Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13
Topics
Discussion
Announcements
 

Compensating, Mediating, and Moderating Effects of School Climate on Academic Achievement Gaps in Israel


by Ruth Berkowitz , Hagit Glickman , Rami Benbenishty, Elisheva Ben-Artzi , Tal Raz, Nurit Lipshtat & Ron Avi Astor - 2015

Background: It is widely agreed among educational researchers and practitioners that schools with positive climates can effectively mitigate the influence of students’ and schools’ socioeconomic status (SES) on academic achievement. Nevertheless, the exact mechanisms by which this occurs are unclear.

Objective: This study aimed to fill that gap, examining student perceptions of school climate, student academic achievement, and student and school SES in Israel to develop a reliable and comprehensive assessment of the role of school climate in the relationship between student and school SES and achievement. Specifically, the study tested whether school climate has an additive contribution to academics beyond students’ and schools’ SES (compensation model), whether the school’s SES influences its social climate, which in turn influences academic achievement (mediation model); or whether the relationship between SES and academics changes across schools with different climates (moderation model).

Research Design: Secondary analysis of a large-scale, nationally representative sample of fifth- and eighth-grade Hebrew-speaking students in public schools in Israel (N = 53,946).

Data Analysis: Hierarchical linear modeling (HLM) was used to examine models with variables both on the student and the school levels. Linear regressions were used to examine student level and school level only models.

Results: School climate had an additive compensation contribution to academic achievements, both on the student and the school levels. School climate moderated the relationship between students’ SES and academic achievements. However, findings did not support the hypothesis that school climate mediated the relationship between SES background and academic achievement, both at the student and school levels.

Conclusions: School climate plays an important role in accounting for achievements, beyond students’ and schools’ SES. Results highlight the need to improve school climate, especially in schools serving communities of low SES, to enhance social mobility and equality of opportunity.



Achievement gaps among racial and ethnic groups, as well as between students of different socioeconomic status (SES), have been documented and studied extensively in countries such as Canada (Caldas, Bernier, & Marceau, 2009), South Africa (Van der Berg, 2008), the United States (Grigg, Donahue, & Dion, 2007), and Israel (Zussman & Tzur, 2008). Achievement gaps have been often explained in terms of disparities in opportunity for educational success among students of different backgrounds and ethnicities (Akiba, LeTendre, & Scribner, 2007; Barton, 2003; Clotfelter, Ladd, & Vigdor, 2005). Researchers from countries striving to improve the achievement gap through high standards and school climate interventions could learn from each other.


Many studies found evidence of a strong relationship between SES and academic achievement (e.g., Jencks et al., 1972; Lee, 2002). In his famous report “Equality of Educational Opportunity,” Coleman (1966) linked student background with academic performance and argued schools cannot do much to improve academic outcomes. Later, scholars who studied schools as organizations pointed out numerous ways in which schools can effectively influence students and schools of lower SES to increase academic proficiency (Ladd & Walsh, 2002). An important, often-discussed central factor in a school’s positive influence on low-SES student academic outcomes is school climate (Brand, Felner, Shim, Seitsinger, & Dumas, 2003; McEvoy & Welker, 2000).


School climate refers to patterns of academic experiences and reflects a school’s norms, goals, and values; interpersonal relationships; teaching, learning, and leadership practices; and organizational structures. A sustainable, positive school climate fosters youth development and learning necessary for a productive, contributing, and satisfying life in a democratic society (National School Climate Council, 2007).


Numerous scholars and policies have been built on the assumption that school climate has an impact on academic outcomes, especially in high-poverty schools (e.g., Berliner, 2008; Bryk & Schneider, 2002; Ladd & Dinella, 2009; Noguera, 2010; Sherblom, Marshall, & Sherblom, 2006). Multiple countries and educational entities are now engaged in policy debates on school climate and social and emotional variables, and how they affect academics (Beller, 2013; Bryk, Sebring, Allensworth, Easton, & Luppescu, 2010). Studies show that a positive school climate, as manifested by teachers’ expectations (Weinstein, Gregory, & Strambler, 2004) and support of students (Ma, Phelps, Lerner, & Lerner, 2009), lack of school violence and students’ disruptive behavior (Robers, Zhang, Truman, & Snyder, 2012), safety (Osher, Dwyer, Jimerson, & Brown, 2012), and student engagement in school (Powers, Bowen, & Rose, 2005), contributes to student proficiency and promotes achievement beyond expected outcomes based on student and school SES. Even so, many of these studies tend to be small scale, not representative, and do not have an array of climate and academic variables.


Further, despite the wide interest and growing empirical evidence indicating the academic benefits garnered by all students who attend schools with positive climates, little is known about the nature of such relationships, as the scientific literature provides diverse, often confusing, descriptions of the mechanism by which positive school climate contributes to student achievement.


Some authors, who have recognized the critical influence of students’ and schools’ SES background on academic achievements, argued that positive school climate has an additive contribution to children’s achievements above the negative influence of low SES on student proficiency (e.g., Brand et al., 2003; McEvoy & Welker, 2000; Schagen & Hutchison, 2003). These authors suggested an explanation describing school climate as compensating for low SES, adding value independently and thus contributing to academic achievements beyond the expected outcomes based on SES background. Theoretically, this explanation suggests that school does matter as it has an effect on students’ achievements. Practically, these compensatory relationships stress the importance of investing resources to promote a positive school climate among all schools, but especially in those serving communities living in poverty, as it could improve students’ proficiency.


Others have shown that a school’s SES influences its social climate, which in turn influences academic achievement (Sebring, Allensworth, Bryk, Easton, & Luppescu, 2006). In some communities, stress linked to poverty, crime, and other social problems makes it more challenging to operate schools in a way that enhances student proficiency (Sebring et al., 2006). Further, previous studies pointed to the relationship between community poverty and school violence (Gottfredson, Gottfredson, Payne, & Gottfredson, 2005; Khoury-Kassabri, Benbenishty, Astor, & Zeira, 2004; Stewart, 2003) as well as between school violence and lower academic competence or self-perceived school performance (Ma et al., 2009; Nansel et al., 2001; Rutter & Maughan, 2002). This explanation describes school climate as mediating the relationship between SES background and academic achievements through factors such as students’ connectedness and engagement with school (Klem & Connell, 2004; McNeely, Nonnemaker, & Blum, 2002; Wang & Holcombe, 2010), students’ feeling of safety (Benbenishty & Astor, 2005), and school violence (Benbenishty, Khoury-Kassabri, Roziner, & Astor, 2005; Dwyer, Osher, & Hoffman, 2000). The underlying premise of the mediation explanation is a significant correlation between SES background and school climate. Practically, a confirmation of this explanation could imply the determinate nature of the relationship between low SES and negative school climate, and thus the unworthiness of any efforts to improve climate in schools of low SES.


Finally, other researchers have shown that the relationship between SES background and academic achievements changes across schools of different climate; in a school with a positive climate, the relationship between SES background and achievement would be weaker compared to schools with a less positive climate (e.g., Sebring et al., 2006). In other words, this explanation suggests that the relationship between student and school SES and academic achievement can be moderated by school climate. Practically, a confirmation of the moderation model would stress the importance of promoting positive school climate, especially in schools serving communities living in poverty, where a great potential for significant improvements in academic achievements exists.


In order to be able to utilize and manipulate school climate to enhance achievements, it is essential to understand how these factors interrelate. Therefore, the current study was undertaken to develop a more reliable and comprehensive assessment of the role of school climate in the relationship between student and school SES and achievement. Using a nationally representative database that includes both school climate and academic variables, this study examined the role and effects of school climate as it pertains to academics. Specifically, the study examined three ways in which school climate and student and school SES might explain students’ and schools’ differences in academic achievement: compensation, mediation, and moderation. The multivariate and nested sample allowed for a more thorough exploration of these long-debated issues. These findings could contribute to theoretical and policy discussions on the role of climate for improving academic outcomes.


Prior research indicates differences between young and older students in terms of school experiences. Younger students share more caring, intimate, and trustful relationships with their teachers, compared with student in higher grades (Author & Author, 2005; Brand et al., 2003), partly because lower-grade students remain with their homeroom teacher throughout most of the day, while older students change classes and are exposed to different teachers (Pianta & Allen, 2008). Caring relationships with teachers enhance engagement at school and equip students with the resources and social support they need to be academically successful (Osher et al., 2012). Further, studies suggest different patterns of violence involvement for students of different ages. An inverse relationship between age and bullying involvement was suggested (Olweus, Limber, & Mihalic, 1999; Smith, Madsen, & Moody, 1999) with lower bullying victimization in high schools than in middle or elementary schools (Sullivan, 2000). Further, inconsistent patterns have been suggested with regard to students’ grade level, and more severe forms of violence such as gang-related activities (e.g. Huff, 2002). Therefore, all models were examined separately for fifth- and eighth-grade students.   


