View in EDS HTML Full Text PDF Full Text

What Do Parents Teach Their Children?--The Effects of Parental Involvement on Student Performance in Dutch Compulsory Education

Saved in:
Bibliographic Details
Title: What Do Parents Teach Their Children?--The Effects of Parental Involvement on Student Performance in Dutch Compulsory Education
Language: English
Authors: Cabus, Sofie J., Ariës, Roel J.
Source: Educational Review. 2017 69(3):285-302.
Availability: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
Peer Reviewed: Y
Page Count: 18
Publication Date: 2017
Document Type: Journal Articles
Reports - Research
Education Level: Secondary Education
Descriptors: Parent Participation, Academic Achievement, Foreign Countries, Birth Order, Educational Attainment, Correlation, Family School Relationship, Family Environment, Socioeconomic Status, Grade Repetition, Immigrants, Mathematics Achievement, Outcomes of Education, Homework, Secondary School Students, Least Squares Statistics, Parent Child Relationship, Regression (Statistics), Robustness (Statistics)
Geographic Terms: Netherlands
DOI: 10.1080/00131911.2016.1208148
ISSN: 0013-1911
Abstract: Theory and evidence indicate that, if family size grows, the younger children will get less parental involvement than the older children. These differences in parental involvement through birth order may impact academic achievement if, and only if, parental involvement is an important determinant of children's educational attainment. The oldest child then benefits the most in terms of educational outcomes. Estimates for the Netherlands show a robust negative relationship between birth order and parental involvement, and significant positive medium to large effects of parental involvement through birth order on various measures of academic achievement. Furthermore, our findings indicate that academic achievement is rooted in a school-supportive home climate, and often created by the mother. However, when it comes to math performance and grade retention, it is better that both parents unduly interfere with school. We also find that parents with low socio-economic status and from immigrant families are as much involved in the education of their children as the average Dutch family, but their involvement is less effective in terms of children's learning outcomes.
Abstractor: As Provided
Number of References: 65
Entry Date: 2017
Accession Number: EJ1131023
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwEEsIkmHzLLGmcKUJ0GS7MtAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDEX5Vq8txP6JJJrVfQIBEICBmm7aGGPKLL01j1bgdneTp6ntgyEA0QKj9dp4vvStdPc0NP0qd4nnW0nmA6jM4axlNxQ1AoJTxC4z73PoxaAVdtIVlK-F0IMYg4GW5ApSSqO-NuemtYzJstiId5YFHoURKGDUA_gk6_nzIRYz8U3CejbhbhpnSsDDz7ksm2jJz57CARKipcM4ciL7Xj8PFirrTnHvKTVWxcd5q3w=
Text:
  Availability: 1
  Value: <anid>AN0121413965;edi01may.17;2019Feb20.14:11;v2.2.500</anid> <title id="AN0121413965-1">What do parents teach their children? – The effects of parental involvement on student performance in Dutch compulsory education. </title> <p>Theory and evidence indicate that, if family size grows, the younger children will get less parental involvement than the older children. These differences in parental involvement through birth order may impact academic achievement if, and only if, parental involvement is an important determinant of children's educational attainment. The oldest child then benefits the most in terms of educational outcomes. Estimates for the Netherlands show a robust negative relationship between birth order and parental involvement, and significant positive medium to large effects of parental involvement through birth order on various measures of academic achievement. Furthermore, our findings indicate that academic achievement is rooted in a school-supportive home climate, and often created by the mother. However, when it comes to math performance and grade retention, it is better that both parents unduly interfere with school. We also find that parents with low socio-economic status and from immigrant families are as much involved in the education of their children as the average Dutch family, but their involvement is less effective in terms of children's learning outcomes.</p> <p>Keywords: Birth order; communication; homework involvement; parents; student achievement</p> <hd id="AN0121413965-2">1. Introduction</hd> <p>It is widely supported that student achievement increases one's chances on school career and labor market success (De Witte et al. [<reflink idref="bib22" id="ref1">22</reflink>]; Rumberger [<reflink idref="bib59" id="ref2">59</reflink>]). There is much less consensus in the literature, however, on how to increase student achievement in the most effective way (e.g. European Commission [<reflink idref="bib28" id="ref3">28</reflink>]; Organization for Economic Cooperation and Development [OECD] [[<reflink idref="bib48" id="ref4">48</reflink>]]). The literature indicates several determinants of academic achievement, including: student motivation (Baumert and Demmrich [<reflink idref="bib11" id="ref5">11</reflink>]; Linnenbrink [<reflink idref="bib39" id="ref6">39</reflink>]); rewards (Deci, Koestner, and Ryan [<reflink idref="bib23" id="ref7">23</reflink>]); school engagement (Wang and Holcombe [<reflink idref="bib63" id="ref8">63</reflink>]); teachers (Klassen and Tze [<reflink idref="bib34" id="ref9">34</reflink>]; Rockoff [<reflink idref="bib55" id="ref10">55</reflink>]); self-regulation strategies for cognitive, motivational and behavioral aspects (Nota, Soresi, and Zimmerman [<reflink idref="bib47" id="ref11">47</reflink>]); socio-economic status (Alexander, Entwisle, and Olson [<reflink idref="bib2" id="ref12">2</reflink>]; Dumont et al. [<reflink idref="bib25" id="ref13">25</reflink>]); classroom composition (Koth, Bradshaw, and Leaf [<reflink idref="bib36" id="ref14">36</reflink>]; Reyes et al. [<reflink idref="bib54" id="ref15">54</reflink>]); peer effects (Rothstein [<reflink idref="bib56" id="ref16">56</reflink>]); competition between schools (Macleod and Urquiola, [<reflink idref="bib42" id="ref17">42</reflink>]); unauthorized truancy (Henry [<reflink idref="bib32" id="ref18">32</reflink>]); crime and citizenship (Lochner and Moretti [<reflink idref="bib40" id="ref19">40</reflink>], [<reflink idref="bib41" id="ref20">41</reflink>]); reasoning abilities (Ariës, Groot, and Maassen van den Brink, [<reflink idref="bib8" id="ref21">8</reflink>]); and health and mortality (Albouy and Lequien [<reflink idref="bib1" id="ref22">1</reflink>]; Groot and Maassen van den Brink [<reflink idref="bib30" id="ref23">30</reflink>]). This article puts particular focus on one determinant of academic attainment that has increasingly attracted attention from many scientists and policy-makers across OECD countries, namely: parental involvement in the education of their children (e.g. European Commission [<reflink idref="bib28" id="ref24">28</reflink>]; US Department of Education [<reflink idref="bib62" id="ref25">62</reflink>]). As parental involvement involves many complex dimensions, previous literature indicate large differences in effects on children's academic achievement (Ariës and Cabus [<reflink idref="bib7" id="ref26">7</reflink>]; Driessen, Smit, and Sleegers [<reflink idref="bib24" id="ref27">24</reflink>]). Wilder ([<reflink idref="bib64" id="ref28">64</reflink>], 390) summarizes:</p> <p>The most prominent parental involvement components in the analyzed meta-analyses were communication between parents and children regarding school, checking and helping with homework, parental educational expectations and aspirations for their children, and attendance and participation in school activities.</p> <p>If one could estimate the benefits of parental involvement in terms of higher academic achievement, not only a definition problem may arise (see Section 3), but also some empirical issues. Parental involvement is clearly (reversely) related to student achievement through its determinants. For example, children with learning difficulties may benefit from increased parental involvement in their homework. As a result, student performance determines the level of homework involvement, so that one may falsely conclude that parental involvement leads to lower student performance. Reversed causality or endogeneity are serious issues of concern when interpreting the relationship between parental involvement and achievement (Angrist and Pischke [<reflink idref="bib5" id="ref29">5</reflink>]). In fact, several reviews of the literature argue that causal empirical evidence on the positive relationship between parental involvement and student success is ultimately scarce (e.g. Ariës and Cabus [<reflink idref="bib7" id="ref30">7</reflink>]; Avvisati, Besbas, and Guyon [<reflink idref="bib9" id="ref31">9</reflink>]; Avvisati et al. [<reflink idref="bib10" id="ref32">10</reflink>]; Fan and Chen [<reflink idref="bib29" id="ref33">29</reflink>]; Hotz and Pantano [<reflink idref="bib33" id="ref34">33</reflink>]; Patall, Cooper, and Robinson [<reflink idref="bib50" id="ref35">50</reflink>]; Price [<reflink idref="bib52" id="ref36">52</reflink>], [<reflink idref="bib53" id="ref37">53</reflink>]). The main contribution of this article to the previous literature is then an evaluation of the effects of parental involvement on student academic achievement, rather than estimating the association between both. Doing so, it is proposed in this article to use birth order of the child as an instrumental variable (IV). Fairly recent evidence (Damian and Roberts [<reflink idref="bib21" id="ref38">21</reflink>]; Price [<reflink idref="bib53" id="ref39">53</reflink>]) increasingly supports the idea that birth order is an appropriate instrument for parental involvement. The identification strategy benefits from unique and rich data of about 9000 12-year old students and their parents in one country (the Netherlands). The data allow us to define parental involvement in the two distinct ways that deal with the school-supportive home climate (i.e. homework and parent–child communication on education), and include information on birth order of the child (see Section 3). Information on student and parent characteristics are also provided, such as education of the parents and income, native tongue and cultural differences of the parents, and which one of the parents mostly decides on school matters. For the outcome variable "student performance", we have detailed information on test scores from the national exam (CITO). CITO test scores measure children's proficiency at the end of primary education (on average 12 year olds) in a standardized way with respect to math, language and information processing. These exams are considered highly reliable and valid and, above all, owing to its standardized measurement, are comparable across students and schools (Zijsling et al. [<reflink idref="bib65" id="ref40">65</reflink>]). Additionally, we also use grade retention over the primary education school career as an outcome variable. We consider grade retention as closely related to the set of study results over primary education, and as a proxy for being at-risk of school failure (e.g. school dropout) (Snow [<reflink idref="bib61" id="ref41">61</reflink>]).