Family Background and Parent–School Interactions in Parent Involvement for At-Risk Preschool Children With Disabilities

Abstract

Using data (N = 1,350) from the Head Start Family and Childhood Experiences Survey, this study examined sociodemographic predictors of parent involvement in educationally enriching activities at home for low-income children with disabilities compared with children without disabilities. Analyses examined whether associations were moderated by aspects of parent–school interactions. Parent involvement was greater for White compared with Black and Hispanic parents of all children. Higher parental education related to greater involvement at the end of the year only for parents of children with disabilities. Parent perceptions of teacher support and school communication differentially moderated associations between predictors and parent involvement for children with and without disabilities. Results inform an individualized approach to fostering involvement among low-income parents of children with disabilities in early education.

Keywords

early childhood parent involvement low-income disability Head Start

Parents may engage their preschoolers in educationally enriching activities at home in a variety of ways, such as reading books, visiting libraries, engaging the child in chores/errands, and playing counting games or reading books that feature numbers (Ramani, Rowe, Eason, & Leech, 2015; Roberts, Jergens, & Burchinal, 2005; Schick, 2014). A wealth of research demonstrates advantages for parents’ at-home involvement in early education, including gains in print knowledge and expressive language (Schick, 2014), receptive language (Roberts et al., 2005), and overall reading ability in kindergarten (Hammer, Farkas, & Maczuga, 2010). Parents also promote early math skills through activities such as reading books with numbers (Ramani et al., 2015), and social-emotional skills through positive parent–child interactions like talking about the school day or playing games (Sheridan, Knoche, Edwards, Bovaird, & Kupzyk, 2010). This study examined factors related to parents’ at-home involvement in educationally enriching activities (home-based parent involvement [PI]) for children with disabilities enrolled in Head Start (HS), including the influence of parents’ interactions with teachers on correlates of PI.

Head Start is the nation’s largest, federally funded early childhood education program for low-income children (Improving Head Start for School Readiness Act, 2007). The 2016 Head Start Performance Standards require that programs have provisions for involving families in all aspects of program services, including purposeful efforts to engage parents in their child’s learning (Head Start Performance Standards, 2016). Preschoolers with disabilities represent at least 10% of HS enrollment. The current study used data from the Family and Child Experiences Survey (FACES; Malone et al., 2013), a nationally representative study of 3,149 children enrolled in HS in 2009. Within this sample, children with disabilities were identified as having speech/language impairment (3.39%), autism/pervasive developmental delay (.62%), intellectual disability/cognitive impairment (.25%), emotional/behavioral disability (.06%), sensory impairment (.50%), or multiple impairments (.39%). All percentages are weighted to adjust for the FACES complex sampling design.

Although PI in general educational activities (e.g., telling stories, playing counting games) is less frequently studied for children with disabilities, PI in specific interventions for preschoolers with disabilities is consistently associated with positive child outcomes. Examples include decreased behavioral problems among preschoolers with attention deficit/hyperactivity disorder (ADHD; Bor, Sanders, & Markie-Dadds, 2002) and improved social skills for preschoolers with autism (Coolican, Smith, & Bryson, 2010). Parents’ school-based involvement (e.g., volunteering in classroom, requesting services for child) is also related to higher early reading skills for preschool and school-age deaf children (Calderon, 2000) and fewer years in special education for preschoolers at risk of developmental and learning difficulties (Miedel & Reynolds, 1999).

Hoover-Dempsey and Sandler’s model of PI (Walker, Wilkins, Dallaire, Sandler, & Hoover-Dempsey, 2005) posits that PI, including home-based PI, is influenced by parents’ perceptions of invitations to become involved from the school and teacher. We examined three variables that represented school invitations for involvement as potential moderators of associations between family background characteristics and PI: communication from the HS center about child progress, HS centers’ support of parent–school cooperation, and teacher support provided to parents.

PI for Low-Income Families

Economic and sociodemographic factors common to families in HS may influence PI (Anderson & Minke, 2007), though some of this research has focused on parents’ school- rather than home-based involvement, and associations between background characteristics of parents of children with disabilities and home-based involvement have seldom been studied. Relevant factors include marital status as lower levels of home-based PI are reported by single parents (Fantuzzo, Tighe, & Childs, 2000). Low-income parents are also likely to work full-time, perhaps in more than one job, to meet financial demands. Working parents have been found to volunteer less frequently in HS classrooms (Castro, Bryant, Peisner-Feinberg, & Skinner, 2004) possibly due to time scarcity that reduces time for involvement (Weiss et al., 2003). Other research suggests that relatively higher educated parents engage in more school-based involvement, though parental education was not related to home-based involvement (Fantuzzo et al., 2000). In a sample of African American HS parents, economic and neighborhood stressors were associated with less home- and school-based involvement among low-income parents (Waanders, Mendez, & Downer, 2007).

Low-income parents may also be more at risk for experiencing depression, which may be linked to lower levels of PI. Mothers of kindergarteners in a primarily low-income White and African American sample who reported greater symptoms of depression engaged in fewer home learning activities (Kohl, Lengua, & McMahon, 2000). Alternatively, Lamb-Parker et al. (2001) found that while nearly half of HS mothers in their sample reported frequent symptoms of depression, symptoms were not related to levels of school-based PI per mother and teacher report. More evidence is needed to clarify the relations between parental characteristics and home-based involvement, including whether relations differ for young children with and without disabilities.

Another consideration is the racial and ethnic diversity of HS. The majority (66%) of HS parents are from non-White racial backgrounds (Office of Head Start, 2016). While there are differences within and between parents from different racial backgrounds, research suggests parents from minority backgrounds may experience unique challenges to involvement when compared with parents from non-Hispanic, White backgrounds. Kim’s (2009) review identified several barriers to school-based PI that appear to occur more often for racial minority parents than for White parents, such as teacher’s perceptions about the efficacy and capacity of minority parents, school friendliness, and positive communication. More specifically, Latino and African American parents may have expectations and conceptualizations of what constitutes PI that differ from predominantly White school staff, in addition to potentially experiencing an imbalanced power dynamic with school personnel (Cooper, 2009; Guerra & Nelson, 2013). For example, Guerra and Nelson’s (2013) review of more than 20 years of research of working-class Latino parents revealed that Latino parents may not view it is part of their role to intervene in the school setting because they view teachers as the experts in this domain. The extent that race and ethnicity play a role specifically in HS parents’ home-based involvement is less clear though, given the diversity of the HS population, the racial and ethnic backgrounds of parents is an important consideration.

