Academic buoyancy and academic resilience: Exploring ‘everyday’ and ‘classic’ resilience in the face of academic adversity

Abstract

Academic buoyancy has been defined as a capacity to overcome setbacks, challenges, and difficulties that are part of everyday academic life. Academic resilience has been defined as a capacity to overcome acute and/or chronic adversity that is seen as a major threat to a student’s educational development. This study is the first to examine the extent to which (a) academic buoyancy and academic resilience are distinct (but correlated) factors, and (b) academic buoyancy is more relevant to low-level negative outcomes (anxiety, uncertain control, failure avoidance), whereas academic resilience is more relevant to major negative outcomes (self-handicapping, disengagement). The findings, based on 918 Australian high school students from nine schools, showed that academic buoyancy and academic resilience represented distinct factors sharing approximately 35% variance. Also, academic buoyancy was more salient in negatively predicting low-level negative outcomes whereas academic resilience was more salient in negatively predicting major negative outcomes. In supplementary analyses, the effect of academic buoyancy on low-level negative outcomes tended to be direct, whereas the effect of academic buoyancy on major negative outcomes was mediated by academic resilience. Implications for practice and research are discussed.

Keywords

Academic buoyancy academic resilience engagement motivation

Martin and Marsh (2009) defined academic buoyancy as an ability to deal with everyday academic setback and challenge (i.e. minor adversity; see also Putwain, Connors, Symes, & Douglas-Osborn, 2012) and academic resilience as an ability to deal with chronic and/or acute academic adversity (i.e. major adversity). Academic adversity increases the likelihood of disengagement from school and is associated with reduced educational achievement and attainment (Covington, 1992). ‘Failing students’ (MacDonald, 2007) have little or no access to the ‘structure of opportunities’ (Roberts, 1995) available to other students. As a result, they are systematically disconnected from adaptive school and post-school pathways (Martin & Marsh, 2009). Despite the theoretical importance of these constructs, empirical investigation of their differentiation, ordering, and relationship to salient outcomes is needed.

Differentiation and ordering of academic buoyancy and academic resilience

Martin and Marsh (2008a, 2008b, 2009) proposed detailed examples of how academic buoyancy and academic resilience may be differentiated from an applied perspective. For example, whereas academic resilience is relevant to chronic underachievement, academic buoyancy is relevant to isolated poor grades and patches of poor performance; whereas academic resilience is relevant to self-system debilitation in the face of chronic failure, academic buoyancy is relevant to threats to self-confidence as a result of negative feedback; whereas academic resilience is relevant to truancy and disaffection from school, academic buoyancy is relevant to dips in motivation and engagement; and, whereas academic resilience is relevant to alienation from school or opposition to teachers, academic buoyancy is relevant to relatively minor negative interactions with teachers (Martin & Marsh, 2008a, 2008b, 2009).

In addition to being differentiated as separate constructs from an applied perspective, contentions have also been made about the ordering of academic buoyancy and academic resilience. Martin and Marsh (2009) proposed that an ability to deal with everyday academic adversity should better position the student when more substantial adversity presents. This suggests an ordering of academic buoyancy and academic resilience such that academic buoyancy predicts academic resilience and academic resilience predicts major outcomes. In some ways this is in line with bottom-up approaches to self-system factors and processes. Such approaches argue for low-level, specific factors (e.g. self-concept) giving rise to broader constructs (e.g. self-esteem) (e.g. Shavelson, Hubner, & Stanton, 1976; Trautwein, 2003). The present study is an opportunity to test this in relation to academic buoyancy and resilience.

