Children's Achievement in Early Elementary School: Longitudinal ...

Journal of Educational Psychology 2004, Vol. 96, No. 4, 723–730

Copyright 2004 by the American Psychological Association 0022-0663/04/$12.00 DOI: 10.1037/0022-0663.96.4.723

Children’s Achievement in Early Elementary School: Longitudinal Effects of Parental Involvement, Expectations, and Quality of Assistance Michelle M. Englund, Amy E. Luckner, Gloria J. L. Whaley, and Byron Egeland

This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

University of Minnesota, Twin Cities Campus In this prospective, longitudinal study, the authors examined the relations among parental behaviors, parental expectations, and children’s academic achievement. Participants were 187 low-income children and their mothers, studied from birth of the child through 3rd grade. Mothers’ quality of instruction prior to school entry had significant direct effects on IQ and indirect effects on achievement in 1st and 3rd grades. Parental expectations in 3rd grade had significant direct effects on parental involvement in 3rd grade. Children’s achievement in 1st grade had significant direct effects on parental involvement and expectations in 3rd grade. Parental involvement in 3rd grade had a significant direct effect on achievement in 3rd grade. Results suggest that early parenting factors are important for children’s academic achievement.

tion is not clear from contemporaneous measures, and some researchers (e.g., Shumow & Miller, 2001), when examining longitudinal data, have found that previous achievement predicts parental involvement rather than the opposite. Other researchers have reported mixed results (Deslandes, Royer, Turcotte, & Bertrand, 1997; Fan, 2001; Singh et al., 1995; Sui-Chu & Willms, 1996), including no evidence of a direct effect of parental involvement on children’s academic achievement (Bobbett, French, Achilles, & Bobbett, 1995; Keith, Reimers, Fehrmann, Pottebaum, & Aubey, 1986; Okpala, Okpala, & Smith, 2001) and even negative relations between these two variables (Deslandes et al., 1997; Sui-Chu & Willms, 1996). These discrepant results probably reflect, at least in part, varying definitions of parental involvement. Operational definitions of parental involvement used in research include communication between parents and teachers (Deslandes et al., 1997); parents’ communication with their children regarding school issues (Keith et al., 1986); number of hours parents volunteer in their children’s schools (Okpala et al., 2001); parental involvement in school activities, such as conferences (Bobbett et al., 1995; Izzo et al., 1999; Miedel & Reynolds, 1999; Stevenson & Baker, 1987); attendance at parent meetings organized by the school (Shaver & Walls, 1998); parental involvement in school-related activities with their children at home, for example, helping with homework (Shumow & Miller, 2001; Singh et al., 1995; Sui-Chu & Willms, 1996); and parental expectations regarding their children’s educational attainment (Keith et al., 1998). In a few studies, researchers have looked at parental involvement globally, merging various types of involvement into one variable (Keith et al., 1996, 1998; Miliotis, Sesma, & Masten, 1999), with substantial study-to-study variation in the measures included in such composites. Moreover, the inclusion of different dimensions in one composite parental involvement variable makes it difficult to determine which components of parental involvement are driving the association between parental involvement and academic achievement. When researchers have attempted to disentangle the various components of parental involvement (Fan, 2001; Fan & Chen,

Increasing parental school involvement has been an important part of local, state, and national education initiatives (e.g., Goals 2000; National Education Goals Panel, 1999). To implement these initiatives and increase academic success, many schools are making concerted efforts to reach out to parents and involve them in their children’s education. According to the National Center for Education Statistics (1998), most public elementary schools offer a range of opportunities for parental involvement, including open houses (97%), parent–teacher conferences (92%), and volunteer activities (90%). This focus on parental involvement has its roots in research demonstrating a positive association between parental involvement and academic achievement. Parental involvement, both at home and at school, is moderately but significantly related to children’s academic achievement (see, e.g., Christenson, Rounds, & Gorney, 1992; Fehrmann, Keith, & Reimers, 1987; Izzo, Weissberg, Kasprow, & Fendrich, 1999; Keith, Keith, Quirk, CohenRosenthal, & Franzese, 1996; Keith et al., 1998; Miedel & Reynolds, 1999; Shaver & Walls, 1998; Stevenson & Baker, 1987). This relation has been found fairly consistently regardless of the sample demographics (e.g., age, ethnicity, sex, or socioeconomic status [SES]) or measure of achievement (e.g., achievement tests, grade point average, grades). Parental involvement is an important predictor of children’s achievement in school. Even though some researchers have found a relation between parental involvement and achievement, the direction of this rela-

Michelle M. Englund, Amy E. Luckner, Gloria J. L. Whaley, and Byron Egeland, Institute of Child Development, University of Minnesota, Twin Cities Campus. Preparation of this work and the research described herein were supported by National Institute of Mental Health Grant R01 MH 40864. An earlier version of this article was presented in April 2001 at the biennial meeting of the Society for Research in Child Development, Minneapolis, Minnesota. Correspondence concerning this article should be addressed to Byron Egeland, Institute of Child Development, University of Minnesota, 51 East River Road, Minneapolis, MN 55455. E-mail: [email protected] 723


