scale (CES-D)	0.86
Parental Acceptance and Rejection Questionnaire (PARQ): Mother Was Loving/Accepting	0.95
PARQ: Mother Was Harsh and Rejecting	0.95
PARQ: Father Was Loving/Accepting	0.97
PARQ: Father Was Harsh and Rejecting	0.94
Parent Stress Inventory (PSI): General Stress	0.88
PSI: Stress Related to Parenting	0.95
Parenting
Maternal Self Rating: Didactic	0.86
Maternal Self Rating: Nurturing	0.83
Maternal Self Rating: Autonomy Support	0.78
Mother's Rating of Child's Father: Didactic & Nurturing	0.98
Mother's Rating of Child's Father: Autonomy Support	0.85
Mother's Rating of Ideal Father: Didactic & Nurturing	0.98
Mother's Rating of Ideal Father: Autonomy Support	0.8
Modernity: Traditional Values/Respect for Authority	0.77
Modernity: Value children having their own point of view	0.77
Modernity: Belief that children will naturally misbehave unless taught to behave	0.55
Child Development 36 mo
Social Skills Rating System (SSRS)-Positive Social Behaviors	0.95
SSRS-Negative: Disrupts, Aggressive, Loner	0.86

1References for all scales are included at the end of this article

TABLE 4
ANTECEDENTS OF EHS PARTICIPATION: PSYCHOSOCIAL VARIABLES ASSESSED WHEN CHILD WAS SIX MONTHS OLD N = 20-40

Sample: Only families randomly assigned to EHS	Fair (+) Attendance vs. Poor Attendance	Degree of Involvement With EHS Social Service Staff
Exposure to Violence
Total Community Violence Experienced: Past 5 Years	-.41*	-.32t
Total Domestic Violence Experienced: Past Year	-.46*	-.40*
Witnessed/Aware of Domestic Violence toward Others	-.39*	-.36t
Psychosocial Variables
PARQ: Harshness/Rejection by Father (family of origin)	-.40t
PARQ: Love/Acceptance by Father (family of origin)	.32t
Maternal Efficacy		.44*
Cultural Variables
Modernity: Belief that it is good for children to have their own point of view		.38t
Modernity: Belief that children will naturally misbehave unless taught to behave		-.47*
Support from Baby's Father
Living With Partner (husband/living as married)	.36*
Baby's Father is a Caretaker	.35*	.35*
Social Support from Baby's Father	.33t

* p < .05, t p < .1

Outcomes of Participation

Measures tested as outcomes of participation included measures from the national evaluation that were collected when children were 14 and 24 months old, and 36 month Bayley MDI scores, and measures collected in the local outcome research when children were 14, 24 and 36 months of age. Fourteen-month assessments may occur too early to be considered outcome measures. However, most families at the Teen Aid EHS site completed their tenure at Teen Aid High School before their children turned two years old. Thus for one of our two EHS centers, the 14-month assessment was often the last assessment that occurred during families intensive period of EHS participation. Therefore, we are including measures from the 14-month assessment in the table of outcomes presented below.

Children’s cognitive development was associated with both program participation variables at each age milestone, as the MDI-Participation correlations below demonstrate. Moderately strong effects for participation on children’s social development were also found (SSRS-Positive Social Behaviors). Moderately strong program effects on children’s language were demonstrated by correlations between participation and the Vineland Communication Domain and the Child Quality of Communication dyad rating.

Parental domains significantly associated with participation included quality of parent-child interaction, quality of parenting, discipline strategies, parenting stress, psychological well-being, and social support. Table 5 presents correlates with program participation. Only correlations that reached at least a trend toward significance are reported.

