Determinants of academic achievement: systematic review of 25 years of meta-analyses1

Condicionantes del rendimiento académico: revisión sistemática de 25 años de meta-análisis

https://doi.org/10.4438/1988-592X-RE-2022-398-552

Belén Gutiérrez-de-Rozas

https://orcid.org/0000-0003-4210-3270

Esther López-Martín

https://orcid.org/0000-0002-0367-2019

Universidad Nacional de Educación a Distancia

Elvira Carpintero Molina

https://orcid.org/0000-0003-1223-6857

Universidad Complutense de Madrid

Abstract

This work is a continuation of the review carried out by Sipe and Curlette (1997), which synthesized the results of 103 meta-analyses published between 1984 and 1993 aimed at studying the variables that influenced academic performance. Knowing the aspects that enhance or hinder students’ academic performance is key to improving it. Therefore, in this paper we perform a review of 80 meta-analyses published between 1994 and 2019 with 127 effect sizes that have analyzed the relationship between personal, family, school and teacher variables and students’ academic performance. The results provide an overview of the characteristics of the meta-analyses identified in relation to their search process, the selection and coding of the primary studies, their methodology, and the characteristics of the selected studies. Also, an estimate of the effect size of each of the determinants of academic performance is calculated from the 127 effect sizes distributed by these meta-analyses. The above shows that the personal variables that have the greatest influence on academic performance are prematurity, student’s previous performance, intelligence, and health. Among the family factors, the absence of the father, mistreatment received by the family environment and socioeconomic status stand out. The school aspects that have shown the greatest weight on students´ results are classroom climate, measures to reduce misbehavior and school organization. Finally, among the variables associated with the teacher, the teachers´ own characteristics, their relationship with the students and the quality of teaching have demonstrated to be the most important. For all these reasons, the review conducted in this paper in relation to the determinants of academic performance will facilitate the adoption of better decisions when addressing its improvement.

Key words: Academic achievement, Academic failure, Systematic review, Meta-analysis, Meta-synthesis

Resumen

Este trabajo supone una continuación de la revisión realizada por Sipe y Curlette (1997), en la que se sintetizaron los resultados de 103 meta-análisis publicados entre 1984 y 1993 destinados a estudiar las variables que influían en el rendimiento académico. Conocer los aspectos que potencian o dificultan el rendimiento académico de los estudiantes resulta clave para poder favorecer su mejora y, por ello, en este estudio se realiza una revisión de los meta-análisis publicados entre 1994 y 2019 que han analizado la relación entre variables personales, familiares, escolares y docentes y el rendimiento académico del alumnado. Los resultados proporcionan una visión general de las características de los 80 meta-análisis identificados en relación con su proceso de búsqueda, selección y codificación de los estudios primarios, el procedimiento metodológico seguido y las características de los estudios primarios seleccionados. Asimismo, a partir de los 127 tamaños del efecto reportados por estos meta-análisis, se estima un tamaño del efecto global para cada uno de los condicionantes del rendimiento académico. Lo anterior permite observar cómo las variables personales que ejercen una mayor influencia en el rendimiento académico son la prematuridad, el rendimiento previo del alumnado, su inteligencia y su salud. Entre los factores familiares destacan la ausencia del padre, el maltrato recibido por parte del entorno familiar y el estatus socioeconómico. Los aspectos escolares que han demostrado tener un mayor peso sobre los resultados de los estudiantes han sido el clima del aula, las medidas de reducción del mal comportamiento y la organización escolar. Por último, entre las variables asociadas al profesor destacan sus propias características, su relación con los estudiantes y la calidad de la docencia. Por todo ello, la presente revisión contribuye a identificar los principales condicionantes del rendimiento académico, lo cual facilitará la adopción de decisiones adecuadas a la hora de abordar su mejora.

Palabras clave: Rendimiento académico, Fracaso escolar, Revisión sistemática, Meta-análisis, Meta-síntesis

Introduction

While deepening in the concept of academic achievement may seem a simple task due to its familiarity, this term encompasses a great complexity both in its definition and in its evaluation (Bentley, 1966; Stevenson, 2021; York et al., 2015). Said complexity is not only due to the fact that academic achievement can cover a wide range of educational outcomes, ranging from the acquisition of a diploma to the students’ moral development (York et al., 2015), but also to its relation to some elements that are difficult to quantify (Mozammel et al., 2021). Moreover, the term academic achievement has a number of interchangeable expressions –such as academic performance or academic success– that make its definition and operationalization even more complex worldwide (Stevenson, 2021). In addition, the ambiguity that characterizes academic achievement is also related to the different perspectives from which success, in general, can be approached (Kumar & Lal, 2014).

Consequently, academic achievement can be considered as a multidimensional concept that evidences the learnings of students at different levels. These learnings are not only linked to the contents acquired by the students, but also to their cognitive, emotional, social, and physical development (Kumar & Lal, 2014). Thus, in general terms, academic achievement shows the level of mastery achieved by students in relation to a series of previously established and diverse learning standards (Robinson & Biran, 2006). According to Fan and Chen (2001), said learning standards range from global indicators –such as permanence in compulsory secondary education or grades– to indicators linked to students’ aspirations or to their academic self-concept, also considering more specific elements –such as the results obtained in standardized tests on a specific subject–.

Research on the determinants of academic performance

Regardless of the approach adopted in the conceptualization and assessment of academic performance, there is no doubt that the level of academic achievement of students is one of the main indicators of the quality of education systems. Therefore, the improvement of education systems requires to deepen in the aspects that influence educational outcomes.

Traditionally, students´ intelligence has been considered the most important conditioning factor of academic performance, being the most studied personal variable in educational and psychological scientific research (Ali & Ara, 2017; Ferragut & Fiero, 2012; Gunawardena et al., 2017; Smedsrud et al., 2019). However, more recent investigations seem to confirm that, although intelligence explains an important part of academic performance, there are numerous factors that, being closely interrelated, contribute to explain the variability of educational outcomes (Akbas-Yesilyurt et al., 2020; Bhowmik, 2019; McCoach et al., 2017; Nisar & Mahmood, 2017; Olmos Rueda & Mas Torelló, 2013).

The large number of empirical studies that have analyzed how these variables predict and explain student learning generates the need to carry out review studies that allow to identify the main determinants of academic performance and their associated effects. For this reason, meta-analyses summarising the empirical evidence on the factors that influence educational outcomes have been conducted since the past century. Said meta-analyses consist on systematic reviews and statistical procedures that provide a quantitative estimate of the mean effect of a variable on the basis of the findings derived from previous studies (Russo, 2007). Also, although less commonly, meta-syntheses on the predictive capacity of certain variables on academic performance have been published, allowing the results from meta-analyses to be compared and summarized (Higgins, 2016).

A meta-synthetic investigation of reference in the field of academic achievement is the review published by Hattie (2017), who analyzed the influence of students´ own characteristics, their families, and various aspects of schools on academic achievement. In his research, the author highlighted the positive influence of some personal variables such as previous high academic achievement and self-efficacy, as well as the pernicious influence of boredom, depression, use of minority languages, superficial motivation, sleep problems, attention deficit hyperactivity disorder and hearing difficulties. The author also demonstrated the positive effects that certain family variables such as home environment and socio-economic status, as opposed to corporal punishment, excessive television consumption, or benefitting from welfare policies, have on academic performance. Moreover, Hattie (2017) observed the influence that school and teacher variables have on academic performance, highlighting the positive effects of teacher effectiveness and the negative influence of aspects such as student suspension, excessively long summer holidays or school changes.

The meta-synthetic work published by Sipe and Curlette (1997) should also be mentioned. In their investigation, the authors conducted a synthesis of 103 meta-analyses published between 1984 and 1993 which were aimed at studying the variables that influenced academic performance. The research is centered on the influence of different personal, family, school, and teacher aspects on students’ academic performance. Also, it provides an in-depth overview of the characteristics of the meta-analyses on which it is based –evidencing the major role of motivation, personal skills, home environment, quality of teaching and classroom social group –.

With the aim of providing an updated overview of the factors that condition the educational outcomes achieved by students and of the characteristics of the meta-analyses that study these factors, this research consists of a systematic review of the meta-analyses that have synthesized the effect of personal, family, school and teacher aspects on academic performance over the last 25 years. Thus, the present study is a continuation of the review carried out by Sipe and Curlette (1997).