724

ENGLUND, LUCKNER, WHALEY, AND EGELAND

2001; Izzo et al., 1999; Shumow & Miller, 2001; Singh et al., 1995; Sui-Chu & Willms, 1996), the results have been somewhat conflicting. Both Izzo et al. (1999) and Sui-Chu and Willms (1996) found that parental involvement at home, rather than at school, was associated with higher levels of academic achievement. Shumow and Miller (2001) found the opposite, however; parental involvement at home was negatively related to achievement, whereas parental involvement at school was positively related to achievement. In addition, some researchers examining various components of parental involvement have found that parental expectations for their children’s educational attainment have the strongest relation to academic achievement compared with other types of parental involvement, such as parent– child communication and parent participation in school activities (Fan, 2001; Fan & Chen, 2001; Singh et al., 1995). Other researchers have looked specifically at the relation between parental expectations for their children’s educational attainment and children’s academic achievement (Gill & Reynolds, 1999; Goldenberg, Gallimore, Reese, & Garnier, 2001; Halle, Kurtz-Costes, & Mahoney, 1997; Okagaki & Frensch, 1998; Seginer, 1983). The results in this literature, however, are also contradictory. For instance, Halle et al. (1997) found in a sample of third and fourth graders that parental expectations predicted children’s achievement after controlling for earlier achievement. Goldenberg et al. (2001), on the other hand, found in a longitudinal study of kindergarten through sixth-grade children that children’s achievement predicted parental expectations but that parental expectations had no significant effects on children’s achievement over time. Parental expectations appear to be different from other definitions of parental involvement in that parental expectations are beliefs, whereas other measures of parental involvement focus on actual behaviors, such as attending parent–teacher conferences or helping with homework. It is possible that parents’ beliefs influence their behaviors, but it is important to distinguish between these two types of constructs when considering their influences on academic achievement. To further complicate the relation between these variables, other factors in addition to previous achievement may relate to both parental expectations for their children’s educational attainment and parental involvement in their children’s schooling. Parents’ educational level, for example, has been found to significantly predict both parental involvement (Keith et al., 1996, 1998; Shumow & Miller, 2001; Stevenson & Baker, 1987) and parental expectations (Gill & Reynolds, 1999; Singh et al., 1995). Those parents who have attained higher levels of education have higher expectations for their children’s educational attainment and are more involved in their children’s education (both at school and at home) than parents who have lower levels of education. Additionally, researchers have found that children’s achievement and intellectual abilities are affected by parent– child interactions prior to school entry, especially parents’ behaviors in problem-solving situations (Connell & Prinz, 2002; Pianta & Egeland, 1994; Pianta, Erickson, Wagner, Kreutzer, & Egeland, 1990; Pianta & Harbers, 1996; Pianta, Nimetz, & Bennett, 1997). For example, Pianta and Harbers (1996) found that parent– child interactions, as measured in a problem-solving task at children’s entry to kindergarten, explained a significant amount of variance in achievement test scores through fourth grade. Parents who scaffold learning experiences and provide assistance to their children when

needed early in children’s lives may prepare their children for school entry and provide a basis for them to benefit from educational activities. Obviously, the relation among parents’ actions, parents’ beliefs, and children’s behavior has not been clearly established. Although there has been some longitudinal research (e.g., Keith et al., 1996, 1998) detailing relations among these variables, involvement or expectations were measured at only one point in time. It is important, when attempting to understand relational processes, to measure not only achievement but also parental involvement and expectations across time. Longitudinal researchers who have included multiple time points have examined either parental involvement and achievement (see, e.g., Izzo et al., 1999) or parental expectations and achievement (see, e.g., Goldenberg et al., 2001) only, not all three factors in combination. The purpose of the current study was to examine longitudinally the interactional process that occurs over time among parental behavior (quality of instruction prior to school entry and parental involvement with school), parental expectations, and children’s academic achievement. Given the potential effects of both parental involvement and parental expectations on earlier achievement, it is important to examine these variables early in children’s lives. Some researchers have suggested that parental involvement and parental expectations may have a stronger influence on younger children’s achievement compared with that of children in the later years of elementary school through high school (Rogala, 2001; Singh et al., 1995). Furthermore, early experiences with parents in problem-solving situations prior to school entry may play an important role in later academic achievement. Early experiences may be especially important in that the scaffolding parents provide for their children in problem-solving situations may prepare children for later learning and facilitate children’s development of early adaptive skills, such as how to draw on resources to support them in solving problems. These possibilities necessitate examining the relations among the variables under consideration early in children’s schooling. In this study, it was hypothesized that (a) parental involvement with their children’s school, parental expectations of their children’s educational attainment, and children’s actual achievement are related; (b) parents’ educational level relates to both their actual behavior (quality of instruction and involvement with school) and their expectations for their children; (c) parents’ quality of instruction prior to school entry has direct effects on children’s IQ and achievement early in school; and (d) parental expectations have direct effects on parental involvement, parental involvement with school has direct effects on children’s academic achievement, and academic achievement has direct effects on both parental expectations and parental involvement in a cumulative process over time.