TABLE 5
OUTCOMES OF PARTICIPATION: FAMILIES WITH FAIR-TO-EXCELLENT ATTENDANCE VERSUS CONTROL GROUP FAMILIES
N = 40-86

Sample: Families randomly assigned to EHS with "Poor Attendance" were excluded from this sample.	EHS Fair (+) Attendance vs. Control Group	Worked with Social Service Staff
Variables from the National Study (n = 70-86)
Cognitive Development: MDI (Bayley) Scales
MDI 36 mo	.32***	.25**
MDI 24 mo	.28*	.25*
MDI 14 mo	.37**	.31**
Parenting
14 mo High chair: Warm	.30**	.19t
14 mo High chair: Positive Regard	.25*
14 mo High chair: Sensitivity	.32**	.22t
24 mo Discipline: Prevent-Distract	.31**	.30**
24 mo Mild Discipline Only	.28*	.31**
14 mo Discipline: Prevent-Distract	.31**	.30**
24 mo HOME: Non-punitive		.25t
14 mo Parent-Child Play	.21 t	.19 t
14 mo Reading at bedtime	.25*
Mental Health
14 mo Depression (CES-D)	-.27**	-.24*
14 mo Parent Mastery	.23*	.26*
24 mo Parental Stress (PSI)	-.30**	-.27*
24 mo PSI Parent-Child Dysfunctional Interaction	-.40***	-.36**
Variables from the Local Study (n=40-84)
Dyadic Interaction (Coded from Videotaped Interaction)
Maternal Didactic Responsiveness	.22 t	.23*
Dyad Mutuality	.27*	.30*
Child Quality of Communication	.29*	.28*
Child Positive Affect	.29*	.30*
Self-Rated Psychosocial Variables and Parenting
Psychosocial 36 mo
Practical Support	.28*
Emotional Support	.25t
Advice and Guidance Support	.43**	33*
Empathy	.26*
Parenting Stress	-.27*	-.26*
Maternal Self Rating: Didactic		.22*
Maternal Self Rating: Autonomy Support		.25 t
Mother Rated Child Development
SSRS-Positive Social Behaviors	.42*	.38*
Vineland-Communication	.34t	.35*
Vineland-Motor	.27t

** p < .01, * p < .05, t p < .1

Correlates Of Children’s Cognitive Development

A wide range of variables demonstrated significant association with children’s cognitive development, as indexed by MDI scores. The quality of parenting, the quality of parent-child interaction, parent psychological well-being, social support, quality of EHS program engagement, cultural values and other domains of child development all demonstrated significant correlations with MDI scores.

Observational measures of quality of parenting showed substantial associations with cognitive development at 24 and 36 months. Quality of parenting at six months (Maternal Language, Play, and Emotional Availability) was a fairly strong predictor of MDI scores at 24 months and 36 months. Similar findings were found for quality of parenting at 14 months. Maternal Rich Language (coded only at 14 months) was a moderately strong predictor of child cognitive development at 24 months. The high-chair parent-child observation measures from the 14-month national evaluation battery also demonstrated a pattern of correlations with cognitive development at all three age milestones.

The quality of parent-child interaction, as measured by Dyad Mutuality, was significantly associated with cognitive development at 24 and 36 months. Dyad Mutuality at six months was predictive of 36 month MDI. Dyad Mutuality at 24 months was associated with MDI scores at both 24 months and MDI 36 months. Dyad Mutuality at 36 months was associated with the 36 month MDI.

Self-rated parenting measures demonstrated a pattern of findings similar to the observational parenting measures. Mothers’ ratings of their teaching (Maternal Self Rating: Didactic), nurturing (Maternal Self Rating: Nurturing) and Autonomy Support (Maternal Self Rating: Autonomy Support) were all associated with cognitive development at 36 months.

Father involvement was associated with children’s cognitive development. Mothers’ ratings of babies’ fathers’ autonomy support was significantly associated with cognitive development at 36 months. Mothers’ reports of time fathers spent with their children and time spent playing, reading or talking to baby were associated with cognitive development at 14 months. Interestingly, there was a very strong relationship between how much mothers valued father involvement--Ideal Father: Didactic/Nurturing and Ideal Father: Autonomy Support—and cognitive development.

The relationship between children’s cognitive development and the quality of their home environment was demonstrated by positive correlations between HOME observations (total HOME score) at 14 months and cognitive development at 14 and 24 months.

