Short-Term Dynamics: Transitions in First-Year Undergraduates’ Motivational Profiles Across a Semester
YiJiang1,2✉Emailyjiang@dep.ecnu.edu.cn
LinjiaZhang1
XinyiShi1
1Department of Educational PsychologyEast China Normal UniversityShanghaiChina
2Faculty of Education, Department of Educational PsychologyEast China Normal University3663 North Zhongshan Road200062ShanghaiChina
Yi Jiang*, Linjia Zhang, Xinyi Shi
Department of Educational Psychology, East China Normal University, Shanghai, China
* Correspondence should be addressed to Yi Jiang, Faculty of Education, Department of Educational Psychology, East China Normal University, 3663 North Zhongshan Road, Shanghai, 200062, China (e-mail: yjiang@dep.ecnu.edu.cn; ORCiD: 0000-0003-4440-1722)
A
Data Availability
The data from this study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate:
A
A
All participants provided informed consent before taking part in this study. They were informed about the purpose of the research, their right to withdraw at any time without consequences, and the confidentiality of their responses. A
A
The study adhered to ethical guidelines and received approval from the Institutional Review Board of East China Normal University for human participants.Consent for publication:
The results presented in this manuscript have not been published elsewhere, nor are they under consideration (from any of the authors) by another publisher.
Competing interests:
The authors declare no conflict of interest.
A
Funding:
This research was supported by the National Social Science Foundation for Education of China (BIA220069).
A
Author Contribution
YJ: Supervision, Conceptualization, Methodology, Writing- Reviewing and Editing; LZ: Data curation, Formal analysis, Writing- Original draft preparation; XS: Data curation, Formal analysis, Writing- Original draft preparation. All authors have read and agreed to the published version of the manuscript.
Acknowledgements:
Not applicable.
Abstract
A
Based on situated expectancy-value theory and a person-centered approach, we examined the stability and changes in motivational, defined by self-efficacy, task values, and perceived costs, among first-year undergraduates enrolled in an introductory psychology course during their first semester. Using latent profile and latent transition analyses with 232 Chinese undergraduates, we identified three consistent motivational profiles across three time points: Burdened (25–32%), Average-all (51–52%), and Positively motivated (17–24%). Approximately 54% of students remained in the same profile across the semester. While shifts toward less favorable profiles were more common in the first half of the semester, transitions to more favorable profiles increased in the latter half. Gender did not significantly predict the likelihood of transition. These findings offer important implications for timing and tailoring motivational interventions to support undergraduates’ academic learning.Keywords:
situated expectancy-value theory
latent transition analysis
undergraduates
Short-Term Dynamics: Transitions in First-Year Undergraduates’ Motivational Profiles Across a Semester
A
A
1 Introduction
The first semester of undergraduate study represents a critical transitional period during which students must adapt to a new academic environment, often accompanied by shifts in their motivational beliefs (Putwain & Sander, 2016). Situated Expectancy-Value Theory (SEVT), a prominent theoretical framework in motivation research, emphasizes the importance of understanding students’ motivational beliefs during key developmental phases, especially when they are making decisions about future educational and career pathways (Eccles & Wigfield, 2020, 2023). Recent studies grounded in SEVT suggest that students concurrently hold beliefs about expectancy, task values, and perceived costs, which together form their motivational profiles (Jiang & Zhang, 2023; Lee et al., 2022; Perez et al., 2019). Longitudinal, person-centered approaches are particularly well-suited to tracing nuanced changes in motivation profiles over time, providing valuable insights into motivational development that are crucial for promoting long-term academic success (Xie et al., 2022).
Despite the growing body of longitudinal, person-oriented research in educational settings, several important gaps remain. First, existing studies often emphasize competence beliefs and positive task values, while overlooking the role of perceived costs (e.g., Lazarides et al., 2019, 2020). Given the importance of perceived costs in predicting students’ avoidance behaviors and maladaptive academic outcomes (Jiang et al., 2018, 2020; Perez et al., 2014), it is essential to examine how motivational profiles, including the contribution of perceived costs, evolve during transitional periods. Second, much of the existing research on developmental shifts in expectancy-value-cost profiles has focused on secondary school students (Dietrich & Lazarides, 2019; Vinni-Laakso et al., 2022; Widlund et al., 2024) or on the transition from primary to secondary education (Raufelder et al., 2022). In contrast, relatively little is known about how these profiles develop among first-year undergraduates, despite the significance of this period for students’ academic adjustment and long-term motivation. Therefore, we aimed to address these gaps by examining how first-year undergraduate students’ motivational profiles, comprising self-efficacy, task values, and perceived costs, change across three time points during their first semester. Additionally, we investigate whether gender plays a role in predicting profile transitions, contributing to a more comprehensive understanding of motivational development in early undergraduate education.
1.1 Situated expectancy-value theory
The Situated Expectancy-Value Theory (SEVT) is a widely recognized framework that examines how students’ competence-related and value-related beliefs shape their academic performance and future career aspirations (Eccles & Wigfield, 2020; Wigfield & Cambria, 2010). Within this framework, expectancies for success refer to individuals’ perceptions of their likely performance on future tasks (Eccles & Wigfield, 2002). Self-efficacy, which measures students’ confidence in their academic abilities, is both conceptually and empirically aligned with expectancies for success (Bong & Skaalvik, 2003; Eccles & Wigfield, 2020). SEVT also highlights four key dimensions of value-related belief that shape students’ motivation: interest value (The enjoyment derived from a task); utility value (The perceived relevance of a task for achieving personal goals); attainment value (How a task aligns with an individual’s sense of identity); and perceived cost (The negative aspects of engaging in a task). These components can be categorized according to their potential positive or negative influence on the overall valuation of a task.
Among these dimensions, perceived cost has historically received less attention compared to the positive task values, especially prior to 2015. However, it is increasingly recognized as critical to understanding why students may lower their ambitions or develop avoidance motivation (Conley, 2012; Jiang et al., 2018, 2020). Researchers have identified multiple facets of cost, including negative appraisals of the effort required to complete the task (effort cost), foregone opportunities to engage in other valued tasks (opportunity cost), ego threats associated with potential task failure (psychological cost), and negative affective perceptions associated with task engagement (emotional cost; Eccles et al., 1983; Wigfield et al., 2017; Wigfield & Eccles, 2020). Recent studies highlight that examining these distinct facets of cost provides a more comprehensive understanding of student motivation (Gaspard et al., 2020; Lee et al., 2022; Perez et al., 2019). Therefore, focusing exclusively on positive task values may offer only a partial view of the factors that influence students’ academic decisions and behaviors.
