![[PDF]](/pb-assets/Icons/adobe-pdf-icon-1498649896243.png){kind=icon}
Abstract
In explaining academic achievement, school climate and social belonging (connectedness, identification) emerge as important variables. However, both constructs are rarely explored in one model. In the current study, a social psychological framework based on the social identity perspective (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) is introduced that provides a way to integrate these two areas of enquiry. Using this framework, the current study (N = 340 grade 7 and 9 students) investigates: (a) school climate and social identification as distinct predictors of academic achievement; and (b) social identification as a mediator of the school climate and achievement relationship. Achievement in reading, numeracy and writing was assessed by a national standardized test. The three variables most significantly associated with achievement were parental education, socio-economic status, and school identification. In line with predictions, school identification fully mediated the relationship between school climate and academic achievement in numeracy and writing, but not reading. The research highlights the importance of feeling psychologically connected to the school as a group for academic success.
Academic achievement is of central concern in the education domain. Over many years a range of variables has been identified that are related to achievement including socio-economic status, parental education, student cognitive abilities, school belonging or connectedness, and school climate (e.g., Hattie, 2013). Thapa, Cohen, Guffey, and Higgins-D’Allessandro (2013) point out that many jurisdictions are focused on school reform in order to improve well-being and achievement outcomes for young people and that school climate is a central construct of interest. Of the many predictors of achievement, factors that relate to the school social environment (such as climate and belonging) can be directly influenced by school policy and practices (Thapa et al., 2013). However, it has been difficult to fully assess these factors because school climate and school belonging or connection, are rarely jointly examined.
In the current study, the social identity perspective (Tajfel & Turner, 1979; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) is used as a framework to help integrate existing work. School climate and school identification (connectedness, belonging, relatedness) are conceptualized as distinct but related concepts and were investigated in the same model to explain academic achievement (e.g., Roeser, Midgley, & Urdan, 1996; Wang & Holcombe, 2010). Furthermore, school identification was investigated as a potential mediator in helping to explain how school climate factors come to impact on achievement (Bizumic, Reynolds, Turner, Bromhead, & Subasic, 2009; Loukas, Suzuki & Horton, 2006).
In addition to this conceptual contribution, the current study has a range of methodological advantages over related research. Much of the previous research in the educational context has used less objective measures of academic achievement (e.g., self-reported grades, grades, and state-based tests; see below e.g., Crosnoe, Johnson, & Elder, 2004; Ma & Klinger, 2000). Furthermore, often the focus is on one aspect of achievement such as mathematics performance (see Moos, 1987). The current study addresses these issues as it utilizes a national standardized measure of academic achievement; the Australian National Assessment Program–Literacy and Numeracy (NAPLAN), which assesses three main areas of academic performance: Reading, writing, and numeracy. In addition, other key predictors of achievement (i.e., socio-economic status, parental education) are controlled for, making the current research robust and informative. It is possible to assess the contribution of school climate and school identification in explaining achievement over and above the contribution of these other factors. The model that is investigated, then, is comprehensive, adding to the body of evidence regarding academic achievement. Before the research and findings are discussed, important findings relating to school climate, belonging and connectedness and social identity are outlined in more detail.
School climate and academic achievement
Around the world an increasing number of government departments of education are focused on school climate as a fundamental aspect of school reform and improvement. It is at the forefront of efforts in addressing the achievement gap between different sub-groups of students, student well-being, and bullying (e.g., Thapa et al., 2013; Turner, Reynolds, Lee, Subasic, & Bromhead, 2014). School climate is a broad, multi-dimensional concept that refers to social aspects of the learning environment including school members’ interactions and relationships, shared values and norms, and the personal development and growth of the members (Anderson, 1982; Moos, 1978; Ramelow, Currie, & Felder-Puig, 2015; Thapa et al., 2013). It is typically measured through self-report, where members of the school community are asked in a survey to report their perceptions about school functioning in relation to relationships, academic emphasis and shared values and rules.
One reason for the growing interest in school climate is its’ well-established relationship with achievement (e.g., Hoy & Hannum, 1997; McEvoy & Welker 2000; Ma & Klinger, 2000; Sherblom, Marshall, & Sherblom, 2006; Stewart, 2008; Wang, Haertel, & Walberg, 1997) and health and well-being outcomes (for a review see Thapa et al., 2013). An early example of the relationship with achievement is provided by Moos and Moos (1978) who assessed school climate and found that students’ perceptions of relationships within the school (affiliation and teacher support) were positively correlated with their mean grades. To give a more recent example, using a nationally representative sample of Americans, Crosnoe et al. (2004), showed that stronger teacher-student relations (an aspect of school climate) was associated with higher academic achievement (assessed using self-reported grades).
School climate can also act as a protective factor for disadvantaged students. Hopson and Lee (2011) found students who experienced high poverty had lower academic achievement. Importantly, when disadvantaged students reported a positive school climate, school grades were more aligned with peers from higher income families. Likewise, O’Malley, Voight, Renshaw, and Eklund (2015) found in a survey of approximately 500,000 students from 900 Californian schools that regardless of family structure (two-parent or one-parent) there was a significant relationship between positive school climate ratings and self-reported grade point average.
Research using more standardized measures of achievement (North American state-based tests) rather than grades, also shows a relationship between school climate and achievement in mathematics, science, reading, and writing. Ma and Klinger (2000) assessed school climate through the sub-scales of disciplinary climate (rule clarity, perceived fairness of rules, consistency of application and disruptive behaviour), academic press, and parental involvement in learning. Disciplinary climate had the most consistent significant effects on achievement in all areas other than reading (controlling for factors such as gender, ethnicity, family structure, and socio-economic status). Using a longitudinal design, Esposito (1999) found that for children from low-income families, school climate predicted achievement in mathematics and reading (assessed by standardized tests and controlling for parental education and socio-economic status).
