454 Kocak,O. Goksu, I, Goktas, Y. (2021). The factors affecting academic achievement: a systematic review of meta analyses. International Online Journal of Education and Teaching (IOJET), 8(1). 454-484. Received : 27.07.2020 Revised version received : 27.10.2020 Accepted : 29.10.2020 THE FACTORS AFFECTING ACADEMIC ACHIEVEMENT: A SYSTEMATIC REVIEW OF META ANALYSES Research Article Ömer Koçak Atatürk University kocakomer@atauni.edu.tr İdris Göksu Mardin Artuklu University idrisgoksu@artuklu.edu.tr Yüksel Göktaş Atatürk University yukselgoktas@atauni.edu.tr Ömer Koçak is an Assistant Professor at the Department of Information and Document Management of Ataturk University. His research interests are educational technologies, interactive whiteboards, and cartoons in early childhood education. İdris Göksu is an Assistant Professor at the Department of Educational Sciences of Mardin Artuklu University. His research interests include educational technologies, instructional design, digital media, mobile learning and distance learning. Yüksel Göktaş is a full Professor at the Department of Computer Education and Instructional Technologies Department of Ataturk University. His research interests are 3D virtual worlds, augmented reality, and research in instructional technologies. Copyright by Informascope. Material published and so copyrighted may not be published elsewhere without the written permission of IOJET. mailto:kocakomer@atauni.edu.tr mailto:idrisgoksu@artuklu.edu.tr mailto:yukselgoktas@atauni.edu.tr http://orcid.org/0000-0001-7756-7169 http://orcid.org/0000-0002-7120-6562 http://orcid.org/0000-0002-7341-2466 International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 455 THE FACTORS AFFECTING ACADEMIC ACHIEVEMENT: A SYSTEMATIC REVIEW OF META ANALYSES Ömer Koçak kocakomer@atauni.edu.tr İdris Göksu idrisgoksu@artuklu.edu.tr Yüksel Göktaş yukselgoktas@atauni.edu.tr Abstract The study aimed to identify the factors and to demonstrate their effects on academic achievement in various publications that utilized meta-analyses. For this purpose, the meta- analyses publications on the Web of Science-All Database till 2018 were reviewed. In the study, the systematic review method was adopted. Following a related review, 169 meta- analyses were included in the scope of the study. The effects of 254 variables on academic achievement were investigated, and consequently, 427 effect sizes were found in total. Variables obtained from meta-analyses with the effect sizes between -.799 and 3.170 were examined in nine categories. The results revealed that the number of variables evaluated in the categories of psychological, socio-economic, socio-demographic and individual characteristics, learning theories and teaching strategies, and family was bigger than other categories. Keywords: Academic achievement, academic success, meta-analysis, meta-review, systematic review 1. Introduction Education is a phenomenon for which nations have developed policies for centuries and which they have laid special emphasis to leave the next generations a more sustainable world. According to the results of PISA 2015, TIMMS 2015 and PIRLS 2016- in which more than 60 countries participated- Estonia, Finland, Japan, Singapore, Russian Federation, and Chinese Taipei are among the most successful countries (OECD, 2018; TIMMS & PIRLS, n.d.). The data concerning some of those countries for the year 2014 demonstrated that Japan allocated 3.6% of gross domestic product (GDP) in the budget to education while Estonia allocated 5.5% and Finland allocated more than 7% to education. In addition to that such country as the US, the UK, France, and Germany- which are in the category of developed nations- spend approximately 5% of GDP on education (UNESCO, 2019). Those countries which spend a considerable amount of their GDP aim to raise capable and successful students who are to build their future (Steenbergen-Hu, Makel, & Olszevski-Kubilius, 2016; UNESCO, 2019). All nations aim to raise such capable students by means of state schools or private schools. One of the factors coming into mind primarily is academic achievement when such concepts as school, student, education and teaching are discussed (Voltmer & von Salisch, 2017). Candidates are firstly compared in terms of academic achievement in transition into an upper-level education or in employment (von Stumm, Hell, & Chamorro-Premuzic, 2011). Thus, academic achievement is important and determining the future education of individuals and job mailto:kocakomer@atauni.edu.tr mailto:idrisgoksu@artuklu.edu.tr mailto:yukselgoktas@atauni.edu.tr Kocak. Goksu,Goktas 456 opportunities for them (Flashman, 2012). Educators and researchers try several ways and new methods to investigate academic achievement- which is considered very important for the future of generations in all age groups (Marques, Gallagher, & Lopez, 2017). Academic achievement and variables influencing academic achievement in positive and negative ways are analyzed by researchers so that the investments made can attain the goals (Credé & Kuncel, 2008); this is because one of the goals of teaching is to raise academically well-equipped individuals. Yet, it is not so easy as it is though because there are several variables influential in and correlated with academic achievement. There are many factors with positive or negative effects on or correlated with students’ academic achievement such as learning-teaching methods (Donker, de Boer, Kontos, Dinath van Ewjk, & van der Werf, 2014), the circumstances students are in (White, 1982), factors stemming from students (Perera & DiGiacomo, 2013), factors stemming from school and teachers (Rivkin, Hanushek, & Kain, 2005), physical activities (Alvarez-Bueno et al., 2017), parents attitudes towards students (Jeynes, 2017; Pinquart, 2016) and students’ medical status (Galland et al., 2015). Especially effective factors are thought to be influential in and determiner of academic achievement (Credé, Tynan, & Harms, 2017). Research into learning and teaching processes is necessary to find whether or not the teaching method used or the variable considered is influential in learning and whether or not students gain the targeted academic capabilities (Vo, Zhu, & Diep, 2017). Such research demonstrates the effects of a number of variables on academic achievement (Gajda, Karwowski, & Beghetto, 2017). It is also thought that review studies (e.g. meta-analyses) about the variables affecting academic achievement which will guide practitioners are important. Studies in which review studies are combined are also important. Such studies are known as a second-order meta-analysis, overviews of reviews, systematic reviews of reviews or as a meta- analysis of meta-analyses (Polanin, Maynard, & Dell, 2017). Also, it can be defined as a quantitatively summarizing meta-analyses addressing a similar topic or research question (Tamim, Bernard, Borokhovski, Abrami, & Schmid, 2011). A review of the literature demonstrated that the number of studies analyzing the meta-analyses focusing on academic achievement at all stages of education was limited (Hattie, 2009, 2015; Schneider & Preckel, 2017; Sipe & Curlette, 1996; Steenbergen-Hu et al., 2016; Tamim et al., 2011). Besides, the fact that the studies except for Hattie (2009) generally focus on one category or on one stage of education makes this current study different from others. Hattie (2015), in a study analyzing 1200 meta-analyses, found 195 effect sizes (ESs) influencing academic achievement. Thus, the effect size changed between -.42 and 1.65. Hattie (2015) divided the variables influencing academic achievement into such categories as students, administration, school, peers, home, and teachers. Another study analyzing 103 meta-analyses and summarizing the variables correlated with academic achievement found effect sizes ranging between -.03 and 1.15 (Sipe & Curlette, 1996). While some of the studies investigated the variables influential in academic achievement only for one stage of education, some of the studies investigated the variables influential only in a certain category. Schneider and Preckel (2017), for instance, investigating the variables related to academic achievement in higher education, had access to 38 meta- analyses and found that 105 ESs ranged between -.52 and 1.91 and they also found that there were high correlations between social interaction in courses and academic achievement. Steenbergen-Hu et al. (2016), in another analysis of meta-analyses, compiled meta-analyses that included stages of education from primary education to high school education in the analyses and divided students into groups according to their capabilities, academically accelerated them and thus analyzed the effects on academic achievement. As a result of the compilation, it was found that grouping students with special education, forming groups in the classroom and forming groups of students who are at different levels would affect academic achievement in positive ways. It was claimed, however, that forming groups between classes International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 457 at the same level did not have any effects on academic achievement. Another variable, academic acceleration, was found to have medium level of positive effects. Tamim et al. (2011), compiling the meta-analyses analyzing the effects of educational technologies on learning, found that the effect size was .33 on average. Richardson, Abraham, and Bond (2012) performed meta-analyses on the psychological factors influencing the academic achievement of university students and they found that the size of correlations was between -0.24 and .59. They also found that the correlation between self-efficacy and academic achievement was high. It became important for meta-analyses and for review studies to bring several studies together and thus to make evaluations since reliability problems influencing the result in experimental and relational studies concerning the increase in academic achievement. This review study was believed to contribute to relevant literature in that it would lead researchers to conduct researches on academic achievement, and would increase awareness of educators, parents, students, and administrators in terms of factors related to academic achievement. In this context, the study aimed to compile the variables in meta-analyses examining the effect on or their correlations with academic achievement, as well as to draw a general framework on variables connected with academic achievement by classifying them into categories. 2. Method The study adopted review method, and analyzed some meta-analyses in terms of their effects on and correlations with academic achievement. Academic achievement was measured with school degree and standardized tests, and studies that were correlated with academic achievement or whose effects on academic achievement were clearly demonstrated in meta- analyses were included in this paper. The flow described by Moher, Liberati, Tetzlaff, Altman and PRISMA Group (2009) for choosing studies was used at the stage of reviewing the studies. It is thought that Web of Science (WOS) database was preferred for reviewing the studies. WOS, the oldest database, has a strong scope of reference and contains good quality studies (Aghaei Chadegani et al., 2013). In addition to that, it was observed in some review studies that only WOS database was used at the stage of choosing the studies to be considered (Chamberlain et al., 2012; Chen, Yang, Yang, Jiang, & Zhou, 2014; Zhang et al., 2016). 