1.2 Expectancy-value profiles through a person-centered lens
Traditionally, researchers examining the relationships between expectancy-value beliefs, performance, and educational choices have relied on variable-centered approaches. While these approaches provide valuable insights into the general linear associations among these constructs, they often fail to capture how expectancies, task values, and perceived costs are organized within individuals. In contrast, person-centered approaches, such as latent profile analysis (LPA), offer a complementary perspective by examining the nature, extent, and functions of combined motivational beliefs across different groups of individuals. As highlighted by Wormington and Linnenbrink-Garcia (2017), LPA is particularly effective for analyzing students’ multifaceted motivation, allowing researchers to investigate how combinations of competencies, task values, and perceived costs interact to shape the learning process in a nuanced way.
Empirical research has revealed heterogeneous patterns in students’ combined expectancy-value-cost beliefs. Common patterns include profiles with moderate levels across all beliefs, profiles with high competence-related beliefs and task values combined with low perceived costs, and profiles with low competence-related beliefs and task values combined with high perceived costs (e.g., Perez et al., 2019; Lee et al., 2022). Additionally, mixed motivational profiles have been identified, such as those characterized by low perceived costs together with low competence-related beliefs and task values (Jiang & Zhang, 2023), or profiles with simultaneously high levels of competence-related beliefs, task values, and perceived costs (Conley, 2012, Watt et al., 2019). These diverse patterns highlight the complexity and variability of student motivation.
In recent years, researchers have embraced longitudinal person-centered approaches, such as latent transition analyses (LTA), to examine the stability and volatility of expectancy-value-cost combinations over time. These approaches are particularly valuable because changes in motivation are often more complex than simple increases or decreases, especially when motivation is viewed as a multifaceted construct (Xie et al., 2022). Some studies have focused primarily on competence-related beliefs and positive task values, often neglecting perceived costs (Fryer & Ainley, 2019; Lazarides et al., 2019, 2020). For example, Fryer and Ainley (2019) identified three distinct motivational subgroups among college students, namely Low, Mid, and High, based on reported levels of self-efficacy, self-concept, interest, and utility value. These subgroups remained relatively stable over two semesters, although some transitions occurred between the Low and Mid groups. Similarly, Lazarides et al. (2019) identified a motivational profile during the transition from grade 9 to grade 10 that they labeled the Highly confident, hardly interested profile, characterized by high self-concept, interest, and utility value. This profile, however, was less stable compared to others.
As the significance of perceived costs in academic performance has gained more attention, researchers have increasingly incorporated various types of cost into motivational profile analyses. Studies have identified cost-oriented profiles and documented related motivational changes over time. For instance, Dietrich and Lazarides (2019) identified four distinct profiles among secondary school students: High motivation, Balanced above-average motivation, Average motivation, and Low motivation (characterized by high perceived costs). These profiles remained relatively stable between grades 9 and 10. Similarly, Vinni-Laakso et al. (2022) investigated the stability of motivational profiles from grades 7 to 8, finding that the Moderate motivation and cost profile was the most consistent. Notably, students in profiles such as High motivation with low cost, Low motivation with high cost, or High motivation with high cost often transitioned to the Moderate motivation and cost profile, reflecting both positive and negative shifts influenced by emotional cost.
Other studies have confirmed the general stability of motivational profiles, while noting occasional shifts. For example, Raufelder et al. (2022) studied students transitioning from primary to lower secondary school and identified four profiles: Positively engaged, Struggling ambitious, Disengaged relaxed, and Disengaged strained. While most profiles remained stable, the Positively engaged profile showed a greater tendency to shift toward disengaged profiles over time. In a more recent study, Widlund et al. (2024) identified four motivational profiles among ninth-grade students: Positively ambitious, Struggling ambitious, Indifferent, and Maladaptive. Some students in the Positively ambitious profile transitioned to the Struggling ambitious profile, while others from the Struggling ambitious profile shifted to Indifferent or moved from Indifferent to Maladaptive. Positive changes were also observed, with some students transitioning from Struggling ambitious to Positively ambitious and others moving from Maladaptive to Indifferent.
While research has provided important insights into the dynamic nature of expectancy-value-cost profiles and their shifts over time, significant gaps remain in the study of undergraduate students, particularly first-year students in non-Western cultural contexts. Additionally, many studies have focused primarily on transitions between two time points, leaving the variation in transition patterns across different periods largely unexplored. Understanding these patterns could offer valuable insights into motivational stability and shifts across broader student populations, providing a more detailed and nuanced perspective on how these changes unfold.
1.3 Gender disparities in profile development
Gender differences have long been a central focus of SEVT (Eccles & Wigfield, 2023), which was originally developed to explain disparities in students’ achievement-related choices. For example, SEVT was designed to address the question of why girls are less likely than boys to take advanced math courses in high school or to pursue careers in math and science (Eccles, 1984; Eccles & Wigfield, 2023). In the context of longitudinal motivational profile transitions, understanding the role of gender is equally critical. Specifically, it is important to explore whether male or female students are more likely to transition into less favorable motivational profiles.
Despite its significance, research examining the effect of gender on specific motivational transitions remains limited, leaving gaps in our understanding of how gender influences the stability of motivation profiles. The existing empirical findings are inconsistent. For instance, Dietrich and Lazarides (2019) found that boys were more likely to remain in a High motivation profile throughout the school year, a stable pattern that was less common for girls. In contrast, Widlund et al. (2024) reported no significant gender effects on the probabilities of transitioning between profiles. These mixed results underscore the need for further research into how gender influences the dynamics of motivational profile changes over time.
1.4 The present study
Building on the valuable insights of previous person-centered studies, while addressing their limitations, we employed LTA to investigate the development of students with more adaptive motivation profiles compared to those with less adaptive profiles. Specifically, we aimed to explore the combinations of self-efficacy, task values (i.e., intrinsic, utility, and attainment value), and perceived costs (i.e., effort, opportunity, ego, and emotional cost) among Chinese first-year undergraduate students. Our investigation focused on the stability and transitions of these profiles across three time points within the first semester. Additionally, we explored whether gender could predict changes in profile membership. The research was guided by the following questions: (1) What types of expectancy-value-cost profiles can be identified at three time points during the first semester? We anticipated identifying various motivation profiles, though we did not hypothesize a specific number; (2) How many students experience changes in their motivation profiles throughout the semester? We hypothesized that profiles would be relatively stable, although we expected some students to transition to different profiles; (3) Are there notable gender disparities in longitudinal profile paths, including stable paths and transitions? We approached this question without a specific hypothesis, leaving it open-ended for exploration.
