Psychometric Properties of the Teachers’ Sense of Efficacy Scale for Chinese Special Education Teachers

Abstract

Teacher self-efficacy plays an important role in special education; yet, the literature on the measurement of teacher self-efficacy in this field is limited. This study investigated the psychometric properties of the Teachers’ Sense of Efficacy Scale with a sample of Chinese special education teachers (N = 401, 24.19% male). Confirmatory factor analysis indicated that the Chinese special educational version of the Teacher's Sense of Efficacy Scale (CS-TSES), modified from the Asian model by Ruan, J., Nie, Y., Hong, J., Monobe, G., Zheng, G., Kambara, H., & You, S. (2015). Cross-cultural validation of teachers’ sense of efficacy scale in three Asian countries. Journal of Psychoeducational Assessment, 33(8), 769-779, showed an excellent fit. Additionally, correlations were observed between the results of the CS-TSES and both the Multidimensional Scale of Perceived Social Support and the Satisfaction with Life Scale, providing criterion-related validity. In sum, the findings indicated that the TSES is applicable to special education teachers in the Chinese context.

Keywords

special education teachers confirmatory factor analysis teachers’ sense of efficacy scale teacher self-efficacy China

Introduction

Teacher self-efficacy is generally viewed as a type of self-efficacy (Bandura, 1997). It is defined as teachers’ confidence in their individual capability to influence student learning (Klassen, Tze, Betts, & Gordon, 2011). Recently, an increasing number of studies of teacher self-efficacy have emerged (Zee & Koomen, 2016), and more and more evidence has been found that teacher self-efficacy plays a critical role in explaining and resolving teaching and learning problems (Klassen et al., 2011; Kleinsasser, 2014; Tschannen-Moran, Hoy, & Hoy, 1998; Zee & Koomen, 2016). Teacher self-efficacy significantly predicts the psychological well-being of teachers (Burgueño, Sicilia, Medina-Casaubón, & Alcaraz-Ibañez, 2019; Zee & Koomen, 2016). Developed out of Bandura’s perceived self-efficacy theory, it is also associated with many social variables that generally contribute to positive mental health outcomes. For example, empirical evidence widely supports a medium-to-large correlation between social support and self-efficacy (Chang, Yuan, & Chen, 2018; Nwoke, Onuigbo, & Odo, 2017; Paukert et al., 2010).

Compared with teachers of typically developing children (or general teachers), special education teachers face an array of professional challenges, such as high stress levels (Wisniewski & Garigiulo, 1997), highly disruptive student behaviors (Hopman, Tick et al., 2018; Hopman, van Lier et al., 2018), the complex demands of difficult classes (Lavian, 2012), and high-risk work-related threats and violence (Pihl, Grytnes, & Andersen, 2018; Rasmussen, Hogh, & Andersen, 2013). Together with the physiological challenges associated with teaching students with special needs, these issues may have a great impact on teachers’ self-efficacy. Thus, it is vital to study special education teachers’ self-efficacy. Indeed, a number of studies of the self-efficacy of special education teachers have already been conducted (Coladarci & Breton, 1997; Hopman, Tick et al., 2018; Hopman, van Lier et al., 2018; Leyser, Zeiger, & Romi, 2011; Meijer & Foster, 1988; Podell & Soodak, 1993; Sarıçam & Sakız, 2014; Shillingford & Karlin, 2014; Viel-Ruma, Houchins, Jolivette, & Benson, 2010). However, few have discussed ways of measuring the self-efficacy of special education teachers.

Experts, scholars, and educational authorities are paying increasing attention to special education, particularly the development of special education teachers. However, few studies have focused on special education teachers in China, and the literature on the self-efficacy of Chinese special education school teachers is even more limited. International studies of the self-efficacy of Chinese special education teachers are also scarce. One exception was a study of Chinese special education teachers by Minghui, Lei, Xiaomeng and Potmesilc (2018), who found relationships between teacher efficacy, work engagement, and social support. They also reported that several sociodemographic factors significantly predicted teachers’ self-efficacy. To complement these findings, further research is needed on the self-efficacy of Chinese special education teachers, including assessments of the validity of relevant self-efficacy measurements.