The study extends prior ecological nested empirical work (Benbenishty & Astor, 2005; Chen & Astor, 2011a, 2011b; Chen & Astor, 2012; Marachi, Astor, & Benbenishty, 2007) by examining how school climate influences the relationship between student and school SES and achievement. Because students who attend the same school are exposed to the same surroundings and experiences, influence one another, and are more likely to share the same organizational and structural experiences (such as teaching methods, teachers, and school climate), we used a multilevel approach that accounted for micro (within-school, or student-level) and macro (between-school, or school-level) factors. Research questions included:



(1) Does school climate compensate for the influence of student background on achievement?

(2) Does school climate moderate the relationship between student SES and achievement, such that the relationship is more pronounced for schools with a less positive climate?

(3) Does school climate mediate the relationship between student SES and achievement?


All models were examined at both student and school levels.


THE ISRAELI EDUCATION SYSTEM  


The Israeli education system reflects the different populations it serves. Arabs and Jews are highly segregated in schools. The Hebrew school system consists of three main sectors: nonreligious public schools, which are attended by about 70% of Jewish students; religious public schools, attended by about 20% of Jewish students; and ultra-Orthodox “independent schools.” Ultra-Orthodox schools were not included in the current study because they do not take the Meitzav tests and surveys. In Israel, there are significant achievement gaps between students of high socio-economic background and others (Zussman & Tzur, 2008).


METHOD


GENERAL STUDY DESIGN


In the present study, datasets from two consecutive years (2008 and 2009) were used to create a census of half of all fifth- and eighth-grade students who attended schools in the official public school system supervised by the Israeli Ministry of Education (half the country is collected each year). In Israel, the grade level system is not homogenous across schools: some “primary” schools are comprised of first through eighth grades and others first through sixth grades only, while seventh through 12th grades may belong to either one school or split into middle and high. Therefore, in order to avoid confusions, the present study refers to grades instead of school level.


Due to multiple cultural and structural differences between Hebrew- and Arabic-language schools, this study focuses only on the larger group of Hebrew-language schools. The total sample consisted of 53,946 students who attended 902 Hebrew-language schools (response rate ranged between 88% and 92%).


MEASUREMENTS


The National Authority for Measurement and Evaluation in Education (RAMA) runs a large-scale national education monitoring system, Growth and Efficiency Measures of Schools (in Hebrew, Meitzav), for fifth- and eighth-grade students. This unique system provides information on both students’ educational achievement and numerous variables measuring perceptions of school climate. All schools in Israel were stratified by the Ministry of Education into four equal groups (hereafter, clusters), each cluster designed to represent all schools in Israel. According to the Meitzav regulation, every year all fifth- and eighth-grade students in two clusters are given achievement tests and school climate questionnaires. One cluster is tested in mathematics and native language (Hebrew or Arabic), and the other cluster is tested in science and English as a foreign language. It should be noted that in this system findings on these grade levels are assumed to be representative of the school as a whole. This assumption has not been tested empirically.


The Meitzav includes two main sub-scales: achievements and school climate.


The Meitzav Achievement Tests


The Meitzav achievement tests examine the extent to which students gained required knowledge in four subjects. The current study used only mathematics test scores because they seemed to most reliably represent knowledge acquired at school and are less influenced by knowledge acquired at home. Mathematics test scores are measured on a scale with a mean of 500 and a standard deviation of 100 (for further information, see the National Authority for Measurement and Evaluation in Education, n.d.).


The Meitzav School Climate


The Meitzav school climate sub-scales include 54 items measuring different aspects of schools’ social climate. Some of the items were found by the Ministry of Education research team as having low reliability in measuring school climate (e.g., students’ overall satisfaction with the school); other items were only asked on the 2008 school climate questionnaire and thus were missing information on most of the sample. Therefore, in the present study we omitted 17 items. Because the item pool was different than the original pool, the remaining 37 items were subjected to an exploratory principal components factor analysis with Varimax factor rotation yielding a nine-factor solution (eigenvalue > 1), explaining 59.49 % of the total variance. A second-order principal components factor analysis with Varimax rotation was conducted on the nine factor scores (factor scores were computed by averaging the standardized items scores for each factor). The second-order analysis yielded three factors (eigenvalue > 1), explaining 59.56% of the total variance. The first factor consisted of items referring to student–teacher relationships; the second factor consisted of items referring to risky behavior; and the third factor consisted of items referring to school violence.  


Positive student–teacher relationship (α = .875). This factor included 19 items referring to seven issues: teacher feedback to promote student proficiency, teacher belief in student ability to succeed, close and caring relationships between students and teachers, fair relationships between students and teachers, high expectations and academic pressure for students, and school efforts to prevent violence. Higher scores on the composite measure indicate a more positive students-teacher relationship.  


Risky peer behavior (α = .474). This factor included six items referring to three issues: proper student behavior in the classroom, vandalism at school, and gangs at school. Higher scores on the composite measure indicate more risky peer behavior.


School violence (α = .600). This factor included 12 items referring to three issues: student victimization via direct forms of violence, student victimization via indirect forms of violence, and a perceived lack of protection at school. Higher scores on the composite measure indicate more school violence.


SES Background


SES was determined according to the Social Deprivation Index (SDI), computed by the Ministry of Education. SES describes the relative personal SES of the students’ families based on parental education, periphery neighborhood of residence, income, home country, and immigration from poor countries. The current study used this measure on both the student and school levels.


School-level SES background represents the mean SES percentiles of all students in a school. The values range from 1 to 10; higher values indicate better SES backgrounds.     


Personal student-level SES background was computed by subtracting the school measure from the student measure. Therefore, the personal SES measure represents the student SES relative to other students in his school.


Religious Educational Stream


Hebrew-speaking public schools belong to either the religious or nonreligious educational stream. In addition to differences in curricula and structure (many religious schools are gender specific), these schools tend to differ in students’ SES (students in the religious public tend to come from weaker SES backgrounds (Shavit & Blank, 2012). To take this into account, we controlled for the schools’ religious educational stream.



Table 1. Descriptive Statistics for the Study Variables (N = 59,946)

 

5th grade (n=27,878)

8th grade (n=25,923)

 

Variables

M

SD

M

SD

Range

Student SES*

.02

1.77

.16

1.67

-9 - 9

Mathematics test scores

513.06

98.14

511.26

102.25

 

School SES*

4.62

2.18

4.41

2.12

1-10

Positive student–teacher relationship

     

Teachers make sure students understand them when they teach

4.19

1.02

3.19

1.19

a1-5

Teachers believe in my ability to succeed

4.48

.85

3.95

1.03

a1-5

Teachers let me believe that I can do well at school

4.29

.99

3.67

1.10

a1-5

I have close and good relationships with most of the teachers

3.92

1.07

3.36

1.11

a1-5

When I am sad I feel comfortable to speak to one of my teachers

3.19

1.40

2.25

1.23

a1-5

Most teachers care about me, and not only with regards to school work

3.79

1.20

3.13

1.20

a1-5

Most teachers care about how I feel in general and at school

3.74

1.17

2.93

1.17

a1-5

My school makes great efforts to prevent and solve the problem of violence

4.11

1.14

3.36

1.11

a1-5

During breaks there are always teachers who make sure that there is no violence at the school premises

4.49

.94

2.25

1.23

a1-5

When there is violence at school, teachers know about it

4.09

1.00

3.13

1.20

a1-5

In my school teachers treat students decently

4.04

1.12

3.31

1.13

a1-5

There are no students in my class who are more favorable on the teachers*

3.54

1.45

2.68

1.34

a1-5

There are no students in my class, that teachers always treat them badly, regardless of what they do*

4.01

1.30

3.14

1.36

a1-5

Most teachers in my school treat students with respect

4.26

.97

3.64

1.07

a1-5

Most teachers respect my thoughts and feelings

3.88

1.10

3.21

1.10

a1-5

There are no teachers in my school who disrespect students opinion*

3.99

1.24

3.17

1.27

a1-5

Most teachers expect their students to make efforts to do well at school

4.47

.82

4.07

.93

a1-5

Most teachers expect their students to try and improve their grades

4.44

.84

4.08

.91

a1-5

Most teachers expect their students to invest time and effort at their school work

4.41

.83

3.90

.95

a1-5

Risky peer behavior

     

In my class students treat teachers disrespectfully

3.54

.94

3.10

.94

a1-5

Very often students are noisy in the classroom and interrupt the class*

2.65

1.25

2.34

1.09

a1-5

In my class students talk back at the teachers*

2.79

1.33

2.35

1.14

a1-5

Teachers have to wait a long time before students are quiet at the beginning of lessons*

2.93

1.26

2.62

1.12

a1-5

During the last month children broke and ruined school property*

2.57

.62

2.26

.72

b1-3

There are gangs of students in my school who bully, pick and hurt other kids*

3.10

1.30

3.18

1.36

a1-5

School violence

     

During the last month I got kicked, hit or punched by a student who wished to hurt me

1.34

.58

1.20

.47

b1-3

During the last month a student used a stone, stick, chair or other object to hurt me

1.08

.31

1.06

.26

b1-3

During the last month a student beat me hard

1.14

.40

1.06

.28

b1-3

During the last month a student threatened to hurt me at school or after school

1.17

.45

1.08

.31

b1-3

During the last month a student blackmailed me for money, food or other valuables.  