2 Method
2.1 Participants
A sample of first-year undergraduate students enrolled in an introductory-level psychology course at a public university in Shanghai, China, participated in the study. Data on students’ self-efficacy, task values and perceived costs were collected at three points during their first semester via online questionnaires, which students completed independently.
In the first wave (T1), conducted during the 7th week of the fall semester, 209 students participated. Six weeks later, during the 13th week (T2), 206 students took part in the second wave. In the third wave (T3), conducted during the 19th week, 190 students participated. The final sample consisted of 232 students (Mage = 19.40, SD = 0.96, 28.4% male), with each student contributing data at least once to maximize the sample size for more accurate and generalizable results (Zhang et al., 2023; Zuidema et al., 2023). The study was approved by the Institutional Review Board for human participants at the authors’ university.
2.2 Measures
All survey items were written in Chinese and were based on six-point Likert scales ranging from 1 (completely disagree) to 6 (completely agree). Scales originally developed in English were translated and back translated following the process recommended by Brislin (1970). Sample items are presented in Table 1.
Variable | # of items | Sample item |
|---|---|---|
Self-efficacy | 6 | I’m certain that I can understand what is taught in the introductory psychology course. |
Interest value | 4 | I enjoy studying introductory psychology. |
Utility value | 4 | What I am learning in the introductory psychology course is relevant to my life. |
Attainment value | 4 | It is important for me to learn introductory psychology. |
Effort cost | 3 | Doing well in introductory psychology requires more effort than I want to put into it. |
Opportunity cost | 3 | I have to give up other activities that I like to do well in the introductory psychology course. |
Ego cost | 3 | Others would think worse of me if I failed to do well in the introductory psychology course. |
Emotional cost | 3 | Studying the introductory psychology course scares me. |
2.2.1 Self-efficacy
Six items measuring self-efficacy were adopted from Bong (2008). The reliability coefficients for the three measurement waves were αs = .92, .92, and .95, respectively.
2.2.2 Task values
Items measuring task value beliefs were adopted from Jiang and Zhang (2023). There were twelve items assessing three types of task value, with four items each for interest value (αs = .95, .96, and .96 for the three measurement waves, respectively), utility value (αs = .93, .97, and .94 for the three measurement waves, respectively), and attainment value (αs = .86, .92, and .93 for the three measurement waves, respectively).
2.2.3 Perceived costs
Twelve items measuring students’ perceptions of cost were adopted from Jiang et al. (2018). These items covered four facets: effort cost (αs = .88, .91, and .88 for the three measurement waves, respectively), opportunity cost (αs = .96, .98, and .97 for the three measurement waves, respectively), ego cost (αs = .94, .93, and .95), and emotional cost (αs = .81, .90, and .94 for the three measurement waves, respectively).
2.2.4 Covariate
Gender was self-reported by students, with males coded as 0 and females coded as 1, and was included as a covariate in the analyses.
2.3 Statistical analysis
All analyses were conducted using Mplus 8.10. The model parameters were estimated using the Maximum Likelihood Estimator (MLR) with Full Information Maximum Likelihood (FIML) estimation to handle missing data (Enders, 2010). Confirmatory factor analyses (CFA) were initially conducted on all measures to verify the factor structure of the constructs and obtain factor scores for subsequent profile analyses. Before the profile analysis, longitudinal CFAs were conducted to test measurement invariance across the three time points, ensuring that students interpreted self-efficacy, task values, and perceived costs consistently over time. We specified a series of nested models with increasing invariance constraints and examined changes in goodness-of-fit statistics. First, we fitted a configural invariance model, with identical loading patterns but no parameter constraints. We then tested for weak (invariant factor loadings) and strong (invariant factor loadings and item intercepts) measurement invariance across the three waves. Invariance was considered retained if the change in CFI was less than 0.01 and RMSEA increased by no more than 0.015 (Chen, 2007; Cheung & Rensvold, 2002).
Model building using LTA began with identifying the best-fitting latent profiles by conducting a series of LPA separately for each time point. Several statistical indicators were used to identify the optimal number of profiles, following recommendations by Nylund et al. (2007): Model Log-Likelihood (LL), Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Sample-Size Adjusted BIC (aBIC), with lower values indicating better fit. The Vuong-Lo-Mendell-Rubin Likelihood Ratio Test (VLMR) and Bootstrapped Likelihood Ratio Test (BLRT) were also used, where a significant p value indicates a better fit for the k-profile model compared to the k-1 model. Additionally, entropy values greater than 0.70 indicate accurate classification. Beyond statistical indicators, conceptual interpretability and the practical significance of profile sizes were also evaluated to ensure a parsimonious solution in which each profile contained a substantial proportion of the sample.
In the next phase, LTA was used to examine transitions in profile membership across the semester, based on the cross-sectional LPA solutions (Widlund et al., 2024). To assess the role of gender in predicting these transitions, the BCH-LTA approach proposed by Asparouhov and Muthén (2021) was employed. The BCH method follows a two-step procedure: first, the latent profile measurement model is estimated and BCH weights are computed, accounting for the standard errors of the latent profile variable. In the second step, these weights are incorporated into an auxiliary model to examine the predictive effect of gender on profile transitions. The LTA model specification was guided by the syntax and procedures outlined in Widlund et al. (2024).
3 Results
3.1 Preliminary results
Descriptive statistics, reliability estimates, and correlations for all measures at each time point are presented in Table 2. As shown in Table 3, the results of the CFAs support measurement invariance over time, indicating that the constructs were measured consistently across the three time points (Chen, 2007; Cheung & Rensvold, 2002).