Method Participants The participants were drawn from an ongoing longitudinal study of children and their families. Participants were firstborn children (N ⫽ 187; 102 male, 85 female) and their mothers, who were recruited through Minneapolis, Minnesota, public health clinics during the third trimester of pregnancy. Mothers were all low income at the time of recruitment, with 59% single at the time of the birth of their children and 3% divorced,

CHILDREN’S ACHIEVEMENT IN EARLY ELEMENTARY SCHOOL separated, or widowed when their children were born. Approximately 66% of the children were Caucasian, 12% were African American, and 19% were mixed or other minority race (3% were missing father data). The age of the mothers at the birth of their children ranged from 12 years to 34 years (M ⫽ 20.63, SD ⫽ 3.75).

Procedures We examined data from birth through third grade, including interviews with children’s parents and teachers, laboratory observations with mothers and children, and intelligence tests with the children. Early elementary school data were gathered in first and third grades.


Measures Mother’s education. Each mother’s level of education was determined by asking the mothers, at the birth of their children, to identify the number of years of education they had completed. Mother’s quality of instruction. When the children were 42 months old, the quality of each mother’s instruction was assessed in a videotaped laboratory procedure in which mother-and-child pairs participated in a set of four developmentally appropriate problem-solving situations. Two independent coders viewing all four tasks rated the instructional behavior of all of the mothers on a 7-point scale, rating how well the mother structured the situation and coordinated her behavior to the child’s activity and needs for assistance. The scale ranged from 1, indicating poor quality of instruction (uninvolved and/or unstructured), to 7, indicating effective instruction throughout the session. This measure, used in a number of previous studies to assess various dimensions of early parent– child interactions, including quality of instruction, correlated with subsequent scores on standardized achievement tests (see, e.g., Connell & Prinz, 2002; Pianta & Egeland, 1994; Pianta et al., 1990, 1997; Pianta & Harbers, 1996). The rating scores for both observers were summed to form a composite rating (range ⫽ 2 to 14). Interrater reliability was determined by computing a single-measure intraclass correlation coefficient (absolute agreement) between the raters. The intraclass correlation for the final composite score was .87. Child’s IQ. Child’s IQ was assessed at 64 months of age using an abbreviated version of the Wechsler Preschool and Primary Scale of Intelligence (WPPSI; Wechsler, 1967). The total score of four subtests (vocabulary, comprehension, block design, and animal house) was used for data analysis. Each of the subtests correlates with the full scale IQ at vocabulary ⫽ .70, comprehension ⫽ .65, block design ⫽ .61, and animal house ⫽ .53 (Sattler, 1992; Wechsler, 1967). A higher correlation results when the four subtest scores are combined and correlated with the full scale (Wechsler, 1967). Additionally, Sattler (1992) reported that correlations between the WPPSI and Wechsler Intelligence Scale for Children—Revised (verbal, performance, and full scale; Wechsler, 1974) completed 11 years later ranged from .73 to .86. Parental expectations. Parental expectations were assessed at Grades 1 and 3. During a semistructured interview, parents were asked the openended question “How far do you think your child will go in school?” A 5-point scale to code parents’ responses ranging from will not complete high school to will go to graduate or professional school was developed. Such expected educational attainment has often been used as a measure of parental expectations in elementary as well as high school (see, e.g., Fan, 2001; Gill & Reynolds, 1999; Goldenberg et al., 2001; Halle et al., 1997; Seginer, 1983; Singh et al., 1995). The open-ended nature of the question allowed for variability in parents’ responses and therefore required some judgment on the part of the one coder. For purposes of reliability, a second independent rater was randomly assigned 16% of the sample for each grade. Interrater agreement was determined by computing a single-measure intraclass correlation coefficient of .78 for both grades combined. Parental involvement. Level of parental involvement with school was assessed at Grades 1 and 3 by interviews with each child’s teacher at each