There is much evidence, then, supporting the relationship between school climate and academic achievement. This work, though, also has limitations (Anderson, 1982; Cohen, McCabe, Michelli, & Pickeral, 2009; Thapa et al., 2013). There is conceptual overlap between the school climate construct and other factors such as school belonging or school connectedness (Blum, 2005; Wilson, 2004) and these factors are also related to academic achievement. Bryan et al., (2012) found that attachment to school had a direct effect on mathematics achievement scores. Niehaus, Rudasill, and Rakes (2012) examined sixth graders in high poverty neighborhoods and found that students’ perception of school support declined across sixth grade. However, those sixth graders who maintained school connection had higher academic achievement. It is unclear how these bodies of research on connectedness and school climate fit together because the constructs are rarely investigated jointly.
It is also unclear how school climate impacts student behaviour such as achievement. The design of effective and efficient interventions to improve school climate is aided through more precise understandings of how and when school climate (e.g., practices, policies) impacts student functioning (e.g., engagement, well-being, achievement). In the current study, a social psychological analysis of school identification is introduced which serves to clarify the relationship between school climate and belonging/connectedness, and helps explain how school climate factors impact individual student achievement.
How does school climate affect achievement? The role of school identification
Psychological theories underpin explanations of how school factors and the subjective experience of school life impact student functioning. Drawing on developmental and motivational accounts, it is argued that a positive school climate enables the formation of strong interpersonal attachments and emotional bonds between students and care-taker adults (e.g., Comer & Emmons, 2006), feelings of belonging (Baumeister & Leary, 1995) and the satisfaction of human needs for relatedness (e.g., Deci, Vallerand, Pelletier, & Ryan, 1991). Finn (1989) also argues that student success at school is related to a sense of belongingness (i.e., bonding, attachment, affiliation) and a valuing of school-relevant goals (see also Goodenow, 1992; Osterman, 2000). So to the degree that school climate affords strong relationships and social bonds, there also should be more positive student experiences, including higher academic achievement.
In social psychology, particularly the social identity perspective, which refers to social identity theory (SIT; Tajfel & Turner, 1979) and self-categorization theory (SCT; Turner et al., 1987), terms such as ‘affiliation’, ‘belonging’, and ‘emotional attachment’ are also used, but the underpinning psychological processes are connected to the psychology of group membership. It is argued that individual psychology (and behaviour) is transformed as self-categorization shifts from the level of the personal identity (‘I’ and ‘me’), towards identifying as a group member (social identity as ‘we’ and ‘us’), so that ‘I’ now act in line with ‘our’ shared goals, values and interests (Turner et al., 1987).
When people identify with a particular group, the norms, values, and beliefs that define the group are internalized (there is self-stereotyping in relation to this ‘meaning’ or ‘content’). As a result the stronger one’s identification with a particular group (and its salience in the context of interest), the more likely it is that one will behave in line with the norms, values, and beliefs that define the group (e.g., following school rules; Reynolds, Subasic, & Tindall, 2015). The school identification construct can help explain how the characteristics of the group can come to shape the behaviour of group members.
Translating these ideas to the school context, school climate can be conceptualized as capturing people’s assessments of the qualities or characteristics of the school as a group. School identification relates to the cognitive and emotional significance of this group, for the individual student. Most importantly, though, it is argued that psychological attachment to the school as a group and associated internalization of school norms could be considered as a process or mechanism that underlies how school climate comes to affect individual student emotions, attitudes, and behaviors (e.g., Bizumic et al., 2009; Loukas et al., 2006; Reynolds, Subasic, Bromhead, & Lee, 2017; Reynolds, Subasic, Lee, Bromhead, & Tindall, 2015). As a result, school (social) identification takes on new importance and could help integrate work on connectedness, bonding, relatedness, and attachment through the theoretical analysis of the psychological group. It also offers an explanation as to how it is that school climate impacts school outcomes such as academic achievement.
Drawing links between school climate, school belonging (attachment, relatedness), and the analysis of the psychological group provides a more integrated theoretical analysis where school climate and school identification are related but distinct concepts, each playing a different role in explaining achievement. Bizumic and colleagues (2009), for example, examined the school outcomes of student engagement, well-being, and peer aggression. They found that school identification was a significant predictor of these outcome variables over and above demographic factors and school climate. Furthermore, they found that school identification was a significant mediator of the relationship between school climate and these outcome variables. There was limited evidence for social identification as a moderator. It was not the case that the impact of school climate on school outcomes was significant only for high identifiers compared to low identifiers. The results, rather, provided evidence that school identification was a general mediating mechanism between school climate and school outcomes.
There has been limited work, though, looking at social identity ideas in the context of achievement. Most of the work in organizational and educational contexts has focused on constructs such as trust, respect, and well-being rather than productivity and performance (achievement; Haslam, 2004). Some work in the university context is encouraging, showing that deep approaches to learning (when a student is motivated to understand rather than to simply reproduce the material presented), were positively related to student identity, and social identity positively predicted academic achievement (Bliuc, Ellis, Goodyear, & Hendres; 2011; Smyth, Mavor, Platow, Grace, & Reynolds, 2013). Building on these findings, the present study is novel in a number of respects including investigating: (a) the role of both school climate and school identification (bonding, connectedness) in explaining academic achievement; (b) whether school climate impacts achievement through school identification; and (c) achievement using a standardized measure across three key learning domains.
The present study
As is outlined in Figure 1, it is expected that: 1) school climate (path a) and school identification (path c) will be positively associated with academic achievement having taken into account other critical factors such as socio-economic status and parental education; 2) positive school climate will explain stronger school identification (path b); and, 3) the association between school climate and academic achievement will be significantly mediated by school identification through an indirect effect (path d).
Method
Participants
The participants in this study were 340 Australian students in grades seven and nine from two schools in a city with a population of approximately 350,000 as part of a broader project (Reynolds, Bizumic, Subasic, Melsom, & McGregor, 2007). The schools participated voluntarily and the sample comprised 60% of the enrolled students in the relevant grades. The response rate reflects student absences on the survey days and NAPLAN testing days and difficulties matching students across the two phases and two separate data sets (i.e., survey and administration data). Within the sample, 19% and 28% of the students at each school spoke languages other than English at home compared to the overall Australian mean of 18% (Australian Bureau of Statistics, 2013). The students’ mean age was 13.72 years with 1.10 standard deviation (SD) and ranged from 11- to 16-years-old. The mean time students had spent at the particular school was 2.61 years with 1.06 SD with a range from over one year to less than nine years. A slight majority of students were female (N = 189, 55.6%). The questionnaire data was matched to school records of parental education, socio-economic status and academic achievement on NAPLAN for reading, writing, and numeracy assessed in May each year.