1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 SE T1 | -- | |||||||||||||||||||||||
2 ITV T1 | .47 | -- | ||||||||||||||||||||||
3 UTV T1 | .48 | .74 | -- | |||||||||||||||||||||
4 ATV T1 | .58 | .73 | .79 | -- | ||||||||||||||||||||
5 EFC T1 | .10 | .02 | .05 | .06 | -- | |||||||||||||||||||
6 OPC T1 | − .09 | − .27 | − .25 | − .20 | .46 | -- | ||||||||||||||||||
7 EGC T1 | − .04 | − .11 | − .15 | − .11 | .33 | .45 | -- | |||||||||||||||||
8 EMC T1 | − .27 | − .37 | − .38 | − .34 | .43 | .63 | .69 | -- | ||||||||||||||||
9 SE T2 | .47 | .31 | .31 | .38 | − .01 | − .12 | − .16 | − .32 | -- | |||||||||||||||
10 ITV T2 | .39 | .61 | .52 | .53 | − .06 | − .30 | − .18 | − .40 | .61 | -- | ||||||||||||||
11 UTV T2 | .32 | .51 | .56 | .59 | − .02 | − .28 | − .13 | − .37 | .57 | .76 | -- | |||||||||||||
12 ATV T2 | .39 | .45 | .45 | .54 | .09 | − .16 | − .05 | − .28 | .61 | .74 | .84 | -- | ||||||||||||
13 EFC T2 | − .12 | − .18 | − .22 | − .22 | .39 | .31 | .26 | .36 | − .18 | − .19 | − .12 | − .03 | -- | |||||||||||
14 OPC T2 | − .08 | − .23 | − .24 | − .23 | .24 | .41 | .21 | .29 | − .13 | − .24 | − .20 | − .07 | .63 | -- | ||||||||||
15 EGC T2 | − .15 | − .19 | − .24 | − .20 | .10 | .30 | .39 | .42 | − .20 | − .31 | − .26 | − .18 | .49 | .66 | -- | |||||||||
16 EMC T2 | − .21 | − .29 | − .25 | − .28 | .18 | .36 | .33 | .51 | − .31 | − .40 | − .38 | − .28 | .44 | .57 | .78 | -- | ||||||||
17 SE T3 | .46 | .37 | .36 | .47 | .08 | − .14 | − .07 | − .19 | .39 | .41 | .30 | .36 | − .14 | − .13 | − .12 | − .18 | -- | |||||||
18 ITV T3 | .46 | .42 | .37 | .50 | .08 | − .07 | − .08 | − .17 | .30 | .49 | .33 | .36 | − .27 | − .21 | − .19 | − .27 | .72 | -- | ||||||
19 UTV T3 | .47 | .40 | .46 | .54 | .02 | − .19 | − .02 | − .20 | .34 | .46 | .47 | .47 | − .18 | − .14 | − .10 | − .20 | .76 | .72 | -- | |||||
20 ATV T3 | .41 | .39 | .43 | .50 | .12 | − .10 | − .02 | − .12 | .32 | .42 | .40 | .42 | − .13 | − .14 | − .09 | − .16 | .86 | .69 | .84 | -- | ||||
21 EFC T3 | .09 | .01 | − .02 | .03 | .42 | .37 | .16 | .31 | .06 | .09 | .05 | .13 | .27 | .19 | .14 | .14 | .26 | .10 | .17 | .29 | -- | |||
22 OPC T3 | − .04 | − .15 | − .18 | − .17 | .25 | .51 | .21 | .38 | − .03 | − .13 | − .18 | − .06 | .32 | .36 | .26 | .29 | .00 | − .13 | − .10 | .00 | .63 | -- | ||
23 EGC T3 | − .05 | − .09 | − .14 | − .12 | .17 | .29 | .37 | .43 | − .10 | − .12 | − .16 | − .10 | .26 | .20 | .39 | .37 | .00 | − .10 | − .07 | .00 | .38 | .56 | -- | |
24 EMC T3 | − .17 | − .24 | − .28 | − .27 | .19 | .40 | .35 | .52 | − .23 | − .30 | − .31 | − .21 | .37 | .33 | .40 | .47 | − .17 | − .30 | − .24 | − .16 | .45 | .67 | .71 | -- |
α | .92 | .95 | .93 | .86 | .88 | .96 | .94 | .81 | .92 | .96 | .97 | .92 | .91 | .98 | .93 | .90 | .95 | .96 | .94 | .93 | .88 | .97 | .95 | .94 |
M | 4.49 | 4.71 | 4.85 | 4.88 | 3.86 | 2.76 | 2.86 | 2.68 | 4.34 | 4.57 | 4.81 | 4.74 | 3.65 | 2.87 | 2.88 | 2.74 | 4.84 | 5.00 | 5.10 | 4.99 | 3.95 | 2.99 | 3.02 | 2.62 |
SD | 0.72 | 0.88 | 0.78 | 0.69 | 0.89 | 1.01 | 1.02 | 0.86 | 0.77 | 0.91 | 0.81 | 0.77 | 1.03 | 1.17 | 1.12 | 1.05 | 0.76 | 0.76 | 0.69 | 0.72 | 1.09 | 1.24 | 1.16 | 1.17 |
| Note. SE = Self-efficacy; ITV = Interest value; UTV = Utility value; ATV = Attainment value; EFC = Effort cost; OPC = Opportunity cost; EGC = Ego cost; EMC = Emotional cost. All correlation coefficients with an absolute value greater than .14 are statistically significant. | ||||||||||||||||||||||||
Measurement models | RMSEA [90% CI] | CFI | TLI | SRMR | χ2 | df |
|---|---|---|---|---|---|---|
Self-efficacy | ||||||
1 Configural invariance | 0.069 [0.058, 0.080] | 0.933 | 0.923 | 0.045 | 278.26 | 132 |
2 Metric invariance | 0.069 [0.058, 0.080] | 0.928 | 0.922 | 0.076 | 300.67 | 142 |
3 Scalar invariance | 0.076 [0.066, 0.086] | 0.908 | 0.907 | 0.080 | 354.44 | 152 |
Task values | ||||||
1 Configural invariance | 0.056 [0.050, 0.062] | 0.933 | 0.925 | 0.050 | 959.41 | 558 |
2 Metric invariance | 0.055 [0.049, 0.061] | 0.933 | 0.926 | 0.061 | 981.66 | 576 |
3 Scalar invariance | 0.057 [0.051, 0.062] | 0.927 | 0.923 | 0.068 | 1034.16 | 594 |
Costs | ||||||
1 Configural invariance | 0.063 [0.057, 0.069] | 0.922 | 0.907 | 0.060 | 1012.67 | 528 |
2 Metric invariance | 0.062 [0.056, 0.067] | 0.923 | 0.911 | 0.059 | 1023.22 | 544 |
3 Scalar invariance | 0.063 [0.058, 0.069] | 0.917 | 0.906 | 0.060 | 1079.91 | 560 |
| Note. RMSEA = root mean square error of approximation; CFI = comparative fit index; TLI = Tucker- Lewis index; SRMR = standardized root mean square residual; df = degrees of freedom; CI = confidence interval. Strong measurement invariance was observed across the three survey waves, except for self-efficacy, which exhibited weak measurement invariance. | ||||||
3.2 Profiles of undergraduates’ motivational beliefs
Using factor scores obtained from the initial CFAs, LPA models specifying between two and seven profiles were tested. Among these, the three-profile model provided the best fit across all three measurement waves (see Table 4). Although the BIC continued to decrease as more profiles were added, the rate of reduction leveled off after the three-profile model. Additionally, adding additional profiles resulted in smaller group sizes that were difficult to justify both statistically and theoretically. The VLMR value further indicated that additional profiles did not yield a significantly better model fit. The three-profile solution was also supported by high entropy values at each time point (T1 = 0.92, T2 = 0.93, T3 = 0.95), indicating clear and reliable classification of individuals into distinct profiles.