Method

This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, as well as its bias control procedures.

The search and selection processes are described below. The inclusion criteria, the coding procedure and the analysis of the coded information are also described in the following sections.

Search procedure

The search for articles was performed in the two main international databases with multidisciplinary coverage: Web of Science and Scopus. ERIC and APA PsycInfo (EBSCOhost) databases, which are specialized in education and psychology, respectively, were also used.

Given that the purpose of this search was to identify meta-analyses aimed at analyzing the effect of personal, family, school and teacher variables on academic performance, a search equation that combined both terms (meta-analysis and academic performance) was used using the Boolean operator “AND” (Table I).

TABLE I. Terms used in the search equation

Meta-analysis

Academic achievement

“meta analysis” OR “meta-analysis” OR “metaanalysis” OR “meta-analytic” OR “meta analytic” OR “metanalytic” OR “meta synthesis” OR “meta-synthesis” OR “metasynthesis” OR “qualitative synthesis” OR “systematic review” OR “systematic literature review” OR “systematic scoping review” OR “systematic qualitative review” OR “systematic quantitative review” OR “systematic meta-review” OR “systematic critical review” OR “systematic mapping review” OR “systematic search and review” OR “systematic integrative review”

“academic* achievement*” OR “academic* performance*” OR “academic* outcome*” OR “academic* success*” OR “academic* competence*” OR “academic* attain*” OR “academic* improvement*” OR “academic* output*” OR “academic* learning*” OR “school* performance*” OR “school* outcome*” OR “school* achievement*” OR “scholastic* achievement*” OR “education* outcome*” OR “education* achievement*” OR “education* attain*” OR “education* improvement*” OR “education* output*” OR “education* performance*” OR “student* achievement*” OR “student* competence*” OR “student* attain*” OR “student* improvement*” OR “student* output*” OR “student* outcome*” OR “student* learning*” OR “student* performance*” OR “performance* level*” OR “learning* outcome*” OR “learning* attain*” OR “learning* achievement*” OR “learning* performance*” OR “achievement* gain*”.

In order to complement and update Sipe and Curlette´s (1997) findings, this search was limited to articles published between January 1994 and December 2019, so that evidence for the 25 years after those years considered in said study could be provided. This process was carried out on October 27, 2020, and resulted in the retrieval of a total of 1230 records. Of these records, 235 came from APA PsycInfo, 187 were from ERIC, 405 belonged to Scopus and 403 were obtained from the Web of Science.

Elegibility criteria

Taking the inclusion criteria proposed by Sipe and Curlette (1997) as a reference, the following inclusion criteria were established for the selection of the studies included in this synthesis:

Together with the above, it should be noted that only studies published in scientific paper format and in the English or Spanish language were considered.

Selection process

The study selection process began by eliminating duplicates, which resulted in a total of 537 unique records. After discarding all documents published in a language other than English or Spanish, or in a format different to a scientific paper, the sample was reduced to 425 articles.

A review of the title and abstract was then performed. From this review, 295 publications that did not meet the aforementioned inclusion criteria were excluded. To avoid selection bias, each of the records was reviewed independently by two researchers, with an agreement rate of 91.43%. This percentage reflects the relationship between the number of agreements and the total number of articles reviewed. Finally, the full text review of the 130 articles considered in the previous phase resulted in a final selection of the 80 articles included in the present meta-synthesis. Graph I shows the flow diagram of the search process and the study selection following PRISMA guidelines (Moher et al., 2009).

GRAPH I. Flow diagram of the study selection process

Coding of variables

A data extraction sheet based on the Sipe and Curlette´s (1997) coding was used to code the information derived from each of the selected meta-analyses. Specifically, variables related to the search process, selection and coding of the primary studies, methodological characteristics of the meta-analyses, characteristics of the primary studies, dependent and independent variables involved, and results obtained were considered.

For the search process and selecting and coding of the studies, information on the following aspects was collected:

The independent variables considered in each of the meta-analyses were also collected and, on the basis of the classification established by Hattie (2009)2, they were classified according to the categories listed in Table II.

TABLE II. Categories considered for the classification of the independent variables

Category

Subcategory

Indicator

Student

Attitudes and dispositions

Attitude to school subjects

Cognitive processes and self-regulation*

Concentration, persistence, and engagement

Emotional intelligence*

Happiness and well-being*

Personality influences

Procrastination and boredom*

Background

Creativity

Intelligence*

Prior achievement

Free time use*

Media use*

Physical attributes

Ethnicity

Exercise

Gender (female)

Health

Sleep*

Prematurity

Other (crossed laterality) *

Family

Family structure

Non-resident fathers (father in prison)

Home environment

Parental involvement in learning

Socioeconomic and cultural status

Cultural capital*

Socioeconomic status

Well-being*

Child maltreatment*

Teacher

Professional development

Professional development

Quality of teaching

Quality of teaching

Teacher characteristics*

Teacher characteristics*

Teacher-student relationships

Teacher-student relationships

School

Classroom compositional effects

Class size

Decreasing disruptive behavior measures*

Mainstreaming

Single-sex classes*

Classroom influences

Climate of the classroom: classroom management

Peer influences

Principals and school leaders

Principals and school leaders

School compositional effects

Out-of-school curriculum experiences

Summer vacation effect

School organization*

Types of schools

Charter schools

Religious schools

* Subcategories and indicators marked with an asterisk have been added to those proposed by Hattie (2009).

Finally, information of each meta-analysis on (1) the estimated mean effect size and (2) the number of effect sizes from which said effect size was estimated was collected.

Data analysis procedure

Based on the coded information, we first analyzed the extent to which the 80 meta-analyses included the selected aspects related to the search process, selection and coding of the primary studies. We also analyzed the methodological procedure used. The main characteristics of the primary studies included in these meta-analyses were also analyzed. For this purpose, the frequencies of appearance and their respective percentages were calculated. Likewise, the main descriptive statistics (minimum, maximum, mean and standard deviation) were estimated for the number of studies included in these reviews.

Secondly, the influence on academic performance of the independent variables extracted from the meta-analyses was analyzed. Specifically, the 127 mean effect sizes reported by the 80 meta-analyses were synthesized according to each of the categories, subcategories and indicators in Table II. The process followed to achieve this purpose consisted of three stages:

FIGURE I. Transformation applied to convert effect sizes to R

Results

Description of the search process, selection and coding of the studies

The results show that, although a high percentage of meta-analyses (80%) did not specify the protocol used (see Table III), PRISMA was the most widespread procedure in these studies (13.75%).

Regarding the search sources, all the authors used databases in their search, being ERIC (68.75%) and PsycInfo (65%) the most widely used. The ancestry method was selected as a secondary search method in more than half of the studies (65%), with the search in specific journals being the least used complementary procedure for the identification of primary studies (12.5%).

Regarding the study selection process, 97.5% of the meta-analyses indicated the inclusion criteria; however, only half of them (45%) detailed the exclusion criteria. Differences were also observed in the degree of specification of the keywords used in the search, since, although 60% of the authors reported the search terms used, only 38.75% provided the complete search equation.

Another remarkable aspect is that in only 17.5% of the meta-analyses more than one researcher intervened in the selection of the studies. Also, the agreement index between researchers was calculated on 5% of occasions. This percentage is higher for the coding of variables, since 67.5% of the meta-analyses reported the participation of more than one researcher in this process. Of these, 40% provided an index of agreement between coders.

TABLE III. Description of the search, selection and coding process of the studies in the 80 meta-analyses considered

Description of the search process, selection and coding of the studies

Yes

Percentage

Protocol

PRISMA

11

13.75%

Other

5

6.25%

Not specified

64

80.00%

Search sources

Databases

80

100.00%

WoS

23

28.75%

Scopus

6

7.50%

ERIC

55

68.75%

PsycInfo

52

65.00%

Medline

10

12.50%

PubMed

11

13.75%

ProQuest Dissertations and Theses

19

23.75%

Google Scholar

20

25.00%

Other

58

72.50%

Ancestry

52

65.00%

Specific journals

10

12.50%

Grey literature

26

32.50%

Other

10

12.50%

Reviews and previous studies

3

3.75%

Books and reports

2

2.50%

Hand search

5

6.25%

Study selection process

Inclusion criteria are specified

78

97.50%

Exclusion criteria are specified

36

45.00%

The search years are specified

66

82.50%

The keywords used are specified

48

60.00%

The search equation used is included

31

38.75%

Controlling for bias in the quality of the studies

27

33.75%

The selection of studies is carried out by several researchers

14

17.50%

The index of agreement between researchers is calculated

4

5.00%

Over 80 %

2

2.50%

Over 90 %

2

2.50%

Variable coding

Information on the coding of variables is provided

68

85.00%

The coding of variables is carried out by several researchers

54

67.50%

The agreement index between coders is calculated

32

40.00%

Over 70 %

3

3.75%

Over 80 %

6

7.50%

Over 90 %

23

28.75%

Methodological characteristics

Considering the methodological characteristics of the meta-analyses (Table IV), publication bias was calculated in 68.75% of the systematic reviews. Funnel plot was the most commonly used procedure (31.25%), followed by fail-safe N (26.25%) and trim and fill (25%).