Minor and major negative academic phenomena

One contention about academic buoyancy and academic resilience is that they relate to qualitatively distinct academic outcomes. Academic buoyancy is relevant to reducing minor negative outcomes and academic resilience is relevant to reducing major negative outcomes. Recent multidimensional work has identified different levels of negative academic phenomena. Specifically, the Motivation and Engagement Wheel (Martin, 2007, 2009) is a framework that explicitly organizes negative academic dimensions in terms of the degree of their negative/maladaptive valence. In terms of low-level negative engagement, the Wheel comprises anxiety, failure avoidance, and uncertain control. These are referred to as ‘impeding’ motivation and engagement factors (Martin, 2007, 2009). In terms of major negative engagement, the Wheel comprises self-handicapping and disengagement. These are referred to as ‘maladaptive’ motivation and engagement factors (Martin, 2007, 2009).

The Wheel is assessed using the Motivation and Engagement Scale (MES; Martin, 2010). The MES consists of motivation and engagement subscales congruent with the factors in the Wheel (e.g. anxiety, failure avoidance, uncertain control, self-handicapping, and disengagement). Consistent with the bottom-up perspective on academic buoyancy and academic resilience (with the lower-level buoyancy hypothesized to underpin a more substantial resilience construct), these parts of the Wheel can be seen in similar ways, with the lower-level impeding factors underpinning more maladaptive factors in the form of self-handicapping and disengagement (e.g. Covington, 1992). To the extent that this alignment is the case, it may be hypothesized that the low-level predictors (buoyancy) and outcomes (impeding factors) will be more closely associated and the more substantial predictors (resilience) and outcomes (maladaptive factors) will be more closely associated. Indeed, some theorists have moved beyond bottom-up and top-down models to argue for horizontal models of self along similar lines (Marsh & Yeung, 1998).

Purpose of the study

Previous theoretical work has suggested a distinction between ‘everyday’ academic resilience and ‘classic’ academic resilience (Martin & Marsh, 2009). Within an Australian high school context, this investigation tests two central propositions of this theoretical work: (a) that academic buoyancy and academic resilience are distinct (but correlated) factors; and (b) that academic buoyancy should be more salient in negatively predicting low-level negative outcomes (impeding engagement in the form of anxiety, failure avoidance, uncertain control) whereas academic resilience should be more salient in negatively predicting major negative outcomes (maladaptive engagement in the form of self-handicapping and disengagement). A supplementary analysis explores a hypothesized ordering of academic buoyancy to academic resilience.

Method

Sample and procedure

A total of 918 high school students participated, ranging from junior high (11- to 14-years-old; 53% of the sample) to senior high (15- to 19-years-old; 47% of the sample). Students attended nine high schools in four major cities on the east coast of Australia. Participating schools were of mixed levels of achievement (though, slightly higher in socio-economic status and achievement than the national average). Four schools were co-educational, three were single-sex girls’ schools, and two were single-sex boys’ schools. The sample consisted of 42% females and 58% males. The average age was 14.54 (SD = 1.53) years. In total, 14% of students spoke a language other than English at home. With few exceptions, students in attendance on the day of the testing participated in the survey. Teachers administered the survey during class time. The survey and the rating scale were explained and then a sample item was presented. Students completed the survey on their own and they returned the survey at the end of class.

Students in this sample were selected on the basis of having experienced at least one form of major academic adversity in the past year (see Appendix in Supplemental Materials for academic adversity items; Academic Risk and Resilience Scale [ARRS]). Because academic resilience is defined in the presence of major academic adversity, the only students eligible to answer academic resilience items are those who have experienced major academic adversity. Thus, unlike studies involving academic buoyancy that do not need to screen the sample (because ‘everyday’ adversity is relevant to all students), studies of academic resilience will necessarily involve more selective samples.

Materials

Independent variables

There were two independent variables in the study: Academic buoyancy and academic resilience.

Academic buoyancy

Academic buoyancy was assessed using the Academic Buoyancy Scale (ABS; Martin & Marsh, 2008a, 2008b). The ABS comprises four items (e.g. ‘I’m good at dealing with setbacks at school–– e.g. negative feedback on my work, poor result’). The items are rated from 1 (‘Strongly Disagree’) to 7 (‘Strongly Agree’). Prior research has demonstrated unidimensionality, invariance as a function of age, ethnicity and gender, reliability, approximately normal distribution, and significant associations with numerous educational outcomes (Martin & Marsh, 2008a, 2008b; Putwain et al., 2012). The Academic Buoyancy Scale is available from the author on request.