The interplay of factors influencing PI for low-income families is complex. However, when low-income parents are involved when their children are young, this can provide a boost to children’s academic outcomes and have a potentially even greater impact than PI activities (e.g., volunteering at school, having educational resources at home, telling stores, parent–child art projects) for their higher income peers (Penner, 2018). The Office of Head Start recognizes this crucial opportunity and thus includes PI as a major tenet of HS programs (CFR § 1302.50). In fact, evidence suggests that HS parents spend more time engaging in at-home learning activities than at-risk parents whose children do not attend HS (Gelber & Isen, 2013).

PI for Preschoolers With Disabilities

Despite findings indicating similar benefits from school-based PI for preschoolers with specific disabilities (e.g., Calderon, 2000) and parental involvement in specific treatment protocols (e.g., Corcoran & Dattalo, 2006), limited research has examined what factors are associated with home-based PI for young children with disabilities. Identifying factors that are related to lower or higher levels of home-based PI is of particular importance among low-income families of preschoolers with disabilities such as those served by HS. In addition to the risks associated with being from a low-income background, the learning, social/emotional, or behavioral challenges of preschoolers with disabilities puts them at additional risk for school failure, which could be potentially mitigated by parents’ efforts to support learning through everyday learning opportunities.

Research on children with disabilities suggests similar trends in associations between parental background characteristics and parents’ involvement in children’s education, though these studies largely focus on specific groups (e.g., students with autism) and/or older, elementary-age students. Among elementary students with disabilities, research suggests lower levels of school-based PI among non-White parents (Frew, Zhou, Duran, Kwok, & Benz, 2013) and lower rates of participation in Individualized Education Program (IEP) meetings for parents with lower levels of education (Jones & Gansle, 2010). Parents’ school-based involvement also appears to be more frequent among higher versus lower socioeconomic status (SES) families as well as married or cohabitating versus single parents (Frew et al., 2013). Notably, in a sample of 95 mothers of children ages 3 to 7 years diagnosed with autism spectrum disorder, Benson, Karlof, and Siperstein (2008) found higher SES to be related to home- but not school-based involvement. The number of hours worked by mothers in this study was not related to any type of educational involvement, though connections between parental employment status and school-based PI are indicated by studies of children without disabilities (e.g., Castro et al., 2004). Maternal depression has not been explicitly studied in relation to PI for children with disabilities, though elevated levels of depressive symptoms are well-documented among parents of young children with a broad range of disabilities (e.g., Alvarez, Meltzer-Brody, Mandel, & Beeber, 2015) and depressive symptoms have been shown to dampen effects of parent-focused intervention such as parent training for children aged 2.5 to 6.5 years with behavioral concerns (Dempsey, McQuillin, Butler, & Axelrad, 2016).

Parent–Teacher Relationships and PI

HS’s framework emphasizes two-way communication and information sharing between parents and program staff as well as collaboration with families to identify unique needs and strengths (CFR § 1302.50). Parent–school relationships and interactions influence PI at home and in school (Anderson & Minke, 2007). Moreover, teachers’ attitudes favoring PI relate to their efforts to involve parents (Swick & McKnight, 1989) even when parents face multiple potential barriers to involvement (e.g., being single, less educated, employed; Epstein & Dauber, 1991). Studies of elementary, middle, and high school students indicate that parents’ decisions to become involved at school are influenced by their perceptions of the teacher and school’s communications and attitudes about their involvement (Overstreet, Devine, Bevans, & Efreom, 2005), though this association is less clear for home-based involvement.

Parent–school relationships may be particularly influential in PI for children with disabilities. Given and the complex needs of children with disabilities, parents may more frequently seek consultation and feedback from educators to foster educationally enriching activities at home. However, extant research has again focused more so on school- rather than home-based involvement. In a representative sample of 9,747 elementary students with disabilities, frequency of school communication was associated with higher levels of school-based PI regardless of primary disability category (Frew et al., 2013). Interviews with parents of preschoolers exhibiting behavioral and/or social-emotional concerns conducted by Koivunen, Alst, Ocasio, and Allegra (2017) likewise highlight the importance of parent–teacher connections in gaining buy in from parents to participate in the child’s treatment program. In a sample of children with autism, the degree to which school staff provided opportunities, supported, and encouraged involvement was associated with PI at home and at school (Benson et al., 2008).

Three research questions were addressed in the current study:

Research Question 1 (RQ1): What sociodemographic and parent characteristics are associated with PI for children with disabilities when they enter HS, and do they differ from that of children without disabilities?

Research Question 2 (RQ2): Does PI at the beginning of the first HS year as well as sociodemographic and parent characteristics relate to PI at the end of the first year of HS for children with disabilities, and are these associations different from that of children without disabilities?

Research Question 3 (RQ3): Are associations between sociodemographic and parent characteristics and PI at the end of the first HS year moderated by aspects of parent–school interactions (i.e., parent–school cooperation, school communication, and teacher support) for children with and without disabilities?

Method

Sample and Participants

Sampling design

The FACES 2009 Cohort study used a complex multistage sampling design (see the FACES user’s guide for sampling procedures; Malone et al., 2013) to create a nationally representative sample of 3- and 4-year-old children participating in HS. Data were collected in fall 2009 through spring 2011 (4-year cohort) or 2012 (3-year cohort). The complete data set included 3,149 children.