</p> <p>This article proceeds as follows. Theory and evidence on the relationship between birth order and parental involvement are discussed in Section 2. Section 3 presents the data and descriptive statistics, and Section 4 the results. Section 5 deals with the robustness analyses. Section 6 concludes.</p> <hd id="AN0121413965-3">2. Preferential treatment of first-borns</hd> <p></p> <hd id="AN0121413965-4">2.1. Theory</hd> <p>Early literature discussed several mechanisms that may lead to the "preferential" treatment of first-borns compared with higher ranked children. These mechanisms are not only economically, but also biologically based. As the main economic reason, Behrman and Taubman ([<reflink idref="bib14" id="ref42">14</reflink>]) point to the household budget constraint. While parents tend to evenly engage themselves in the education of all of their children, they have to increasingly trade off their time investments in the children with household expenses in response to growing family size. If the family size increases, parents may have to adjust downwards their level of expenditures per child, while expenditures on the first-born child are already foregone. It is in this respect that Becker and co-authors earlier discussed the tradeoff between quantity and quality of children. Their model is often referred to as the QQ-model in the literature (Becker and Lewis [<reflink idref="bib12" id="ref43">12</reflink>]; Becker and Tomes [<reflink idref="bib13" id="ref44">13</reflink>]). Quality is defined as partly endowed and partly under (parents') control. They argue that: "It is sufficient to recognize that an increase in the quantity of children raises the cost or shadow price of the quality of children, and vice versa" (Becker and Tomes [<reflink idref="bib13" id="ref45">13</reflink>], S143). Behrman and Taubman ([<reflink idref="bib14" id="ref46">14</reflink>]) as follows discuss several biological reasons – on which evidence is scarce. They argue that higher order children may have lower genetic endowments (i.e. innate ability), because the older the mother, the likelier birth deficits happen. Or small family households may dislike parenting because of a difficult first-born or second-born child, and consequently, decide against having another child. Hanushek ([<reflink idref="bib31" id="ref47">31</reflink>]) adds that first-born children spend their first years in life as an only child in the family. Consequently, the child is more adult oriented and will imitate the parents more.</p> <p>A theoretical framework on preferential treatment of first-borns is also offered by Blundell, Chiappori, and Meghir ([<reflink idref="bib19" id="ref48">19</reflink>]). Within their framework, the authors add the decision power of the wife (relative to her partner) to the household budget constraint. This decision power may well denote the "bargaining power", i.e. the relative ability of the woman to exert influence in collective household decisions on investments, such as investments in the child. It is shown that, if the woman (man) gains in decision power relative to the man (woman) in the household, then this increases (decreases) investments in the child. Investments in the child can be partly captured as expenditures on goods, such as nutrition and clothing for the child, but also partly in the time spent with the child, briefly denoted by parental involvement.</p> <hd id="AN0121413965-5">2.2. Proposition</hd> <p>From Becker and Tomes ([<reflink idref="bib13" id="ref49">13</reflink>]); Behrman and Taubman ([<reflink idref="bib14" id="ref50">14</reflink>]); and Blundell, Chiappori, and Meghir ([<reflink idref="bib19" id="ref51">19</reflink>]), a testable proposition can be derived, namely: if the family size increases, then this requests not only more expenditures, but also more parental involvement, while, <emph>ceteris paribus</emph> parents' labor supply, the budget constraint remains the same. As such, relatively to expenditures, it becomes more expensive to invest in (additional) time spent with the child. Therefore, if family size increases, the child with the highest rank will get less parental involvement in education than the child with the lowest rank. These differences in parental involvement by birth order may impact academic achievement if, and only if, parental involvement in education is an important determinant of educational attainment. If so, then the oldest child benefits the most in terms of better educational outcomes than the younger child(ren).</p> <hd id="AN0121413965-6">2.3. Findings from the previous literature</hd> <p>Previous literature mainly focused on empirical estimates of the QQ-model (Cáceres-Delpiano [<reflink idref="bib20" id="ref52">20</reflink>]; Li, Zhang, and Zhu [<reflink idref="bib38" id="ref53">38</reflink>]). Angrist and co-authors (Angrist and Evans [<reflink idref="bib3" id="ref54">3</reflink>]; Angrist, Lavy, and Schlosser [<reflink idref="bib6" id="ref55">6</reflink>]) analyze the effect of increasing quantity of children on labor supply of both women and men. They take advantage of the sibling gender mix in households with at least two children in order to estimate an instrumental variables strategy. Hereby, Angrist and Evans ([<reflink idref="bib3" id="ref56">3</reflink>]) exploit parental preferences for sibling gender composition; and later also combine new sources of exogenous variation in family size in order to explore robustness of using twins as an instrument (e.g. Black, Devereux, and Salvanes [<reflink idref="bib17" id="ref57">17</reflink>]). Angrist, Lavy, and Schlosser ([<reflink idref="bib6" id="ref58">6</reflink>]) confirm the "stereotype" that childbearing impacts labor supply of especially lower educated women, but not of men. They additionally check for robustness of their results by using households with second-birth twins. By comparing households with three children with households with second-birth twins, the authors make advantage of the fact that a third-born child is always younger than the third-born twin. Angrist and Evans ([<reflink idref="bib3" id="ref59">3</reflink>]) conclude that it takes up till the child's age of 13 before the labor market consequences of bearing a third child disappears.</p> <p>Cáceres-Delpiano ([<reflink idref="bib20" id="ref60">20</reflink>]) and Lee ([<reflink idref="bib37" id="ref61">37</reflink>]) also used multiple births as an instrument for family size in order to explore its impact of child quantity and child wellbeing. Cáceres-Delpiano ([<reflink idref="bib20" id="ref62">20</reflink>]) finds results in line with the QQ-model of Becker and Lewis ([<reflink idref="bib12" id="ref63">12</reflink>]) and Becker and Tomes ([<reflink idref="bib13" id="ref64">13</reflink>]): twin births reduce preferential treatment of higher order children and also negatively impact the school career (e.g. decreased likelihood to attend private schools). Like in Angrist and Evans ([<reflink idref="bib3" id="ref65">3</reflink>]), Cáceres-Delpiano ([<reflink idref="bib20" id="ref66">20</reflink>]) identifies effects on mother's labor supply. The author also adds evidence to the literature that parents have an increased likelihood to divorce owing to bearing twins. Lee ([<reflink idref="bib37" id="ref67">37</reflink>]) uses preferences for sons in the particular context of South Korea as an instrument for family size. The author argues that, as long as parents do not abort girls, the gender of the first-born child is a good predictor of having more than one child.</p> <p>Price ([<reflink idref="bib52" id="ref68">52</reflink>], [<reflink idref="bib53" id="ref69">53</reflink>]) presents new valuable insights on preferential treatment of first-borns by explicitly focusing on the parents' time spent with their children. The author argues that first-borns receive a preferential treatment of about 20 to 30 minutes of quality time more each day than their higher order siblings. Hereby, he accounts for differences between families and children's ages. The author shows that the estimated effect is mainly driven by the age of the first-born child: parents spent equal time with each child at any given point in time, but increasingly less time when children grow older. The second-born child will receive less quality time when he/she reaches the same age as the older sibling.</p> <hd id="AN0121413965-7">3. Data</hd> <p></p> <hd id="AN0121413965-8">3.1. Student and parent characteristics</hd> <p>The empirical application deals with one country, namely: the Netherlands. We use data (Voortgezet Onderwijs Cohort Leerlingen, VOCL'99) from a representative cohort of 9126 students who enrolled in Grade 7 of secondary education in 1999–2000. Notwithstanding that we present results from a cohort of students enrolled in Dutch secondary education in 1999–2000, we argue from previous literature that the results are still relevant and valid for the current situation. In the systematic literature review of Ariës and Cabus ([<reflink idref="bib7" id="ref70">7</reflink>]) it is argued that the way parents engage in children's homework or learning for tests is correlated with student performance. However, these parental strategies for teaching their children are hard to influence or change. Moreover, the authors also indicate that the pre- and post-2001 estimates in the previous literature on the relationship between parental involvement and children's learning outcomes did not significantly change. In addition, Wilder ([<reflink idref="bib64" id="ref71">64</reflink>], 393) also argues that it did not matter how parental involvement in the education of their children was defined, "The results of the meta-synthesis indicated that this relationship was positive regardless of how parental involvement was defined." There is no reason to believe that the effect of parental involvement on children's learning outcomes drastically changed over time.</p> <p>At Grade 7, most students are 12 year-olds. These students constitute 721 classrooms that are part of 122 different schools. The data contain a rich set of student and parent characteristics, are partly composed of administrative data (i.e. the exam results on CITO), and partly of questionnaires (i.e. the questions on parental involvement). The descriptive statistics of these data are provided in the Supplemental Material, available online.</p> <p>The outcome variables of interest in this article are: (<reflink idref="bib1" id="ref72">1</reflink>) the test results on the standardized CITO exams of math, language, and information processing; and (<reflink idref="bib2" id="ref73">2</reflink>) a history of grade retention over the school career in primary education. CITO are national exams performed at the end of primary education. The test results of the CITO exam are used to give advice to the student and his parents on school type and secondary education track (pre-university level, general secondary level, and pre-vocational secondary level). About 80% of students follow the advice from the school based on CITO. The test is considered highly reliable and valid and, above all, owing to its standardized measurement, is comparable across students and schools (Zijsling et al. [<reflink idref="bib65" id="ref74">65</reflink>]).