The statistics mainly extracted from the primary studies were correlations (80%), means and standard deviations (25%) and regression coefficients (13.75%). The main procedure for the calculation of the effect size was R (47.5%), followed by estimation of the standardized mean difference (38.75%) and Fisher’s z (13.75%).

The model used for the estimation of effect sizes was specified in 90% of the meta-analyses, with the random-effects model prevailing over the fixed-effects model (63.75% and 11.25%, respectively). In addition, most of the studies selected evaluated the heterogeneity of the effect size (85%), with Q (62.5%) and I2 (42.5%) being the most commonly used procedures for this purpose.

Finally, the small number of studies reporting the presence or absence of outliers (22.5% and 6.25%, respectively) is noteworthy. In contrast, the confidence interval for the effect size was provided in most of the meta-analyses (85%).

TABLE IV. Description of the methodological procedure followed in the 80 meta-analyses considered

Methodological characteristics

Yes

Percentage

Control of
publication bias

Publication bias is calculated

55

68.75%

Fail-safe N

21

26.25%

Funnel plot

25

31.25%

Spearman rank-order correlation

5

6.25%

Trim and fill

20

25.00%

Egger´s test

17

21.25%

Begg and Mazumdar rank correlation test

4

5.00%

Kendall’s rank correlation

9

11.25%

Moderator analyses

4

5.00%

Other

13

16.25%

Statistics
extracted from primary studies

Correlations

64

80.00%

Means and standard deviations

20

25.00%

Beta

11

13.75%

Odds ratio

4

5.00%

Other

28

35.00%

Procedure for calculating effect sizes*.

Fisher’s z

11

13.75%

Standarized mean difference (Cohen’s d or Hedges’ g)

31

38.75%

Log odds ratio

9

11.25%

R

38

47.50%

Estimation of the mean effect size

Confidence interval is reported

68

85.00%

The presence of outliers is reported

18

22.50%

The ausence of outliers is reported

5

6.25%

The type of estimated model is specified

72

90.00%

Fixed effects model

9

11.25%

Random effects model

51

63.75%

Fixed effects and random effects models

12

15.00%

Heterogeneity analysis

Heterogeneity between effect sizes is evaluated.

68

85.00%

The type of procedure used to assess
heterogeneity is specified

65

81.25%

Q

50

62.50%

I2

34

42.50%

Tau2

5

6.25%

Other

6

7.50%

* Some of the meta-analyses used more than one procedure in the estimation of mean effect sizes.

Characteristics of studies included in meta-analyses

The mean number of primary studies included in the meta-analyses is 58.28, ranging from 2 to 310 publications (Table V). No geographical limitation was established for the primary studies in most cases (81.25%), so the majority of meta-analyses included studies carried out in any country.

Considering the educational stages on which the systematic reviews focused, most of these studies were based on primary investigations that were performed with populations of students from various stages. The highest prevalence was for studies which focused on kindergarten, primary and secondary education (28.75%), followed by meta-analyses that considered primary and secondary education and university (20%).

Finally, with regard to the dependent variable, most of the selected meta-analyses analyzed the effect of personal, family, school and teacher characteristics on students’ overall performance (92.5%), while the remaining 8.75% studied academic performance in a specific academic subject.

TABLE V. Description of the characteristics of the studies included in the 80 meta-analyses considered

Minimum

Maximum

Mean

Std. Deviation

Number of studies included in the meta-analysis

2

310

58.725

59.58

Geographical limitation

N

Percentage

-

-

No

65

81.25%

-

-

Yes

15

18.75%

-

-

Educational stage

N

Percentage

-

-

Kindergarten and primary

1

1.25%

-

-

Kindergarten, primary and secondary

23

28.75%

-

-

Kindergarten, primary, secondary and university

8

10.00%

-

-

Primary

3

3.75%

-

-

Primary and secondary

16

6.25%

-

-

Primary, secondary and university

16

20.00%

-

-

Secondary

5

6.25%

-

-

Secondary and university

5

6.25%

-

-

Measure of DV*

N

Percentage

-

-

General

74

92.50%

-

-

Specific

7

8.75%

-

-

*In one of the meta-analyses, the mean effect size is estimated both for studies that considered specific performance and for those that considered general performance.

Effects of the variables considered on academic performance

This section describes the main variables related to academic performance, taking as a reference the categories considered in Table II. In general terms, the results show the high effect that teacher characteristics have on academic performance in comparison to other variables, with a mean effect size of 0.25. In contrast, the mean effect size for student characteristics was 0.08, and for family and school variables, 0.06. However, according to Hattie (2009), these effect sizes encompass a great internal complexity derived from the diversity of variables that compose them and from the variation in the effect sizes associated with each of them. Due to this, they should be interpreted with caution. Consequently, our study is centered in the effects associated with each of the individual indicators, examining said effects in more detail.

Effects of student characteristics on academic performance

Although the mean effect size for the relationship between students’ characteristics and their academic performance is 0.08, there are remarkable differences in the mean effect sizes associated with the variables that conform this category (Table VI). First, the effect size of the factors associated with background stands out, being positively related to academic performance (r ̅ = 0.34). More specifically, intelligence and previous academic performance have proven to be the aspects most closely linked to educational results, both showing mean effect sizes that, according to Cohen (1992), are medium-high (r ̅ = 0.40 and r ̅ = 0.34, respectively).

Attitudes and dispositions have an overall effect size of 0.16. However, some components of this subcategory, such as cognitive processes and self-regulation, concentration, persistence and engagement, and emotional intelligence, have mean effect sizes equal to or greater than 0.2. Regarding the effect of personality influences, it is worth noting that, despite the fact that certain personality types are negatively related to academic performance, the effect sizes for some others are high (r ̅ = 0.50). By contrast, procrastination and boredom have an inverse relationship with academic performance (r ̅ = -0.15).

Finally, physical attributes and free time use in media are negatively associated with academic performance, although the overall effect sizes for both categories are close to zero. Of note, however, are effect sizes for lack of health (r ̅ = -0.29) and prematurity (r ̅ = -0.32), these being the physical attributes with the most pernicious effect on academic performance.

TABLE VI. Synthesis of the effect of student characteristics on academic performance

Mean

Minimum

Maximum

N summary effect sizes

N effects

Attitudes and
dispositions

.16

-.16

.50

33

-

Attitude to school subjects

.12

-

-

1

29

Cognitive processes and self-regulation

.20

.07

.40

9

2,296

Concentration, persistence, and engagement

.22

.11

.29

6

584

Emotional intelligence

.20

.20

.20

2

1,350

Happiness and well-being

.16

-

-

1

151

Personality influences

.16

-.08

.50

12

884

Procrastination and
boredom

-.15

-.16

-.13

2

103

Background

.34

.22

.54

4

-

Creativity

.22

1

782

Intelligence

.40

.25

.54

2

62

Prior achievement

.34

-

-

1

11

Free time use

-.07

-.16

.08

7

-

Media use

-.07

-.16

.08

7

206

Physical attributes

-.07

-.39

.31

19

-

Ethnicity

.09

-

-

1

87

Exercise

-.01

-.18

.31

3

28

Sleep

.05

-.14

.16

6

99

Gender (female)

.06

-.00

.11

2

538

Health

-.29

-.39

-.11

3

87

Prematurity

-.32

-.36

-.27

3

N/A

Other (cross laterality)

-.02

-

-

1

27

TOTAL STUDENT

.08

-.39

.54

63

-

Effect of family characteristics on academic performance

As in the previous section, although the students’ family characteristics have a small mean effect on academic achievement when considered as a whole (r ̅ = 0.06) (Table VII), the mean effect sizes for each of the subcategories also vary for each of the categories. The fact that the father is away from home and, more specifically, in a situation of internment in a penitentiary center, presents the greatest negative mean effect on academic performance (r ̅ = -0.36). Although this effect comes from a single meta-analysis, it can be affirmed that this situation of absence increases the risk of low achievement among students.