Because FACES data were collected using a complex, multistage sampling design rather than a traditional simple random sample, specific analytical procedures were used to avoid biased parameter estimates and incorrect standard errors due to sampling design (Hahs-Vaugh, McWayne, Bulotsky-Shearer, Wen, & Faria, 2011). The design-based approach estimates a single-level model using sampling weights as well as cluster and stratum design variables to account for the sampling design. Based on the recommendation of the FACES user’s guide (Malone et al., 2013), a design-based approach was deemed the best method for addressing the complex sampling design in the current study.

The design-based approach was employed using the SURVEYREG procedure available in the software program SAS 9.3 (SAS Institute, 2011). In the analysis, the stratum (Strata = STRAT) and primary sampling unit (cluster = PSU; HS programs) variables were specified to adjust the standard errors to account for multistage sampling. In addition, the child longitudinal weight variable (weight = PRA12WT) was applied to adjust parameter estimates for differential probabilities of selection and response in the child-level outcome variables.

Empirical sample

The current study used data from the 3-year-old cohort (fall 2009 and spring 2010, T1 and T2, respectively). A total of 1,849 children were assessed in 2009 and 1,659 were assessed in spring 2010. See Table 1 for weighted and unweighted sample statistics.

Table 1.

Descriptive Statistics.

Variable	Weighted			Unweighted
Variable	n	M	SE	n	M	SD
T1 number of parent activities	1,645	11.39	0.07	1,838	11.34	2.05
T2 number of parent activities	1,519	11.86	0.07	1,530	11.81	1.90
T1 crime witness/victim index	1,631	0.75	0.05	1,820	0.75	1.24
T1 parent depression	1,629	5.00	0.21	1,819	5.09	6.09
T2 teacher support (parent)	1,499	3.83	0.01	1,510	3.83	0.54
T2 school communication (parent)	1,501	1.17	0.01	1,512	1.16	0.42
T2 cooperation (teacher)	1,670	4.36	0.04	1,928	4.34	0.69
	Frequency		Percentage	Frequency		Percentage
T1 poverty status
Not below threshold	639		37.61	709		38.57
Below poverty threshold	1,006		62.39	1,129		61.43
T1 child has IEP or IFSP
Yes	94		5.98	103		5.67
No	1,534		94.02	1,713		94.33
T1 child’s gender
Female	828		49.25	932		49.34
Male	867		50.75	957		50.66
T1 mother’s race
White, non-Hispanic	396		25.86	459		24.31
Minority	1,298		74.14	1,429		75.69
T1 mother’s employment status
Not employed	803		53.08	914		53.11
Employed	737		46.92	807		46.89
T1 mother’s education level
Less than high school degree	522		33.87	587		33.26
High school degree/GED/some college	947		50.07	1,061		66.74
Bachelor’s or higher	115		7.06	117		6.63
T1 single parent status
Not a single parent	777		49.90	863		48.95
Single parent	800		50.10	900		51.05

Note. Estimates were weighted using the PRA12WT variable and standard errors are calculated for the weighted values. T1 = fall 2009; T2 = spring 2010; IEP = Individualized Education Program; IFSP = Individualized Family Service Plan; GED = General Educational Development.

Missing data

The primary sources of missing data were on the outcome variable only (T2 PI; N = 156). FACES sampling weights account for selection into the sample, attrition, and participant nonresponse (i.e., missing data at the instrument level; Malone et al., 2013). Although weights were adjusted for missing data prior to the data being released for large-scale analysis, the current study evaluated whether there was potential for bias due to attrition. Specifically, we evaluated whether demographic and child outcome variables at T1 significantly predicted which cases had missing data at T2. Results from analyses using SURVEYLOGISTIC in SAS 9.3 showed that none of the demographic variables in fall 2009 were significantly related to whether or not the outcome variable was available in spring 2010.

Measures

Home-based PI

PI at home was measured using items drawn from the broadly used National Household Education Survey (NHES; National Center for Education Statistics [NCES], 2003) and Home Observation for Measurement of the Environment–Short Form (HOME-SF; Bradley, Corwyn, McAdoo, & García Coll, 2001). Parents reported whether educationally enriching activities were completed with the child in the week prior to the parent interview (variable PnPWkAc2). A total of 13 yes/no items were included. Parents indicated whether they told stories; taught letters, words, or numbers; taught songs or music; worked on arts and crafts; played a game, sport, or exercised; took the child along during errands like the post office or bank; involved the child in household chores; talked about what happened in HS; talked about TV or videos; played counting games like singing songs with numbers; played a board or card game; played with blocks; and/or counted different things with the child. Item-level data were not available, thus total scores were included for T1 and T2.

Parent–school relationships

A series of indicators measured parent–school relationships via interviews with parents and HS teachers. The first item evaluated teachers’ perceptions about the degree to which the HS center promoted parent–school cooperation (Cooperation). Teachers responded using a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Parents rated communication from the HS program regarding children’s progress (School Communication) on a 3-point scale ranging from 1 (does it very well) to 3 (doesn’t do it at all) and how frequently they felt the HS teacher supported them as parents (Teacher Support) on a 4-point scale ranging from 1 (never) to 4 (always). All parent–school relationship variables were measured at the second time point.

Child disability status

Parent report of whether the child currently had an IEP in the fall (T1) was used as a measure of child disability status.

Sociodemographic and parent predictors

Sociodemographic factors included neighborhood stress (i.e., number of times parents reported being a witness/victim of crime), poverty status, and minority status. Additional variables included child age and maternal employment status, marital status, education, and depression (Center for Epidemiological Studies–Depression [CES-D] short form score). All predictor variables were measured at the first time point.

Data Analysis

The outcome variable for our analyses was parental involvement at the end of prekindergarten (T2; spring 2010). Predictors included the sociodemographic and parent background variables, PI at the beginning of prekindergarten (T1; fall 2009), and the three moderating variables measuring parent–school relationships (Cooperation, School Communication, and Teacher Support). Prior to modeling, we evaluated the unweighted data for heteroscedasticity and multicollinearity and our results showed no evidence of either.

