Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2007 Dec 14.
Published in final edited form as: School Psych Rev. 2006;35(4):645–662.

Ethnicity and Language Contributions to Dimensions of Parent Involvement

Shuk Wa Wong 1, Jan N Hughes 1
PMCID: PMC2139999  NIHMSID: NIHMS35316  PMID: 18084632

Abstract

This study examined ethnic and language group differences on dimensions of parent-rated and teacher-rated parent involvement after adjusting for the influence of family socioeconomic factors. A total of 179 teachers and 481 parents provided information on parent school involvement for a sample of ethnically and linguistically diverse first-grade children attending one of three school districts in Texas. Four groups were examined: White, Black, Hispanic-English speaking, and Hispanic-Spanish speaking. Exploratory and confirmatory factor analysis supported four parent-reported involvement dimensions (positive perceptions about school, communication, parent-teacher shared responsibility, and parent school-based involvement) and three teacher-reported dimensions (alliance, general parent involvement, and teacher initiation of involvement). Data generally supported the hypothesized ethnic and language group differences in parent involvement and the moderating effect of dimension of parent involvement on group differences. Implications for school psychologists are discussed.

Despite the effort of policy makers and educators to close the achievement gap between racial and ethnic minority students and Euro-American students, significant discrepancies remain. For example, national data indicate that in 1999, 18% of Black students and 13% of Hispanic or Latino students in kindergarten through twelfth grade had repeated at least one grade. These retention rates are higher than the 9% of White students who had repeated a grade (National Center for Education Statistics, 2003). Ethnic differences were also apparent in suspension or expulsion and high school completion rates. In 1999, 35% of seventh- to twelfth-grade Black students had been suspended or expelled for disciplinary reasons at some point in their school years, compared to 20% of Hispanic students and 15% of Whites (National Center for Education Statistics, 2003). In 2003, 88% of Black students and 62% Hispanic students completed high school, significantly lower than the 94% of Whites (National Center for Education Statistics, 2005).

For the past two decades, studies have provided convincing evidence that parents make significant contributions to their children's school outcomes (for review, see Fan & Chen, 2001; Jeynes, 2003). Results indicate that when parents participate at school and encourage or assist learning at home, children tend to be more successful at all grade levels. Specifically, parent participation in education is associated with increased student achievement, better school attendance, increased achievement motivation, reduced dropout rate, better emotional adjustment, and improved social behavior and interactions with peers (Fan & Chen, 2001; Hill et al., 2004; Izzo, Weissberg, Kasprow, & Fendrich, 1999; Marcon, 1999). Furthermore, the benefits of parent involvement in their children's schooling accrue to all ethnic groups (Jeynes, 2003).

Although empirical studies offer strong support for the view that parent involvement is associated with higher student achievement, it is important to acknowledge that the relationship between parent involvement and student achievement is most likely bidirectional (Englund, 2001). That is, parents adjust their involvement in response to student achievement levels. Also, parent involvement may be manifested differently at different ages and in different cultural contexts, and both age and context may moderate the effect of parent involvement on achievement (Hill et al., 2004). Thus, conclusions about the role of parent involvement may not be equally applicable for all groups.

Much educational research has examined why some parents become involved in their children's education and others do not (Grolnick, Benjet, Kurowski, & Apostoleris, 1997; Hoover-Dempsey & Sandler, 1997). Much of this research has investigated the role of race and ethnicity in parent involvement, with varying results. On one hand, Black and Hispanic parents, when compared with their White counterparts, are found to have less contact with their children's schools (Floyd, 1998). Ethnic minority status is associated with teacher ratings of lower levels and quality of parent involvement (Hill et al., 2004; Kohl, Weissberg, Reynolds, & Kasprow, 1994). Teachers and principals tend to attribute lower levels of parent involvement among ethnic minority parents to a lack of motivation to cooperate, a lack of concern for their children's education, and a lower value placed on education (Lopez, 2001).

Other research findings indicate that such attributions regarding parent involvement levels are erroneous. For example, a survey done by Chavkin and Williams (1993) of 682 Black, 506 Hispanic, and 1,779 Anglo parents across six southwestern states found that Black and Hispanic parents not only strongly agreed with the importance of being involved in their children's education but also expressed a strong interest in assuming various parent involvement roles (e.g., program supporter or home tutor). Although ethnic minority parents express a strong desire to be actively involved in their children's education, they are more likely than are majority parents to believe the school is responsible for initiating efforts and creating opportunities for parent involvement in school (Chavkin & Williams, 1993). Chavkin and Williams concluded that beliefs about responsibility for initiating parent involvement may explain why ethnic minority parents, relative to nonminority parents, participate less in school-based parent involvement activities (Chavkin & Williams, 1993). In addition, several researchers have identified contextual barriers to minority and low-income family involvement in school, including transportation, child care, and work schedules (see Boethel, 2003).

Several characteristics of the extant literature on ethnic and socioeconomic (SES) differences in parent involvement make it difficult to reach definitive conclusions (Fan & Chen, 2001; Fantuzzo, Tighe, & Childs, 2000). First, most of what we know about the role of parent involvement in student achievement is based on parents' or teachers' perceptions of parent involvement rather than observed differences in parent involvement behaviors. Second, inconsistencies in conceptualizing and measuring parent involvement render it difficult to integrate results across studies. Even when investigators examine the same aspect of parent involvement, they often measure it differently (Fan & Chen, 2001; Reynolds, 1991). Third, most existing measures of parent involvement are not psychometrically robust (Kohl, Lengua, McMahon, & The Conduct Problems Prevention Research Group, 2000). Fourth, most studies rely on single-reporter ratings. A generally low correspondence among informants of parent involvement renders it hard to integrate research findings provided by different sources (Reynolds, 1991). A related issue is the tendency for parents to inflate reports of parent involvement because they recognize such activities are desirable (Nord, Lennon, Liu, & Chandler, 1999). Fifth, most studies investigating ethnic differences in parent involvement have confounded ethnicity with other socioeconomic variables, such as parents' education level, parents' employment status, and family income. It is thus hard to separate the effect of ethnicity from that of SES (Hill, 2001). Last, levels of parent involvement are not at all uniform within or across minority groups. Regarding differences across minority groups, Hispanic parents, for example, often report lower levels of school involvement than Black parents (Griffith, 1998). Differences within ethnic groups are likely to be important, too. For example, parents subsumed under the label “Hispanic” or “Latino” are likely to differ on many variables that would be expected to influence parent involvement in school, such as acculturation and language. Attempts to integrate research findings collected from different ethnic minority groups under the name of minority parents will be misleading.