Academic resilience

Academic resilience was assessed using the Academic Risk and Resilience Scale (ARRS) implemented for the first time in this study. Respondents were first asked to indicate ‘yes’ or ‘no’ to a series of major academic adversity items. These major adversities include repeating a grade, failing a subject, school suspension, school expulsion, a learning disability, and the like. They are drawn from factors identified in various literatures focusing on young people and academic risk (e.g. Coleman & Hagell, 2007; Finn & Rock, 1997; Lucio, Hunt, & Bornovalova, 2012; Martin & Marsh, 2009; Rutter, 2006). Students answering ‘yes’ to one or more of these academic adversities were then asked four academic resilience items that related to these major adversities (e.g. ‘I’m good at dealing with these types of setbacks’). Items were rated on a 1 (‘Strongly Disagree’) to 7 (‘Strongly Agree’) scale. The Academic Risk and Resilience Scale is presented in the Appendix of Supplemental Material.

Dependent variables

The study also used five dependent variables: Anxiety, failure avoidance, uncertain control, self-handicapping, and disengagement. Consistent with recent multidimensional perspectives on motivation and engagement (Reschly & Christenson, 2012), academic engagement outcomes are separated into impeding engagement outcomes (i.e. relatively low-level ‘everyday’ negative engagement) and maladaptive engagement outcomes (i.e. relatively major negative engagement). The former outcomes are represented by anxiety, failure avoidance, and uncertain control; the latter outcomes are represented by self-handicapping and disengagement. All five measures are drawn from the Motivation and Engagement Scale (MES; Martin, 2010), comprise four items each, are rated on a 1 (‘Strongly Disagree’) to 7 (‘Strongly Agree’) scale, are psychometrically sound, and shown to be reliable and invariant as a function of gender, ethnicity, age, and ability (Martin, 2007, 2009).

Impeding academic engagement

Anxiety (e.g. ‘When exams and assignments are coming up, I worry a lot’) has two parts: Feeling nervous and worrying. Feeling nervous is the uneasy or sick feeling students get when they think about their schoolwork, assignments, or exams. Worrying is their fear of not doing very well in their schoolwork, assignments, or exams.

Failure avoidance (e.g. ‘Often the main reason I work at school is because I don’t want to disappoint my parents’) is evident when the main reason students do their schoolwork is to avoid doing poorly or to avoid being seen to do poorly.

Uncertain control (e.g. ‘I’m often unsure how I can avoid doing poorly at school’) assesses students’ uncertainty about how to do well or how to avoid doing poorly.

Maladaptive academic engagement

Self-handicapping (e.g. ‘I sometimes don’t study very hard before exams so I have an excuse if I don’t do so well’) refers to students’ tendency to do things that reduce their chances of success at school. Examples are putting off doing an assignment or wasting time while they are meant to be doing their schoolwork or studying for an exam.

Disengagement (e.g. ‘I’ve pretty much given up being involved in things at school’) refers to students’ inclination to give up in particular school subjects or in school generally. Students high in disengagement tend to accept failure and behave in ways that reflect helplessness.

The covariates

It is important to understand academic buoyancy and academic resilience controlling for covariates that may share variance with the two central constructs and with the outcome factors. Insofar as there exists shared variance between covariates and buoyancy/resilience, or between covariates and academic outcome factors, it is important to account for their presence in the modeling.

Socio-demographics

A number of socio-demographic covariates were included in order to partial out their influence and thereby better establish relatively unique effects of academic buoyancy and academic resilience. Socio-demographic variables were gender, age, language background, and socio-economic status (SES). For language background, students were asked if they spoke English (0) or another language (1–non-English speaking background, NESB) at home. Gender was coded 0 for girls and 1 for boys. Age was a continuous variable. SES was assessed using students’ home postcode and based on the Australian Bureau of Statistics (ABS) relative advantage/disadvantage index, with higher scores reflecting higher SES. Gender (Martin & Marsh, 2008a), language background, age and SES indicators (Martin, 2007, 2009; Rutter, 2006) have been found to significantly relate to academic buoyancy, resilience, and/or engagement.