The first and second models evaluated the main effects of predictor variables on parental involvement at T1 and T2, respectively. The third model, used to evaluate our third research question, included interactions between parent–school relationship, parent background, and sociodemographic variables. Interactions were examined using a top-down approach (Ryoo, 2011). We began with the most complex model that included all possible interactions of interest and iteratively removed interactions that were nonsignificant (p > .05). In the final model, significant effects were evaluated for local effect size, using Cohen’s f² and following the procedures described in Selya, Rose, Dierker, Hedeker, and Mermelstein (2012). “Small” effects were f² ≥ 0.02 and < .15, “medium” effects were f² ≥ 0.15 and < .35, and “large” effects were f² ≥ 0.35 (Cohen, 1992).

All variables were centered to aid in parameter interpretation. Continuous variables were centered at the sample average. Discrete variables were coded so the reference group was comprised of non-Hispanic, White unemployed non–single parent mothers, who were living at or below the poverty threshold and had less than a high school education, whose female children completed the child assessments. Based on this method, estimated regression effects can be interpreted as the change in the predicted outcome (parental involvement) for every one-unit change in a particular variable, holding all other variables constant at their centering point.

Results

To evaluate the consistency of effects across children with and without disabilities, parallel analyses based on these groups were conducted using the DOMAIN statement in SURVEYREG. This procedure permits simultaneous parallel analyses across groups and adjusts variances across subpopulations (SAS Institute, 2011). Post hoc simple slopes tests were used to probe significant interactions, controlling for family-wise error using the Bonferroni–Holm procedure (Holm, 1979). The three p values for the simple slopes tests were ordered from smallest to largest and multiplied by 3, 2, and 1, respectively. These adjusted p values (p_BH) were sequentially compared with .05 until the first nonsignificant test (i.e., p_BH > .05) was identified.

RQ1: Sociodemographic and Parent Variables Related to PI at HS Entry

Table 2 provides the unstandardized parameter estimates for this model. For children without disabilities (n = 1,445), there were significant main effects for mother’s race (B = −0.57, p < .01, f² = .01), neighborhood crime/victimization (B = 0.12, p < .03, f² = .00), and education (overall f² = .02; less than high school degree: B = 0.64, p < .03; high school degree/General Educational Development [GED]/some college: B = 0.59, p < .01). Thus, PI was lower among mothers who were minorities and higher for mothers who experienced less neighborhood crime/victimization and had less than a high school education. For children with disabilities (n = 84), there was a significant main effect for education (overall f² = .04; less than high school degree: B = 1.22, p < .05), such that mothers who had not completed high school were more involved.

Table 2.

Sociodemographic and Parent Variables Related to PI at Head Start Entry and After 1 Year of Head Start.

Parameter	No IEP				IEP
Parameter	Estimate	SE	p	f ²	Estimate	SE	p	f ²
RQ1
Intercept	11.33	0.24	<.01		11.00	0.49	<.01
T1 mother’s race (minority)	−0.57	0.16	<.01	.01	0.33	0.38	.40
T1 child’s gender (male)	−0.02	0.12	.90		0.29	0.36	.80
T1 parent depression	−0.01	0.01	.12		0.04	0.03	.17
T1 crime witness/victim index	0.12	0.05	<.05	.00	−0.04	0.16	.75
T1 poverty status (below threshold)	0.07	0.13	.59		−0.31	0.40	.44
T1 mother’s employment status (employed)	−0.10	0.13	.43		0.07	0.34	.83
T1 mother’s education level (Ref.—college degree)				.02				.04
Less than high school degree	0.64	0.27	<.05		1.22	0.60	<.05
High school degree/GED/some college	0.59	0.17	<.01		0.44	0.55	.43
T1 single parent status (single parent)	0.19	0.10	.07		0.29	0.39	.47
RQ2
Intercept	11.90	0.14	<.01		9.31	1.21	<.01
T1 number of parent activities	0.53	0.02	<.01	.49	0.29	0.15	.06
T1 mother’s race (minority)	−0.19	0.11	.09		0.62	0.46	.18
T1 child’s gender (male)	−0.13	0.09	.16		0.56	0.49	.26
T1 parent depression	−0.00	0.00	.67		−0.01	0.03	.76
T1 crime witness/victim index	0.02	0.04	.51		0.20	0.16	.23
T1 poverty status (below threshold)	0.00	0.09	.99		0.42	0.51	.41
T1 mother’s employment status (employed)	0.08	0.11	.44		−0.07	0.44	.88
T1 mother’s education level (Ref.—college degree)				.01
Less than high school degree	0.04	0.17	0.81		1.68	1.28	.19
High school degree/GED/some college	0.24	0.09	<.01		1.61	1.06	.13
T1 single parent status (single parent)	−0.15	0.09	.12		0.55	0.44	.21

Note. RQ1 = correlates at Head Start entry. RQ2 = correlates after 1 year of Head Start. Estimates were weighted using the PRA12WT variable. N = 1,445 (non-IEP) and N = 84 (IEP). Continuous variables were mean centered. The reference group for categorical variables was non-Hispanic, White females with a college education or higher, living above the poverty threshold in a two parent family. “Small” effects are defined as f² ≥ 0.02 to < .15, “medium” effects defined as f² ≥ 0.15 to < .35, and “large” effects defined as f² ≥ 0.35 (Cohen, 1992). Shaded rows indicate coefficients statistically significant at p < .05. PI = parent involvement; IEP = Individualized Education Program; GED = General Educational Development.

RQ2: Background Variables, PI at Entry, and PI After 1 Year of HS

Table 2 provides the unstandardized parameter estimates for the second model. In the non-IEP group (n = 1,271), there were significant main effects for PI at the beginning of the HS year (B = 0.53, p < .01, f² = .49) and education (overall f² = .01; high school degree/GED/some college: B = 0.24, p < .01). PI at the beginning of the year positively related to PI at the end of the year. Involvement was also higher among mothers who completed high school, a GED, or some college. There were no significant main effects for the IEP group (n = 79), thus none of the background characteristics or entry level PI were related to PI at the end of the year.