</p> <p>The average test scores on CITO of the full sample are 13.04 (math), 13.07 (language), and 12.53 (information processing). And 12.29% of students have a history of grade retention in primary education. Note that we standardize these outcomes variables in Section 4, so that the estimation output can be interpreted as standardized coefficients.</p> <p>About one in every two students is male, and 87.45% are Dutch (i.e. born in the Netherlands). Their parents are also most likely born in the Netherlands, married (87.42%) and Catholic (32.51%) or atheist (32.52%). About 82% of parents have at least a certificate of secondary education; 45% of parents have an annual household income higher than 27,200 euros.</p> <p>Students are part of a family with two to three children. One in every two students of our sample is first in line, 31.59% second in line, 12.69% third in line, and 6.33% fourth (or further) in line. Important to note is that in our data each group of children having the same birth order comes from different families. As such, observed differences in birth order are observed across, not within, families. As a result of using cross-section between-family data, each rank represents a "healthy" mix of student and family characteristics, including a lot of variation in age of the mother and socio-economic status of the family. For example, it is perfectly possible that a child with birth order 2 was born when his/her mother was 27, while another child with birth order 2 was born when his/her mother was 22. Accordingly, we observe a small and insignificant (<emph>p</emph>-value = 0.3170) correlation of 0.0105 between birth order and socio-economic status. The correlation between family size and socio-economic status is equal to 0.0528 and significant at 1% level.</p> <hd id="AN0121413965-9">3.2. Measures of parental involvement</hd> <p>The literature offers multiple definitions of parental involvement (see, for example, Driessen, Smit, and Sleegers [<reflink idref="bib24" id="ref75">24</reflink>]; Epstein [<reflink idref="bib26" id="ref76">26</reflink>]; Epstein and Sheldon [<reflink idref="bib27" id="ref77">27</reflink>]; McNeal [<reflink idref="bib45" id="ref78">45</reflink>]; Phtiaka [<reflink idref="bib51" id="ref79">51</reflink>]). In order to make parental involvement measureable, we are in need of a clear and unambiguous definition of what constitutes parental involvement. In this article we define two different types of parental involvement, namely: (<reflink idref="bib1" id="ref80">1</reflink>) involvement in homework; and (<reflink idref="bib2" id="ref81">2</reflink>) parent–child communication on school matters. The former type of involvement is considered an active form, as parents invest (additional) time in doing homework with their children (Phtiaka [<reflink idref="bib51" id="ref82">51</reflink>]). It also addresses McNeal's ([<reflink idref="bib45" id="ref83">45</reflink>]) concept of cultural capital. And in line with Epstein ([<reflink idref="bib26" id="ref84">26</reflink>]) and Epstein and Sheldon ([<reflink idref="bib27" id="ref85">27</reflink>]), parental involvement in homework deals with creating a supportive learning environment at home. Contrary, the latter type of involvement, parent–child communication, is rather passive, as parents can talk with the child on school matters, for example, over dinner. It does not necessarily request time investment, but instead reflects the school-supportive home climate (McNeal [<reflink idref="bib45" id="ref86">45</reflink>]).</p> <p>Both measures of parental involvement are included in the VOCL questionnaires. Factor analysis was used in order to construct scales from the underlying questions. The scale parental involvement in homework is a composite measure of questions (to parents) dealing with helping, hearing, controlling, and encouraging homework. The scale reliability coefficient Cronbach's alpha of the unstandardized items is equal to 0.7479. The scale parent–child communication is a composite measure of questions (to parents) dealing with asking (talking with) the child about results on exams, school matters, or what the child has learned at school. The scale reliability coefficient Cronbach's alpha of the unstandardized items is equal to 0.7045.</p> <hd id="AN0121413965-10">4. Estimates on parental involvement</hd> <p></p> <hd id="AN0121413965-11">4.1. Ordinary least squares (OLS)-estimates</hd> <p>We wish to estimate the effect of parental involvement on student achievement. The estimates with respect to the CITO test scores are significantly negative, and with respect to grade retention significantly positive. The effect sizes are small. These findings are counter-intuitive, as they indicate that parental involvement is negatively associated with student achievement, and positively associated with grade retention. Therefore, we argue that the relationship rather runs from academic achievement to parental involvement (i.e. reversed causality). It is in this respect that the literature indicates that students with rather low academic achievement also increasingly receive help and attention at home (Fan and Chen [<reflink idref="bib29" id="ref87">29</reflink>]; Hotz and Pantano [<reflink idref="bib33" id="ref88">33</reflink>]; Silinskas et al. [<reflink idref="bib60" id="ref89">60</reflink>]).</p> <hd id="AN0121413965-12">4.2. IV-estimates</hd> <p></p> <hd id="AN0121413965-13">4.2.1. Discussion of the assumptions</hd> <p>This article explores how we can deal with the aforementioned issues of endogeneity and reversed causality by using birth order as an instrument for parental involvement. There are a couple of assumptions underlying the use of instrumental variables in order to claim the estimates to be causal (Angrist, Imbens, and Rubin [<reflink idref="bib4" id="ref90">4</reflink>]). From the literature and the identification tests later, we argue that these assumptions can be met. First, consider the exclusion restriction. Birth order is randomly assigned, and, additionally, the child cannot affect its rank in the family. Birth order is, as such, exogenous to the child. In addition, fairly recent evidence from the previous literature (Damian and Roberts [<reflink idref="bib21" id="ref91">21</reflink>]; Price [<reflink idref="bib52" id="ref92">52</reflink>], [<reflink idref="bib53" id="ref93">53</reflink>]), increasingly supports the idea that birth order is an appropriate instrument for parental involvement. Recent findings from a representative sample of 377,000 US high school students indicate that personality or intelligence scores are not (or only negligible) affected by birth order (Damian and Roberts [<reflink idref="bib21" id="ref94">21</reflink>]). The authors chose for a between-family design and a set of similar control variables as in our present study (age, gender, family size, indicators for socio-economic status and family structure) in order to control for potential confounding factors. Note that in our study, we deal with exactly the same set-up as Damian and Roberts ([<reflink idref="bib21" id="ref95">21</reflink>]) – i.e. we only have (the birth order of) one child per family, also called a between-family design. As such, observed differences in parental involvement with increasing birth order are measured between (not within) families. Damian and Roberts ([<reflink idref="bib21" id="ref96">21</reflink>]) argue that birth order is not an important determinant of personality and intelligence, and, as such, this research supports the hypothesis that the main effects of birth order on student performance come from variations in parental involvement.</p> <p>Second, according to the theoretical and empirical studies with respect to family size and parental involvement (or other household outcomes), birth order should have a negative relationship with parental involvement, so that also the IV-assumption of monotonicity is met (Section 2). Theory indicates several mechanisms that drive this negative relationship, including: preferential treatment of first-borns (Behrman and Taubman [<reflink idref="bib14" id="ref97">14</reflink>]; Price [<reflink idref="bib52" id="ref98">52</reflink>]); the quantity–quality tradeoff (Becker and Tomes [<reflink idref="bib13" id="ref99">13</reflink>]); or parents' labor supply and bargaining power of the wife (Blundell, Chiappori, and Meghir [<reflink idref="bib19" id="ref100">19</reflink>]). Although family size and birth order are closely related to each other, they are still conceptually distinct. Family size is a constant for every child of the household and captures unobserved information at the household level. On the other hand, birth order is an individual (child) level instrument that differs between siblings and, controlling for family size in a multivariate regression, it is unrelated to parents' background or fertility decisions (Belmont and Marolla [<reflink idref="bib15" id="ref101">15</reflink>]; Black, Devereux, and Salvanes [<reflink idref="bib18" id="ref102">18</reflink>]). As such, we wish to estimate the effect of birth order on academic attainment through its (negative) association with parental involvement and conditional on (at least) family size. By making the estimates conditional on family size, we can avoid that birth order captures unobserved household level information.</p> <hd id="AN0121413965-14">4.2.2. First-stage estimates</hd> <p>Consider the first-stage regression of the most simple model specification that does not consider household income, decision power, or student–parent characteristics:</p> <p>(<reflink idref="bib1" id="ref103">1</reflink>)</p> <p>Graph</p> <p>where denotes parental involvement and , the rank of the child, the instrument. Note that we include dummy variables for each rank instead of a continuous variable <emph>R</emph>, as we do not wish to make assumptions on the linearity of the relationship between rank of child and parental involvement (for a discussion on linearity, see also Angrist and Evans [<reflink idref="bib3" id="ref104">3</reflink>]; Black, Devereux, and Salvanes [<reflink idref="bib18" id="ref105">18</reflink>]; Hanushek [<reflink idref="bib31" id="ref106">31</reflink>]). The basic first-stage model specification controls for family size in order to control for the relationship between rank of the child and family size.