A low mean effect size was observed with respect to parental involvement in learning (r ̅ = 0.09). However, this effect varies greatly depending on the specific aspects of this family involvement, with mean effect sizes ranging from -0.16 to 0.36.

The mean effect size of the socioeconomic and cultural status of the students is 0.14. Although the mean effect size of socioeconomic status is slightly higher than that corresponding to cultural capital, the effects are medium-low in both cases. Finally, the lack of well-being of the children, concretized in situations of maltreatment, presents a mean effect size that can be considered as medium-low (r ̅ = -0.15).

TABLE VII. Synthesis of the effect of family characteristics on academic achievement

Mean

Minimum

Maximum

N summary effect sizes

N effects

Family structure

-.36

-

-

1

-

Non-resident fathers (father in prison)

-.36

-

-

1

13

Home environment

.09

-.16

.35

18

-

Parental involvement in learning

.09

-.16

.35

18

> 1,804*

Socioeconomic and cultural status

.14

.07

.27

5

-

Cultural capital

.13

.10

.16

2

345

Socioeconomic status

.15

.07

.27

3

981

Well-being

-.15

-.32

.19

3

-

Child maltreatment

-.15

-.32

.19

3

105

TOTAL FAMILY

.06

-.36

.35

27

-

* Two of the meta-analyses did not report the number of effects from which the mean effect size was estimated.

Effect of teacher characteristics on academic achievement

Teacher characteristics analyzed in the selected meta-analyses are positively linked to student academic achievement when considered as a whole (r ̅ = 0.22) (Table VIII). Among them, quality of teaching is the most strongly linked to the students’ results. While the overall effect for that subcategory is medium (r ̅ = 0.29), the mean effect size values for some aspects of teacher quality –such as teacher self-regulation– are notably larger (r ̅ = 0.44).

Similarly, although overall the mean effect size for teacher characteristics can be considered as medium-low (r ̅ = 0.21), some specific characteristics, such as leadership, present higher values.

TABLE VIII. Synthesis of the effect of teacher-associated variables on academic performance

Mean

Minimum

Maximum

N summary effect sizes

N effects

Professional
development

.12

-

-

1

-

Professional development

.12

-

-

1

11

Quality of teaching

.29

.10

.44

3

-

Quality of teaching

.29

.10

.44

3

> 98*

Teacher characteristics

.21

.19

.26

2

-

Teacher characteristics

.21

.19

.26

2

1,076

Teacher-student
relationships

.16

-

-

1

-

Teacher-student relationships

.16

-

-

1

N/A

TOTAL TEACHERS

.23

.10

.44

7

-

* One of the meta-analyses does not report the number of effects from which the mean effect size is estimated.

Effect of school characteristics on academic achievement

The results show that the mean effect size for school characteristics is 0.06 (Table VIII). Moreover, there is little variability among the second-level subcategories, which have overall effect sizes that, in general, can be considered as low.

Regarding the different subcategories, the mean effect size for principals and school leaders is 0.14. However, there are remarkable differences in the mean effect sizes reported depending on the aspects of leadership considered in each of the meta-analyses, with values ranging from r ̅ = 0.04 to r ̅ = 0.49.

The mean effect size for the school compositional effects is 0.12, with school organization (school culture) having the highest mean effect size within this subcategory (r ̅ = 0.25).

The subcategories related to the classroom –classroom compositional effects and classroom influences– present mean effect sizes close to zero. Within the former, the negative mean effect of the measures aimed at reducing disruptive behavior (school suspension) stands out (r ̅ = -0.21). In relation to classroom influences, the mean effect size for the association between classroom management and academic achievement (r ̅ = 0.24) is remarkable, reaching a value of 0.42 in one of the selected studies. By contrast, peer influence (bullying) is negatively related to academic achievement, presenting a mean effect size of -0.13.

Finally, the types of school show a negative mean effect size on academic achievement, although there are differences within the subcategory. Thus, a small but negative mean effect size is observed for charter schools (r ̅ = -0.09), while the mean effect size is positive for religious schools (r ̅ = 0.13).

TABLE IX. Synthesis of the effect of school-associated variables on academic achievement

Mean

Minimum

Maximum

N summary effect sizes

N effects

Classroom
compositional effects

.02

-.21

.10

10

-

Class size

.10

-

-

1

120

Decreasing disruptive behavior

-.21

-

-

1

43

Mainstreaming

.06

-

-

1

143

Single-sex classes

.04

.02

.06

7

114

Classroom influences

.05

-.14

.42

4

-

Climate of the classroom: classroom management

.24

.05

.42

2

N/A

Peer influences

-.13

-.14

-.12

2

58

Principals and school leaders

.14

.04

.49

8

-

Principals and school leaders

.14

.04

.49

8

426

School compositional effects

.11

.04

.23

4

-

Out-of-school curriculum experiences

.09

-

-

1

3

School organization

.23

-

-

1

25

Summer vacation effect

.06

.04

.09

2

63

Types of schools

-.03

-.14

.13

4

-

Charter schools

-.09

-.14

.01

3

> 244*

Religious schools

.13

-

-

1

N/A

TOTAL SCHOOL

.06

-.21

.49

30

-

* One of the meta-analyses did not report the number of effects from which the mean effect size was estimated.

Conclusions

The present meta-synthesis, which is proposed as a continuation of Sipe and Curlette´s (1997) work, was aimed at analyzing the relationship between personal, family, school and teacher characteristics and students’ academic achievement. Specifically, we have synthesized the results of 80 meta-analyses published between 1994 and 2019, which provided 127 effect sizes.

In their meta-synthesis, Sipe and Curlette (1997) noted that the Glass procedure, followed by Hedges, was the most commonly used for conducting meta-analyses. However, the most used method in the selected studies of our research was PRISMA. Since it was first published in 2009, it did not appear in the review conducted by these authors (Moher et al., 2009).

An evolution in the search procedures is also observed. Only 84% of the meta-analyses provided information on the search process in the study by Sipe and Curlette (1997), in contrast to the 100% of articles on which this meta-synthesis is based. Furthermore, the most commonly used procedure in the meta-analyses carried out before 1994 was ancestry (68%). It is also noteworthy that 32% of them did not use the computer as a search tool; this contrasts with the widespread use of information and communication technologies today (Dobrota et al., 2012). However, the high number of authors using ERIC –which constitutes the main database specialized in education– is an aspect that coincides with the work done by Sipe and Curlette (1997).

There have also been notable advances towards a greater description and detail of the search processes. This is a very important issue given that replicability constitutes one of the paths to confirm the validity of a new scientific finding (National Academies of Sciences, Engineering, and Medicine, 2019). Sipe and Curlette (1997) identified that many details about the search procedures were not present in the selected meta-analyses, thus hindering the replicability of the studies. For example, only 29% of the meta-analyses indicated the start year and 26% the end year, while 82.5% of the meta-analyses included in our synthesis provided this information. Similarly, whereas in the previous review only 27% of the meta-analyses listed the keywords used, this percentage rises to 60% in our work. Advances are also observed in the information provided on the variables coded, from being described in less than half of the meta-analyses prior to 1994, to being described in 85% of the studies included in this meta-synthesis. Furthermore, whereas in the former revision only 20% of the selected meta-analyses used two coders in the study selection process, this percentage has now risen to 67.5%. There has also been a notable increase in the information provided on the rate of agreement, rising from 3.26% to 40%.

With regards to the methodological procedures, there has been a notable increase in the proportion of meta-analyses reporting the confidence interval: 85% of the meta-analysis in this synthesis compared to the 22% reported by Sipe and Curlette (1997). This fact could be explained by the greater difficulties in performing statistical calculations prior to the development of new techniques, in contrast to the present existence of computer technology and the widespread accessibility of specific data analysis software, all of which has led to a rapid evolution in statistical methodology in recent years (Barreto-Villanueva, 2012; Sagaró & Zamora, 2019). Also, similar values are observed in both works in relation to the percentage of studies reporting the presence of outliers, with these values hovering at around 25% in both cases (26% vs 22.5% in the present work).