Recently, researchers have begun to investigate whether parent involvement may be manifested differently by different ethnic and SES groups (Eccles & Harold, 1996; Hill et al., 2004) Parents from different ethnic groups may define parent involvement differently and may demonstrate different levels of involvement depending on the type of involvement opportunity (e.g., home based vs. school based). Regarding overall level of parent involvement, White parents were higher than ethnic minority parents (particularly Black and Hispanic mothers; Zellman & Waterman, 1998). However, when specific dimensions of parent involvement were investigated, the results were more complex. Whereas White and Black parents reported greater involvement at school than did Hispanic parents (Sheldon, 2002; Steinberg, Lamborn, Dornbusch, & Darling, 1992), Black parents reported similar to or greater levels of involvement at home than did either White or Hispanic parents (Eccles & Harold, 1996; Ho & Willms, 1996; Sheldon, 2002; Watkins, 1997).

Although Hispanic parents endorse similar levels of importance to education and attitudes towards parent involvement as compared to White and Black parents (Chavkin & Williams, 1993; Tinkler, 2002), they often report the least involvement in school (Steinberg, Lamborn, Dornbusch, & Darling, 1992). Researchers consider the lack of English proficiency one of the major obstacles to some Hispanic parents' participation in their children's education (Tinkler, 2002). To examine the moderating effect of language on Hispanic parents' involvement in education, the Hispanic parents in this study were divided into two groups based on whether they predominantly spoke Spanish or English.

Purpose of This Study

The purpose of this study was to investigate ethnic group differences on parent-reported and teacher-reported parent involvement in school on different dimensions of involvement, controlling for the influence of family socioeconomic variables. Toward that purpose, the first study objective was to develop and test parent-report and teacher-report measures of relevant dimensions, or types, of parent involvement. The second study objective was to examine whether ethnic and language group differences were moderated by the dimension of parent involvement. It was hypothesized that White parents would demonstrate a significantly higher level of involvement than the other three groups (Black, Hispanic-English speaking, and Hispanic-Spanish speaking), especially on teacher-report measures. Black parents were expected to demonstrate a higher level of involvement than the two Hispanic parent groups, and Hispanic parents who spoke English were expected to demonstrate a higher level of involvement than Hispanic parents who spoke Spanish. Furthermore, group differences on parent-reported, home-based involvement were expected to be smaller than group differences on parent-reported, school-based involvement dimensions.

Method

Participants

Participants were parents and teachers of first-grade children attending one of three ethnically diverse school districts (1 urban, 2 small city) in central and southeast Texas. School District A (student population = 13,558) had an ethnic distribution of 38% White, 37% Hispanic, 25% Black, and 1% other. District B (student population = 24,429) had an ethnic distribution of 35% White, 30% Hispanic, 30% Black, and 5% other. District C (student population = 7,424) had an ethnic distribution of 67% White, 12% Hispanic, 12% Black, and 9% other. Participants were drawn from a larger sample of children participating in a longitudinal study examining the effect of grade retention on academic achievement. Participants were recruited across two sequential cohorts in first grade during the fall of 2001 and 2002. Children were eligible to participate in the larger study if they scored below the median score on a state-approved, district-administered measure of literacy administered either in May of kindergarten or September of first grade and had not been previously retained in first grade. However, only children whose ethnic background was Black, Hispanic, or White were eligible to participate in this study. Of 1,320 eligible children (25% Black, 37% Hispanic, 34% White), written parental consent was obtained for 750 (57%).

Children with and without parent consent for participation did not differ on age, gender, ethnicity, free or reduced-cost lunch status or literacy test scores. Children with consent were somewhat more likely to be limited English proficient (68%) than were children without consent [32%; Pearson χ2(l) =16.218, p < .001]. Questionnaires containing the parent involvement scales were sent to parents and teachers in the spring of first grade.

Parent respondents

Questionnaires were mailed to parents of first-grade students across each of the two recruitment years. A total of 481 (64%) parents returned completed questionnaires. Children of responding and nonresponding parents did not differ on age, gender, limited English proficiency status, free or reduced-cost lunch status, or district literacy test scores. Children of responding parents were somewhat less likely to be Black or Hispanic (33%) than children of nonresponding parents [67%; Pearson χ2(2) = 11.432, p =.003].

Of the 481 participating parents, 106 (22%) reported being Black, 163 (34%) being Hispanic, and 212 (44%) being White. Of 163 Hispanic parents, 71 (44%) spoke only Spanish or Spanish better than English, based on parent response to an acculturation item on the questionnaire. Table 1 provides descriptive statistics of the parent sample as a function of parent ethnicity. English-speaking Hispanic parents were younger than White parents, and English-speaking Hispanic parents obtained a higher score on acculturation than did Spanish-speaking Hispanic parents.

Table 1.

Descriptive Statistics of the Parent Sample as a Function of Parent Ethnicity

Variable White Black English-Speaking Hispanic Spanish-Speaking Hispanic All
Sample size 212 106 92 71 481
Mean age (SD) 36 (6.8) 33 (8.8) 31(6.4) 33 (5.8) 34 (7.2)
Relationship to child
 Mother 190(89.6%) 95 (89.6%) 77 (83.7%) 54(76.1%) 416(86.5%)
 Father 22(10.4%) 3(2.8%) 12(13.0%) 16(22.5%) 53 (11.0%)
 Grandparent and other - 8 (7.5%) 3 (3.3%) 1(1.4%) 12 (2.5%)
Education level
 Less than high school 1(0.5%) 5 (4.7%) 11(12%) 42(59.2%) 59(12.3%)
 At least high school 211 (99.5%) 100(94.3%) 81(88%) 28(39.4%) 420(87.3%)
Employment status
 Unemployed 52 (24.5%) 41(38.7%) 18 (19.6%) 17 (23.9%) 128 (26.6%)
 Employed 150 (70.8%) 62 (58.5%) 71(77.2%) 49 (69.0%) 332 (69.0%)
Mean acculturation (SD) 4.0 (.45) 2.4 (.63)
Open in a new tab

Teacher respondents

A total of 648 (86%) teacher questionnaires were received from 179 teachers. Children with and without complete data on teacher questionnaires did not differ on any demographic variables at baseline, with one exception. Children with complete data were somewhat more likely to qualify as limited English proficient (55%) than children without complete data (45%). The teacher sample was predominantly White (n = 147, 82%). The highest educational level was a bachelor's degree for 49% of teachers. Teachers' mean number of years of teaching experience was 8.7 (SD = 4.8).

Development and Testing of Measures of Parent Involvement in School

Conceptualizing parent involvement in school

Consistent with current conceptualizations (e.g., Epstein, 1995; Grolnick et al., 1997; Kohl et al., 2000), parent involvement in school refers to the efforts made by parents or primary caretakers that directly support the academic success of their children or administrative needs of their children's schools as well as perceptions of the quality of home-school interactions. Parent involvement practices can be initiated by parents or by teachers and can be based at home or in school. Different measures for parent and teacher report were developed based on previous literature indicating that parents and teachers are most reliable informants on different dimensions of parent involvement (Kohl et al., 2000; Reynolds, 1991). For example, teachers are typically not aware of parent-child interactions that occur in the home.