Prior achievement

Academic achievement was based on students’ results in annual nation-wide assessment of numeracy and literacy (National Assessment Program in Literacy and Numeracy, NAPLAN) administered by the Australian Curriculum and Assessment and Reporting Authority (ACARA). In this nationally-standardized test, school students receive a score for literacy and for numeracy. In this study, an achievement factor was formed through averaging students’ literacy and numeracy scores.

Academic adversity

It is likely that academic adversity is correlated with academic engagement outcomes, with greater adversity associated with higher impeding and maladaptive engagement scores. Thus, adversity was included as a covariate. As described above, in the Academic Risk and Resilience Scale (ARRS), respondents are asked to indicate ‘yes’ or ‘no’ to a series of major academic adversity items. These major adversities are presented in Appendix A. An academic adversity score was calculated by tallying the number of ‘yes’ responses.

Data analysis

Measurement properties were assessed using confirmatory factor analysis (CFA) to test factor structure, reliability (Cronbach’s alpha) to assess internal consistency, and skewness and kurtosis to indicate distributional properties. The hypothesized model was assessed using structural equation modeling (SEM). CFA and SEM were performed using Mplus version 6.12 (Muthén & Muthén, 2010). In terms of goodness of fit, the root mean square error of approximation (RMSEA) and the comparative fit index (CFI) are presented. For RMSEAs, values at or less than 0.08 and 0.05 reflect close and excellent fits respectively (see Schumacker & Lomax, 1996). For CFI, values at or greater than 0.90 and 0.95 reflect acceptable and excellent fits respectively (McDonald & Marsh, 1990). Maximum likelihood with robustness to non-normality and non-independence of observations (MLR; Muthén & Muthén, 2010) was the method of estimation used. Less than 5% of the data were missing, and subsequently imputed using the Expectation Maximization (EM) Algorithm (within LISREL 8.80; Jöreskog & Sörbom, 2006).

A supplementary SEM explored a process model in which academic buoyancy predicts academic resilience, and both academic buoyancy and academic resilience predict outcomes. In this model, indirect effects (academic buoyancy on outcomes via academic resilience) are also examined. Indirect effects are based on bootstrapped standard errors (with 1000 draws) (Shrout & Bolger, 2002). MLR is not appropriate for indirect bootstrapping models and so the present study implemented maximum likelihood (ML) as the method of estimation here (Muthén & Muthén, 2010).

Results

Descriptive statistics, distribution, and reliability

Means and standard deviations are consistent with prior research (Martin, 2007, 2009; Martin & Marsh, 2008a, 2008b; presented in Table 1 within the Supplemental Materials are scale means, standard deviations, distributions [skewness, kurtosis], and reliability). Skewness and kurtosis values are indicative of approximately normal distributions. Reliability for all factors range between 0.73 and 0.90 (mean Cronbach’s α = 0.80), suggesting internal consistency.

Confirmatory factor analysis

The hypothesized two-factor (academic buoyancy and academic resilience) model was tested using confirmatory factor analysis (CFA). CFA yielded a good fit to the data, χ² (19, N = 918) = 106.51, p < 0.001, CFI = 0.95, RMSEA = 0.071. Mean factor loadings for academic resilience are acceptable, ranging from 0.75 to 0.89 (mean loading of 0.83), as are loadings for academic buoyancy, ranging from 0.60 to 0.67 (mean loading of 0.64). A subsequent CFA was then conducted to ascertain that the two factors reflect a better fit than a one-factor model in which all buoyancy and resilience items load on the one factor. This one-factor model fit the data poorly, χ² (20, N = 918) = 318.43, p < 0.001, CFI = 0.83, RMSEA = 0.127.

