Abstract
This study aims to revise the emotion and motivation self-regulation questionnaire(EMSR-Q) and test its validity among Chinese middle school students. A total of 780 middle school students were selected for item analysis based on the classical test theory (CTT) and multidimensional item response theory (MIRT) respectively and exploratory factor analysis. 615 middle school students were selected for confirmatory factor analysis and criterion validity test. The full sample was used for the reliability test. The results show that the Chinese version of EMSR-Q retains 19 items. The Pearson correlation coefficients and the corrected item-total correlation coefficients between each item and the total score of each dimension are greater than 0.4. The independent sample t-test for high and low groups was significant. The results of the confirmatory factor analysis support the second-order second-factor model of the English version of EMSR-Q. There are significant positive correlations between the dimension of the Chinese version of EMSR-Q and the dimensions of the relevant scale (p < 0.01). The internal consistency reliability coefficient ranged from 0.671 to 0.861 and the split-half reliability coefficients ranged from 0.726 to 0.809. The discrimination and difficulty of the 19 items are good, with the discrimination coefficients α of the items greater than 1.177 and the difficulty coefficient β range of the five options is within [-3,3]. The results show that the revised Chinese version of EMSR-Q indicators meets the psychometric requirements and can be used to measure the structure and characteristics of Chinese middle school students’ emotion and motivation self-regulation. It can explore the possible influence of emotion and motivation self-regulation on academic development, which provides an important research tool for promoting the intervention research of emotion and motivation self-regulation, with a broad prospect of educational application.
Keywords: Emotion and motivation self-regulation, Academic, Middle school student, Reliability, Validity
Introduction
Self-regulation in learning refers to the process of cognitive, motivational, emotional, and behavioral regulation of learners in completing specific learning tasks [1]. Emotion regulation in learning is the process of adjustment in which individuals intentionally or automatically respond to their emotions, when to produce emotions and how to express and change emotional experience in the process of listening, homework and review [2]. It includes conscious or unconscious changes in all aspects of emotional response [2]. Motivation adjustment in learning is that individuals consciously take certain activities to intervene, manage and control their own will, so as to complete a specific learning activity or goal [3]. Studies have shown that emotion self-regulation and motivation self-regulation in learning play an important role in students’ learning behavior and learning performance [4, 5]. Mindfulness, cognitive reappraisal and goal approach motivation used in learning were positively correlated with academic engagement and academic performance [6–8].
However, emotion and motivation do not act on learning alone. They often exist simultaneously and affect learning together. Studies have shown that cognitive evaluation, task value and emotion can jointly affect academic engagement and academic performance [9, 10].Therefore, some studies [11] have investigated the role of motivation in emotion regulation from the perspective of synergy between the two. They take the overall operation process of the two as a variable which can regulate learning and try to clarify the regularity of the influence of emotion and motivation self-regulation on students’ learning.
Emotion and motivation self-regulation [11] refers to the self-regulation style configured by the frequently repeated emotions and goal orientation when individuals deal with challenging experiences in the academic environment. The theoretical basis of emotion and motivation self-regulation is Boekaerts’s dual processing theory [12] and Elliot’s trichotomy goal orientation theory [13].
The dual processing model of Boekaerts suggests that the self-regulation system in learning has two self-regulation pathways, namely, the path of mastery and the path of well-being. The individuals who adopt the path of mastery tend to focus on the learning task in order to master knowledge and skills. The path of mastery is activated when individuals feel positive emotions or their personal interests consistent with the current task goal. Individuals who adopt the path of well-being are more self-concerned, whose goals are to avoid self-threat, resource loss and self-harm. When individuals feel negative emotions or their personal goals conflict with the task goals in the learning process, the path of well-being will be activated. Individuals achieve three goals of self-regulation through the two paths. One is to achieve the goal of growing knowledge and skills through the path of mastery. The second is to achieve the goal of preventing self-threat and resource loss through the path of well-being. The third is to change from the path of well-being to the path of mastery in order to protect one’s academic goal commitment [12]. The dual processing theory believes that the influence of emotion on study is determined by students’ goal orientation. Students activate different self-regulation paths according to the type of goals they pursue, and these goals in turn are activated by the emotion caused by the task [14].
Elliot’s trichotomy goal orientation theory suggests that achievement goals are divided into mastery goal, performance approach goal and performance avoidance goal. Individuals who hold mastery goal will focus on grasping and understanding of learning tasks and developing their abilities. Those who hold performance approach goal will show their abilities in learning tasks. Those who hold avoidance approach goal will avoid exposing their deficiencies in their learning tasks [13]. The combination of these two theories correlates and organizes strategies for emotion and motivation self-regulation, which allows goals to play a central role in the form of self-regulation and the organization of emotion [15].
Based on the two theoretical viewpoints, Alonso-Tapia et al. divided emotion and motivation self-regulation into avoidance self-regulation style (ASRS) and learning self-regulation style (LSRS), and proposed a second-order two-factor model of emotion and motivation self-regulation (Fig. 1) [11]. Among them, avoidance self-regulation styles include negative self-regulation of stress, avoidance oriented self-regulation, and performance oriented self-regulation. Learning self-regulation styles include performance oriented self-regulation, positive regulation of motivation and process oriented self-regulation. Given previous research results, Performance oriented self-regulation may have both positive and negative effects on academic performance, so performance oriented self-regulation exists in both styles of self-regulation [11].
Fig. 1.
Second-order two-factor model of emotion and motivation self-regulation [11]
Based on the second-order two-factor structural model, Alonso-Tapia et al. compiled the Emotion and Motivation Self-Regulation Questionnaire(EMSR-Q) with 20 items [11]. EMSR-Q was specifically designed to identify emotion and motivation self-regulation processes during academic tasks [11]. The questionnaire was divided into two second-order factors of avoidance self-regulation style (ASRS) and learning self-regulation style (LSRS), which included five first-order factors of negative self-regulation of stress, avoidance oriented self-regulation, performance oriented self-regulation, positive regulation of motivation and process oriented self-regulation. The Cronbach α coefficients of avoidance self-regulation style (ASRS) and learning self-regulation style (LSRS) were 0.84 and 0.77 in Spanish middle school students. The α coefficients of the five dimensions ranged from 0.69 to 0.79. Later, Alonso-Tapia et al. further revised the questionnaire, and the test results showed that the reliability coefficients of the first and second order factors of the revised scale ranged from 0.66 to 0.90 in the samples of Colombian and Spanish college students. It conforms to the theoretical model and has good structural validity [15].