1.02

.18

1.02

.17

b1-3

During the last month students tried to persuade other kids not to talk to me and not be friends with me

1.26

.54

1.12

.38

b1-3

During the last month a student spread rumors about me to hurt me

1.24

.52

1.19

.46

b1-3

During the last month I was boycotted, a group of students did not want to play or speak with me

1.11

.36

1.04

.22

b1-3

Sometimes I am afraid to go to school because of violence

1.67

1.12

1.43

.84

a1-5

I feel safe and protected at school

4.13

1.08

3.73

1.10

a1-5

There are places at school in which I am afraid to get

1.56

1.09

1.70

1.09

a1-5

Sometimes I prefer to stay in the classroom during the break because I am afraid of being hurt

1.42

.95

1.30

.74

a1-5

* Original item was reversed

a measured on a scale :1= strongly disagree; 2= disagree; 3= agree a little; 4= agree; 5=strongly agree

b measured on a scale: 1= never; 2= once or twice; 3= three times or more





Table 2. Correlations for Study Variables (N = 59,946)

5th grade (n=27,878)

 


Mathematics test scores

Student SES

Religious educational stream

school SES

Positive student–teacher relationship

Risky peer behavior

School violence

Mathematics test scores

--

0.21***

-.09***

0.18***

0.03***

-0.07***

-0.15***

Student SES

0.21***

--

-.03***

0.02***

-0.02***

0.00

0.02***

Religious educational stream

-0.09***

-.03***

--

-0.17***

-0.08***

-0.02**

0.02***

School SES

0.18***

0.02***

-.17***

--

0.09***

-0.01*

0.08

Positive student–teacher relationship

0.03***

-0.02***

-.08***

0.09***

--

-0.36***

-0.23***

Risky peer behavior

-0.07***

0.00

-.02**

-0.01*

-0.36

--

0.37***

School violence

-0.15***

0.02***

0.02***

0.08

-0.23***

0.37***

--

8th grade (n=25,923)

Mathematics test scores

--

.23***

-.03***

.25***

-.08***

-.01

-.01

Student SES

.23***

--

.00

.04***

-.01*

.03***

.03***

Religious educational stream

-.03***

.00

--

-.11***

.02**

-.08***

-.05***

School SES

.25***

.04***

-.11***

--

-.04***

-.02***

.02**

Positive student–teacher relationship

-.08***

-.01*

.02**

-.04***

--

-.30***

-.17***

Risky peer behavior

-.01

.03***

-.08***

-.02***

-.30***

--

.31***

School violence

-0.01

.03***

-.05***

.02**

-.17***

.31***

--

Note: SES = socioeconomic status; Religious educational stream measured as 0 = secular, 1 = religious

*p < .05, **p < .01, ***p < .001


DATA ANALYSIS


We used hierarchical linear modeling (HLM), an analytic technique used to study data organized hierarchically in multiple levels, such as students nested within schools. HLM analysis allows the partitioning of variance and covariance components among levels (Raudenbush & Bryk, 2002). In the current study, the first level variables included student SES, student perceptions of school climate, and student achievement. The second level included school characteristics: school SES and religious educational stream (secular or religious). Three hierarchical linear models predicting mathematics test scores were tested: Model 1 included school SES, religious educational stream, and student SES. Model 2 included the three school climate factors, and Model 3 included interaction factors between school climate factors and student SES. Finally, linear regressions were used to examine school level only models. Analyses were conducted using HLM (7.0) and SPSS (17.0) statistical programs.


The cross-sectional nature of the data should be noted. Because student test scores and school climate variables were collected at one point in time, other variables that might explain the patterns and relationships in the current study should be considered.


RESULTS


SCHOOL VARIANCE OF TEST SCORES


To test whether HLM analysis is justified for these data, the first step of our analysis involved developing a fully unconditional, two-level model that provided useful preliminary information regarding between- and within-school variance in student test scores. A considerable percentage of the variance in student scores was between-school variance (16.64% and 20.10% for fifth- and eighth-grade, respectively) (see Table 3). Thus, the between-school variance in student scores was large enough to warrant HLM analyses, which involved separate student-level and school-level analyses.


Table 3. Fully Unconditional HLM for Partitioning Variance in Mathematics Test Scores

 

Grade

Between-School Variance

Within-School Variance

ICCa

χ2

df

5

1596.37*

8259.47

16.64%

5280.02

576

8

2197.77*

8705.74

20.10%

5602.34

315

Note. HLM = hierarchical linear modeling; ICC = intraclass correlation.

aICC calculated by dividing between-school variance by sum of between-school and within-school variance

*p < .001


SES AND TEST SCORES: A SCHOOL CLIMATE COMPENSATION HYPOTHESIS


The hypothesis that positive school climate would have a significant additive contribution to student achievement beyond the influence of student and school SES was tested on two levels.  


Student Level


These models tested whether and to what extent students who reported higher positive school climate also achieved higher test scores, beyond student and school SES levels. Model 1 included school SES, religious educational stream, and student SES. Model 2 included the three school climate factors (Table 4). The intercept represented the average student achievement, and the slope represented the strength of the relationship between background variables and achievement—the steeper the slope (reflected by B coefficients), the stronger the relationship. Findings for student- and school-level compensation models are also presented in Figure 1.


Figure 1. Explained variance in students’ and schools’ mathematics test scores by student and school SES and religious educational stream, and school climate

[39_17989.htm_g/00002.jpg]


Figure 2. Eighth-grade student test scores as a function of student SES at different levels of positive student–teacher relationships (N = 25,923).


[39_17989.htm_g/00003.jpg]


The results of Model 1 indicated that religious educational stream and SES variables together explained 6.29% of the variance within schools in test scores in fifth grade and 7.44% in eighth grade. Inspection of the unique contributions of each predictor revealed that both SES variables were significant for both fifth and eighth grades, suggesting that students with higher personal SES or those from schools with higher SES achieved better scores compared to students with lower personal SES or those from schools with lower SES. In addition, fifth-grade student achievement in secular schools was significantly higher than the achievement of those in religious schools (p < .001). However, this was not the case for eighth-grade students.


More importantly, the results of Model 2 indicated that school climate factors accounted for an additional 2.52% (for fifth grade) and 2.41% (for eighth grade) of the explained within-schools variance related to student test scores. However, for fifth graders, only school violence was significantly related to test scores, such that higher reports of school violence were negatively related to test scores. In contrast, for eighth-grade students, only positive relationships with teachers and risky peer behavior contributed to the test score variance, such that higher levels of positive relationships with teachers and lower risky peer behavior were associated with higher test scores.


School Level


All variables were now averaged across students within schools. To test the influence of mean school climate, beyond school SES and religious educational stream factors, on school average test scores, two multiple linear hierarchical regressions were conducted separately for fifth- and eighth-grade schools. School SES and religious educational stream data were entered in Step 1 and the three school climate factors were added in Step 2. Results indicated that reports of school violence were associated with school mean test scores, above and beyond the contribution of SES, for both fifth (β = -.12, t (586) = -2.80, p = .005) and eighth grades (β = -.13, t (316) = -2.47, p = .014). Therefore, a positive school climate, as manifested by low levels of school violence, has a compensating positive contribution to mathematics test scores for both fifth- and eighth-grade schools, beyond the contribution of school SES and religious educational stream variables.


SES AND TEST SCORES: A SCHOOL CLIMATE MEDIATION HYPOTHESIS


The hypothesis that school climate mediates the relationship between student and school SES and test scores was tested both on student and school levels. The HLM results appear in Table 4 (Models 1 and 2 only).


Student Level


These models tested whether student perceptions of school climate mediate the relationship between student SES and test scores. To test mediation, the student SES effect coefficients in Model 1, which did not include school climate factors, were compared to those in Model 2, which included school climate factors. A significant decrease in the contribution of the student SES effects from Model 1 to Model 2 would indicate that the relationship between student SES and test scores is mediated by school climate. Full mediation would be indicated by significant student SES coefficients in Model 1, insignificant student SES coefficients in Model 2, and a significant decrease in the student SES coefficients’ from Model 1 to Model 2. Partial mediation would be indicated by a similar pattern of results, except that the significant decrease in the student SES coefficients’ contribution to the prediction of student achievement is coupled with a significant SES effect in Model 2 (Baron & Kenny, 1986; MacKinnon, 2008). A Sobel test was used to evaluate the significance of the decrease in the SES effects across models (Sobel, 1982). The Sobel test revealed an insignificant reduction in the effects of student SES after school climate factors were added to the regression for both fifth and eighth grades (ps > .05). Therefore, the mediation hypothesis was not confirmed by the data at the student level.




