The Impact of Parental Involvement and Expectations on Elementary School Students' Academic Achievement

This study is a correlational research that aims to define the relationship between parental involvement, parental expectations and elementary school students’ academic success. In defining the sample of the study, it was aimed to ensure the diversity of socioeconomic level and sociocultural status. Totally, 420 parents have voluntarily participated in the research after the purpose of the study explained to them. The data was gathered with the Scale for School and Family Involvement in Elementary Schools, subscale of parents developed by Sheldon and Epstein’s (2007) which has been adapted into Turkish by Oğuz (2012). While there was only one dependent variable of the study (academic success), there are six independent variables (parents’ educational status, family SES, parents’ gender, child’s gender, parental involvement, parental expectation). Within the research, independent samples t-test and Chi-square independence test were used. Moreover, logistic regression analysis was performed in order to define the power of predictability of the variables. Results showed that, there are not significantly statistical relations between academic success and parents’ gender, children’s gender, parental involvement at school. On the other hand, it has been concluded that parents’ educational status, family SES, parental expectation and parental involvement at home are significantly correlated with children’s academic success. It was seen that the most important predictor of a child’s academic success is parental expectations. Based on the results, recommendations were developed for parents, schools and researchers.


The Impact of Parental Involvement and Expectations on Elementary School Students' Academic Achievement
Theorists investigating human behavior from the perspective of contextual features emphasize that the personal characteristics of an individual, as well as his/her interactions with others, play an important role in the acquisition of behavior. Vygotsky (1978) (Valsiner, 1988). Parental cognition is important since it influences all the behaviors of parents supporting a child's education and development. The present study examines parental cognition (specifically, parental expectation) in terms of academic achievement and uses "parents' educational status," "family socioeconomic status (SES)," and "parental involvement (PI)" as variables.

Parental Expectation
In their meta-analysis studies, Fan and Chen (2001) and Jeynes (2007) argued that the most important parental variable playing an effective role in academic achievement is the parent's expectations regarding the educational performance of a child. Carpenter (2008) defined parental expectation, which is one of the parental cognitions, as predictions and judgments towards the child's future academic achievement. These expectations include all the judgments and beliefs about a child's performance at school or the level of education he/she can complete as well as the behaviors supporting them. However, parental expectations are different from parental aspirations related to the child's education.
Educational expectations are defined as judgments and beliefs with more realistic bases for the future, whereas aspirations are defined as the wishes expressed by parents based on social norms regarding the importance of schooling and their own aims (Yamamoto & Holloway, 2010).
Expectations have an important effect on parental behaviors related to both allocating resources to support the child at a certain academic level and evaluating the child's academic competencies. Moreover, the parents' actions addressing the academic achievement mediate them to establish new expectations on their children or reconstruct the current expectations (Valsiner, 1988 (Daymaz, 2012;Davis-Keen, 2005;Davis-Kean et al., 2003;Yamamoto & Holloway, 2010), child's gender (Jacobs et al., 2005;Räty et al., 2002), and previous achievements of a child (Englund et al., 2004;Neuenschwander et al., 2007).

Parental Involvement (PI)
The environment and conditions provided by parents at home are as important as the school in the academic achievement of a child. information about the progression of students and taking necessary measures by receiving ontime information about the problems that the child faces (Hornby, 2011;Jeynes, 2011). On the other hand, school-based PI includes practices such as meetings with teachers, voluntary help for teachers on school trips, and making efforts to meet the needs of schools.