Studies on students’ emotion self-regulation and motivation self-regulation in learning have been continuously promoted in China, covering the pathways of emotion self-regulation and motivation self-regulation on study and intervention strategies [16–18]. However, there is still a lack of research on the synergistic effect of emotion self-regulation and motivation self-regulation in China, and a lack of emotion and motivation self-regulation measurement tools for Chinese middle school students. Based on the current education practice for students’ emotion and motivation regulation effect is unsatisfactory, in order to investigate the middle school students’ emotion and motivation self-regulation and give middle school students effective guidance, we need to have a suitable scale for Chinese middle school students’ emotion and motivation self-regulation. EMSR-Q has been validated in secondary and university groups in Colombia and Spain. The validity test is required whether EMSR-Q is applicable to Chinese middle school students. For this, the study revised and examined the EMSR-Q to develop a measurement tool applicable to Chinese middle school students. On one hand, we test the cross-cultural stability of the emotion and motivation self-regulation second-factor model and expand the application scope and influence of EMSR-Q. On the other hand, with the help of EMSR-Q, it is beneficial to deepen the study of the synergistic effect of emotion self-regulation and motivation self-regulation for Chinese middle school students.
Current researches help to reveal emotion and motivation self-regulation intrinsic structure and characteristics in the Chinese educational environment. They find different types of emotion and motivation self-regulation styles and further clarify the intrinsic influencing mechanism of emotion and motivation on Chinese middle school students’ learning. They expand self-regulation learning theory, exploring and developing the intervention model which can scientifically guide students how to learn. In terms of practical application, EMSR-Q has for the first time focused on the role of motivation in emotion regulation among Chinese middle school students from a collaborative perspective. This questionnaire can help teachers, parents and students themselves to understand the characteristics of emotion and motivation self-regulation in learning and know the possible effects of emotion and motivation self-regulation on academic development. Therefore, students can adjust their emotion and motivation self-regulation, provide tool support for targeted tutoring and intervention. It can be used to detect the effect of emotion and motivation self-regulation training and finally improve their academic performance.
In conclusion, this study aims to verify the rationality of the theoretical model of emotion and motivation self-regulation in Chinese middle school students and to examine the cultural applicability of EMSR-Q among Chinese middle school students. Firstly, this study conducted an item analysis based on the Classical Test Theory (CTT), which was based on the process of previous questionnaire revision [18–20]. Later, exploratory factor analysis and confirmatory factor analysis were performed to explore the factor structure of EMSR-Q and to determine the extent to which the observed variables were associated with the latent factors. Therefore, referring to the dimensions of criterion scales used in the English version of EMSR-Q [11], this study analyzed the correlation between the relevant dimensions of emotion and motivation self-regulation with mastery goal, performance approach goal, performance avoidance goal, external and internal motivation in learning and academic emotion regulation strategy. Then we conducted internal consistency reliability and split-half reliability analysis of EMSR-Q. Finally, the item analysis was performed based on the Multidimensional Item Response Theory (MIRT). Through the above revision process, we tested whether the revised Chinese version of EMSR-Q met the psychometric requirements, which provided an empirical basis for the use of EMSR-Q in Chinese middle school students.
Method
Sample
The first sampling: the purpose is to evaluate and improve the translation content of the questionnaire. Convenient sampling was used to investigate Chinese middle school students (12–18 years). Informed consents were obtained from all individual participants and their parents/legal guardians prior to their participation in the study. 185 samples were collected in a city of Anhui Province in China, among which 170 were valid. 15 samples which were obviously chaotic and answered in a regular way were excluded. 63 students were boys, 97 students were girls, and 10 students did not report their gender. 82 students were from junior school and 88 students were from senior school.
The second sampling: The purpose is to test the psychometric indicators of the questionnaire. Convenient sampling was used to investigate Chinese middle school students (12–18 years). Informed consents were obtained from all individual participants and their parents/legal guardians prior to their participation in the study. The samples in Anhui Province in China were selected. Anhui Province is located in the central region of China, which is at the north and south intersection. 1450 questionnaires were collected in three cities in the northern, central and southern parts of Anhui Province in China, among which 1395 were valid. 55 samples which were obviously chaotic and answered in a regular way were excluded. In the data analysis, 1395 participants were randomly divided into two groups: Sample 1 included 780 participants. 375 students were boys, 322 students were girls, and 83 students did not report their gender. 371 students were from junior school and 409 students were from senior school (see Table 1). Sample 1 was used to complete item analysis based on CTT and MIRT. Sample 2 included 615 participants. 250 students were boys, 286 students were girls, and 79 students did not report their gender. 304 students were from junior school and 311 students were from senior school. Sample 2 was used for confirmatory factor analysis and criterion validity test. The whole sample was used for internal consistency reliability and split-half reliability test [20, 21].
Table 1.
Basic information of Sample
| Gender | Grade | ||||
|---|---|---|---|---|---|
| Boy | Girl | Did Not Report | Junior School | Senior School | |
| Full sample(1395) | 625 | 608 | 162 | 675 | 720 |
| Group 1 of sample(780) | 375 | 322 | 83 | 371 | 409 |
| Group 2 of sample(615) | 250 | 286 | 79 | 304 | 311 |
Instruments
The Emotion and Motivation Self-Regulation Questionnaire (EMSR-Q). The study used the Emotion and Motivation Self-Regulation Questionnaire (EMSR-Q) compiled by Alonso-Tapia et al. [11] with 20 items. The questionnaire is divided into five dimensions: avoidance oriented self-regulation, performance oriented self-regulation, negative self-regulation of stress, positive regulation of motivation and process oriented self-regulation. Each dimension contains 4 items. The questionnaire is used by five-level Likert scoring, in which 1 means completely inconsistent and 5 means completely consistent. All the items are adopted positive scoring. The dimension score is obtained by averaging all the item scores on each dimension. The higher the dimension score, the more orientation that the student is to adopt on the dimension. In this study, the internal consistency coefficients of the five dimensions of this questionnaire were 0.788, 0.671, 0.819, 0.785, and 0.752, respectively.