Sipe and Curlette (1997) also provided information on the procedures used to calculate the heterogeneity of the effect size, detecting that 13 publications (12.6%) used the Q test. This aspect contrasts with 62.5% of the meta-analyses that used the Q test in our study. Moreover, since the Q test only reports the presence or absence of heterogeneity, I2 is an interesting complement for its quantification (Huedo-Medina, 2006). In our study, 42.5% employed this procedure.

Our results show that there is also a greater use of fail-safe N to calculate publication bias, since the percentage has increased from 9% to 26.25%. This increase is in line with the findings of Heenee (2010), who detected an exponential increase in the use of fail-safe N in meta-analyses between 1979 and 2008. However, our study reveals that other procedures –such as the funnel plot (31.25%) and trim and fill (25%)– are nowadays used to a greater extent than fail-safe N.

Considering the results derived from the effect sizes of student variables3, Sipe and Curlette (1997) identified the highest mean effects for motivational aspects, followed by those related to student skills. These results are in partial agreement with those obtained in the present research, where both student background (r ̅ = 0.34) and student persistence, concentration and engagement (r ̅ = 0.22) are the most strongly related to the personal aspects of academic performance. Hattie’s (2017) findings are also in this line, since he observed that the variables linked to these aspects presented mean effect sizes close to d = 0.5 (r ̅ = 0.24). Leisure time use is also presented as a student variable related to performance in Sipe and Curlette’s (1997) study, although their mean effect size comes from a single paper. In their synthesis, studies on leisure time use have been found in relation to media use, which is negatively linked to student achievement. This may be associated with the large amount of time spent on media not only during adolescence but also at very early ages (Hadders-Algra, 2020; Spina et al., 2021). Beyond these findings, our research has also demonstrated the importance of cognitive processes and self-regulation, emotional intelligence, health and non-prematurity in academic performance.

With respect to family characteristics, although Sipe and Curlette (1997) only studied the home environment, their results are consistent with those obtained in this paper, being the variable with the smallest mean effect size of all those considered. In this vein, although Hattie (2017) did not provide an overall effect size either for family characteristics in general, or for home environment in particular, he reported higher mean effect sizes than those obtained in our synthesis for the categories of parental involvement (r ̅ = 0.24; versus r ̅ = 0.09) and socioeconomic status (r ̅ = 0.25; versus r ̅ = 0.15). These results are also consistent with the investigation of Castro et al. (2015), who found medium effects on the variables related to parent-child communication.

In relation to the factors associated with teachers, Sipe and Curlette (1997) highlighted the effect of quality of instruction. This variable not only presented one of the largest mean effect sizes in our synthesis
(r ̅ = 0.29), but also yielded a similar result to that reported in the work by Hattie (2017)4 (r ̅ = 0.24). Our findings also demonstrated the influence of teacher characteristics on students’ academic performance. However, it should be recalled that our study has excluded from its analysis any research directly related to the effect of specific interventions or methodologies. It is possible that personal or behavioral variables of teachers, as well as classroom management variables, may be directly implicated in many of those studies.

Finally, although Sipe and Curlette (1997) only considered the influence of the classroom social group within the scope of school factors, its low effect size is again consistent with our results for this category. However, we have also detected other variables with higher mean effect sizes, such as the climate of the classroom (classroom management), the school organization (school culture) and the pernicious role of measures to reduce disruptive behavior.

Although the aim of this work was to temporarily extend the research carried out by Sipe and Curlette (1997), we have also identified new personal, family, school and teacher variables that influence students’ academic performance.

The comparison of the results of both studies shows that, although some personal variables –such as cognitive and attitudinal characteristics–, or the quality of teaching have historically maintained their status of predictors of academic performance, the most recent research is considering and demonstrating the role that variables like family involvement, socio-economic status or the climate and culture of classrooms and schools have on academic achievement. Therefore, this study shows an evolution in the explanatory factors of academic performance. Although, in some cases, this evolution might be due to changes in present-day societies, on most occasions, it may be a consequence of an evolution in the variables considered and in the approaches adopted by the scientific community.

Our results thus provide a holistic and updated overview of the factors that may influence students’ academic performance. This constitutes an opportunity for achieving the goal of giving fair and quality education to all students (Iglesias-Díaz & Romero-Pérez, 2021; Vera Sagredo et al. 2021) and for designing and implementing educational policies and interventions. Said interventions would be aimed, on the one hand, at strengthening those factors that contribute to improving academic performance and, on the other hand, at neutralizing the negative effects of the variables identified as pernicious. As shown in our study, said variables are related to having the father in prison, facing situations of abuse as a child, having health problems, or making excessive use of technology, as well as to being influenced by the peer group or receiving measures to reduce misbehavior.

Furthermore, systematized evidence on predictors of academic performance provides an opportunity for international organizations to access updated research on the factors that have proven their influence on student performance. This may help to facilitate the updating and inclusion of new variables in international assessments.

Moreover, our results evidence a methodological improvement in the procedures employed, which incorporate greater rigor in the techniques and specific search processes. However, as Sipe and Curlette (1997) pointed out, the main limitation of meta-analyses and, consequently, of meta-synthesis, is that it is likely that there are variables with influence on academic performance that have not been incorporated in systematic reviews. Similarly, it should be noted that, in a meta-synthesis, it is not possible to have information about aspects such as the controlled variables or the procedures and instruments used by the primary studies, nor is it to ensure homogeneity in the definition of the variables by these studies. Therefore, when interpreting the results, it is necessary to consider that meta-syntheses echo the limitations of the meta-analyses contained in them. Furthermore, when analyzing the findings, it should be remembered that this type of research does not reflect the interactions between variables, but it rather establishes the basis for the aspects that should be considered in confirmatory studies.

In this sense, meta-syntheses such as the one presented here provide solid evidence to draw a comprehensive map of the variables that influence academic performance and to establish the basis for a deeper understanding of the relationships between them.

References

References identified with an asterisk (*) constitute the sample of meta-analyses included in this meta-synthesis.

*Adelantado-Renau, M., Moliner-Urdiales, D., Cavero-Redondo, I., Beltran-Valls, M. R., Martínez-Vizcaíno, V., & Álvarez-Bueno, C. (2019). Association between screen media use and academic performance among children and adolescents: a systematic review and meta-analysis. JAMA pediatrics, 173(11), 1058-1067. https://doi.org/10.1001/jamapediatrics.2019.3176

Akbas-Yesilyurt, F., Kocak, H., & Yesilyurt, M. E. (2020). Spatial models for identifying factors in student academic achievement. International Journal of Assessment Tools in Education, 7(4), 735-752. https://doi.org/10.21449/ijate.722460

Ali, S., & Ara, A. (2017). Intelligence as a determinant of academic achievement: A comparative study of high achievers and underachievers. International Journal of Humanities and Social Sciences (IJHSS), 6(6), 79-88.

*Allotey, J., Zamora, J., Cheong-See, F., Kalidindi, M., Arroyo-Manzano, D., Asztalos, E., van der Post, J. A. M., Mol, B. W., Moore, D., Birtles, D., Khan, S. K., & Thangaratinam, S. (2018). Cognitive, motor, behavioural and academic performances of children born preterm: a meta-analysis and systematic review involving 64 061 children. BJOG: An International Journal of Obstetrics & Gynaecology, 125(1), 16-25. https://doi.org/10.1111/1471-0528.14832

Barreto-Villanueva, A. (2012). El progreso de la Estadística y su utilidad en la evaluación del desarrollo. Papeles de Población, 18(73), 241-271.

*Bektas, F., Çogaltay, N., Karadag, E., & Ay, Y. (2015). School culture and academic achievement of students: A meta-analysis study. The Anthropologist, 21(3), 482-488. https://doi.org/10.1080/09720073.2015.11891837

Bentley, J. C. (1966). Creativity and academic achievement. The Journal of Educational Research, 59(6), 269-272.