A variety of measures of social support were associated with child cognitive development. Total emotional support and advice and guidance support at 36 months were associated with 36-month MDI scores. Support mothers received from babies’ fathers at 14, 24 and 36 months predicted 36-month MDI scores. Support from mother at 14 and 36 months was associated with 36-month MDI scores.

Program involvement variables were associated with child cognitive development (only families randomly assigned to EHS were included in this analysis). Five program involvement variables—Social Support from EHS staff, “What I Got from EHS: Growth as a Parent,” “What I Got from EHS: Family-Program Bond,” “What I Got from EHS: Child Development,” and the Working Alliance Inventory Goal Disagreement subscale (a negative measure of involvement)—were associated with cognitive development at 14 and 36 months. The stronger pattern of findings for 14 months is likely due to the fact that many families at the Teen Aid site completed their stay at the EHS center by the time their children were 14 months of age.

Measures of parent’s emotional well-being were significantly associated with children’s cognitive development. Symptoms of posttraumatic stress disorder yielded significant negative correlations with MDI scores at 24 and 36 months. Parenting stress was negatively associated with cognitive development at 36 months. Harsh, rejecting parenting by fathers, in mothers’ families of origin, was negatively associated with cognitive development at all three age milestones. The quality of mothering in mothers’ families of origin was associated with MDI scores at 14 and 24 months.

Other aspects of child development also demonstrated significant association with cognitive development. Social development showed a strong correlation with cognitive development (SSRS Positive Behaviors, Negative Behaviors; Vineland Social Domain). Other indices of social behavior, including negative behaviors with parent on observational measures also yielded significant correlations with cognitive development at 24 and 36 months. Mother’s ratings of children’s distractibility, difficult temperament, and difficult behavior were associated with lower MDI scores at 36 months. Children’s health, as rated by their mothers, was associated with cognitive development at 36 months. As would be expected, language development was strongly associated with MDI scores.

Table 6 presents the magnitude and statistical significance of the correlates of children’s cognitive development. All families in the study are included in these analyses. Only correlations that reached at least a trend toward significance are reported.

