Teacher support is a well-established predictor of students' academic motivation in science classes. However, less is known about how students' personal beliefs shape their responsiveness to supportive teaching. Situated within a science curriculum addressing local environmental issues, this study investigated whether high school students' ecocentric beliefs, reflecting environmental concern and viewing humans as interconnected with ecological systems, moderate the relationship between teacher support and students' motivation in science classes. A total of 285 high school students (46.9% female; M age = 16.18, SD = 1.20) completed surveys both prior to and six months after the introduction of a science curriculum that highlights real-world relevance. Measures included perceptions of teacher support, ecocentric beliefs, and science class motivation (interest, importance, and utility value). Results revealed that teacher support was generally associated with increases in all three motivational outcomes over time. However, these associations varied among students: supportive teaching predicted increased motivation only for those who prioritized environmental concerns. Findings highlight the importance of value alignment in science education. Ecocentric beliefs amplified the motivational impact of supportive teaching, suggesting instruction is more effective when it resonates with students' personal environmental concerns and underscoring the need for differentiated strategies to engage all students.
Although researchers investigating randomized controlled trials (RCTs) may collect fidelity of implementation data, this information may not be optimally used in analyzing data to yield causal effect estimates. An approach to obtaining causal effects, though underutilized by educational researchers, uses instrumental variable (IV) estimation. This tutorial explains why and how instrumental variables work; differentiates compliance types; illustrates how IVs can be used to properly deal with issues related to noncompliance and dosage effects; provides R syntax to estimate models using two-stage least squares regression, structural equation modeling, and Bayesian regression; and includes a sample writeup. To support more widespread understanding and use in RCTs with issues of noncompliance/nonadherence, this tutorial does not use equations or specialized notation.
The underrepresentation of women in many academic disciplines is a well-documented phenomenon. Various interventions have been developed in the STEM field to address this. However, the extent to which these findings can be transferred to other disciplines remains an open question. In the present research, we address this gap and develop a tailored intervention to empower women in a male-dominated discipline outside of STEM - namely in philosophy. To this end, a Communal Goal Intervention was adapted to philosophy. Across three Pretests (N-Total = 114), communal cues typical for philosophy were gathered. An experimental study (N = 479) showed that variations in the presented working activities can affect the communal goals and attributes associated with philosophy. Building on this, a Communal Goal Intervention was tested experimentally with philosophy students (N = 260). The communal goal intervention significantly increased students' perceived suitability regarding abilities necessary for philosophy. This was further associated with a heightened intention to persist and a reduced intention to quit their studies. These findings generalize the application of Communal Goal Interventions from STEM to a discipline in the humanities. The results underscore the potential of targeted interventions to empower women in male-dominated disciplines without disadvantages for men.
As schools face record absenteeism, needed are ways of indexing forces that motivate youths' attendance. In this paper, we introduce eudaimonic utility (EU): the perception that an environment can help someone become the person they are meant to be. An environment with high EU is seen as a place where one can (a) pursue their purpose in life, (b) grow, (c) bring out their best, and (d) live authentically. To propose this new theoretical construct, we set out to articulate EU's position at the crossroads of motivation science and positive psychology. Then, to offer the field a tool to begin its empirical study, we established the Eudaimonic Utility Scale through factor analyses among adolescents enrolled in an extracurricular program (Study 1: N = 146, M[SD] = 16.24[1.22] years/Study 2: N = 170, M[SD] = 16.02[1.10] years). Program and school EU converged with most context-matched motivation measures; school EU correlated with higher self-reported grades and well-being; and, when anticipated program EU outpaced school EU, lower psychosocial health was likely. Together, EU offers programs and schools a novel and interdisciplinary way for tracking impact, as well as aligning their mission with the young people they are built to serve.
Researchers have increasingly investigated the relations between teacher motivation and teacher actions; however, like most motivation research, teacher motivation studies commonly focus on independent relations between motivational and action constructs. Such studies reveal little about the complex motivational processes that underlie teachers' situated behavior. The current study investigated the motivational processes undergirding one preservice teacher's situated, moment-by-moment actions while teaching an inquiry lesson in a high school history classroom. Guided by the Dynamics Systems Model of Role Identity theoretical framework, we analyzed video and written-reflection data. Results suggested that the preservice teacher's emergent actions were framed by her shifting role identity within the classroom's changing cultural activity system. These findings highlight the complex, dynamic, and socially-situated interplay of the motivational processes that undergird the preservice teacher's unfolding specific instructional actions. The findings support a view of situated actions emerging from the role identities of agents within a particular cultural activity system. We discuss the implications of this perspective for motivational theory, research, and practice.
This study examines how the functional role of pedagogical content in gameplay and the number of players affect learning outcomes and situational interest in an educational board game. We conducted research with 183 engineering students using four versions of the same game. The versions varied according to (a) whether processing the pedagogical content was required to succeed in the game (surface exposure vs. elaborative processing) and (b) the number of players (single-player vs. cooperative three-player). Results showed the effect of the elaborative processing on learning, whereas the effect of the number of players on interest was not statistically significant. These findings suggest that the educational effectiveness of games depends on how game mechanics structure learners' cognitive engagement with instructional content.