Materials and procedure
A questionnaire was administered to all the students present during scheduled class times in the last term of the school year. In line with national guidelines, the ethics review committee did not require parental consent given the low risk nature of the research and students being deemed capable of offering consent (see Section 4.2.9 National Health and Medical Research Council Guidelines). The research team and teachers assisted the students and the questionnaire included a range of other measures (see Bizumic et al., 2009). All of the psychological measures required the student to indicate the level to which they agreed or disagreed with the statement on a Likert scale from 1 (Disagree strongly) to 7 (Agree strongly). The constructs relevant to the current research and demographics were as follows:
School climate (α = 0.75)1 included nine items concerning general school climate of shared values and approach related to rules and academic learning (e.g., ‘Students and staff have similar views about what being at this school should be like’, ‘When students misbehave and act-up staff tend to respond in the same kinds of ways’, ‘Staff work hard to make sure students learn and improve’). The measure of school climate was developed for this project drawing on relevant previous research (see Bizumic et al., 2009; Bizumic, Reynolds, & Meyer, 2012). A confirmatory factor analysis (CFA) showed that the nine items significantly indicated one school climate construct with factor loadings from 0.316 to 0.698. The CFA model fit was χ2 (23, N = 661) = 36.604; Comparative Fit Index (CFI) = 0.986; Tucker-Lewis Index (TLI) = 0.978; root mean squared error of approximation (RMSEA) = 0.030; the maximum likelihood (ML)-based standardized root mean squared residual (SRMR) = 0.026. A two-factor model and three-factor model were also tested but Chi-square difference test showed one factor was superior. The two models were also inferior in terms of model fit and cross-loadings of the items.
School identification (four items, α = 0.90)2 was assessed by items such as ‘Being part of this school is important to me’ and ‘I identify with this school’ (based on Doosje, Ellemers, & Spears, 1995; Haslam, Oakes, Reynolds, & Turner, 1999). Although school identification can incorporate a range of sub-factors (commitment, self-esteem e.g., Ellemers, Kortekaas, & Ouwerkerk, 1999), the current measure is widely used and correlates with other measures (Postmes, Haslam, & Jans, 2012). In the current study it was also validated with the measurement sub-model in the main structural equation modelling (SEM).
Student demographic information included gender, age, and the number of years they had attended the target school.
The student questionnaire data was combined with administrative school record data including:
Socio-economic status (SES) using the Index of Relative Socioeconomic Disadvantage (IRSD) from The Australian Bureau of Statistics (ABS). This index measures the relative social and economic hardship in the area each student lives (Newman & Kopras, 2004) and a score is given to each student (Australian Bureau of Statistics, 2009). This is a standardized relative measure with a mean of 1,000 and standard deviation of 100.
Parental education, with a university degree and above coded 1 and all other forms of education (e.g. Grade 12 certificate, trade qualifications, other types of certificates) coded 0.
NAPLAN scores assessing students’ academic achievement for reading, writing, and numeracy. The NAPLAN scale ranges from 0 to 1000 score. The tests showed good reliability (for grade seven for reading α = 0.87, writing 0.96, numeracy 0.92 and for grade nine, 0.89, 0.96, 0.93 respectively). The validity continues to be tested (Grasby, Byrne, & Olson, 2015; Harris et al., 2013). (Australian Curriculum, Assessment and Reporting Authority, 2012).3
Analytical strategy
First, data cleaning, missing data analysis and descriptive analyses were done with IBM SPSS while Mplus 6.0 was used to conduct Structural Equation Modelling (SEM).4 Second, mediation analyses of the three main achievement variables were conducted with SEM. The other covariate variables of age, gender, and the length of enrolment at the school were controlled. In total four paths for each mediation model were estimated and interpreted (MacKinnon, 2007). As outlined in Figure 2 the paths were as follows: (a) the direct path from school climate to achievement in domains of reading, numeracy and writing; (b) the path from school climate to school identification; (c) the path from school identification to achievement; and (d) the indirect path from school climate to achievement through school identification. The significance of the indirect path provides evidence for the mediation of school identification. Recent statistical developments indicate that complete mediation can be inferred if path b and c are significant with a significant indirect effect (path d) and a non-significant direct path c. If the direct path was not significant, the significant indirect path was interpreted as evidence for complete mediation. But if the direct path was also significant, then the significant indirect path was interpreted as evidence of partial mediation (MacKinnon, 2007, p. 70).
Figure 2. Structural Equation Modelling for numeracy showing the mediation effect of school identification with the inclusion of other individual-level covariates. SCHOOL IDENT: School Identification. SID: School Identification item. SES: socio-economic status. All error terms and school identification item factor loadings are omitted for simplicity. *p < .05, **p < .01.
We modelled each of the three achievement domains of reading, writing, and numeracy separately, because domain specificity has been observed in the literature (Hinnant, O’Brien, & Ghazarian, 2009; Marsh, Trautwein, Lüdtke, Köller, & Baumert, 2005) with specific effects of predictors (e.g., gender, school socio-economic status) on different subject domains such as science, language, and math (e.g., Ma & Klinger, 2000). The covariates of age, gender, the number of years they had attended the school, SES, parental education were controlled as covariate predictors of achievement. Moreover, applying SEM for the mediation models, the variance and covariance matrices of all the study variables were estimated.
Results
The means for academic achievement scores of reading, writing, and numeracy are presented in Table 1 along with the correlations between variables. Covariate variables mostly showed small correlations for each domain of reading, writing, and numeracy. School climate was not significantly correlated with academic achievement scores, so possible indirect effects via a mediator variable (Hayes, 2009, p. 415) were investigated, in line with hypotheses. It was the case that school identification showed a significant positive relationship with numeracy and writing. In addition, there was a strong intercorrelation between school climate and school identification (r = 0.57, p < 0.01) in line with previous research that has also shown these constructs to be related, but distinct (Bizumic et al., 2009).
|
Table 1. Descriptive statistics and correlations.