2.1. Inclusion and Exclusion Criteria In the study, a comprehensive review of related literature was conducted on the “Web of Science-All Databases” by using the phrases (“academic achievement" or "academic success", or "academic outcomes" or "academic performance") and (meta analysis OR meta-analysis) in the title, abstract and keywords parts of the studies to reach the studies. The studies published in English before 2018 were filtered and, as a consequence, 538 studies in total were reached out. They were downloaded by the researchers, their abstracts were examined, and 246 of them were excluded from analyses. After analyzing the full-text articles, 123 of them were also excluded from the content and 169 articles in total were included in analyses. The process of selecting the studies is shown in Figure 1. Besides, meta-analyses conducted within the scope of the same issue or the same research question can also contain the same primary studies. In that case, the overlap is taken into consideration to attain the validity and reliability of review studies (Tamim et al., 2011). It was found that the rate of overlap in meta-analyses considering the same variable was below 25% - in a method similar to the one used by Wilson and Lipsey (2001), and therefore any of the meta-analyses were not excluded. Kocak. Goksu,Goktas 458 2.2. Data Analysis This study followed the stages of PRISMA guide in the whole process from the stage of searching the electronic database to analyzing the data and reporting them. Additionally, Google Sheets application was used in analyzing the data and thus Figure 2 was prepared. We used this application because it offers researchers the opportunity to work simultaneously. Random effects model was taken into consideration in studies that presented all effect sizes unless otherwise stated. Weighted ES was regarded as the base in studies that presented the correlation values unless the opposite was stated. Following the searching made in the criteria described below, 538 studies were reached. All databases were searched on the WoS Being published before the date 01.01.2018 ✓ Key words determined (("academic achievement" OR "academic success" OR "academic outcomes" OR "academic performance") AND (meta analysis OR meta- analysis)) were searched in the title, abstract and key words parts ✓ Searching was made in English. • After examining the abstracts, 246 articles with the following properties were excluded from analysis; ✓ Studies which were not described as Meta-analysis ✓ Studies which did not analyse the correlations of variables with academic achievement and effects of variables on academic achievement The abstracts of the 538 studies were examined 169 studies which were found to be appropriate to the scope of the study were included in analyses (See Appendix C). In c lu d e d E li g ib il it y Id e n ti fi c a ti o n • After examining the full text articles, 123 articles with the following properties were excluded from analysis ✓ Studies which were not Meta-analysis ✓ Studies which did not analyse the correlations of variables with academic achievement and effects of variables on academic achievement ✓ Studies whose effect size could not be determined were excluded from analysis The full texts of the 292 studies were examined S c r e e n in g Figure 1. Inclusion and exclusion process International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 459 Figure 2. Codes created in Google Sheets Meta-analyses analyzing the same variable in the same type of ES were brought together to be able to present more meaningful findings. In accordance with this principle, the overall ES of the studies which were independent of each other but which investigated the same type of ES was calculated according to mean. The overall ES was not presented in some of the studies. Because more than one ESs were divided into such categories as content (science, verbal, and math) and level (pre-school, elementary school) and thus analyzed, the mean of the values of the variable was again calculated. Pearson correlation (r) was transformed into Cohen’s d from effect size in studies analyzing the correlations between variables. The formula for changing correlations into Cohen’s d proposed by Borenstein, Hedges, Higgins, and Rothstein (2009) was used in changing the correlations into ES (See Formula 2). For variables with more than one correlation values, first Cohen’s d was found for each size and then ES mean was calculated. In some studies, Fisher’s Z correlation coefficient was given as the correlation coefficient. They were first transformed into Pearson’s correlation coefficient (r) through Fisher Z-Transformation Table and then into Cohen’s d with Formula 2. Eta square (η2) was transformed into Cohen’s d with Formula 3 (Cohen, 1988). The effects sizes turned into Cohen’s d were shown in tables in the Findings part of the study. 𝑑 = 2𝑟 √1 − 𝑟2 Formula 2. Transforming correlation (r) into Cohen’s d 𝑓 = √ η2 (1 − η2) 𝑑 = 2 ∗ 𝑓 Formula 3. Transforming Eta-Square (η2) into Cohen’s d Because the ES of Hedge’s g was not transformed into the ES of Cohen’s d in the meta- analyses, the ES of Hedge’s g was presented as it was. Besides, the ES of Glass’ Delta (Glass’s Δ), which could not be transformed into another type of ES, was given as it was. In addition to that, in some studies, the type of ES was not described, and therefore they were labelled as “ES”. 2.3. Dividing the Variables into Categories The variables which were found to be correlated with or to have effects on academic achievement in meta-analyses published before 2018 were divided into nine categories such as Psychological Characteristics”, “Learning Theories and Teaching Strategies”, “SES&SDC and Individual Characteristics”, “Family”, “Teacher”, “Special Education”, “School”, “Educational Technology” and “Violence”; and each category was presented under sub- headings. The categories were created by considering similar review studies in the literature (Engin-Demir, 2009; Hattie, 2015; Richardson et al., 2012; Sarier, 2016; Schneider & Preckel, 2017; Zaff et al., 2017). All the variables found in this study were categorized by five experts separately and the variables were put into the most suitable category. The full texts of the relevant studies were re-examined for the variables fitting in more than one category, and thus they were put into the most suitable category with consensus. Intraclass Correlation Coefficient (ICC) analysis was performed to ensure inter-rater reliability at the stage of creating categories. The average measures ICC was .86 and according to Koo and Li (2016), it indicates good Kocak. Goksu,Goktas 460 reliability. The definitions of the variables in the categories were prepared to benefit from the relevant studies so that their purpose could be reflected correctly and so that this study could be interpreted correctly by readers, and they are presented in Appendix B. 2.4. Interpreting the Effect Size Kelley and Preacher (2012) point out that ES is statistical data and that it represents the quantitative reflection of the size of phenomena containing a problem. The ESs in the findings obtained in this study were interpreted in Table 1 which was created in accordance with Sawilowsky (2009) and Cohen (1988). Besides, some of the meta-analyses examined the correlations between academic achievement and a number of variables. Thus, readers can evaluate the ESs obtained for those variables as relational values rather than as experimental effects. Table 1. Comment rules of the ES Definition ES Range References Very small ES ≥ .01 and < .2 Sawilowsky, 2009 Small ES ≥ .2 and < .5 Cohen, 1988; Sawilowsky, 2009 Medium ES ≥ .5 and < .8 Cohen, 1988; Sawilowsky, 2009 Large ES ≥ .8 and < 1.2 Cohen, 1988; Sawilowsky, 2009 Very Large ES ≥ 1.2 and < 2.0 Sawilowsky, 2009 Huge ES ≥ 2.0 Sawilowsky, 2009 3. Results This study, examining 169 meta-analyses, found that there were 254 different variables influencing academic achievement and that 427 different ESs were identified in relation to those variables. The findings obtained were described under ten headings. The variables with effects sizes ranging between huge and small were shown in the tables in the findings section, and the variables with effect size of very small were given in Appendix A. The ESs found for some of the variables were shown in tables separately because they were of different types and because they could not be transformed into Cohen’s d. Readers need to know that the relevant variables and ESs are the results of meta-analyses having different qualities and limitations and they need to consider it in interpreting the results given in the tables below and in making comparisons. The fact that no limitations were put on the levels of education in including the relevant meta-analyses in this study is also supportive of this necessity. Different colors were also used in tables to show the moves from the positive to the negative in the tables so that the ES levels could be understood better. 3.1. Psychological Characteristics A number of psychological variables influencing academic achievement and having effect sizes between small and huge are shown in Table 2. Accordingly, 87 ES values were found for 64 variables in 25 meta-analyses. On examining Table 2, it is evident that self-efficacy (ES= 1.173) and academic emotions positive low-arousal (ES= .812) have positive and huge effects on academic achievement. The variables with medium effects on academic achievement are self-assigned minimal goal standards (ES= .747), academic self-efficacy (ES= .735), recognition of emotions in faces (ES= .676), effort regulation (ES= .676), academic emotions comprising positive high arousal (PHA) (ES= .657), attitude towards the course (ES= .638), motivation (ES= .558), study attitude (ES= .539) and academic motivation (ES= .532), respectively. It is remarkable that academic extrinsic motivation (ES= .020) has very small effects on academic achievement (see Appendix A). International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 461 Table 2. Variables related to psychological characteristics Variables References SI PS IES OES ES 1 Self-efficacy Sarier (2016) 2000-2015 62 8 1 1.173 2 Academic emotions PLA Lei & Cui (2016) 2005-2016 35 39 1 .812 3 Self-assigned minimal goal standards Richardson et al. (2012) 1997-2010 217 13 1 .747 4 Academic self-efficacy Richardson et al. (2012) Robbins et al. (2004) 1997-2010 After 1984 217 109 67 18 2 .735† (.652, .817) 5 Emotion knowledge Voltmer & von Salisch (2017) NA 49 84 1 .676 6 Effort regulation Richardson et al. (2012) 1997-2010 217 19 1 .676 7 Academic emotions comprising PHA Lei & Cui (2016) 2005-2016 35 39 1 .657 8 Attitude towards the course Sarier (2016) 2000-2015 62 26 1 .638 9 Motivation Fong et al. (2017a); Sarier (2016) NA 2000-2015 95 62 106 9 2 .558† (.345, .771) 10 Study habits and attitudes-study attitude Credé & Kuncel (2008) 1980-2005 344 37 1 .539 11 Academic motivation Robbins et al. (2004) After 1984 109 17 1 .532 12 Critical thinking Fong et al. (2017b) Richardson et al. (2012) Ross et al. (2013) 1976-2014 1997-2010 1980-2011 23 217 41 27 9 41 3 .498† (min: .303, max: .652) 13 Study habits and attitudes-study motivation Credé & Kuncel (2008) 1980-2005 344 25 1 .473 14 Hope Marques et al. (2017) NA 29 24 1 .387 15 Need for cognition Richardson et al. (2012) 1997-2010 217 5 1 .387 16 Self-regulation Fong et al. (2017a) NA 95 57 1 .366 17 Elaboration Richardson et al. (2012) 1997-2010 217 12 1 .366 18 Meta cognition Richardson et al. (2012) 1997-2010 217 9 1 .366 19 Psychotherapy Baskin et al. (2010) 1980-2008 83 27 1* .360 20 Identified regulation Taylor et al. (2014) 1956-2013 18 11 1* .350 21 Concentration Richardson et al. (2012) 1997-2010 217 12 1 .324 22 Academic goals Robbins et al. (2004) After 1984 109 34 1 .314 23 Intrinsic motivation Taylor et al. (2014) Richardson et al. (2012) 1956-2013 1997-2010 18 217 10 22 2* .308† (.270, .345) 24 Help seeking Richardson et al. (2012) 1997-2010 217 8 1 .303 25 Goal commitment Richardson et al. (2012) 1997-2010 217 10 1 .303 26 Self-esteem Sarier (2016) Richardson et al. (2012) 2000-2015 1997-2010 62 217 6 21 2 .294† (.181, .408) 27 Grit Credé et al. (2016) NA 73 76 2 .293† (.283, .303) 28 Self-perception Fong et al. (2017a) NA 95 108 1 .262 29 Locus of control Richardson et al. (2012) 1997-2010 217 13 1 .262 30 Active coping Clarke (2006) 1980-2001 40 6 1 .242 31 Mastery approach goals Wirthwein et al. (2013) Huang (2012) 1980-2011 Until 2008 180 151 209 19 2 .232† (.201, .262) 32 Self-concept Ma & Kishor (1997) 1966-1993 143 89 1*** .230 33 Optimism Richardson et al. (2012) 1997-2010 217 6 1 .221 34 Instutional commitment Robbins et al. (2004) After 1984 109 11 1 .217 --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 52 Depression Richardson et al. (2012) 1997-2010 217 17 1 -.201 53 Anxiety Fong et al. (2017a) Richardson et al. (2012) NA 1997-2010 95 217 50 29 2 -.217† (-.494, .060) 54 External regulation Taylor et al. (2014) 1956-2013 18 11 1* -.220 55 Work-avoidance Wirthwein et al. (2013) 1980-2011 180 25 1 -.221 56 Academic Stress Richardson et al. (2012) 1997-2010 217 4 1 -.242 57 Performance avoidance goals Wirthwein et al. (2013) Huang (2012) Richardson et al. (2012) 1980-2011 Until 2008 1997-2010 180 151 217 109 19 31 4 -.257† (min: -.283, max: -.242) 58 Stress Richardson et al. (2012) 1997-2010 217 8 1 -.262 59 Subsequent depression Huang (2015) Until 2012 43 50 1 -.303 60 Boredom Tze et al. (2016) 1990-2013 29 21 1 -.324 61 Study habits and attitudes-study anxiety Credé & Kuncel (2008) 1980-2005 344 22 1 -.345 62 Academic emotions NHA Lei & Cui (2016) 2005-2016 35 39 1 -.364 63 Self Handcaping Schwinger et al. (2014) Until 2013 36 49 1 -.473 64 Amotivation Taylor et al. (2014) 1956-2013 18 7 1* -.610 65 Academic emotions NLA Lei & Cui (2016) 2005-2016 35 39 1 -.799 *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d, †: Mean of more than one effect size of the same type, NLA: Negative low-arousal, NHA: Negative high-arousal, PHA: Positive high-arousal, PLA: Positive low-arousal, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 462 An examination of variables influencing academic achievement in negative ways demonstrated that NLA (ES= -.799) and amotivation (ES= -.610) had medium effects. According to Table 2, most of the variables having negative influences represent negativity whereas performance-avoidance goals (ES= -.257) and external regulation (ES= -.220) are positive variables- which is remarkable. 