Table 4. HLM Models 1 and 2 Predicting Mathematics Test Scores (N = 59,946)

 

Effect

5th-Grade (n = 27,878)

8th-Grade (n = 25,923)

 

Model 1

Model 2

Model 1

Model 2

 

B

SE

t

B

SE

t

B

SE

t

B

SE

t

Intercept

564.90**

5.50

102.60

564.83**

5.51

102.51

568.59**

7.94

71.65

568.64**

7.94

71.59

School SES

7.29**

0.76

9.65

7.31**

0.76

9.63

12.11**

1.10

10.98

12.11**

1.10

10.96

Religious educational stream

-15.45**

3.63

-4.27

-15.50**

3.64

-4.26

-5.32

4.99

-1.07

-5.35

5.00

-1.07

Student SES

11.89**

1.57

7.55

11.40**

1.56

7.29

23.72**

1.93

12.28

23.50**

1.90

12.38

School Climate Factors

            

Positive student–teacher relationship

--

--

--

-2.10

3.57

-0.59

--

--

--

15.36**

3.10

4.95

Risky peer behavior

--

--

--

0.63

3.44

0.18

--

--

--

-14.81**

4.12

-3.59

School violence

--

--

--

-10.50*

3.24

-3.24

--

--

--

3.49

4.65

0.75

Between-school variance

1224.50**

1234.61**

1539.69***

1549.07***

Within-school variance

7739.52

7534.38

8057.63

7848.36

Between-school explained variance

23.3%

22.76%


29.94%

29.51%

Within-school explained variance

6.29%

8.81%

7.44%

9.85%

Within-school explained variance change

--

2.52%

--

2.41%

Note. HLM = hierarchical linear modeling; ICC = intraclass correlation; SES = socioeconomic status. First-level predictors are centered on school mean; Religious educational stream measured as 0 = secular, 1 = religious.

*p < .01, **p < .001



School Level


These models tested whether the relationship between school SES and school mean test scores is mediated by school climate at the school level. As for the compensation hypothesis, we used two multiple linear hierarchical regressions to test it, for fifth- and eighth-grade schools. In each regression, school SES and religious educational stream data were entered in Step 1, and the three school climate factors were added in Step 2. To test mediation we compared school SES in Step 1, which did not include school climate factors, to the SES coefficient in Step 2, which included school climate factors. Full and partial mediation effects were tested similarly to the student level. Similarly to the student level, the Sobel test revealed an insignificant reduction in the effects of school SES after school climate factors were added to the regression for both fifth- and eighth-grade schools (ps > .05). Therefore, the mediation hypothesis was not confirmed by the data at the school level.


SES AND TEST SCORES: A SCHOOL CLIMATE MODERATION HYPOTHESIS


The hypothesis that the relationship between SES and test scores depends on school climate was tested on three levels: student, student and school, and school.


Student Level


To test whether student perceptions of school climate moderate the relationship between student SES and test scores, we added a third model to the original student level two-model HLM presented in Table 5. Model 3 included interaction terms between school climate factors and student SES. As can be seen, none of the interactions reached significance, indicating that there was no evidence for a moderation effect of school climate, both for fifth- and eighth-grade students.




Table 5. HLM Models 2 and 3 Predicting Mathematics Test Scores (N = 59,946)

 

Effect

5th-Grade (n = 27,878)

8th-Grade (n = 25,923)

 

Model 2

Model 3

Model 2

Model 3

 

B

SE

t

B

SE

t

B

SE

T

B

SE

t

Intercept

564.83**

5.51

102.51

564.83**

5.51

102.50

568.64**

7.94

71.59

568.62**

7.94

71.61

School SES

7.31**

0.76

9.63

7.31**

0.76

9.63

12.11**

1.10

10.96

12.11**

1.10

10.96

Religious educational stream

-15.50**

3.64

-4.26

-15.50**

3.64

-4.26

-5.35

5.00

-1.07

-5.33

5.00

-1.07

Student SES

11.40**

1.56

7.29

10.93**

1.69

6.46

23.50**

1.90

12.38

24.51**

2.16

11.33

Positive studentteacher relationship

-2.10

3.57

-0.59

-1.99

3.57

-0.56

15.36**

3.10

4.95

11.05**

3.29

3.36

Risky peer behavior

0.63

3.44

0.18

0.50

3.42

0.15

-14.81**

4.12

-3.59

0.55

2.56

0.21

School violence

-10.50*

3.24

-3.24

-10.79*

3.26

-3.31

3.49

4.65

0.75

-2.02

2.39

-0.85

Student SES x  School Climate Factors (Interactions)

            

Student SES x  Positive studentteacher relationship

--

--

--

2.60

2.69

0.97

--

--

--

3.71

3.15

1.18

Student SES x  Risky peer behavior  

--

--

--

-0.06

2.55

-0.02

--

--

--

-568.62

7.94

-71.61

Student SES x  School violence

--

--

--

-1.58

2.28

-0.69

--

--

--

-12.11

1.10

-10.96

Between-school variance

1234.61**

1235.67***

1549.07***

1501.41***

Within-school variance

7534.38

7502.22

7848.36

7807.59

Between-school explained variance

22.76%

22.59%

29.51%

31.68%

Within-school explained variance

8.81%

9.16%

9.85%

10.31%

Within-school explained variance change

2.52%

.35%

2.41%

.46%

Note. HLM = hierarchical linear modeling; ICC = intraclass correlation; SES = socioeconomic status. First-level predictors are centered on school mean; Religious educational stream measured as 0 = secular, 1 = religious.

*p < .01, **p < .001


Cross-level Interactions


These models tested whether the relationship between student SES and test scores decreased in schools with positive climate. First, we examined whether the student SES and test scores slope variance was significant. A significant variance would justify the examination of cross-level interactions. Findings revealed significant student SES and test scores slopes variances both for fifth graders (B = 28.42, p < .001) and eighth graders (B = 4.71, p < .001). These findings justified further examination of interaction effects between student SES, school climate, and test scores.


The HLM analysis of this study used the slopes-as-outcomes model to explore interactions between student- and school-level predictors. This model tested the moderating effect of school-level school climate factors on the relationship between student SES and student-level test scores. Findings presented in Table 6 indicate that for fifth-grade students, school-level factors did not moderate the relationship between student SES and test scores. For eighth graders there was a significant interaction between student-level SES and positive student–teacher relationships. As can be seen in Figure 1, this interaction was manifested by a decreased student SES effect on test scores (i.e., smaller achievement gap) when positive relationships with teachers in the school increase. 


Table 6. HLM Predicting Student Test Scores including School-Level School Climate (N = 53,946

Effect

5th-Grade (n = 27,878)

8th-Grade (n = 25,923)

 

Model 4

Model 4

 

B

SE

t

B

SE

t

Intercept

559.59***

5.93

94.33

557.80***

9.33

59.76

School SES

6.47***

0.80

8.07

11.55***

1.12

10.27

Religious educational stream

-14.46***

3.72

-3.89

-10.23*

5.15

-1.99

Student SES

12.10***

1.60

7.58

20.61***

2.04

10.08

Student Level

      

Positive student–teacher relationship

2.66

9.07

0.29

-23.87*

11.80

-2.02

Risky peer behavior

9.74

9.55

1.02

29.40*

12.68

2.32

School violence

-33.57**

10.90

-3.08

-41.00*

17.57

-2.33

Student SES × Positive student–teacher relationship (school level)

-1.04

2.11

-0.49

-7.22***

2.28

-3.17

Student SES × Risky peer behavior (school level)

-3.10

2.17

-1.43

1.60

2.35

0.68

Student SES × School violence (school level)

-0.25

2.55

-0.10

-1.05

3.06

-0.34

Between-school variance

1154.03***

1451.75***

Within-school variance

7531.50

7847.17

Between-school explained variance

27.70%

33.94%

Within-school explained variance

8.81%

9.86%

Note. HLM = hierarchical linear modeling; SES = socioeconomic status. First-level predictors are centered on school mean; Religious educational stream measured as 0 = secular, 1 = religious.

*p < .05, **p < .01, ***p < .001



School Level


In these models the moderation effect of school climate on the SES–test scores relationship was tested with both SES and school climate at school level. As all variables were at school level, simple hierarchical linear regressions were utilized.


In each regression, school SES, religious educational stream, and school climate factors were entered in Step 1, and the interaction terms of school-level school climate factors and school SES were added in Step 2.


Results for fifth-grade schools revealed a non-significant interaction between school SES and climate, indicating that school climate does not moderate the relationship between school SES and average test scores of the school (F change (3,577) = 1.89; p = .130). Further, for eighth-grade schools, the school SES and climate interactions added 1% to the explained variance in test scores (F change (3,307) = 2.52; p = .048). However, none of the interactions was significant, suggesting only an overall, non-specific moderating effect of school climate on the SES–test scores relationship.