725

grade level. In most cases, participants had different teachers each year. Interviews were scheduled during the spring quarter of the school year. Each year, teachers were asked (a) whether they knew the parents, (b) whether the parents demonstrated concern and interest in the child’s schoolwork, and (c) whether the parents participated in parent conferences or other meetings with the teacher. Two independent raters coded teachers’ responses on a 5-point scale, with 1 being not involved and 5 being very involved. Additionally, 25% of the sample for each grade, randomly selected, was coded for reliability. The intraclass correlation for both grades combined was .82. Child’s achievement. In the spring of first and third grades, teachers were asked to rate the child’s overall academic progress compared with typical students in the same grade. Teachers rated each child’s progress on a 5-point scale ranging from very poor to outstanding. This measure allowed for uniform assessment of academic progress across various grading systems (e.g., letter grades vs. satisfactory/unsatisfactory ratings) used within different schools. In previous work, similar measures predicted school grades (see, e.g., Stevenson & Baker, 1987). Additionally, compared with standardized achievement test scores, teachers’ ratings of performance may be a more accurate and sensitive measure of children’s actual classroom achievement (Keith et al., 1998; Stevenson & Baker, 1987).

Data Analysis Plan We developed a path analysis model to reflect the hypothesized relations among the variables. Path analysis is an important methodological tool for disentangling the effects of various predictors and understanding how a group of variables interrelates in complex patterns. This technique allows for the examination of theory-driven models that identify not only possible direct pathways but also indirect paths contributing to the prediction of a specified dependent variable (Maruyama, 1998). Direct effects establish a direct relation from one variable to another, whereas indirect effects indicate that the relation between two variables is mediated by one or more additional variables (Maruyama, 1998). Because of the correlational nature of our data, causality could not be determined; rather, our intent was to understand how the variables interrelated over time and to test our proposed model. The proposed model (see Figure 1) includes the expected direct and indirect effects of the variables over time. Mothers’ educational level was expected to have direct effects on mothers’ quality of instruction, children’s IQ, parental expectations in first grade, and parental involvement with school in first grade. Mothers’ quality of instruction prior to school entry was expected to have direct effects on later IQ and children’s achievement in first grade. Children’s IQ was expected to have direct effects on parental expectations, parental involvement, and children’s achievement in first grade. We further expected parental expectations, parental involvement, and children’s achievement to have significant effects on each other over time such that (a) parental expectations would have direct effects on parental involvement and children’s achievement at the same grade, (b) parental involvement would have direct effects on children’s achievement at the same grade, and (c) children’s achievement in first grade would have direct effects on both parental expectations and parental involvement in third grade. We also compared the model fit of the full model with the model fit of an alternative model that does not include parental involvement, parental expectations, or mother’s quality of instruction. This alternative model includes mother’s educational level, children’s IQ, and children’s achievement in first grade and third grade. Comparing these two models allowed us to assess whether adding the parenting variables to the model contributed significantly to the model fit.



726

Figure 1. Path analysis examining mother’s education, mother’s quality of instruction, child’s IQ, child’s achievement, parent’s expectations, and parent’s involvement.

Results Correlations To examine the relations between the variables, we calculated Pearson product–moment correlations (see Table 1). Means and standard deviations were also noted. As hypothesized, all zeroorder correlations among the variables were statistically significant at p ⬍ .05.

Path Analysis All analyses involved 187 cases. The path analysis was tested using the statistical software program LISREL 8.5. Missing data points (6%) for the cases were estimated by using estimation maximization (EM) in PRELIS (Schafer, 1997). Multiple imputa-

tion under EM uses an iterative procedure to estimate missing data. This method for handling missing data is superior to other missingdata approaches, such as listwise or pairwise deletion, in that it provides more accurate standard errors and test statistics (Allison, 2002). EM performs well with samples as small as 50 (Graham & Schafer, 1999). The full model fits the data significantly better than the alternative model that eliminates the parenting variables of interest, ⌬␹2(12, N ⫽ 187) ⫽ 25.11, p ⬍ .05. Results of the path analysis for the full model are in Figure 2. The unstandardized, standardized, and statistical significance levels for the path model estimates are shown in Table 2. The overall model is an acceptable fit of the data, ␹2(15, N ⫽ 187) ⫽ 34.62, p ⬍ .01, root-mean-square error of approximation ⫽ .08, comparative fit index ⫽ .96, goodness-of-fit index ⫽ .96, normed fit index ⫽

Table 1 Correlations, Means, and Standard Deviations for Variables Under Study (N ⫽ 187) Variable

1

2

3

4

5

6

7

8

9

1. Mother’s education 2. Quality of instruction 3. WPPSI 4. Parental involvement 1st grade 5. Parental involvement 3rd grade 6. Parental expectations 1st grade 7. Parental expectations 3rd grade 8. Academic achievement 1st grade 9. Academic achievement 3rd grade M SD

— .36 .37 .35 .43 .48 .27 .25 .27 11.83 1.78

— .46 .26 .26 .23 .27 .33 .31 8.91 2.81

— .31 .22 .38 .38 .54 .46 43.37 8.65

— .34 .21 .15 .24 .10 3.56 0.98

— .32 .29 .27 .33 3.34 1.10

— .52 .28 .29 3.25 1.11

— .30 .32 3.48 1.05

— .61 2.97 1.07

— 2.91 1.01

Note. All correlations are significant at p ⬍ .05. WPPSI ⫽ Wechsler Preschool and Primary Scale of Intelligence.