Teachers’ Sense of Efficacy Scale

The Teacher’s Sense of Efficacy Scale (TSES, also named the Ohio State Teacher Efficacy Scale) is a widely used measure of teacher self-efficacy developed by Tschannen-Moran and Hoy (2001). In their original study, the authors reviewed the teacher self-efficacy measurements developed in the last century and noted the following problems with previous instruments: (a) an ambiguous conception of teacher self-efficacy, (b) inconsistencies in the teacher self-efficacy factor structure and measurement testing (e.g., studies of the famous Teacher Efficacy Scales yielded different results), and (c) a lack of balance between specificity and generality (Tschannen-Moran & Hoy, 2001). In response, the authors developed a new measurement of teacher self-efficacy—the TSES. The final results yielded four versions of the TSES with adequate psychometric properties: long-form, short-form, one-factor, and three-factor (i.e., Long-1, Long-3, Short-1, and Short-3). The authors identified three separate but related factors of TSES: student engagement, classroom management, and instructional strategies. These three dimensions of efficacy correspond to three important elements of a teacher’s working life. In suggesting that the three-factor pattern was appropriate for in-service teachers, the authors noted that the three-factor structure was less distinct for preservice teachers, for whom the best solution was a single-factor version.

Subsequent research reached a consensus that the three-factor pattern was appropriate for in-service teachers in various contexts (Fives & Buehl, 2009; Karami, Mozaffari, & Nourzadeh, 2019; Klassen et al., 2009; Ruan et al., 2015; Tsigilis, Koustelios, & Grammatikopoulos, 2010), but the findings regarding the factorial pattern for preservice teachers have been mixed. The recommendation of a single factor was supported by some researchers (Duffin, French, & Patrick, 2012; Fives & Buehl, 2009), but other studies suggested that the three-factor pattern was more appropriate for preservice teachers (Burgueño et al., 2019).

Staying within the three-factor pattern, Ruan et al. (2015) recently proposed an 11-item version of the three-factor model by removing one item (Item 16, “How well can you establish a classroom management system with each group of students?”) from the original three-factor model because that item seemed unsuitable for contexts in which group work is not a norm. Subsequently, the invariance of this model was confirmed in three Asian countries: Japan, Korea, and China. The model was recently also found to be applicable in a Spanish context (Burgueño et al., 2019).

Literature on TSES among Special Education Teachers

The above studies were conducted with samples of general teachers. Despite the TSES being widely used in the field of special education (Guo, Dynia, Pelatti, & Justice, 2014; Hopman, Tick et al., 2018; Hopman, van Lier et al., 2018; Sarıçam & Sakız, 2014; Shillingford & Karlin, 2014), few researchers have attempted to validate its use with special education teachers. To date, no peer-reviewed publication has focused on validating the TSES among Chinese special education teachers. This is problematic because special education teachers may have a unique understanding of the TSES items based on the particular needs of their students. It also remains unclear whether the three-factor structure widely observed among general teachers is also the most appropriate for special education teachers. Specific evidence of validity should be provided when test scores may differ in meaning between groups (American Educational Research Association, American Psychological Association, & National Council on Measurement in Education, 2014, p. 65). In fact, based on the results of reliability analysis, some studies have removed one item (i.e., Item 22, “assist families in helping their children do well in school”) of the student engagement dimension (the factor that mainly relates to students’ psychological status; e.g., “motivate students who show low interest in school work” and “get students to believe they can do well in school work”) from the short version of the TSES for special educators (Hopman, Tick et al., 2018; Hopman, van Lier et al., 2018). This item has generally passed structural tests with samples of general teachers (Burgueño et al., 2019; Duffin et al., 2012; Fives & Buehl, 2009; Karami et al., 2019; Klassen et al., 2009; Tsigilis et al., 2010). In general education, poor academic achievement is strongly associated with students’ psychological issues, such as low academic motivation (Bruinsma, 2004; Covington, 2000; Eccles & Wigfield, 2002), stress (Hirvonen, Yli-Kivistö, Putwain, Ahonen, & Kiuru, 2019), and anxiety (Hill et al., 2016). The main challenge within the broad task of helping families to enhance students’ achievement is mainly addressed to their children’s psychological problems (the student engagement dimension), which accounts for why Item 22 should be salient for the student engagement factor among general teachers. In special education, however, parents generally lack knowledge of their children’s special needs and how to meet them (Hunter, Blake, Simmons, Thompson, & Derouin, 2019). Special education teachers therefore often undertake the extra task of training parents, implying that there might be additional tasks beyond addressing students’ psychological problems involved in helping families to enhance students’ achievement. Consequently, for special education teachers, Item 22 might need to be loaded on instructional strategies alongside student engagement. Evidence is needed to clarify this issue.

This Study

The present study investigated the factor structure and psychometric properties of the TSES with a sample of Chinese special education teachers. It sought to extend the limited literature on teacher self-efficacy among special education teachers and to provide an appropriate measurement approach for use in the Chinese context. The findings further our understanding of the nature of teacher self-efficacy and offer suggestions for improving its measurement. More specifically, we investigated the scale’s internal structure and internal consistency and its criterion-related validity (see Standards 1.13, 1.16, American Educational Research Association et al. (2014)).