Fraction reduction (e.g., 4/6 to 2/3) is an important canonicalization rule for understanding equivalent fractions and solving fraction problems. However, relatively little is known about its cognitive consequences. We predicted that people would always reduce reducible fractions in general cognitive tasks as a legacy of this canonicalization rule, even when doing so hurts performance, but that they will be more flexible in mathematical tasks where they draw on a broad repertoire of strategies. Two experiments found evidence for these predictions. The general cognitive task was a fraction span task where participants remembered a sequence of fractions verbatim. Participants performed worse when remembering sequences that include reducible fractions (e.g., 6/7, 2/4, 5/7, 3/9) versus sequences that did not (e.g., 9/4, 2/7, 3/5, 6/7). The mathematical task was a fraction comparison task where participants compared pairs of fractions and judged which one was greater. In this task, people only reduced fractions when doing so was task-relevant and enabled use of an efficient componential strategy for comparing fractions with common numerators or denominators (e.g., 8/14 vs. 5/7). Importantly, they avoided reducing fractions when doing so blocked the use of the componential strategy (e.g., 4/14 vs. 5/14). These findings broaden research on rational number understanding and have implications for the impact of fraction instruction.
Understanding heterogeneity in a population through mixture modeling offers valuable insights that are not directly observable using traditional analytic approaches. Many social science studies involve a diverse range of participant characteristics, and using a model approach that captures heterogeneity across one or more of these constructs can deepen the understanding of the population. This paper introduces a mixture modeling approach that uses two latent class variables to examine relationships across domains. We compare this new approach to a more traditional latent profile analysis (LPA), where we estimate a single mixture model with both English and Spanish language and cognition indicators from a sample of 330 Latine DLLs. Next, we compare its results to a model we refer to as the joint occurrence latent profile analysis, where we model heterogeneity in language proficiency, followed by heterogeneity in cognition proficiency, and link the two models to examine their joint cross-domain relations. The joint occurrence approach reveals distinct patterns of language and cognitive proficiency that provide new insight into cross-domain variability among bilingual learners. The methodological framework and detailed code shared in the Supplementary Materials aim to make this approach accessible to researchers seeking to model multidimensional heterogeneity in applied settings.
The purpose of this study was to empirically test a Classroom Assessment and Self-Regulated Learning (CA:SRL) model that strategically fuses self-regulated learning (SRL) prompts into assessment tasks to improve student performance in high-school computer science (CS) classes. While many researchers have articulated the synergistic merging of SRL processes and assessment for learning (AfL) practices into new frameworks, no study has tested such a framework by manipulating SRL versus no SRL prompt in assessment tasks to improve student performance. Using an experimental design with repeated measures, 119 high-school CS students were randomly assigned to the Experimental (n = 64) or Control (n = 55) condition and underwent two cycles of performance assessments of computational thinking. Results showed that the Experimental group, which received SRL prompts, performed significantly less well than the Control group during the first cycle, but experienced gains in the second cycle, while the Control group did not. Students who underestimated their performance were more likely to perform poorly compared to their more confident peers. The educational implications of this study are articulated, with suggestions for further research.
This study examines how a theory-driven psychological intervention enhances online learning outcomes through motivation and self-regulation. Grounded in self-determination theory and social cognitive theory, a 16-week quasi-experimental study with 376 undergraduates in a blended management course tested an LMS-embedded intervention featuring personalized reminders, progress visualization with goal setting, and digital badges. Compared to controls, the experimental group showed significantly greater engagement (study time, task completion, platform activity) and superior academic performance (Cohen's d = 0.64). The intervention boosted intrinsic and extrinsic motivation, reduced amotivation, and strengthened self-regulated learning. Mediation analysis showed intrinsic motivation and self-regulation jointly accounted for over half of the total effect. Findings demonstrate that psychologically supportive online interventions improve outcomes by fulfilling autonomy, competence, and relatedness needs.
Math anxiety is widespread and can have many consequential effects on one's life. Extensive research has reported a strong negative association between math anxiety and math achievement. This relationship impacts students of all grade levels. Two of the most common short interventions to mitigate math anxiety and bolster math achievement are arousal reappraisal and expressive writing. The present preregistered study sought to replicate and extend seminal findings in the arousal reappraisal and expressive writing literature to improve math performance. Compared to a control condition in which participants wrote about what they did yesterday, participants who learned about how arousal can be used to facilitate rather than hinder performance (arousal reappraisal) or expressed their worries about an upcoming math assessment (expressive writing) failed to improve their performance on a math assessment compared to the control condition. Furthermore, these interventions did not affect participants' subjective assessments of their performance on the math task or their discomfort with it. No differences were observed when considering participants' prior math anxiety levels, the difficulty of the math problems, or socioeconomic status. Teachers and practitioners may seek a "quick fix," but overcoming math anxiety will likely require a consistent and concentrated effort.