Having established the measurement properties of the buoyancy and resilience scales, a full CFA was conducted in which all eleven factors in the model were estimated. For academic buoyancy, academic resilience, anxiety, failure avoidance, uncertain control, self-handicapping and disengagement, latent factors were estimated using items as indicators. Age, gender, achievement, SES, language background, and academic adversity, were single-item indicators and thus estimated without error. This model provided a good fit to the data, χ² (455, N = 918) = 1061.12, p < 0.001, CFI = 0.93, RMSEA = 0.038.

Correlations relevant to academic buoyancy and academic resilience

Table 2 in Supplemental Material presents correlations derived from the 11-factor CFA described above; for brevity, correlations central to the hypothesized model (i.e. relevant to academic buoyancy and academic resilience) are reported here. All other correlations are available in Table 2; see Supplemental Materials. These correlations show academic buoyancy is significantly correlated with academic resilience (r = 0.59, p < 0.001). Academic buoyancy is also significantly correlated with anxiety (r = −0.56, p < 0.001), failure avoidance (r = −0.18, p < 0.001), uncertain control (r = −0.31, p < 0.001), self-handicapping (r = −0.12, p < 0.01), and disengagement (r = −0.21, p < 0.001). Academic resilience is significantly correlated with anxiety (r = −0.27, p < 0.001), failure avoidance (r = −0.17, p < 0.001), uncertain control (r = −0.32, p < 0.001), self-handicapping (r = −0.20, p < 0.001), and disengagement (r = −0.31, p < 0.001).

Structural equation modeling

SEM was used to estimate a model in which (a) academic buoyancy and academic resilience predict anxiety, failure avoidance, uncertain control, self-handicapping, and disengagement, and (b) socio-demographics, academic adversity, and prior achievement are covariates alongside buoyancy and resilience, also predicting the five dependent measures. The model fit the data well, χ² (455, N = 918) = 1061.12, p < 0.001, CFI = 0.93, RMSEA = 0.038. Table 3 presents all findings and Figure 1 presents central findings (see Supplemental Materials for Table 3 and Figure 1).

Figure 1.

Significant SEM effects for academic buoyancy and academic resilience.

Beyond the variance explained by socio-demographics, academic adversity and prior achievement, academic buoyancy significantly predicted anxiety (β = −0.64, p < 0.001), failure avoidance (β = −0.13, p < 0.05), and uncertain control (β = −0.20, p < 0.01)––but not self-handicapping or disengagement. Academic resilience significantly predicted uncertain control (β = −0.13, p < 0.05), self-handicapping (β = −0.15, p < 0.01), and disengagement (β = −0.25, p < 0.001)–– but not anxiety or failure avoidance. Hence, consistent with hypotheses, academic buoyancy more strongly predicted (low-level) impeding phenomena whereas academic resilience more strongly predicted (major) maladaptive phenomena.

Supplementary analyses: Ordering of academic buoyancy and academic resilience

A two-step application of the Baron and Kenny (1986) mediation criteria was used to test Martin and Marsh’s (2009) proposed ordering of buoyancy and resilience. The first step involved SEM in which academic buoyancy predicts all five negative engagement measures (alongside all covariates). This yielded good fit to the data, χ² (345, N = 918) 846.96, p < 0.001, CFI = 0.92, RMSEA = 0.040. Academic buoyancy significantly predicted anxiety (β = −0.58, p < 0.001), failure avoidance (β = −0.19, p < 0.001), uncertain control (β = −0.28, p < 0.001), self-handicapping (β = −0.09, p < 0.05), and disengagement (β = −0.15, p <.001).