Parent report measure

The Parent Reported Involvement Measure was used to collect parents' self-report on their involvement in their children's education. The measure was initially derived from a pool of 32 items. Twenty-six items were drawn from the Parent-Teacher Involvement Questionnaire (Kohl et al., 2000), which covers four dimensions of parent-teacher partnership: teacher relationship quality factor, parent involvement, parent's endorsement of school, and parent-teacher contact, and which has demonstrated good construct validity. Six additional items were created to cover parents' perceived self-efficacy and teacher and parent roles, because role construction and self-efficacy perceptions influence parent involvement behaviors (Deslandes & Bertrand, 2005). An exploratory principal components factor analysis based on the first cohort of parents (n = 273) yielded a four-factor solution accounting for 49.7% of the variance that fit the theoretical model well. Confirmatory factor analysis conducted with parents of Cohort 2 (n = 337) using the maximum likelihood estimation method provided an adequate fit to the data [χ2 (333) = 542.36; p < .001; CFI = 0.92; Comparative Fit Index (CFI) root mean square error of approximation = 0.04]. The four scales were Positive Perceptions about School (α= .93), Communication (α = .72), Parent-Teacher Shared Responsibilities (α = .72), and Parent School-Based Involvement (α = .72). Appendix A provides factor loadings, means, and standard deviations for each item and eigenvalue, explained variance, and Cronbach's alpha for the four parent involvement factors of the Parent Reported Involvement Measure from the exploratory factor analysis.

Appendix A.

Factor Loadings, Means, and Standard Deviations for Each Item and Eigenvalue, Explained Variance, and Cronbach’s Alpha for the Four Factors of Parent-Reported Involvement Measure (n = 273)

Factor Loadings
Factor and Item Fl F2 F3 F4 M SD
Factor 1 (Positive Perceptions About School)
 13. Parent feels child’s teacher cares about child. .82 .33 .00 .00 4.22 0.95
 26. Child’s school is doing a good job of preparing children for their futures. .81 −. 18 .16 .17 4.36 0.79
 25. Parent has confidence in people at child’s school. .80 −.24 .15 .18 4.39 0.72
 23. Child’s school is a good place for child to be. .80 −. 26 .12 .19 4.53 0.66
 24. Staff at child’s school is doing good things for child. .79 −. 27 .12 .15 4.45 0.67
 16. Parents feels child’s teacher pays attention to parent’s suggestions. .76 .40 .00 .00 3.88 1.07
 15. Parent feels comfortable talking with child’s teacher about child. .75 .46 .00 .00 4.23 0.99
 12. Parent enjoys talking with child’s teacher. .74 .48 .10 .00 4.17 0.10
 14. Parent thinks child’s teacher is interested in getting to know parent. .71 .45 .00 .00 3.63 1.18
 11. Parent feels welcome to visit child’s school. .66 .00 .14 .00 4.36 0.86
Factor 2 (Communication)
 2. Child’s teacher has called parent. .00 .61 .13 .22 2.06 0.84
 17. Parent asks child’s teacher questions or make suggestions about child. .26 .61 .25 .00 3.72 1.10
 3. Parent has written child’s teacher. .00 .57 .17 .15 2.40 0.97
 1. Parent has called child’s teacher. .00 .55 .16 .32 2.22 0.93
 4. Child’s teacher has written parent. .00 .54 .00 .00 2.98 1.06
Factor 3 (Parent-Teacher Shared Responsibility)
 28. Parent is responsible for solving child’s learning problem at school. .10 .15 .70 .00 4.50 0.69
 30. Parent is responsible for solving child’s behavior problem at school. .15 .16 .67 .00 4.70 0.65
 21. Parent makes sure that child gets homework done. .00 .00 .62 .00 4.69 0.62
 27. Parent makes a difference in child’s success at school. .00 .12 .60 .00 4.56 0.79
 19. Parent helps child at home with subjects that child has difficulty. .00 .00 .56 .00 4.33 0.87
 32. Parent is prepared to help child. .00 −. 10 .53 .14 4.27 0.87
 29. Teacher is responsible for solving child’s learning problem at school. .15 .17 .46 −. 11 4.25 0.77
 31. Teacher is responsible for solving child’s behavior problem at school. .00 .15 .43 .00 3.93 0.94
Factor 4 (Parent School-Based Involvement)
 7. Parent has visited child’s school for a special event. .00 .21 .00 .79 2.50 0.79
 9. Parent has attended a parent-teacher conference. .00 .19 .00 .77 2.17 0.67
 8. Parent has been invited to attend a parent-teacher conference. .00 .00 .00 .69 2.30 0.68
 6. Parent has been invited to child’s school for a special event. .14 .12 .14 .59 2.68 0.80
 10. Parent has attended PTA or PTO meetings. .15 .00 .15 .48 1.61 0.78
 22. Parent volunteers at child’s school. .20 .23 .29 .44 2.32 1.32
Eigenvalue 6.15 3.33 3.28 3.15
Percentage explained variance 19.20 10.40 10.30 9.80
Cronbach’s alpha .93 .72 .72 .72
Open in a new tab

Note. PTA/PTO = Parent-Teacher Association or Parent-Teacher Organization.

Teacher report measure

The Teacher Reported Involvement Measure was used in this study to collect teachers' ratings on parent involvement in education. The measure was initially derived from a pool of 28 items. Twenty-one items were adapted from the Parent-Teacher Involvement Questionnaire-Teacher Report (Conduct Problems Prevention Research Group, 1995; Kohl et al., 2000), which covers four dimensions of parent-teacher partnership: teacher relationship quality factor, parent involvement, teachers' perceptions of parent's value of education, and parent-teacher contact. Seven items were adapted from the Joining Scale of the Parent-Teacher Relationship Scale-Teacher Form (Vickers & Minke, 1995) based on research suggesting that relationship quality may be more predictive of student achievement than parent involvement behaviors (Hughes, Gleason, & Zhang, 2005). An exploratory factor analysis based on the first cohort of teachers of 311 first-grade children yielded a three-factor solution that accounted for 57.31% of the variance. The three factors were Alliance (α = .90), General Parent Involvement (α = .85), and Teacher Initiation (α = .71). Confirmatory factor analysis conducted on questionnaires completed by teachers of children in the second cohort (n = 296) found that the model provided an adequate fit to the data, [χ2(167) = 394.38; p < .001; CFI = 0.94; root mean square error of approximation = 0.07]. Appendix B provides factor loadings, means, and standard deviations for each item and eigenvalue, explained variance, and Cronbach's alpha for the three parent involvement factors of the Teacher Reported Involvement Measure based on the exploratory factor analysis. A composite score for each of the parent-report and teacher-report parent involvement scales was computed as the mean of the item scores for each scale.

Appendix B.

