Pre-service teachers’ empathy and attitudes toward inclusive education—The chain mediating role of teaching motivation and inclusive education efficacy

PLOS One | https://doi.org/10.1371/journal.pone.0321066 April 29, 2025 11 / 18 was determined based on Chin’s (1998) recommendation that the value should be greater than 0.1 [58]. As shown in Table 4, only Hypothesis 2 had an R2 value of less than 0.1, which suggests that empathy has a low explanatory power for teaching motivation. The effect size (f2) determines the magnitude of the omitted construct’s influence on a specific endogenous construct, with thresholds of 0.02, 0.15, and 0.35 representing small, medium, and large effects, respectively. As shown in Table 4, the f2 values for direct effect hypotheses 1, 2, 3, 5, 6, and 8 all exceed 0.02, indicating a moderate effect [59]. PLS-Predict The PLS prediction procedures were conducted following the methodology proposed by Shmueli et al. (2019) to evaluate the model’s out-of-sample projecting capabilities [60]. The PLS analysis’s root mean squared error (RMSE) was a crucial criterion for assessing predictive performance. This RMSE was compared to the values obtained from a linear model (LM) RMSE, as Danks and Ray (2018) suggested [61]. The expectation was that the PLS analysis would yield lower prediction errors compared to the naïve benchmark (PLS-LM), indicating a medium predictive power. The results in Table 5 reveal that the minority of indicators in the PLS-SEM analysis yield higher prediction errors compared to the naive LM benchmark, thus this indicates a medium predictive power. Furthermore, the data in Table 5 indicates that all Q2 values are higher than 0, suggesting sufficient pre- dictive relevance.

Fig 2. Hypothesis testing. https://doi.org/10.1371/journal.pone.0321066.g002