Based on the research discussed above, we proceeded on the following basis: (1) that the evidence supports a multidimensional structure instead of a single-factor structure for special education teachers (in-service), and that the three-factor pattern (i.e., student engagement, instructional strategies, and classroom management) that has been widely observed in general teachers (in-service) could be also found in special education teachers (in-service) and (2) that Item 22, which was found to be appropriate for loading on student engagement for general teachers, should be loaded on instructional strategies for special education teachers.

Methods

Participants and Procedure

The data were collected as follows. First, we made a list of special education schools in mainland China from which participants could be recruited. Next, the research assistants contacted the principals of the schools and informed them of the purpose of the study. With the principals’ permission, a comprehensive set of materials (comprising a detailed description of the study and the intended use of the results, an informed consent form, a sociodemographic information questionnaire, the TSES, the Multidimensional Scale of Perceived Social Support (MSPSS), and the Satisfaction with Life Scale (SWLS)) was given to each volunteer teacher. Finally, completed copies of the questionnaires were collected by the research assistants. The study was approved by the university’s ethics committee.

The initial sample comprised 432 full-time special education teachers. Of these, 401 fully responded to the TSES (92.82%), 23 partially responded (5.32%), and eight did not respond to any items (1.85%). Overall, nine participants responded to fewer than 50% of the TSES items and were therefore excluded (Hair, Black, Babin, & Anderson, 2010). The final sample comprised 423 special education teachers, of whom 108 self-identified as biologically male (25.53%) and 315 as female (74.47%). The teachers’ average teaching experience was 11.75 years (SD = 7.77), and 69.2% of the participants reported more than 5 years of teaching experience. According to the Ministry of Education of the People’s Republic of China, there are 58,656 full-time teachers of special education schools in China, and 43,351 (73.91%) are female. Therefore, the gender balance of our sample was representative of the population of interest.

Measurements

TSES

The TSES (Tschannen-Moran & Woolfolk-Hoy, 2001), originally named the Ohio State Teacher Self-Efficacy Scale, comprises 24 items designed to assess people’s perceptions of their own teaching effectiveness. Respondents give their answers on a 9-point scale ranging from 1 (nothing) to 9 (a great deal). Higher scores indicate that the respondents hold more positive beliefs about their teaching. As the existing Chinese version of the TSES in the literature was developed with a sample of general teachers in mainstream schools (Ruan et al., 2015), we decided to retranslate the measure for this study focused on special education teachers. The TSES was translated from its original English into Chinese through translation and back-translation. Two professional translators translated the TSES into Chinese, and the translated scale was then translated into English by another professional translator and a Chinese psychologist with two years’ experience of visiting an English-speaking country. This widely used procedure ensures that translated items retain their inherent meaning and intent (e.g., Gana, Daigre, & Ledrich (2013), Negy, Snyder, & Diáz-Loving (2004), Valdivia-Salas, Martín-Albo, Zaldivar, Lombas, & Jiménez (2017)). The procedure was repeated until a consensus was reached on the linguistic equivalence of the items, as assessed by the authors.

The detailed factor structure of the five previous models examined in this study is as follows. “Long-1”: one factor (Items 1–24). “Short-1”: one factor (Items 4, 6, 9, 11, 13, 15, 16, 18, 20, 22, and 23). “Long-3”: student engagement (Items 1, 2, 4, 6, 9, 12, 14, and 22), instructional strategies (Items 7, 10, 11, 17, 18, 20, 23, and 24), and classroom management (Items 3, 5, 8, 13, 15, 16, 19, and 21). “Short-3”: student engagement (Items 4, 6, 9, and 22), instructional strategies (Items 11, 18, 20, and 23), and classroom management (Items 3, 13, 15, and 16). Asian model (Ruan et al., 2015): student engagement (Items 4, 6, 9, and 22), instructional strategies (Items 11, 18, 20, and 23), and classroom management (Items 3, 13, and 15). In addition, given our hypothesis regarding Item 22, two types of revised three-factor pattern models (based on Long-3, Short-3, and the Asian) were proposed and examined: (a) Item 22 was revised to be loaded on student engagement and instructional strategies simultaneously (i.e., Long-3-R1, Short-3-R1, and the Asian-R1) and (b) Item 22 was revised to be loaded on instructional strategies but not on student engagement (i.e., Long-3-R2, Short-3-R2, and Asian-R2).

MSPSS

The MSPSS (Zimet, Dahlem, Zimet, & Farley, 1988) was also used in this study. This scale consists of 12 items (e.g., “I can talk about problems with my family”) rated on a 7-point scale (1 = very strongly disagree and 7 = very strongly agree). Higher scores indicate greater social support. Evidence from a previous study supported the use of the Chinese version of the MSPSS (Wang, Wan, Huang, Huang, & Kong, 2017). In the present study, the values of Cronbach’s alpha for the total scores and three subscales were .907, .827, .879, and .829.

SWLS

The SWLS (Diener, Emmons, Larsen, & Griffin, 1985) was adopted to measure life satisfaction. This scale consists of 5 items (e.g., “I am satisfied with my life”) rated on a 7-point scale (1 = totally disagree and 7 = totally agree). Higher scores indicate greater life satisfaction. Evidence from previous studies supports the use of the Chinese version of the SWLS (e.g., Bai, Wu, Zheng, & Ren (2011)). In the present study, the Cronbach’s alpha was .905.

Statistical Analysis

Missing data were primarily handled using a multiple imputation approach (with 10 categorically imputed datasets), following the recommendation of Newman (2014). The previous five TSES models (Long-1, Long-3, Short-1, Short-3, and the Asian model) and the revised three-factor pattern models (i.e., the Long-3-R1, Long-3-R2, Short-3-R1, Short-3-R2, Asian-R1, and Asian-R2) were tested using confirmatory factor analysis (CFA) with a maximum likelihood mean-adjusted estimator. The model appropriateness was evaluated using the root mean square error of approximation (RMSEA), the standardized root mean square residual (SRMR), the weighted root mean square residual (WRMR), the comparative fit index (CFI), and the Tucker–Lewis index (TLI). Values of CFI and TLI > .95, RMSEA < .06, and SRMR < .06 were taken to represent an adequate fit (Fried et al., 2016; Hu & Bentler, 1999; Orri et al., 2018; Pagliaccio et al., 2016; Strand, Gotch, French, & Beaver, 2019), while an adequate value for WRMR was <1.00 (Sass, 2011; Strand et al., 2019; Yu, 2002). Factor loading higher than .30 was considered salient.

After identifying the appropriate model, we selected values of Cronbach’s alpha, average variance extracted (AVE), and McDonald’s coefficient omega (McDonald, 1999) as indices for the internal consistency of each subscale. The generally used value of .80 for Cronbach’s alpha was treated as a cutoff for acceptable internal consistency (Cho & Kim, 2014; Wong et al., 2012); AVE values higher than .50 are appropriate (Burgueño et al., 2019; Hair et al., 2010); and omega values higher than .70 were considered to indicate good internal consistency (Ham et al., 2017; Huang, Reinke, Thompson, & Herman, 2018; Zucoloto, Maroco, & Campos, 2014).

Finally, we examined the criterion-related validity using two-tailed Pearson correlation analyses with the appropriate model, the MSPSS, and the SWLS. The strength of the correlations (r) was as follows: r < .10, low; .10 < r < .30, medium; .30 < r < .70, large; and r > .70, very high (Cohen, 1988). Given the observed associations of self-efficacy with life satisfaction and social support in the literature, the criterion-related validity would be supported if the correlations with the MSPSS and SWLS were positive and significant, whereas it would not be supported if the correlations were nonsignificant. All of the statistical analyses were performed using SPSS 24.0 and Mplus 7.0.

Results

Descriptive Statistics

Table 1 presents the mean and standard deviation for each item of the TSES, as well as the Z-scores for skewness and kurtosis. Both skewness and kurtosis were determined by indexes, with 0 (zero) as the central value for a perfectly normal distribution. The skewness and kurtosis values ranged nearly between −1 and +1, indicating that each item had a normal distribution (Curran, West, & Finch, 1996; Martínez Ortega, Gomà-i-Freixanet, & Valero, 2017).

Table 1.

Descriptive Statistics for the TSES.

Item	M	SD	Skewness	Kurtosis
TSES1	5.81	1.78	−.30	−.30
TSES2	5.82	1.58	−.40	−.01
TSES3	6.68	1.53	−.45	.13
TSES4	6.05	1.69	−.34	−.14
TSES5	7.30	1.53	−.96	1.08
TSES6	6.38	1.58	−.44	.20
TSES7	6.58	1.41	−.34	−.02
TSES8	6.66	1.41	−.35	−.04
TSES9	6.21	1.62	−.38	.02
TSES10	6.31	1.48	−.47	.35
TSES11	6.32	1.33	−.23	−.40
TSES12	5.78	1.64	−.35	.31
TSES13	7.00	1.46	−.47	−.09
TSES14	5.93	1.66	−.42	.04
TSES15	6.66	1.60	−.50	.02
TSES16	6.44	1.61	−.38	−.24
TSES17	6.33	1.60	−.54	.40
TSES18	5.68	1.65	−.53	.14
TSES19	6.46	1.56	−.34	−.21
TSES20	6.55	1.49	−.42	.14
TSES21	6.42	1.45	−.38	.06
TSES22	6.40	1.55	−.48	.12
TSES23	6.02	1.48	−.50	.83
TSES24	6.61	1.50	−.47	−.14

Note. SD = standard deviation; TSES = Teachers’ Sense of Efficacy Scale. Item numbers correspond to the original item numbers (Tschannen-Moran & Hoy, 2001).

CFA

As shown in Table 2, the three-factor models were found to be more appropriate than the single-factor models. The SRMRs for the existing models (i.e., Long-1, Long-3, Short-1, Short-3, and Asian) were adequate (ranging from .040 to .057 < .06). However, among the remaining indices for the existing models, only the CFI for the Asian model was acceptable (CFI = .952 > .95). Therefore, none of the previously proposed models were deemed satisfactory. In contrast, the revised models had better fit (i.e., Long-3-R1, Long-3-R2, Short-R1, Short-R2, Asian-R1, and Asian-R2), and the fit indices for the Asian-R1 and Asian-R2 models were all adequate (i.e., CFI and TLI > .95, RMSEA and SRMR < .06, and WRMR < 1.00). The fit indices were similar for the R1 and R2 models (i.e., between Long-3-R1 and -R2, between Short-3-R1 and -R2, and between Asian-R1 and -R2). Furthermore, in the Long-R1, Short-R1, and Asian-R1 models, the factor loadings of Item 22 on student engagement were clearly not salient (|.106|, |.152|, |.158| < .30), whereas the other factor loadings (including Item 22 on instructional strategies) were salient (> .30). These results did not support Item 22 being loaded on both instructional strategies and student engagement (as in the R1 models) but did support Item 22 being loaded on instructional strategies but not on student engagement (as in the R2 models). Asian-R2 was thus preferred among the models.

Table 2.

Model Fit for Existing Models.

	Number of dimensions	Item-number	S-B scaled χ²	DF	CFI	TLI	RMSEA	SRMR	WRMR
Long-1	1	24	885.819	252	.852	.838	.077	.057	1.402
Short-1	1	12	210.919	54	.909	.889	.083	.051	1.211
Long-3	3	24	733.948	249	.887	.874	.068	.053	1.310
Long-3-R1	3	24	686.383	248	.897	.886	.065	.050	1.234
Long-3-R2	3	24	687.618	249	.897	.886	.065	.050	1.234
Short-3	3	12	156.401	51	.939	.921	.070	.044	1.067
Short-3-R1	3	12	121.660	50	.958	.945	.058	.040	.975
Short-3-R2	3	12	123.223	51	.958	.946	.058	.040	.984
Asian	3	11	114.955	41	.952	.934	.065	.040	.946
Asian-R1	3	11	77.055	40	.975	.966	.047	.033	.799
Asian-R2	3	11	78.715	41	.975	.966	.047	.033	.817

Note. DF = degrees of freedom; CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual; WRMR = weighted root mean square residual. The Long-1, Long-3, Short-1, and Short-3 models were developed by Tschannen-Moran and Hoy (2001); the Asian model was proposed by Ruan et al. (2015).

R1. Item 22 was revised to be loaded on the student engagement and instructional strategies simultaneously (the Long-3-, Short-3-, and the Asian-R1).

R2. Item 22 was revised to be loaded on the instructional strategies but not on the student engagement (the Long-3-, Short-3-, and the Asian-R2).

The detailed structure of Asian-R2 is shown in Figure 1. All factor loadings were higher than .30. Given the relatively high correlation between instructional strategies and student engagement (.841), a two-factor model based on the items of Asian-R2 was examined (Factor 1: combining instructional strategies with student engagement and Factor 2: classroom management). The results showed that the fit for the two-factor model was inadequate (i.e., CFI = .948, TLI = .933, RMSEA = .066, SRMR = .040, and WRMR = .960) and poorer than that of the original three-factor model, thus supporting the three-factor pattern. Therefore, the Asian-R2 model was viewed as the Chinese special educational version of the TSES (CS-TSES) and used in the subsequent analysis.

Figure 1.

Structure of the Chinese special educational version of the Teachers’ Sense of Efficacy Scale.

Internal Consistency Reliability

The Cronbach’s alpha values for student engagement (.804), instructional strategies (.836), and classroom management (.785) were near the acceptable minimum; the AVE values for the three subscales reached appropriate (.571, .515, and .561), and the omegas measuring group variance were as follows: student engagement, .799; instructional strategies, .840; and classroom management, .792. Overall, these indices indicated a satisfactory level of internal consistency for each subscale.

Criterion-Related Validity

Pearson correlation analysis was conducted with the MSPSS and the SWLS (see Table 3) to seek the criterion-related validity of CS-TSES. The results showed that the three CS-TSES factors were significantly correlated with the MSPSS and the SWLS in a medium to fairly large range of rs (.209–.419, ps < .01), with all correlations positive (rs > 0). These theory-consistent relationships supported the criterion-related validity of the CS-TSES.

Table 3.

Correlations between the CS-TSES and MSPSS.

Correlation		Student engagement	Instructional strategies	Classroom management
CS-TSES	Student engagement	1.00
	Instructional strategies	.679	1
	Classroom management	.608	.609	1
SWLS		.327	.273	.206
MSPSS	Total MSPSS	.370	.419	.287
	Family support	.301	.384	.282
	Friend support	.327	.372	.251
	Significant other support	.325	.324	.210

Note. CS-TSES = Chinese special educational version of the Teacher's Sense of Efficacy Scale; MSPSS = multidimensional scale of perceived social support; SWLS = satisfaction with life scale. All correlations were significant at <.01

Discussion

The purpose of this study was to validate the TSES as a measure of the teacher self-efficacy of Chinese special education teachers. To this end, we examined whether several previous TSES models fitted our sample of Chinese special education teachers, and finally, we obtained the appropriate factor solution, the CS-TSES.

Factor Structure of the CS-TSES

We tested five previous models, with the results providing greater support for the three-factor pattern than the single-factor pattern. This finding was in line with previous findings among general teachers, which have shown that the one-factor model is suitable only for preservice teachers (Duffin et al., 2012; Tschannen-Moran & Hoy, 2001), whereas a three-factor model best fits in-service teachers (Fives & Buehl, 2009; Karami et al., 2019; Klassen et al., 2009; Tsigilis et al., 2010). It suggests that the factor structure of teacher efficacy is multidimensional for in-service teachers in special education, as has been shown to be the case for general in-service teachers. Among the existing models, the Asian model demonstrated the best fit. However, none of the existing models yielded an adequate fit, diverging from the results of most previous researches (Burgueño et al., 2019; Duffin et al., 2012; Klassen et al., 2009; Tsigilis et al., 2010). In particular, the results differed from those of a study with a sample of Chinese general teachers (Ruan et al., 2015). We hypothesized that the inconsistent finding was due to the meanings of Item 22 being different for special education teachers than for general teachers. Supporting this hypothesis, the revised Asian model (i.e., the CS-TSES) fit well with our sample when Item 22 was loaded on instructional strategies rather than student engagement. The results echoed the findings in Hopman, Tick et al. (2018) and Hopman, van Lier et al. (2018).

Despite the change affecting Item 22, the CS-TSES had a similar three-factor pattern to the models derived in previous studies (Fives & Buehl, 2009; Karami et al., 2019; Klassen et al., 2009; Tsigilis et al., 2010). The results supported the construct-level consistency of the factor structure of the TSES between special education teachers and general in-service teachers. The correlation between the three factors was found to be high in our CFA model, as found in other recent studies (Burgueño et al., 2019; Ruan et al., 2015). Given that the evidence from these and the present study supports the three-factor pattern, these high correlations might result from the CFA model overestimating the correlation within factors (Marsh, Morin, Parker, & Kaur, 2014).

Overall, the results provided evidence that the CS-TSES is appropriate for Chinese special education teachers. Consistent with the results of previous studies supporting the adequate fit of the TSES (Burgueño et al., 2019; Duffin et al., 2012; Klassen et al., 2009; Ruan et al., 2015), these findings once again support the adequacy of the psychometric properties of the TSES.

Internal Consistency of the CS-TSES

The values of Cronbach’s alpha and McDonald’s omega were satisfactory for all of the subscales, indicating that the CS-TSES can be used for comparison in different dimensions. Likewise, previous studies have widely reported that the TSES has high internal consistency (e.g., Burgueño et al. (2019) and Ruan et al. (2015)). These findings suggest that the TSES possesses a good and stable internal consistency across various groups of teachers in various contexts.

Criterion-Related Validity of the CS-TSES

The results for our sample of Chinese special education teachers indicated that the correlations between the CS-TSES and MSPSS were significant and in the low-to-medium range. These results are in line with the findings reported in the literature (Chang et al., 2018; Nwoke et al., 2017; Paukert et al., 2010). In particular, this study found that the correlation of instructional strategies with MSPSS was relatively higher than with the other two factors. This might be explained by students’ performance playing a more salient role in classroom management and student engagement than in instructional strategies and having little association with teachers’ perceived social support. Furthermore, the correlations between the CS-TSES subscales and the SWLS were all significant. This is consistent with previous studies suggesting that teachers’ self-efficacy is associated with their psychological well-being (Burgueño et al., 2019; Troesch & Bauer, 2017). The findings support the accuracy of the CS-TSES in measuring teacher self-efficacy with our sample. These findings were consistent with those of previous studies demonstrating the validity of the TSES by revealing its correlations with other scales (Burgueño et al., 2019; Tschannen-Moran & Hoy, 2001).

Contributions

As expected, this study provided evidence of the validity of the TSES with our sample. A similar three-factor pattern was observed among both general and special education teachers (in-service), supporting the assumption that teacher self-efficacy has general characteristics. Furthermore, the study enriches the scarce literature on the measurement of teacher self-efficacy of special education teachers and the literature on the validation of the TSES.

Tschannen-Moran et al. (1998) noted that “one of the most perplexing issues in the measurement of efficacy beliefs is determining the level of specificity” (p. 28). No consensus has yet been achieved on how to evaluate special education teachers’ self-efficacy. Some studies have applied specific instruments to explore teacher self-efficacy in special education contexts (Coladarci & Breton, 1997; Dawson & Scott, 2013; Meijer & Foster, 1988; Zhang, Wang, Stegall, Losinki, & Katsiyannis, 2017). Arguing for the use of specific instruments, Zhang et al. (2017) highlighted the technical problems that have occurred when general instruments have been used in special education contexts (p. 42). However, these issues may have arisen from researchers’ failure to validate general instruments for use with special education teachers before deploying them. Indeed, we found that several of the structures suited to general teachers did not show a good fit to our sample of special education teachers. Nevertheless, we obtained a factor solution (the CS-TSES) with an excellent fit by slightly revising a factor structure for general teachers (the Asian model). Our findings indicated that there might be not salient differences in teacher self-efficacy between general teachers and special education teachers, suggesting that it may be valid to apply general scales in special education contexts. These results are encouraging for researchers who wish to use general teacher self-efficacy instruments to assess teacher self-efficacy in special education (Guo et al., 2014; Hopman, Tick et al., 2018; Hopman, van Lier et al., 2018; Leyser et al., 2011; Minghui et al., 2018; Sarıçam & Sakız, 2014; Shillingford & Karlin, 2014; Viel-Ruma et al., 2010). Based on these findings, future studies should continue to explore the best measures of special education teacher efficacy with attention to both specific and general instruments.

The study also has some practical significance. Importantly, it extends the literature on special education teachers, a group at high risk of encountering negative psychological challenges (Brunsting, Sreckovic & Lane, 2014; Wisniewski & Gargiulo, 1997). In particular, special educators who work with students with severe conditions, such as certain cognitive disabilities, are very likely to perceive a long-term lack of obvious teaching effectiveness (Browder et al., 2008; Kurth, Born, & Love, 2016; Ruppar, Neeper, & Dalsen, 2016). Indeed, researchers have long recognized the risk of attrition faced by special education teachers (Billingsley, 2004; Boe & Cook, 2006; Hagaman & Casey, 2017; Stempien & Loeb, 2002). The present study may suggest practical strategies for addressing related difficulties. For instance, the findings suggest that the TSES can be effectively used to investigate the teacher self-efficacy of special education teachers in cross-group contexts. Comparing special education teachers with other groups of teachers may offer insights that help to improve special education teachers’ psychological well-being. Additionally, this study observed the relationship between social support, life satisfaction, and teacher self-efficacy. Similar to Minghui et al. (2018), we found that it is of great value to create a social environment with high recognition of special education.

Improving the self-efficacy of special education teachers may also benefit their instructive practices, as they are the main supporters of instruction-based education. This study found that the TSES has good psychometric properties with a Chinese sample. The CS-TSES may serve as a culturally sensitive measure of Chinese special education teachers’ self-efficacy that helps to identify and meet their needs and thereby solve practical problems.

Limitations and Future Directions

This study has filled a gap in the literature by examining the psychometric properties of self-efficacy measurement among special education teachers in a Chinese context. However, several limitations of the study should be noted and addressed in future research. First, the sample comprised only Chinese special education teachers; future studies should recruit samples of special education teachers from multiple countries. Second, although the study provided evidence of construct validity and internal consistency of the CS-TSES, it did not examine convergent validity. Third, the study did not provide evidence of stability (e.g., test–retest reliability), which could be a direction for future research. Fourth, because of the small number of males (108) in the sample, we did not test the gender invariance of CS-TSES. Future research should conduct invariance analysis with a larger sample size (and balanced numbers of men and women). Last, two of the three factors were measured using only three items. As three is the bare minimum number of items per factor, with four or more items preferred, future researchers could investigate a more appropriate version of TSES for administering to special education teachers.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Guangdong 13th five-year Education Science Project (No. 2018GXJK143).