Significance and Purpose of the Research
In the literature, besides the individual factors affecting academic achievement, and parental characteristics have also been frequently discussed in recent years. Family SES, parents' gender, parental educational status, PI, and parental cognitions (expectation, attribution, motivational beliefs about involvement, etc.) are among the variables related to parents and family.
In previous studies carried out in different cultures, it was emphasized that the gender of parents plays an effective role in academic achievement, and this is related to the differences in the parents' cognition regarding the subject. Accordingly, it can be seen that mothers take more responsibility and are more effective in their child's education when compared to fathers (Deslandes, 2009;Hornby, 2011;Kim, 2018). It is understood that this applies even to countries such as Sweden, where there are many legal regulations on gender equality (Gurdal, 2015). Studies have revealed that the difference is substantially related to the traditional gender roles attributed to women and men. Another remarkable finding concerns the effect of parents' educational status and family SES on academic achievement. The assumption that a child of a family with a high socio-economic level will have greater academic success has been emphasized for many years in the literature. It is emphasized in the literature that PI is one of the most important variables affecting the academic achievement of students at any educational level (Englund et al., 2004;Fan & Chen, 2001;Hill & Tyson, 2009;Şad, 2012;Weber, 2010). Furthermore, it can also be seen that several parent-related variables affect the level and quality of PI. The first one of these variables is parents' educational status. It has been shown that as the education level of parents decreased, PI also decreased (Ramirez, 2003;LaRocque et al., 2011;Flynn, 2007). This finding can be explained by the limited perception of parents about their role or efficacy in PI (Amatea & West-Olatunji, 2007;Hoower-Dempsey et al., 2005;LaRocque et al., 2011;Ramirez, 2003). Besides this, it has been reported that home-and school-based PI practices have different effects on children at different education levels (Çubukçu & Girmen, 2006;Flaxman & Inger, 1991;Vaden-Kiernan & Chandler, 1996). For students at the elementary school level, school-based PI positively affects the student's motivation about learning and schooling as well as their academic achievement (Englund et al., 2004;Gonzales-DeHass et al., 2005;Pavalache-Ilie & Tirdia, 2015). However, this effect is not valid for high school students. Accordingly, it was revealed that PI at home has a positive effect on school attendance and motivation for learning as well as academic achievement at the high school level (Clinton & Hattie, 2013;Gonzales-DeHass et al., 2005;Lam & Ducreux, 2013;Sarpkaya, 2007).
Studies have shown that the children of parents with high expectations are more successful at school when compared to the children of parents with low expectations (Kirk et al., 2011;Patrikakou, 2008;Yan & Lin, 2005;Wilder, 2014). Morera et al. (2015)  study examining its direct effect on academic achievement. Accordingly, the present study aims to examine the effects of PI and parental expectations on the academic achievement of students aged between 10 and 13. The research questions are as follows: 1. Is there a significant relationship between parents' gender, students' gender, and academic achievement?
2. Is there a significant relationship between family SES, parents' educational status, parents' expectations, and academic achievement?
3. Are family SES, parents' educational status, PI, and parents' expectation significant predictors of academic achievement?
4. How accurately can the variables of family SES, parents 'educational status, PI, parents' expectations distinguish students with low and high academic achievement?

Methodology
This study is a correlational study aiming to examine whether academic achievementdependent variable-is predicted by the parents' educational status, family SES, parents' and child's gender, PI, and parental expectation -independent variables-.

Participants
A stratified sampling method was used in the present study. After the first interviews with the relevant units in the Kastamonu Provincial Directorate of National Education, residential areas differing in terms of socioeconomic and socio-cultural characteristics were determined. As in determining the regions, socioeconomic and sociocultural level differences were taken into consideration in determining the primary and secondary schools, where data would be collected. In calculating the sample size, the reliability level was set to be 95% and the mean deviation value was set at five. Thus, the sample size was calculated to be 375.
However, it was decided to increase the sample size to 450 in order to handle the possible missing data. After the sample size was determined, the students were distributed equally according to their grade levels.
Following this step, the relevant school principals were contacted; parents were informed about the purpose of the research and contact information of the researchers (by principals, teachers, or school counselors). The scales were then delivered to parents through students. On the first page of the scale form, information about the researchers and the purpose of the study was given along with researcher contact information. In addition, it was stated that attendance in the study was completely voluntary and that filling out the scales would be considered as acceptance for attendance. Excluding 30 invalid scales (filled incompletely or with the same answers always marked) from 450 forms submitted, 420 forms were used for analyses. Table 1 provides information about the participants whose answers were valid. 3. High (3100-5000 TL / 5000 TL and more) 139 33.1 As can be seen in Table 1, mothers constituted the majority of participants. Moreover, although there was a difference between the parents' educational statuses, they were mostly university graduates. Finally, the family SES data suggested that most of the participants were in the low-income group.