</p> <p>Table 1 presents the results of the first-stage regression as specified in Equation (<reflink idref="bib1" id="ref107">1</reflink>). Note that each model clusters the standard error at the level of the classroom (Murnane and Willett [<reflink idref="bib46" id="ref108">46</reflink>]). The results indicate a clear negative relationship between birth order and parental involvement. The size of the effect increases with birth order. One can easily control for student and parent characteristics (including household income) by adding covariates to Equation (<reflink idref="bib1" id="ref109">1</reflink>). These estimates are also presented in the second column of Table 1. The results indicate that the estimates of birth order are robust to including control variables. Several identification tests have been performed (the under-identification test, the weak identification test, and the over-identification test). These or additional statistics on the validity of IV are available from the authors on request. In summary, the statistics show that: (<reflink idref="bib1" id="ref110">1</reflink>) the equation is well-identified; (<reflink idref="bib2" id="ref111">2</reflink>) the instrument has significant explanatory power; and (<reflink idref="bib3" id="ref112">3</reflink>) the instrument is uncorrelated with the error term. However, these latter results should be interpreted with caution, as using a flexible birth order specification still does not provide us with multiple instruments (necessary for the over-identification test).</p> <p>Table 1. Results of the first stage regression (parental involvement in homework = standardized scale).</p> <p> <ephtml> <table><thead><tr valign="top"><td /><td>Without control variables</td><td>With control variables</td></tr></thead><tbody><tr valign="top"><td><italic>Instrumental variable (IV)</italic></td><td /><td /></tr><tr valign="top"><td>First in line = reference</td><td /><td /></tr><tr valign="top"><td>Second in line</td><td char=".">−0.1362<sup>***</sup></td><td char=".">−0.1330<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0251)</td><td char=".">(0.0245)</td></tr><tr valign="top"><td>Third in line</td><td char=".">−0.1773<sup>***</sup></td><td char=".">−0.1667<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0363)</td><td char=".">(0.0354)</td></tr><tr valign="top"><td>Fourth in line</td><td char=".">−0.2188<sup>***</sup></td><td char=".">−0.1992<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0534)</td><td char=".">(0.0516)</td></tr><tr valign="top"><td /><td /><td /></tr><tr valign="top"><td>Family size</td><td char=".">−0.0425<sup>***</sup></td><td char=".">−0.0313<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0132)</td><td char=".">(0.0133)</td></tr><tr valign="top"><td><italic>IV and matching</italic></td><td /><td /></tr><tr valign="top"><td>First in line = reference</td><td /><td /></tr><tr valign="top"><td>Second in line</td><td char=".">−0.1431<sup>***</sup></td><td char=".">−0.1369<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0249)</td><td char=".">(0.0243)</td></tr><tr valign="top"><td>Third in line</td><td char=".">−0.1783<sup>***</sup></td><td char=".">−0.1665<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0365)</td><td char=".">(0.0355)</td></tr><tr valign="top"><td>Fourth in line</td><td char=".">−0.2074<sup>***</sup></td><td char=".">−0.2083<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">(0.0537)</td><td char=".">(0.0523)</td></tr><tr valign="top"><td>Family size</td><td char=".">−0.0408</td><td char=".">−0.0321</td></tr><tr valign="top"><td /><td char=".">(0.0131)</td><td char=".">(0.0133)</td></tr><tr valign="top"><td>Number of clusters</td><td>721</td><td>721</td></tr><tr valign="top"><td>Number of observations</td><td>9126</td><td>9126</td></tr></tbody></table> </ephtml> </p> <p>1 Note: Robust standard errors between brackets.</p> <hd id="AN0121413965-15">4.2.3. Second-stage estimates</hd> <p>In the second-stage regression, we estimate the effect of parental involvement on standardized test scores that only comes through the effect of instrument, birth order. We then may write:</p> <p>(<reflink idref="bib2" id="ref113">2</reflink>)</p> <p>Graph</p> <p>where the estimate of interest is <emph>θ.</emph> The results from the second-stage regressions as specified in Equation (<reflink idref="bib2" id="ref114">2</reflink>) are presented in Table 2. Using the birth order of the child as an instrumental variable, the second stage results show a clear positive effect of parental involvement on academic achievement. The effect sizes of math (0.6690), language (0.9494), and information processing (1.3206) are medium to large, whereas the effect size of grade retention (−0.3212) is small.</p> <p>Table 2. Summary of the estimation output (parental involvement in homework = standardized scale).</p> <p> <ephtml> <table><thead><tr valign="top"><td /><td>Math</td><td>Language</td><td>Information processing</td><td>Grade retention</td></tr></thead><tbody><tr valign="top"><td>OLS</td><td>−0.1367<sup>***</sup></td><td>−0.1743<sup>***</sup></td><td>−0.1879<sup>***</sup></td><td>−0.1704<sup>***</sup></td><td>−0.1583<sup>***</sup></td><td>−0.1710<sup>***</sup></td><td>0.0609<sup>***</sup></td><td>0.0519<sup>***</sup></td></tr><tr valign="top"><td /><td>(0.0118)</td><td>(0.0106)</td><td>(0.0116)</td><td>(0.0109)</td><td>(0.0114)</td><td>(0.0108)</td><td>(0.0112)</td><td>(0.0103)</td></tr><tr valign="top"><td>IV</td><td>0.6690<sup>***</sup></td><td>0.6490<sup>***</sup></td><td>0.9494<sup>***</sup></td><td>1.0209<sup>***</sup></td><td>1.3206<sup>***</sup></td><td>1.3710<sup>***</sup></td><td>−0.3212<sup>**</sup></td><td>−0.3507<sup>**</sup></td></tr><tr valign="top"><td /><td>(0.2020)</td><td>(0.2003)</td><td>(0.2389)</td><td>(0.2454)</td><td>(0.2798)</td><td>(0.2905)</td><td>(0.1463)</td><td>(0.1392)</td></tr><tr valign="top"><td>IV and matching</td><td>0.6647<sup>***</sup></td><td>0.6833<sup>***</sup></td><td>0.9955<sup>***</sup></td><td>1.0838<sup>***</sup></td><td>1.3410<sup>***</sup></td><td>1.4068<sup>***</sup></td><td>−0.3008<sup>**</sup></td><td>−0.3067<sup>**</sup></td></tr><tr valign="top"><td /><td>(0.2010)</td><td>(0.1999)</td><td>(0.2458)</td><td>(0.2507)</td><td>(0.2824)</td><td>(0.2904)</td><td>(0.1390)</td><td>(0.1355)</td></tr><tr valign="top"><td>Control Variables</td><td>No<sup>a</sup></td><td>Yes<sup>b</sup></td><td>No<sup>a</sup></td><td>Yes<sup>b</sup></td><td>No<sup>a</sup></td><td>Yes<sup>b</sup></td><td>No<sup>aa</sup></td><td>Yes<sup>b</sup></td></tr><tr valign="top"><td>Number of clusters</td><td>721</td><td>721</td><td>721</td><td>721</td><td>721</td><td>721</td><td>721</td><td>721</td></tr><tr valign="top"><td>Number of observations</td><td>9126</td><td>9126</td><td>9126</td><td>9126</td><td>9126</td><td>9126</td><td>9103</td><td>9103</td></tr></tbody></table> </ephtml> </p> <ulist> <item>2 No control variables except family size.</item> <item>3 These control variables include: family size, gender, age, ethnicity, marital status, religion, language, culture, education, and income.</item> <item>4 Robust standard errors clustered at the level of the class between brackets.</item> </ulist> <hd id="AN0121413965-16">4.2.4. IV-estimates with matching of parents</hd> <p>In a third and final model specification, we additionally suggest matching of the parents who are not or only to a limited extent involved in the child's homework to parents who frequently are. Propensity score matching is most suitable in this respect, as it would only compare parents whose household situation is (almost) identical based on the observed parent characteristics (Rubin [<reflink idref="bib57" id="ref115">57</reflink>], [<reflink idref="bib58" id="ref116">58</reflink>]). Hereby, we can check whether households with low versus high levels of involvement in school matters are different based on their background characteristics. In order to match parents, we first create a dummy variable from the frequency variable with respect to parental involvement in homework. Parents who indicate "no help" (45%) on this variable receive the value of 0, and 1 if otherwise.</p> <p>The Probit results on which matching of the parents are based (cf. propensity score matching; Rubin [<reflink idref="bib58" id="ref117">58</reflink>]) are available from the authors upon request. Figure 1 plots the overlap in the covariate distributions between low versus high levels of parental involvement in homework. We observe a strong overlap in parent characteristics, indicating that high and low levels of parental involvement in homework can be observed in households with very different socio-economic states. Results of the "IV and matching" model specification are presented in Table 1. We conclude that the IV approach with matching yield comparable estimates to the IV approach without matching. These findings confirm that our IV-results are robust to matching of parents, making matching of parents unnecessary in further robustness checks.</p> <p>Graph: Figure 1. Distribution of the propensity scores of parents helping and not helping in homework.</p> <hd id="AN0121413965-17">5. Robustness analysis</hd> <p>The estimates from the robustness analysis are summarized in Table 3. Each model controls for student–parent characteristics. Results from first-stage regressions are not presented in tables in this article, but they are available from the authors.</p> <p>Table 3. Summary of the estimation output (parental involvement = standardized scale).</p> <p> <ephtml> <table><thead><tr valign="top"><td /><td /><td>Model 1 </td><td>Model 2</td><td>Model 3 </td><td>Model 4</td><td>Model 5</td></tr><tr valign="top"><td>Full model</td><td>Family size > 1</td><td>Both parents</td><td>Only mother</td><td>Only father</td></tr></thead><tbody><tr valign="top"><td>1. Math</td><td /><td /><td /><td /><td /></tr><tr valign="top"><td>  Parent–child</td><td>0.6490<sup>***</sup></td><td>0.6379<sup>***</sup></td><td>0.7572<sup>***</sup></td><td>0.5870<sup>**</sup></td><td>−0.2763</td></tr><tr valign="top"><td>  homework</td><td>(0.2003)</td><td>(0.1936)</td><td>(0.2857)</td><td>(0.2826)</td><td>(0.5950)</td></tr><tr valign="top"><td>  Parent–child</td><td>0.7358<sup>***</sup></td><td>0.6859<sup>***</sup></td><td>0.5397<sup>**</sup></td><td>0.7832<sup>**</sup></td><td>−0.2994</td></tr><tr valign="top"><td>  Communication</td><td>(0.2178)</td><td>(0.2075)</td><td>(0.2352)</td><td>(0.3531)</td><td>(1.9393)</td></tr><tr valign="top"><td>2. Language</td><td /><td /><td /><td /><td /></tr><tr valign="top"><td>  Parent–child</td><td>1.0209<sup>***</sup></td><td>1.0298<sup>***</sup></td><td>1.0691<sup>***</sup></td><td>0.9987<sup>***</sup></td><td>−0.3070</td></tr><tr valign="top"><td>  homework</td><td>(0.2454)</td><td>(0.2409)</td><td>(0.3386)</td><td>(0.3473)</td><td>(0.5519)</td></tr><tr valign="top"><td>  Parent–child</td><td>0.8543<sup>***</sup></td><td>0.9637<sup>***</sup></td><td>0.5621<sup>**</sup></td><td>1.0031<sup>**</sup></td><td>−0.2312</td></tr><tr valign="top"><td>  Communication</td><td>(0.2288)</td><td>(0.2353)</td><td>(0.2270)</td><td>(0.3873)</td><td>(1.8355)</td></tr><tr valign="top"><td>3. Information processing </td></tr><tr valign="top"><td>  Parent–child</td><td>1.3710<sup>***</sup></td><td>1.3640<sup>***</sup></td><td>1.3375<sup>***</sup></td><td>1.3147<sup>***</sup></td><td>−0.2398</td></tr><tr valign="top"><td>  homework</td><td>(0.2905)</td><td>(0.2808)</td><td>(0.3912)</td><td>(0.4106)</td><td>(0.5550)</td></tr><tr valign="top"><td>  Parent–child</td><td>1.1948<sup>***</sup></td><td>1.2676<sup>***</sup></td><td>0.8404<sup>***</sup></td><td>1.3216<sup>***</sup></td><td>0.3422</td></tr><tr valign="top"><td>  Communication</td><td>(0.2797)</td><td>(0.2822)</td><td>(0.2795)</td><td>(0.4632)</td><td>(2.1077)</td></tr><tr valign="top"><td>4. Grade retention</td></tr><tr valign="top"><td>  Parent–child</td><td>−0.3507<sup>**</sup></td><td>−0.3250<sup>**</sup></td><td>−0.4019<sup>**</sup></td><td>−0.3097</td><td>−0.3854</td></tr><tr valign="top"><td>  homework</td><td>(0.1392)</td><td>0.1347</td><td>(0.1892)</td><td>(0.2215)</td><td>(0.6619)</td></tr><tr valign="top"><td>  Parent–child</td><td>−0.3315<sup>**</sup></td><td>−0.2998<sup>**</sup></td><td>−0.3168<sup>*</sup></td><td>−0.3684</td><td>−0.5320</td></tr><tr valign="top"><td>  Communication</td><td>(0.1539)</td><td>(0.1505)</td><td>(0.1766)</td><td>(0.2551)</td><td>(2.2854)</td></tr><tr valign="top"><td>Number of clusters</td><td>721</td><td>721</td><td>718</td><td>715</td><td>285</td></tr><tr valign="top"><td>Number of observations</td><td>9126</td><td>7741</td><td>4426</td><td>3844</td><td>378</td></tr></tbody></table> </ephtml> </p> <p>5 Notes: Specification IV without matching and standard errors clustered at the level of the class. Each model also controls for all student and parent characteristics. These control variables include: family size, gender, age, ethnicity, marital status, religion, language, culture, education, and income.