TABLE 6

CORRELATES OF CHILDREN'S COGNITIVE DEVELOPMENT N= 31-104

	Cognitive Development at Age Milestones
	MDI 14 mo	MDI 24 mo	MDI 36 mo
National Survey Variables (N = 50-104)
Parenting
14 mo High Chair: Warm		.34**	.28*
14 mo High Chair: Positive regard	.25*	.35**	.34**
14 mo High Chair: Sensitivity		.28*
14 mo discipline-remove object	.17 t	.22 t
14 mo HOME Total	.26**	.24*
Mental Health
14 mo Depression (CES-D)	-.28**	-.19 t	-.23*
14 mo Parent Mastery	.17 t		.32**
24 mo Parental Stress			-.26 t
24 mo PSI Dysfunctional Parent-Child Interaction	-.48***	-.54***	.43**
Quality Of Parent Child Relationship
Dyad Ratings And Language Codes (N = 31-84)
36 mo Dyad Mutuality			.29**
36 mo Child: Participation with Caregiver			.32**
36 mo Child: Low Emotional Regulation/Aggressive			-.24*
Behavior toward Caregiver
36 mo Child: Quality of Communication			.40***
36 mo Child: Positive Affect			.22*
24 mo Dyad Mutuality		.43***	.30**
24 mo Child Language Quality		.53***	.38***
24 mo Child: Low Emotional Regulation/Aggressive		-.31**	-.22 t
Behavior toward Caregiver
24 mo Child Persistence			.24 *
14 mo Maternal Language, Play, and Emotional		.36**	.27 t
Availability
14 mo Maternal Intrusiveness & Rigidity		-.36 *
14 mo Maternal Rich Language Factor		.42***
6 mo Maternal Language, Play, and Emotional		.43**	.46***
Availability
6 mo Dyad Mutuality			.28*
Psychosocial and Parenting Variables from NYU survey (n = 31-66)
Social Support: Ehs Program, Family, Baby's Father
Support from all EHS staff1	.25(*)	.30(*)
Working Alliance Inventory Goal Disagreement1			-.39*
What I Got from EHS: Growth as a Parent1	.35*		.27 t
What I Got from EHS: Family-Program Bond1	.31t
What I Got from EHS: Child Development1	.36*
36 mo Emotional Support			.27*
36 mo Advice and Guidance Support			.33**
36 mo Support from baby's father			.27*
24 mo Support from baby's father			.30**
14 mo Support from mother	.35*		.35*
14 mo Support from father			.23 t
14 mo Support from baby's father			.26 t
Parent Mental Health
36 mo PTSD symptoms (IES)			-.39 ***
24 mo PTSD symptoms (IES)		-.43***	-.37 **
36 Parenting Stress-General (PSI)			-.39 **
36 Parenting Stress--From Child (PSI)			-.32*
Family of Origin: Mother was Loving & Accepting	.48**
Family of Origin: Mother was Harsh & Rejecting		-.30*
Family of Origin: Father was Harsh & Rejecting	-.28t	-.39*	-.39**
Parenting (Self-Rated)
Maternal Self Rating: Didactic			.23t
Maternal Self Rating: Nurturing			.20t
Maternal Self Rating: Autonomy Support			.25*
Mother Rating of Father's Autonomy Support			.25*
Mother Rating of Ideal Father: Didactic/Nurturing		.28*	.50***
Mother Rating of Ideal Father: Autonomy Support			.28*
14 mo Time Father spends with child	.31*
14 mo Time Father spends playing, reading or talking to baby	.34**
Child Temperament
Distractibility (DOTS)			-.29*
Difficult temperament (CHQ)			-.30*
Child Social Development
SSRS-Positive Social		.29*	.50***
SSRS-Negative: Disrupts, Argues, Loner			-.33*
Child Health Questionnaire -Negative Behavior		-.29*	-.32*
Vineland Social Development Domain		.44**	.58***
Child Health
Mother's rating of child's general health (CHQ)		-.38**	-.31**

1 (EHS families only) ***p < .001, ** p < .01, * p < .05, t p < .1

Summary and Discussion

The findings from this study demonstrated that participation in the Educational Alliance’s Early Head Start was negatively affected by exposure to community and domestic violence. It seems likely that this would be the case in other EHS programs as well. Programs may need to make greater outreach efforts to overcome the barriers to participation created by exposure to violence.

Further research is needed to better understand the mechanisms by which exposure to violence suppresses participation. It may be that families do not want their situation to come to light because of feelings of shame, or because of fears of legal action that may result in their custodial rights being threatened. Further research into this question could guide better outreach efforts to these vulnerable families.

Similarly, traditional cultural values may also present barriers to participation. Families may fear that their values will not be respected, and that their authority with their children will be undermined. What approaches would be most effective in this situation is an open question. Is it best if differences are openly acknowledged in a climate of respect for different value systems? Can parents who are members of the traditionally oriented culture, and who have bonded with the EHS program play a role in building bridges to other families? Further research is needed to evaluate different approaches to effective outreach to families whose values are more traditional than those of the EHS program.

Father involvement supported participation. It was somewhat surprising that father support was such a robust predictor of participation; especially given how few of the wide range of potential predictors tested yielded significant effects. However, the importance of fathers has been consistently demonstrated in the EHS father research initiative. Further research is needed to explore the mechanisms through which father support promotes participation.

The pattern of findings generated from parent’s experiences in their families of origin was very interesting. When parents experienced their fathers as harsh and rejecting, attendance was likely to be lower. Conversely, when parents experienced their fathers as loving and accepting, attendance was likely to be higher. Similar scales (PARQ) also tapped parents’ feelings about the relationships they had with their mothers when they were growing up; but maternal acceptance and rejection was not associated with attendance. One possible explanation for this set of findings is that the group means for the maternal acceptance dimension was considerably higher than for fathers’ acceptance and rejection, and the father harshness and rejection mean was distinctly greater than the maternal harshness and rejection mean. In other words, most parents reported fairly benign relationships with their mothers when they were growing up, but many parents experienced rejection and harshness from their fathers. Therefore the father relationship dimensions had greater variance and thus greater statistical power. In a much larger sample, the maternal dimensions might have also predicted EHS program participation, but they lacked the statistical power to generate significant correlations in this sample.