Achievement Goal Oriented Questionnaire. The study adopted the Achievement Goal Oriented Questionnaire compiled by Elliot and Church [22] with a total of 16 items. It can be divided into three subscales: performance approach goal, performance avoidance goal and mastery goal. There are 6 items in the dimension of the performance approach goal, 5 items in the dimension of the performance avoidance goal and 5 items in the mastery goal. The questionnaire is used by five-level Likert scoring, in which 1 means completely inconsistent and 5 means completely consistent. All the items are adopted positive scoring. The subscale score is obtained by averaging all the item scores on each dimension. The subscale score is obtained by averaging all the item scores on each dimension. The higher the subscale score, the more orientation that the student is to adopt on the dimension. The Chinese version of the questionnaire [23] was used in this study. The internal consistency coefficient of the questionnaire in this study was 0.773, and the internal consistency coefficients of the three subscales were 0.826, 0.722, and 0.811, respectively.
Self-regulated Learning Strategies Questionnaire for Middle School Students. The study adopted a subscale of extrinsic motivation, academic emotion regulation strategies and intrinsic motivation in the Self-regulated Learning Strategies Questionnaire for Middle School Students [24] with 10 items. The extrinsic motivation dimension consists of 4 items, which measures the degree of students’ learning for performance. The academic emotion regulation strategy dimension consists of 4 items, which measures the performance of coping with emotion; The intrinsic motivation dimension consists of 2 items, which measures learning interest. The questionnaire is used by seven-level Likert scoring, in which 1 means completely inconsistent and 7 means completely consistent. The subscale score is obtained by averaging all the item scores on each dimension. The higher the subscale score, the more orientation that the student is to adopt on each dimension. The internal consistency coefficient of the scale in this study was 0.770, and the internal consistency coefficients of the three subscales were 0.832, 0.797, and 0.632, respectively.
Procedures
Referring to the translation procedures adopted by Zhu et al. [19], we used the following steps to translate the English EMSR-Q into Chinese and carry out data analysis later:
Step One: Translation and Back-translation of the scale. Two experienced English translators were asked to independently translate the 20 items of EMSR-Q into Chinese. According to the differences between translators, two translators were invited again to evaluate and discuss the two translations. Later, another two experienced English translators who were blinded to the source questionnaire were invited to translate the questionnaire. These two translators were not professional psychology researchers, so we believe that they had a similar understanding of the questionnaire to most Chinese participants. The translated questionnaire was compared with the original questionnaire in order to prevent ambiguities due to different understandings in the translation process. If the item has the same expression or meaning with the original questionnaire, the translation of the item is acceptable. If the item is different from the expression or meaning of the original questionnaire, the item should be discussed again. The Chinese translation version of the questionnaire was finally determined.
Step Two: Cross-cultural Adjustment of the scale. the Chinese version of the questionnaire was formed by evaluating the cultural adaptability and modifying the content by psychology professors and graduate students and forming a preliminary Chinese version of EMSR-Q.
Step Three: Evaluation and Improvement of Items. 170 valid student samples in a city in Anhui province in China were collected to evaluate and improve the translation of the questionnaire. According to the questions asked on the questionnaire, such as " Which items don’t you understand? Please mark them out and explain why you don’t understand them.” Ask participants to assess whether the meaning of the items is clear. For the items that are not clear, items would be modified according to the suggestions of the subjects and the re-discussion of the experts by considering their cultural adaptability.
Step Four: Formal Investigation. 1395 valid student samples which were collected in three cities of Anhui province in China were used to analyze the structure of the questionnaire. A formal questionnaire was formed after screening out unsuitable items by item analysis, exploratory factor analysis and the analysis of reliability and validity.
Data analysis
In this study, SPSS 27.0 was used to complete item analysis, exploratory factor analysis and reliability and validity test based on Classical Test Theory (CTT). R software was used to complete the item analysis based on Multidimensional Item Response Theory (MIRT). Confirmatory factor analysis was performed with Mplus 8.0.
In this study, the correlation coefficient method and the extreme grouping method were used in the item analysis of CTT. The correlation coefficient method takes the Pearson correlation coefficients between each item and each dimension as the index. If the total correlation coefficient is greater than 0.4, the differentiation is acceptable. The extreme grouping method used the significant results of the independent sample t-test of the high and low groups of each dimension (highest 27% and lowest 27%) as the item discrimination index and p < 0.05 indicates a high degree of item differentiation [25, 26]. For exploratory factor analysis, principal components were used to extract the factors and promax oblique rotation was used to determine the factor load matrix. KMO value greater than 0.9 was used as the standard that results of Bartlett’s test of spherical are significant and the sample data is suitable for exploratory factor analysis. A common degree greater than 0.3 and a factor load greater than 0.4were used as the standard to retain the items. For CFI and TLI greater than 0.9, χ 2 / df less than 0.5, SRMR and RMSEA less than 0.1, the statistical model fits the data well [27]. Dimensions of mastery goal, performance approach goal, performance avoidance goal, external and internal motivation in learning and academic emotion regulation strategy were used as the criteria of EMSR-Q. Significant correlation means criterion validity is good. Internal consistency reliability and split-half reliability were adopted and the reliability coefficient greater than 0.6 indicates that the questionnaire has good internal consistency reliability. Item analysis of MIRT occured in the framework of IRT. The differentiation value below 0.64 means the discrimination is unacceptable. The differentiation value between 0.65 and 1.34 means the discrimination is moderate. The differentiation value between 1.35 and 1.69 means the discrimination is high. The differentiation value higher than 1.70 means the discrimination is very high [28]. the difficulty of all items, within [-3,3], ranges from small to large, indicating that the difficulty is acceptable [29].
Results
Sample characteristics
In this study, 1450 questionnaires were collected in three cities in the northern, central and southern parts of Anhui Province in China, among which 1395 were valid. The age of participants ranges from 12 to 18. 625 students were boys (44.8%). 608 students were girls (43.6%). 142 students (11.6%) did not report their gender. 675 junior school students (48.4%) and 720 senior school students (51.6%).
Item analysis
The Pearson correlation coefficients and the corrected item-total correlation coefficients between each item and the total score of each dimension were used as the item identification index. As can be seen from Table 2, the scores of each item are highly correlated with the total scores of each dimension (r > 0.62, p < 0.01). Except for the corrected item-total correlation coefficients (CITC) of item 7 and item 11 are slightly less than 0.4, the corrected item-total correlation coefficients (CITC) of the other items within the dimension are greater than 0.4. The Pearson correlation coefficients and the corrected item-total correlation coefficients between each item and the total score of each dimension indicate that each item has a good correlation with the dimension, which show that all the items can measure the content of the dimension.
Table 2.