Bhowmik, M. K. (2019). Ethnic Minority Young People’s Education in Hong Kong: Factors Influencing School Failure. En Gube, J. & Gao, F. Education, Ethnicity and Equity in the Multilingual Asian Context (pp. 179-195). Springer. https://doi.org/10.1007/978-981-13-3125-1_11

Borenstein, M., Hedges, L. V., Higgins, J. P., & Rothstein, H. R. (2010). A basic introduction to fixed-effect and random-effects models for meta-analysis. Research Synthesis Methods, 1(2), 97-111. https://doi.org/10.1002/jrsm.12

*Bücker, S., Nuraydin, S., Simonsmeier, B. A., Schneider, M., & Luhmann, M. (2018). Subjective well-being and academic achievement: A meta-analysis. Journal of Research in Personality, 74, 83-94. https://doi.org/10.1016/j.jrp.2018.02.007

*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

*Chang, D. F., Chien, W. C., & Chou, W. C. (2016). Meta-analysis approach to detect the effect of student engagement on academic achievement. ICIC Express Letters, 10(10), 2241-2246.

*Chin, J. M. C. (2007). Meta-analysis of transformational school leadership effects on school outcomes in Taiwan and the USA. Asia Pacific Education Review, 8(2), 166-177. https://doi.org/10.1007/BF03029253

Cohen, J. (1992). A power primer. Psychological bulletin, 112(1), 155. https://doi.org/10.1037/0033-2909.112.1.155

*Cortés Pascual, A., Moyano Muñoz, N., & Quilez Robres, A. (2019). The relationship between executive functions and academic performance in primary education: review and meta-analysis. Frontiers in Psychology, 10, 1582. https://doi.org/10.3389/fpsyg.2019.01582

*Costa, A., & Faria, L. (2018). Implicit theories of intelligence and academic achievement: A meta-analytic review. Frontiers in Psychology, 9, 829. https://doi.org/10.3389/fpsyg.2018.00829

*Danişman, Ş., Güler, M., & Karadağ, E. (2019). The effect of teacher characteristics on student achievement: A meta-analysis study. Croatian Journal of Education, 21(4), 1367-1398.

*Dewald, J. F., Meijer, A. M., Oort, F. J., Kerkhof, G. A., & Bögels, S. M. (2010). The influence of sleep quality, sleep duration and sleepiness on school performance in children and adolescents: A meta-analytic review. Sleep medicine reviews, 14(3), 179-189. https://doi.org/10.1016/j.smrv.2009.10.004

*Dent, A. L., & Koenka, A. C. (2016). The relation between self-regulated learning and academic achievement across childhood and adolescence: A meta-analysis. Educational Psychology Review, 28(3), 425-474. https://doi.org/10.1007/s10648-015-9320-8

Dobrota, M., Jeremic, V., & Markovic, A. (2012). A new perspective on the ICT Development Index. Information Development, 28(4), 271-280. https://doi.org/10.1177/0266666912446497

*Ergen, B., & Kanadli, S. (2017). The effect of self-regulated learning strategies on academic achievement: A meta-analysis study. Eurasian Journal of Educational Research, 17(69), 55-74. https://doi.org/10.14689/ejer.2017.69.4

*Erickson, M. (2013). Examining a decade of reading and mathematics student achievement among primary and secondary traditional public school and charter school students: A meta-analytic investigation. Journal of College Teaching and Learning 10(4), 213. https://doi.org/10.19030/tlc.v10i4.8118

*Fan, X., & Chen, M. (2001). Parental involvement and students’ academic achievement: A meta-analysis. Educational psychology review, 13(1), 1-22. https://doi.org/10.1023/A:1009048817385

*Ferguson, C. J. (2015). Do angry birds make for angry children? A meta-analysis of video game influences on children’s and adolescents’ aggression, mental health, prosocial behavior, and academic performance. Perspectives on psychological science, 10(5), 646-666. https://doi.org/10.1177/1745691615592234

Ferragut, M., & Fierro, A. (2012). Inteligencia emocional, bienestar personal y rendimiento académico en preadolescentes. Revista Latinoamericana de Psicología, 44(3), 95–104.

*Ferrero, M., West, G., & Vadillo, M. A. (2017). Is crossed laterality associated with academic achievement and intelligence? A systematic review and meta-analysis. PloS one, 12(8), e0183618. https://doi.org/10.1371/journal.pone.0183618

*Fitzpatrick, D., & Burns, J. (2019). Single-track year-round education for improving academic achievement in US K-12 schools: Results of a meta-analysis. Campbell Systematic Reviews, 15(3), e1053. https://doi.org/10.1002/cl2.1053

*Fry, D., Fang, X., Elliott, S., Casey, T., Zheng, X., Li, J., Florian, L., & 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. https://doi.org/10.1037/edu0000133

*Gardella, J. H., Fisher, B. W., & Teurbe-Tolon, A. R. (2017). A systematic review and meta-analysis of cyber-victimization and educational outcomes for adolescents. Review of Educational Research, 87(2), 283-308. https://doi.org/10.3102/0034654316689136

Gunawardena, S., de Zoysa, P., Jayasinghe, S., Manathunge, A., Alles, H., Shenoy, V., Chamba, T., & de Silva, L. (2017). Selected correlates associated with test anxiety among 14-16 year olds in a Colombo district school. Sri Lanka Journal of Child Health, 46(2), 117-121. https://doi.org/10.4038/sljch.v46i2.8266

Hadders-Algra, M. (2020). Interactive media use and early childhood development. Jornal de Pediatria, 96(3), 273-275. https://doi.org/10.1016/j.jped.2019.05.001

Hattie, J. (2009). Visible learning: A synthesis of 800+ meta-analyses on achievement. Routledge.

Hattie, J. (2017). Visible Learning plus. 250+ Influences on Student Achievement. https://visible-learning.org/wp-content/uploads/2018/03/250-Influences-Final-Effect-Size-List-2017_VLPLUS.pdf

*He, J., Chen, X., Fan, X., Cai, Z., & Huang, F. (2019). Is there a relationship between body mass index and academic achievement? A meta-analysis. Public Health, 167, 111-124. https://doi.org/10.1016/j.puhe.2018.11.002

Heene, M. (2010). A brief history of the fail safe number in applied research. Arxiv, 1–8. http://arxiv.org/ftp/arxiv/papers/1010/1010.2326.pdf

Higgins, S. (2016). Meta-synthesis and comparative meta-analysis of education research findings: some risks and benefits. Review of Education, 4(1), 31-53. https://doi.org/10.1002/rev3.3067

*Huang, C. (2012). Discriminant and criterion-related validity of achievement goals in predicting academic achievement: A meta-analysis. Journal of Educational Psychology, 104(1), 48. https://doi.org/10.1037/a0026223

*Huang, C. (2018). Social network site use and academic achievement: A meta-analysis. Computers & Education, 119, 76-83. https://doi.org/10.1016/j.compedu.2017.12.010

Huedo-Medina, T. B., Sánchez-Meca, J., Marín-Martínez, F., & Botella, J. (2006). Assessing heterogeneity in meta-analysis: Q statistic or I² index?. Psychological methods, 11(2), 193. https://doi.org/10.1037/1082-989X.11.2.193

Iglesias-Díaz, P., & Romero-Pérez, C., (2021). Aulas afectivas e inclusivas y bienestar adolescente: una revisión sistemática. Educación XX1, 24(2), 305-350. https://doi.org/10.5944/educXX1.28705

*Jeynes, W. H. (2005). A meta-analysis of the relation of parental involvement to urban elementary school student academic achievement. Urban Education, 40(3), 237-269. https://doi.org/10.1177/0042085905274540

*Jeynes, W. H. (2007). The relationship between parental involvement and urban secondary school student academic achievement: A meta-analysis. Urban Education, 42(1), 82-110. https://doi.org/10.1177/0042085906293818

*Jeynes, W. (2010). The relationship between Bible literacy and behavioral and academic outcomes in urban areas: A meta-analysis. Education and Urban Society, 42(5), 522-544. https://doi.org/10.1177/0013124510366648

*Jeynes, W. H. (2012). A meta-analysis on the effects and contributions of public, public charter, and religious schools on student outcomes. Peabody Journal of Education, 87(3), 305-335. https://doi.org/10.1080/0161956X.2012.679542

*Jeynes, W. H. (2015). A meta-analysis: The relationship between father involvement and student academic achievement. Urban Education, 50(4), 387-423. https://doi.org/10.1177/0042085914525789