Maternal efficacy (when babies were six months old) was positively associated with involvement with EHS social service staff. (There was no effect for maternal efficacy on children’s attendance.) Maternal efficacy is an indicator of how much confidence a parent has in confronting the challenges of raising a young child. Thus this finding suggests it was easier for more confident mothers to engage, or “open up” with EHS social service staff. Hopefully, EHS family workers are well aware that mothers who are less confident or secure in their mothering abilities are likely to hold back from involvement.

Families who apply for EHS and then do not participate in the program can easily be forgotten. EHS programs are fully occupied by serving the families that actively seek their services. Understandably, programs are unlikely to devote their energies to pursuing families who may appear uninterested or unmotivated. Findings in this study though suggest that families who withdraw may do so for very different reasons, with very different implications. Therefore it seems very important for programs to understand as deeply as possible the individual reasons behind withdrawal and low involvement. When families withdraw because there is not a good fit between the child-rearing values of the program and of the family, there is no cause for immediate alarm for the safety and well-being of the child or the family. However, programs may question whether they are sufficiently inviting and inclusive toward all segments of the communities they serve when this is the reason for family withdrawal.

When family withdrawal from an EHS program is related to the lack of father involvement, it is possible that the underlying issue is that the mother lacks the support and resources necessary to sustain involvement. At least in New York City, the tasks of bringing a child to the EHS center and picking up the child again can involve complex and time consuming travel arrangements. Greater family resources make attendance and involvement more likely. Recent research on father involvement makes clear that father involvement translates into greater family resources. (Ongoing EHS research on father involvement may reveal more sophisticated explanations of how fathers affect family childcare decisions.) Thus EHS programs might be alert to the lack of father involvement as an indicator that families new to EHS may need extra attention and support if they are to maintain attendance and involvement.

The most ominous reason (of those uncovered in this study) for a family to withdraw from EHS is exposure to violence. Children and families in these situations are clearly at high risk. Of course EHS programs cannot always know whether domestic violence or community violence is a dynamic in a family’s withdrawal. But EHS staff could explicitly address the question to themselves as to whether any warning signs of violence were evident when families “disappeared.” Further research is needed to explore the magnitude of this problem; and, if necessary, to increase EHS awareness of its dimensions.

Outcomes of Participation

The Educational Alliance’s Early Head Start program demonstrated a wide range of benefits for child development, parenting, and parental psychological well-being. The literature on early intervention programs demonstrates that “Two-Generation” program models are necessary to provide benefits to both children and parents. The Educational Alliance’s EHS program sought to provide direct services to both children and parents, and the data supports the view that the Educational Alliance EHS program was an effective Two-Generation Program.

Child development benefits were found in the realms of cognitive, social, and language development. Cognitive development benefits were manifested at each age milestone. Effect sizes were of moderate strength (average r = .32). The effect size for social development was also moderately strong (SSRS r = .42, .38). Similar effect sizes for communication gains were also found (Ratings of the Quality of Children’s Communication, Vineland Communication Domain, average r = .32).

A range of parenting variables yielded significant correlations with program participation. Small but significant effect sizes were found for observational measures including the high-chair scales, and HOME, and local coded videos (Maternal Didactic Responsive Factor). Survey ratings of parental discipline and parenting also yielded significant correlations with participation. Confidence in these findings is increased by the multi-method nature of the data—both observational measures and survey measures supported the benefits of program participation on parenting and parent discipline strategies.

Program participation showed moderately strong effects on self-report measures of psychological well-being. Dimensions showing significant program effects included emotional distress, depression, parenting stress and social support. Effect sizes ranged from small to moderately strong.