DISCUSSION


Education systems throughout the world are characterized by achievement gaps among students of different ethnicities and SES backgrounds, which often result from inequality of opportunity for academic success among certain groups of students. The negative individual and societal consequences of low academic achievement have aroused concern and great interest in this issue worldwide (Barton, 2003; Ladson-Billings, 2006; Lee, 2002; Nam & Haung, 2009). It is widely agreed among educational researchers and practitioners that school climate contributes significantly to the effectiveness of schools (National School Climate Council, 2007). Nevertheless, less is known about the exact mechanisms by which positive school climate affects academics.  


The scientific literature offers diverse, often confusing, explanations for the interrelation between SES background, school climate, and achievements. Some have argued that positive school climate contributes to academics above and beyond the negative influence of low SES on student proficiency (e.g., Brand et al., 2003; Schagen & Hutchison, 2003), suggesting that school climate has a positive additive contribution and thus compensates for low SES. Others have shown that a school’s SES influences its social climate, which in turn influences academic achievement (Khoury-Kassabri et al., 2004), suggesting that school climate mediates the relationship between SES background and academic achievements through school connectedness, school safety, positive relationships, and so forth. Finally, some authors have shown that the relationship between SES background and academic achievement changes across schools of different climate, suggesting that the relationship between student and school SES and academic achievement can be moderated by school climate (Sebring et al., 2006). Each suggested explanation for these interrelationships has different theoretical and practical implications.  


Using a nationally representative database gathered by the Israeli Ministry of Education that included school climate and academic variables, this study was undertaken to develop a more reliable and comprehensive assessment of the role of school climate in the relationship between student and school SES and achievement.


In summary, findings indicate that school climate had an additive compensation contribution to academic achievements, both on the student and school levels. School climate moderated the relationship between students’ SES and academic achievements. However, findings did not support the hypothesis that school climate mediated the relationship between SES background and academic achievement, both at the student and school levels.  


BETWEEN-SCHOOL VARIABILITY IN ACHIEVEMENT


The significant intraclass correlation coefficients (ICCs) indicated that schools, as units of measurement, were important for predicting variance in academic achievement, even after controlling for students’ personal background variables. This finding corresponded with the underlying assumptions of prior ecological nested empirical work that schools are affected by outer contexts such as the neighborhood and community, but also have a significant influence on and unique contribution to student outcomes (Benbenishty & Astor, 2005).


SES AND ACADEMIC ACHIEVEMENTS


In line with past research (e.g., Borman & Dowling, 2010), findings from the current study indicate that SES is a major predictor of academic outcomes, as we found strong indications that schools with higher SES classification had students with better average test scores compared with other schools. Furthermore, the findings reveal within-school inequalities between students of higher SES and their peers of other SES. This is an unusual finding, as most studies have not directly measured the SES of individual students. Rather, many studies have relied on student ethnicity (e.g., Lee, 2002; Powers et al., 2005) or school-level factors, such as the percentage of students eligible for free or reduced-price meals at school (e.g., Clotfelter et al., 2005; Robers et al., 2012; Stewart, 2008), to determine SES background. Rather than assuming that only between-school differences are related to academic achievements and therefore to social inequality, these analyses based on students’ personal SES show how teachers and schools may also hold responsibility for inequalities between lower and higher SES students attending the same school.


These findings imply that investing more resources in schools with low SES would not necessarily moderate within-school achievement gaps between different SES students. This suggests a need for further in-depth research, perhaps through mixed-method studies, to better understand how student SES influences their academic achievements, and how within-school achievement gaps could be moderated. For example, it has been proposed that teacher expectations contribute to the achievement gap between students of different backgrounds within the same school (Borman & Dowling, 2010; McKown & Weinstein, 2008). Varying teacher expectations could be addressed by developing a more positive school climate that promotes an anti-discrimination policy and enhances tolerance for students of different backgrounds, ethnicities, special needs, and so forth.


SES, SCHOOL CLIMATE, AND ACADEMIC ACHIEVEMENTS


In line with previous research (Gregory & Weinstein, 2004), findings suggest that school climate positively contributes to student achievement beyond their SES status, and therefore could compensate (to a degree) for the negative influence of low SES. Although our design cannot determine causality, these findings suggest that school climate could narrow achievement gaps between different SES students. Therefore, investing resources to promote a positive climate, in which students experience their school as a supportive and safe environment and enjoy positive relationships with their teachers, is of great importance. The findings may suggest that professionals and policymakers should be encouraged to develop and apply intervention programs that promote positive, caring, and respectful student–teacher relationships and to promote feelings of safety at school. These interventions would not only improve student satisfaction with school, but could also promote academic proficiency, especially among students of low SES backgrounds.


Comparisons between grade levels showed school climate had a greater contribution to variance in test scores for fifth-grade students compared to eighth-grade students. The findings correspond with past research indicating a greater contribution of school climate to younger students’ achievements compared to older students (e.g., Ryan & Patrick, 2001; Sirin, 2005). This may be explained in light of more positive school experiences of younger students, compared to older students, in terms of school violence and relationships with teachers (Benbenishty & Astor, 2005; Pianta & Allen, 2008; Thapa, Cohen, Guffey, & Higgins-D’Alessandro, 2013). Nevertheless, past research has indicated that achievement gaps between students of different SES backgrounds grow wider as students grow older (Gilboa, 2010). Therefore, intervention should focus on improving school climate at all school levels but especially in higher grades, where climate tends to be less positive, to compensate for the growing achievement gaps.  


The school-level compensating model revealed further important insights regarding student reports on violence and feelings of insecurity at school. Although our measure of school violence was not always significantly associated with achievement at the student level, its contribution to variance in achievement at the school level was significant for both grade levels. These findings may suggest that a student’s personal experience of victimization does not necessarily influence his or her academic achievement. Rather, students in violent schools where a general feeling of insecurity prevails achieved lower scores. This finding mirrors past research showing that insecurity and fear among students are influenced by the collective experience rather than personal victimization (Astor, Benbenishty, Zeira, & Vinokur, 2002; Astorr, Benbenishty, Vinokur, & Zeira, 2006; Kitsantas, Ware, & Martinez-Arias, 2004; Mijanovich & Weitzman, 2003). The findings may point to the need for planning and implementing intervention programs that consider the whole school community, as opposed to focusing on personal characteristics of the victims or perpetrators of school violence (De Pedro, Astor, Gilreath, & Benbenishty, 2013). Accordingly, Vreeman and Carroll (2007), who examined studies evaluating school-based anti-bullying efforts, concluded that the most promising results were reported for whole-school anti-bullying efforts. Such multidisciplinary approaches establish school-wide rules for bullying, teacher training, close contact with parents, and the involvement of community members. In the context of today’s greater accountability and demands for children’s achievement, our findings suggest that improving feelings of safety at school would not only make school a more pleasant place for students, but it could actually improve average test scores and enhance academic outcomes. Schools, especially those serving children from low-income families receiving special funds, are under tremendous pressure to abide by state and federal academic regulations. It is therefore logical to consider climate as an integral element of achievement, a part of the academic picture (Cohen, McCabe, Michelli, & Pickeral, 2009, p. 191).


The hypothesis that a school’s SES influences its social climate, which in turn influences academic achievement, was not supported in the current study, both at the student and school levels. The underlying assumption of this mediation model is that students and schools with lower SES experience lower levels of school climate, and this adverse climate impinges on productive learning and results in lower scores (Syvertsen, Flanagan, & Stout, 2009; Willms, 2006). Nevertheless, our study did not find evidence for such a phenomenon—students and schools with low SES experienced school climate levels that were not different systematically from students and schools with higher SES. These findings suggest that it is possible to maintain and promote a positive climate in schools of low SES and those located in poor and violent neighborhoods.


The notion of promoting a positive climate in schools of low SES is supported by previous research findings indicating that several schools had the same levels of school violence, regardless of different SES backgrounds of their students (Olweus, 1993). Similarly, another study (Astor & Benbenishty, 2005) showed that schools of similar socio-demographic background had different levels of violence and school climate.


A study on school violence and theoretically atypical schools (Astor, Benbenishty, & Estrada, 2009) supported the idea that positive school climate can be enhanced in schools embedded in poor or violent neighborhoods. Case studies of such atypical peaceful schools have emphasized the leadership of a visionary, inspiring, and strong principal who can mobilize students, teachers, and staff to create a caring, inclusive, and nurturing school environment that will eventually support and enhance positive outcomes for all students (Astor et al., 2009). It is possible and important, therefore, to sustain a positive school climate, even in schools located in communities of low SES.