CHILDREN’S ACHIEVEMENT IN EARLY ELEMENTARY SCHOOL

727

Figure 2. Results of path analysis examining mother’s education, mother’s quality of instruction, child’s IQ, child’s achievement, parent’s expectations, and parent’s involvement. N ⫽ 187. Bolded paths are significant at p ⬍ .05. ␹2(15, N ⫽ 187) ⫽ 34.62, p ⬍ .01, root-mean-square error of approximation ⫽ .080, comparative fit index ⫽ .96, goodness-of-fit index ⫽ .96, normed fit index ⫽ .93.

.93, and accounts for 41% of the variance in children’s achievement in third grade. Mothers’ educational level at their children’s birth significantly predicted mothers’ quality of instruction in a teaching task at 42 months (␤ ⫽ .35, t ⫽ 5.17) and their children’s IQ at 64 months (␤ ⫽ .23, t ⫽ 3.45). Mothers’ educational level also significantly predicted both parental involvement with school in first grade (␤ ⫽ .27, t ⫽ 3.43) and parental expectations in first grade (␤ ⫽ .40, t ⫽ 5.95) after controlling for children’s IQ at 64 months. In addition, mothers’ educational level had statistically significant indirect effects on parental involvement in third grade (␤ ⫽ .19, t ⫽ 4.93), parental expectations in first grade (␤ ⫽ .08, t ⫽ 2.90) and third grade (␤ ⫽ .27, t ⫽ 5.71), IQ (␤ ⫽ .13, t ⫽ 3.78), and children’s achievement in first grade (␤ ⫽ .25, t ⫽ 5.00) and third grade (␤ ⫽ .20, t ⫽ 5.21). Mothers’ quality of instruction had a significant direct effect on children’s IQ (␤ ⫽ .38, t ⫽ 5.55) and significant indirect effects on children’s achievement in first grade (␤ ⫽ .18, t ⫽ 4.35) and third grade (␤ ⫽ .16, t ⫽ 3.68). Children’s IQ had significant direct effects on parental expectations in first grade (␤ ⫽ .23, t ⫽ 3.45), parental involvement in first grade (␤ ⫽ .21, t ⫽ 2.82), and children’s achievement in first grade (␤ ⫽ .45, t ⫽ 6.17) and significant indirect effects on parental expectations in third grade (␤ ⫽ .19, t ⫽ 4.14), parental involvement in third grade (␤ ⫽ .17, t ⫽ 3.97), and children’s achievement in third grade (␤ ⫽ .31, t ⫽ 6.13). Although neither parental involvement with their children’s school in first grade nor parental expectations for their children’s

educational attainment in first grade predicted first-grade achievement after controlling for IQ, first-grade achievement did add significant variance to the prediction of parental involvement in third grade (␤ ⫽ .14, t ⫽ 2.01) and parental expectations in third grade (␤ ⫽ .17, t ⫽ 2.53). Parental expectations in third grade added significant variance to the prediction of parental involvement with school in third grade (␤ ⫽ .21, t ⫽ 3.00) after controlling for children’s achievement in first grade. Additionally, parental involvement in third grade (␤ ⫽ .15, t ⫽ 2.52) added significantly to the prediction of children’s achievement in third grade after controlling for previous achievement. Furthermore, parental expectations in first grade (␤ ⫽ .11, t ⫽ 2.30) and mothers’ quality of instruction at 42 months (␤ ⫽ .16, t ⫽ 3.68) had indirect effects on children’s achievement in third grade.

Discussion We proposed a model examining the relations between parental beliefs, parental behavior, and children’s achievement through early elementary school. We hypothesized that parents’ educational level would relate to parents’ quality of instruction, children’s IQ, parents’ involvement with school, and parents’ expectations for their children’s educational attainment. We also proposed that parents’ quality of instruction prior to school entry would have direct effects on children’s IQ and achievement early in school. Furthermore, we anticipated that a cumulative process would occur whereby (a) parental expectations would directly


728


Table 2 Unstandardized, Standardized, and Significance Levels for the Full Model (Standard Errors in Parentheses; N ⫽ 187) Path model estimates