RQ3: Moderation of Parent–School Relationships in Associations Between Background Variables and PI

Teacher support

Table 3 contains the unstandardized parameter estimates for the model results. There were several significant main effects for the IEP group (n = 76), which are displayed in Table 3. There were also several significant interactions. Teacher support moderated the association between education and PI (overall f² = .07; high school degree/GED/some college: B = −1.62, p < .01. ESTIMATE statements and simple slopes tests failed to produce consistent values for the highest education group due to low sample sizes. The simple slopes of teacher support on PI were −0.59 (p_BH > .05) and 1.11 (p_BH < .01) at low and medium levels of education, respectively. Parents whose children had IEPs and who completed high school/some college were more likely to engage in PI as teacher support increased compared with similar parents who did not.

Table 3.

Moderation of Parent–School Relationships in Associations Between Background Variables and PI.

Parameter	No IEP				IEP
Parameter	Estimate	SE	p	f ²	Estimate	SE	p	f ²
Intercept	11.92	0.14	<.01		8.62	1.08	<.01
T1 number of parent activities	0.53	0.02	<.01	.50	0.42	0.09	<.01	.25
T1 mother’s race (minority)	−0.19	0.11	0.09		0.61	0.36	.09
T1 child’s gender (male)	−0.13	0.09	0.16		0.46	0.41	.27
T1 parent depression	0.53	0.02	<.01		−0.02	0.02	.30
T1 crime witness/victim index	0.00	0.01	0.77		0.35	0.15	<.05	.15
T1 poverty status (below threshold)	0.02	0.04	0.52		1.00	0.47	<.05	.11
T1 mother’s employment status (employed)	0.00	0.10	1.00		−0.33	0.42	.44
T1 mother’s education level (Ref.—college degree)				.01				.37
Less than high school degree	−0.09	0.16	0.56		2.65	1.20	<.05
High school degree/GED/some college	0.23	0.09	0.01		2.02	1.02	.05
T1 single parent status (single parent)	−0.14	0.09	0.14		1.23	0.41	<.01	.08
T2 cooperation	0.08	0.08	0.36		−0.30	0.29	.31
T2 school communication	0.17	0.21	0.41		−1.72	0.65	<.05	.21
T2 teacher support	0.14	0.16	0.38		1.02	0.27	<.01	.18
Depression × School Communication	−0.01	0.01	0.49		0.18	0.04	<.01	.14
Crime Witness/Victim × Teacher Support	0.03	0.06	0.65		−0.54	0.16	<.01	.07
Single Parent Status × Teacher Support	−0.14	0.16	0.38		−3.32	0.78	<.01	.03
School Communication × Education								.07
School Communication × Education (no high school)	−0.12	0.25	0.64		0.48	0.80	.55
School Communication × Education (high school/some college)	−0.03	0.27	0.92		1.93	0.75	<.05
Teacher Support × Education				.00				.04
Teacher Support × Education (no high school)	1.11	0.40	0.01		−1.62	0.47	<.01
Teacher Support × Education (high school/some college)	0.19	0.18	0.29		.	.	.

Note. Estimates were weighted using the PRA12WT variable. Estimates are not provided for the High School/Some College × Teacher Support interaction due to insufficient data. Continuous variables were mean centered. The reference group for categorical variables was non-Hispanic, White females with a college education or higher, living above the poverty threshold in a two parent family. “Small” effects are defined as f² ≥ 0.02 to < .15, “medium” effects defined as f² ≥ 0.15 to < .35, and “large” effects defined as f² ≥ 0.35 (Cohen, 1992). Shaded rows indicate coefficients statistically significant at p < .05. PI = parent involvement; IEP = Individualized Education Program; GED = General Educational Development.

Teacher support also moderated associations in the IEP group between single parent status and PI (B = −3.32, p < .01, f² = .03). Simple slopes of single parent status on PI were −2.09 (p_BH < .01) and 4.55 (p_BH < .01) at low and high values of teacher support, respectively. Single parents whose children had IEPs were less likely to engage in PI as teacher support increased. Coparenting parents were more likely to engage in PI as teacher support increased. An association between neighborhood stress and PI was also moderated by teacher support in the IEP group (B = −0.54, p < .01, f² = .07). Simple slopes of neighborhood stress as measured by the crime/victimization index on PI were −0.19 (p_BH > .05) and 0.88 (p_BH < .01) at high and low values of teacher support, respectively. Parents who witnessed/were victim to low levels of crime were more likely to engage in PI when teacher support increased compared with similar parents who witnessed/were victim to high levels of crime.

Significant main effects for the non-IEP group (n = 1,243) are displayed in Table 3. Teacher support moderated the association between parent education and PI (overall f² < .00; less than high school degree B = 1.11, p < .01). Simple slopes for each education group were calculated by recoding the data and conducting three separate analyses in which each group served as the comparison group (Aiken, West, & Reno, 1991). The simple slopes of teacher support on PI were 1.25 (p_BH < .01) and 0.14 (p_BH >.05) at low and high levels of education, respectively. Parents who did not complete high school were more likely to engage in PI as teacher support increased compared with similar parents who completed a high school degree/some college and college degree or higher.

School communication

For the IEP group, school communication moderated associations between maternal depression and PI (B = 0.18, p < .01, f² = .14). Simple slopes of depression on PI were 0.07 (p_BH < .01) and −0.12 (p_BH < .01) at high and low values of school communication, respectively. Parents who reported high levels of depression were more likely to engage in PI when school communication was reported as high, compared with similar parents who reported low levels of depression. School communication also moderated associations between education and PI (f² = .07; high school degree/GED/some college: B = 1.93, p < .05). Parents who did not complete high school were less likely to engage in PI when communication was low, compared with similar parents who completed high school or some college. ESTIMATE statements and simple slopes tests failed to produce consistent values for the highest education group due to low sample sizes. The simple slopes of school communication on PI were −1.24 (p_BH < .01) and −1.72 (p_BH < .01) at low and medium levels of education, respectively.