In this study, we examine the impact of worked examples-based supplemental materials on elementary students' motivation for math. Previously published findings from a randomized controlled trial showed that higher dosage of using the worked examples materials demonstrated significantly higher math achievement on standardized tests compared to a control group (McGinn et al., 2024). In this follow-up analysis, we examine the relation between an understudied construct-perceived functionality of errors-and concurrent math achievement, along with other more commonly studied constructions (i.e., perceptions of math importance, math self-concept, interest in math). We also examine motivation constructs assessed within the randomized controlled trial to determine whether the materials had similarly positive effects on math motivation. Results demonstrated that students' perceived functionality of errors is indeed significantly and positively related to concurrent math achievement, even when accounting for other well-established motivation constructs. Students who used the worked examples supplemental materials did not statistically significantly differ in their motivational beliefs compared to a problem-solving control group. Unexpected findings, such as a negative association between perceptions of math importance and concurrent math achievement are discussed.
Errorful generation-making (often) incorrect guesses about unfamiliar information-can enhance learning if accompanied by corrective feedback. Four experiments investigated how feedback processing influences the retention of obscure word definitions. Participants processed feedback via non-generative (reading, copying), elaborative (studying examples or mnemonics), or generative (rephrasing, example generation) methods. In Experiments 1, 2, and 4, rephrasing feedback consistently yielded better learning than reading or copying. In Experiment 3, rephrasing and example generation outperformed keyword mnemonics. Experiment 4 revealed that rephrasing improved learning even without errorful generation. Evidence for studying provided examples was mixed. Overall, these findings demonstrate that generative processing facilitates learning, and for errorful generation, the impact of feedback is determined not just by its presence, but how it is engaged with.
One goal of motivation science is to create classroom climates that support students' success. Workforce demands and retention problems in STEM point to a need for research on how to support the motivational needs of students. Motivational climate, i.e., students' perceptions of their learning environments' motivating or demotivating qualities, is key in shaping students' subsequent motivation. Extant research does not provide clarity about how motivational climate should be conceptualized and measured, especially in large post-secondary lectures. Existing measures of motivational climate are not comprehensive and have limited validity evidence to support use in the post-secondary setting. To address these challenges, we used think-aloud interviews to examine the response processes post-secondary students undertake to answer items. Our findings reveal threats to validity regardless of setting and illuminate how existing items are irrelevant to the large lecture setting. We offer insight into what practices contributed to positive motivational climate and guidance on writing items to measure motivational climate in the STEM post-secondary lecture setting.
Generative artificial intelligence (AI) offers new opportunities for individualized education, yet its effectiveness may vary across learners. This study examined whether ChatGPT improves nursing students' performance in creating care plans and whether intelligence and language proficiency moderate this potential effect. A total of 221 nursing students completed two care plan tasks in randomized order-one with ChatGPT-4 support and one without. Expert raters, blind to condition, evaluated care plan quality, and students' intelligence and language proficiency were assessed using standardized measures. Latent change score modeling revealed that ChatGPT-assisted care plans were rated significantly higher in quality than unassisted ones (d = 0.85). Intelligence moderated this benefit, such that students with higher intelligence showed greater improvement in creating nursing care plans through the use of ChatGPT. In contrast, language proficiency predicted overall task performance, but did not moderate ChatGPT-related benefits. These findings indicate that while ChatGPT can substantially enhance nursing students' performance in complex, domain-specific writing tasks, it may also risk reinforcing existing achievement differences associated with intelligence. Implications for the equitable integration of generative AI in education are discussed.
Learning engagement is situational and context-dependent. Reconceptualizing situational engagement as consisting of both "trait" and "state" aspects can be theoretically and practically informative. The purpose of this study was to examine undergraduate students' situational engagement as they learn in out-of-classroom contexts, including how much variance in behavioral, cognitive, and affective engagement is at the trait- versus state-levels. Another aim was to identify student- and context-related factors that predict engagement at each level. The study employed multilevel Latent State-Trait (LST) modeling with Experience Sampling Method (ESM) data of 327 learning events from 57 undergraduate students. Results showed that more of the variance in engagement was consistent across learning events. Trait self-efficacy was significantly associated with trait engagement, whereas state self-efficacy was associated with state engagement. Time management was a significant predictor of trait engagement, whereas academic workload and presence of a deadline were predictors of state engagement. These results have implications for learning supports that can be provided to enhance situational engagement during out-of-classroom learning.
This tutorial offers practical guidance on causal mediation analysis for observational studies, emphasizing the critical alignment of research questions, causal estimands, and statistical models. Through educational research examples, the tutorial addresses analytical challenges, such as distinguishing between full and partial mediation, controlling for confounders, and avoiding collider bias. The tutorial demonstrates how to employ appropriate statistical methods (e.g., path analysis or regression modeling), contrasting correct analytical strategies with common misspecifications arising from misaligned estimands or statistical approaches. The tutorial illustrates the decomposition of a total effect into direct and indirect components to clarify causal mechanisms. Directed acyclic graphs (DAGs) are introduced as tools for visualizing causal assumptions, guiding variable selection, and ensuring alignment across the research question, estimand, and analysis.