</p> <hd id="AN0121413965-18">5.1. Parent–child communication</hd> <p>First, consider the results of using another definition/measure of parental involvement, namely: the parent–child communication on school matters at home. This measure could be especially interesting, because it takes parents less effort to ask children about their stories of the day than helping them with homework, but, nonetheless, also creates a school-supportive home climate. We summarize the first-stage estimates of birth order as follows (first in line as the reference category): −0.12 significant at 1% level (second in line); −0.08 significant at 1% level (third in line); and −0.23 significant at 1% level (fourth in line). Overall, the magnitude of these first-stage estimates are somewhat smaller than those presented in Table 1, but it does not alter the relationship between birth order and parental involvement. It is also worth noting that the conclusions from the identification tests remain unchanged.[<reflink idref="bib1" id="ref118">1</reflink>] These statistics are available from the authors upon request. The second-stage estimates of Table 3 indicate that parent–child communication is effective in boosting academic performance. The results of the full model specification indicate medium to large effect sizes with respect to math (0.7358), language (0.8543), and information processing (1.1948). With respect to grade retention, the effects size (−0.3315) is small and significant at 5% level. In general, the estimates with respect to parent–child communication are relatively lower than those with respect to homework involvement.</p> <hd id="AN0121413965-19">5.2. At least two children in the family</hd> <p>Angrist and Evans ([<reflink idref="bib3" id="ref119">3</reflink>]) argue for a model specification that includes households with at least two children. As such, we exclude households with only one child from the data and then re-estimate Equations (<reflink idref="bib1" id="ref120">1</reflink>) and (<reflink idref="bib2" id="ref121">2</reflink>). The results of the second-stage regression are highly comparable in magnitude compared to the full model specification in the first column of Table 3. As such, we conclude that our estimates are robust to this model specification. Furthermore, we also have checked whether our findings change when one would report the results separately for families with two, three, or four children. For example, we ran a regression estimating the effects of parental involvement in homework on math performance keeping only observations for families with two children. The second-stage results from such regression is equal to 0.5472 significant at 5% level (<emph>N</emph> = 3552). We repeated this approach for families with three children (second-stage result equal to 0.6596 significant at 5% level; <emph>N</emph> = 2642) and for families with four or more children (second-stage result equal to 0.9068 not significant; <emph>N</emph> = 932).</p> <hd id="AN0121413965-20">5.3. Decision power of the wife</hd> <p>We observe that most often only the wife (42.51%) makes the educational choices, or both parents (49.08%) (variable "decision power"). A small share of fathers (4.18%) indicates that they have the decision power on these school matters, and not the wife. However, mostly mothers (69.6%) filled in the VOCL questionnaires for parents, whereas only a small share of fathers (23.9%), and a very small share of caregivers (0.4%) did so. Therefore, this self-reported variable should be interpreted with caution. Following Blundell, Chiappori, and Meghir ([<reflink idref="bib19" id="ref122">19</reflink>]), we estimate three models (Models 3–5) that consider the decision power of the wife (relative to her partner) on the education of their child. We then may estimate Equation (<reflink idref="bib2" id="ref123">2</reflink>) separately by mother, father, and both parents, as to only compare families wherein only the mother (Model 3), only the father (Model 5), or both parents (Model 4) make the decisions on their child's education. Table 3 presents the results of the second-stage regression. Our findings indicate a special role of the mother in the family, as, overall, her involvement in communication with the child on school matters clearly significantly boosts the academic performance of the child, especially on language and information processing. These findings are in line with Björklund, Lindahl, and Lindquist ([<reflink idref="bib16" id="ref124">16</reflink>]). However, when it comes to math performance and grade retention, it is better that both parents unduly interfere with school.</p> <hd id="AN0121413965-21">5.4. Differential effectiveness</hd> <p>Table 4 presents the second-stage results by immigrant status, gender and socio-economic status. The results from Table 4 show that largest effects can be found for Dutch female students with relatively high socio-economic status. Contrary, we cannot present significant results for immigrant and low socio-economic status students. In combination with previous estimates, we conclude that low socio-economic status and immigrant families are as much involved in the education of their children as the average Dutch family (see Table 3, specification "IV and matching"), but their involvement is less effective in terms of children's learning outcomes. Hence, the insignificant results for these two subgroups. These findings can be explained by the differential effectiveness of parents teaching or helping strategies (Ariës and Cabus [<reflink idref="bib7" id="ref125">7</reflink>]), but also because of parental competence to help with homework (Dumont et al. [<reflink idref="bib25" id="ref126">25</reflink>]).</p> <p>Table 4. Second-stage results by immigrant status, gender and socio-economic status (SES).</p> <p> <ephtml> <table><thead><tr valign="top"><td> </td><td>Model A</td><td>Model B</td><td>Model C</td><td>Model D</td><td>Model E</td><td>Model F</td><td>Model G</td></tr><tr valign="top"><td>Full model</td><td>Only Dutch</td><td>Only immigrants</td><td>Only Boys</td><td>Only girls</td><td>Low SES</td><td>High SES</td></tr></thead><tbody><tr valign="top"><td><italic>IV: Second stage</italic></td><td /><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Math</td><td char=".">0.6490<sup>***</sup></td><td char=".">0.6683<sup>***</sup></td><td char=".">0.0222</td><td char=".">0.4191**</td><td char=".">0.6481**</td><td char=".">0.5795</td><td char=".">0.6575<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">−0.2003</td><td char=".">0.1892</td><td char=".">0.3994</td><td char=".">0.1869</td><td char=".">0.3295</td><td char=".">1.1603</td><td char=".">0.1919</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Language</td><td char=".">1.0209<sup>***</sup></td><td char=".">0.9304<sup>***</sup></td><td char=".">−0.1747</td><td char=".">0.7464<sup>***</sup></td><td char=".">1.1947<sup>***</sup></td><td char=".">0.2795</td><td char=".">1.0315<sup>***</sup></td></tr><tr valign="top"><td /><td char=".">−0.2454</td><td char=".">0.2183</td><td char=".">0.4028</td><td char=".">0.2335</td><td char=".">0.4315</td><td char=".">1.0474</td><td char=".">0.2282</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Information</td><td char=".">1.3710<sup>***</sup></td><td char=".">1.2802<sup>***</sup></td><td char=".">−0.7527</td><td char=".">1.1069<sup>***</sup></td><td char=".">1.2945<sup>***</sup></td><td char=".">0.2795</td><td char=".">1.0315<sup>***</sup></td></tr><tr valign="top"><td>processing</td><td char=".">−0.2905</td><td char=".">0.2617</td><td char=".">0.5274</td><td char=".">0.2831</td><td char=".">0.4459</td><td char=".">1.0474</td><td char=".">0.2282</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Grade</td><td char=".">−0.3507**</td><td char=".">−0.4090<sup>***</sup></td><td char=".">0.3857</td><td char=".">−0.4426<sup>***</sup></td><td char=".">−0.0839</td><td char=".">−0.8972</td><td char=".">−0.3126<sup>***</sup></td></tr><tr valign="top"><td>retention</td><td char=".">−0.1392</td><td char=".">0.1345</td><td char=".">0.4377</td><td char=".">0.1623</td><td char=".">0.2080</td><td char=".">1.1658</td><td char=".">0.1294</td></tr><tr valign="top"><td>Number of clusters</td><td>721 </td><td>717 </td><td>495 </td><td>717 </td><td>707 </td><td>618 </td><td>719 </td></tr><tr valign="top"><td>Number of observations</td><td>9126 </td><td>7981 </td><td>1145 </td><td>4588 </td><td>4538 </td><td>1652 </td><td>7451 </td></tr></tbody></table> </ephtml> </p> <p>6 Notes: Specification IV without matching and standard errors clustered at the level of the class. Each model also controls for all student and parent characteristics. These control variables include: family size, gender, age, ethnicity, marital status, religion, language, culture, education, and income. High socio-economic status was mainly based on parents' education (at least upper secondary education or higher). As such, parents with no diploma (ISCED 0–2) are considered having low socio-economic status.</p> <hd id="AN0121413965-22">5.5. Reduced form (RF) regression and indirect least squares (ILS)</hd> <p>The estimates of reduced form (RF) regression and indirect least squares (ILS) are presented in Table 5. If birth order only works through parental involvement, then, first, we should observe that student performance decreases with increasing birth order. The RF coefficients for Math test scores indicate a small and significant negative direct effect of birth order on math test scores (effect size = −0.0852). This negative sign and the magnitude of the effect remain unchanged when considering performance on language and information processing. Logically, we observe the opposite positive sign for grade retention. If birth order only works through parental involvement, then, second we should observe that ILS yields similar results as in the case of second-stage IV. The results indicate that ILS coefficients and second-stage IV coefficients are exactly the same.