Cooperation

There were no main effects for parent–school cooperation on PI in the model, thus interaction effects were not tested for this variable.

Discussion

This study investigated associations between family and sociodemographic characteristics and home-based PI for HS children with disabilities compared with those without, and the influence of parent–school interactions in these associations. Findings add to the scant literature available regarding relations between family background, parent–school interactions, and home-based PI in education for a vulnerable group: low-income children with disabilities.

Study findings highlight how parent–school interactions (i.e., parental perceptions of teacher support and communication from the HS program) help explain the relationship between familial background and involvement at home in the education of young children with disabilities. Consistent with Hoover-Dempsey and Sandler’s model of PI (Walker et al., 2005), associations between background characteristics and PI for children with an IEP varied when parental perceptions of teacher support and communication were considered. Mothers of children with disabilities who completed high school reported more involvement as support from the HS teacher increased, compared with similar mothers who did not complete high school. Parents of children with disabilities with higher education may reap greater benefits from teacher support than those with less education, given the need for more specialized approaches to supporting the learning of their children across at home. Other findings suggest this is also the case for school-based PI; parents with more than a high school education are more active participants in IEP meetings than those who have completed high school or less (Jones & Gansle, 2010).

It should be noted that these results do not account for the many possible barriers that may affect PI among relatively less educated parents, such as higher poverty levels (Fantuzzo et al., 2000). It is vital that teachers ensure that efforts to involve parents are accessible to parents regardless of education level (e.g., considering reading level of materials provided to parents) and that other potential barriers are taken into account, such as the provision of information about low-cost or free enriching activities that parents can access regardless of financial burdens.

Results also indicated that mothers of children with disabilities who were single were less likely to be involved as teacher support increased, compared with mothers of children with disabilities who were coparenting or married. Single mothers may have perceived increased levels of support from the HS teacher as a sign that the child was receiving adequate support in school, and allocated their time to other family activities or household demands as they were the only caregivers available in the household. Likewise, Grolnick, Benjet, Kurowski, and Apostoleris (1997) found that teacher attitudes about school-based PI had a reduced positive effect at more difficult compared with more optimal familial context (i.e., life stress, family resources). Grolnick et al. did not include a measure of home-based PI, thus the current findings add to the notion that there may be diminishing returns on teachers’ efforts to foster PI as parents face increasingly difficult life stressors.

Findings also suggest that parents of children with disabilities who infrequently witnessed or fell victim to crime were more likely to be involved as teacher support increased, compared with similar parents who more often witnessed or fell victim to crime. Parents of children with disabilities face more stressors and therefore may be more vulnerable to the effects of neighborhood risks, leaving fewer psychological and time resources for engaging in educationally focused activities.

Another moderator, school communication, buffered negative associations between home-based PI and maternal depression as well as lower maternal education for children with disabilities. Given the deleterious effects of maternal depression on children’s social (Alvarez et al., 2015), developmental, and cognitive (Petterson & Albers, 2001) outcomes, it is noteworthy that effective school communication may help increase involvement among depressed mothers. Moreover, results suggest that through receipt of information from the HS program, mothers who may have had limited knowledge about how to support learning may have been better able to identify opportunities to engage their children in educationally focused activities at home.

Overall, parent perceptions of parent–school interactions altered more associations between background variables and home-based PI for children with disabilities compared with those without. This trend suggests that parental opinions regarding how well they are being supported by and receiving information from the child’s HS program is particularly essential to foster this type of educational involvement among parents of young low-income children with disabilities. Of note, teacher perceptions of how well the HS program promoted cooperation between parents and the program did not moderate any associations. It is possible that parental rather than teacher perceptions are relatively more influential in a parent’s decision to seek and engage in activities to promote learning at home; though prospective, experimental research is needed to explore this notion among young children with disabilities.

Some background variables were consistently associated with home-based PI for children with and without disabilities. For instance, lower involvement was found among Black and Hispanic families at the beginning but not the end of the HS year. Somewhat similarly, Hindman, Miller, Froyen, and Skibbe (2012) found lower levels of home-based involvement among Hispanic/Latino mothers at the beginning of HS, but higher involvement at the end of the year, compared with White, Black, Asian, Native American, and multiracial parents. Previous studies support the notion that Latino and African American families may have ideas of how to be involved that are different from schools’ conceptualizations (Cooper, 2009; Guerra & Nelson, 2013). The increase in involvement over the HS school year may reflect an increasingly mutual understanding between families and HS staff about how to encourage learning in the home setting. Home-based involvement was also higher for relatively less educated mothers of children when they entered HS, consistent with past research focused on school-based involvement (Castro et al., 2004). However, for children without disabilities, mothers with higher levels of education reported more involvement at the end of the year. For children with disabilities in HS, participation in programming may foster PI equally among parents, reducing maternal education-level differences in PI by year’s end.

Limitations and Future Research

First, the correlational nature of analyses prevents conclusions about causal relationships between PI and family and sociodemographic characteristics. However, our results highlight the need for experimental work that may speak to causal relationship between these factors. Second, data from any longitudinal study, particularly one in which a large proportion of children are at risk due to economic disadvantage, are likely to be missing due to nonresponse and attrition. Although the sample had missing data, the use of survey design variables such as the sampling weights mitigated potential concerns related to generalizability and nonresponse bias. Third, the difference in sample size between the IEP and non-IEP groups may have contributed to some instability in the model due to the relatively smaller sample size of children in the IEP group. Yet, findings regarding the IEP group emphasize the need for additional, focused work.

Recommendations for HS and Early Childhood Special Educators

HS’s Family and Community Engagement standards urge HS staff to make PI a priority. The strategies emphasize the need for coordinated efforts toward family engagement across HS teachers, home visitors, and family services staff as well as integrated strategies to engage families (CFR § 1302.50). Recent research suggests positive yield from these efforts; participation in HS has been associated with greater PI including engaging in home literacy and math activities as well as taking children to cultural activities (Gelber & Isen, 2013).