Factor Loadings, Means, and Standard Deviations for Each Item and Eigenvalue, Explained Variance, and Cronbach’s Alpha for the Three Factors of Teacher-Reported Involvement Measure (n = 311)

Factor Loadings
Factor and Item F1 F2 F3 M SD
Factor 1 (Alliance)
 18. Teacher can talk to and feel heard by parent. .88 .17 .09 3.95 1.02
 20. Parent has shared goals with school. .88 .20 .02 3.94 1.06
 15. Mutual understanding. .87 .15 .03 3.97 0.96
 16. Similar expectations of child. .86 .15 .03 3.98 1.04
 14. Parent respects teacher. .83 .09 .01 4.21 0.80
 13. Teacher respects parent. .83 .12 −. 02 4.30 0.86
 21. Education important to child’s family. .82 .29 −. 05 3.99 1.04
 19. Teacher comfortable discussing child problems with parent. .77 .07 .16 4.21 0.90
 17. Parent interested in knowing teacher. .75 .34 .09 3.35 1.11
 12. Difficult communication (reversed coding). .68 .20 .01 1.97 1.13
 28. How often parent encourages child’s positive attitude toward education. .64 .50 −. 12 3.25 1.30
 27. How involved is parent in child’s education. .60 .51 −. 02 3.16 1.32
Factor 2 (General Parent Involvement)
 7. Parent has visited child’s school for a special event. .36 .75 −. 04 2.16 0.75
 26. How often parent volunteers at school. .32 .71 −.03 1.77 1.16
 5. Parent stopped by to talk to teacher. .32 .66 .14 2.32 0.92
 11. Parent has attended PTA/PTO meetings. .34 .66 .02 1.67 0.69
 1. Parent has called teacher. .17 .64 .20 2.05 0.83
 25. How often parent sends books or objects to class. .43 .64 .06 1.92 1.08
 24. How often parent asks or makes suggestions. .38 .60 .29 2.65 1.14
 3. Parent has written teacher. .12 .59 .08 2.23 0.79
 2. Teacher has called parent. −.19 .47 .36 2.39 0.71
 9. Parent has attended a parent-teacher conference. .21 .46 −. 02 1.94 0.46
 6. Parent has been invited to school for a special event. .04 .46 .23 2.66 0.64
 8. Parent has been invited to attend a parent-teacher conference. −.26 .29 .14 2.11 0.39
Factor 3 (Teacher Initiation)
 22. How often teacher tells parent when concerned. .12 .12 .91 3.66 1.10
 23. How often teacher tells parent when worried. .12 .08 .90 3.41 1.23
 4. Teacher has written parent. −.05 .25 .40 3.08 0.98
Eigenvalue 10.96 3.28 1.80
Percentage explained variance 30.34 18.97 8.00
Cronbach’s alpha .90 .85 .71
Open in a new tab

Note. PTA/PTO = Parent-Teacher Association or Parent-Teacher Organization.

Ethnic and Language Group Differences

Overview of design and variables

Parents indicated their ethnic membership on a questionnaire with the following choices: Native American, Asian, Black, Hispanic, White, and others. Only White, Black, and Hispanic parents were included in the analyses. Hispanic parents were further divided into English-speaking and Spanish-speaking groups based on the level of language proficiency they designated on the parent questionnaire. Parents who indicated speaking only English or English better than Spanish were assigned to the English-speaking group. Similarly, parents who indicated speaking only Spanish or Spanish better than English were assigned to the Spanish-speaking group. Four levels of ethnicity permits the testing of no more than three orthogonal contrasts (Stevens, 2002). Three orthogonal contrasts were used to identify sources of ethnic group differences. The first contrast compared majority parents with minority parents, whereas the second contrast compared Black parents with Hispanic parents. The last contrast compared English-speaking Hispanic parents with Spanish-speaking Hispanic parents.

SES

Parent education and employment status were entered as covariates in the analyses. Parents provided data on the highest educational and employment level of any adult in the household. Education was coded on an 8-point scale (1 = elementary school; 4 = GED; and 8 = PhD or equivalent) and employment status was coded on a 3-point scale (0= unemployed; 1 = employed part time; 2= employed full time).

Acculturation

Hispanic parents completed the Parent Acculturation Scale (Balcazar, Castro, & Krull, 1995), which covers language use, the ethnicity of one's social network, country in which one spent one's childhood, and pride in one's Hispanic heritage (α= .78).

Results

A one-way multivariate analysis of covariance (MANCOVA) examined the first hypothesis concerning ethnic group differences in parent involvement dimensions after controlling for parent education and parent employment status. A repeated measures analysis of covariance (ANCOVA) was used to examine the second hypothesis concerning the interaction effect of the parent involvement scale on parent ethnicity to account for ethnic differences in parent involvement. For both analyses, covariates included parent education and parent employment status. Three planned contrasts were used to determine if there were significant differences among the four ethnic groups. Presentation of the results is organized into four sections: (a) testing of assumptions, (b) relations between study variables, (c) differences in parent involvement by ethnic group, and (d) ethnic differences in parent involvement as moderated by type of involvement scale.

Testing of Assumptions

Because MANCOVA and repeated measures ANCOVA were selected to address the two research questions, the data set was examined for violations to essential assumptions associated with the application of these two statistical methods, including multivariate normality, outliers, homogeneity of variances and covariances, and sphericity. Results indicated that most of the essential assumptions associated with the application of ANCOVA and MANCOVA were met in this data set. As a correction to the violation of homogeneity of covariances and sphericity, the Pillai's trace significance test and the Greenhouse-Geisser epsilon were used to interpret the results of MANCOVA and repeated measures ANCOVA, respectively.

Relations Among Study Variables

Pearson product-moment correlation was used to explore the relations among the four parent-rated parent involvement sub-scales and the three teacher-rated parent involvement subscales, respectively. Significant positive correlations were found among the four parent-rated parent involvement scales (r's = .24–.38) and the three teacher-rated parent involvement scales (r = .12–.59). However, the magnitude of the associations was rather low, suggesting that the seven sub-scales were sufficiently independent to justify analyzing them separately (see Table 2 for details).

Table 2.

Pearson r Correlations Between Parent Involvement Subscales

Subscale 1 2 3 4 5 6
Parent ratings
 1. Positive Perceptions
 2. Communication .29**
 3. Shared Responsibility .34** .27**
 4. School-Based Involvement .27** .38** .24**
Teacher ratings
 5. Alliance .27** .11* .14** .23**
 6. General Parent Involvement .14** .23** .16** .30** .59**
 7. Teacher Initiation .04 .12* .08 .06 .12** .33**
Open in a new tab
*

p < .05.

**

p < .01.

A series of η2 and Pearson product-moment correlation tests was conducted to examine Mont-Reynaud the relations between parent ethnicity, parent education, and parent employment status. Parent ethnicity was significantly positively associated with parent education (η2 = .22) but was not significantly related to parent employment status (η2 = .01). Parent education was significantly related to parent employment status; however, the association was not strong (r = .11, p = .009; see Table 3).

Table 3.

Relations Between Parent Ethnicity, Parent Education, Parent Employment Status, and Parent Involvement Scales

Parent Ratings
Teacher Ratings
Variable Education Employment Status pp cm sr si al gi ti
Ethnicitya .22 .01 .01 .03 .09 .00 .04 .09 .00
Educationb .11** .07 .11* .20** .02 .11* .19** −. 03
Employment statusb −.04 .05 −. 12* −. 04 .04 −. 07 −. 07
Open in a new tab

Note. pp = Positive Perceptions about School; cm = Communication; sr = Parent-Teacher Shared Responsibility; si = Parent School-Based Involvement; al = Alliance; gi = General Parent Involvement. ti = Teacher Initiation.

a

Values represent η2

b

Values represent Pearson product-moment correlation coefficients.