Data Collection
When collecting data on academic achievement, the parents could not be asked about their children's year-end GPAs since the study data were collected during the spring semester. The Parents Scale for School and Family Involvement in Elementary Schools, which was developed by Sheldon and Epstein (2007) and adapted to Turkish by Oğuz (2012)

Data Analysis
Before analyzing the data, normal distribution curves and skewness and kurtosis with Kolmogorov-Smirnov (K-S) test values were examined in order to determine the missing data related to responses and to determine whether the data were distributed normally. According to the results presented in Table 2, it can be seen that, although the data were not normally distributed, skewness and kurtosis values are determined to be more reliable sources of information for normality. In the present study, skewness and kurtosis values were reported to be within the acceptable limit, which was +/-1.50 (Tabachnick & Fidell, 2007). In this regard, the Chi-square independence test was used in order to determine the correlation between dependent and independent variables. In Chi-square tables, the Phi value was used when the categories of the variables were 2X2 and Cramer's V value was used when the variables were 2X4. Additionally, an independent samples t-test was used to determine whether there was a significant difference between the mean values of academic achievement and PI at school and PI at home.
In this study, logistic regression analysis was performed. As Palant (2016)  assumption is that the model that formed as a result of logistic regression analysis, is sensitive toward the extreme values in the data. The last assumption is that logistic regression is a type of multiple regression (Çokluk, 2010). Thus, it is sensitive toward a high correlation between independent/predictor variables because high correlation causes multi-collinearity problems between the predictor variables (Çokluk et al., 2014). Eigenvalues state indices, and variance rates, which were determined to examine the multi collinearity problem between the variables, are presented in Table 3. Table 3. As seen in Table 3, state index values are lower than 30, and there is no value that is significantly higher than others. Moreover, the highest variance of each predictor variable is loaded to a separate eigenvalue. From this aspect, it can be seen that every predictor variable explains another dimension of the variance (Çokluk, 2010). The standard errors, tolerance, VIF values of the predictor variables, and binary correlations between the variables are presented in Table 4. As seen in Table 4, the tolerance values of all predictor variables are higher than .1.

Eigenvalues, state indexes and variance rates
Moreover, the VIF values are lower than 10. Çokluk (2010) stated that tolerance values higher than .1 are an intended situation in multicollinearity problems. This finding suggested that there was no such problem, and the logistic regression analysis could now be performed.

Correlations Between Dependent Variable and Independent Variables
According to the results of the Chi-square test, no significant correlation was found between academic achievement and parents' gender (x²=.026; p>.05) or the child's gender (x²=1,49; p>.05). In addition, it was determined that mothers' mean values were higher regarding both high and low academic achievements. In analyses of the children's gender, it was determined that female or male gender did not cause any difference in academic achievement. Significant correlations were found between the variables of parents' educational status, family SES, parental expectation, and academic achievement. Moreover, according to the results of the independent samples t-test conducted in order to reveal whether academic achievement varies depending on PI at school, no statistically significant difference was found between the variables (t=-1,61; p>.05). The eta square value, which was used in order to calculate the effect size of this finding, was calculated to be η2=0.03.
Considering the difference between mean scores, it was determined that the values were close to each other and, thus, PI at school did not cause any significant difference in academic achievement.
The Chi-square test results showed that there was a significant correlation between academic achievement and parents' education level (x²=17.20; p<.05). The results are presented in Table 5. The results showed that there was a significant difference between being in the high achievement groups of children with poorly educated parents and children with highly educated parents. Similarly, a significant correlation was specified between academic achievement and family SES (x²=6.10; p>.05). The results are presented in Table 6. It is determined that there is a significant correlation between the parents' expectations and academic achievement (x²=40,24; p<.05). The results are given in Table 7. As can be seen in Table 7, when parents' expectations are at moderate to high levels, the ratios between their children's being at low or high achievement levels vary significantly.
Moreover, when parents' expectation levels are low or weak, a decrease was observed in their children's rate of being in a high achievement state.
Finally, the results of the independent samples t-test showed that there was a statistically significant difference between the variables (t=-3.92; p<.01). The analysis results are presented in Table 8. The eta square value, which was used in calculating the effect size of the determined difference, was calculated to be η2=0.12. This result indicates that, contrary to PI at school, PI at home increases academic achievement. Furthermore, the positive correlation between PI at home and high academic achievement shows the importance of PI activities such as helping with homework and projects, checking homework, reading books together, talking about courses, discussing the day spent in school, etc.