Investigating school identification as the mechanism through which school
Climate affects academic achievement
A mediation SEM was conducted to examine the relations between school climate, school identification, and reading scores. The four-item school identification construct was examined with a measurement model estimating the item factor loadings. School climate was examined as observed scale scores given limitations with the sample size, but its construct validity with the nine items had already been confirmed through CFA as described above. The known covariates of age, gender, years spent at the current school, SES (IRSD), and parental education were also included in the model. To accommodate non-normality issues in the psychological variables, the MLM was utilized with the Mplus option. MLM, also known as Satorra-Bentler chi-square, refers to maximum likelihood parameter estimates with standard errors and a mean-adjusted chi-square test statistic that are robust to non-normality (Muthén & Muthén, 1998–2010, p. 533).
Results for reading also showed that the model fit to the data was satisfactory χ2 (28, N = 286) = 43.345; CFI = 0.981; TLI = 0.973; RMSEA = 0.044; SRMR = 0.040) but neither school climate nor school identification showed any significant effects. Certain covariates significantly predicted reading scores. Reading scores were higher for older students (ß = 0.237, p < 0.01), newer students to the current school (ß = −.137, p < 0.01), students who had a more advantaged socio-economic status (ß = .237, p < 0.01), and students whose parents had higher education levels (ß = 0.237, p < 0.01).
For numeracy the model fit also was satisfactory χ2 (28, N = 286) = 47.442; CFI = 0.976; TLI = 0.965; RMSEA = 0.049; SRMR = 0.041) and explained 16.2% of the variance in students’ scores (see Figure 2). The MLM estimator was used and significant effects were evident for the covariates of age (ß = 0.121, p < 0.05), SES (ß = .151, p < 0.01), and parental education (ß = 0.258, p < 0.01). School climate did not significantly predict numeracy but there was a significant relationship with school identification (ß = 0.587, p < 0.01). Results indicated that school climate significantly predicted numeracy achievement through school identification (ß = 0.217, p < 0.01). The results indicated an indirect effect from school climate through school identification on numeracy scores (ß = 0.127, p < 0.01).
Students’ writing scores were examined in a mediation SEM and the model fitted the data well χ2 (28, N = 286) = 42.597; CFI = 0.982; TLI = 0.974; RMSEA = 0.043; SRMR = 0.040). Higher results were obtained for female students (ß = 0.229, p < 0.01) and older students (ß = 0.129, p < 0.05). Students with a more advantaged socio-economic status and more educated parents were more likely to score higher (ß = 0.211, 0.159 respectively, p < 0.01). School climate did not affect writing scores but there was a significant relationship with school identification (ß = 0.587, p < 0.01). Interestingly, school identification showed a stronger effect on writing than the parental education covariate. The indirect effect from school climate to writing (β = .163, p < 0.01) through school identification was significant (ß = 0.104, p < 0.01) and 15.3% of the variance in writing scores was explained by this mediation model.
Discussion
School identification specifically relates to one’s psychological connection and self-definition as a school member (Bizumic et al., 2009). It was predicted that school identification would be significantly related to academic achievement and perhaps, mediate the relationship between school climate and achievement. In two learning domains, numeracy and writing, the hypothesized relationships were all supported. Importantly this research is one of only a few studies to assess the relationship between school climate and academic achievement using a standardized measure. The fact that NAPLAN is a national measure of achievement strengthens the importance of the current findings.
The three variables most significantly associated with achievement were parental education, socio-economic status, and school identification. The results for parental education and socio-economic status (IRSD) replicated established findings (see Thapa et al., 2013). More specifically, numeracy performance was positively related to age, socio-economic status, parental education, and school identification. Furthermore, while no direct effect was found for school climate, complete mediation was evident through school identification with a significant indirect effect on numeracy scores. Likewise, for writing, better performance was related to being female, being older, higher socio-economic status, and having more educated parents as well as school identification. Again, there was no direct effect of school climate, but a positive indirect effect emerged through school identification, indicating complete mediation in explaining writing scores. Reading performance was positively related to age, tenure at school, parental education, and socio-economic status. Yet no effects were evident for school climate or school identification in explaining reading scores. In previous research, reading has been found to be less affected by school climate (e.g., Ma & Klinger, 2000).
School identification emerges as an important predictor of academic achievement both directly and by creating an indirect effect of school climate on achievement. There is evidence that identification with the school accounted for significant variance in academic achievement for numeracy and writing with medium to large effect sizes (Cohen, 1992), even after taking into account a large number of variables previously identified as critical to academic success. School life and perceptions of the climate of the school are most likely to impact student achievement through the psychological mechanism where students also identify with the school as a meaningful and self-defining reference group. These findings should encourage further integrated work on school climate and school identification, which captures school belonging and connectedness dimensions (Bryan et al., 2012; Loukas et al., 2006; Niehaus et al., 2012).
A distinctive feature of the current research is that on theoretical grounds school climate and school identification were explored as intertwined, but separate conceptual constructs (Bizumic et al., 2009). Although providing conceptual clarity, this strategy may have weakened the association between the school climate measure used in this research and academic achievement. A closer examination of previous work reveals that in many cases it was only specific school climate sub-factors that were significantly related to achievement. For example, Stewart (2003) found a small correlation with grade point average for school cohesion, however, the school attachment factor did not show any significant correlation. For Ma and Klinger (2000), disciplinary climate emerged as being particularly important as a predictor of academic achievement. There is more work to be done in identifying which aspects of school climate may be particularly important in explaining achievement.