Total correlation and decision values of each item of the emotion and motivation self-regulation scale (n = 780)
| Item | r | t | CITC | Item | r | t | CITC | Item | r | t | CITC |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | 0.81** | 36.14** | 0.64 | 6 | 0.77** | 27.28** | 0.57 | 11 | 0.66** | 21.90** | 0.39 |
| 2 | 0.74** | 33.01** | 0.47 | 7 | 0.62** | 16.89** | 0.37 | 12 | 0.76** | 27.74** | 0.55 |
| 3 | 0.79** | 35.51** | 0.60 | 8 | 0.79** | 31.92** | 0.60 | 13 | 0.75** | 25.74** | 0.55 |
| 4 | 0.80** | 33.08** | 0.62 | 9 | 0.72** | 21.97** | 0.50 | 14 | 0.77** | 26.54** | 0.58 |
| 5 | 0.75** | 27.18** | 0.52 | 10 | 0.74** | 23.03** | 0.52 | 15 | 0.79** | 26.95** | 0.62 |
Note: **p < 0.01
Subsequently, the total score of Sample 1 (n = 780) was ranked from low to high, with the top 27% of the score was identified as the low group and the bottom 27% of the score was identified as the high group after ranking the total score of Sample 1 (n = 780). Independent sample T-test was conducted for the high and low groups. As can be seen from Table 2, there are significant differences between the high and low groups of 20 items. The results of the independent sample t-test between the high and low 27% groups indicate that all the items are well differentiated and can effectively distinguish different populations.
Exploratory factor analysis
Exploratory factor analysis was carried out for Sample 1 (n = 780). The principal component method was used to extract the factors. The factor load matrix was determined by Promax oblique rotation. The results showed that the KMO value was 0.93 and the Bartlett’s test of sphericity was significant (χ2 = 6334.70, df = 190, p < 0.01), indicating that the sample was suitable for exploratory factor analysis. Based on the standard of the eigenvalue greater than 1, four factors with eigenvalue greater than 1 were extracted. The eigenvalues of the four factors were 7.310, 2.211, 1.236, and 1.033, respectively. The cumulative variance interpretation rate was 58.95%.
Through the analysis of the data results, we found that one of the extracted factors contains all the first-order factors of “positive regulation of motivation” and “process oriented self-regulation” in the original EMSR-Q, and both factors belong to the second order factor of “learning self-regulation style”. Therefore, It is reasonable to believe that the exploratory factor analysis directly extracted the second-order factor and there is an underestimate of the number of factors [30]. Previous studies indicate that the retention of factors in exploratory factor analysis should be first driven by theory. If the theoretical framework for the instrument is sound, we should start with the expectation that we should see that structure in the data [31]. The eigenvalues in the scree plot show inflection points at the 2,3,4 and 5 factors (see Fig. 2). Combining the theoretical structure of the original EMSR-Q and the results of exploratory factor analysis with the scree plot, a 5-factor model is selected to realize the consistency with the original EMSR-Q theoretical hypothesis as shown in Table 3.
Fig. 2.
Scree plot
Table 3.
Factor load and common degree of each item of emotion and motivation self-regulation scale
| Item | Factor load | Communality | ||||
|---|---|---|---|---|---|---|
| Factor 1 | Factor 2 | Factor 3 | Factor 4 | Factor 5 | ||
| 1 | 0.806 | 0.700 | ||||
| 6 | 0.787 | 0.653 | ||||
| 16 | 0.679 | 0.656 | ||||
| 7 | 0.836 | 0.620 | ||||
| 12 | 0.724 | 0.671 | ||||
| 17 | 0.651 | 0.645 | ||||
| 11 | 0.835 | 0.601 | ||||
| 3 | 0.650 | 0.571 | ||||
| 8 | 0.727 | 0.584 | ||||
| 13 | 0.538 | 0.498 | ||||
| 18 | 0.752 | 0.589 | ||||
| 4 | 0.647 | 0.611 | ||||
| 9 | 0.568 | 0.524 | ||||
| 14 | 0.766 | 0.584 | ||||
| 19 | 0.705 | 0.558 | ||||
| 5 | 0.467 | 0.493 | ||||
| 10 | 0.724 | 0.491 | ||||
| 15 | 0.716 | 0.604 | ||||
| 20 | 0.783 | 0.573 | ||||
Note: Factor 1 is avoidance oriented self-regulation, Factor 2 is performance oriented self-regulation, Factor 3 is negative self-regulation of stress, Factor 4 is positive regulation of motivation and Factor 5 is process oriented self-regulation
According to the theoretical hypothesis of the original questionnaire and the results of exploratory factor analysis, Factor 1 was named avoidance oriented self-regulation, Factor 2 was named performance oriented self-regulation, Factor 3 was named negative self-regulation of stress, Factor 4 was named positive regulation of motivation and Factor 5 was named process oriented self-regulation. The avoidance oriented self-regulation means that individuals focus on the task itself and its difficulties. Individuals do not want to do the task but want to finish it as soon as possible in order to avoid the task. Performance oriented self-regulation means that the task is an event that individuals dislike but seek the motivation to complete. Individuals focus on their performance in the task and the possible consequences of quitting or failing at the task. Negative self-regulation of stress means that the task makes individuals stressed and then they will have negative thoughts about their performance and results. Positive regulation of motivation means that the task is considered as a challenge. Individuals are motivated to complete the task from a positive perspective of efficiency and outcome. Process oriented self-regulation means that the task is seen as a challenge which is part of a larger task. Individuals are motivated to complete the task which is not just for succeeding in the task, but for improving their performance during the process [15].
Item 2 in Factor 2, “I’m dead tired… Well, I had to go on to pass” was loaded on Factor 1, but its content was inconsistent with the connotation of Factor 1. In order to ensure the content validity, this item was deleted. With the criteria of common degree greater than 0.3 and factor load greater than 0.4, 19 items were finally retained. The common degree of items ranged from 0.491 to 0.700, the load of other items in the rotating component matrix ranged from 0.467 to 0.836, and the cumulative variance interpretation rate was 59.07%, as shown in Table 3. It should be pointed out that Item 11, “What instructions so long! They only make me confused”, belonged to Factor 1 in the original scale. However, in the exploratory factor analysis of this study, the item was loaded in Factor 3, and the content of this item had a certain overlap with Factor 3. Therefore, the subsequent confirmatory factor analysis would further verify the loading of this article.