*Karadağ, E., Bektaş, F., Çoğaltay, N., & Yalçın, M. (2015). The effect of educational leadership on students’ achievement: A meta-analysis study. Asia Pacific Education Review, 16(1), 79-93. https://doi.org/10.1007/s12564-015-9357-x

*Kates, A. W., Wu, H., & Coryn, C. L. (2018). The effects of mobile phone use on academic performance: A meta-analysis. Computers & Education, 127, 107-112. https://doi.org/10.1016/j.compedu.2018.08.012

*Kim, K. R., & Seo, E. H. (2015). The relationship between procrastination and academic performance: A meta-analysis. Personality and Individual Differences, 82, 26-33. https://doi.org/10.1016/j.paid.2015.02.038

*Kim, K. R., & Seo, E. H. (2018). The relationship between teacher efficacy and students’ academic achievement: A meta-analysis. Social Behavior and Personality: an international journal, 46(4), 529-540. https://doi.org/10.2224/sbp.6554

*Kim, S. W., Cho, H., & Kim, L. Y. (2019). Socioeconomic status and academic outcomes in developing countries: a meta-analysis. Review of Educational Research, 89(6), 875-916. https://doi.org/10.3102/0034654319877155

*Kim, S. W. & Hill, N. E. (2015). Including fathers in the picture: A meta-analysis of parental involvement and students’ academic achievement. Journal of Educational Psychology, 107(4), 919. https://doi.org/10.1037/edu0000023

Kumar, R., & Lal, R. (2014). Study of academic achievement in relation to family environment among adolescents. The International Journal of Indian Psychology, 2(1), 146-155. https://doi.org/10.25215/0201.074

*Kyriakides, L., Christoforou, C., & Charalambous, C. Y. (2013). What matters for student learning outcomes: A meta-analysis of studies exploring factors of effective teaching. Teaching and Teacher Education, 36, 143-152. https://doi.org/10.1016/j.tate.2013.07.010

*Lam, K. K. L., & Zhou, M. (2019). Examining the relationship between grit and academic achievement within K-12 and higher education: A systematic review. Psychology in the Schools, 56(10), 1654-1686. https://doi.org/10.1002/pits.22302

*Lei, H., Cui, Y., & Zhou, W. (2018). Relationships between student engagement and academic achievement: A meta-analysis. Social Behavior and Personality: an international journal, 46(3), 517-528. https://doi.org/10.2224/sbp.7054

*Li, J., Ye, H., Tang, Y., Zhou, Z., & Hu, X. (2018). What are the effects of self-regulation phases and strategies for Chinese students? A meta-analysis of two decades research of the association between self-regulation and academic performance. Frontiers in Psychology, 9, 2434. https://doi.org/10.3389/fpsyg.2018.02434

*Liebowitz, D. D., & Porter, L. (2019). The effect of principal behaviors on student, teacher, and school outcomes: A systematic review and meta-analysis of the empirical literature. Review of Educational Research, 89(5), 785-827. https://doi.org/10.3102/0034654319866133

*Liu, D., Kirschner, P. A., & Karpinski, A. C. (2017). A meta-analysis of the relationship of academic performance and Social Network Site use among adolescents and young adults. Computers in Human Behavior, 77, 148-157. https://doi.org/10.1016/j.chb.2017.08.039

*Liu, J., Peng, P., & Luo, L. (2020). The relation between family socioeconomic status and academic achievement in China: a meta-analysis. Educational Psychology Review, 32(1), 49-76. https://doi.org/10.1007/s10648-019-09494-0

*Lomos, C., Hofman, R. H., & Bosker, R. J. (2011). Professional communities and student achievement–a meta-analysis. School Effectiveness and School Improvement, 22(2), 121-148. https://doi.org/10.1080/09243453.2010.550467

*MacCann, C., Jiang, Y., Brown, L. E., Double, K. S., Bucich, M., & Minbashian, A. (2020). Emotional intelligence predicts academic performance: A meta-analysis. Psychological Bulletin, 146(2), 150. https://doi.org/10.1037/bul0000219

*Madigan, D. J. (2019). A meta-analysis of perfectionism and academic achievement. Educational Psychology Review, 31(4), 967-989. https://doi.org/10.1007/s10648-019-09484-2

*Marker, C., Gnambs, T., & Appel, M. (2018). Active on Facebook and failing at school? Meta-analytic findings on the relationship between online social networking activities and academic achievement. Educational Psychology Review, 30(7), 651-677. https://doi.org/10.1007/s10648-017-9430-6

McCoach, D. B., Yu, H., Gottfried, A. W., & Gottfried, A. E. (2017). Developing talents: A longitudinal examination of intellectual ability and academic achievement. High Ability Studies, 28(1), 7-28. https://doi.org/10.1080/13598139.2017.1298996

*McGuire, A., & Jackson, Y. (2018). A multilevel meta-analysis on academic achievement among maltreated youth. Clinical child and family psychology review, 21(4), 450-465. https://doi.org/10.1007/s10567-018-0265-6

*Miller-Cotto, D., & Byrnes, J. P. (2016). Ethnic/racial identity and academic achievement: A meta-analytic review. Developmental Review, 41, 51-70. https://doi.org/10.1016/j.dr.2016.06.003

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2010). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. International Journal of Surgery, 8(5), 336-341. https://doi.org/10.1016/j.ijsu.2010.02.007

Mozammel, S., Ahmed, U., & Shakar, N. (2021). COVID-19 and online learning: critical insights for academic achievement. Elementary Education Online, 20(4), 1452-1457.

*Murray, J., Farrington, D. P., & Sekol, I. (2012). Children’s antisocial behavior, mental health, drug use, and educational performance after parental incarceration: a systematic review and meta-analysis. Psychological Bulletin, 138(2), 175. https://doi.org/10.1037/a0026407

*Nakamoto, J., & Schwartz, D. (2010). Is peer victimization associated with academic achievement? A meta-analytic review. Social Development, 19(2), 221-242. https://doi.org/10.1111/j.1467-9507.2009.00539.x

National Academies of Sciences, Engineering, and Medicine. (2019). Reproducibility and replicability in science. National Academies Press.

Nisar, N., & Mahmood, K. (2017). Determinants of students’ academic achievement at secondary school level. Bulletin of Education and Research, 39(1), 145-158.

*Noltemeyer, A. L., Ward, R. M., & Mcloughlin, C. (2015). Relationship between school suspension and student outcomes: A meta-analysis. School Psychology Review, 44(2), 224-240. https://doi.org/10.17105/spr-14-0008.1

*Ohtani, K., & Hisasaka, T. (2018). Beyond intelligence: a meta-analytic review of the relationship among metacognition, intelligence, and academic performance. Metacognition and Learning, 13(2), 179-212. https://doi.org/10.1007/s11409-018-9183-8

Olmos Rueda, P., & Mas Torelló, O. (2013). Youth, academic failure and second chance training programmes. Revista Española de Orientación y Psicopedagogía, 24(1), 78-92. https://doi.org/10.5944/reop.vol.24.num.1.2013.11272

*Ouma, C., & Nam, J. (2015). A meta-analysis of gender gap in student achievement in African countries. International Review of Public Administration, 20(1), 70-83. https://doi.org/10.1080/12294659.2014.967372

*Pahlke, E., Hyde, J. S., & Allison, C. M. (2014). The effects of single-sex compared with coeducational schooling on students’ performance and attitudes: A meta-analysis. Psychological Bulletin, 140(4), 1042. https://doi.org/10.1037/a0035740

*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

*Phansikar, M., Ashrafi, S. A., Khan, N. A., Massey, W. V., & Mullen, S. P. (2019). Active commute in relation to cognition and academic achievement in children and adolescents: A systematic review and future recommendations. International Journal of Environmental Research and Public Health, 16(24), 5103. https://doi.org/10.3390/ijerph16245103

*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

*Pinquart, M., & Ebeling, M. (2020). Parental educational expectations and academic achievement in children and adolescents—a meta-analysis. Educational Pshychology Review, 32, 463-480. https://doi.org/10.1007/s10648-019-09506-z

*Poropat, A. E. (2009). A meta-analysis of the five-factor model of personality and academic performance. Psychological Bulletin, 135(2), 322. https://doi.org/10.1037/a0014996

*Poropat, A. E. (2014). A meta-analysis of adult-rated child personality and academic performance in primary education. British Journal of Educational Psychology, 84(2), 239-252. https://doi.org/10.1111/bjep.12019