A direct implication of this research is that researchers, principals, and policy makers will have new insights into the role of the group and social identification in affecting students’ attitudes and behaviour. The message is that strengthening students’ psychological connection to the school is important in improving learning. An important first step is clarifying and strengthening the norms, values, and goals of the school as a group–who we are, what we do, why we do it, and what makes us special (Reynolds, Subasic, Bromhead, & Lee, 2017). Identification can also be enhanced through the use of a process where student groups, along with other sub-groups within the school, feel they have a voice and their views are valued and respected (Haslam, Eggins, & Reynolds, 2003; Tyler & Blader, 2000). It is also the case that warm and respectful socio-emotional relationships between students and teachers can build students’ psychological connection to the school (Lee, Reynolds, Subasic, & Bromhead, 2016). More research is needed investigating specific school-based strategies that aim to build a positive climate and school identification.
Limitations and future directions
The major limitations of this study lies in the sample not being representative and its cross-sectional design which prohibits an assessment of the direction of association between variables. Longitudinal models are needed, which can attempt to demonstrate the causal relationships between variables. Moreover, examining school-level variables using a non-nested design is not ideal given that students from one school are likely to be more similar than students from another school. A multi-level analysis was not possible in the current study given that only two schools were involved, however it has been suggested that a non-nested design can lead to an underestimation of school-level effects (Lee, 2000). The inclusion of a larger sample of schools is an important next step in this program of research.
In this research a general unidimensional measure of school climate was used that included items assessing shared values, rule clarity, and consensus in implementation and teacher support for learning, but there are many sub-factors that comprise school climate. In future research, other measures of school climate should also be incorporated and investigated using similar models. Ideally school climate should not be assessed via the inclusion of sub-scales or as a general overall construct, but through the use of measurement sub-models within SEM (Brand, Felner, Shim, Seitsinger, & Dumas, 2003; Lee, Reynolds, Subasic, Bromhead, Lin, Marinov, & Smithson, 2017).
Counter to previous work we did not find a relationship between school climate and academic achievement. The reason may be that school climate and school identification were measured separately. If the school identification items were included within the school climate construct, a stronger climate and achievement relationship may have been found, but the critical explanatory role of school identification would have been overlooked. Other mediation-moderator models (e.g., gender, age, socio-economic status) can also be explored which could more precisely identify the conditions under which school climate will be effective in bringing about change in student learning and behaviour at school.
A strength of the current research is the inclusion of a nationally administered measure of academic achievement (NAPLAN). There is widespread agreement that NAPLAN is an effective measure of student ability in reading, writing, and numeracy. Yet it should also be pointed out that there is emerging evidence that NAPLAN scores in the Australian Capital Territory (ACT) in Australia may be skewed towards the upper end of academic achievement level (Ferrari, 2011). Thus, the associations in this data may only apply to relatively high achieving students and needs to be examined in more varied contexts, with a broader range of participants in larger samples.
Despite these limitations, the results of the present study offer evidence for one possible mechanism through which schools can influence their students’ academic success. School identification does appear to be playing an important and significant role in adding additional explanatory value over traditional measures in understanding academic achievement. Reynolds et al. (2017) also call for more systematic research exploring whether learning itself is a social influence process which depends on the information source (teachers) being perceived as a member of the ingroup (one of ‘us’). The social identity perspective with its emphasis on group processes opens up a range of new avenues for research and practice.
Conclusion
Understanding what propels students towards achievement—and what may hinder their progress—is a central concern for education research. The current article contributes to this endeavour by highlighting the central role that school identification plays in fostering achievement both independently and in mediating the impact of school climate. The results suggest that school identification added significant explanatory power over more traditionally studied variables in school research such as age, gender, years at school, parental education, socio-economic status, and school climate. Educational reform programs and practices may benefit from an approach that serves to build the psychological connection between school members. It is hoped that a more systematic integration of group processes and associated social identity within the educational domain will open up new avenues to understanding school-relevant behaviours and also help shape such behaviours. Clarifying the definitions of who ‘we’ are and strengthening connection to ‘us’ as a school group are fundamental to academic success.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Australian Research Council and Directorate of Education and Training through the Linkage grant scheme (LP0883652).
Notes
1
Average Variance Extracted (AVE = 0.35) and Composite Reliability (CR = 0.29) were lower than the recommended threshold (Bagozzi & Yi, 1988; Fornell & Larcker, 1981), yet other model fit indices in the CFA were satisfactory.
2
AVE = 0.68; CR = 0.86.
3
The national mean and standard deviations (in brackets) for grade 7 in 2010 were for reading 546.0 (68.4), writing, 533.5 (72.9), numeracy, 547.8 (72.4). For grade 9 the results were 573.7 (66.2), 567.7 (81.4), 585.1 (70.4) respectively. The sample means for the two grades were higher and the SDs smaller than the national statistics.
4
Missing data analyses showed a missing rate less than 3% ranging from 0.6 % to 2.6% for school climate and school identification items. The missing rates were very low and missed completely at random. Little’s MCAR tests were all non-significant (p = 0.453 ∼ 0.989) confirming there was no systematic missing patterns in the data. To secure better estimates and include as many cases as possible for SEM analyses the school climate and identification missing items were imputed with the Expectation Maximisation (EM, Gold & Bentler, 2000) method with SPSS 20. Missing data imputation is the standard technique for SEM and provides less biased estimation of the coefficients, compared to deleting cases with missing data (Baraldi & Enders, 2010; Little & Rubin, 2014; Peugh, & Enders, 2004). Meanwhile the missing cases of the demographic covariates of age, gender, and the years spent at the school were deleted list-wise in all the analyses.