3.2. Learning Theories and Teaching Strategies 48 variables were identified in 43 meta-analyses in total which were examined. 82 ESs were calculated between the variables and academic achievement. An examination of Table 3 made it clear that 18 variables had positive and very large effects on academic achievement, but 6 variables had positive and very small effects on academic achievement (See Appendix A). According to Table 3, creative drama (ES= 1.453), constructivist learning (ES= 1.391), learning strategy instruction (ES= 1.250) and collaborative learning (ES= 1.230) have very large effects on academic achievement. Those variables are followed by 4MAT model (ES= 1.168), conceptual change text (ES= 1.160), multiple intelligence (ES= 1.077), Kolb learning styles model (ES= 1.067), mind mapping techniques (ES= 1.057), learning style (ES= 1.029), constructivist learning (ES= 1.003), Dunn and Dunn learning style model (ES= 1.001), project- based learning (ES= .997), peer learning (ES= .900), graphic organizers (ES= .897), cognitive learning strategies (ES= .876), perceptual learning styles (ES= .870) and portfolio (ES= .831). International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 463 Table 3. Variables related to learning theories and teaching strategies Variables References SI PS IES OES ES 1 Creative Drama Ulubey & Toraman (2015) Batdi & Batdi (2015) 1997-2015 2000-2014 65 40 65 40 2*** 1.453† (1.225, 1.680) 2 Constructivist learning Erisen & Gunay (2015) 2001-2013 27 27 1** 1.391 3 Learning strategy instruction Donker et al. (2014) 2000-2012 58 95 1** 1.250 4 Collaborative learning Kalaian & Kasim (2017) 1987-2014 19 1 1** 1.230 5 4MAT model Kanadli (2016) 2004-2014 30 10 1* 1.168 6 Conceptual change text Armagan et al. (2010) 1995-2010 42 42 1* 1.160 7 Multiple intelligence Bas (2016) 1998-2014 75 75 1** 1.077 8 Kolb learning styles model Kanadli (2016) 2004-2014 30 2 1* 1.067 9 Mind mapping techniques Batdi (2015b) 2005-2013 15 10 1** 1.057 10 Learning style Kanadli (2016) 2004-2014 30 29 1* 1.029 11 Constructivist learning Ural & Bümen (2016) 2002-2012 77 27 1* 1.003 12 Dunn and Dunn learning style model Kanadli (2016) Lovelace (2005) 2004-2014 1980-2000 30 68 3 168 2* 1.001† (.670, 1.331) 13 Project based learning Ayaz & Söylemez (2015) 2002-2013 41 42 1* .997 14 Peer learning Kalaian & Kasim (2017) 1987-2014 19 1 1** .900 15 Graphic organizers Kansızoğlu (2017) 2000-2016 70 70 1** .897 16 Cognitive learning strategies Kim et al. (2008) 1990-2006 50 21 5* .876† (min: .510, max: 1.550) 17 Perceptual learning styles Kanadli (2016) 2004-2014 30 13 1* .870 18 Portfolio Başol & Erbay (2017 1990-2017 24 46 1* .831 19 Student respond cards Randolph (2007) Until 2005 18 NA 2* .730† (.380, 1.080) 20 Study habits and attitudes-aggregate measures Credé & Kuncel (2008) 1980-2005 344 107 1 .699 21 Strategy instruction Ardasheva et al. (2017) 2008-2014 37 90 8* .660† (min: .130, max: 1.230) 22 Educational interventions de Boer et al. (2014) 2000-2011 58 93 5** .646† (min: .360, max: 1.250) 23 Brain based learning Gozuyesil & Dikici (2014) 1999-2011 31 42 1** .640 24 Various innovative learning methods Kalaian & Kasim (2017) 1987-2014 19 21 1** .590 25 Cooperative learning Capar & Tarim (2015) 1988-2010 26 36 1** .590 26 Problem-based learning Kalaian & Kasim (2017) Dagyar & Demirel (2015) 1987-2014 19 16 2** .560† (.290, .830) 27 Self-regulated learning Dent & Koenka (2015) Dignath et al. (2008) Dignath & Büttner (2008) 2000-2010 1992-2006 1992-2006 61 48 74 81 102 136 4 .545† (min: .408, max: .620) 28 Study skills Credé & Kuncel (2008) 1980-2005 344 87 1 .516 29 Small-group learning Springer et al. (1999) After 1980 37 116 1* .510 30 Strategic approach to learning Richardson et al. (2012) 1997-2010 217 15 1 .473 31 Universal SEL programmes Sklad et al. (2012) 1995-2008 75 10 1* .460 32 Services learning Conway et al. (2009) Until 2008 78 19 1* .430 33 Blended learning Vo et al. (2017) 2001-2017 40 51 1** .385 34 Peer assisted learning Ginsburg-Block et al. (2006) NA 36 26 1** .350 35 Homework Fan et al. (2017) Cooper et al. (2006) Bas et al. (2017) 1986-2015 1987-2003 NA 28 32 11 61 69 11 3 .335† (min: .229, max: .453) 36 Phonics reading instruction Jeynes (2008) 1966-2000 22 NA 1** .300 37 POGIL Walker & Warfa (2017) NA 21 21 1** .290 38 Deep approach to learning Richardson et al. (2012) 1997-2010 217 23 1 .283 39 Universal SEL programmes Wigelsworth et al. (2016) NA 89 15 1** .280 40 Universal SEL programmes Durlak et al. (2011) 1970-2017 213 35 1*** .270 41 Peer tutoring Leung (2015) Richardson et al. (2012) NA 1997-2010 72 217 72 4 2* .261† (.260, .262) 42 Experimental interventions Braithwaite & Corr (2016) NA 47 14 1** .259 43 Study habits and attitudes-deep processing Credé & Kuncel (2008) 1980-2005 344 28 1 .242 44 Learning goal orientation Richardson et al. (2012) 1997-2010 217 60 1 .201 --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 52 Surface approach to learning Richardson et al. (2012) 1997-2010 217 22 1 -.366 *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type SEL: Social and emotional learning, POGIL: Process-oriented guided inquiry learning, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 464 On examining the variables with negative effects, it was found that the surface approach to learning (ES= -.366) had small effects on academic achievement. Although negative effect size was also found in relation to cross-age tutoring, the mean effect size (ES= .107) was very small but it was positive- as can be seen in Appendix A. besides, it was also remarkable that cooperative learning (ES= -.140) had negative effects- despite very small effects- on achievement (See Appendix A). It became apparent that no learning theories or teaching strategies apart from the surface approach to learning and cooperative learning had negative effects on academic achievement. 3.3. Family 34 variables and 79 ES values affecting academic achievement were found from 20 meta- analyses in relation to the family. Family-related variables and the effect sizes for them are shown in Table 4. Accordingly, parents’ attitudes and behaviors (ES= .873), parental expectations (ESCohen’s d= .865) and academic socialization parental involvement (ES= .847) had large effects on academic achievement. However, variables such as parental expectations (ESHedges’ g= .730), parental involvement in homework (ES= .720), maternal employment (ES= .699), parental attendance and participation (ES= .668), parental involvement types (other) (ES= .622), parental involvement (ES= .538), communication parental involvement (ES= .530) and education level of father (ES= .519) have medium effects. It was observed that the education level of fathers affected academic achievement more than the education level of the mother (ES= .324). International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 465 Table 4. Variables related to family Variables References SI PS IES OES ES 1 Parents attitudes and behaviors Sarier (2016) 2000-2015 62 6 1 .873 2 Parental expectations Fan & Chen (2001) NA 25 10 1 .865 3 Academic socialization parental involvement Hill & Tyson (2009) 1985-2006 50 16 1 .847 4 Parental expectations Jeynes (2005) Jeynes (2007) 1975-2000 1972-2002 41 52 NA NA 2** .730† (.580, .880) 5 Parental involvement in homework Jeynes (2003) 1988-1999 20 NA 1*** .720 6 Maternal employment Goldberg et al. (2008) Until 2005 68 57 1 .699 7 Parental attendance and participation Fan & Chen (2001) NA 25 7 1 .668 8 Parental involvement types (other) Fan & Chen (2001) NA 25 53 1 .622 9 Parental involvement Ma et al. (2016) Hill & Tyson (2009) Sarier (2016) Fan & Chen (2001) After 1990 1985-2006 2000-2015 NA 46 50 62 25 100 32 6 92 4 .538† (min: .080, max: 1.183) 10 Communication parental involvement Jeynes (2003) 1988-1999 20 NA 1*** .530 11 Education level of father Sarier (2016) 2000-2015 62 5 1 .519 12 Parental reading Jeynes (2005) 1975-2000 41 NA 1** .420 13 Communication parental involvement Fan & Chen (2001) NA 25 10 1 .391 14 School based parental involvement Hill & Tyson (2009) 1985-2006 50 21 1 .387 15 Parental involvement Jeynes (2003) 1988-1999 20 NA 6*** .378† (min: .220, max: .480) 16 Parental involvement Goldman & Burke (2017) Jeynes (2005) Jeynes (2007) Jeynes (2012) Jeynes (2015) Jeynes (2016) Until 2015 1975-2000 1972-2002 1964-2006 NA 1970-2013 8 41 52 51 28 42 8 NA NA 51 NA NA 9** .361† (min: -.080, max: .740) 17 Parental style Jeynes (2005) Jeynes (2007) 1975-2000 1972-2002 41 52 NA NA 2** .355† (.310, .400) 18 Authoritative parenting style Pinquart (2016) Until 2015 308 29 1 .345 19 Specific parental involvement Jeynes (2005) Jeynes (2007) 1975-2000 1972-2002 41 52 NA NA 2** .340† (.290, .390) 20 Education level of mother Sarier (2016) 2000-2015 62 5 1 .324 21 Specific parental involvement Jeynes (2003) 1988-1999 20 NA 2*** .305† (.300, .310) 22 Mother involvement in education Hill et al. (2015) 1980-2013 52 23 1 .303 23 Parental reading Jeynes (2003) 1988-1999 20 NA 2*** .300† (.210, .390) 24 Parental responsiveness (warmth) Pinquart (2016) Until 2015 308 53 1 .283 25 Parental expectations Jeynes (2003) 1988-1999 20 NA 3*** .280† (min: -.290, max: .620) 26 Home-based parental involvement (activities at home) Hill & Tyson (2009) 1985-2006 50 19 1 .242 27 Communication parental involvement Jeynes (2005) Jeynes (2007) 1975-2000 1972-2002 41 52 NA NA 2** .240† (.240, .240) 28 Father involvement in education Jeynes (2015) 1974-2012 66 66 2 .222† (.160, .283) 29 Autonomy granting parenting Pinquart (2016); Vasquez et al. (2016) Until 2015 1986-2011 308 36 17 29 2 .221† (.221, .221) 30 Behavioral control parenting Pinquart (2016) Until 2015 308 45 1 .221 31 Parental style Jeynes (2003) 1988-1999 20 NA 2*** .215† (-.010, .440) --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 44 Psychological control parenting Pinquart (2016) Until 2015 308 21 1 -.221 45 Harsh control parenting Pinquart (2016) Until 2015 308 12 1 -.324 *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES Psychological control parenting (ES= -.324) and harsh control parenting (ES= -.221) had small effects on academic achievement- as can be seen in Table 4. On the other hand, it was remarkable that behavioral control parenting (ES= .221) had small but positive effects. Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 466 3.4. SES & SDC and Individual Characteristics 45 variables related to SES, SDC and students’ individual characteristics which were obtained from 35 meta-analyses and 87 ES values were found. On examining Table 5, it is clear that the majority of the variables (N=35) have small or very small effects. The other variables have medium (N=6), large (N=2), and very large (N=2) effects. According to Table 5, grade retention (ES= 1.616) and college admissions test (ES= 1.416) can be said to have very large effects on academic achievement. Besides, high school GPA (ES= .907) and class attendance (ES= .899) have large effects on academic achievement. It was found that ACT/SAT (ES= .792), bible literacy (ES= 0.730), general intelligence (ES= .610), socio-economic status (ES= .547), conscientiousness (ES= .539) and level examination (ES= .515) had medium effects. Conscientiousness, one of the variables with the highest number of overall effect size (OES=7) had medium effect (ES= .539) and extraversion (See Appendix A) had very small effect (ES= -.011) on academic achievement. Openness, which ranked second according to number of overall effect size (OES=6) was found to have small effect (ES= .281) on academic achievement. International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 467 Table 5. Variables related to SES & SDC and individual characteristics Variables References SI PS IES OES ES 1 Grade retention Allen et al. (2009) 1990-2007 22 207 1 1.616 2 College admissions test Kreiter & Kreiter (2007) After 1991 12 NA 3 1.416† (min: 1.283, max: 1.540) 3 High School GPA Robbins et al. (2004) After 1984 109 30 1 .907 4 Class attendance Credé et al. (2010) 1927-2009 90 33 1 .899 5 ACT/SAT Robbins et al. (2004) After 1984 109 31 1 .792 6 Bible literacy Jeynes (2010) 1970-2007 11 3 1 .730 7 General intelligence Von Stumm et al. (2011) Poropat (2009) NA Until 2007 11 80 NA 47 2 .610† (.473, .747) 8 Socio-economic status White, K. R. (1982) Sarier (2016) Sirin (2005) Robbins et al. (2004) Strenze (2007) NA 2000-2015 1990-2000 After 1984 1929-2003 101 62 58 109 NA 620 11 102 13 27 5 .547† (min: .221, max: 1.094) 9 Conscientiousness Vedel (2014) Poropat (2009) Poropat (2014a) Richardson et al. (2012) Trapmann et al. (2007) O’Connor & Paunonen (2007) Von Stumm et al. (2011) 1996-2013 Until 2007 NA 1997-2010 After 1980 1991-2006 NA 20 80 12 217 58 23 11 21 138 23 69 41 23 NA 7 .539† (min: .387, max: .953) 10 Level examinations Peers & Johnston (1994) 1954-1983 20 120 3* .515† (min: .350, max: .620) 11 Number acuity Chen & Li (2014) Until 2013 36 35 1 .494 12 Shifting ability Yeniad et al. (2013) Until 2011 18 34 2 .484† (.430, .539) 13 Study habits Credé & Kuncel (2008) 1980-2005 344 102 1 .473 14 Time/study management Richardson et al. (2012) 1997-2010 217 7 1 .451 15 Creativity Gajda et al. (2017) NA 120 782 1 .451 16 Within-language oral proficiency Prevoo et al. (2016) NA 88 4 1 .451 17 Typical intellectual engagement Von Stumm et al. (2011) NA 11 NA 1 .408 18 Financial support Robbins et al. (2004) After 1984 109 5 1 .398 19 Study habits and attitudes- metacognition Credé & Kuncel (2008) 1980-2005 344 7 1 .366 20 National mathematics performance Wang & Lin (2009) NA 16 28 1*** .350 21 Emotional intelligence Ranjbar et al. (2017) Richardson et al. (2012) Perera & DiGiacomo (2013) NA 1997-2010 1980-2011 23 217 40 23 14 74 3 .336† (min: .283, max: .408) 22 Eveningness Preckel et al. (2011) 1989-2010 21 6 1 .324 23 Private tutoring expenditure Nam et al. (2017) NA 16 275 1 .303 24 Attributions Fong et al. (2017a) NA 95 52 1 .283 25 Openness Vedel (2014) O’Connor & Paunonen (2007) Poropat (2009) Poropat (2014a) Richardson et al. (2012) Trapmann et al. (2007) 1996-2013 1991-2006 Until 2007 NA 1997-2010 After 1984 20 23 80 12 217 58 21 23 113 22 52 41 6 .281† (min: .120, max: .797) 26 Study habits Sarier (2016) 2000-2015 62 5 1 .279 27 Academic-related skills Robbins et al. (2004) After 1984 109 33 1 .260 28 Social involvement Robbins et al. (2004) After 1984 109 33 1 .250 --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 46 Procrastination Kim & Seo (2015) Richardson et al. (2012) 1984-2014 1997-2010 33 217 82 10 2 -.357† (-.451, -.262) *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type GPA: Grade Point Average, ACT/SAT: American college testing/scholastic assessment test, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES Procrastination (ES= -.357) had small effects and negative effects on academic achievement (See Table 5) but some of the variables (extraversion, crossed laterality, neuroticism, sex (being male) and cross-language relations between oral proficiency and academic achievement) had Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 468 Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) very small and negative effects (See Appendix A). On examining the variables related to sex, being female (ES= .140) was found to have very small and positive effects but being male (ES= -.160) was found to have very small and negative effects on academic achievement (See Appendix A) - which was also remarkable. 3.5. Teacher The findings of the effects of variables related to teachers on students’ academic achievement are shown in Table 6. In relation to the effects of those variables on academic achievement, 16 variables and 17 effect sizes were found in 10 meta-analyses. Teachers’ judgments of students’ academic achievement (ES= 1.622) and students’ intelligence (ES= 1.540) were found to have very large effects on academic achievement. Teacher conscientiousness (self) (ES= .699) was, however, found to have medium effects on academic achievement. Teacher openness (ES= .473), closeness in teacher-child relationship (ES= .430), teacher support (ES= .355), teacher emotional stability (ES= .324), positive teacher-student relationship (ES= .324) and instructor leadership (ES= .267) had small effects on academic achievement. Table 6. Variables related to teacher Variables References SI PS IES OES ES 1 Teachers' judgments of students' academic achievement Sudkamp et al. (2012) 1989-2009 75 73 1 1.622 2 Teachers' judgments of students' intelligence Machts et al. (2016) NA 33 106 1 1.540 3 Teacher conscientiousness (self) Poropat (2014b) NA 16 22 1 .699 4 Teacher openness Poropat (2014b) NA 16 14 1 .473 5 Closeness in teacher–child relationship Nurmi (2012) Until 2011 19 7 1 .430 6 Teacher support Givens Rolland (2012) 1991-2011 49 7 1 .355 7 Teacher emotional stability Poropat (2014b) NA 16 17 1 .324 8 Positive teacher-student relationship Roardo et al. (2011) 1990-2011 99 61 1 .324 9 Instructor leadership Balwant (2016) Sarier (2016) Until 2015 2000-2015 22 62 7 19 2 .267† (.209, .324) --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 14 Negative teacher-student relationship Roardo et al. (2011) 1990-2011 99 28 1 -.366 15 Child dependency in teacher-child relationship Nurmi (2012) Until 2011 19 2 1 -.387 16 Conflicts in teacher–child relationship Nurmi (2012) Until 2011 19 10 1 -.408 *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES On examining the variables with negative effects on academic achievement, it was found that variables such as conflicts in a teacher-child relationship (ES= -.408), child dependency in teacher-child relationship (ES= -.387) and negative teacher-student relationship (ES= -.366) had at least small effects. No other teacher-related variables having negative effects on academic achievement were found. On the other hand, it was remarkable that such positive variables as teacher extraversion, academic integration, classroom management strategies and programs and teacher agreeableness had very small effects on academic achievement (See Appendix A). 3.6. School In the study, 14 variables and 17 effect sizes were found related to school in 14 meta- analyses. Classroom-based physical activities (ES= 2.987) were found to have huge effects on academic achievement. School-based interventions that target executive function (ES= .641) and physical activities (ES= .539) in general had medium effects. According to Table 7, leadership of school director (ES= .498), attending a religious school (ES= .250), building condition (ES= .242) and full-day kindergarten (ES= .240) have small effects. On the other hand, it was found that career education interventions, academic admission interviews, International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 469 institutional integration, and first-year seminars had very small effects on academic achievement (See Appendix A). Table 7. Variables related to school Variables References SI PS IES OES ES 1 Classroom-based physical activity Watson et al. (2017) Until 2017 16 10 1 2.987 2 School-based interventions that target executive function Jacob & Parkinson (2015) 2000-2015 67 104 2 .641† (.629, .652) 3 Physical activity Álvarez-Bueno et al. (2017) Until 2016 26 23 1 .539 4 School culture Sarier (2016); Bektas et al. (2015) 2000-2015 2004-2014 62 25 6 25 2 .498† (.473, .523) 5 Leadership of school director Sarier (2016) 2000-2015 62 6 1 .362 6 Attending a religious school Jeynes (2002) 1970-2002 15 NA 1*** .250 7 Building condition Gunter & Shao (2016) NA 215 NA 1 .242 8 Full-day kindergarten Cooper et al. (2010) 1979-2009 40 43 1* .240 --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 14 School mobility Mehana & Reynolds (2004) 1975-1994 45 NA 2 -.748† (-.797, -.699) *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES On examining the variables with negative effects, it was found that school mobility (ES= - .748) was remarkable. While school mobility had medium effects, school-based mentoring for adolescents had negative and very small effects (See Appendix A). 3.7. Educational Technology In the study, 10 variables that were considered in the category of educational technologies were obtained from 18 meta-analyses. An examination of Table 8 makes it clear that computer- aided teaching (ES= 1.690), the material used in classroom instruction (ES= 1.269) and one- to-one laptop programs (ES= 1.249) have positive and huge effects on academic achievement. In addition to that, information technology (ES= .507) was found to have medium effects on academic achievement. It was also found that technology (ES= .456) in general, augmented reality (ES= .360) and audience response system (ES= .249) had small effects on academic achievement. Table 8. Variables related to educational technology Variables References SI PS IES OES ES 1 Computer-aided teaching Christmann et al. (1997) Yesilyurt (2011) NA 2002-2010 27 25 52 54 2*** 1.690† (.209, 3.170) 2 Material use in classroom instruction Kablan et al. (2013) 2000-2012 57 57 1* 1.269 3 One-to-one laptop programs Zheng et al. (2016) 2001-2015 10 NA 1* 1.249 4 Computer-aided teaching Palavan & Sunğur (2017) Batdi (2015a) Demir & Basol (2014) Thomas et al. (2013) 2002-2014 2006-2014 NA Until 2011 60 78 40 50 60 78 40 55 4** .842† (min: .175, max: 1.162) 5 Computer-aided teaching Zheng (2016) Camnalbur & Erdogan (2008) 2004-2015 1998-2007 29 78 NA NA 2* .744† (.438, 1.050) 6 Information technology Lim & Chang (2003) After 1990 58 52 1* .507 7 Technology Chauhan (2017) 2000-2017 122 212 1* .456 8 Augmented reality Yılmaz & Batdı (2016) 2005-2016 12 NA 1* .360 9 Audience response system Castillo-Manzano et al. (2016) Hunsu et al. (2016) 2008-2012 NA 33 53 53 41 2** .249† (.210, .288) --- 3 more variables with effect size between "-.20" and ".20" are given in Appendix A --- *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 470 Only video game and social network site use, of the variables related to educational technologies influential in academic achievement, were found to have negative but very small effects on academic achievement (See Appendix A). 3.8. Special Education In the study, 16 variables and 30 effect sizes were found from 15 meta-analyses related to special education. The findings of the variables having effects on academic achievement are shown in Table 9. According to Table 9, without reading disability (ES= 1.266) has very large effects. Besides, it is the variable with the biggest number of overall ESs (OES=5). The variables such as reading instruction on reading skills of the student with/at risk of behavioral disorder (ES= 1.020) and enrichment programs on gifted students (ES= .960) were also found to have medium effects on academic achievement. The other variables affecting academic achievement were facilitation of school re-entry of children with cancer (ES= .591), school- based interventions for attention deficit hyperactivity disorder (ES= .565) and gifted education programs for gifted ethnic minority students (ES= .251), respectively; and they had small effects. An examination of the variables related to special education demonstrated that only three of the negative effects and that the rest had positive effects on academic achievement at various levels. Table 9. Variables related to special education Variables References SI PS IES OES ES 1 Without reading disability Kudo et al. (2015) 1957-2013 48 109 5* 1.266† (min: .830, max: 1.850) 2 Reading the instruction on reading skills of the student with/at risk of behavioral disorder Benner et al. (2010) 1970-2010 24 NA 1** 1.020 3 Enrichment programs on gifted students Kim (2016) 1985-2014 13 13 1** .960 4 Facilitation of school re-entry and peer acceptance of children with cancer Helms et al. (2016) NA 6 3 1** .591 5 School-based interventions for attention deficit hyperactivity disorder DuPaul & Eckert (1997) 1971-1995 60 55 2^ .565† (.310, .820) 6 Gifted education programs on gifted ethnic minority students Henfield et al. (2017) 1983-2014 13 13 1** .251 --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 14 Sleep-disordered breathing Galland et al. (2015) Until 2015 16 16 1* -.300 15 Neurocognitive sequelae of treatment for childhood acute lymphocytic leukemia Campbell et al. (2007) 1980-2004 28 41 3** -.530† (min: -.600, max: -.420) 16 Emotional/behavioral disturbance Reid et al. (2004) 1961-2000 25 101 1* -.690 *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type ^: Glass’s Delta, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES On examining the variables affecting academic achievement in negative ways, the variables emotional/behavioral disturbance (ES= -.690) and neurocognitive sequelae of treatment for childhood acute lymphocytic leukemia (ES= -.530) were found to come to the fore. According to Table 9, those variables had medium effects on academic achievement. Another variable having a negative effect was sleep-disordered breathing (ES= -.300). 