*

p < .05.

**

p < .01.

A series of ,η2 and Pearson product-moment correlation tests was conducted to explore the associations between the seven parent involvement subscales and the independent (parent ethnicity) and covariates. Parent education was the only variable that was consistently statistically correlated with parent involvement subscales (r = .11−.20, p.≤ 05).

Differences in Parent Involvement by Ethnic Group

A series of MANCOVAs and univariate ANCOVAs with three planned orthogonal contrasts of parent ethnicity as the independent variables was conducted to determine if a main effect of ethnicity existed after parent education and parent employment status were statistically controlled. Results are reported by the source of ratings.

Parent ratings

In the first MANCOVA, the dependent variables consisted of positive perceptions about school, communication, parent-teacher shared responsibility, and parent school-based involvement. The independent variable, parent ethnicity, was entered in the model through three planned orthogonal contrasts using the Helmert approach. The first contrast compared majority (White) and minority (Black and Hispanic) parents. The second contrast compared Black and Hispanic parents. The third contrast compared English-speaking and Spanish-speaking Hispanic parents. Parent education and parent employment status were entered as covariates in the model.

Results of the first MANCOVA indicated a significant main effect of ethnicity for parent ratings of parent involvement [Pillai’s trace = 0.08, F(12, 1320) = 3.11, p < .001, η2 .03] after parent education and parent employment status were controlled. In the univariate analyses, significant main effects of ethnicity were obtained for communication [F(3, 441) = 4.61, p = .003, η2 = .03] and shared responsibility [F(3, 441) = 9.14, p < .001, η2 = .06]. Contrast results indicated that White parents reported a significantly higher level of parent-teacher shared responsibility than Black and Hispanic parents (p = .002). Black parents reported significantly higher levels of communication and parent-teacher shared responsibility than Hispanic parents (p ≤ .001). English-speaking Hispanic parents reported a significantly higher level of parent-teacher shared responsibility than Spanish-speaking Hispanic parents (p = .001). Table 4 shows the means and multivariate, univariate, and contrast results for three planned comparisons on parent-rated parent involvement scores.

Table 4.

Means and Multivariate, Univariate, and Contrast Results for Three Planned Comparisons on Parent-Rated Parent Involvement Scores

Means for the Groups
Subscale White (n = 202) Black (n = 98) Hispanic-English (n = 87) Hispanic-Spanish (n = 60)
Positive Perceptions 4.24 (0.751) 4.28 (0.639) 4.15 (0.719) 4.04 (0.574)
Communication 2.67 (0.563) 2.85 (0.803) 2.60 (0.584) 2.42 (0.759)
Shared Responsibility 4.47 (0.389) 4.47 (0.469) 4.32 (0.504) 4.00 (0.597)
School-Based Involvement 2.27 (0.452) 2.32 (0.603) 2.23 (0.544) 2.20 (0.597)
df M F p
Multivariate testa 12/1320 — 3.107 .000
Univariate tests
 Positive Perceptions 3/441 0.488 0.997 .394
 Communication 3/441 1.934 4.606 .003
 Shared Responsibility 3/441 1.864 9.142 .000
 School-Based Involvement 3/441 1.135 0.489 .690
Contrast Results (p Values)
Subscale Nonminority vs. Minority Black vs. Hispanic Hispanic-English vs. Hispanic-Spanish
Positive Perceptions .502 .101 .476
Communication .602 .001 .422
Shared Responsibility .002 .000 .001
School-Based Involvement .907 .247 .969
Open in a new tab

Note: Standard deviations are given in parentheses.

a

Pillai's trace test was applied.

Teacher ratings

In the second MANCOVA, only the dependent variable changed and consisted of the Alliance, General Parent Involvement, and Teacher Initiation of Involvement scales. Results of the second MANCOVA indicated a significant main effect of ethnicity for teacher ratings of parent involvement [Pillai’s trace = 0.12, F = (9,1425) = 6.52, p < .001, η2 = .04] after parent education and parent employment status were controlled. In the univariate analyses, significant main effects of ethnicity were obtained for alliance [F(3, 475)= 10.87, p <.001, η2 = .06] and general parent involvement [F(3, 475 = 13.43, p < .001, η2 = .081]. Contrast results indicated that White parents received higher ratings of general parent involvement than Black and Hispanic parents (p < .001) whereas Hispanic parents received higher levels of alliance and general parent involvement than Black parents (p < .001). Table 5 shows the means and multivariate, univariate, and contrast results for three planned comparisons on teacher-rated parent involvement scores.

Table 5.

Means and Multivariate, Univariate, and Contrast Results for Three Planned Comparisons on Teacher-Rated Parent Involvement Scores

Means for the Groups
Subscale White (n = 202) Black (n = 110) Hispanic-English (n = 92) Hispanic-Spanish (n = 72)
Alliance 3.83 (0.683) 3.39 (0.732) 3.80 (0.628) 3.70 (0.623)
General Parent Involvement 2.34 (0.512) 1.94 (0.424) 2.22 (0.533) 2.06 (0.540)
Teacher Initiation 3.49 (0.806) 3.48 (0.794) 3.51 (0.933) 3.31 (0.909)
df M F p
Multivariate testa 9/1425 - 6.516 .000
Univariate tests
 Alliance 3/475 4.814 10.873 .000
 General Parent
 Involvement 3/475 3.303 13.427 .000
 Teacher Initiation 3/475 0.938 1.317 .268
Contrast Results (p Values)
Subscale Nonminority vs. Minority Black vs. Hispanic Hispanic-English vs Hispanic-Spanish.
Alliance .343 .000 .732
General Parent Involvement .000 .000 .251
Teacher Initiation .268 .460 .072
Open in a new tab

Note: Standard deviations are given in parentheses.

a

Pillai's trace test was applied.

Ethnic Differences in Parent Involvement as Moderated by Type of Involvement Scale

Two repeated measures ANCOVAs with thiee planned orthogonal contrasts of parent ethnicity as the independent variable were conducted to determine if an interaction effect of ethnicity and scale existed after parent education and parent employment status were controlled. These results are reported by the source of ratings.

Parent ratings

In the first repeated measures ANCOVA, the within-subjects variable was scale and the between-subjects variable was parent ethnicity, which was entered through three planned comparisons involving comparison of majority (White) and minority (Black and Hispanic) parents, Black and Hispanic parents, as well as English-speaking and Spanish-speaking Hispanic parents, respectively. Parent education and parent employment status were entered as covariates in the model.