References
| Anderson, C. S. (1982) The search for school climate: A review of the research. Review of Educational Research 52: 368–420. doi: 10.1177/0013131X7300900204. Google Scholar | SAGE Journals | ISI | |
| Australian Bureau of Statistics. (2009). An introduction to socio-economic indexes for areas (SEIFA) 2006. (Information Paper - 2039.0). Retrieved from http://www.abs.gov.au/ausstats/abs@.nsf/mf/2039.0/. Google Scholar | |
| Australian Bureau of Statistics (2013). National regional profile: Australian Capital Territory. Retrieved from http://www.abs.gov.au/AUSSTATS/abs@nrp.nsf/Latestproducts/8Population/People12007-2011?opendocument&tabname=Summary&prodno=8&issue=2007-2011. Google Scholar | |
| Australian Curriculum, Assessment and Reporting Authority. (2012). National Assessment Program: Literacy and Numeracy. Retrieved from http://www.naplan.edu.au/. Google Scholar | |
| Baraldi, A. N., Enders, C. K. (2010) An introduction to modern missing data analyses. Journal of School Psychology 48(1): 5–37. doi: 10.1016/j.jsp.2009.10.001. Google Scholar | Crossref | Medline | ISI | |
| Baumeister, R. F., Leary, M. R. (1995) The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin 117(3): 497. doi: 10.1037/0033-2909.117.3.497. Google Scholar | Crossref | Medline | ISI | |
| Bliuc, A. M., Ellis, R. A., Goodyear, P., Hendres, D. M. (2011) Understanding student learning in context: Relationships between social identity, perceptions of the learning community, approaches to learning and academic performance. European Journal of Psychology of Education 23(3): 417–433. doi: 10.1007/s10212-011-0065-6. Google Scholar | Crossref | ISI | |
| Blum, R. W. (2005) A case for school connectedness. The Adolescent Learner 62(7): 16–20. Google Scholar | |
| Bizumic, B., Reynolds, K. J., Meyers, B. (2012) Predicting social identification over time: The role of group and personality factors. Personality and Individual Differences 53(4): 453–458. doi: 10.1016/j.paid.2012.04.009. Google Scholar | Crossref | ISI | |
| Bizumic, B., Reynolds, K. J., Turner, J. C., Bromhead, D., Subasic, E. (2009) The role of the group in individual functioning: School identification and the psychological well-being of staff and students. Applied Psychology: An International Review 58(1): 171–192. doi: 10.1111/j.1464-0597.2008.00387.x. Google Scholar | Crossref | ISI | |
| Brand, S., Felner, R., Shim, M., Seitsinger, A., Dumas, N. (2003) Middle school improvement and reform: Development and validation of a school-level assessment of climate, culture pluralism, and school safety. Journal of Educational Psychology 95(3): 570–588. doi: 10.1037/0022-0663.95.3.570. Google Scholar | Crossref | ISI | |
| Bryan, J., Moore-Thomas, C., Gaenzle, S., Kim, J., Lin, C. H., Na, G. (2012) The effects of school bonding on high school seniors’ academic achievement. Journal of Counselling and Development 90(4): 467–480. doi: 10.1002/j.1556-6676.2012.00058.x. Google Scholar | Crossref | ISI | |
| Cohen, J. (1992) A power primer. Psychological Bulletin 112(1): 155. doi: 10.1037/0033-2909.112.1.155. Google Scholar | Crossref | Medline | ISI | |
| Cohen, J., McCabe, L., Michelli, N. M., Pickeral, T. (2009) School climate: Research, policy, practice, and teacher education. The Teachers College Record 111(1): 180–213. Retrieved from https://schoolclimate.org/climate/documents/policy/School-Climate-Paper-TC-Record.pdf. Google Scholar | ISI | |
| Comer, J. P., Emmons, C. (2006) The research program of the Yale child study centre school development program. The Journal of Negro Education 75(3): 353–372. Retrieved from http://www.jstor.org/stable/40026808. Google Scholar | |
| Crosnoe, R., Johnson, M. K., Elder, G. H. (2004) School size and the interpersonal side of education: An examination of race/ethnicity and organizational context. Social Science Quarterly 85(5): 1259–1274. doi: 10.1111/j.0038-4941.2004.00275.x. Google Scholar | Crossref | ISI | |
| Deci, E. L., Vallerand, R. J., Pelletier, L. G., Ryan, R. M. (1991) Motivation and education: The self-determination perspective. Educational Psychologist 26(3–4): 325–346. doi: 10.1207/s15326985ep2603&4_6. Google Scholar | Crossref | ISI | |
| Doosje, B., Ellemers, N., Spears, R. (1995) Perceived intragroup variability as a function of group status and identification. Journal of Experimental Social Psychology 31(5): 410–436. https://doi.org/10.1006/jesp.1995.1018. Google Scholar | Crossref | ISI | |
| Ellemers, N., Kortekaas, P., Ouwerkerk, J. W. (1999) Self-categorisation, commitment to the group and group self-esteem as related but distinct aspects of social identity. European Journal of Social Psychology 29(2–3): 371–389. doi: 10.1002/(SICI)1099-0992(199903/05)29:2/3<371::AID-EJSP932>3.0.CO;2-U. Google Scholar | Crossref | ISI | |
| Esposito, C. (1999) Learning in urban blights: School climate and its effect on the school performance of urban, minority, low-income children. School Psychology Review 28(3): 365–377. Retrieved from http://search.proquest.com/docview/219648298?pq-origsite=gscholar. Google Scholar | ISI | |
| Ferrari, J. (2011, October 15-16). NAPLAN no-shows ‘skewing results’. The Weekend Australian, p. 5. Google Scholar | |
| Finn, J. D. (1989) Withdrawing from school. Review of Educational Research 59(2): 117–142. doi: 10.3102/00346543059002117. Google Scholar | SAGE Journals | ISI | |
| Goodenow, C. (1992) Strengthening the links between educational psychology and the study of social contexts. Educational Psychologist 27(2): 177–196. doi: 10.1207/s15326985ep2702_4. Google Scholar | Crossref | ISI | |
| Gold, M. S., Bentler, P. M. (2000) Treatments of missing data: A Monte Carlo comparison of RBHDI, iterative stochastic regression imputation, and expectation-maximization. Structural Equation Modeling 7(3): 319–355. doi: 10.1207/s15328007sem0703_1. Google Scholar | Crossref | ISI | |