The study also tested whether the relationship between student SES and test scores was weakened in schools with a positive climate. Findings suggested that achievement gaps between students of different SES in the same school decreased in schools characterized by positive climate, especially those in which students experienced highly supportive and positive relationships with teachers. These schools somehow encourage close and caring student–teacher relationships, which support feelings of safety (Whitlock, 2006) and a strong connection to the school (Blum, 2005). These schools may also have inspiring leaders that encourage care and support for all students, which in turn enhances student outcomes because they feel they have someone to turn to in need (Astor et al., 2009). Research on the interrelations between student- and school-level characteristics and academic outcomes is not well established (Bradshaw, Sawyer, & O’Brennan, 2009). Findings from the current study that suggested some school factors could moderate the relationship between SES and achievement highlight the need for further research to better identify those influential structural or interpersonal aspects of schools. Qualitative studies could be more perceptive of how school climate is experienced by students, teachers, and other school personnel, and therefore allow a more comprehensive understanding of how school climate affects achievement. Furthermore, longitudinal study designs would be helpful in exploring causality; for example, how perceptions of school climate at one point in time are related to student outcomes, such as test scores, at another point in the future.


In conclusion, findings from the current study stress the importance of promoting positive climate among all schools, and especially those serving communities living in poverty. Findings cast a different light upon the common argument that family and school backgrounds are the only important factors for understanding student outcomes. Rather, it seems that schools do matter, and that a positive climate in the schools may have a profound impact on students’ academic achievements, and therefore on the social mobility of students of low SES. Policymakers are encouraged, therefore, to invest more resources to enhance positive relationships and safety in the schools to promote equality of opportunities for all students.


STUDY LIMITATIONS


There are several limitations to this study that warrant consideration. A comprehensive evaluation of school climate should recognize parent and school personnel voices, in addition to students’ voices (National School Climate Council, 2007). Indeed, the Meitzav routinely collects other voices to evaluate school climate (teachers and school leaders). However, due to length limitations, the current study was based solely on students’ reports. Further, a school climate assessment should include all the dimensions that color and shape the process of teaching and learning, and educators’ and students’ experiences in the school building (Cohen, Pickeral, & McCloskey, 2008). Although the present study examined many central aspects of school climate, including safety, relationships, and engagement, the available data did not include additional characteristics of school life, such as the characteristics of school’s external environment. Future studies are encouraged to carry out broader-scope evaluations, acknowledging teachers’ and parents’ voices in addition to students’ voices, and to include the many aspects of school climate to allow more comprehensive analyses.


Finally, the internal consistency for risky peer behavior was low; therefore, conclusions from analyses based on this factor should be considered tentative.


References


Akiba, M., LeTendre, G. K., & Scribner, J. P. (2007). Teacher quality, opportunity gap, and national achievement in 46 countries. Educational Researcher, 36, 369–387. doi:10.3102/0013189X07308739


Astor, R. A., Benbenishty, R., & Estrada, J. N. (2009). School violence and theoretically atypical schools: The principal’s centrality in orchestrating safe schools. American Educational Research Journal, 46, 423–461. doi:10.3102/0002831208329598


Astor, R. A., Benbenishty, R., Vinokur, A. D., & Zeira, A. (2006). Arab and Jewish elementary school students’ perceptions of fear and school violence: Understanding the influence of school context. British Journal of Educational Psychology, 76, 91–118. doi:10.1348/000709905X37307


Astor, R. A., Benbenishty, R., Zeira, A., & Vinokur, A. (2002). School climate, observed risky behaviors, and victimization as predictors of high school students’ fear and judgments of school violence as a problem. Health Education & Behavior, 29, 716–736. doi:10.1177/109019802237940


Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182. doi:10.1037/0022-3514.51.6.1173


Barton, P. E. (2003). Parsing the achievement gap: Baselines for tracking progress. Princeton, NJ: Educational Testing Service. Retrieved from http://www.ets.org/Media/Research/pdf/PICPARSING.pdf


Beller, M. (2013). Assessment in the service of learning: Theory and practice. Ramat Gan, Israel: National Authority for Measurement and Evaluation in Education.


Benbenishty, R., Khoury-Kassabri, M., Roziner, I., & Astor, R. (2005). The relationship between school climate, school violence and academic achievements [in Hebrew]. Jerusalem, Israel: Hebrew University, Paul Baerwald School of Social Work.


Benbenishty, R., & Astor, R. A. (2005). School violence in context: Culture, neighborhood, family, school, and gender. New York, NY: Oxford University Press.


Berliner, D. C. (2008). Letter to the president. Journal of Teacher Education, 59, 252–256. doi:10.1177/0022487108317788


Blum, R. (2005). School connectedness: Improving the lives of students. Baltimore, MD: Johns Hopkins University, Bloomberg School of Public Health. Retrieved from Center for Effective Collaboration and Practice website: http://cecp.air.org/download/MCMonographFINAL.pdf


Borman, G., & Dowling, M. (2010). Schools and inequality: A multilevel analysis of Coleman’s equality of educational opportunity data. Teachers College Record, 112, 1201–1246.


Bradshaw, C. P., Sawyer, A. L., & O’Brennan, L. M. (2009). A social disorganization perspective on bullying-related attitudes and behaviors: The influence of school context. American Journal of Community Psychology, 43, 204–220. doi:10.1007/s10464-009-9240-1


Brand, S., Felner, R., Shim, M., Seitsinger, A., & Dumas, T. (2003). Middle school improvement and reform: Development and validation of a school-level assessment of climate, cultural pluralism, and school safety. Journal of Educational Psychology, 95, 570–588. doi:10.1037/0022-0663.95.3.570


Bryk, A. S., & Schneider, B. L. (2002). Trust in schools: A core resource for improvement. New York, NY: Russell Sage Foundation.


Bryk, A. S., Sebring, P. B., Allensworth, E., Easton, J. Q., & Luppescu, S. (2010). Organizing schools for improvement: Lessons from Chicago. Chicago, IL: University of Chicago Press.


Caldas, S. J., Bernier, S., & Marceau, R. (2009). Explanatory factors of the Black achievement gap in Montréal’s public and private schools: A multivariate analysis. Education and Urban Society, 41, 197–215. doi:10.1177/0013124508325547


Chen, J.-K., & Astor, R. A. (2011a). School engagement, risky peers, and student–teacher relationships as mediators of school violence in Taiwanese vocational versus academically oriented high schools. Journal of Community Psychology, 39, 10–30. doi:10.1002/jcop.20413


Chen, J.-K., & Astor, R. A. (2011b). Students’ personal traits, violence exposure, family factors, school dynamics and the perpetration of violence in Taiwanese elementary schools. Health Education Research, 26, 150–166. doi:10.1093/her/cyq083


Chen, J.-K., & Astor, R. A. (2012). School variables as mediators of personal and family factors on school violence in Taiwanese junior high schools. Youth & Society, 44, 175–200. doi:10.1177/0044118X12448145


Clotfelter, C. T., Ladd, H. F., & Vigdor, J. (2005). Who teaches whom? Race and the distribution of novice teachers. Economics of Education Review, 24, 377–392. doi:10.1016/j.econedurev.2004.06.008


Cohen, J., McCabe, L., Michelli, N. M., & Pickeral, T. (2009). School climate: Research, policy, practice, and teacher education. Teachers College Record111(1), 180–213.


Cohen, J., Pickeral, T., & McCloskey, M. (2008). The challenge of assessing school climate. Educational Leadership, 66(4). Retrieved from www.ascd.org/publications/educational_leadership/dec08/vol66/num04/toc.aspx


Coleman, J. S. (1966). Equality of educational opportunity. Washington, DC: U.S. Government Printing Office.


De Pedro, K., Astor, R.A., Gilreath, T., & Benbenishty, R. (2013). Victimization among students in military-connected schools: The role of school climate. Paper accepted for presentation at the American Educational Research Association Conference, San Francisco, California


Dwyer, K. P., Osher, D., & Hoffman, C. C. (2000). Creating responsive schools: Contextualizing early warning, timely response. Exceptional Children, 66, 347–365.


Gilboa, Y. (2010). The growing gaps in achievements: Lesson from the Meitzav exams (in Hebrew). Jerusalem, Israel: Van Leer Jerusalem Institute. Retrieved from http://www.vanleer.org.il/sites/files/product-pdf/research_policy10.pdf


Gottfredson, G. D., Gottfredson, D. C., Payne, A. A., & Gottfredson, N. C. (2005). School climate predictors of school disorder: Results from a national study of delinquency prevention in schools. Journal of Research in Crime & Delinquency, 42, 412–444. doi:10.1177/0022427804271931


Gregory, A., & Weinstein, R. S. (2004). Connection and regulation at home and in school: Predicting growth in achievement for adolescents. Journal of Adolescent Research, 19, 405–427. doi:10.1177/0743558403258859


Grigg, W., Donahue, P., & Dion, G. (2007). The nation’s report card: 12th-grade reading and mathematics 2005 (NCES Publication 2007-468). Retrieved from the National Center for Education Statistics website: http://nces.ed.gov/nationsreportcard/pdf/main2005/2007468.pdf


Huff, C. R. (2002). Gangs in America (3rd ed.). Thousand Oaks, CA: Sage Publishing.