Recommendations and resources are available through the Office of HS to provide guidance in the implementation of these standards. Programs may assess the extent that they are meeting the needs of diverse families, including families of children with disabilities, through the Integrating Strategies for Program Progress self-assessment (https://eclkc.ohs.acf.hhs.gov/sites/default/files/pdf/integrated-strategies-for-program-progress-ppce.pdf). Gaps in parent engagement efforts may be addressed through home visits and meetings with parents. For instance, HS educators can assist parents in establishing learning goals for families that are consistent with family preferences, language, culture, and other unique characteristics. The process of getting to know families and their needs also assists HS programs in tailoring programming to be responsive to families’ preferences and interests. HS programs are further encouraged to provide information such as parenting education through a partnership approach in which parents’ knowledge and perspectives are incorporated. Finally, HS educators should assist families in obtaining support from mental health consultants or other community agencies to address significant issues that challenge parent–child relationships and hinder parental involvement efforts.

For all early childhood special educators, a partnership approach is beneficial in that it recognizes parents’ role as their children’s first teachers and provides parents the opportunity to learn about how to best support children’s learning. This approach is particularly important when working with parents of children with disabilities who may experience additional barriers to involvement such as depression, lower education, and single parenthood. For more information on these strategies, please see the Families as Learners and Families as Lifelong Educators outcomes at https://eclkc.ohs.acf.hhs.gov/family-engagement/article/understanding-family-engagement-outcomes-research-practice-series.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Aiken

L. S.

West

S. G.

Reno

R. R.

(1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: SAGE.

Alvarez

S. L.

Meltzer-Brody

Mandel

Beeber

(2015). Maternal depression and early intervention: A call for an integration of services. Infants & Young Children, 28, 72–87. doi:10.1097/IYC.0000000000000024

Anderson

K. J.

Minke

K. M.

(2007). Parent involvement in education: Toward an understanding of parents’ decision making. The Journal of Educational Research, 100, 311–323.

Benson

Karlof

K. L.

Siperstein

G. N.

(2008). Maternal involvement in the education of young children with autism spectrum disorders. Autism, 12, 47–63. doi:10.1177/1362361307085269

Bor

Sanders

M. R.

Markie-Dadds

(2002). The effects of the triple P-positive parenting program on preschool children with co-occurring disruptive behavior and attentional/hyperactive difficulties. Journal of Abnormal Child Psychology, 30, 571–587. doi:10.1023/A:1020807613155

Bradley

R. H.

Corwyn

R. F.

McAdoo

H. P.

García Coll

(2001). The home environments of children in the United States part I: Variations by age, ethnicity, and poverty status. Child Development, 72, 1844–1867.

Calderon

(2000). Parental involvement in deaf children’s education programs as a predictor of child’s language, early reading, and social-emotional development. Journal of Deaf Studies and Deaf Education, 5, 140–155.

Castro

D. C.

Bryant

D. M.

Peisner-Feinberg

E. S.

Skinner

M. L.

(2004). Parent involvement in Head Start programs: The role of parent, teacher and classroom characteristics. Early Childhood Research Quarterly, 19, 413–430. doi:10.1016/j.ecresq.2004.07.005

Cohen

J. E.

(1992). Power primer. Psychological Bulletin, 112, 155–159.

10.

Coolican

Smith

I. M.

Bryson

S. E.

(2010). Brief parent training in pivotal response treatment for preschoolers with autism. Journal of Child Psychology and Psychiatry, 51, 1321–1330. doi:10.1111/j.1469-7610.2010.02326.x

11.

Cooper

C. W.

(2009). Parent involvement, African American mothers, and the politics of educational care. Equity & Excellence in Education, 42, 379–394. doi:10.1080/10665680903228389

12.

Corcoran

Dattalo

(2006). Parent involvement in treatment for ADHD: A meta-analysis of the published studies. Research on Social Work Practice, 16, 561–570. doi:10.1177/1049731506289127

13.

Dempsey

McQuillin

Butler

A. M.

Axelrad

M. E.

(2016). Maternal depression and parent management training outcomes. Journal of Clinical Psychology in Medical Settings, 23, 240–246. doi:10.1007/s10880-016-9461-z

14.

Epstein

J. L.

Dauber

S. L.

(1991). School programs and teacher practices of parent involvement in inner-city elementary and middle schools. The Elementary School Journal, 91, 289–305. doi:10.1086/461656

15.

Fantuzzo

Tighe

Childs

(2000). Family Involvement Questionnaire: A multivariate assessment of family participation in early childhood education. Journal of Educational Psychology, 92, 367–376. doi:10.1037/0022-0663.92.2.367

16.

Frew

L. A.

Zhou

Duran

Kwok

O.-M.

Benz

M. R.

(2013). Effect of school-initiated parent outreach activities on parent involvement in school events. Journal of Disability Policy Studies, 24, 27–35.

17.

Gelber

Isen

(2013). Children’s schooling and parents’ behavior: Evidence from the Head Start impact study. Journal of Public Economics, 101, 25–38. doi:10.1016/j.jpubeco.2013.02.005

18.

Grolnick

W. S.

Benjet

Kurowski

C. O.

Apostoleris

N. H.

(1997). Predictors of parent involvement in children’s schooling. Journal of Educational Psychology, 89, 538–548.

19.

Guerra

P. L.

Nelson

S. W.

(2013). Latino parent involvement: Seeing what has always been there. Journal of School Leadership, 23, 424–455.

20.

Hammer

C. S.

Farkas

Maczuga

(2010). The language and literacy development of Head Start children: A study using the Family and Child Experiences Survey database. Language, Speech, and Hearing Services in Schools, 41, 70–83. doi:10.1044/0161-1461(2009/08-0050)

21.

Head Start Program Performance Standards, 45 C.F.R. Chapter XIII § 1302.50 (2016).

22.

Hahs-Vaughn

D. L.

McWayne

C. M.

Bulotsky-Shearer

R. J.

Wen

Faria

A. M.

(2011). Methodological considerations in using complex survey data: An applied example with the Head Start Family and Child Experiences Survey. Evaluation Review, 35, 269-303.