Prediction of Academic Achievements of Elementary School Students
A regression model including the independent variables that had statistically significant correlations with academic achievement was formed according to the results of the analyses presented in the previous section. This model consists of four independent variables: PI at home, parents' educational status, parents' expectations, and family SES. Findings obtained from the logistic regression analysis are presented in Tables 9, 10, 11, and 12. Table 9.

Summary of the Target Model
Step 954 ,148 ,199 According to Table 9, predictor variables explained 14.8-19.9% of the variance regarding academic achievement. Table 10.

Hosmer and Lemeshow Test
Step Chi-Square Sd P 1 6,982 8 ,539 According to the table above, it can be seen that the Hosmer and Lemeshow test result is not significant (p>.05). Thus, the model has an acceptable fit. As seen in Table 11, the formed regression model accurately classified 49.2% of the students with low achievement and 84.4% of the students with high achievement.

Coefficient Estimates of the Target Model Variables
Step possibility of a child with highly educated parents is almost three times higher when compared to the children of parents with low to moderate educational levels (Exp(B)=2.968).

Discussion, Conclusion and Recommendations
In the present study, the findings revealed that there was no relationship between academic achievement and parents' gender, which might be because of the majority of participants in the present study were mothers. Secondly, the difference may also be because of some traditional gender roles which they address the responsibilities concerning the education and development of children are mostly given to mothers. This may be due to a belief that mothers may have greater awareness and knowledge in this respect. For instance, Lee et al. (2007) noted that the roles of mothers in explaining the antisocial behaviors and social skills of children and those of fathers in explaining characteristics of children such as cognitive development and entrepreneurship may be more important than those of the other parent. In the present study, it was also determined that there was no relationship between children's gender and academic achievement. This may be related to the fact that the majority of the parents participating in the study had daughters and that the number of parents with sons was limited. However, it has also been stated in the literature that there was no significant difference between the genders in terms of general intelligence and achievement and that there may only be differences in specific subject areas (Batyra, 2017;Liu & Lynn, 2015).
In parallel with the literature, it was found that there was a positive and significant relationship between parents' educational status and academic achievement. Şirin (2005) stated that parents' educational status has a more important impact on academic achievement when compared to family's SES. However, in this study, a significant positive relationship was found between family's SES and academic achievement. This result could be interpreted as meaning that income level is important in terms of access to resources in Turkey, and therefore that a high family income positively affects the academic achievement of a child. In his study, Şirin (2005) found that students' grade level, ethnic origin, some student characteristics, and school environment affected the degree of correlation between SES and academic achievement.
PI was examined as home-and school-based participation in the present study. A significant relationship was found between academic achievement and PI practices at home, whereas no statistically significant relationship was found between academic achievement and school-based participation practices. There may be several possible explanations for this. The first may be that parents do not receive invitations or support from their schools and/or teachers for participation. Hoover-Dempsey et al. (2005) stated that parents' perception of being invited is a significant motivational belief influencing school-based participation. Another possible reason is that PI practices at school were insufficient. As both the Turkish and international literature emphasizes, PI is mostly performed at limited levels and as limited practices in all educational levels (Epstein, 2008;Çubukçu & Girmen, 2006;Graue & Brown, 2003;Hornby & Lafaele, 2011;Sabancı, 2009). The last possible reason is related to the age group included in the present study. The literature on PI suggests that adolescents do not take kindly to school-based parental participation since they are in a period of identity-searching and experiencing vital physical changes (Comer & Haynes, 1991;Patrikakou, 2008;Deslandes, 2009;Hornby, 2011;Stahl, 2012). Studies have shown that PI practices at home, especially those based on giving positive feedback about students' educational performance or guiding them for their future decisions and establishing mutual communication, have a very significant and positive impact on academic achievement (Gonzalez-DeHass et al., 2005;Hickman et al., 1995;Hill & Tyson, 2009;Kuperminc et al., 2007).