Time point | # of profiles | LL | #par | AIC | BIC | aBIC | Entropy | VLMR | BLRT | N smallest class |
|---|---|---|---|---|---|---|---|---|---|---|
Time 1 | ||||||||||
2 | -1869.15 | 25 | 3788.30 | 3874.47 | 3795.24 | 0.82 | 0.05 | < .001 | 112, 48.3% | |
3 | -1728.03 | 34 | 3524.07 | 3641.26 | 3533.49 | 0.92 | 0.26 | < .001 | 48, 20.7% | |
4 | -1643.88 | 43 | 3373.76 | 3521.97 | 3385.68 | 0.95 | 0.31 | < .001 | 3, 1.3% | |
5 | -1570.58 | 52 | 3245.17 | 3424.40 | 3259.58 | 0.96 | 0.23 | < .001 | 3, 1.3% | |
6 | -1508.17 | 61 | 3138.33 | 3348.58 | 3155.25 | 0.93 | 0.19 | < .001 | 2, 0.9% | |
7 | -1457.99 | 70 | 3055.97 | 3297.24 | 3075.38 | 0.94 | 1.00 | < .001 | 3, 1.3% | |
Time 2 | ||||||||||
2 | -2033.99 | 25 | 4117.97 | 4204.14 | 4124.90 | 0.92 | 0.06 | < .001 | 80, 34.5% | |
3 | -1903.22 | 34 | 3874.44 | 3991.63 | 3883.86 | 0.93 | 0.09 | < .001 | 19, 8.2% | |
4 | -1788.47 | 43 | 3662.94 | 3811.15 | 3674.86 | 0.97 | 0.65 | < .001 | 16, 6.9% | |
5 | -1662.73 | 52 | 3429.47 | 3608.70 | 3443.89 | 0.98 | 0.03 | < .001 | 1, 0.4% | |
6 | -1604.78 | 61 | 3331.55 | 3541.80 | 3348.46 | 0.98 | 0.81 | < .001 | 2, 0.9% | |
7 | -1550.47 | 70 | 3240.94 | 3482.21 | 3260.35 | 0.98 | 0.51 | < .001 | 1, 0.4% | |
Time 3 | ||||||||||
2 | -1993.38 | 25 | 4036.76 | 4122.93 | 4043.69 | 0.93 | < .001 | < .001 | 64, 27.6% | |
3 | -1741.11 | 34 | 3550.23 | 3667.41 | 3559.65 | 0.95 | < .001 | < .001 | 50, 21.6% | |
4 | -1650.43 | 43 | 3386.85 | 3535.06 | 3398.78 | 0.96 | 0.01 | < .001 | 18, 7.8% | |
5 | -1564.06 | 52 | 3232.11 | 3411.34 | 3246.53 | 0.94 | 0.42 | < .001 | 17, 7.3% | |
6 | -1510.43 | 61 | 3142.85 | 3353.10 | 3159.76 | 0.94 | 0.36 | < .001 | 17, 7.3% | |
7 | -1468.40 | 70 | 3076.80 | 3318.07 | 3096.21 | 0.94 | 0.17 | < .001 | 17, 7.3% | |
| Note. LL = model log likelihood; #par = number of free parameters; AIC = Akaike information criterion; BIC = Bayesian information criterion; aBIC = Sample-Size adjusted BIC; VLMR = Vuong-Lo-Mendell-Rubin likelihood ratio test; BLRT = Bootstrapped likelihood ratio test. The selected model values are bolded. | ||||||||||
Across the three time points, the motivational profiles identified were highly consistent and were labeled as Burdened (characterized by low self-efficacy and task values, and relatively high perceived costs, T1: 27%, T2: 32%, T3: 25%), Average-all (all expectancy-value-cost beliefs were at moderate levels, T1: 52%, T2: 51%, T3: 52%), and Positively motivated (defined by high self-efficacy and task values, and low perceived costs, T1: 21%, T2: 17%, T3: 24%).
3.3 Stability and changes in profile membership
Based on the results of the LPAs, we proceeded with the three-profile solution for the subsequent LTA. Figure 1 illustrates the time-invariant motivational profiles and the transition probabilities are resented in Table 5. The results reveal distinct patterns of stability and change across the semester. Between T1 and T2, profile membership was relatively stable for students in the Burdened (76.0%) and Average-all (71.0%) profiles, while stability was lower for the Positively motivated profile (62.4%). Between T2 and T3, the Average-all (72.7%) and Positively motivated (65.0%) profiles remained relatively stable, whereas the Burdened profile exhibited reduced stability (59.8%).