*Preckel, F., Lipnevich, A. A., Schneider, S., & Roberts, R. D. (2011). Chronotype, cognitive abilities, and academic achievement: A meta-analytic investigation. Learning and Individual Differences, 21(5), 483-492. https://doi.org/10.1016/j.lindif.2011.07.003

*Purdie, N., & Hattie, J. (1999). The relationship between study skills and learning outcomes: A meta-analysis. Australian Journal of Education, 43(1), 72-86. https://doi.org/10.1177/000494419904300106

*Rebelo, M. A. B., Rebelo Vieira, J. M., Pereira, J. V., Quadros, L. N., & Vettore, M. V. (2019). Does oral health influence school performance and school attendance? A systematic review and meta-analysis. International Journal of Paediatric Dentistry, 29(2), 138-148. https://doi.org/10.1111/ipd.12441 Robinson, J., & Biran, M. (2006). Discovering self: Relationships between African identity and academic achievement. Journal of Black Studies, 37(1), 46-68. https://doi.org/10.1177/0021934704273149

*Ruff, R. R., Senthi, S., Susser, S. R., & Tsutsui, A. (2019). Oral health, academic performance, and school absenteeism in children and adolescents: A systematic review and meta-analysis. The Journal of the American Dental Association, 150(2), 111-121. https://doi.org/10.1016/j.adaj.2018.09.023

*Ruiz-Hermosa, A., Álvarez-Bueno, C., Cavero-Redondo, I., Martínez-Vizcaíno, V., Redondo-Tébar, A., & Sánchez-López, M. (2019). Active commuting to and from school, cognitive performance, and academic achievement in children and adolescents: A systematic review and meta-analysis of observational studies. International Journal of Environmental Research and Public Health, 16(10), 1839. https://doi.org/10.3390/ijerph16101839

Russo MW. (2006) How to review a metaanalysis. Gastroenterol Hepatol, 3(8), 637–642.

Sagaró, N. M., & Zamora, L. (2019). Evolución histórica de las técnicas estadísticas y las metodologías para el estudio de la causalidad en ciencias médicas. Medisan, 23(3), 534-556.

*Shin, I. S., & Chung, J. Y. (2009). Class size and student achievement in the United States: A meta-analysis. KEDI Journal of Educational Policy, 6(2), 3-19.

*Shulruf, B. (2010). Do extra-curricular activities in schools improve educational outcomes? A critical review and meta-analysis of the literature. International Review of Education, 56(5), 591-612. https://doi.org/10.1007/s11159-010-9180-x

*Shulruf, B., Keuskamp, D., & Brake, D. (2010). The impact of course-taking on academic achievements a systematic review and Meta analysis. Procedia-Social and Behavioral Sciences, 2(2), 3401-3406. https://doi.org/10.1016/j.sbspro.2010.03.523

Sipe, T. A., & Curlette, W. L. (1997). 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

*Sirin, S. R. (2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75(3), 417-453. https://doi.org/10.3102/00346543075003417

*Sisk, V. F., Burgoyne, A. P., Sun, J., Butler, J. L., & Macnamara, B. N. (2018). To what extent and under which circumstances are growth mind-sets important to academic achievement? Two meta-analyses. Psychological Science, 29(4), 549-571. https://doi.org/10.1177/0956797617739704

Smedsrud, J., Nordahl-Hansen, A., Idsøe, E. M., Ulvund, S. E., Idsøe, T., & Lang-Ree, O. C. (2019). The associations between math achievement and perceived relationships in school among high intelligent versus average adolescents. Scandinavian Journal of Educational Research, 63(7), 1041-1055. https://doi.org/10.1080/00313831.2018.1476406

Spina, G., Bozzola, E., Ferrara, P., Zamperini, N., Marino, F., Caruso, C., Antilici, L, & Villani, A. (2021). Children and adolescent’s perception of media device use consequences. International Journal of Environmental Research and Public Health, 18(6), 3048. https://doi.org/10.3390/ijerph18063048

Stevenson, M. N. (2021). Homework and Academic Achievement: A Meta-Analysis Examining Impact [Tesis Doctoral]. Universidad de Dayton.

*Szumski, G., Smogorzewska, J., & Karwowski, M. (2017). Academic achievement of students without special educational needs in inclusive classrooms: A meta-analysis. Educational Research Review, 21, 33-54. https://doi.org/10.1016/j.edurev.2017.02.004 *Tan, C. Y. (2017). Examining cultural capital and student achievement: Results of a meta-analytic review. Alberta Journal of Educational Research, 63(2), 139-159 https://doi.org/10.1016/j.edurev.2019.100289

*Tan, C. Y., Peng, B., & Lyu, M. (2019). What types of cultural capital benefit students’ academic achievement at different educational stages? Interrogating the meta-analytic evidence. Educational Research Review, 28, 100289.

*Tonetti, L., Natale, V., & Randler, C. (2015). Association between circadian preference and academic achievement: A systematic review and meta-analysis. Chronobiology international, 32(6), 792-801. https://doi.org/10.3109/07420528.2015.1049271

Tufanaru, C., Munn, Z., Stephenson, M., & Aromataris, E. (2015). Fixed or random effects meta-analysis? Common methodological issues in systematic reviews of effectiveness. International journal of evidence-based healthcare, 13(3), 196-207. https://doi.org/10.1097/XEB.0000000000000065

*Tze, V. M., Daniels, L. M., & Klassen, R. M. (2016). Evaluating the relationship between boredom and academic outcomes: A meta-analysis. Educational Psychology Review, 28(1), 119-144. https://doi.org/10.1007/s10648-015-9301-y

*Uysal, S., & Sarier, Y. (2018). Meta-Analysis of school leadership effects on student achievement in USA and Turkey. Cypriot Journal of Educational Sciences, 13(4), 590-603. https://doi.org/10.18844/cjes.v13i4.3539

*Uysal, S., & Sarier, Y. (2019). Teacher leadership effects on student achievement and student satisfaction: A Meta-analysis of the studies published in Turkey and the USA. Croatian Journal of Education: Hrvatski časopis za odgoj i obrazovanje, 21(3), 989-1010. https://doi.org/10.15516/cje.v21i3.3257

*Vasquez, A. C., Patall, E. A., Fong, C. J., Corrigan, A. S., & Pine, L. (2016). Parent autonomy support, academic achievement, and psychosocial functioning: A meta-analysis of research. Educational Psychology Review, 28(3), 605-644. https://doi.org/10.1007/s10648-015-9329-z

Vera Sagredo, A., Cerda Etchepare, G., Aragón Mendizábal, E., & Pérez Wilson, C., (2021). Rendimiento académico y su relación con variables socioemocionales en estudiantes chilenos de contextos vulnerables. Educación XX1, 24(2), 375-398. https://doi.org/10.5944/educXX1.28269

*Voyer, D., & Voyer, S. D. (2014). Gender differences in scholastic achievement: a meta-analysis. Psychological Bulletin, 140(4), 1174. https://doi.org/10.1037/a0036620

York, T. T., Gibson, C., & Rankin, S. (2015). Defining and measuring academic success. Practical Assessment, Research, and Evaluation, 20(1), 5.

*Zaboski II, B. A., Kranzler, J. H., & Gage, N. A. (2018). Meta-analysis of the relationship between academic achievement and broad abilities of the Cattell-Horn-Carroll theory. Journal of School Psychology, 71, 42-56. https://doi.org/10.1016/j.jsp.2018.10.001

Contact address: Belén Gutiérrez-de-Rozas, Universidad Nacional de Educación a Distancia. Facultad de Educación. Departamento de Métodos de Investigación y Diagnóstico en Educación II. Calle Juan del Rosal, 14, C.P. 28040 Madrid, España. E-mail: bgutierrezderozas@edu.uned.es

1 This research has been conducted under the support of the Ayudas para la Formación de Profesorado Universitario (FPU).

2 According to the needs derived from the variables identified in our study, 3 new subcategories and 14 indicators were added to the categories proposed by Hattie (2009). Thus, while Hattie established 22 subcategories and 66 indicators, the variables in our study have been classified according to the categories identified in Table II.

3 Sipe and Curlette (1997) did not provide results for all the categories established in this meta-synthesis.

4 In Hattie’s (2018) study, teacher quality was measured through student perception.