</p> <p>Table 5. Reduced form regression, indirect least squares (ratio), and IV-estimates.</p> <p> <ephtml> <table><thead><tr valign="top"><td> </td><td> </td><td>1. Math</td><td> </td><td>2. Language</td><td>3. Information processing</td><td>4. Grade retention</td></tr></thead><tbody><tr valign="top"><td>Reduced form (RF)</td><td>−0.0852<sup>***</sup></td><td>−0.0878<sup>***</sup></td><td>−0.1313<sup>***</sup></td><td>0.0143<sup>***</sup></td></tr><tr valign="top"><td /><td /><td>(0.0153)</td><td>(0.0149)</td><td>(0.0152)</td><td>(0.0047)</td></tr><tr valign="top"><td>First stage (FS)</td><td>−0.0885<sup>***</sup></td><td>−0.0885<sup>***</sup></td><td>−0.0885<sup>***</sup></td><td>−0.0885<sup>***</sup></td></tr><tr valign="top"><td /><td /><td>(0.0148)</td><td>(0.0148)</td><td>(0.0148)</td><td>(0.0148)</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Ratio: RF/FS</td><td>0.9632</td><td>0.9929</td><td>1.4836</td><td>−0.1618</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>IV-estimates</td><td>0.9632<sup>***</sup></td><td>0.9929<sup>***</sup></td><td>1.4837<sup>***</sup></td><td>−0.1618</td></tr><tr valign="top"><td /><td /><td>(0.2533)</td><td>(0.2624)</td><td>(0.3210)</td><td>(0.0604)</td></tr><tr valign="top"><td /><td /><td /><td /><td /><td /></tr><tr valign="top"><td>Number of clusters</td><td>721</td><td>721</td><td>721</td><td>721</td></tr><tr valign="top"><td>Number of observations</td><td>9126</td><td>9126</td><td>9126</td><td>9103</td></tr></tbody></table> </ephtml> </p> <p>7 Note: Output using a linear specification for birth order. Robust standard errors clustered at the level of the class between brackets. Each model controls for family size.</p> <hd id="AN0121413965-23">6. Conclusion</hd> <p>This article discusses the effects of parental involvement on academic achievement. Birth order has been used as an instrument for parental involvement in order to account for endogeneity issues. The previous literature argues a robust negative effect of birth order of the child on parental involvement (e.g. Angrist and Evans [<reflink idref="bib3" id="ref127">3</reflink>]; Becker and Tomes [<reflink idref="bib13" id="ref128">13</reflink>]; Price [<reflink idref="bib52" id="ref129">52</reflink>]). Our first-stage findings are in line with the results of previous studies, namely: first-borns benefit from preferential treatment. We confirm that older children, on average, receive more parental involvement than the youngest of the family. The estimated effect sizes are relatively small, but do steadily increase with growing family size. Moreover, the first-stage estimates are highly robust to several model specifications, including matching of the parents, using two different definitions of parental involvement, and adding control variables. The validity of the instrument is also discussed in this article, and meets the conventions.</p> <p>For the second stage, we estimate positive effects of parental involvement in homework on academic achievement with medium to large effect sizes for math (0.66), language (0.99), and information processing (1.34), and small effect sizes −0.32 for grade retention. Furthermore, we observe that academic performance is rooted in a school-supportive home climate that is often created by the mother. Parental homework involvement positively influences language skills the most. Using another definition of parental involvement, namely parent–child communication on school matters, does not change our conclusions. The second-stage estimates are robust to various model specifications and control variables.</p> <p>Our study has limitations. Mammen ([<reflink idref="bib43" id="ref130">43</reflink>]) argues that not only child quantity matters, but also the sibling gender composition. The author finds that fathers tend to invest more time in the children when there is a son in the family compared to all-girls families. McGuire and Shanahan ([<reflink idref="bib44" id="ref131">44</reflink>]) also refer to the differential effects of siblings, birth order, and the parent–child relationship (compared to siblings) on child outcomes across context and family types. However, the data did not include information on gender of the siblings, so that we could not account for this. Second, we could not control for age of the parents at birth, too. Older age of the parents at birth may be associated with lower child endowments (Behrman and Taubman [<reflink idref="bib14" id="ref132">14</reflink>]; Price [<reflink idref="bib52" id="ref133">52</reflink>]). However, evidence from Statistics Netherlands (cbs.nl, 2015) indicates several general trends in parents' child wish. The average age of the mother giving birth to her first child increases with her own fathers' level of education and parents' family composition. If fathers' level of education is rather high, then women will postpone their child wish. And women coming from families with multiple siblings are more likely to become mothers at an earlier age. Additionally, women who grew up in single parent families will start earlier motherhood, while choosing for smaller families. These general trends imply that parents with high socio-economic status choose for smaller families with the mother being relatively old, and parents with low socio-economic status choose for bigger families with the mother being relative young. Based on these general trends, one can expect a lot of variation in mothers' age at first-born and later born children within and across families, depending on socio-economic status and family history. The potential association between birth order and socio-economic status then rather comes from decisions on family size. Especially for these reasons, we controlled for family size and indicators of socio-economic status in the multivariate regression.</p> <p>To conclude, some implications for policy are discussed. The policy debate could further discuss the extent to which schools have a role in encouraging parental involvement in education. School policy could establish better didactic partnerships with the parents by taking the position of the child in the household into parent–teacher discussions. National policy could focus more on the fathers of the household, as they are, compared to the mothers, less actively making decisions about their children's education. Following Kluve and Tamm ([<reflink idref="bib35" id="ref134">35</reflink>]), parental leave regulations can play an important role.</p> <hd id="AN0121413965-24">Disclosure statement</hd> <p>No potential conflict of interest was reported by the authors.</p> <hd id="AN0121413965-25">Supplemental data</hd> <p>The supplementary material for this paper is available online at <ulink href="http://dx.doi.org/10.1080/00131911.2016.1208148">http://dx.doi.org/10.1080/00131911.2016.1208148</ulink>.</p> <hd id="AN0121413965-26">Acknowledgements</hd> <p>This paper benefitted from comments and discussions with Wim Groot, Henriette Maassen van den Brink, Joris Ghysels, TIER seminar participants, and participants of the 5th international workshop on Applied Economics of Education (IWAEE 2014). We also express our gratitude to the referees of <emph>Educational Review</emph>. The usual caveat applies.</p> <ref id="AN0121413965-27"> <title> Note </title> <blist> <bibl id="bib1" idref="ref22" type="bt">1</bibl> <bibtext> These statistics with regard the basic model specification include: the under-identification test (Chi-square = 37.150; <emph>p</emph>-values = 0.0000); the weak identification test (<emph>F</emph> =13.960 > 10.000); and the over-identification test using birth order as dummy variables (Hansen <emph>J</emph> statistics = 5.063; <emph>p</emph>-values = 0.0795).</bibtext> </blist> </ref> <ref id="AN0121413965-28"> <title> References </title> <blist> <bibtext> Albouy, Valerie, and Laurent Lequien. 2009. "Does Compulsory Education Lower Mortality?" Journal of Health Economics 28 (1): 155–168. doi:10.1016/j.jhealeco.2008.09.003.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref12" type="bt">2</bibl> <bibtext> Alexander, Karl L., Doris R. Entwisle, and Linda Steffel Olson. 2007. "Lasting Consequences of the Summer Learning Gap." American Sociological Review 72 (2): 167–180. doi:10.1177/000312240707200202.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref54" type="bt">3</bibl> <bibtext> Angrist, Joshua D., and William N. Evans. 1998. "Children and Their Parents' Labor Supply: Evidence from Exogenous Variation in Family Size." The American Economic Review 88 (3): 450–477. doi:10.1086/653830.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref90" type="bt">4</bibl> <bibtext> Angrist, Joshua D., Guido W. Imbens, and Donald B. Rubin. 1996. "Identification of Causal Effects Using Instrumental Variables." Journal of the American Statistical Association 91 (434): 444–455. doi:10.1080/01621459.1996.10476902.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref29" type="bt">5</bibl> <bibtext> Angrist, Joshua David, and Jörn-Steffen Pischke. 2009. Mostly Harmless Econometrics: An Empiricist's Companion. vol. 1. Princeton, NJ: Princeton University Press.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref55" type="bt">6</bibl> <bibtext> Angrist, Joshua D., Victor Lavy, and Analia Schlosser. 2010. "Multiple Experiments for the Causal Link between the Quantity and Quality of Children." Journal of Labor Economics 28 (4): 773–824. 10.1086/653830</bibtext> </blist> <blist> <bibl id="bib7" idref="ref26" type="bt">7</bibl> <bibtext> Ariës, Roel J. F. J., and Sofie J. Cabus. 2015. "Parental Homework Involvement Improves Test Scores? A Review of the Literature." Review of Education 3 (2): 179–199. doi:10.1002/rev3.3055.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref21" type="bt">8</bibl> <bibtext> Ariës, Roel Jacobus, Wim Groot, and Henriette Maassen Brink. 2015. "Improving Reasoning Skills in Secondary History Education by Working Memory Training." British Educational Research Journal 41 (2): 210–228. doi:10/1002.berj3142.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref31" type="bt">9</bibl> <bibtext> Avvisati, Francesco, Bruno Besbas, and Nina Guyon. 2011. "Parental Involvement in School: A Literature Review." Revue D'économie Politique 120 (5): 759–778. doi:10.3917/redp.205.0759.</bibtext> </blist> <blist> <bibtext> Avvisati, Francesco, M. Gurgand, N. Guyon, and E. Maurin. 2014. "Getting Parents Involved: A Field Experiment in Deprived Schools." Review of Economic Studies 81 (3): 57–83. doi:10.1093/restud/rdt027.</bibtext> </blist> <blist> <bibtext> Baumert, Jürgen, and Anke Demmrich. 2001. "Test Motivation in the Assessment of Student Skills: The Effects of Incentives on Motivation and Performance." European Journal of Psychology of Education 16 (3): 441–462. doi:10.1007/BF03173192.