Fig. 1
The Time-Invariant Motivational Profiles with Factor Mean Scores
Latent Profile T2 | |||
|---|---|---|---|
Burdened | Average-all | Positively motivated | |
Latent Profile T1 | |||
Burdened | 0.760 | 0.227 | 0.014 |
Average-all | 0.199 | 0.710 | 0.091 |
Positively motivated | 0.017 | 0.359 | 0.624 |
Latent Profile T3 | |||
Burdened | Average-all | Positively motivated | |
Latent Profile T2 | |||
Burdened | 0.598 | 0.310 | 0.092 |
Average-all | 0.107 | 0.727 | 0.166 |
Positively motivated | 0.000 | 0.350 | 0.650 |
When transitions did occur, Burdened students were most likely to transition to the Average-all profile from T1 to T2 (22.7%) and from T2 to T3 (31.0%). Average-all students most commonly shifted to the Burdened profile from T1 to T2 (19.9%) and to the Positively motivated profile (16.6%) from T2 to T3. Positively motivated students were most likely to move to the Average-all profile in both transitions, with 35.9% from T1 to T2 and 35.0% from T2 to T3. Figure 2 illustrates the detailed shifts in membership from the beginning (T1) to midterm (T2) and from midterm to the end of the semester (T3). Overall, 53.8% of students remained in the same motivational profile across all three time points, suggesting moderate stability. Descriptive shifts in profile size over time still reveal dynamic patterns, from T1 to T2, the number of students in the Burdened profile increased (from 62 to 74), while the number in the Positively motivated profile declined (from 49 to 39). Between T2 and T3, this trend reversed, with the Burdened profile decreasing to 57 students and the Positively motivated profile rising to 55. These findings indicate that while many students remained stable in their motivational beliefs, others experienced both positive and negative shifts across the semester.
Fig. 2
Shifts in Membership Transitions Across Three Time Points and Profiles
3.4 Associations with gender
As shown in Table 6, the LTA model incorporating gender as a covariate revealed that gender did not significantly predict any of the transition probabilities, as indicated by the odds ratios (OR). This suggests that no significant gender-based differences were observed in students’ motivational profile changes over time.
Latent Profile T2 | |||
|---|---|---|---|
Burdened | Average-all | Positively motivated | |
Latent Profile T1 | |||
Burdened | 1.000 [1.000, 1.000] | 0.989 [0.500, 1.957] | 1.008 [0.418, 2.434] |
Average-all | 1.011 [0.511, 2.001] | 1.000 [1.000, 1.000] | 1.020 [0.436, 2.386] |
Positively motivated | 0.992 [0.411, 2.393] | 0.981 [0.419, 2.295] | 1.000 [1.000, 1.000] |
Latent Profile T3 | |||
Burdened | Average-all | Positively motivated | |
Latent Profile T2 | |||
Burdened | 1.000 [1.000, 1.000] | 1.129 [0.540, 2.362] | 0.730 [0.319, 1.672] |
Average-all | 0.886 [0.423, 1.852] | 1.000 [1.000, 1.000] | 0.646 [0.314, 1.332] |
Positively motivated | 1.370 [0.598, 3.137] | 1.547 [0.751, 3.187] | 1.000 [1.000, 1.000] |
4 Discussion
We adopted a longitudinal, person-centered approach to examine the stability and change in first-year undergraduates’ motivational profiles, defined by self-efficacy, task values, and perceived costs, across three time points during their first semester in an introductory psychology course. We also explored whether gender predicted profile membership transitions. Our analysis identified three consistent motivational profiles: Burdened, Average-all, and Positively motivated. Approximately half of the students remained in the same profile throughout the semester. Notably, shifts to less favorable profiles were more frequent during the first half of the semester, whereas positive transitions became more common in the second half. No significant predictive effect of gender on profile transitions was found.
4.1 Motivational profiles and their overall stability
Using LTA, we identified three distinct motivational profiles among the participants. The Burdened profile was characterized by the lowest levels of self-efficacy and task values, combined with the highest levels of perceived costs. In contrast, the Positively motivated profile, the smallest group, exhibited the highest self-efficacy and task values and the lowest perceived costs. The most common profile across all three time points was the Average-all profile, with moderate levels of all motivational beliefs. These profiles are consistent with previous person-centered research conducted among middle school and secondary students (Dietrich & Lazarides, 2019; Vinni-Laakso et al., 2022; Widlund et al., 2024).
Moreover, the LTA results revealed that students’ profile membership was moderately stable across the semester, with approximately 54% of students remaining in the same profiles across all three time points. This level of stability is lower than studies that tracked profiles over longer periods and two time points, where stability rates reached up to 82% (Widlund et al., 2024). However, our findings align with other research using three-wave designs, where the addition of a third measurement point often reveals greater volatility (Raufelder et al., 2022). These fluctuations likely reflect students’ ongoing adjustment to the academic and social demands of college life (Hilpert & Marchand, 2018; Kaplan & Garner, 2020). Accordingly, short-term analyses provide important insights into how motivational beliefs evolve during this transitional phase (Beymer & Rosenzweig, 2023).
4.2 Shifts in membership stability across early and late semester periods
Consistent with previous LTA studies based on two time points (Vinni-Laakso et al., 2022; Widlund et al., 2024), our three-wave analysis revealed both stability and change in motivational profiles across the semester. However, we also observed distinct patterns between the early and later parts of the semester.
During the first half of the semester, the Burdened profile emerged as the most stable, while the Positively motivated profile was least stable, with many students transitioning to less favorable profiles. Notably, many Positively motivated students moved to the Average-all profile, and Average-all students were likely to shift to the Burdened profile. These trends are in line with prior findings from STEM contexts, where freshmen’s self-efficacy and value beliefs often decline, and perceived costs rise during the early semester (Benden & Lauermann, 2022; Lee et al., 2024). This may reflect that students face considerable adjustment challenges during this period, possibly due to unfamiliar academic demands and competing priorities (Lee et al., 2024).
In contrast, the second half of the semester showed more positive transitions. Students in the Burdened profile most often transitioned to Average-all, and those in Average-all shifted toward Positively motivated. These changes may reflect students’ increasing familiarity with course expectations, greater confidence from accumulated feedback, and improved time management (Putwain & Sander, 2016; Lee et al., 2024). The findings underscore the second half of the semester as a critical window for intervention, where students may be more receptive to support that fosters motivation and resilience.
Understanding when and how students' motivational profiles change offers practical implications for identifying those at risk and providing timely support. The contrast in transitions between early and later semester points to the need for targeted interventions at key moments, especially around midterms, to help students recover from early setbacks and reinforce adaptive motivational patterns.
4.3 Gender and changes in profile membership
Our findings showed that gender was not a significant predictor of motivational profile transitions. Transition probabilities were largely homogeneous across male and female students, consistent with previous research (Widlund et al., 2024). This suggests that the profile patterns and shifts observed are broadly applicable to both genders.
Two possible explanations may account for this result. First, the course context of introductory psychology is generally more gender-balanced and may not exhibit the same disparities observed in STEM fields, where gender gaps in motivation are well-documented (Eccles & Wigfield, 2023). Second, the higher proportion of female students in our sample may have influenced variability and statistical power. Given the limited research on gender effects in motivational transitions, further studies in varied academic domains and with balanced samples are needed to better understand the role of gender in motivational development.