Eliminating the families whose children did not attend the program raises the question of bias. Indeed, the first set of analyses reported in this paper demonstrated that low participation was not a random phenomenon. Families with higher levels of exposure to domestic violence and community violence were less likely to participate in the EHS program. Families with higher father involvement were more likely to participate in the EHS program. Neither exposure to violence or father involvement were significantly associated with children’s cognitive development in this study, thus it is unlikely that the pattern of findings that emerged were substantially affected by lower levels of these variables in the EHS program families. Further, exposure to violence at baseline was a predictor of sample attrition for control group members, as well as a predictor of lower levels of program participation. This attrition probably balances out some of the effects of eliminating “poor attendees” from the EHS group. However, it is possible that some of the parent domain gains may have been enhanced by the exclusion of the “poor-attendance” group.

One purpose of this initial investigation was to identify factors to test as potential moderators of program effects. Exposure to violence, cultural child-rearing values (Modernity) and father involvement have emerged as candidates for inclusion in future analyses.

In summary, the data support the effectiveness of the Educational Alliance’s Early Head Start program in promoting child and family development, for those who actually participated in the program. But it seems unlikely that this range of benefits could have emerged from an analysis that did not take the substantial rates of program nonparticipation into account. An analysis of those who actually participated in the program is crucial for answering the research question “What benefits can be reasonably expected from participating in Early Head Start?”

Correlates of Children’s Cognitive Development

A wide range of factors was associated with children’s cognitive development, including the quality of parenting and the quality of parent-child interaction, parents’ emotional well-being and social support, children’s social development, children’s health, and the quality of families’ involvement with the EHS program. Some, or all, of these dimensions may have been pathways to children’s cognitive gains at the Educational Alliance’s EHS program. Indeed, parenting, parent-child relationship, and parents’ social support and psychological well-being were all positively affected by program participation. Of course, the direction of causality is ambiguous, and likely not uni-directional in these reported associations. But our purpose here was not to test mediators through path analyses, but to illustrate that children’s cognitive development is embedded in multiple levels of systems, at the child, family, and program levels. The implication of these findings is that early intervention programs are likely to be increasingly effective, to the degree that they are able to address each level of the system in which children’s cognitive development is embedded.

References for Scales in Local Research

Abidin, R. (1986). Parenting Stress Index (2nd ed.). Charlottesville, VA: Pediatric Psychology Press.

Gresham, R., and Elliott, S. (1990). The Social Skills Rating System. Circle Pines, Minnesota: American Guidance Service.

Horvath A., and Grrenberg, L. (1989). Development and Validation of the Working Alliance Inventory. Journal of Counseling Psychology, 36(2) 223-233.

Horowitz, M. (1984). A Cross Validation of the Impact of Events Scale. Journal of Consulting and Clinical Psychology, 51(5) 188-194.

Landgraff, J. (1996a) The Child Health Questionnaire: Manual and Interpretation Guide. Boston: The Health Institute.

MacPhee, D., Benson, J. and Bullock, D. (1986). Influences on maternal self-perceptions. Paper presented at the Fifth Biennial International Conference on Infant Studies, April, Los Angeles.

Mehrabian, A., and Epstein, N. (1971). A measure of emotional empathy. Journal of Personality, 40, 525-543.

Rohner, G. (1986). The warmth dimension: Foundations of parental-acceptance rejection theory. San Francisco: Sage Press.

Schemer, E., and Edgerton, M. (1985). Parental and child correlates of parental modernity. in I. Sigel (ed.), Parental belief systems: The psychological consequences for children, 287-318.

Teti, D. and Gelfand. D. (1991). Behavioral Competence among mothers of infants in the first year: The mediational role of maternal self-efficacy. Child Development, 62, 918-929.

Vaux, A., and Harrison, D. (1985). Social network characteristics associated with support satisfaction and perceived support. American Journal of Community Psychology, 13(3) 245-268.

 

 

Table of Contents | Previous | Next