</bibtext> </blist> <blist> <bibtext> Becker, Gary S, and H. Gregg Lewis. 1973. "On the Interaction between the Quantity and Quality of Children." Journal of Political Economy 81(2), Part 2: New Economic Approaches to Fertility: S279–S288.</bibtext> </blist> <blist> <bibtext> Becker, Gary S., and N. Tomes. 1976. "Child Endowments and the Quantity and Quality of Children." Journal of Political Economy 84(4), Part 2: Essays in Labor Economics in Honor of H. Gregg Lewis: S143–S162.</bibtext> </blist> <blist> <bibtext> Behrman, Jere R., and Paul Taubman. 1986. "Birth Order, Schooling, and Earnings." Journal of Labor Economics 4 (3): S121–S145. 10.1086/298124</bibtext> </blist> <blist> <bibtext> Belmont, Lillian, and Francis A. Marolla. 1973. "Birth Order, Family Size, and Intelligence a Study of a Total Population of 19-Year-Old Men Born in the Netherlands is Presented." Science 182 (4117): 1096–1101. doi:10.1126/science.182.4117.1096.</bibtext> </blist> <blist> <bibtext> Björklund, Anders, Lena Lindahl, and Matthew J. Lindquist. 2010. "What More than Parental Income, Education and Occupation? An Exploration of What Swedish Siblings Get from Their Parents." The BE Journal of Economic Analysis & Policy 10 (1): 1–38.</bibtext> </blist> <blist> <bibtext> Black, Sandra E., Paul J. Devereux, and Kjell G. Salvanes. 2005. "The More the Merrier? The Effect of Family Size and Birth Order on Children's Education." The Quarterly Journal of Economics 120 (2): 669–700.</bibtext> </blist> <blist> <bibtext> Black, Sandra E., Paul J. Devereux, and Kjell G. Salvanes. 2011. "Older and Wiser? Birth Order and IQ of Young Men." CESifo Economic Studies 57 (1): 103–120. doi:10.1093/cesifo/ifq022.</bibtext> </blist> <blist> <bibtext> Blundell, Richard, Pierre-André Chiappori, and Costas Meghir. 2005. "Collective Labor Supply with Children." Journal of Political Economy 113 (6): 1277–1306. doi:10.1086/491589.</bibtext> </blist> <blist> <bibtext> Cáceres-Delpiano, Julio. 2006. "The Impacts of Family Size on Investment in Child Quality." Journal of Human Resources 41 (4): 738–754. doi:10.3368/jhr.XLI.4.738.</bibtext> </blist> <blist> <bibtext> Damian, Rodica Ioana, and Brent W. Roberts. 2015. "The Associations of Birth Order with Personality and Intelligence in a Representative Sample of US High School Students." Journal of Research in Personality 58: 96–105. doi:10.1016/j.jrp.2015.05.005.</bibtext> </blist> <blist> <bibtext> De Witte, Kristof, Sofie J. Cabus, Geert Thyssen, Wim Groot, and Henriette Maassen van den Brink. 2013. "A Critical Review of the Literature on School Dropout." Educational Research Review 10 (2013): 13–28. doi:10.1016/j.edurev.2013.05.002.</bibtext> </blist> <blist> <bibtext> Deci, Edward L., Richard Koestner, and Richard M. Ryan. 2001. "Extrinsic Rewards and Intrinsic Motivation in Education: Reconsidered Once Again." Review of Educational Research 71 (1): 1–27. doi:10.3102/00346543071001001.</bibtext> </blist> <blist> <bibtext> Driessen, Geert, Frederik Smit, and Peter Sleegers. 2005. "Parental Involvement and Educational Achievement." British Educational Research Journal 31 (4): 509–532. doi:10.1080/01411920500148713.</bibtext> </blist> <blist> <bibtext> Dumont, Hanna, Ulrich Trautwein, Oliver Lüdtke, Marko Neumann, Alois Niggli, and Inge Schnyder. 2012. "Does Parental Homework Involvement Mediate the Relationship between Family Background and Educational Outcomes?" Contemporary Educational Psychology 37 (1): 55–69. doi:10.1016/j.cedpsych.2011.09.004.</bibtext> </blist> <blist> <bibtext> Epstein, Joyce L. 1995. "School/Family/Community Partnerships: Caring for the Children We Share." Phi Delta Kappan 76 (9): 701–712.</bibtext> </blist> <blist> <bibtext> Epstein, Joyce L., and Steven B. Sheldon. 2002. "Present and Accounted for: Improving Student Attendance through Family and Community Involvement." The Journal of Educational Research 95 (5): 308–318. doi:10.1080/00220670209596604.</bibtext> </blist> <blist> <bibtext> European Commission. 2010. Europe 2020. a European Strategy for Smart, Sustainable and Inclusive Growth. Brussels: European Commission.</bibtext> </blist> <blist> <bibtext> Fan, Xitao, and Michael Chen. 2001. "Parental Involvement and Students' Academic Achievement: A Meta-Analysis." Educational Psychology Review 13 (1): 1–22. doi:10.1023/A:1009048817385.</bibtext> </blist> <blist> <bibtext> Groot, Wim, and Henriëtte Maassen Van Den Brink. 2007. "The Health Effects of Education." Economics of Education Review 26 (2): 186–200. doi:10.1016/j.econedurev.2005.09.002.</bibtext> </blist> <blist> <bibtext> Hanushek, Eric A. 1992. "The Trade-off between Child Quantity and Quality." Journal of Political Economy 100 (1): 84–117.10.1086/261808</bibtext> </blist> <blist> <bibtext> Henry, Kimberly L. 2007. "Who's Skipping School: Characteristics of Truants in 8th and 10th Grade." Journal of School Health 77 (1): 29–35. doi:10.1111/j.1746-1561.2007.00159.x.</bibtext> </blist> <blist> <bibtext> Hotz, V. Joseph, and Juan Pantano. 2015. "Strategic Parenting, Birth Order, and School Performance." Journal of Population Economics 28 (4): 911–936. doi:10.1007/s00148-015-0542-3.</bibtext> </blist> <blist> <bibtext> Klassen, Robert M., and Virginia M. C. Tze. 2014. "Teachers' Self-Efficacy, Personality, and Teaching Effectiveness: A Meta-Analysis." Educational Research Review 12: 59–76. doi:10.1016/j.edurev.2014.06.001.</bibtext> </blist> <blist> <bibtext> Kluve, Jochen, and Marcus Tamm. 2013. "Parental Leave Regulations, Mothers' Labor Force Attachment and Fathers' Childcare Involvement: Evidence from a Natural Experiment." Journal of Population Economics 26 (3): 983–1005. doi:10.1007/s00148-012-0404-1.</bibtext> </blist> <blist> <bibtext> Koth, Christine W., Catherine P. Bradshaw, and Philip J. Leaf. 2008. "A Multilevel Study of Predictors of Student Perceptions of School Climate: The Effect of Classroom-Level Factors." Journal of Educational Psychology 100 (1): 96.10.1037/0022-0663.100.1.96</bibtext> </blist> <blist> <bibtext> Lee, Jungmin. 2008. "Sibling Size and Investment in Children's Education: An Asian Instrument." Journal of Population Economics 21 (4): 855–875. doi:10.1007/s00148-006-0124-5.</bibtext> </blist> <blist> <bibtext> Li, H., J. Zhang, and Y. Zhu. 2008. "The Quantity-quality Trade-off of Children in a Developing Country: Identification using Chinese Twins." Demography 45 (1): 223–243. doi:10.1353/dem.2008.0006.</bibtext> </blist> <blist> <bibtext> Linnenbrink, Elizabeth A. 2005. "The Dilemma of Performance-Approach Goals: The Use of Multiple Goal Contexts to Promote Students' Motivation and Learning." Journal of Educational Psychology 97 (2): 197–213. doi:10.1037/0022-0663.97.2.197.</bibtext> </blist> <blist> <bibtext> Lochner, Lance, and Enrico Moretti. 2004a. "The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports." American Economic Review 94 (1): 155–189. 10.1257/000282804322970751</bibtext> </blist> <blist> <bibtext> Lochner, Lance, and Enrico Moretti. 2004b. "Does Education Improve Citizenship? Evidence from the United States and the United Kingdom." Journal of Public Economics 88 (9–10): 1667–1695.</bibtext> </blist> <blist> <bibtext> MacLeod, W. Bentley, and Miguel Urquiola. 2009. Anti-Lemons: School Reputation and Educational Quality. No. w15112. Cambridge, MA: National Bureau of Economic Research. doi: 10.3386/w15112.</bibtext> </blist> <blist> <bibtext> Mammen, Kristin. 2011. "Fathers' Time Investments in Children: Do Sons Get More?" Journal of Population Economics 24 (3): 839–871. doi:10.1007/s00148-009-0272-5.</bibtext> </blist> <blist> <bibtext> McGuire, Shirley, and Lilly Shanahan. 2010. "Sibling Experiences in Diverse Family Contexts." Child Development Perspectives 4 (2): 72–79. doi:10.1111/j.1750-8606.2010.00121.x.</bibtext> </blist> <blist> <bibtext> McNeal, Ralph B. 2001. "Differential Effects of Parental Involvement on Cognitive and Behavioral Outcomes by Socioeconomic Status." The Journal of Socio-Economics 30 (2): 171–179. doi:10.1016/S1053-5357(00)00100-1.</bibtext> </blist> <blist> <bibtext> Murnane, Richard J., and John B. Willett. 2010. Methods Matter: Improving Causal Inference in Educational and Social Science Research. New York: Oxford University Press.</bibtext> </blist> <blist> <bibtext> Nota, Laura, Salvatore Soresi, and Barry J. Zimmerman. 2004. "Self-Regulation and Academic Achievement and Resilience: A Longitudinal Study." International Journal of Educational Research 41 (3): 198–215. doi:10.1016/j.ijer.2005.07.001.</bibtext> </blist> <blist> <bibtext> Organization for Economic Cooperation and Development (OECD). 1999. Preparing Youth for the 21st Century: The Transition from Education to the Labour Market. Paris: OECD.</bibtext> </blist> <blist> <bibtext> Organization for Economic Cooperation and Development (OECD). 2009. Education at a Glance: OECD Indicators. Paris: OECD.</bibtext> </blist> <blist> <bibtext> Patall, Erika A., Harris Cooper, and Jorgianne Civey Robinson. 2008. "Parent Involvement in Homework: A Research Synthesis." Review of Educational Research 78 (4): 1039–1101. doi:10.3102/0034654308325185.</bibtext> </blist> <blist> <bibtext> Phtiaka, Helen. 1996. "Each To 'his' Own?: Home/School Relations in Cyprus." Forum of Education 51 (1): 47–59.</bibtext> </blist> <blist> <bibtext> Price, Joseph. 2008. "Parent-Child Quality Time Does Birth Order Matter?" Journal of Human Resources 43 (1): 240–265. doi:10.3368/jhr.43.1.240.</bibtext> </blist> <blist> <bibtext> Price, Joseph. 2010. "The Effects of Parental Time Investments: Evidence from Natural within-Family Variation." Unpublished Manuscript. Brigham Young University.</bibtext> </blist> <blist> <bibtext> Reyes, M. R., M. A. Brackett, S. E. Rivers, M. White, and P. Salovey. 2012. "Classroom Emotional Climate, Student Engagement, and Academic Achievement." Journal of Educational Psychology 104 (3): 700–712. doi:10.1037/a0027268.</bibtext> </blist> <blist> <bibtext> Rockoff, Jonah E. 2004. "The Impact of Individual Teachers on Student Achievement: Evidence from Panel Data."The American Economic Review 94 (2): 247–252. 10.1257/0002828041302244</bibtext> </blist> <blist> <bibtext> Rothstein, Jesse. 2006. "Good Principals or Good Peers? Parental Valuation of School Characteristics, Tiebout Equilibrium, and the Incentive Effects of Competition among Jurisdictions." The American Economic Review 96 (4): 1333–1350. 10.1257/aer.96.4.1333</bibtext> </blist> <blist> <bibtext> Rubin, Donald B. 1974. "Estimating Causal Effects of Treatments in Randomized and Nonrandomized Studies." Journal of Educational Psychology 66 (5): 688–701.10.1037/h0037350</bibtext> </blist> <blist> <bibtext> Rubin, Donald B. 2006. Matched Sampling for Causal Effects. Cambridge: Cambridge University Press. 10.1017/CBO9780511810725</bibtext> </blist> <blist> <bibtext> Rumberger, Russell W. 2011. Why Students Dropout of High School and What Can Be Done. Cambridge, MA: Harvard University Press. 10.4159/harvard.9780674063167</bibtext> </blist> <blist> <bibtext> Silinskas, Gintautas, P. Niemi, M. K. Lerkkanen, and J. E. Nurmi. 