Confirmatory factor analysis
This study conducted confirmatory factor analysis on Sample 2 (n = 615). Confirmatory factor analysis tested the first-order five-factor model (M1), second-order two-factor model (M3) in the Chinese version of EMSR-Q and first-order five-factor model (M2) and second-order two-factor model (M4) of the English version of EMSR-Q in the exploratory factor analysis in this study. The results are shown in Table 4. According to previous studies, for non-nested models, when ΔBIC is 6–10, the model with smaller BIC is strongly supported, and when ΔBIC is greater than 10, the model with smaller BIC is more strongly supported [32]. In this study, ΔBIC between M1 and M2 was 1891.944 and ΔBIC between M3 and M4 was 1900.959, which was much higher than 10. With the model with smaller AIC had better fit, ΔAIC between M1 and M2 was 1878.679 and ΔAIC between M3 and M4 was 1887.694, suggesting that whether the first-order model or the second-order model was adopted, the model fit was relatively better when Item 2 was deleted and Item 11 was included in Factor 3. So Item 2 was deleted and Item 11 was included in Factor 3.
Table 4.
Comparison of model fitting index for confirmatory factor analysis of emotion and motivation self-regulation scale (n = 615)
| Model | χ 2 | df | CFI | TLI | SRMR | RMSEA(90%CI) | BIC | RMSEA | AIC |
|---|---|---|---|---|---|---|---|---|---|
| M1 | 427.097 | 142 | 0.94 | 0.93 | 0.044 | 0.057(0.051–0.063) | 33623.690 | 0.057 | 33327.441 |
| M2 | 596.787 | 160 | 0.92 | 0.90 | 0.052 | 0.067(0.061–0.072) | 35515.634 | 0.067 | 35206.120 |
| M3 | 441.672 | 145 | 0.94 | 0.93 | 0.046 | 0.058(0.052–0.064) | 33619.001 | 0.058 | 33336.017 |
| M4 | 620.378 | 163 | 0.92 | 0.90 | 0.053 | 0.068(0.062–0.073) | 35519.960 | 0.068 | 35223.711 |
Note: **p < 0.01. M1 is the first-order five-factor model of the Chinese version of EMSR-Q. M2 is the first-order five-factor model of the English version of EMSR-Q. M3 is the second-order two-factor model for the Chinese version of EMSR-Q. M4 is a second-order two-factor model of the English version of EMSR-Q
M1 and M3 were compared to verify the first-order five-factor structure and second-order two-factor structure of the emotion and motivation self-regulation questionnaire. The fitting results of the model are shown in Table 4, and the fitting index of the two models is relatively ideal. Considering that the indexes of the two models were close and the second-order model has 3 fewer parameters (Δdf = 3) than the first-order model, following the principle of frugality and the original scale’s theory and the structure of the measurement model, this study tended to support the second-order two-factor model. The final second-order two-factor model fitted well, χ2/df = 3.05, TLI = 0.93, CFI = 0.94, SRMR = 0.046, RMSEA = 0.058. The load of each item ranged from 0.504 to 0.815 in their respective dimensions (p < 0.001).
Criterion validaty
In this study, the criterion validity test was conducted for Sample 2 by using dimensions of performance-approach goal, performance-avoidance goal and mastery goal, as well as extrinsic motivation, academic emotion regulation strategy and intrinsic motivation in self-regulated learning strategy were used as the criterion of connotation related dimension of EMSR-Q. The correlation coefficients between the dimensions of EMSR-Q and the criteria are shown in Table 5. The results showed that there was a significant positive correlation between avoidance oriented self-regulation and performance avoidance goal (r = 0.209, p < 0.01) and a significant negative correlation between avoidance oriented self-regulation and intrinsic motivation (r=-0.339, p < 0.05) and academic emotion regulation strategy (r=-0.409, p < 0.05). There was no significant correlation between avoidance oriented self-regulation and extrinsic motivation (r = 0.068, p > 0.05). Performance oriented self-regulation was positively correlated with a performance-approach goal (r = 0.281, p < 0.01) and extrinsic motivation (r = 0.285, p < 0.01), and negatively correlated with intrinsic motivation (r=-0.146, p < 0.01). There was a significant negative correlation between negative self-regulation of stress and academic emotion regulation strategies (r=-0.414, p < 0.01). Positive regulation of motivation was positively correlated with performance-oriented goal (r = 0.188, p < 0.01), mastery goal (r = 0.678, p < 0.01), academic emotion regulation strategy (r = 0.510, p < 0.01), extrinsic motivation (r = 0.085, p < 0.05), intrinsic motivation (r = 0.486, p < 0.01), and negatively correlated with performance-avoidance goal (r=-0.197, p < 0.01); Process oriented self-regulation was positively correlated with mastery goal (r = 0.692, p < 0.01), academic emotion regulation strategies (r = 0.530, p < 0.01) and intrinsic motivation (r = 0.491, p < 0.01). The results showed that the Chinese version of EMSR-Q has good criterion validity.
Table 5.
Correlation between dimensions of emotion-motivation self-regulation scale and dimensions of criterion scale (n = 615)
| Performance-approach goal | Performance-avoidance goal | Mastery goal | Extrinsic motivation | Academic emotion regulation strategies | Intrinsic motivation | |
|---|---|---|---|---|---|---|
| Avoidance oriented self-regulation | - | 0.209** | - | 0.068 | -409** | -0.339** |
| Performance oriented self-regulation | 0.281** | - | - | 0.285** | -0.163** | -0.146** |
| Negative self-regulation of stress | - | - | - | - | -0.414** | - |
| Positive regulation of motivation | 0.188** | -0.197** | 0.678** | 0.085* | 0.510** | 0.486** |
| Process oriented self-regulation | - | - | 0.692** | - | 0.530** | 0.491** |
Note: **p < 0.01;*p < 0.05
Reliability
In this study, the internal consistency reliability and split-half reliability tests were carried out on the whole sample. The test results showed that the internal consistency reliability coefficients of each dimension ranged from 0.671 to 0.861 and the split-half reliability coefficients ranged from 0.726 to 0.809, indicating the good internal consistency reliability and split-half reliability of the Chinese version of EMSR-Q.
The item difficulty and differentiation parameters
After the items of each dimension were determined, item analysis based on Multidimensional Item Response Theory (MIRT) was conducted again on Sample 1 (n = 780). The results showed that the Chinese version of EMSR-Q fitted well(TLI = 0.967, CFI = 0.975, SRMR = 0.073, RMSEA = 0.057). We calculate item-fit statistics for each item within its respective dimension. The results showed the value of RMSEA less than 0.05 for all items, indicating a good fit for all items. We use Q3 statistics to check for local item dependence (LID). The results show Q3 residual correlations between all items less than 0.3, indicating good local independence for all items.