The second step involved SEM in which academic buoyancy predicts academic resilience and both (alongside covariates) predict all five negative engagement outcomes. This model fit the data well, χ² (455, N = 918) = 1061.12, p < 0.001, CFI = .93, RMSEA = 0.038. For the most part, predictive paths operated as hypothesized. Academic buoyancy directly predicted academic resilience (β = 0.59, p < 0.001). Academic buoyancy also directly predicted anxiety (β = −0.64, p < 0.001), failure avoidance (β = −0.13, p < 0.05), and uncertain control (β = −0.20, p < 0.01)––but its direct links with self-handicapping and disengagement dropped out. Academic resilience directly predicted self-handicapping (β = −0.15, p < 0.01) and disengagement (β = −0.25, p < 0.001). Although it modestly predicted uncertain control (β = −0.13, p < 0.05), it did not predict anxiety or failure avoidance.

When testing for indirect effects of academic buoyancy on outcomes via academic resilience using bootstrapping (1000 draws), only one effect from academic buoyancy to the low-level negative outcomes is mediated by academic resilience: A relatively small effect for academic buoyancy → academic resilience → uncertain control (β = −0.07, p < 0.05). However, when estimating indirect effects from academic buoyancy to major negative outcomes via academic resilience, effects are larger: academic buoyancy → academic resilience → self-handicapping (β = −0.09, p < 0.01) and academic buoyancy → academic resilience → disengagement (β = −0.15, p < 0.001). Thus, the effect of academic buoyancy on low-level negative outcomes tends to be direct, whereas the effect of academic buoyancy on major negative outcomes is mediated by academic resilience.

Discussion

The present findings, based on a large sample of Australian high school students, showed that academic buoyancy and academic resilience represented separate factors that shared approximately 35% variance. Measurement findings further showed that a two-factor model fit the data better than a one-factor model, supporting previous speculation academic buoyancy and academic resilience represent distinct constructs (Martin & Marsh, 2009). Also consistent with hypotheses and previous conceptual contentions (Martin & Marsh, 2009), academic buoyancy was more salient in predicting low-level negative outcomes (anxiety, uncertain control, failure avoidance) whereas academic resilience was more salient in predicting major negative outcomes (self-handicapping, disengagement). Furthermore, the effect of academic buoyancy on low-level negative outcomes tends to be direct, whereas the effect of academic buoyancy on major negative outcomes is mediated by academic resilience, consistent with bottom-up (buoyancy predicting resilience) and horizontal (low-level predictors and outcomes more closely associated and substantial predictors and outcomes more closely associated) approaches to the self-system (Marsh & Yeung, 1998; Shavelson et al., 1976; Trautwein, 2003).

In relation to intervention, Waxman, Huang, and Pedron (1997) report that there are ‘alterable processes or mechanisms that can be developed and fostered for all students’ (p. 137). Psychologists and teachers, thus, may be encouraged to identify academic risk in the classroom and at school, help students better identify academic risk, minimize academic risk over which they have some control, and foster and sustain attributes in students that may assist them deal with these risks (see Morales, 2000 regarding such a resilience cycle). This aligns with research emphasizing the need to consider resilience a dynamic process that reflects the interaction of the context and the individual (see Rutter, 2006 for reviews). Indeed, given the preliminary evidence on the possible ordering of buoyancy and resilience, practitioners may consider prioritizing academic buoyancy as an ongoing approach to dealing with ‘everyday’ academic adversity and as a means to successfully position the student for academic resilience should major academic adversity arise. The Academic Risk and Resilience Scale ([ARRS]; see Supplemental Materials) provides a basis for assessing students’ academic risk.

Although these findings are located in the Australian school context and relate to students and schoolwork, practical implications may be generalizable beyond this specific context. In relation to the national context, recent research in the UK has found that academic buoyancy explains unique variance in student processes and outcomes (Putwain et al., 2012). In relation to diverse school subjects, other UK research has found shared variance in academic buoyancy across different school subjects and distinct links to external validity correlates (Malmberg, Hall, & Martin, in press). Beyond students, other work has demonstrated the relevance of buoyancy in the school context amongst school personnel (Parker & Martin, 2009). There is, then, growing evidence showing the applicability of academic buoyancy across various settings.