In the present study, logistic regression analysis was performed in order to determine the predictive power of variables regarding academic achievement. Accordingly, as Fan and Chen (2001), Jeynes (2007), and Yamamoto and Holloway (2010) also revealed, it was found that the most important predictor of academic achievement is high parental expectation.
Moreover, even moderate expectations were found to be a better predictor than all other variables (parents' and children' gender, family SES, parents' educational status, and PI at home and school) related to academic achievement. As Lenz (1999) has also highlighted, it was found that the possible reason for the effectiveness of expectations is mutual, open, and direct communications between child and parents. Lenz (1999) also stated that children demonstrate positive emotional and behavioral changes when they feel that they are supported and assisted by adults. Accordingly, high expectations basically affect a child's perception for her/his-self. Positive perceptions cause children to make efforts to be successful and, therefore, to make their parents happy. In previous studies, it was determined that this strong relationship between academic achievement and expectation also has a spiral aspect.
Accordingly, it was found that academic achievement increases as the coherence between parental expectations and children's expectations about their own achievement increases. It seems that as the academic achievement increases, parental expectations are also getting higher (Goldenberg et al., 2001;Hoover-Dempsey & Sandler, 2007;Rubie-Davies et al., 2010 Moreover, as Başar (2001) and Carney-Hall (2008) emphasized, unrealistically high expectations may cause the child to feel excessive pressure and responsibility and to experience failure and disappointment. In the present study, it was revealed that family's SES, parents' educational status, PI practices at home, and parents' educational expectations are important variables in explaining academic achievement. At the same time, the most outstanding result of the present study is that parental expectations were found to be the most important predictor of academic achievement when compared to all other variables.
The results of the present study offer several more insights for parents, principals, and researchers. As mentioned above, the most important predictor of academic achievement is the expectations of parents about their child's education. For this reason, it is important for parents to monitor their children's academic development and achieve the necessary knowledge to help them establish realistic expectations about their children and discuss them with their children. However, for the parents to establish realistic expectations among themselves and with their children about school and education, communication with teachers is an important information source. Besides that, it is especially important that parents with a low educational status are supported by teachers, schools, and other institutions so that they can access information sources about their children's development and education.
Since the features and characteristics of each school and family are different, it is important to analyze possible risk factors for the planning, implementation, and effectiveness of PI programs. In the analysis, to ensure the effectiveness of programs, it is also necessary to focus not only on school-based but also on involvement practices at home. As mentioned above, parents may need support, training, develop communication skills, and information enabling them to create realistic expectations about their children. In this regard, principals and teachers have a significant role in creating an effective PI environment for all.
Finally, with reference to the results obtained from this research, there are several issues that have been handled in detail by the researchers. In the Turkish literature, it is believed that studies should examine academic success within the scope of parental cognitions and academic socialization through qualitative research methods and techniques in detail. In conclusion, it has been argued that parental expectation is an important predictor of academic achievement. Therefore, it is necessary to present the relationship of this variable to academic achievement in further detail through longitudinal studies to be conducted at different educational levels.

Statement for Ethical Issues
We confirm that all aspect of the research covered in this manuscript that has involved human has been conducted with the ethical rules. The authors also declared that they did not use any references other than those given in the References section and the manuscript has not been sent to another journal for review or publication.