As shown in Table 6, differentiation α of Factor 1 ranged from 2.302 to 2.701, α of Factor 2 ranged from 1.177 to 2.648,α of Factor 3 ranged from 1.786 to 2.218,α of Factor 4 ranged from 1.919 to 2.117. α of Factor 5 ranged from 1.677 to 2.692. The difficulty of all items which are within [-3,3] ranges from small to large, which well reflects the orderly classification characteristics of the Likert 5 scale. These results show that the items of the Chinese version of EMSR-Q have good differentiation and difficulty.
Table 6.
Item analysis of emotion and motivation self-regulation scale (based on MIRT) (n = 780)
| Item | α | β values for each option | |||
|---|---|---|---|---|---|
| β1 | β2 | β3 | β4 | ||
| 1 | 2.701 | -0.633 | 0.219 | 0.709 | 1.619 |
| 6 | 2.269 | -0.165 | 0.56 | 1.16 | 1.805 |
| 16 | 2.302 | -0.859 | 0.049 | 0.569 | 1.465 |
| 7 | 1.177 | -3.037 | -1.569 | -0.805 | 0.905 |
| 12 | 2.648 | -1.614 | -0.856 | -0.253 | 0.893 |
| 17 | 2.071 | -1.544 | -0.694 | -0.179 | 1.258 |
| 11 | 1.786 | -1.217 | -0.258 | 0.381 | 1.635 |
| 3 | 2.163 | -1.025 | -0.221 | 0.366 | 1.529 |
| 8 | 2.213 | -0.911 | 0.049 | 0.649 | 1.721 |
| 13 | 2.140 | -0.026 | 0.73 | 1.302 | 1.871 |
| 18 | 2.218 | -0.859 | 0.131 | 0.829 | 1.962 |
| 4 | 2.117 | -1.694 | -0.572 | 0.097 | 1.464 |
| 9 | 1.925 | -2.344 | -1.331 | -0.629 | 0.586 |
| 14 | 2.097 | -2.127 | -1.015 | -0.234 | 1.026 |
| 19 | 1.919 | -1.565 | -0.578 | 0.009 | 1.131 |
| 5 | 1.899 | -1.942 | -1.098 | -0.498 | 0.732 |
| 10 | 1.677 | -2.333 | -1.382 | -0.815 | 0.684 |
| 15 | 2.692 | -2.096 | -1.284 | -0.637 | 0.596 |
| 20 | 2.019 | -2.095 | -1.300 | -0.669 | 0.596 |
Discussion
This study aims to revise the Chinese version of EMSR-Q in order to provide a specialized assessment tool for further studies of emotion and motivation self-regulation among Chinese middle school students. The results showed that EMSR-Q has good psychometric properties for the Chinese middle school students. Specifically, the Chinese version of EMSR-Q has a factor structure consistent with the English version of EMSR-Q. It has good criterion validity. The internal consistency reliability of each dimension of the Chinese version of EMSR-Q is similar to that of the English version of EMSR-Q [11], and each item of the Chinese version of EMSR-Q has good difficulty and differentiation.
In this study, the factor structure of the Chinese version of EMSR-Q was analyzed. Firstly, the revised Chinese version of EMSR-Q was found that it contains five dimensions consistent with the English version of EMSR-Q by exploratory factor analysis, including avoidance oriented self-regulation, performance oriented self-regulation, negative self-regulation of stress, positive regulation of motivation and process oriented self-regulation. The dimension division is consistent with the original theoretical model. Factor loadings ranged from 0.467 to 0.836. Commonality ranged from 0.491 to 0.700. The cumulative variance interpretation was 59.07%. The results showed that the factor structure of EMSR-Q has cross-cultural stability for Chinese middle school students, which is in line with the theoretical conception of Alonso-Tapia et al. [11] based on the dual processing theory [12] and the trichotomy goal orientation theory [13].
According to the results of exploratory factor analysis, the revised Chinese version of EMSR-Q retained 19 items. Item 2, “I’m dead tired… Well, I had to go on to pass”, belonged to the factor of performance oriented self-regulation in the English version of EMSR-Q. In this study, the load was on the factor of avoidance oriented self-regulation. This showed that Chinese middle school students might pay more attention to the boredom of the academic task contained in the context, rather than the pursuit of academic achievement when they tried to understand this item. Considering that its content is inconsistent with the connotation of avoidance oriented self-regulation, the item is deleted in order to ensure the validity of the content. Item 11, “What instructions so long! They only make me confused”, belonged to the factor of avoidance-oriented self-regulation in the original scale, but in this study, it was loaded on the factor of negative self-regulation of stress. This may be because of cultural differences. This may indicate that there is some ambiguity in the connotation of the item. For Chinese middle school students, when reading the item, it is easier to focus on the pressure caused by not understanding the teacher’s explanation rather than the boredom and avoidance of academic tasks. The cultural differences between Chinese and foreign students’ understanding and perception of this item can be explained from the perspective of the psychological distance between teacher and students and the goals of academic examination. Firstly, the principle of feudal moral conduct in the ancient feudal society, which is a hierarchy, reflects in the relationship between teachers and students as “students must obey their teachers”.The teacher is the noblest. There is a strong hierarchical concept of the relationship between teachers and students. The traditional culture make a deep historical imprint of hierarchical concept on the relationship between teachers and students in people’s hearts. Even some experienced teachers seem to be more concerned about their authoritative role in the classroom and are less willing to practice the student-centered teaching model emotionally [33]. When the teacher instructs too long to make students hard to understand, the authoritative role of the teachers make students dare not to seek help from teachers. However, driven by the goal of academic examination, students have to face the current academic task, which invisibly transformed into a source of stress. Students are more likely to feel pressure rather than want to escape. Therefore, the differences between Chinese and Western cultures should be taken into account in the process of revision, and the item should be placed in a more appropriate dimension according to the statistical results. The confirmatory factor analysis further compared the model of the English version of EMSR-Q and the Chinese version of EMSR-Q. The data fitting results also proved that the adjustment of items in exploratory factor analysis is better fitting than the attribution of items in the English version of EMSR-Q. The corresponding relationship between the adjusted items and the dimensions is more in line with the Chinese cultural background and contributes to a more accurate understanding of the emotion and motivation self-regulation of Chinese middle school students.