There are a number of limitations to consider when interpreting findings. First, the data are self-reported. Future research should collect data in other ways, including observing students’ responses to setback and the implications for their academic outcomes. Also, the present data are cross-sectional and thus longitudinal data are needed to support preliminary claims about the ordering of buoyancy and resilience. It might also be important to understand the cumulative effect of risk on academic buoyancy and academic resilience and the precise extent to which each construct relates to different amounts of cumulative risk. In recent research, it seems the presence of two risk factors is sufficient to predict academic failure (Lucio et al., 2012) and this might therefore signal a need to focus on academic resilience more than academic buoyancy.

Further, although academic adversity was used as a covariate to control for different levels of adversity in the study, future research might examine if different levels of adversity differentially impact resilience and its effects. It is also important to note that academic buoyancy and resilience items are similar, leading to potential bias in parameters. Future research might extend the present preliminary work to investigate various alternative empirical approaches (e.g. correlated residuals for parallel items) in the buoyancy-resilience process. Also, although the academic buoyancy items have been validated in previous research by others (Putwain et al., 2012), there is a need for similar validation of the academic risk and resilience items. Similarly, follow-up work (e.g. student interviews) might examine the fidelity of the two scales to ascertain the extent to which students differentiate them in the intended way.

Conclusion

This study has identified academic buoyancy and academic resilience as two distinct adversity-related constructs. After controlling for socio-demographic, prior achievement, and adversity factors, the study also identified unique positive roles of academic buoyancy and academic resilience in explaining student outcomes. These findings hold implications for researchers investigating students’ capacity to deal with everyday academic setback as well as more substantial academic adversity. Results are also relevant to school-based practitioners seeking to assist students deal with the highs and lows that characterize academic life.

Footnotes

Note

Thanks are extended to Brad Papworth, Harry Nejad, Farideh Nejad, and Marianne Mansour for data collection and data entry, Gregory Liem for data management, the Australian Research Council for funding, and participating schools and students.

Author biography

Andrew J. Martin, BA (Hons), MEd (Hons), PhD, is Professorial Research Fellow at the University of Sydney specializing in motivation, engagement, achievement, and quantitative research methods. He is also Honorary Senior Research Fellow in the Department of Education, University of Oxford and President Elect, International Association of Applied Psychology–Division 5 Educational, Instructional, and School Psychology. Andrew is an Associate Editor of the British Journal of Educational Psychology, and an Editorial Board member for two international journals (Journal of Educational Psychology and Contemporary Educational Psychology).

References

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.

Coleman

Hagell

(2007) Adolescence, risk, and resilience, London: John Wiley.

Covington

M. V.

(1992) Making the grade: A self-worth perspective on motivation and school reform, Cambridge: Cambridge University Press. doi: 10.1017/CBO9781139173582.

Finn

J. D.

Rock

D. A.

(1997) Academic success among students at risk for school failure. Journal of Applied Psychology 82: 221–234. doi: 10.1037/0021-9010.82.2.221.

Jöreskog

K. G.

Sörbom

(2006) LISREL 8.80, Chicago, IL: Scientific Software International.

Lucio

Hunt

Bornovalova

(2012) Identifying the necessary and sufficient number of risk factors for predicting academic failure. Developmental Psychology 48: 422–428. doi: 10.1037/a0025939.

MacDonald

(2007) Social exclusion, risk, and young adulthood. In: Coleman

Hagell

(eds) Adolescence, risk, and resilience, London: John Wiley.

Malmberg, L.-E., Hall, J., & Martin, A. J. (in press). Academic buoyancy in secondary school: Exploring patterns of convergence in mathematics, science, English and physical education. Learning and Individual Differences.

Marsh

H. W.

Yeung

(1998) Top-down, bottom-up, and horizontal models: The direction of causality in multidimensional, hierarchical self-concept models. Journal of Personality and Social Psychology 75: 509–527. doi: 10.1037/0022-3514.75.2.509.

10.