Jencks, C., Smith, M., Acland, H., Bane, M. J., Cohen, D., Hintis, H., . . . Michelson, S. (1972). Inequality: A reassessment of the effect of family and schooling in America. New York, NY: Basic Books.


Khoury-Kassabri, M., Benbenishty, R., Astor, R. A., & Zeira, A. (2004). The contributions of community, family, and school variables to student victimization. American Journal of Community Psychology, 34, 187–204. doi:10.1007/s10464-004-7414-4


Kitsantas, A., Ware, H. W., & Martinez-Arias, R. (2004). Students’ perceptions of school safety: Effects by community, school environment, and substance use variables. Journal of Early Adolescence, 24, 412–430. doi:10.1177/0272431604268712


Klem, A. M., & Connell, J. P. (2004). Relationships matter: Linking teacher support to student engagement and achievement. Journal of School Health, 74, 262–273. doi:10.1111/j.1746-1561.2004.tb08283.x


Ladd, G. W., & Dinella, L. M. (2009). Continuity and change in early school engagement: Predictive of children’s achievement trajectories from first to eighth grade? Journal of Educational Psychology101(1), 190–206. doi:10.1037/a0013153


Ladd, H. F., & Walsh, R. P. (2002). Implementing value-added measures of school effectiveness: Getting the incentives right. Economics of Education Review, 21, 1–17. doi:10.1016/S0272-7757(00)00039-X


Ladson-Billings, G. (2006). From the achievement gap to the education debt: Understanding achievement in U.S. schools. Educational Researcher, 35(7), 3–12. doi:10.3102/0013189X035007003


Lee, J. (2002). Racial and ethnic achievement gap trends: Reversing the progress toward equity? Educational Researcher, 31(1), 3–12. doi:10.3102/0013189X031001003


Ma, L., Phelps, E., Lerner, J. V., & Lerner, R. M. (2009). The development of academic competence among adolescents who bully and who are bullied. Journal of Applied Developmental Psychology, 30, 628–644. doi:10.1016/j.appdev.2009.07.006


MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. New York, NY: Lawrence Erlbaum.


Marachi, R., Astor, R. A., & Benbenishty, R. (2007). Effects of student participation and teacher support on victimization in Israeli schools: An examination of gender, culture, and school type. Journal of Youth and Adolescence, 36, 225–240. doi:10.1007/s10964-006-9136-4


McEvoy, A., & Welker, R. (2000). Antisocial behavior, academic failure, and school climate: A critical review. Journal of Emotional and Behavioral Disorders, 8, 130–140. doi:10.1177/106342660000800301


McKown, C., & Weinstein, R. S. (2008). Teacher expectations, classroom context, and the achievement gap. Journal of School Psychology, 46, 235–261. doi:10.1016/j.jsp.2007.05.001


McNeely, C. A., Nonnemaker, J. M., & Blum, R. W. (2002). Promoting school connectedness: Evidence from the National Longitudinal Study of Adolescent Health. Journal of School Health, 72, 138–146. doi:10.1111/j.1746-1561.2002.tb06533.x


Mijanovich, T., & Weitzman, B. C. (2003). Which “broken windows” matter? School, neighborhood, and family characteristics associated with youths’ feelings of unsafety. Journal of Urban Health, 80, 400–415. doi:10.1093/jurban/jtg045


Nam, Y., & Haung, J. (2009). Equal opportunity for all? Parental economic resources and children’s educational attainment. Children and Youth Services Review, 31, 625–634. doi:10.1016/j.childyouth.2008.12.002


Nansel, T. R., Overpeck, M., Pilla, R. S., Ruan, W. J., Simons-Morton, B., & Scheidt, P. (2001). Bullying behaviors among US youth: Prevalence and association with psychosocial adjustment. Journal of the American Medical Association, 285, 2094–2100. doi:10.1001/jama.285.16.2094


National Authority for Measurement and Evaluation in Education. (n.d.). School climate and pedagogical environment questionnaires: General description of students’ questionnaires (in Hebrew). Retrieved from the National Authority for Measurement and Evaluation in Education website: http://cms.education.gov.il/EducationCMS/Units/Rama/Meitzav/Aklim_f.htm


National School Climate Council. (2007). The school climate challenge: Narrowing the gap between school climate research and school climate policy, practice guidelines and teacher education policy. Retrieved from the Education Commission of the United States website: http://www.ecs.org/school-climate


Noguera, P. (2010, June 14). A new vision for school reform. The Nation. Retrieved from http://www.thenation.com/article/new-vision-school-reform


Olweus, D. (1993). Bullying at school: What we know and what we can do. Malden, MA: Blackwell.


Olweus, D., Limber, S., & Mihalic, S. F. (1999). Blueprints for violence prevention, book nine: Bullying prevention program. Boulder, CO: Center for the Study and Prevention of Violence.


Osher, D., Dwyer, K. P., Jimerson, S. R., & Brown, J. A. (2012). Developing safe, supportive, and effective schools: Facilitating student success to reduce school violence. In S. R. Jimerson, A. B. Nickerson, M. J. Mayer, & M. J. Furlong (Eds.), Handbook of school violence and school safety: International research and practice (2nd ed., pp. 27–44). New York, NY: Routledge.


Pianta, R. C., & Allen, J. P. (2008). Building capacity for positive youth development in secondary school classrooms: Changing teachers’ interactions with students. In M. Shinn & H. Yoshikawa (Eds.), Toward positive youth development: Transforming schools and community programs (pp. 21–39). New York, NY: Oxford University Press.


Powers, J. D., Bowen, G. L., & Rose, R. A. (2005). Using social environment assets to identify intervention strategies for promoting school success. Children & Schools, 27, 177–187. doi:10.1093/cs/27.3.177


Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models: Applications and data analysis methods (2nd ed.). Thousand Oaks, CA: Sage.


Robers, S., Zhang, J., Truman, J., & Snyder, D. T. (2012). Indicators of school crime and safety: 2011. Washington, DC: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics and U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from http://bjs.ojp.usdoj.gov/content/pub/pdf/iscs11.pdf


Rutter, M., & Maughan, B. (2002). School effectiveness findings 1979–2002. Journal of School Psychology, 40, 451–475. doi:10.1016/S0022-4405(02)00124-3


Ryan, A. M., & Patrick, H. (2001). The classroom social environment and changes in adolescents’ motivation and engagement during middle schools. American Educational Research Journal, 38, 437–460. doi:10.3102/00028312038002437


Schagen, I., & Hutchison, D. (2003). Adding value in educational research: The marriage of data and analytical power. British Educational Research Journal, 29, 749–765. doi:10.1080/0141192032000133659


Sebring, P. B., Allensworth, E., Bryk, A. S., Easton, J. Q., & Luppescu, S. (2006). The essential supports for school improvement. Chicago, IL: Consortium on Chicago School Research at the University of Chicago.


Shavit, Y., & Blank, C. (2012). School discipline and achievement in Israel. In R. Arum & M. Velez (Eds.), Improving learning environments: School discipline and student achievement in comparative perspective (pp. 104–136). Palo Alto, CA: Stanford University Press.


Sherblom, S. A., Marshall, J. C., & Sherblom, J. C. (2006). The relationship between school climate and math and reading achievement. Journal of Research in Character Education4(1–2), 19–32.


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


Smith, P. K., Madsen, K. C., & Moody, J. C. (1999). What causes the age decline in reports of being bullied at school? Towards a developmental analysis of risks of being bullied. Educational Research, 41(3), 267–285.


Sobel, M. E. (1982). Asymptotic confidence intervals for indirect effects in structural equation models. In S. Leinhardt (Ed.), Sociological methodology 1982 (pp. 290–313). San Francisco, CA: Jossey-Bass.


Stewart, E. A. (2003). School social bonds, school climate, and school misbehavior: A multilevel analysis. Justice Quarterly, 20, 575–604. doi:10.1080/07418820300095621


Stewart, E. B. (2008). School structural characteristics, student effort, peer associations, and parental involvement: The influence of school- and individual-level factors on academic achievement. Education and Urban Society, 40, 179–204. doi:10.1177/0013124507304167


Sullivan, K. (2000). What we know about bullying. In K. Sullivan (Ed.), The anti-bullying handbook (pp. 9–39). New York, NY: Oxford University Press.


Syvertsen, A. K., Flanagan, C. A., & Stout, M. D. (2009). Code of silence: Students’ perceptions of school climate and willingness to intervene in a peer’s dangerous plan. Journal of Educational Psychology, 101, 219–232. doi:10.1037/a0013246


Thapa, A., Cohen, J., Guffey, S., & Higgins-D’Alessandro, A. (2013). A review of school climate research. Review of Educational Research83(3), 357–385.


Van der Berg, S. (2008). How effective are poor schools? Poverty and educational outcomes in South Africa. Studies in Educational Evaluation, 34, 145–154. doi:10.1016/j.stueduc.2008.07.005


Vreeman, R. C., & Carroll, A. E. (2007). A systematic review of school-based interventions to prevent bullying. Archives of Pediatrics & Adolescent Medicine, 161(1), 78–88.