3.9. Violence In the study, 6 variables and 10 effect sizes were found from 6 meta-analyses about the effects of violence-related variables on academic achievement. The findings are shown in Table 10. Accordingly, the variables with the highest effects were peer victimization (ES= - .292), physically aggressive or violent behavior (ES= -.289) and cyber-victimization (ES= - .283). Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 471 Table 10. Variables related to violence Variables References SI PS IES OES ES --- Variables with effect size between "-.20" and ".20" are given in Appendix A --- 4 Cyber-Victimization Gardella et al. (2017) 1989-2017 7 7 1 -.283 5 Physically aggressive or violent behavior Savage et al. (2017) NA 28 28 1 -.289 6 Peer Victimization Nakamoto & Schwartz (2010) 1978-2007 33 31 5 -.292† (min: -.430, max:-.100) *: Cohen’s d, **: Hedges’ g, ***: Type of ES is unclear, Others: Converted to Cohen’s d †: Mean of more than one effect size of the same type, SI: Search Interval, NA: Not available, PS: Number of primary studies, IES: Number of independent ES, OES: Number of overall ES It is remarkable in Table 10 that all three variables in the Table have medium effects in addition to being close to each other and having negative effects. As in the mean effect size of the variable with the biggest number of overall effect sizes, peer victimization (OES=5). Child sexual abuse, another variable related to violence had negative and very small effects on academic achievement (See Appendix A). On the other hand, it was also found that juvenile delinquency intervention and understanding of and ability to deal with racism had positive and very small effects (See Appendix A). 3.10. The Most Effective Variables on Academic Achievement In the study, 254 variables influencing academic achievement were found in 169 meta- analyses. 427 effect sizes were found for the variables which were considered in nine categories. The distribution of the variables according to positive and negative effect sizes are shown in Figure 3. The fact that the number of variables in Figure 3 was 272 stemmed from the fact that some of the variables had different types of effect sizes. Figure 3. Distribution of variables by effect size According to Figure 3, there are no variables with large, very large and huge negative effects. In addition to that, the majority of the variables have small or very small effects and the number of variables with small and positive effects is greater than others. On examining the variables in the categories shown in Figure 4 on the basis of levels, it became apparent that there were several variables with small effects in the category of psychological characteristics whereas there were variables with large and medium effects in the category of learning theories and teaching strategies. 17 58 23 91 5 40 24 12 1 0 10 20 30 40 50 60 70 80 90 100 Negative Positive Very Small Small Medium Large Very Large Huge Negative Effect Positive Effect (+) Effect Size 0 Effect Size (+) Kocak. Goksu,Goktas 472 Figure 4. Distribution of categories by effect size Variables with large, very large and huge effects were identified by considering all the categories, and they are shown in Table 11 below. Accordingly, it is evident that classroom- based physical activity is the variable with huge effect. Table 11. Variables with the highest effects on academic achievement Category: Variables ES H u g e School: Classroom-based physical activity. 2.987 V e ry L a rg e Educational Technologies: Computer-aided teaching, the material used in classroom instruction, one-to-one laptop programs. between 1.230 and 1.690 Learning Theory and Teaching Strategies: Creative drama, constructivist learning, learning strategy instruction, collaborative learning. SES, SDC and Individual Characteristics: Grade retention, college admissions test. Special Education: Without reading disability. Teacher: Teachers' judgments of students' academic achievement, teachers' judgments of students' intelligence. L a rg e Family: Parents attitudes and behaviors, parental expectations, academic socialization parental involvement. between .812 and 1.173 Learning Theory and Teaching Strategies: 4MAT model, conceptual change text, multiple intelligence, Kolb learning styles model, mind mapping techniques, learning style, constructivist learning, Dunn and Dunn learning style model, project-based learning, peer learning, graphic organizers, cognitive learning strategies, perceptual learning styles, portfolio Psychological Characteristics: Self-efficacy, academic emotions PLA SES, SDC and Individual Characteristics: High school GPA, class attendance Special Education: Reading instruction on reading skills of the student with/at risk of behavioral disorder, enrichment programs on gifted students. As is clear from Table 11, there are no variables related to family and psychological characteristics and having very large and huge effects. A general exanimation of the table International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 473 shows that there are several variables related to learning theory and teaching strategies with large effects. Besides, it is remarkable that there are no variables related to violence in the Table. Another thing remarkable in the Table is that there are no variables with negative and huge, very large and large effects on academic achievement. 4. Discussion This study made a systematic compilation of variables that were considered in meta- analyses as related to or as having effects on academic achievement, and it presented them in nine categories. It was noticed in the categorization of the variables that some of the variables also fitted in different categories and therefore they were placed in the most consistent categories on the basis of researchers’ views. The variables were transformed into Cohen’s d effect size so that the findings obtained in the meta-analyses could be interpreted and discussed. The types of effect size which could not be transformed were given as they were. However, Cohen’s d effect size was used in general in the study. Despite this, it was impossible to compare all the ESs and to discuss them because of the great number of variables, because of the fact that this study included only one level that most of the variables were obtained from one meta-analysis and that meta-analyses had limitations. Therefore, readers are expected to make their own inferences for the conclusions we have reached but we cannot discuss. 254 variables and 427 effect sizes were found from the meta-analyses considered in nine categories in this study. Hattie (2009), in a study analyzing 800 meta-analyses, identified 138 variables related to academic achievement. The researcher, in a study, conducted later and analyzed 1200 meta-analyses, did not give the number of variables clearly but found 195 ESs in total (Hattie, 2015). Schneider and Preckel (2017) found 105 variables from 38 meta- analyses related to higher education. This current study, however, found greater number of variables than the above-mentioned studies. The major reason for it was that meta-analyses related to all levels of teaching were included in the scope of this study. Besides, the fact that the number of meta-analyses increased in recent years was also a factor influential in it. Schneider and Preckel (2017), distinguishing two main categories related to teaching and students, formed sub-categories such as presentation, technology, extra-curricular learning activities, meaningful learning and assessment in the category of teaching variables. The researchers formed the sub-categories of intelligence, preliminary achievement, strategy, motivation, character and contextual in the category of students. Hattie (2015) divided the variables influencing academic achievement into such categories as students, administration, school, peers, home, and teachers. Another study considered the variables influential in academic achievement in such categories as family, individual characteristics and school (Engin-Demir, 2009). This study, on the other hand, distinguished nine categories labelled as “psychological characteristics”, “learning theories and teaching strategies”, “SES & SDC and individual characteristics”, “family”, “teacher”, “special education”, “school”, “educational technologies” and “violence”. The reason for distinguishing different categories in this study was that the study contained a great number of variables and that its scope was broader. Richardson et al (2012) divided psychological variables into five categories as personal characteristics, motivational factors, self-regulated learning strategies, students’ approaches towards learning and psycho-social contextual effects in their meta-analysis investigating the correlations between psychological variables related to university students and academic achievement. On examining the effect sizes of the variables in those categories, it was found that the variables were mostly positively correlated with academic achievement, but that there were also variables which were negatively correlated. This study also found that there were variables with negative effects on academic achievement but that most of the variables had positive effects. Richardson et al (2012) found that the correlations of the variables with academic achievement were between -.24 and .59. In this study, however, the effects of Kocak. Goksu,Goktas 474 psychological characteristics on academic achievement were between -.799 and 1.173. The difference might have stemmed from the fact that Richardson et al (2012) analyzed the correlations between academic achievement and variables but that this study analyzed the effects of variables on academic achievement. The values were almost doubled when the correlation coefficients were transformed into effect sizes (Cohen, 1988). Besides, the fact that the number of variables identified in this study was great could also be a significant factor in the difference. Hattie (2009) found 138 variables that were correlated with academic achievement and found that almost all effect sizes related to the variables were positive. The researcher also found that the variables related to teaching methods were positively correlated with academic achievement. Hattie (2015) found that the effect sizes influencing academic achievement ranged between -.42 and 1.65. The effect sizes found in this current study, however, ranged between -799 and 2.987. The fact that the number of variables identified in this study was greater than the ones identified in Richardson et al (2012) and in Hattie (2009, 2015 stemmed from the fact that this study is more up to date. It was found that the variables in the category of educational technologies affected academic achievement between -.161 and 1.69. Tamim et al (2011), on the other hand, found that educational technologies had small effects (ES= .33) on learning. Some of the variables which were found remarkable in this study are discussed below on the basis of categories. Psychological. On examining the results in detail, it was found that self-efficacy, one of the psychological characteristics, had