23.

Hindman

A. H.

Miller

A. L.

Froyen

L. C.

Skibbe

L. E.

(2012). A portrait of family involvement during Head Start: Nature, extent, and predictors. Early Childhood Research Quarterly, 27, 654–667. doi:10.1016/j.ecresq.2011.11.002

24.

Holm

(1979). A simple sequentially rejective multiple test procedure. Scandinavian Journal of Statistics, 6, 65–70.

25.

Improving Head Start for School Readiness Act, 42 U.S.C. § 9801 (2007).

26.

Jones

B. A.

Gansle

K. A.

(2010). The effects of a mini-conference, socioeconomic status, and parent education on perceived and actual parent participation in individual education program meetings. Research in the Schools, 17(2), 23–38.

27.

Kim

(2009). Minority parental involvement and school barriers: Moving the focus away from deficiencies of parents. Educational Research Review, 4, 80–102.

28.

Kohl

G. O.

Lengua

L. J.

McMahon

R. J.

(2000). Parent involvement in school: Conceptualizing multiple dimensions and their relations with family and demographic risk factors. Journal of School Psychology, 38, 501–523. doi:10.1016/S0022-4405(00)00050-9

29.

Koivunen

Alst

D. V.

Ocasio

Allegra

(2017). Understanding engagement in mental health services for preschool children: An analysis of teacher, clinician, and parent perspectives. Early Childhood Education Journal, 45, 313–320. doi:10.1007/s10643-016-0799-5

30.

Lamb-Parker

Piotrkowski

C. S.

Baker

A. J. L.

Kessler-Sklar

Clark

Peay

(2001). Understanding barriers to parent involvement in Head Start: A research-community partnership. Early Childhood Research Quarterly, 16, 35–51. doi:10.1016/S0885-2006(01)00084-9

31.

Malone

Lepidus Carlson

Aikens

Moiduddin

Kopack Klein

West

. . . Rall

(2013). Head Start Family and Child Experiences Survey (FACES): 2009 Cohort [United Stages] user guide (ICPSR 34558). Washington, DC: Administration for Children and Families, U.S. Department of Health and Human Services.

32.

Miedel

W. T.

Reynolds

A. J.

(1999). Parent involvement in early intervention for disadvantaged children: Does it matter? Journal of School Psychology, 37, 379–402. doi:10.1016/S0022-4405(99)00023-0

33.

National Center for Education Statistics. (2003). National Household Education Surveys Program 2001 electronic codebook and data file (NCES 2003-078) Washington, DC: U. S. Department of Education. National Center for Education Statistics.

34.

Office of Head Start. (2016). Head Start program facts: Fiscal year 2016. Retrieved from https://eclkc.ohs.acf.hhs.gov/about-us/article/head-start-program-facts-fiscal-year-2016

35.

Overstreet

Devine

Bevans

Efreom

(2005). Predicting parental involvement in children’s schooling within an economically disadvantaged African American sample. Psychology in the Schools, 42, 101–111. doi:10.1002/pits.20028

36.

Penner

E. K.

(2018). Early parenting and the reduction of educational inequality in childhood and adolescence. The Journal of Educational Research, 111, 213–231. doi:10.1080/00220671.2016.1246407

37.

Petterson

S. M.

Albers

A. B.

(2001). Effects of poverty and maternal depression on early child development. Child Development, 72, 1794–1813.

38.

Ramani

G. B.

Rowe

M. L.

Eason

S. H.

Leech

K. A.

(2015). Math talk during informal learning activities in Head Start families. Cognitive Development, 35, 15–33. doi:10.1016/j.cogdev.2014.11.002

39.

Roberts

Jergens

Burchinal

(2005). The role of home literacy practices in preschool children’s language and emergent literacy skills. Journal of Speech Language and Hearing Research, 48, 345–359. doi:10.1044/1092-4388(2005/024)

40.

Ryoo

J. H.

(2011). Model selection with the linear mixed model for longitudinal data. Multivariate Behavioral Research, 46, 598–624.

41.

SAS Institute. (2011). SAS/STAT® 9.3 User’s Guide. Cary, NC: Author.

42.

Schick

A. R.

(2014). Home-school literacy experiences of Latino preschoolers: Does continuity predict positive child outcomes? Journal of Applied Developmental Psychology, 35, 370–380. doi:10.1016/j.appdev.2014.05.006

43.

Selya

A. S.

Rose

J. S.

Dierker

L. C.

Hedeker

Mermelstein

R. J.

(2012). A practical guide to calculating Cohen’s f2, a measure of local effect size, from PROC MIXED. Frontiers in Psychology, 3, Article 111. doi:10.3389/fpsyg.2012.00111

44.

Sheridan

S. M.

Knoche

L. L.

Edwards

C. P.

Bovaird

J. A.

Kupzyk

K. A.

(2010). Parental engagement and school readiness: Effects of the Getting Ready intervention on preschool children’s social-emotional competencies. Early Education and Development, 21, 125–156. doi:10.1080/10409280902783517

45.

Swick

K. J.

McKnight

(1989). Characteristics of kindergarten teachers who promote parent involvement. Early Childhood Research Quarterly, 4, 19–29. doi:10.1016/S0885-2006(89)90038-0

46.

Waanders

Mendez

J. L.

Downer

J. T.

(2007). Parent characteristics, economic stress and neighborhood context as predictors of PI in preschool children’s education. Journal of School Psychology, 45, 619–636. doi:10.1016/j.jsp.2007.07.003

47.

Walker

J. M. T.

Wilkins

A. S.

Dallaire

J. R.

Sandler

H. M.

Hoover-Dempsey

K. V.

(2005). Parental involvement: Model revision through scale development. The Elementary School Journal, 106, 85–104.

48.

Weiss

H. B.

Mayer

Kreider

Vaughan

Dearing

Hencke

Pinto

(2003). Making it work: Low-income working mothers’ involvement in their children’s education. American Educational Research Journal, 40, 879–901. doi:10.3102/00028312040004879