Wang, M.-T., & Holcombe, R. (2010). Adolescents’ perceptions of school environment, engagement, and academic achievement in middle school. American Educational Research Journal, 47, 633–662. doi:10.3102/0002831209361209


Weinstein, R. S., Gregory, A., & Strambler, M. J. (2004). Intractable self-fulfilling prophecies: Fifty years after Brown v. Board of Education. American Psychologist, 59, 511–520. doi:10.1037/0003-066X.59.6.511


Whitlock, J. L. (2006). Youth perceptions of life at school: Contextual correlates of school connectedness in adolescence. Applied Developmental Science, 10, 13–29. doi:10.1207/s1532480xads1001_2


Willms, J. D. (2006). Learning divides: Ten policy questions about the performance and equity of schools and schooling systems (UIS Working Paper No. 5). Montreal, Canada: UNESCO Institute for Statstics. Retrieved from http://unesdoc.unesco.org/images/0014/001470/147066e.pdf


Zussman, N., & Tsur, S. (2008). The effect of Israeli students’ socioeconomic background on their achievements in the matriculation examinations (in Hebrew). Jerusalem, Israel: Bank of Israel, Research Department. Retrieved from http://www.boi.org.il/deptdata/mehkar/seker84/surv84_6.pdf





Cite This Article as: Teachers College Record Volume 117 Number 7, 2015, p. 1-34
https://www.tcrecord.org ID Number: 17989, Date Accessed: 3/8/2021 7:14:39 AM

Purchase Reprint Rights for this article or review
 
Article Tools
Related Articles

Related Discussion
 
Post a Comment | Read All

About the Author
  • Ruth Berkowitz
    University of Southern California, School of Social Work
    E-mail Author
    RUTH BERKOWITZ, PhD, post-doctoral researcher at the USC School of Social Work. Her main research interest is the school setting as an arena to form the lives and well-being of children in normative settings and as influencing social justice and social mobility, school climate, students’ academic and socio-emotional outcomes, military-connected students, students with learning and attention deficit disorders, school evidence-based practices, schools as learning organizations. Publications include:

    Berkowitz, R., DePedro, N.K., Couture, J., & Benbenishty, R. (2014). Military Parents' Perceptions of Public School Support for Their Children. Children & Schools.

    Berkowitz, R. (2013). Student and teacher responses to violence in school: The divergent views of bullies, victims, and bully-victims. School Psychology International.


  • Hagit Glickman
    National Authority for Measurement and Evaluation
    E-mail Author
    Hagit Glickman, PhD, is the General-Director of the National Authority for Measurement and Evaluation in Education (RAMA). In recent years she involves in a large variety of research projects in the area of education; including designing and analyzing large scale assessments and surveys, cross sectional studies and longitudinal studies, evaluation of educational programs, and developing applied statistical and psychometric methodologies. She has a Ph.D. in Statistic from the Hebrew University of Jerusalem (since 2000). Publications include:

    Glickman, H., Lipshtat N. (2013) Ability Groups in Lower Secondary Mathematics.

    Glickman, H., Lipshtat N., Raz T., & Ratner D. (2011) Does a National System of "Small-Group" Learning Improve Outcomes? An Analysis of the "New Horizons" Education Reform.


  • Rami Benbenishty
    Bar Ilan University
    E-mail Author
    RAMI BENBENISHTY is a professor of social work at Bar Ilan University. He is studying school climate and child protection, victimization and well-being. Publications include:

    Benbenishty, R., Jedweb, M., Chen, C. Glasser, S., Slutzky, H. Siegal, G., Lavi-Sahar, Z., Lerner-Geva, L. (in press). Predicting the decisions of hospital based child protection teams to report to child protective services, police and community welfare services. Child Abuse & Neglect.

    Benbenishty, R., & Schmid, H. (2013). Public attitudes toward the identification and reporting of alleged maltreatment cases among social groups in Israel. Children and Youth Service Review, 35, 332–339.

    Gilreath, T., D., Astor, R. A., Cederbaum, J. A., Atule, H., & Benbenishty, R. (in press). Prevalence and correlates of victimization and weapon carrying among military and non-military connected youth in southern California. Preventive Medicine.


  • Elisheva Ben-Artzi
    Bar-Ilan University
    E-mail Author
    ELISHEVA BEN-ARTZI is a Senior Lecturer at the Center for Academic Studies and Bar-Ilan University, Israel and former Head of Experimental Program at the Psychology Department, Bar-Ilan University, Israel. Main research interest include: dyslexia, temporal-order processes in speech comprehensions, false-memories. Publications include:

    Ben-Artzi, E., Fostick. L., & Babkoff, H. (2011). Auditory temporal processes in the elderly. Audiology Research, 1, 21-23.

    Faust, M., Ben-Artzi, E., & Vardi, N. (2012). Semantic processing in native and second language: Evidence from hemispheric differences in fine and coarse semantic coding. Brain and Language, 123, 228–233.

    Fostick, L., Ben-Artzi, E., & Babkoff, H. (2013). Aging and speech perception: Beyond hearing threshold and cognitive ability. Journal of Basic and Clinical Physiology and Pharmacology, 24, 175–183.


  • Tal Raz
    National Authority for Measurement and Evaluation in Education in Israel
    E-mail Author
    TAL RAZ is the director of projects' evaluation in the national authority for measurement and evaluation in education in Israel (RAMA). Her research interests include the measurement of school climate; evaluation of large scale reforms in education; interrelations between school climate and scholastic achievement. Publications include:

    Weisenberg, M., Raz, T., & Hener, T. (1998). The influence of film- induced mood on pain perception. Pain, 76, 365–375.

    Vakil, E., Raz, T., Lev, D.A. (2010). Probing the brain substrates of cognitive processes responsible for context effects on recognition memory. Aging, Neuropsychology, and Cognition, 17,(5), 519–544.

    Glickman, H., Lipshtat, N., Raz T., & Ratner, D. (2011). Does a National System of "Small-Group" Learning Improve Outcomes? An Analysis of the "New Horizons" Education Reform.


  • Nurit Lipshtat
    National Authority for Measurement and Evaluation in Education in Israel
    E-mail Author
    NURIT LIPSHTAT is director of longitudinal study design field in the National Authority for Measurement and Evaluation in Education in Israel (RAMA) where she is responsible for data analysis and methodological design. She studies large scale data sets (such as the Mitzav, and PISA) and intervention evaluation. Publications include:

    Greenbaum, L., Rigbi, A., Lipshtat, N., Cilia, R., Tesei S, Asselta, R., Djadett,i R., Goldwurm, S. and Lerer, B. (2013). Association of nicotine dependence susceptibility gene, CHRNA5, with Parkinson's disease age of onset: Gene and smoking status interaction. Parkinsonism Related Disorder, 19(1), 72–6.

    Gorfine, M., Lipshtat, N., Freedman L., & Prentice R. L. (2007). Linear measurement error models with restricted sampling. Biometrics, 63, 137–142.

    Glickman, H., Lipshtat N. (2013) Ability Groups in Lower Secondary Mathematics.


  • Ron Avi Astor
    University of Southern California, School of Social Work
    E-mail Author
    RON AVI ASTOR is the Thor Professor of social urban development at the USC School of Social Work and USC Rossier School of Education. His work examines the role of the physical, social-organizational and cultural contexts in schools related to different kinds of school violence (e.g., sexual harassment, bullying, school fights, emotional abuse, weapon use, teacher/child violence). Currently, Astor is applying knowledge gained from these prior studies to improve school climate and reduce risky behaviors in military-connected schools. His work has been funded by the Department of Defense Educational Activity, National Institutes of Mental Health, H.F. Guggenheim Foundation, National Academy of Education/Spencer Foundation, William T. Grant Foundation, Israeli Ministry of Education, a Fulbright Senior Scholar Fellowship, University of Michigan, USC and the W.K. Kellogg Foundation. Publications include:

    Astor, R.A., De Pedro, K., Gilreath, T., Esqueda, M., & Benbenishty, R. (2013). The promotional role of community, school, family, and peer contexts for military students in wartime. Clinical Child and Family Psychology Review. DOI 10.1007/s10567-013-0139-x

    Gilreath, T.D., Astor, R.A., Cederbaum, J.A., Atuel, H.R., & Benbenishty, R. (2013). School violence and victimization among military connected youth. Preventive Medicine. http://dx.doi.org/10.1016/j.ypmed.2013.12.002

    Gilreath, T.D., Astor, R.A., Estrada, J.N., Benbenishty, R., & Unger, J. B. (2014). School victimization and substance use among adolescents in California. Prevention Science. doi: 10.1007/s11121-013-0449-8


 
Member Center
In Print
This Month's Issue

Submit
EMAIL

Twitter

RSS