large effects (ES= 1.173). Schneider and Preckel (2017), on the other hand, concluded that self-efficacy had very large effects (ES=1.81) on students’ academic achievement. Another study found through analyses for the factor of motivation that self-efficacy belief (ES= .822) had large effects (Sipe & Curlette, 1996). That the most effective variable in the category of psychological characteristics was self-efficacy in this study was consistent with the results obtained in the literature. The fact that the number of variables having negative effects in the category of psychological characteristics was also a significant finding obtained in this study. The variables academic emotions NLA (ES= -.799) and amotivation (ES= .610) - which had high negative effects on academic achievement- were also available in this category. Hattie (2015) found that the variable with the greatest negative effect on academic achievement was depression (ES= -.42). Lei and Cui (2016) state that NLA's emotions also contain depression. It was remarkable in this study that the effect sizes found for motivation were similar for motivation (ES= .558) and for amotivation (ES= -.610). Amotivation is expected to have negative effects on academic achievement as motivation is expected to have positive effects on academic achievement. Credé et al. (2017) state that mostly affective factors are the factors affecting and determining academic achievement- which is supportive of the conclusion that psychological characteristics are among the important variables affecting academic achievement. Learning theories and teaching strategies. It was remarkable that the number of variables related to learning theories and teaching strategies which had negative effects was small. The reason for it could be that this study included meta-analyses investigating the effects of a number of independent variables in its scope to improve academic achievement. In addition to that, it might have also stemmed from the fact that some of the results were obtained by calculating the averages for the positive and negative effects sizes of the same variables. That the number of variables in the category of learning theories and teaching strategies having large effects was greater than the variables in other categories was an important result. The situation indicated that learning theories and teaching strategies were stronger in predicting academic achievement. Those were the results consistent with the ones obtained in Hattie (2009) and in Kulik and Kulik (1989). It was found in this category that the variables creative drama, constructivist learning, learning strategy instruction, and collaborative learning had very large effects. Creative drama- which is used physically or orally and which provides students with International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 475 opportunities to practice and to gain experience (Bailey, 1997) - was the variable with the most significant effect in this study. However, the fact that it can have certain negative effects (Batdı & Batdı, 2015) as a teaching strategy on academic achievement due to its difficulty to implement in the classroom should be taken into consideration. That the ES found in the meta- analysis (Erisen & Gunay, 2015) done on constructivist learning- which reflected the constructivist philosophy, one of the approaches which shaped the educational paradigm through the end of the 20th century- was very large indicated that constructivist philosophy would continue shaping educational processes. It was found in Sipe and Curlette (1996) that direct instruction- which did not overlap with constructivist philosophy- was one of the five largest effects sizes. Yet, the fact that the study is not recent should be taken into consideration. It was concluded that collaborative learning - among learning theories and teaching strategies- (Kalaian & Kasim, 2017) had very large effects on achievement. McMaster and Fuchs (2003), in their qualitative review study, emphasized that collaborative learning was influential in academic achievement. Family. It became apparent in relation to the category of a family that parents’ attitudes and behaviors, their expectations and involvement had large effects on academic achievement. A similar review study also concluded that there were correlations between variables related to family and academic achievement (Shute, Hansen, Underwood, & Razzouk, 2011). In a review study Boonk, Gijselaers, Ritzen, and Brand-Gruwel (2018) analysed articles published recently and concerning all levels of teaching found correlations between parental involvement and parental expectations. There are several studies emphasizing that parental involvement and parental expectations were influential in academic achievement. Considering the fact that parental involvement is parents’ right, responsibility and also their social need (Castro et al., 2015); it is natural that there are so many studies demonstrating that parents are so influential in their children’s academic achievement. The importance of parental involvement is apparent in those studies; yet, it should not be forgotten that the obstacles in front of parental involvement may stem from parents themselves and from students, teachers or from a number of social factors- as pointed out by Hornby and Blackwell (2018). It can also be said that parental involvement causes negative effects in some cases; because it was concluded in this study that the variables psychological control parenting and harsh control parenting had small and negative effects on academic achievement. Although parents’ involvement in the process of their children’s education is important, it can be said that involvement without pressure is important. SES, SDC, and individual characteristics. It was remarkable in this category that grade retention and college admissions tests had very large effects on academic achievement. It was found in this study that while grade retention did not have significant effects on academic achievement in the short term in higher education, it had negative effects on schooling outcomes in the long term (Cockx, Picchio, & Baert, 2019) - a result which was not overlapping with the result obtained in this study that grade retention had very large effects. The fact that the study conducted by Cockx et al (2019) included in its scope only higher education but that the meta-analysis (Allen, Chen, Wilson &Hughes, 2009) examined in this study included primary and secondary education only indicated that the effects of grade retention on academic achievement could differ according to levels of education. Besides, the circumstances in the process of learning and teaching should also be taken into consideration in evaluating the conclusion that grade retention promotes academic achievement. In other words, it can be said that grade retention can promote students’ achievement only if appropriate learning and teaching occurs. It is claimed that otherwise, students can even drop out of school (Roderick, 1994). In addition to that, another interpretation could be that grade retention can affect achievement significantly in the following year in the case of students who fail. Another variable, SES, was found to have medium effects on academic achievement. In relation to SES, Kocak. Goksu,Goktas 476 Marks (2017) claimed that the emphasis frequently laid on the educational outcomes of this variable in studies was out of place and that it had very small effects when students’ cognitive skills and previous achievements were considered. Hattie (2009) argues that teachers in addition to students’ individual abilities were significant determiners in situations of students’ learning. Teacher. In this category, it was found that teachers’ positive judgments about students’ academic achievement and intelligence had very large effects on their academic achievement. Teachers’ judgments about their students are the main sources in students’ attributional processes and in the development of their self-concept (Machts, Kaiser, Schmidt, & Möller, 2016); and therefore, they are thought to influence the processes of decision making about students. However, teachers’ prejudices can be influenced by several variables in this process. Kaiser, Südkamp, and Möller (2017) claim that teachers’ judgments about students’ academic achievement can be influenced by their demographic properties as well as by their achievement. On evaluating the effects of teachers’ judgments on academic achievement, it should not be ignored that those judgments can also be related to class properties (Kaiser et al., 2017). Classroom management strategy, a variable that is thought to be important for teachers, had positive but very small effects on students’ academic achievement. It was demonstrated in the literature that classroom management affected learning directly and that it was more effective than some of the policies affecting learning indirectly (Wang & Haertel, 1993). This is a conclusion which does not overlap with our conclusion that classroom management strategies have very small effects on academic achievement. Another element remarkable in this category was the availability of several variables about teacher-student relations (closeness in a teacher-child relationship, positive teacher-student relationship, negative teacher-student relationship, child dependency in teacher-child relationship, conflicts in teacher-child relationship). Those variables were found to have small effects. Positiveness and closeness in teacher-student relations affect academic achievement in positive ways, but negativeness, conflicts or dependency in relations affects academic achievement in negative ways. In addition to that, a review study concluded that teacher-student relations had correlations with such variables as students’ psychological states, involvement in school, academic achievement, expelling from school and dropping out of school (Quin, 2017). This situation shows that teacher-student relations occupy an important place in school life. School. It was found in this category that the variable classroom-based activities had huge effect on academic achievement and that the variable had the biggest ES found in this study. The huge ES found in this study demonstrated the strength of the correlations and supported the conclusions reached in previous studies. In a review study, Donnelly et al. (2016) analyzed 73 studies and found that there were correlations between physical activities and academic achievement. Especially studies centering on elementary level lay emphasis on elements hindering classroom-based activities. Accordingly, factors such as time consumption and curriculum pressures in addition to lack of resources, space and competence can hinder the efficiency of classroom-based activities (Knudsen, Skovgaard, & Bredahl, 2018). Another variable remarkable in this category was school mobility. It was found that school mobility had negative and medium effects on academic achievement. Even though it was the result obtained on the basis of only one meta-analysis (Mehana & Reynolds, 2004), it is known that there are many studies concluding that school mobility has negative effects on students’ academic achievement (Herbers, Reynolds, & Chen, 2013). Besides, the longitudinal study conducted in 25 years by Herbers et al. (2013) indicated that individuals who experienced in nursery school or in K12 were more likely to fail to graduate from school on time, to fail to get a popular job and to have symptoms of depression. International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 477 Educational technology. It became apparent that computer-aided teaching and the use of educational technologies in classes in general affected students’ academic achievement in positive ways. Considering the fact that educational technologies are included in educational environments so as to improve teaching and learning processes, it can be said that the result we had obtained was the expected result. This result did not overlap with the one obtained in Tamim et al. (2011) - who had found that educational technologies had small effects on academic achievement- because the majority of the variables in this study had medium and large effects. Some of the variables in this category had negative and very small effects on academic achievement. The fact that variables such as video games