Unstandardized

Standardized

t

Mother’s education 3 Mother’s quality of instruction Mother’s education 3 IQ Mother’s quality of instruction 3 IQ Mother’s education 3 1st-grade parent expectations IQ 3 1st-grade parent expectations Mother’s education 3 1st-grade parent involvement IQ 3 1st-grade parent involvement 1st-grade parent expectations 3 1st-grade parent involvement Mother’s quality of instruction 3 1st-grade achievement IQ 3 1st-grade achievement 1st-grade parent expectations 3 1st-grade achievement 1st-grade parent involvement 3 1st-grade achievement 1st-grade parent expectations 3 3rd-grade parent expectations 1st-grade parent involvement 3 3rd-grade parent expectations 1st-grade achievement 3 3rd-grade parent expectations 1st-grade parent involvement 3 3rd-grade parent involvement 1st-grade achievement 3 3rd-grade parent involvement 1st-grade achievement 3 3rd-grade achievement 3rd-grade parent expectations 3 3rd-grade parent involvement 3rd-grade parent involvement 3 3rd-grade achievement 3rd-grade parent expectations 3 3rd-grade achievement

.35 (.07) .23 (.07) .38 (.07) .40 (.07) .23 (.07) .27 (.08) .21 (.07) .00 (.08) .08 (.07) .45 (.07) .08 (.07) .06 (.06) .47 (.07) .01 (.06) .17 (.07) .27 (.07) .14 (.07) .53 (.06) .21 (.07) .15 (.06) .12 (.06)

.35 .23 .38 .40 .23 .27 .21 .00 .08 .45 .08 .06 .47 .01 .17 .27 .14 .53 .21 .15 .12

5.17 3.45 5.55 5.95 3.45 3.43 2.82 0.01 1.22 6.17 1.13 0.97 7.15 0.20 2.53 4.04 2.01 8.84 3.00 2.52 1.91

Note.

The t value is a test of significance such that t ⬎ 1.96 is equivalent to p ⬍ .05.

affect parental involvement and children’s achievement, (b) parental involvement would directly affect children’s achievement, and (c) children’s achievement would directly affect both parental expectations and involvement. Overall, the proposed model accounts for 41% of the variance in children’s achievement in third grade and provides a significantly better fit than an alternative model that does not include the parenting variables. Mother’s quality of instruction, parental involvement in school, and parental expectations have an effect on children’s achievement in third grade over and above mother’s educational level, children’s IQ, and children’s previous achievement. Our results are consistent with previous findings documenting relations among parents’ educational level and parental involvement and expectations (see, e.g., Gill & Reynolds, 1999; Keith et al., 1996, 1998; Shumow & Miller, 2001; Singh et al., 1995). We found that mothers with higher (vs. lower) educational attainment provided more support for their children in problem-solving situations in preschool, had higher expectations of educational attainment for their children in first grade, and were more involved in their children’s school in first grade. In support of the assumption that parent– child interactions early in children’s lives provide a foundation for later educational activities, mothers’ quality of instruction when their children were 42 months old added significant variance to the prediction of IQ and had significant indirect effects on achievement in both first and third grades. Mothers in our sample who provided appropriate structure for their children and coordinated their own behavior to their children’s activities in problem-solving tasks in early childhood had children with higher levels of IQ compared with mothers who either were uninvolved with their children or did not provide appropriate structure for their children in these laboratory tasks. Higher IQ in turn led to higher academic achievement in first grade. High academic achievement in first grade then led to higher

parental expectations, greater parental involvement, and higher achievement in third grade. High academic achievement early in school may contribute to a process that supports high academic achievement at later ages. Although the results do not fully support a cumulative process among parents’ beliefs, parents’ behavior, and children’s achievement, they do suggest the possibility of bidirectional relations among these variables. When children in our sample had higher (vs. lower) achievement in first grade, their parents were more likely to have higher educational expectations for them and were more involved with their children’s school in third grade. In third grade, parents with higher (vs. lower) expectations for their children’s educational attainment were more likely to be involved in their children’s school. Those parents who were more involved in third grade had higher achieving children in third grade even after controlling for prior achievement. These effects are small, however. Additional research that examines the relation between parents’ beliefs, parents’ behavior, and children’s achievement in other data sets and over longer periods of time could provide further evidence of the hypothesized cumulative process between these variables. Further support for this bidirectional process would help to explain the seemingly divergent findings from previous work examining the relation between parental involvement and children’s achievement. We are cautious in interpreting these correlational data. Causality cannot be determined. Additionally, we examined only whether the model accurately reflected the relation among the variables under investigation. Other variables not included in this model may add to an understanding of how parents influence their children’s achievement in school. Furthermore, parental involvement was measured on the basis of teachers’ reports of parental involvement in school activities, such as parent– child conferences, and did not include parents’ involvement with their children at home


