Study the mean, standard deviation, and pairwise product-moment correlation coefficients of the variables as shown in Table 1.
Table 1
Correlation coefficients of the observed variables
Variables
|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
12
|
1. Simple Epistemic
|
-
|
|
|
|
|
|
|
|
|
|
|
|
2. Determined Epistemic
|
0.13**
|
-
|
|
|
|
|
|
|
|
|
|
|
3. Source of Epistemic
|
0.25***
|
0.29***
|
-
|
|
|
|
|
|
|
|
|
|
4. Epistemic Dialectics
|
0.14**
|
0.44***
|
0.35***
|
-
|
|
|
|
|
|
|
|
|
5. Curiosity
|
0.13**
|
0.06
|
-0.02
|
0. 10*
|
-
|
|
|
|
|
|
|
|
6. Boredom
|
-0.24***
|
-0.01
|
-0.05
|
-0.03
|
-0.46***
|
-
|
|
|
|
|
|
|
7. Confusion
|
-0.11*
|
-0.02
|
-0.07
|
-0.03
|
0.05
|
0.11*
|
-
|
|
|
|
|
|
8. Anxiety
|
-0.01
|
0.11*
|
-0.05
|
0.06
|
0.31***
|
-0.28***
|
0.33***
|
-
|
|
|
|
|
9. Excitement
|
-0.03
|
-0.23**
|
-0.21**
|
-0.18**
|
0.22**
|
0.05
|
0.07
|
0.12**
|
-
|
|
|
|
10. Cognitive Engagement
|
0.09
|
0.13**
|
0.04
|
0.19**
|
0.15**
|
-0.05
|
0.02
|
0.05
|
-0.01
|
|
|
|
11. Emotional Engagement
|
0.12**
|
0.02
|
-0.09
|
0.07
|
0.21**
|
-0.17**
|
0.03
|
0.08
|
0.07
|
0.25***
|
-
|
|
12. Behavioral Engagement
|
0.01
|
0.11*
|
0.01
|
0.12*
|
0.11*
|
-0.04
|
0.03
|
0.05
|
-0.11*
|
0.31***
|
0.19**
|
-
|
N
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
558
|
M
|
3.33
|
4.17
|
4.05
|
4.22
|
3.28
|
2.18
|
2.87
|
3.57
|
2.19
|
3.52
|
3.46
|
3.33
|
SD
|
0.51
|
0.46
|
0.57
|
0.36
|
0.88
|
0.47
|
0.69
|
0.75
|
0.92
|
0.53
|
0.46
|
0.48
|
Note: * stands for p < 0.05; ** stands for p < 0.01; *** stands for p < 0.001 |
Based on Baron and Kenney [35] suggestion, the overall model fit: χ2(13, N = 558) = 52.74, p < 0.001; χ2/df = 4.06; RMSEA = 0.04; SRMR = 0.03, indicating a good overall model fit. In terms of within-model structural fit: the combination reliability of epistemic cognition and student engagement is 0.62 and 0.42 respectively; average variance extracted is 0.33 and 0.21, indicating a good within-structural fit. Considering basic fit, overall model fit, and within-model structural fit, the study results show a significant direct effect of epistemic cognition on student engagement among college students. The standardized regression coefficient of epistemic cognition on student engagement is 0.35, p < 0.001, indicating that epistemic cognition positively influences student engagement. Hypothesis 1 of this study is supported.
Followed the suggestions of Collier[36], we used χ2, χ2/df, RMSEA, and SRMR as evaluation indicators to assess the overall model fit. The statistical results of the overall model fit with positive emotions as a mediator are: χ2 = 133.89, p < 0.001; χ2/df = 3.47; RMSEA = 0.054; SRMR = 0.049. Regarding the test of within-model structural validity, the combination reliability (CR) of epistemic cognition, positive epistemic emotions, and student engagement is 0.66, 0.71, and 0.69 respectively, and the average variance extracted (AVE) is 0.52, 0.59, and 0.55, indicating a good fit of the within-structural model. The study diagram of the mediation effect pathway of epistemic cognition, positive epistemic emotions, and student engagement among college students is shown in Fig. 1.
According to Figure 1, the standardized regression coefficient of epistemic cognition on positive epistemic emotions is 0.16 (p < 0.05), indicating support for the H2 of this study. Furthermore, the standardized regression coefficient of positive epistemic emotions on student engagement is 0.26 (p < 0.001), showing a significant positive influence of positive epistemic emotions on student engagement after controlling for the effect of epistemic cognition. Based on Baron and Kenny's [35] criteria for assessing mediation effects, the mediation effect is significant, with a value of 0.042. To test the mediation effects, this study employed the Bootstrap method with bias-corrected confidence intervals using AMOS, conducting 5,000 iterations. The results of the mediation effects are presented in Table 2. The mediated effect of epistemic cognition through positive epistemic emotions on student engagement is 0.040, with a 95% confidence interval of [0.007, 0.096], indicating a significant mediation effect. H4 of this study is supported.
Table 2
the mediation effect models of epistemic emotions
Mediation Effect Model
|
Direct Effect
|
Indirect Effect
|
95% Confidence Interval
|
p
|
Conclusion
|
Lower Bound
|
Upper Bound
|
EC → PEE → SE
EC → NEE → SE
|
0.293
0.345
|
0.040
0.012
|
0.007
-0.004
|
0.096
0.059
|
0.012
0.151
|
Partial Mediation Effect
|
No Mediation Effect
|
Note: Presented in standardized regression coefficients. EC = Epistemic Cognition. PEE = Positive Epistemic Emotions. SE = Student Engagement. NEE = Negative Epistemic Emotions. |
The statistical results of the overall model fit with negative emotions as a mediator are: χ2 = 151.81, p < 0.001; χ2/df = 5.88; RMSEA = 0.08; SRMR = 0.07. As the fit of the models was not ideal, modifications were suggested based on the Modification Index provided by AMOS. After the modifications, the overall model fit for the model with negative emotions as a mediator showed significant results with 92.66 (p > 0.05), but χ2/df = 3.68, RMSEA = 0.06, and SRMR = 0.05, the criteria for acceptable fit, demonstrating good overall model fit for the mediation effect.
Regarding the test of within-model structural validity, the combination reliability (CR) of epistemic cognition, negative epistemic emotions, and student engagement is 0.65, 0.52, and 0.56 respectively; the average variance extracted (AVE) is 0.44, 0.39, and 0.43, indicating a good fit of the within-structural model. The study diagram of the mediation effect pathway of epistemic cognition, negative epistemic emotions, and student engagement among college students is shown in Figure 2.
In Fig. 2, the standardized regression coefficient of epistemic cognition on negative epistemic emotions is -0.11 (p < 0.05). H3 got support for this study. However, the standardized regression coefficient of negative epistemic emotions on student engagement is -0.09 (p > 0.05), suggesting that after controlling for the effect of epistemic cognition, negative epistemic emotions do not have a significant impact on student engagement. According to Baron and Kenny (1986)[35] criteria for assessing mediation effects, this indicates that there is no mediation effect. The mediated effect of epistemic cognition through negative epistemic emotions on student engagement is 0.001, with a 95% confidence interval of [-0.004, 0.059], which does not reach the significance level. H5 of this study is not supported.