4.4 Limitations and future directions
This study has several limitations that should be considered when interpreting the results. First, while our findings align with previous research conducted in various educational contexts and countries, the relatively small sample size and the specific Chinese context limit the generalizability of the results. Another limitation is that our analyses focused on motivational profile transitions within an introductory psychology course. This does not rule out the possibility of domain-specific variations that may emerge across different academic disciplines. Research has shown that motivational dynamics can vary significantly between subjects, such as STEM fields and the humanities (Harackiewicz et al., 2008; Simpkins et al., 2006), which could uniquely shape students’ motivational profiles. Further exploration of these domain-specific variations is necessary to develop tailored educational interventions that foster optimal motivational climates across various academic contexts.
Several limitations should be acknowledged. First, although our findings align with research conducted in other educational settings, the present study was conducted at a single Chinese university with a relatively small sample, which may limit the generalizability of the results. Second, our analyses focused on a single academic domain which may not reflect domain-specific variations in motivational dynamics. Research has shown that students’ motivation profiles can differ substantially across disciplines, such as STEM versus humanities (Harackiewicz et al., 2008; Simpkins et al., 2006). Future studies should explore discipline-specific motivational trajectories and how these may require tailored intervention strategies. Additionally, future work could investigate the influence of other contextual factors such as academic workload and social support that may help explain shifts in profile membership.
5 Conclusion
This longitudinal, person-oriented study provides insight into how college freshmen’s motivational profiles evolve over the course of a semester. By integrating expectancy beliefs, task values, and perceived costs, the study contributes to a deeper understanding of the situated and dynamic nature of motivation as conceptualized by SEVT (Raufelder et al., 2022). A key strength of this study is its use of three time points, which revealed distinct patterns of stability and change across the early and later phases of the semester. The findings highlight the importance of timing interventions to support students during critical transitions.
References
Asparouhov T, Muthen B. (2021). Auxiliary variables in mixture modeling: Using the BCH method in Mplus to estimate a distal outcome model and an arbitrary secondary model. Mplus Web Notes: 21, Version 11. https://www.statmodel.com/examples/webnotes/webnote21.pdf
Benden DK, Lauermann F. Students’ motivational trajectories and academic success in math-intensive study programs: Why short-term motivational assessments matter. J Educ Psychol. 2022;114(5):1062–85. https://doi.org/10.1037/edu0000708.
Beymer PN, Rosenzweig EQ. Predictors of short-term trajectories of students’ expectancy-value motivational beliefs in introductory calculus courses. Contemp Educ Psychol. 2023;75:102236. https://doi.org/10.1016/j.cedpsych.2023.102236.
Bong M. Effects of parent-child relationships and classroom goal structures on motivation, help-seeking avoidance, and cheating. J Exp Educ. 2008;76(2):191–217. https://doi.org/10.3200/JEXE.76.2.191-217.
Bong M, Skaalvik E. Academic self-concept and self-efficacy: How different are they really? Educational Psychol Rev. 2003;15:1–40. https://doi.org/10.1016/j.lindif.2009.05.004.
Brislin RW. Back-translation for cross-cultural research. J Cross-Cult Psychol. 1970;1(3):185–216. https://doi.org/10.1177/135910457000100301.
Chen FF. Sensitivity of goodness of fit indexes to lack of measurement invariance. Struct Equation Modeling: Multidisciplinary J. 2007;14(3):464–504. https://doi.org/10.1080/10705510701301834.
Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing measurement invariance. Struct Equation Modeling: Multidisciplinary J. 2002;9(2):233–55. https://doi.org/10.1207/s15328007sem0902_5.
Conley AM. Patterns of motivation beliefs: Combining achievement goal and expectancy-value perspectives. J Educ Psychol. 2012;104(1):32–47. https://doi.org/10.1037/a0026042.
Dietrich J, Lazarides R. Gendered development of motivational belief patterns in mathematics across a school year and career plans in math-related fields. Front Psychol. 2019;10:1472. https://doi.org/10.3389/fpsyg.2019.01472.
Eccles JS, Adler TF, Futterman R, Goff SB, Kaczala CM, Meece JL, Midgley C, Spence JT, editors. Achievement and achievement motivation (75–146). W. H. Freeman.
Eccles JS. (1984). Sex differences in achievement patterns. In T. Sonderegger, editor, Nebraska Symposium on Motivation (Vol. 32, pp. 97–132). University of Nebraska Press.
Eccles JS, Wigfield A. Motivational beliefs, values, and goals. Ann Rev Psychol. 2002;53:109–32. https://doi.org/10.1146/annurev.psych.53.100901.135153.
Eccles JS, Wigfield A. From expectancy-value theory to situated expectancy-value theory: A developmental, social cognitive, and sociocultural perspective on motivation. Contemp Educ Psychol. 2020;61:101859. https://doi.org/10.1016/j.cedpsych.2020.101859.
Eccles JS, Wigfield A. Expectancy-value theory to situated expectancy-value theory: Reflections on the legacy of 40 + years of working together. Motivation Sci. 2023;9(1):1–12. https://doi.org/10.1037/mot0000275.
Enders CK. Applied missing data analysis. Guilford; 2010.
Fryer LK, Ainley M. Supporting interest in a study domain: A longitudinal test of the interplay between interest, utility-value, and competence beliefs. Learn Instruction. 2019;60:252–62. https://doi.org/10.1016/j.learninstruc.2017.11.002.
Gaspard H, Jiang Y, Piesch H, Nagengast B, Jia N, Lee J, Bong M. Assessing students’ values and costs in three countries: Gender and age differences within countries and structural differences across countries. Learn Individual Differences. 2020;79:101836. https://doi.org/10.1016/j.lindif.2020.101836.
Harackiewicz JM, Durik AM, Barron KE, Linnenbrink-Garcia L, Tauer JM. The role of achievement goals in the development of interest: Reciprocal relations between achievement goals, interest, and performance. J Educ Psychol. 2008;100(1):105–22. https://doi.org/10.1037/0022-0663.100.1.105.
Hilpert JC, Marchand GC. Complex systems research in educational psychology: Aligning theory and method. Educational Psychol. 2018;53(3):185–202. https://doi.org/10.1080/00461520.2018.1469411.