| Grasby, K. L., Byrne, B., Olson, R. K. (2015) Validity of large-scale reading tests: A phenotypic and behaviour–genetic analysis. Australian Journal of Education 59(1): 5–21. doi: 10.1177/0004944114563775. Google Scholar | SAGE Journals | ISI | |
| Harris, P., Chinnappan, M., Castleton, G., Carter, J., De Courcy, M., Barnett, J. (2013) Impact and consequence of Australia’s National Assessment Program-Literacy and Numeracy (NAPLAN)-using research evidence to inform improvement. TESOL in Context 23(1/2): 30–52. Availability: http://search.informit.com.au/documentSummary;dn=885404083796362;res=IELAPA. Google Scholar | |
| Haslam, S. A. (2001; 2nd ed. 2004). Psychology in organizations: The social identity approach. London and Thousand Oaks, CA: Sage. Google Scholar | |
| Haslam, S. A., Eggins, R. A., Reynolds, K. J. (2003) The ASPIRe model: Actualizing social and personal identity resources to enhance organizational outcomes. Journal of Occupational and Organizational Psychology 76(1): 83–113. doi: 10.1348/096317903321208907. Google Scholar | Crossref | ISI | |
| Haslam, S. A., Oakes, P. J., Reynolds, K. J., Turner, J. C. (1999) Social identity salience and the emergence of stereotype consensus. Personality and Social Psychology Bulletin 25(7): 809–818. doi: 10.1177/0146167299025007004. Google Scholar | SAGE Journals | ISI | |
| Hattie, J. (2013). Visible learning: A synthesis of over 800 meta-analyses relating to achievement. Routledge. https://doi.org/10.4324/9780203887332. Google Scholar | |
| Hayes, A. F. (2009) Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs 76(4): 408–420. doi: 10.1080/03637750903310360. Google Scholar | Crossref | ISI | |
| Hinnant, B. J., O’Brien, M., Ghazarian, S. R. (2009) The longitudinal relations of teacher expectations to achievement in the early school years. Journal of Educational Psychology 101(3): 662–670. http://doi.org/10.1037/a0014306. Google Scholar | Crossref | Medline | ISI | |
| Hopson, L. M., Lee, E. (2011) Mitigating the effect of family poverty on academic and behavioral outcomes: The role of school climate in middle and high school. Children and Youth Services Review 33(11): 2221–2229. http://dx.doi.org/10.1016/j.childyouth.2011.07.006. Google Scholar | Crossref | ISI | |
| Hoy, W. K., Hannum, J. W. (1997) Middle school climate: An empirical assessment of organizational health and student achievement. Educational Administration Quarterly 33(3): 290–311. https://doi.org/10.1177/0013161x97033003003. Google Scholar | SAGE Journals | ISI | |
| Lee, V. E. (2000) Using hierarchical linear modelling to study social contexts: The case of school effects. Educational Psychologist 35(2): 125–141. https://doi.org/10.1207/s15326985ep3502_6. Google Scholar | Crossref | ISI | |
| Lee, E. Reynolds, K.J., Subasic E. & Bromhead, D. (2016). Explaining how the teacher-student relationship impacts on student engagement and well-being: The role of school identification. Submitted manuscript. Google Scholar | |
| Lee, E. Reynolds, K.J., Subasic E., Bromhead, D., Lin, H., Marinov, V., & Smithsom, M. (2017). Development of a dual School Climate and School Identification Measure–Student (SCASIM-St). Contemporary Educational Psychology. Google Scholar | |
| Little, R. J., Rubin, D. B. (2014) Statistical analysis with missing data, London: John Wiley & Sonsdoi: 10.1002/9781119013563. Google Scholar | |
| Loukas, A., Suzuki, R., Horton, K. D. (2006) Examining school connectedness as a mediator of school climate effects. Journal of Research on Adolescence 16(3): 491–502. doi: 10.1111/j.1532-7795.2006.00504.x. Google Scholar | Crossref | ISI | |
| Ma, X., Klinger, D. A. (2000) Hierarchical linear modelling of student and school effects on academic achievement. Canadian Journal of Education 25(1): 41–55. https://doi.org/10.2307/1585867. Google Scholar | Crossref | |
| MacKinnon, D. (2007). Introduction to statistical mediation analysis. CRC Press. Google Scholar | |
| Marsh, H. W., Trautwein, U., Lüdtke, O., Köller, O., Baumert, J. (2005) Academic self-concept, interest, grades, and standardized test scores: Reciprocal effects models of causal ordering. Child Development 76(2): 397–416. https://doi.org/10.1111/j.1467-8624.2005.00853.x. Google Scholar | Crossref | Medline | ISI | |
| McEvoy, A., Welker, R. (2000) Antisocial behavior, academic failure, and school climate: A critical review. Journal of Emotional and Behavioral Disorders 8(3): 130–140. doi:10.1177/106342660000800301. Google Scholar | SAGE Journals | ISI | |
| Moos, R. H. (1978) A typology of junior high and high school classrooms. American Educational Research Journal 15(1): 53–66. https://doi.org/10.3102/00028312015001053. Google Scholar | SAGE Journals | ISI | |
| Moos, R. H. (1987) Person-environment congruence in work, school, and health care settings. Journal of Vocational Behaviour 31(3): 231–247. https://doi.org/10.1016/0001-8791(87)90041-8. Google Scholar | Crossref | ISI | |
| Moos, R. H., Moos, B. S. (1978) Classroom social climate and student absences and grades. Journal of Educational Psychology 70(2): 263–269. doi: 10.1037/0022-0663.70.2.263. Google Scholar | Crossref | ISI | |
| Muthén, B., Muthén, L. (1998–2010) Mplus version 6.1: User’s guide, Los Angeles, CA: Muthén & MuthénRetrieved from https://www.statmodel.com/download/usersguide/Mplus%20Users%20Guide%20v6.pdf. Google Scholar | |
| Newman, G., & Kopras, A. (2004). Socio-economic indexes for electoral divisions: 2001 Census (2003 boundaries). Department of Parliamentary Services. Retrieved from http://202.14.81.88/binaries/library/pubs/cib/2004-05/05cib01.pdf. Google Scholar | |
| Niehaus, K., Rudasill, K. M., Rakes, C. R. (2012) A longitudinal study of school connectedness and academic outcomes across sixth grade. Journal of School Psychology 50(4): 443–460. https://doi.org/10.1016/j.jsp.2012.03.002. Google Scholar | Crossref | Medline | ISI | |
| Osterman, K. (2000) Students’ need for belonging in the school community. Review of Educational Research 70(3): 323–367. https://doi.org/10.3102/00346543070003323. Google Scholar | SAGE Journals | ISI | |