2013. "Children's Poor Academic Performance Evokes Parental Homework Assistance – But Does It Help?" International Journal of Behavioral Development 37 (1): 44–56. doi:10.1177/0165025412456146.</bibtext> </blist> <blist> <bibtext> Snow, Kyle L. 2006. "Measuring School Readiness: Conceptual and Practical Considerations." Early Education and Development 17 (1): 7–41. doi:10.1207/s15566935eed1701_2.</bibtext> </blist> <blist> <bibtext> US Department of Education. 2010. Parent Power: Build the Bridge to Success. Washington, DC: US Department of Education Office of Communications and Outreach. 32.</bibtext> </blist> <blist> <bibtext> Wang, Ming-Te, and Rebecca Holcombe. 2010. "Adolescents' Perceptions of School Environment, Engagement, and Academic Achievement in Middle School." American Educational Research Journal 47 (3): 633–662. doi:10.3102/0002831209361209.</bibtext> </blist> <blist> <bibtext> Wilder, Sandra. 2014. "Effects of Parental Involvement on Academic Achievement: A Meta-Synthesis." Educational Review 66 (3): 377–397. doi:10.1080/00131911.2013.780009.</bibtext> </blist> <blist> <bibtext> Zijsling, D. H., H. Kuyper, M. J. Lubbers, M. P. C. Werf, and van der. 2005. VOCL'99-3: Technisch Rapport. [Technical Report 99-3]. Groningen: GION.</bibtext> </blist> </ref> <aug> <p>By Sofie J. Cabus and Roel J. Ariës</p> <p>Reported by Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib22" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib59" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib28" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib48" firstref="ref4"></nolink> <nolink nlid="nl5" bibid="bib11" firstref="ref5"></nolink> <nolink nlid="nl6" bibid="bib39" firstref="ref6"></nolink> <nolink nlid="nl7" bibid="bib23" firstref="ref7"></nolink> <nolink nlid="nl8" bibid="bib63" firstref="ref8"></nolink> <nolink nlid="nl9" bibid="bib34" firstref="ref9"></nolink> <nolink nlid="nl10" bibid="bib55" firstref="ref10"></nolink> <nolink nlid="nl11" bibid="bib47" firstref="ref11"></nolink> <nolink nlid="nl12" bibid="bib25" firstref="ref13"></nolink> <nolink nlid="nl13" bibid="bib36" firstref="ref14"></nolink> <nolink nlid="nl14" bibid="bib54" firstref="ref15"></nolink> <nolink nlid="nl15" bibid="bib56" firstref="ref16"></nolink> <nolink nlid="nl16" bibid="bib42" firstref="ref17"></nolink> <nolink nlid="nl17" bibid="bib32" firstref="ref18"></nolink> <nolink nlid="nl18" bibid="bib40" firstref="ref19"></nolink> <nolink nlid="nl19" bibid="bib41" firstref="ref20"></nolink> <nolink nlid="nl20" bibid="bib30" firstref="ref23"></nolink> <nolink nlid="nl21" bibid="bib62" firstref="ref25"></nolink> <nolink nlid="nl22" bibid="bib24" firstref="ref27"></nolink> <nolink nlid="nl23" bibid="bib64" firstref="ref28"></nolink> <nolink nlid="nl24" bibid="bib10" firstref="ref32"></nolink> <nolink nlid="nl25" bibid="bib29" firstref="ref33"></nolink> <nolink nlid="nl26" bibid="bib33" firstref="ref34"></nolink> <nolink nlid="nl27" bibid="bib50" firstref="ref35"></nolink> <nolink nlid="nl28" bibid="bib52" firstref="ref36"></nolink> <nolink nlid="nl29" bibid="bib53" firstref="ref37"></nolink> <nolink nlid="nl30" bibid="bib21" firstref="ref38"></nolink> <nolink nlid="nl31" bibid="bib65" firstref="ref40"></nolink> <nolink nlid="nl32" bibid="bib61" firstref="ref41"></nolink> <nolink nlid="nl33" bibid="bib14" firstref="ref42"></nolink> <nolink nlid="nl34" bibid="bib12" firstref="ref43"></nolink> <nolink nlid="nl35" bibid="bib13" firstref="ref44"></nolink> <nolink nlid="nl36" bibid="bib31" firstref="ref47"></nolink> <nolink nlid="nl37" bibid="bib19" firstref="ref48"></nolink> <nolink nlid="nl38" bibid="bib20" firstref="ref52"></nolink> <nolink nlid="nl39" bibid="bib38" firstref="ref53"></nolink> <nolink nlid="nl40" bibid="bib17" firstref="ref57"></nolink> <nolink nlid="nl41" bibid="bib37" firstref="ref61"></nolink> <nolink nlid="nl42" bibid="bib26" firstref="ref76"></nolink> <nolink nlid="nl43" bibid="bib27" firstref="ref77"></nolink> <nolink nlid="nl44" bibid="bib45" firstref="ref78"></nolink> <nolink nlid="nl45" bibid="bib51" firstref="ref79"></nolink> <nolink nlid="nl46" bibid="bib60" firstref="ref89"></nolink> <nolink nlid="nl47" bibid="bib15" firstref="ref101"></nolink> <nolink nlid="nl48" bibid="bib18" firstref="ref102"></nolink> <nolink nlid="nl49" bibid="bib46" firstref="ref108"></nolink> <nolink nlid="nl50" bibid="bib57" firstref="ref115"></nolink> <nolink nlid="nl51" bibid="bib58" firstref="ref116"></nolink> <nolink nlid="nl52" bibid="bib16" firstref="ref124"></nolink> <nolink nlid="nl53" bibid="bib43" firstref="ref130"></nolink> <nolink nlid="nl54" bibid="bib44" firstref="ref131"></nolink> <nolink nlid="nl55" bibid="bib35" firstref="ref134"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1131023
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: What Do Parents Teach Their Children?--The Effects of Parental Involvement on Student Performance in Dutch Compulsory Education
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Cabus%2C+Sofie+J%2E%22">Cabus, Sofie J.</searchLink><br /><searchLink fieldCode="AR" term="%22Ariës%2C+Roel+J%2E%22">Ariës, Roel J.</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Educational+Review%22"><i>Educational Review</i></searchLink>. 2017 69(3):285-302.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Routledge. Available from: Taylor & Francis, Ltd. 530 Walnut Street Suite 850, Philadelphia, PA 19106. Tel: 800-354-1420; Tel: 215-625-8900; Fax: 215-207-0050; Web site: http://www.tandf.co.uk/journals
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 18
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2017
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Secondary+Education%22">Secondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Parent+Participation%22">Parent Participation</searchLink><br /><searchLink fieldCode="DE" term="%22Academic+Achievement%22">Academic Achievement</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Birth+Order%22">Birth Order</searchLink><br /><searchLink fieldCode="DE" term="%22Educational+Attainment%22">Educational Attainment</searchLink><br /><searchLink fieldCode="DE" term="%22Correlation%22">Correlation</searchLink><br /><searchLink fieldCode="DE" term="%22Family+School+Relationship%22">Family School Relationship</searchLink><br /><searchLink fieldCode="DE" term="%22Family+Environment%22">Family Environment</searchLink><br /><searchLink fieldCode="DE" term="%22Socioeconomic+Status%22">Socioeconomic Status</searchLink><br /><searchLink fieldCode="DE" term="%22Grade+Repetition%22">Grade Repetition</searchLink><br /><searchLink fieldCode="DE" term="%22Immigrants%22">Immigrants</searchLink><br /><searchLink fieldCode="DE" term="%22Mathematics+Achievement%22">Mathematics Achievement</searchLink><br /><searchLink fieldCode="DE" term="%22Outcomes+of+Education%22">Outcomes of Education</searchLink><br /><searchLink fieldCode="DE" term="%22Homework%22">Homework</searchLink><br /><searchLink fieldCode="DE" term="%22Secondary+School+Students%22">Secondary School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Least+Squares+Statistics%22">Least Squares Statistics</searchLink><br /><searchLink fieldCode="DE" term="%22Parent+Child+Relationship%22">Parent Child Relationship</searchLink><br /><searchLink fieldCode="DE" term="%22Regression+%28Statistics%29%22">Regression (Statistics)</searchLink><br /><searchLink fieldCode="DE" term="%22Robustness+%28Statistics%29%22">Robustness (Statistics)</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Netherlands%22">Netherlands</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1080/00131911.2016.1208148
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0013-1911
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Theory and evidence indicate that, if family size grows, the younger children will get less parental involvement than the older children. These differences in parental involvement through birth order may impact academic achievement if, and only if, parental involvement is an important determinant of children's educational attainment. The oldest child then benefits the most in terms of educational outcomes. Estimates for the Netherlands show a robust negative relationship between birth order and parental involvement, and significant positive medium to large effects of parental involvement through birth order on various measures of academic achievement. Furthermore, our findings indicate that academic achievement is rooted in a school-supportive home climate, and often created by the mother. However, when it comes to math performance and grade retention, it is better that both parents unduly interfere with school. We also find that parents with low socio-economic status and from immigrant families are as much involved in the education of their children as the average Dutch family, but their involvement is less effective in terms of children's learning outcomes.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Ref
  Label: Number of References
  Group: RefInfo
  Data: 65
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2017
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1131023
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1131023
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/00131911.2016.1208148
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 18
        StartPage: 285
    Subjects:
      – SubjectFull: Parent Participation
        Type: general
      – SubjectFull: Academic Achievement
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Birth Order
        Type: general
      – SubjectFull: Educational Attainment
        Type: general
      – SubjectFull: Correlation
        Type: general
      – SubjectFull: Family School Relationship
        Type: general
      – SubjectFull: Family Environment
        Type: general
      – SubjectFull: Socioeconomic Status
        Type: general
      – SubjectFull: Grade Repetition
        Type: general
      – SubjectFull: Immigrants
        Type: general
      – SubjectFull: Mathematics Achievement
        Type: general
      – SubjectFull: Outcomes of Education
        Type: general
      – SubjectFull: Homework
        Type: general
      – SubjectFull: Secondary School Students
        Type: general
      – SubjectFull: Least Squares Statistics
        Type: general
      – SubjectFull: Parent Child Relationship
        Type: general
      – SubjectFull: Regression (Statistics)
        Type: general
      – SubjectFull: Robustness (Statistics)
        Type: general
      – SubjectFull: Netherlands
        Type: general
    Titles:
      – TitleFull: What Do Parents Teach Their Children?--The Effects of Parental Involvement on Student Performance in Dutch Compulsory Education
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Cabus, Sofie J.
      – PersonEntity:
          Name:
            NameFull: Ariës, Roel J.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2017
          Identifiers:
            – Type: issn-print
              Value: 0013-1911
          Numbering:
            – Type: volume
              Value: 69
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Educational Review
              Type: main
ResultId 1