Results of the first repeated measures ANCOVA indicated significant main effects of scale [F(2.78, 1225.75) = 159.05, p < .001, η2 = .27] and ethnicity [F(3, 441) = 4.98, p =.002, η2 = .03] but a nonsignificant interaction effect of scale with parent ethnicity [F(8.34,1225.75) = 1.74, p = .082, η2 = .01] after parent education and parent employment status were controlled. Thus, parent ratings of the parent involvement subscales were significantly different and parents from different ethnic groups reported their involvement differently. However, ethnic differences in parent ratings of parent involvement were not moderated by the type of parent involvement. Table 6 shows the repeated measures ANCOVA results for parent ratings of parent involvement.

Table 6.

Analysis of Covariance for Parent Ratings of Parent Involvement

Source df F η2 p
Between Subjects
Education (Ed) 1 8.350 .019 .004
Employment (Em) 1 1.482 .003 .224
Ethnicity (Et) 3 4.977 .033 .002
Error 441 (.619)
Within Subjectsa
Scale (S) 2.779 159.054 .265 .000
S × Ed 2.779 1.248 .003 .291
S × Em 2.779 2.824 .006 .042
S × Et 8.338 1.737 .012 .082
Error (S) 1225.750 (.277)
Open in a new tab

Note: Values enclosed in parentheses represent mean square errors.

a

Due to violation of the sphericity assumption, the Greenhouse-Geisser epsilon was used to interpret the within-subjects results.

Teacher ratings

In the second repeated measures ANCOVA, the within-subjects variable was also scale, which consisted of Alliance, General Parent Involvement, and Teacher Initiation. The between-subjects variable was also parent ethnicity, which was entered through the same three planned comparisons used in the first repeated measures ANCOVA. Parent education and parent employment status were also entered as covariates in the model.

Results of the second repeated measures ANCOVA indicated significant main effects of scale [F(I.50, 713.93) = 71.95, p < .001,η2 = .13] and ethnicity [F(3, 475) = 7.31,p <.001, η = .04] as well as a significant interaction effect of scale on parent ethnicity [F(4.51, 713.93) = 5.52, p < .001, η2 = .03] after parent education and parent employment status were controlled. Thus, teacher ratings of the parent involvement subscales were significantly different and teachers differed significantly in their ratings of parents from different ethnic groups. In addition, ethnic differences in parent involvement as reported by teachers were moderated by the type of involvement. Contrast results indicated that the primary source of group differences in the teacher-rated parent involvement measure came from the difference between Black and Hispanic parents. Table 7 shows the repeated measures ANCOVA results for teacher ratings of parent involvement.

Table 7.

Analysis of Covariance for Teacher Ratings of Parent Involvement

Source df F η2 p
Between Subjects
Education (Ed) 1 5.016 .010 .026
Employment (Em) 1 2.993 .006 .084
Ethnicity (Et) 3 7.311 .044 .000
Error 475 (.740)
Within Subjectsa
Scale (S) 1.503 71.954 .132 .000
S × Ed 1.503 10.303 .021 .000
S × Em 1.503 1.358 .003 .255
S × Et 4.509 5.516 .034 .000
Error (S) 713.926 (.439)
Open in a new tab

Note: Values enclosed in parentheses represent mean square errors.

a

Due to violation of the sphericity assumption, the Greenhouse-Geisser epsilon was used to interpret the within-subjects results.

Discussion

The purpose of this study was to investigate ethnic group differences on different dimensions of parent-rated and teacher-rated parent involvement after adjusting for the influence of family socioeconomic factors, and the role of the dimension, or type, of involvement in moderating ethnic differences. A multidimensional measure of parent involvement is necessary to examine racial and ethnic differences on different types of involvement. This study contributes evidence that supports a multidimensional measure of parent-and teacher-reported parent involvement. After exploratory factor analysis on one cohort of students yielded theoretically and conceptually meaningful dimensions of parent involvement, confirmatory factor analysis with an independent sample confirmed the dimensional structure of the parent and the teacher versions. Importantly, the intercorrelations between different dimensions were in the low range, suggesting each dimension measures a distinct aspect of parent involvement. Also, in interpreting these findings it is important to remember that study measures of parent involvement assess perceptions of parent involvement and may not correspond with actual parent involvement behaviors.

Parent Perceptions

With respect to ethnic differences on dimensions of parent involvement, as predicted, Black parents reported more frequent communication with the school than did Hispanic parents. This finding is consistent with previous findings (Ritter, Mont-Reynaud, & Dornbusch, 1993). This finding may reflect that Black parents tend to take a more active approach in their children’s education (Fordham, 1996), whereas previous research reveals that Hispanic parents tend to be more deferential, more trusting, and less comfortable with teachers and schools (Ritter et al., 1993). With respect to beliefs about the respective roles of parents and teachers, White and Black parents reported more shared responsibility than did Hispanic parents; among Hispanic parents, those who spoke more English than Spanish reported more shared responsibility than did parents who spoke more Spanish than English. Spanish-speaking Hispanic parents reported a significantly lower level of shared responsibility than the other three parent groups. That is, the low ratings of Spanish-speaking Hispanic parents account for the largest portion of the between-group differences on parent-rated parent-teacher shared responsibility. This finding might be related to a possible lack of requisite language and instructional skills or familiarity with the American curriculum for Spanish-speaking Hispanic parents (most likely of immigrant status) who may have low efficacy for assisting with their children’s homework assignments, which may be presented in English (Floyd, 1998; Sosa, 1997).

The moderating analysis provides for a statistical test of whether the magnitude of the associations between ethnic and racial group and parent involvement differed statistically. Contrary to expectations, ethnic and racial differences in parent involvement were not moderated by the type of involvement scale for parent ratings. Thus one cannot conclude that ethnic groups differ in their pattern of responses across different scales.

Teacher Ratings

Statistically significant ethnic and racial differences were found for teacher ratings of alliance between parents and teachers and for teacher ratings of parent involvement activities, controlling for education and employment status. Teachers rated their alliance with Black parents lower than for White or Hispanic parents.

It is interesting to note that Black parents are viewed by teachers in this study as the least involved in home-school involvement activities of all four groups, whereas Black parents rate their level of communication with the school and school-based involvement as the highest of all four groups. The finding of low correspondence between parents and teachers is consistent with previous research (Reynolds, 1991) and suggests that parents and teachers are reporting on different aspects of their relationship or give different meaning to the same interactions. It is also possible that both teachers’ stereotypes as well as parent response biases contribute to the discrepancy.

A moderating effect for dimension on the magnitude of the association between ethnic and racial group and involvement was found for teacher ratings. Perhaps because teachers have greater knowledge of the various opportunities for parent involvement they are better able to discriminate among different aspects of parent involvement rather than responding in a more global fashion across dimensions.

Implications

Hispanic parents, especially Spanish-speaking parents, report low levels of communication with the school and a low sense of shared responsibility for their children’s education. Because parent involvement is predictive of achievement for Hispanic parents, as well as for Black and White parents, this finding is troubling. It is possible that a lower sense of parental responsibility for their child’s educational success may translate into less support for the child’s achievement, which may contribute to ethnic patterns of school avoidance that result in nonschool completion. Because this study is cross-sectional, one cannot infer causal associations between parent reports and student achievement. Longitudinal studies are needed to determine whether perceptions of responsibility for educational outcomes are predictive of achievement trajectories for different ethnic groups. Until such studies are conducted, it seems important for school psychologists to help teachers find ways to communicate to Hispanic parents the many ways in which they can help their young children succeed in school, including asking them about their school day, providing books at home, and providing a rich oral language environment.

Teachers report a lower level of alliance with Black parents than with White or Hispanic parents. This finding is of concern based on studies reporting that the quality of the home-school relationship is a stronger predictor of children’s social adjustment and achievement than are measures of parent school involvement behaviors (Hughes et al., 2005; Kohl et al., 1994; Rimm-Kaufman, Pianta, Cox, & Bradley, 2003). Positive connections between parents and teachers constitute social capital, defined as “a task specific construct that relates to the shared expectations and mutual engagement by adults in the active support and social control of children” (Sampson, Morenoff, & Earls, 1999, p. 635).

Several possibilities may explain the lower teacher-reported alliance for Black parents, but this study does not permit a testing of various possibilities. That fewer than 4% of study teachers were Black may contribute to the lower teacher report of alliance for Black parents. When parents do not share a common culture, it is more difficult to establish shared understanding and to build trust. National studies find that schools that serve larger percentages of Black parents are less likely to initiate attempts to involve parents (National Center for Education Statistics, 2006), raising the possibility that teacher and school factors contribute to differences in teachers’ perceptions of the involvement of Black and White parents.

These findings suggest an important role for school psychologists in helping teachers connect with minority parents and families as one means of narrowing the achievement gap between White and minority students. School psychologists can provide professional development programs that enhance teachers’ and administrators’ knowledge and skills for parent involvement. Based on a research synthesis of knowledge and skills that teachers need to work effectively with families, Shartrand, Weiss, Kreider, and Lopez (1997) provided a framework of content areas for teacher preparation. The content areas include (1) general family involvement, (2) general family knowledge, (3) home-school communication, (4) family involvement in learning activities, (5) families supporting schools, (6) schools supporting families, and (7) families as change agents. It is important to realize that teachers are likely to apply their skills in schools that promote broad family involvement policies (Rimm-Kaufman & Pianta, 2000). Thus, it is important for school psychologists to work at the school system level to implement broad-based, empirically supported parent involvement practices.

Limitations and Future Directions

The findings of the current study, while informative, are also limited in certain respects. First, because the study lacked a third source of information on parent involvement activities, such as student report or the use of “real-time” measures of involvement (e.g., daily phone interviews with parents and teachers conducted for random weeks through the year), one cannot compare the validity of either parent or teacher reports. In this regard, students have been found to be reliable reporters of parent academic involvement, at least in the middle grades (Epstein & Sanders, 2002; Reynolds, 1991).

Second, the sample was comprised of relatively low-achieving students. Even though the sample’s average age-based Broad Reading score on the Woodcock Johnson Achievement Test III was well within the average range (M = 96; SD = 17), the sample was not representative of the population of the schools in the study. Similarly, children with consent were somewhat less likely to be Black or Hispanic (33%) than children without consent (67%). Possible selection bias may affect the generalizability of the results.

Third, because SES was defined narrowly (i.e., the highest educational level in the home), the contribution of other elements of family background, such as income or adversity, that may be associated with ethnicity could not be investigated. Fourth, Black and Hispanic parents were less likely to complete the parent involvement questionnaire than were White parents. Because completion of the questionnaire might be indicative of parent involvement, the differential response rate may have distorted ethnic group differences in parent involvement that exist in the population.

Future research with these measures is necessary to establish their associations with objective measures of parent involvement. Recent longitudinal research has documented that ethnicity and SES may moderate the influence of parent involvement on future achievement (Dearing, McCartney, Weiss, Kreider, & Simpkins, 2004; Hill et al., 2004). Thus, future research should also explore the direction and magnitude of change in different dimensions of parent involvement as children progress in school and the longitudinal associations between different dimensions of parent involvement and child outcomes across and within different ethnic and language groups.

References

  1. Balcazar H, Castro G, Krull J. Cancer risk in Mexican American women: The role of acculturation, education, and health related risk factors. Health Education Quarterly. 1995;22(1):61–84. doi: 10.1177/109019819502200107. [DOI] [PubMed] [Google Scholar]
  2. Boethel M. National Center for Family and Community Schools, Southwest Educational Development Laboratory. Austin, TX: 2003. Diversity: School, family, and community connections:Annual synthesis. Retrieved February 18, 2005, from http://www.sedl.org/connections/resources/diversity-synthesis.pdf. [Google Scholar]
  3. Chavkin NF, Williams DL. Minority parents and the elementary school: Attitudes and practices. In: Chavkin NF, editor. Families and schools in a pluralistic society. Albany: SUNY Press; 1993. pp. 73–83. [Google Scholar]
  4. Conduct Problems Prevention Research Group. Technical reportsfor the constructdevelopmentfor the measures for Year 2 outcome analyses. 1995 Unpublished technical report. [Google Scholar]
  5. Dearing E, McCartney K, Weiss HB, Kreider H, Simpkins S. The promotive effects of family educational involvement for low-income children’s literacy. Journal of School Psychology. 2004;42:445–460. [Google Scholar]
  6. Deslandes R, Bertrand R. Motivation of parent involvement in secondary-level schooling. Journal of Educational Research. 2005;98:164–175. [Google Scholar]
  7. Eccles JS, Harold RD. Family involvement in children’s and adolescents’ schooling. In: Booth A, Dunn JF, editors. Family school links: How do they affect educationaloutcomes? Mahwah, NJ: Erlbaum; 1996. pp. 3–34. [Google Scholar]
  8. Englund MM. A path analysis of the relationship between parentalexpectations, parentalinvolvement, and children’s achievement. Paper presented at the biennial meeting of the Society for Research in Child Development; Minneapolis, MN. 2001. [Google Scholar]
  9. Epstein JL. School1family/community partnerships: Caring for the children we share. Phi Delta Kappan. 1995;76(9):701–712. [Google Scholar]
  10. Epstein JL, Sanders MG. Family, school, and community partnerships. In: Bornstein MH, editor. Handbook of parenting. Vol. 5: Practical issues in parenting. Mahwah, NJ: Erlbaum; 2002. pp. 407–437. [Google Scholar]
  11. Fan X, Chen M. Parental involvement and students’ academic achievement: A meta-analysis. Educational Psychology Review. 2001;13(1):1–22. [Google Scholar]
  12. Fantuzzo J, Tighe E, Childs S. Family involvement questionnaire: A multivariate assessment of family participation in early childhood education. Journalof Educational Psychology. 2000;92(2):367–376. [Google Scholar]
  13. Floyd L. Joining hands: A parental involvement program. Urban Education. 1998;33(1):123–135. [Google Scholar]
  14. Fordham S. Blacked out: Dilemmas of race, identity, and success at Capital High. Chicago: University of Chicago Press; 1996. [Google Scholar]
  15. Griffith J. The relation of school structure and social environment to parent involvement in elementary schools. Elementary School Journal. 1998;99(1):53–80. [Google Scholar]
  16. Grolnick WS, Benjet C, Kurowski CO, Apostoleris NH. Predictors of parent involvement in children’s schooling. Journal of Educational Psychology. 1997;89(3):538–548. [Google Scholar]
  17. Hill NE. Parenting and academic socialization as they relate to school readiness: The roles of ethnicity and family income. Journal of Educational Psychology. 2001;93(4):686–697. [Google Scholar]
  18. Hill NE, Castellino DR, Lansford JE, Nowlin P, Dodge KA, Bates JE, Pettit GS. Parent academic involvement as related to school behavior, achievement, and aspirations: Demographic variations across adolescence. Child Development. 2004;75:1491–1509. doi: 10.1111/j.1467-8624.2004.00753.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ho SE, Willms JD. Effects of parent involvement on eighth-grade achievement. Sociology of Education. 1996;69(2):126–141. [Google Scholar]
  20. Hoover-Dempsey KV, Sandler H. Why do parents become involved in their children’s education? Review of Education Research. 1997;67(1):3–42. [Google Scholar]
  21. Hughes JN, Gleason KA, Zhang D. Relationship influences on teachers’ perceptions of academic competence in academically at-risk minority and majority first grade students. Journal of School Psychology. 2005;43:303–320. doi: 10.1016/j.jsp.2005.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Izzo CV, Weissberg RP, Kasprow WJ, Fendrich M. A longitudinal assessment of teacher perception of parent involvement in children’s education and school performance. American Journal of Community Psychology. 1999;27(6):817–839. doi: 10.1023/a:1022262625984. [DOI] [PubMed] [Google Scholar]
  23. Jeynes W. A meta-analysis: The effects of parental involvement on minority children’s academic achievement. Education and Urban Society. 2003;35(2):202–218. [Google Scholar]
  24. Kohl GO, Lengua LJ, McMahon RJ The Conduct Problems Prevention Research Group. Parent involvement in school conceptualizing multiple dimensions and their relations with family and demographic risk factors. Journal of School Psychology. 2000;38(6):501–523. doi: 10.1016/S0022-4405(00)00050-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kohl GO, Weissberg RP, Reynolds AJ, Kasprow WJ. Teacher perceptions of parent involvement in urban elementary schools: Sociodemographic and school adjustment correlates; Paper presented at the annual meeting of the American Psychological Association; Los Angeles, CA. 1994. Aug, [Google Scholar]
  26. Lopez GR. The value of hard work: Lessons on parent involvement from an (im)migrant household. Harvard Educational Review. 2001;71(3):416–437. [Google Scholar]
  27. Marcon RA. Positive relationships between parent school involvement and public school inner-city preschooler’s developmental and academic performance. School Psychology Review. 1999;28:395–412. [Google Scholar]
  28. National Center for Education Statistics. Status and trends in the education of Blacks. 2003 Retrieved October 8, 2005 from http://nces.ed.gov/pubs2003/2003034.pdf.
  29. National Center for Education Statistics. Student effort and educational progress. 2005 Retrieved October 8, 2005, from http://nces.ed.gov/programs/coe/2005/section3/indicator23.asp.
  30. National Center for Education Statistics. Parents’ reports of school practices to provide information to families: 1996 and 2003. 2006 Retrieved February 16, 2006, from: http://www.nces.ed.gov/pubsearch/pubsinfo.asp?pubid=2006041.
  31. Nord CW, Lennon J, Liu B, Chandler K. Home literacy activities and signs of children’semerging literacy, 1993 and 1999 (NCES Publication No. 2000–026) Washington, DC: National Center for Education Statistics; 1999. Retrieved April 24, 2006 from: http://nces.ed.gov/pubsearch/pubsinfo.asp?pubid2000026. [Google Scholar]
  32. Reynolds AJ. Comparing measures of parental involvement and their effects on academic achievement. Early Childhood Research Quarterly. 1991;7:441–462. [Google Scholar]
  33. Rimm-Kaufman SE, Pianta RC. An ecological perspective on the transition to kindergarten: A theoretical framework to guide empirical research. Journal of Applied Developmental Psychology. 2000;21:491–511. [Google Scholar]
  34. Rimm-Kaufman SE, Pianta RC, Cox MJ, Bradley RH. Teacher-rated family involvement and children’s social and academic outcomes in kindergarten. Early Education and Development. 2003;14:179–198. [Google Scholar]
  35. Ritter PL, Mont-Reynaud R, Dornbusch SM. Minority parents and their youth: Concern, encouragement, and support for school achievement. In: Chavkin NF, editor. Families and schools in a pluralistic society. Albany: SUNY Press; 1993. pp. 107–119. [Google Scholar]
  36. Sampson RJ, Morenoff JD, Earls F. Beyond social capital: Spatial dynamics of collective efficacy for children. American Sociological Review. 1999;64:633–660. [Google Scholar]
  37. Shartrand AM, Weiss HB, Kreider HM, Lopez ME. New skills for new schools: Preparing teachers in family involvement. U.S. Department of Education; Washington, DC: 1997. Retrieved February 12, 2006 from www.gse.harvard.edu/hfrp/pubs/onlinepubs/skills/index.html. [Google Scholar]
  38. Sheldon SB. Parents’ social networks and beliefs as predictors of parent involvement. Elementary School Journal. 2002;102(4):301–316. [Google Scholar]
  39. Sosa AS. Involving Hispanic parents in educational activities through collaborative relationships. Bilingual Research Journal. 1997;21(2):1–8. [Google Scholar]
  40. Steinberg L, Lamborn SD, Dornbusch SM, Darling N. Impact of parenting practices on adolescent achievement: Authoritative parenting, school involvement, and encouragement to succeed. Child Development. 1992;63:1266–1281. doi: 10.1111/j.1467-8624.1992.tb01694.x. [DOI] [PubMed] [Google Scholar]
  41. Stevens J. Applied multivariate statisticsfor the social sciences. 4. Mahwah, NJ: Erlbaum; 2002. [Google Scholar]
  42. Tinkler B. A review of literature on Hispanicl Latino parent involvement in K-12 education. 2002 Retrieved October 9, 2005, from http://www.coe.uga.edu/clase/Ed_Resources/latinoparentreport.pdf.
  43. Vickers HS, Minke KM. Exploring parent-teacher relationships: Joining and communication to others. School Psychology Quarterly. 1995;10(2):133–150. [Google Scholar]
  44. Watkins TJ. Teacher communications, child achievement, and parent traits in parent involvement models. Journal of Educational Research. 1997;91(1):3–14. [Google Scholar]
  45. Wellborn JP, Grolnick WS. The Parenting Context Questionnaire. Unpublished manual; University of Rochester: 1988. [Google Scholar]
  46. Zellman GL, Waterman JM. Understanding the impact of parent school involvement on children’s educational outcomes. Journal of Educational Research. 1998;91(6):370–380. [Google Scholar]

RESOURCES