Based on the conceived five-factor model, Alonso-Tapia et al. extracted two higher-order factors, namely avoidance self-regulation style (ASRS) and learning self-regulation style (LSRS). The results show that the Chinese version of EMSR-Q can fit well with both the first-order five-factor models and the second-order two-factor models. The construct validity was good and was in line with the theoretical conception of the English version of EMSR-Q [11], which indicates that EMSR-Q has cross-cultural consistency in the overall structure. The confirmatory factor analysis results also proved that the second-order two-factor structure is better than the first-order five-factor structure. The factor structure model fits the original theoretical hypothesis, which is consistent with the previous studies [11, 15]. This indicates that the Chinese version of EMSR-Q can belong to five different first-order factors or measure two different self-regulation styles.
We examined the validity of EMSR-Q in terms of criterion validity. EMSR-Q theoretically correlates emotion and motivation self-regulation strategies, highlighting the central role the goal plays in emotion self-regulation. Therefore, the dimensions in EMSR-Q should be related to the dimensions of mastery goal, performance-approach goal, performance-avoidance goal, external motivation in learning, internal motivation and academic emotion regulation strategy [15]. The results showed that the dimensions of EMSR-Q are significantly correlated with their relevant criteria, indicating that the Chinese version of EMSR-Q has a good criterion validity, which conforms to the theoretical assumption of the English version of EMSR-Q.
In addition, the internal consistency coefficients α of the five first-order factors and the two second-order factors of the questionnaire are between 0.671 and 0.861. The split-half reliability coefficients are between 0.726 and 0.809. According to the evaluation criterion that the internal consistency coefficients and split-half coefficients are greater than 0.6, the internal consistency reliability and split-half reliability of the Chinese EMSR-Q are good. The reliability is generally consistent with the EMSR-Q in the Spanish middle school students (α is between 0.69 and 0.79) and the Columbia College students (α is between 0.69 and 0.84) [11, 15].
In this study, the items of the questionnaire were analyzed using the classical test theory and the multidimensional item response theory. In CTT, the correlations between the score of each item and each dimension of the Chinese version of EMSR-Q were between 0.62 and 0.81, both of which were greater than 0.40, indicating that all items have good contributions to the factors. There were significant differences in the high and low groups, indicating that the items of the Chinese version of EMSR-Q were well differentiated. In MIRT, the discrimination and difficulty of items were determined by analyzing the α and β parameters of each item. The results show that the discrimination parameter α of all the items is greater than 0.7, and the parameter α of most items is greater than 1.7, indicating that most items have a high degree of discrimination. The results fully distinguished the participants with high or low scores on latent features, which was basically consistent with the analysis results of CTT. The range of the difficulty parameter β of most items is basically within [-3,3], increasing in expected difficulty order. Overall, the items of the revised Chinese version of EMSR-Q have good differentiation and difficulty.
Implications
This study examined the psychometric characteristics of the Chinese version of EMSR-Q. We further explored the cross-cultural stability of the dimension structure in the group of Chinese middle school students and formed a suitable tool to measure emotion and motivation self-regulation for Chinese middle school students. The Chinese version of EMSR-Q lays a foundation for the subsequent study of emotion and motivation self-regulation among Chinese middle school students, which can be used to measure the structure and characteristics of emotion-motivation self-regulation in Chinese middle school students. It provides an important research tool to promote the relationship between emotion and motivation self-regulation and learning behavior, academic performance and intervention of emotion and motivation self-regulation for Chinese middle school students.
Limitations and future directions
There are limitations of this study that need to be pointed out. Firstly, the samples in this study were taken from Anhui Province, China. Anhui province is located in the central region of China, which is in the north and south intersection with the cultural characteristics of integrating the north and the south. However, the applicability of EMSR-Q in more other student groups needs further testing due to the geographical limitations of sampling. In future studies, applicability studies can be further conducted on students in other grades and regions. Secondly, only a cross-sectional study not any longitudinal studies is adopted in the research, making the questionnaire may have some limitations in retest reliability. Retest reliability of the Chinese version of EMSR-Q should be done in future studies. At the same time, we can further explore students’ emotion and motivation self-regulation and its relationship with learning. In addition, we can explore the influence of emotion and motivation self-regulation on their learning behavior and academic performance over time.
Conclusion
The Chinese version of the emotion and motivation self-regulation questionnaire has good reliability and validity, which can be used to measure emotion and motivation self-regulation in Chinese middle school students.
Abbreviations
- EMSRQ
Emotion and Motivation Self-Regulation Questionnaire
- ASRS
Avoidance-oriented self-regulation style
- LSRS
Learning oriented self-regulation style
- CTT
Classical Test Theory
- MIRT
Multidimensional Item Response Theory
- CFI
Comparative Fit Index
- TLI
Tucker-Lewis Index
- RMSEA
Root Mean Square Error of Approximation
- SRMR
Standardized Root Mean Square Residual
- BIC
Bayes Information Criterion
Author contributions
Huanran Wang wrote the main manuscript text. Dongdong Xue provided technical support for data analysis. Xiaozhuang Wang guided this study as a whole.
Funding
This study was supported in part by the grants from the Key Research Base of Humanities and Social Sciences of the Ministry of Education of China (Project Approval number: 22JJD190011).
Data availability
The data and materials that support the findings of this study are not openly available and are available from the corresponding author upon reasonable request.
Declarations
Ethical approval and consent to participate
This study was approved by the Institutional Review Board of Tianjin Normal University. (No. 2024101601) and the study was in compliance with the Declaration of Helsinki for recommendations guiding physicians in biomedical research involving human subjects. Informed consents were obtained from all individual participants and their parents/legal guardians prior to their participation in the study.
Consent for publication
No consent to publish was needed.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Efklides A. Interactions of Metacognition with motivation and affect in self-regulated learning: the MASRL Model. Educational Psychol. 2011;46(1):6–25. [Google Scholar]
- 2.Jacobs SE, Gross JJ. Emotion regulation in education: Conceptual foundations, current applications, and future directions. International handbook of emotions in education. 2014. pp. 183–201.
- 3.Wolters CA. Regulation of motivation: evaluating an underemphasized aspect of self-regulated learning. Educational Psychol. 2003;38(4):189–205. [Google Scholar]
- 4.Przybyl J, Chudak S. University students’ self-regulation in standard and enforced online language learning. Moderna Sprak. 2022;116(1):47–66. [Google Scholar]
- 5.Gehle M, Trautner M, Schwinger M. Motivational self-regulation in children with mild learning difficulties during middle childhood: do they use motivational regulation strategies effectively? J Appl Dev Psychol. 2023;84:101487. [Google Scholar]
- 6.Datu JAD, Yang WP, Lau KW. Does Mindfulness Matter for Cognitive Reappraisal and Academic Engagement? A Cross-lagged Panel Model Study in Filipino High School Students. J Early Adolesc. 2023;43(8):993–1015. [Google Scholar]
- 7.Namaziandost E, Rezai A, Heydarnejad T, Kruk M. Emotion and cognition are two wings of the same bird: insights into academic emotion regulation, critical thinking, self-efficacy beliefs, academic resilience, and academic engagement in Iranian EFL context. Think Skills Creativity. 2023;50:101409. [Google Scholar]
- 8.Hu YY, Wouters P, van der Schaaf M, Kester L. The effects of achievement goal instructions in game-based learning on students’ achievement goals, performance, and achievement emotions. Learn Instruction. 2024;93:101943. [Google Scholar]
- 9.Parker PC, Perry RP, Hamm JM, Chipperfield JG, Pekrun R, Dryden RP, Daniels LM, Tze VMC. A motivation perspective on achievement appraisals, emotions, and performance in an online learning environment. Int J Educational Res. 2021;108:101772. [Google Scholar]
- 10.Umarji O, McPartlan P, Moeller J, Li QJ, Shaffer J, Eccles J. The motivational system of task values and anticipated emotions in daily academic behavior. Motivation Emot. 2021;45(5):599–616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Alonso-Tapia J, Calderón EP, Ruiz MAD. Development and validity of the emotion and motivation self-regulation questionnaire (EMSR-Q). Span J Psychol. 2014;17:e55. [DOI] [PubMed] [Google Scholar]
- 12.Boekaerts M, Corno L. Self-regulation in the classroom: a perspective on assessment and intervention. Appl Psychology: Int Rev. 2005;54(2):199–231. [Google Scholar]
- 13.Elliot AJ. A conceptual history of the achievement goal construct. Handbook of competence and motivation. 2005;16(2005):52–72.
- 14.Boekaerts M. Emotions, emotion regulation, and self-regulation of learning: center for the study of learning and instruction, Leiden University, The Netherlands, and KU Leuven. Handbook of self-regulation of learning and performance. Routledge. 2011. pp. 422–439.
- 15.Alonso-Tapia J, Abello DM, Panadero E. Regulating emotions and learning motivation in higher education students. Int J Emotional Educ. 2020;12(2):73–89. [Google Scholar]
- 16.Le YY, Liu JS, Deng CP, Dai DY. Heart rate variability reflects the effects of emotional design principle on mental effort in multimedia learning. Comput Hum Behav. 2018;89:40–7. [Google Scholar]
- 17.Zhong JJ, Wen J, Li K. Do achievement goals differently Orient Students’ Academic Engagement through Learning Strategy and Academic Self-Efficacy and Vary by Grade. Psychol Res Behav Manage. 2023;16:4779–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Yan VX, Wang L. What predicts quality of learners’ study efforts? Implicit beliefs and interest are related to mastery goals but not to use of effective study strategies. Front Educ. 2021;06:643421. [Google Scholar]
- 19.Zhu TF, Chen CS, Chen SL. Validation of a Chinese version of the five facet mindfulness questionnaire and development of a short form based on item response theory. Curr Psychol. 2023;42(5):4212–24. [Google Scholar]
- 20.Yu EA, Chang EC. Construction of the relational meaning in life questionnaire: an exploratory and confirmatory factor-analytic study of relational meaning. Curr Psychol. 2021;40(4):1746–51. [Google Scholar]
- 21.Zhou Y, Tan DL. The Construction and Initial Application of Chinese College Students’ epistemological beliefs Questionnaire. Front Psychol. 2020;11:54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Elliot AJ, Church MA. A hierarchical model of approach and avoidance achievement motivation. J Personal Soc Psychol. 1997;72(1):218–32. [Google Scholar]
- 23.Jiang JC, Liu HS. Research on the relationship between the achievement goal orientation and the Learning Strategy & the academic performance. Psychol Dev Educ. 2005;(02):56–61.
- 24.Kong BJ, Lu HD. Questionnaire preparation of self-regulated learning strategies for middle school students. J Sichuan Normal Univ (Social Sci Edition). 2012;39(05):129–34. [Google Scholar]
- 25.Wang F, Wu YC, Sun XN, Wang D, Ming WK, Sun XY, Wu YB. Reliability and validity of the Chinese version of a short form of the family health scale. BMC Prim Care. 2022;23(1):108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Ge P, Wang XP, Gao SY, Liu JX, Niu YY, Yan MY, Fan SY, Li QY, Zhang JZ, Sun XA, et al. Reliability and validity of the Chinese version of the Sakata Eating Behavior Scale short form and preliminary analysis of the factors related to the score of the scale. Front Nutr. 2023;10:1076209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Stacchini L, Fonzo M, Catalini A, Di Martino G, Santangelo OE, Menegon T, Cedrone F, Gianfredi V. An Italian validation of the 5-Item attitudes to Mental Illness Questionnaire (AMIQ): a useful Tool for Rapid Assessment of Stigma, Acceptance, and Tolerance. Healthcare. 2024;12(3):395. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Svicher A, Palazzeschi L, Gori A, Di Fabio A. The Gratitude resentment and appreciation test-revised short (GRAT-RS): a multidimensional item response theory analysis in Italian workers. Int J Environ Res Public Health. 2022;19(24):16786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Liang MZ, Tang Y, Chen P, Liang J, Sun Z, Hu GY, Yu YL, Ye ZJ. New resilience instrument for family caregivers in cancer: a multidimensional item response theory analysis. Health Qual Life Outcomes. 2021;19(1):258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Liu QX, Sun B, Yuan CH, Zhou WQ, Zhou CY, Liu CL. Revision of the self-stigma of Depression Scale among Chinese College Students. Psychology: Techniques Appl. 2023;11(12):747–55. [Google Scholar]
- 31.Costello AB, Osborne J. Best practices in exploratory factor analysis: four recommendations for getting the most from your analysis. Practical Assess Res Evaluation. 2019;10(1):7. [Google Scholar]
- 32.Raftery AE. Bayesian model selection in social research. Sociol Methodol. 1995;25:111–63.
- 33.Poole A. Complex teaching realities’ and ‘deep rooted cultural traditions’: barriers to the implementation and internalisation of formative assessment in China. Cogent Educ. 2016;3(1):1156242. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data and materials that support the findings of this study are not openly available and are available from the corresponding author upon reasonable request.