| O’Malley, M., Voight, A., Renshaw, T. L., Eklund, K. (2015) School climate, family structure, and academic achievement: A study of moderation effects. School Psychology Quarterly 30(1): 142–157. https://doi.org/10.1037/spq0000076. Google Scholar | Crossref | Medline | ISI | |
| Peugh, J. L., Enders, C. K. (2004) Missing data in educational research: A review of reporting practices and suggestions for improvement. Review of Educational Research 74(4): 525–556. https://doi.org/10.3102/00346543074004525. Google Scholar | SAGE Journals | ISI | |
| Postmes, T. T., Haslam, S. A., Jans, L. (2012) A single-item measure of social identification: Reliability, validity, and utility. British Journal of Social Psychology 52(4): 597–617. https://doi.org/10.1111/bjso.12006. Google Scholar | Crossref | Medline | ISI | |
| Ramelow, D., Currie, D., Felder-Puig, R. (2015) The assessment of school climate review and appraisal of published student-report measures. Journal of Psychoeducational Assessment 33(8): 731–743. https://doi.org/10.1177/0734282915584852. Google Scholar | SAGE Journals | ISI | |
| Reynolds, K. J., Bizumic, B., Subasic, E., Melsom, K., & MacGregor, F. (2007). Understanding the school as an intergroup system. Grant funded by the Australian Research Council. The Australian National University, Canberra. Google Scholar | |
| Reynolds, K. J., Subasic, E., Bromhead, D., & Lee, E. (2017). The school as a group system: School climate, school identity and school outcomes. In K. Mavor, M. J. Platow, & B. Bizumic (Eds). The self, social identity and education. (pp. 55–69). London: Psychology Press. Google Scholar | |
| Reynolds, K. J. Subasic, E., Lee. E., Bromhead, D., & Tindall, K. (2015). Does education really change us? The impact of social school processes on the person. In K. J. Reynolds, & N. Branscombe (Eds.) (2015) The Psychology of Change: Life Contexts, Experiences, and Identities (pp. 170–186). New York, NY: Psychology Press. https://doi.org/10.4324/9781315735160. Google Scholar | |
| Reynolds, K. J., Subasic, E., Tindall, K. (2015) The problem of behaviour change: From social norms to an ingroup focus. Social and Personality Psychology Compass 9: 45–56. doi: 10.1111/spc3.12155. Google Scholar | Crossref | ISI | |
| Roeser, R. W., Midgley, C., Urdan, T. C. (1996) Perceptions of the school psychological environment and early adolescents’ psychological and behavioural functioning in school: The mediating role of goals and belonging. Journal of Educational Psychology 88(3): 408–422. https://doi.org/10.1037//0022-0663.88.3.408. Google Scholar | Crossref | ISI | |
| 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 Education 4(1–2): 19–31. Google Scholar | |
| Smyth, L., Mavor, K. I., Platow, M., Grace, D., Reynolds, K. J. (2013) Discipline social identification, study norms and learning approach in university students. Educational Psychology 35: 53–72. https://doi.org/10.1080/01443410.2013.822962. Google Scholar | Crossref | ISI | |
| Stewart, E. A. (2003) School social bonds, school climate, and school misbehaviour: A multilevel analysis. Justice Quarterly 20(3): 575–604. http://dx.doi.org/10.1080/07418820300095621. Google Scholar | Crossref | ISI | |
| 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(2): 179–204. doi: 10.1177/0013124507304167. Google Scholar | SAGE Journals | ISI | |
| Tajfel, H., Turner, J. C. (1979) An integrative theory of intergroup conflict. In: Austin, W. G., Worchel, S. (eds) The social psychology of intergroup relations, Monterey, CA: Brooks & Cole, pp. 33–47. doi: 10.1146/annurev.ps.33.020182.000245. Google Scholar | |
| Thapa, A., Cohen, J., Guffey, S., Higgins-D'Alessandro, A. (2013) A review of school climate research. Review of Educational Research 83(2): 357–385. doi: 10.3102/0034654313483907. Google Scholar | SAGE Journals | ISI | |
| Turner, I., Reynolds, K. J., Lee, E., Subasic, E., Bromhead, D. (2014) Well-being, school climate and the social identity process: A latent growth model study of bullying perpetration and peer victimization. School Psychology Quarterly 29(3): 320–335. http://dx.doi.org/10.1037/spq0000074. Google Scholar | Crossref | Medline | ISI | |
| Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., Wetherell, M. S. (1987) Rediscovering the social group: A self-categorization theory, Oxford: Basil Blackwell. Google Scholar | |
| Tyler, T. R., Blader, S. L. (2000) Cooperation in groups: Procedural justice, social identity and behavioral engagement, Philadelphia, PA: Psychology Press. Google Scholar | |
| Wang, M. C., Haertel, G. D., Walberg, H. J. (1997) Learning influences. In: Walberg, H. J., Haertel, G. D. (eds) Psychology and educational practice, Berkley, CA: McCuthan, pp. 199–211. Google Scholar | |
| Wang, M. T., Holcombe, R. (2010) Adolescents’ perceptions of school environment, engagement, and academic achievement in middle school. American Educational Research Journal 47(3): 633–662. doi: 10.3102/0002831209361209. Google Scholar | SAGE Journals | ISI | |
| Wilson, D. (2004) The interface of school climate and school connectedness and relationships with aggression and victimization. Journal of School Health 74(7): 293–299. doi: 10.1111/j.1746-1561.2004.tb08286.x. Google Scholar | Crossref | Medline | ISI |
Author biographies
Katherine J. Reynolds is a Professor of Social Psychology at the Australian National University.
Eunro Lee is a Research Fellow in the Research School of Psychology at the Australian National University.
Isobel Turner recently completed the Doctor of Clinical Psychology in the Research School of Psychology at the Australian National University. Her dissertation focused on mental health and school factors in explaining peer aggression and victimization.
David Bromhead is the Senior Manager of Student Support within the ACT Education Directorate.
Emina Subasic is Lecturer at the University of Newcastle and recently helped the Australian Research Council funded early career fellowship.