Martin

A. J.

(2007) Examining a multidimensional model of student motivation and engagement using a construct validation approach. British Journal of Educational Psychology 77: 413–440. doi: 10.1348/000709906X118036.

11.

Martin

A. J.

(2009) Motivation and engagement across the academic lifespan: A developmental construct validity study of elementary school, high school, and university/college students. Educational and Psychological Measurement 69: 794–824. doi: 10.1177/0013164409332214.

12.

Martin

A. J.

(2010) The Motivation and Engagement Scale, 10th ed. Sydney: Lifelong Achievement Group.

13.

Martin

A. J.

Marsh

H. W.

(2008a) Academic buoyancy: Towards an understanding of students' everyday academic resilience. Journal of School Psychology 46: 53–83. doi: 10.1016/j.jsp.2007.01.002.

14.

Martin

A. J.

Marsh

H. W.

(2008b) Workplace and academic buoyancy: Psychometric assessment and construct validity amongst school personnel and students. Journal of Psychoeducational Assessment 26: 168–184. doi: 10.1177/0734282907313767.

15.

Martin

A. J.

Marsh

H. W.

(2009) Academic resilience and academic buoyancy: Multidimensional and hierarchical conceptual framing of causes, correlates, and cognate constructs. Oxford Review of Education 35: 353–370. doi: 10.1080/03054980902934639.

16.

McDonald

R. P.

Marsh

H. W.

(1990) Choosing a multivariate model: Noncentrality and goodness-of-fit. Psychological Bulletin 107: 247–255. doi: 10.1037/0033-2909.107.2.247.

17.

Morales

E. E.

(2000) A contextual understanding of the process of educational resilience: High achieving Dominican American students and the ‘resilience cycle’. Innovative Higher Education 25: 7–22. doi: 10.1023/A:1007580217973.

18.

Muthén

L. K.

Muthén

B. O.

(2010) Mplus user's guide, Los Angeles, CA: Muthén & Muthén.

19.

Parker

P. D.

Martin

A. J.

(2009) Coping and buoyancy in the workplace: Understanding their effects on teachers' work-related well-being and engagement. Teaching and Teacher Education 25: 68–75. doi: 10.1016/j.tate.2008.06.009.

20.

Putwain

D. W.

Connors

Symes

Douglas-Osborn

(2012) Is academic buoyancy anything more than adaptive coping? Anxiety, Stress and Coping: An International Journal 25: 349–358.

21.

Reschly

A. L.

Christenson

S. L.

(2012) Jingle, jangle, and conceptual haziness: Evolution and future directions of the engagement construct. In: Christenson

S. L.

Reschly

A. L.

Wylie

(eds) Handbook of research on student engagement, New York, NY: Springer.

22.

Roberts

(1995) Youth and employment in modern Britain, Open University Press: Milton Keynes.

23.

Rutter

(2006) Genes and behavior: Nature-nurture interplay explained, London: Blackwell Publishing.

24.

Schumacker

R. E.

Lomax

R. G.

(1996) A beginner's guide to structural equation modeling, Mahwah, NJ: Erlbaum. doi: 10.1080/10705519609540025.

25.

Shavelson

R. J.

Hubner

J. J.

Stanton

G. C.

(1976) Validation of construct interpretations. Review of Educational Research 46: 407–441. doi: 10.2307/1170010.

26.

Trautwein, U. (2003). Schule und Selbstwert. In D. H. Rost (Ed.). Pädagogische Psychologie und Entwicklungspsychologie Bd. 36, Münster et al.: Waxman.

27.

Shrout

P. E.

Bolger

(2002) Mediation in experimental and non-experimental studies: New procedures and recommendations. Psychological Methods 7: 422–445. doi: 10.1037/1082-989X.7.4.422.

28.

Waxman

H. C.

Huang

S. L.

Padron

Y. N.

(1997) Motivation and learning environment differences between resilient and non-resilient Latino middle school students. Hispanic Journal of Behavioral Sciences 19: 137–155. doi: 10.1177/07399863970192003.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB