A Multilevel Person-Centered Examination of Teachers’ Workplace Demands and Resources: Links With Work-Related Well-Being

Abstract

Teachers' healthy and effective functioning at work is impacted by the demands they face and the resources they can access. In this study, person-centered analysis was adopted to identify distinct teacher profiles of demands and resources. We investigated teachers' experiences of two job demands (barriers to professional development and disruptive student behavior), two job resources (teacher collaboration and input in decision-making), and one personal resource (self-efficacy for teaching). Using data from the Teaching and Learning International Survey (TALIS) 2013, the study involved 6,411 teachers from 369 schools in Australia and 2,400 teachers from 154 schools in England. In phase one, latent profile analysis revealed five teacher profiles that were similar across the two countries: the Low-Demand-Flourisher (12%), Mixed-Demand-Flourisher (17%), Job-Resourced-Average (34%), Balanced-Average (15%), and Struggler (21%). The profiles were differently associated with two background characteristics (teacher gender and teaching experience) and two work-related well-being outcomes (job satisfaction and occupational commitment). In phase two, we extended our analysis to the schoollevel to identify school profiles based on the relative prevalence of the five teacher profiles within a school. Indeed, a yield of large scale datasets such as TALIS is that there are sufficient units at the school-level to enable institutional insights, beyond insights garnered at the individual teacher-level. Two school profiles that were similar in both countries were revealed: the Unsupportive school profile (58%) and the Supportive school profile (42%). The Supportive school profile was associated with higher schoolaverage teacher job satisfaction and occupational commitment than the Unsupportive school profile. Taken together, the findings yield knowledge about salient teacher and school profiles, and provide guidance for possible interventions at the teacher-and school level.

Assessing Instructional Cognitive Load in the Context of Students' Psychological Challenge and Threat Orientations: A Multi-Level Latent Profile Analysis of Students and Classrooms

“…For the single-level LPA conducted at the student-level (Level 1; L1), we tested a range of solutions involving 1 through 9 profiles. Following Collie et al ( 2020 ), variances and means were free to differ across profiles and indicator variables; models were estimated using at least 10,000 random start values, with 100 iterations and 1,000 final stage optimizations; and we confirmed that the best log-likelihood value was replicated for each model. Numerous indices were used to assess model fit: for the Consistent Akaike Information Criteria (CAIC), Akaike Information Criteria (AIC), Bayesian Information Criteria (BIC), and sample-size-adjusted Bayesian Information Criteria (SSA-BIC), smaller values reflect better fit.…”

“…Multi-level LPA solutions (1–9 classroom-level profiles) were assessed. Following Collie et al ( 2020 ), each model was estimated using at least 10,000 random start values, 100 iterations, and 1,000 final stage optimizations; replication of the best log-likelihood value was sought for each model; and the best model was selected using the same criteria as the single-level LPA (phase 1), except the p LMR, which is not available for multi-level LPA. After identifying the network of classroom-level profiles (the classroom-level within-network validity aspect), we then examined the network of L2 outcomes (classroom-average persistence, disengagement, and achievement) as a function of profile membership by adding classroom-average outcome variables (using the M plus cluster mean approach) to the final best-fitting model (the classroom-level between-network validity aspect).…”

“…Fourth, our data were cross-sectional which means, for example, that we were unable to determine causal ordering between the profiles and the outcomes, nor whether student and classroom profile memberships change over time. Longitudinal data and modeling (e.g., latent transition analysis) will be an important avenue in future research (Collie et al, 2020 ). Fifth, our study included self-efficacy and growth goals to infer challenge orientation and anxiety and failure avoidance goals to infer threat orientation.…”

“…As noted above, the bulk of research assessing cognitive load in students' academic lives (including LRI research) has been variable-centered. Variable-centered approaches aim to assess relations between factors across individuals (Collie et al, 2020 ). This has been important for conducting classic construct validity work with cognitive load factors and for identifying what cognitive load factors predict or are predicted by other factors (e.g., Leppink et al, 2013 ; Klepsch and Seufert, 2020 ).…”

Caracterización del capital psicológico en el profesorado: una revisión sistemática

“…n los últimos años ha cobrado relevancia el estudio de las capacidades y las fortalezas que pueden mejorar el rendimiento del personal laboral en distintos ámbitos laborales y que caracterizan el comportamiento organizacional positivo (Luthans, 2002(Luthans, , 2012. Del mismo modo, desde la Educación Positiva se destaca la infl uencia de los recursos y competencias del profesorado en su rendimiento y en el de su alumnado (Duckworth, Quinn y Seligman, 2009;Lavy, 2020) y se ponen en valor las destrezas que propician el bienestar y la salud del personal laboral en el ámbito educativo (Collie, Malmberg, Martin, Sammons y Morin, 2020;Wang y Hall 2019).…”