and network site use had negative effects on academic achievement can be considered as an expected result. Young et al. (2012), in their review study, demonstrated that video games had weak effects on academic achievement. Sahin, Gumus, and Dincel (2016) stress that there are negative correlations between addiction to games and academic achievement. Our finding in this study that effect size was negative and very small indicated that there was need for experimental studies to demonstrate more clearly the correlations between video games and academic achievement. Special education. It was remarkable in this category that without reading disability had very large effects. This result showed that having no difficulty in reading affected students’ academic achievement largely. The other variables which were considered in this category and which had large effects on academic achievement were related to students with behavioral disorders and intellectually gifted students. Thus, it was concluded that reading instructions given to students with behavioral disorders to improve their reading skills would have large effects on their academic achievement. Besides, enrichment programs to be used with intellectually gifted children would also have large effects on achievement. The result for the variables in this category demonstrates that the processes followed to a plan and a schedule for students needing special education yield positive results in general. When such programs are not implemented students’ achievement will be affected in negative ways; because emotional/behavioral disturbance - another variable that we identified- was found to have negative and medium effects on students’ achievement. Yet, methods of treatment that students with special status have to receive can have negative effects on academic achievement; because neurocognitive sequelae treatment given to lymphoblastic leukemia patients in this study affected academic achievement in negative ways. Another result remarkable in this study was that sleep-disordered breathing had negative and at least small effects on academic achievement. The fact that some of the studies associate sleep problems with low academic achievement is supportive of this result (Abdulghani et al., 2012; Rasekhi, Pour Ashouri, & Pirouzan, 2016). Violence. It was the category with the smallest number of variables having positive effects on academic achievement. Considering that any type of violence that children can be exposed to (for instance, peer victimization, family violence, community violence) is capable of affecting children’s development in negative ways (Vaillancourt & McDougall, 2013), this finding can be considered as an expected result. Moreover, one of those negative results can befall in academic achievement- which is an important indicator in children’s academic life. This is because the academic achievement of children who are exposed to violence decreases and this situation can raise the probability of grade retention for those children (Fry et al., 2018). Even though the variables such as cyber-victimization, physically aggressive or violent behavior, and peer-victimization had small effects on academic achievement in this study, they need considering because it is claimed that there can be problems in the development of students who are exposed to such situations in the following years (Vaillancourt & McDougall, 2013). 5. Limitations Kocak. Goksu,Goktas 478 Since the findings obtained in this study are the systematic compilations of meta-analyses, they contain the limitations of those meta-analyses. The fact that the ESs whose averages were calculated were found from meta-analyses of differing qualities and scopes should also be considered as a limitation. The major reason for using this method- which is unique in the literature- was to facilitate making interpretations in relation to effects on academic achievement rather than to compare the relevant variables with other variables. Within the scope of this study, only meta-analyses that were published in journals indexed by the WoS database were analyzed. Certain keywords were used in searching on WoS database in accordance with the purpose of this study. Considering this situation as a limitation stemmed from naming the concepts differ in the literature. The fact that some of the variables which were known to be correlated with academic achievement in the literature but which were not encountered in meta-analyses and thus which were not included in this study can also be considered as a limitation. A great number of variables were included in one category although they could have been included in many categories in the process of categorizing them. The effect sizes of some of the variables whose ESs were not given as Cohen’s d were transformed into Cohen’s d by using the methods available in the literature so as to be able to present the variables systematically and in integrity. 6. Conclusion and Recommendations It was important in this study that the variables which were considered important for their effects on academic achievement were compiled from meta-analyses which enabled researchers to bring together the results obtained in quantitative studies and thus to make inferences. This study divided 254 variables and 427 effect sizes into nine categories. The variables had effect sizes ranging between -.799 and 2.987. It was found that there were variables having negative effects in all categories but that the number of the variables was small in categories apart from the category of psychological characteristics. There were variables in the categories of psychological characteristics, school and special education having significant negative effects, but the negative variables in other categories had small effects. It was concluded that there were variables with very large and positive effects in all categories except for the categories of psychological characteristics, family and violence. Many variables attracting researchers’ attention in every category were discussed in the literature. Yet, a great number of variables were presented in tables to make the paper readable. It was thought that the rank of ESs for the variables in relation to each other would not be determined in the evaluation and therefore they should be evaluated from a holistic perspective. The following ten golden recommendations might be made in the light of the results obtained in this study: 1. The variables which were found to affect students’ academic achievement in negative ways could be taken into consideration in planning the learning and teaching processes and the existing conditions could be revised accordingly. 2. It should not be forgotten that teachers’ relations, especially with students and the probable conflicts with them, can influence achievement in negative ways. Additionally, it should be remembered that teachers’ adoption of constructivist approach in teaching, their emphasis on physical activities, making use of educational technologies and their positive attitudes towards students can affect students’ academic achievement significantly. 3. Students’ self-confidence, positive thoughts, no hesitation to learn collaboratively in school, avoiding superficial approaches towards learning and avoiding conflicts with their teachers can be important in their academic achievement. International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 479 4. Parents should consider the fact that their children’s academic achievement might be affected in negative ways if they have conflicts with their children, if they display pressurizing attitudes towards their children or if they use violence against their children. 5. It is considered important for school administrators to create areas for physical activities. 6. It is important for the users of social network sites to resort to authorities from whom they can receive help (school administration, parents, the police, etc.) without wasting any time if they are exposed to cyber violence or peer violence. 7. The fact that grade retention had high ES value in this study indicated that countries that do not use the method in their educational system should develop policies accordingly. 8. School mobility mostly stems from family displacement (Rumberger, 2003), and it is highly probable to affect individuals both cognitively and psychologically. Therefore, parents, company managers, and statesmen should take it into consideration and make decisions or develop policies accordingly. 9. The limited number of similar review studies makes it difficult for us to reach clear conclusions or to make comparisons. Therefore, conducting review studies in a narrower framework would facilitate reaching more clear conclusions about the variables affecting academic achievement. 10. The rapid change of the dynamics of the society in the 21st century modifies students’ teachers’ and parents' needs and expectations and, consequently, it also causes changes in school culture. Thus, the validity of the experimental studies which are performed individually can be lost. In this respect, it is considered important for meta-analyses and systematic reviews to focus on up to date studies in particular (maybe the last decade and no more). 7. Conflict of Interest The authors declare that there is no conflict of interest. 8. Acknowledgement We thank Professor Yavuz Akpınar, Associate Professor Ömer Faruk Ursavaş, and Associate Professor Özlem Baydaş for their comments on the manuscript. Kocak. Goksu,Goktas 480 References Abdulghani, H. M., Alrowais, N. A., Bin-Saad, N. S., Al-Subaie, N. M., Haji, A. M. A., & Alhaqwi, A. I. (2012). Sleep disorder among medical students: Relationship to their academic performance. Medical Teacher, 34(sup1), S37–S41. https://doi.org/10.3109/0142159X.2012.656749 Aghaei Chadegani, A., Salehi, H., Md Yunus, M. M., Farhadi, H., Fooladi, M., Farhadi, M., & Ale Ebrahim, N. (2013). A comparison between two main academic literature collections: Web of Science and Scopus databases. Asian Social Science, 9(5), 18–26. https://doi.org/10.5539/ass.v9n5p18 Allen, C. S., Chen, Q., Willson, V. L., & Hughes, J. N. (2009). Quality of research design moderates effects of grade retention on achievement: A meta-analytic, multilevel analysis. Educational Evaluation and Policy Analysis, 31(4), 480–499. https://doi.org/10.3102/0162373709352239 Álvarez-Bueno, C., Pesce, C., Cavero-Redondo, I., Sánchez-López, M., Garrido-Miguel, M., & Martínez-Vizcaíno, V. (2017). Academic achievement and physical activity: A meta- analysis. Pediatrics, 140(6), 1–14. https://doi.org/10.1542/peds.2017-1498 Bailey, S. (1997). Drama: A powerful tool for social skill development. Disability Solutions, 2(1), 3–15. Batdı, V., & Batdı, H. (2015). Effect of creative drama on academic achievement: A meta- analytic and thematic analysis. Educational Sciences: Theory & Practice, 15(6), 1459– 1470. https://doi.org/10.12738/estp.2015.6.0156 Boonk, L., Gijselaers, H. J. M., Ritzen, H., & Brand-Gruwel, S. (2018). A review of the relationship between parental involvement indicators and academic achievement. Educational Research Review, 24, 10–30. https://doi.org/10.1016/j.edurev.2018.02.001 Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta-analysis. https://doi.org/10.1002/9780470743386 Castro, M., Expósito-Casas, E., López-Martín, E., Lizasoain, L., Navarro-Asencio, E., & Gaviria, J. L. (2015). Parental involvement on student academic achievement: A meta- analysis. Educational Research Review, 14, 33–46. https://doi.org/10.1016/J.EDUREV.2015.01.002 Chamberlain, S. A., Hovick, S. M., Dibble, C. J., Rasmussen, N. L., Van Allen, B. G., Maitner, B. S., … Whitney, K. D. (2012). Does phylogeny matter? Assessing the impact of phylogenetic information in ecological meta-analysis. Ecology Letters, 15(6), 627–636. https://doi.org/10.1111/j.1461-0248.2012.01776.x Chen, H., Yang, Y., Yang, Y., Jiang, W., & Zhou, J. (2014). A bibliometric investigation of life cycle assessment research in the web of science databases. International Journal of Life Cycle Assessment, 19(10), 1674–1685. https://doi.org/10.1007/s11367-014-0777-3 Cockx, B., Picchio, M., & Baert, S. (2019). Modeling the effects of grade retention in high school. Journal of Applied Econometrics, 34(3), 403–424. https://doi.org/10.1002/jae.2670 Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). USA: Lawrence Erlbaum Associates. Credé, M., & Kuncel, N. R. (2008). Study habits, skills, and attitudes: The third pillar supporting collegiate academic performance. Perspectives on Psychological Science, 3(6), 425–453. https://doi.org/10.1111/j.1745-6924.2008.00089.x Credé, M., Tynan, M. C., & Harms, P. D. (2017). Much ado about grit: A meta-analytic synthesis of the grit literature. Journal of Personality and Social Psychology, 113(3), 492– 511. https://doi.org/10.1037/pspp0000102 International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 481 Donker, A. S., de Boer, H., Kostons, D., Dignath van Ewijk, C. C., & van der Werf, M. P. C. (2014). Effectiveness of learning strategy instruction on academic performance: A meta- analysis. Educational Research Review, 11, 1–26. https://doi.org/10.1016/J.EDUREV.2013.11.002 Donnelly, J. E., Hillman, C. H., Castelli, D., Etnier, J. L., Lee, S., Tomporowski, P., … Szabo- Reed, A. N. (2016). Physical activity, fitness, cognitive function, and academic achievement in children: A systematic review. Medicine and Science in Sports and Exercise, 48(6), 1197–1222. https://doi.org/10.1249/MSS.0000000000000901 Engin-Demir, C. (2009). Factors influencing the academic achievement of the Turkish urban poor. International Journal of Educational Development, 29(1), 17–29. https://doi.org/10.1016/j.ijedudev.2008.03.003 Erisen, Y., & Gunay, R. (2015). A meta-analysis into the effectiveness of doctoral dissertations on constructivist learning. Anthropologist, 21(1–2), 202–212. https://doi.org/10.1080/09720073.2015.11891809 Flashman, J. (2012). Academic achievement and its impact on friend dynamics. Sociology of Education, 85(1), 61–80. https://doi.org/10.1177/0038040711417014 Fry, D., Fang, X., Elliott, S., Casey, T., Zheng, X., Li, J., … McCluskey, G. (2018). The relationships between violence in childhood and educational outcomes: A global systematic review and meta-analysis. Child Abuse & Neglect, 75, 6–28. https://doi.org/10.1016/j.chiabu.2017.06.021 Gajda, A., Karwowski, M., & Beghetto, R. A. (2017). Creativity and academic achievement: A meta-analysis. Journal of Educational Psychology, 109(2), 269–299. https://doi.org/10.1037/edu0000133 Galland, B., Spruyt, K., Dawes, P., McDowall, P. S., Elder, D., & Schaughency, E. (2015). Sleep disordered breathing and academic performance: A meta-analysis. Pediatrics, 136(4), e934–e946. https://doi.org/10.1542/peds.2015-1677 Hattie, J. (2009). Visible learning : a synthesis of over 800 meta-analyses relating to achievement. New York: Routledge. Hattie, J. (2015). The applicability of visible learning to higher education. Scholarship of Teaching and Learning in Psychology, 1(1), 79–91. https://doi.org/10.1037/stl0000021 Herbers, J. E., Reynolds, A. J., & Chen, C.-C. (2013). School mobility and developmental outcomes in young adulthood. Development and Psychopathology, 25(2), 501–515. https://doi.org/10.1017/S0954579412001204 Hornby, G., & Blackwell, I. (2018). Barriers to parental involvement in education: an update. Educational Review, 70(1), 109–119. https://doi.org/10.1080/00131911.2018.1388612 Jeynes, W. H. (2017). A meta-analysis: The relationship between parental involvement and Latino student outcomes. Education and Urban Society, 49(1), 4–28. https://doi.org/10.1177/0013124516630596 Kaiser, J., Südkamp, A., & Möller, J. (2017). The effects of student characteristics on teachers’ judgment accuracy: Disentangling ethnicity, minority status, and achievement. Journal of Educational Psychology, 109(6), 871–888. https://doi.org/10.1037/edu0000156 Kalaian, S. A., & Kasim, R. M. (2017). Effectiveness of various innovative learning methods in health science classrooms: a meta-analysis. Advances in Health Sciences Education, 22(5), 1151–1167. https://doi.org/10.1007/s10459-017-9753-6 Kelley, K., & Preacher, K. J. (2012). On effect size. Psychological Methods, 17(2), 137–152. https://doi.org/10.1037/a0028086 Knudsen, L. S., Skovgaard, T., & Bredahl, T. (2018). Understanding and scaffolding Danish schoolteachers’ motivation for using classroom-based physical activity: study protocol for a mixed methods study. BMJ Open, 8(3), e019857. https://doi.org/10.1136/bmjopen- 2017-019857 Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation Kocak. Goksu,Goktas 482 coefficients for reliability research. Journal of Chiropractic Medicine, 15(2), 155–163. https://doi.org/10.1016/j.jcm.2016.02.012 Kulik, J. A., & Kulik, C.-L. C. (1989). The concept of meta-analysis. International Journal of Educational Research, 13(3), 227–340. https://doi.org/10.1016/0883-0355(89)90052-9 Lei, H., & Cui, Y. (2016). Effects of academic emotions on achievement among Mainland Chinese Students: A meta-analysis. Social Behavior and Personality: An International Journal, 44(9), 1541–1553. https://doi.org/10.2224/sbp.2016.44.9.1541 Machts, N., Kaiser, J., Schmidt, F. T. C., & Möller, J. (2016). Accuracy of teachers’ judgments of students’ cognitive abilities: A meta-analysis. Educational Research Review, 19, 85– 103. https://doi.org/10.1016/J.EDUREV.2016.06.003 Marks, G. N. (2017). Is SES really that important for educational outcomes in Australia? A review and some recent evidence. The Australian Educational Researcher, 44(2), 191– 211. https://doi.org/10.1007/s13384-016-0219-2 Marques, S. C., Gallagher, M. W., & Lopez, S. J. (2017). Hope- and academic-related outcomes: A meta-analysis. School Mental Health, 9(3), 250–262. https://doi.org/10.1007/s12310-017-9212-9 McMaster, K. N., & Fuchs, D. (2003). Effects of cooperative learning on the academic achievement of students with learning disabilities: An update of Tateyama-Sniezek’s review. Learning Disabilities Research and Practice, 17(2), 107–117. https://doi.org/10.1111/1540-5826.00037 Mehana, M., & Reynolds, A. J. (2004). School mobility and achievement: a meta-analysis. Children and Youth Services Review, 26(1), 93–119. https://doi.org/10.1016/J.CHILDYOUTH.2003.11.004 Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & PRISMA Group. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6(7), e1000097. https://doi.org/10.1371/journal.pmed.1000097 OECD. (2018). PISA 2015 results in focus. Retrieved from http://www.oecd.org/pisa/pisa- 2015-results-in-focus.pdf Perera, H. N., & DiGiacomo, M. (2013). The relationship of trait emotional intelligence with academic performance: A meta-analytic review. Learning and Individual Differences, 28, 20–33. https://doi.org/10.1016/J.LINDIF.2013.08.002 Pinquart, M. (2016). Associations of parenting styles and dimensions with academic achievement in children and adolescents: A meta-analysis. Educational Psychology Review, 28(3), 475–493. https://doi.org/10.1007/s10648-015-9338-y Polanin, J. R., Maynard, B. R., & Dell, N. A. (2017). Overviews in education research. Review of Educational Research, 87(1), 172–203. https://doi.org/10.3102/0034654316631117 Quin, D. (2017). Longitudinal and contextual associations between teacher–student relationships and student engagement. Review of Educational Research, 87(2), 345–387. https://doi.org/10.3102/0034654316669434 Rasekhi, S., Pour Ashouri, F., & Pirouzan, A. (2016). Effects of sleep quality on the academic performance of undergraduate medical students. Health Scope, 5(3), e31641. https://doi.org/10.17795/jhealthscope-31641 Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’ academic performance: A systematic review and meta-analysis. Psychological Bulletin, 138(2), 353–387. https://doi.org/10.1037/a0026838 Rivkin, S. G., Hanushek, E. A., & Kain, J. F. (2005). Teachers, schools, and academic achievement. Econometrica, 73(2), 417–458. https://doi.org/10.1111/j.1468- 0262.2005.00584.x Roderick, M. (1994). Grade retention and school dropout: Investigating the association. American Educational Research Journal, 31(4), 729–759. International Online Journal of Education and Teaching (IOJET) 2021, 8(1), 454-484. 483 https://doi.org/10.2307/1163393 Rumberger, R. W. (2003). The causes and consequences of student mobility. The Journal of Negro Education, 72(1), 6–21. https://doi.org/10.2307/3211287 Sahin, M., Gumus, Y. Y., & Dincel, S. (2016). Game addiction and academic achievement. Educational Psychology, 36(9), 1533–1543. https://doi.org/10.1080/01443410.2014.972342 Sarier, Y. (2016). The factors that affects students’ academic achievement in Turkey: A meta- analysis study. Journal of Education), 31(3), 609–627. https://doi.org/10.16986/HUJE.2016015868 Sawilowsky, S. S. (2009). New effect size rules of thumb. Journal of Modern Applied Statistical Methods, 8(2), 597–599. https://doi.org/10.22237/jmasm/1257035100 Schneider, M., & Preckel, F. (2017). Variables associated with achievement in higher education: A systematic review of meta-analyses. Psychological Bulletin, 143(6), 565– 600. https://doi.org/10.1037/bul0000098 Shute, V. J., Hansen, E. G., Underwood, J. S., & Razzouk, R. (2011). A review of the relationship between parental involvement and secondary school students’ scademic schievement. Education Research International, 2011, 1–10. https://doi.org/10.1155/2011/915326 Sipe, T. A., & Curlette, W. L. (1996). A meta-synthesis of factors related to educational achievement: a methodological approach to summarizing and synthesizing meta-analyses. International Journal of Educational Research, 25(7), 583–698. https://doi.org/10.1016/s0883-0355(96)00021-3 Steenbergen-Hu, S., Makel, M. C., & Olszewski-Kubilius, P. (2016). What one hundred years of research says about the effects of ability grouping and acceleration on K–12 students’ academic achievement. Review of Educational Research, 86(4), 849–899. https://doi.org/10.3102/0034654316675417 Tamim, R. M., Bernard, R. M., Borokhovski, E., Abrami, P. C., & Schmid, R. F. (2011). What forty years of research says about the impact of technology on learning: A second-order meta-analysis and validation study. Review of Educational Research, 81(1), 4–28. https://doi.org/10.3102/0034654310393361 TIMSS & PIRLS. (n.d.). PIRLS (2016) International results. Retrieved June 11, 2019, from https://timssandpirls.bc.edu/ UNESCO. (2019). Education : Expenditure on education as % of GDP (from government sources). Retrieved April 14, 2019, from http://data.uis.unesco.org/?queryid=181 Vaillancourt, T., & McDougall, P. (2013). The link between childhood exposure to violence and academic achievement: Complex pathways. Journal of Abnormal Child Psychology, 41(8), 1177–1178. https://doi.org/10.1007/s10802-013-9803-3 Vo, H. M., Zhu, C., & Diep, N. A. (2017). The effect of blended learning on student performance at course-level in higher education: A meta-analysis. Studies in Educational Evaluation, 53, 17–28. https://doi.org/10.1016/J.STUEDUC.2017.01.002 Voltmer, K., & von Salisch, M. (2017). Three meta-analyses of children’s emotion knowledge and their school success. Learning and Individual Differences, 59, 107–118. https://doi.org/10.1016/J.LINDIF.2017.08.006 von Stumm, S., Hell, B., & Chamorro-Premuzic, T. (2011). The hungry mind: Intellectual curiosity is the third pillar of academic performance. Perspectives on Psychological Science, 6(6), 574–588. https://doi.org/10.1177/1745691611421204 Wang, M. C., & Haertel, G. D. (1993). What helps students learn? Synthesis of research. Educational Leadership, 51(4), 74. https://doi.org/Article White, K. R. (1982). The relation between socioeconomic status and academic achievement. Psychological Bulletin, 91(3), 461–481. Retrieved from https://eric.ed.gov/?id=EJ266055 Wilson, D. B., & Lipsey, M. W. (2001). The role of method in treatment effectiveness research: Kocak. Goksu,Goktas 484 Evidence from meta-analysis. Psychological Methods, 6(4), 413–429. https://doi.org/10.1037/1082-989x.6.4.413 Young, M. F., Slota, S., Cutter, A. B., Jalette, G., Mullin, G., Lai, B., … Yukhymenko, M. (2012). Our princess is in another castle: A review of trends in serious gaming for education. Review of Educational Research, 82(1), 61–89. https://doi.org/10.3102/0034654312436980 Zaff, J. F., Donlan, A., Gunning, A., Anderson, S. E., McDermott, E., & Sedaca, M. (2017). Factors that promote high school graduation: A review of the literature. Educational Psychology Review, 29(3), 447–476. https://doi.org/10.1007/s10648-016-9363-5 Zhang, J., Yu, Q., Zheng, F., Long, C., Lu, Z., & Duan, Z. (2016). Comparing keywords plus of WOS and author keywords: A case study of patient adherence research. Journal of the Association for Information Science and Technology, 67(4), 967–972. https://doi.org/10.1002/asi.23437