Jiang Y, Zhang L. High school students’ expectancy, value, and cost profiles and their relations with engagement and achievement in Math and English. Learn Individual Differences. 2023;101:102252. https://doi.org/10.1016/j.lindif.2022.102252.
Jiang Y, Kim S, Bong M. The role of cost in adolescent students’ maladaptive academic outcomes. J Sch Psychol. 2020;83:1–24. https://doi.org/10.1016/j.jsp.2020.08.004.
Jiang Y, Rosenzweig EQ, Gaspard H. An expectancy-value-cost approach in predicting adolescent students’ academic motivation and achievement. Contemp Educ Psychol. 2018;54:139–52. https://doi.org/10.1016/j.cedpsych.2018.06.005.
Kaplan A, Garner JK. Steps for applying the complex dynamical systems approach in educational research: A guide for the perplexed scholar. J Experimental Educ. 2020;88(3):486–502. https://doi.org/10.1080/00220973.2020.1745738.
Lazarides R, Dicke A-L, Rubach C, Eccles JS. Profiles of motivational beliefs in math: Exploring their development, relations to student-perceived classroom characteristics, and impact on future career aspirations and choices. J Educ Psychol. 2020;112(1):70–92. https://doi.org/10.1037/edu0000368.
Lazarides R, Dietrich J, Taskinen PH. Stability and change in students’ motivational profiles in mathematics classrooms: The role of perceived teaching. Teach Teacher Educ. 2019;79:164–75. https://doi.org/10.1016/j.tate.2018.12.016.
Lee SY, Christiaans E, Robinson KA. Development in context: Comparing short-term trajectories of expectancy, task values, and costs in four university STEM courses. Contemp Educ Psychol. 2024;78:102282. https://doi.org/10.1016/j.cedpsych.2024.102282.
Lee SY, Friedman S, Christiaans E, Robinson KA. Valuable but costly? University students’ expectancy-value-cost profiles in introductory chemistry courses. Contemp Educ Psychol. 2022;69:102056. https://doi.org/10.1016/j.cedpsych.2022.102056.
Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo Simulation Study. Struct Equation Modeling: Multidisciplinary J. 2007;14(4):535–69. https://doi.org/10.1080/10705510701575396.
Perez T, Cromley JG, Kaplan A. The role of identity development, values, and costs in college STEM retention. J Educ Psychol. 2014;106(1):315–29. https://doi.org/10.1037/a0034027.
Perez T, Wormington SV, Barger MM, Schwartz-Bloom RD, Lee Ykyung, Linnenbrink-Garcia L. Science expectancy, value, and cost profiles and their proximal and distal relations to undergraduate science, technology, engineering, and math persistence. Sci Educ. 2019;103(2):264–86. https://doi.org/10.1002/sce.21490.
Putwain DW, Sander P. Does the confidence of first-year undergraduate students change over time according to achievement goal profile? Stud High Educ. 2016;41(2):381–98. https://doi.org/10.1080/03075079.2014.934803.
Raufelder D, Hoferichter F, Hirvonen R, Kiuru N. How students’ motivational profiles change during the transition from primary to lower secondary school. Contemp Educ Psychol. 2022;71:102117. https://doi.org/10.1016/j.cedpsych.2022.102117.
Simpkins SD, Davis-Kean PE, Eccles JS. Math and science motivation: A longitudinal examination of the links between choices and beliefs. Dev Psychol. 2006;42(1):70–83. https://doi.org/10.1037/0012-1649.42.1.70.
Vinni-Laakso J, Upadyaya K, Salmela-Aro K. Associations between adolescent students’ multiple domain task value-cost profiles and STEM aspirations. Front Psychol. 2022;13:951309. https://doi.org/10.3389/fpsyg.2022.951309.
Watt HM, Bucich M, Dacosta L. Adolescents’ motivational profiles in mathematics and science: Associations with achievement striving, career aspirations and psychological wellbeing. Front Psychol. 2019;10:990. https://doi.org/10.3389/fpsyg.2019.00990.
Widlund A, Tuominen H, Korhonen J. Motivational profiles in mathematics–stability and links with educational and emotional outcomes. Contemp Educ Psychol. 2024;76:102256. https://doi.org/10.1016/j.cedpsych.2024.102256.
Wigfield A, Cambria J. Students’ achievement values, goal orientations, and interest: Definitions, development, and relations to achievement outcomes. Dev Rev. 2010;30(1):1–35. https://doi.org/10.1016/j.dr.2009.12.001.
A
Wigfield A, Eccles JS. (2020). 35 years of research on students’ subjective task values and motivation: A look back and a look forward. In A. J. Elliot, editor, Advances in motivation science (Vol. 7, pp. 161–198). Elsevier. https://doi.org/10.1016/bs.adms.2019.05.002Wigfield A, Rosenzweig EQ, Eccles J. Achievement values: Interactions, interventions, and future directions. In: Elliot A, Dweck C, Yeager D, editors. Handbook of competence and motivation: Theory and application. 2nd ed. Guilford Press; 2017.
Wormington SV, Linnenbrink-Garcia L. A new look at multiple goal pursuit: The promise of a person-centered approach. Educational Psychol Rev. 2017;29:407–45. https://doi.org/10.1007/s10648-016-9358-2.
Xie K, Vongkulluksn VW, Cheng S-L, Jiang Z. Examining high-school students’ motivation change through a person-centered approach. J Educ Psychol. 2022;114(1):89–107. https://doi.org/10.1037/edu0000507.
Zhang L, Jiang Y, Chen S. Longitudinal interrelations among self-efficacy, interest value, and effort cost in adolescent students’ English achievement and future choice intentions. Contemp Educ Psychol. 2023;73:102176. https://doi.org/10.1016/j.cedpsych.2023.102176.
Zuidema PM, Hornstra L, Schuitema J, Poorthuis AM. Attributional profiles: Considering multiple causal attributions for success and failure at the beginning of secondary school. Contemp Educ Psychol. 2023;73:102164. https://doi.org/10.1016/j.cedpsych.2023.102164.
Note
P1 = Burdened (NT1 = 62; NT2 = 74; NT3 = 57); P2 = Average-all (NT1 = 121; NT2 = 119; NT3 = 120); P3 = Positively motivated (NT1 = 49; NT2 = 39; NT3 = 55). From T1 to T3, N = 35 (15.1%) consistently in P1, N = 66 (28.4%) in P2, and N = 24 (10.3%) in P3, respectively.