CHILDREN’S ACHIEVEMENT IN EARLY ELEMENTARY SCHOOL

on school-related matters. We emphasize that only 16% of firstgrade variance and 19% of third-grade variance in parental involvement were accounted for in this model. Obviously, many other factors in addition to children’s achievement and parental expectations affect overall levels and changes in parental involvement with school. Because of the nature of our low-income sample, many of the families experienced instability across this time period (e.g., life stress, mobility, change in employment status, and changes in child’s school) that could have influenced parents’ involvement in their children’s school. In addition, teacher–parent relationships, parents’ work schedules, transportation issues, and school climate have all been found to influence parental involvement over time (Christenson & Sheridan, 2001; Fantuzzo, Tighe, & Childs, 2000; Hoover-Dempsey & Sandler, 1997). It is likely that these factors had some impact on parents’ involvement with school in our sample. Because the sample was relatively small, different models based on sex or ethnicity could not be examined. Future research should attempt to include a larger sample so that sex and ethnic differences can be further explored. Furthermore, although diverse ethnically, the sample was low SES from a large Midwestern city. Other samples should be examined to determine whether similar effects are found across income levels. Additional research exploring the possible influence of parental involvement with school on teachers’ perceptions and interactions with children is needed. Teacher– child interactions do affect children’s achievement (see, e.g., Burchinal, Peisner-Feinberg, Pianta, & Howes, 2002). These interactions may in part be affected by teachers’ perceptions of the children and their parents. These perceptions may be partially shaped by teachers’ interactions with parents. In addition, the long-term effects of early parenting factors on later high school achievement need to be investigated. Longitudinal research has the potential to separate those early parenting factors that are important for later achievement from those that are incidental.

References Allison, P. D. (2002). Missing data. Thousand Oaks, CA: Sage. Bobbett, G. C., French, R. L., Achilles, C. M., & Bobbett, N. C. (1995, November). An analysis of Nevada’s report cards on high school. Paper presented at the annual meeting of the Mid-South Educational Research Association Annual Meeting, Biloxi, MI. Burchinal, M. R., Peisner-Feinberg, E., Pianta, R., & Howes, C. (2002). Development of academic skills from preschool through second grade: Family and classroom predictors of developmental trajectories. Journal of School Psychology, 40, 415– 436. Christenson, S. L., Rounds, T., & Gorney, D. (1992). Family factors and student achievement: An avenue to increase students’ success. School Psychology Quarterly, 7, 178 –206. Christenson, S. L., & Sheridan, S. M. (2001). Schools and families: Creating essential connection for learning. New York: Guilford Press. Connell, C. M., & Prinz, R. J. (2002). The impact of childcare and parent– child interactions on school readiness and social skills development for low-income African American children. Journal of School Psychology, 40, 177–193. Deslandes, R., Royer, E., Turcotte, D., & Bertrand, R. (1997). School achievement at the secondary level: Influence of parenting style and parent involvement in schooling. McGill Journal of Education, 32, 191–207. Fan, X. (2001). Parental involvement and students’ academic achievement:

729

A growth modeling analysis. Journal of Experimental Education, 70, 27– 61. Fan, X., & Chen, M. (2001). Parental involvement and students’ academic achievement: A meta-analysis. Educational Psychology Review, 13, 1–22. Fantuzzo, J., Tighe, E., & Childs, S. (2000). Family Involvement Questionnaire: A multivariate assessment of family participation in early childhood education. Journal of Educational Psychology, 92, 367–376. Fehrmann, P. G., Keith, T. Z., & Reimers, T. M. (1987). Home influence on school learning: Direct and indirect effects of parental involvement on high school grades. Journal of Educational Research, 80, 330 –337. Gill, S., & Reynolds, A. J. (1999). Educational expectations and school achievement of urban African American children. Journal of School Psychology, 37, 403– 424. Goldenberg, C., Gallimore, R., Reese, L., & Garnier, H. (2001). Cause or effect? A longitudinal study of immigrant Latino parents’ aspirations and expectations, and their children’s school performance. American Educational Research Journal, 38, 547–582. Graham, J. W., & Schafer, J. L. (1999). On the performance of multiple imputation with multivariate data with small sample size. In R. H. Hoyle (Ed.), Statistical strategies for small sample research (pp. 1–27). Thousand Oaks, CA: Sage. Halle, T. G., Kurtz-Costes, B., & Mahoney, J. L. (1997). Family influences on school achievement in low-income, African American children. Journal of Educational Psychology, 89, 527–537. Hoover-Dempsey, K. V., & Sandler, H. M. (1997). Why do parents become involved in their children’s education? Review of Educational Research, 67(1), 3– 42. Izzo, C. V., Weissberg, R. P., Kasprow, W. J., & Fendrich, M. (1999). A longitudinal assessment of teacher perceptions and parent involvement in children’s education and school performance. American Journal of Community Psychology, 27, 817– 839. Keith, T. Z., Keith, P. B., Quirk, K. J., Cohen-Rosenthal, E., & Franzese, B. (1996). Effects of parental involvement on achievement for students who attend school in rural America. Journal of Research in Rural Education, 12, 55– 67. Keith, T. Z., Keith, P. B., Quirk, K. J., Sperduot, J., Santillo, S., & Killings, S. (1998). Longitudinal effects of parent involvement on high school grades: Similarities and differences across gender and ethnic groups. Journal of School Psychology, 36, 335–363. Keith, T. Z., Reimers, T. M., Fehrmann, P. G., Pottebaum, S. M., & Aubey, L. W. (1986). Parental involvement, homework, and TV time: Direct and indirect effects on high school achievement. Journal of Educational Psychology, 78, 373–380. Maruyama, G. M. (1998). Basics of structural equation modeling. Thousand Oaks, CA: Sage. 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. Miliotis, D., Sesma, Jr., A., & Masten, A. S. (1999). Parenting as a protective process for school success in children from homeless families. Early Education and Development, 10, 111–133. National Center for Education Statistics. (1998). Statistical analysis report: Parent involvement in children’s education—Efforts by public elementary schools. Retrieved July 17, 2002, from http://nces.ed.gov/pubs98/ 98032.html National Education Goals Panel. (1999). The national education goals report: Building a nation of learners, 1999. Washington, DC: U.S. Government Printing Office. Okagaki, L., & Frensch, P. A. (1998). Parenting and children’s school achievement: A multiethnic perspective. American Educational Research Journal, 35, 123–144. Okpala, C. O., Okpala, A. O., & Smith, F. E. (2001). Parental involvement,


730


instructional expenditures, family socioeconomic attributes, and student achievement. Journal of Educational Research, 95, 110 –115. Pianta, R. C., & Egeland, B. (1994). Predictors of instability in children’s mental test performance at 24, 48, and 96 months. Intelligence, 18, 145–163. Pianta, R. C., Erickson, M. F., Wagner, N., Kreutzer, T., & Egeland, B. (1990). Early predictors of referral for special services: Child-based measures versus mother– child interactions. School Psychology Review, 19, 240 –250. Pianta, R. C., & Harbers, K. L. (1996). Observing mother and child behavior in a problem-solving situation at school entry: Relations with academic achievement. Journal of School Psychology, 34, 307–322. Pianta, R. C., Nimetz, S. L., & Bennett, E. (1997). Mother– child relationships, teacher– child relationships, and school outcomes in preschool and kindergarten. Early Childhood Research Quarterly, 12, 263–280. Rogala, S. (2001, April). Parental involvement and academic achievement in the early years. Poster session presented at the biennial meeting of the Society for Research in Child Development, Minneapolis, MN. Sattler, J. M. (1992). Assessment of children (3rd ed.). San Diego, CA: Author. Schafer, J. L. (1997). Analysis of incomplete multivariate data. Boca Raton, FL: Chapman & Hall/CRC. Seginer, R. (1983). Parents’ educational expectations and children’s aca-

demic achievements: A literature review. Merrill-Palmer Quarterly, 29, 1–23. Shaver, A. V., & Walls., R. T. (1998). Effect of Title I parent involvement on student reading and mathematics achievement. Journal of Research and Development in Education, 31, 90 –97. Shumow, L., & Miller, J. D. (2001). Parents’ at-home and at-school academic involvement with young adolescents. Journal of Early Adolescence, 21, 68 –91. Singh, K., Bickley, P. G., Trivette, P., Keith, T. Z., Keith, P. B., & Anderson, E. (1995). The effects of four components of parental involvement on eighth-grade student achievement: Structural analysis of NELS-88 Data. School Psychology Review, 24, 299 –317. Stevenson, D. L., & Baker, D. P. (1987). The family–school relation and the child’s school performance. Child Development, 58, 1348 –1357. Sui-Chu, E. H., & Willms, J. D. (1996). Effects of parental involvement on eighth-grade achievement. Sociology of Education, 69, 126 –141. Wechsler, D. (1967). Wechsler Preschool and Primary Scale of Intelligence. New York: Psychological Corporation. Wechsler, D. (1974). Manual for the Wechsler Intelligence Scale for Children—Revised. New York: Psychological Corporation.

Received February 12, 2003 Revision received May 27, 2004 Accepted June 30, 2004 䡲

New Editor Appointed for History of Psychology The American Psychological Association announces the appointment of James H. Capshew, PhD, as editor of History of Psychology for a 4-year term (2006 –2009). As of January 1, 2005, manuscripts should be submitted electronically via the journal’s Manuscript Submission Portal (www.apa.org/journals/hop.html). Authors who are unable to do so should correspond with the editor’s office about alternatives: James H. Capshew, PhD Associate Professor and Director of Graduate Studies Department of History and Philosophy of Science Goodbody Hall 130 Indiana University, Bloomington, IN 47405 Manuscript submission patterns make the precise date of completion of the 2005 volume uncertain. The current editor, Michael M. Sokal, PhD, will receive and consider manuscripts through December 31, 2004. Should the 2005 volume be completed before that date, manuscripts will be redirected to the new editor for consideration in the 2006 volume.