ORCID iD

Fei-Fan Pang

References

American Educational Research Association, American Psychological Association, & National Council on Measurement in Education. (2014). Standards for educational and psychological testing. Washington, DC.

Bai

Zheng

Ren

(2011). The psychometric evaluation of the satisfaction with life scale using a nationally representative sample of China. Journal of Happiness Studies, 12(2), 183-197.

Bandura

(1997). Self-efficacy: The exercise of control. New York: W. H. Freeman and Company.

Billingsley

B. S.

(2004). Special education teacher retention and attrition. The Journal of Special Education, 38(1), 39-55.

Boe

E. E.

Cook

L. H.

(2006). The chronic and increasing shortage of fully certified teachers in special and general education. Exceptional Children, 72(4), 443-460.

Browder

Gibbs

Ahlgrim-Delzell

Courtade

G. R.

Mraz

Flowers

(2008). Literacy for students with severe developmental disabilities. Remedial and Special Education, 30(5), 269-282.

Bruinsma

(2004). Motivation, cognitive processing and achievement in higher education. Learning and Instruction, 14(6), 549-568.

Brunsting

N. C.

Sreckovic

M. A.

Lane

K. L.

(2014). Special education teacher burnout: A synthesis of research from 1979 to 2013. Education and Treatment of Children, 37(4), 681-711.

Burgueño

Sicilia

Medina-Casaubón

Alcaraz-Ibañez

Lirola

M.-J.

(2019). Psychometry of the teacher’s sense of efficacy scale in Spanish teachers’ education. The Journal of Experimental Education, 87(1), 89-100.

10.

Chang

C.-W.

Yuan

Chen

J.-K.

(2018). Social support and depression among Chinese adolescents: The mediating roles of self-esteem and self-efficacy. Children and Youth Services Review, 88, 128-134.

11.

Cho

Kim

(2014). Cronbach’s coefficient alpha. Organizational Research Methods, 18(2), 207-230.

12.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

13.

Coladarci

Breton

W. A.

(1997). Teacher efficacy, supervision, and the special education resource-room teacher. The Journal of Educational Research, 90(4), 230-239.

14.

Covington

M. V.

(2000). Goal theory, motivation, and school achievement: An integrative review. Annual Review of Psychology, 51(1), 171-200.

15.

Curran

P. J.

West

S. G.

Finch

J. F.

(1996). The robustness of test statistics to nonnormality and specification error in confirmatory factor analysis. Psychological Methods, 1(1), 16-29.

16.

Dawson

Scott

(2013). Teaching students with disabilities efficacy scale: Development and validation. Inclusion, 1(3), 181-196.

17.

Diener

Emmons

R. A.

Larsen

R. J.

Griffin

(1985). The satisfaction with life scale. Journal of Personality Assessment, 49(1), 71-75.

18.

Duffin

L. C.

French

B. F.

Patrick

(2012). The teachers’ sense of efficacy scale: Confirming the factor structure with beginning pre-service teachers. Teaching and Teacher Education, 28(6), 827-834.

19.

Eccles

J. S.

Wigfield

(2002). Motivational beliefs, values, and goals. Annual Review of Psychology, 53(1), 109-132.

20.

Fives

Buehl

M. M.

(2009). Examining the factor structure of the teachers’ sense of efficacy scale. The Journal of Experimental Education, 78(1), 118-134.

21.

Fried

E. I.

van Borkulo

C. D.

Epskamp

Schoevers

R. A.

Tuerlinckx

Borsboom

(2016). Measuring depression over time…or not? Lack of unidimensionality and longitudinal measurement invariance in four common rating scales of depression. Psychological Assessment, 28(11), 1354-1367.

22.

Gana

Daigre

Ledrich

(2013). Psychometric properties of the French version of the adult dispositional hope scale. Assessment, 20(1), 114-118.

23.

Guo

Dynia

J. M.

Pelatti

C. Y.

Justice

L. M.

(2014). Self-efficacy of early childhood special education teachers: Links to classroom quality and children’s learning for children with language impairment. Teaching and Teacher Education, 39, 12-21.

24.

Hagaman

J. L.

Casey

K. J.

(2017). Teacher attrition in special education: Perspectives from the field. Teacher Education and Special Education: The Journal of the Teacher Education Division of the Council for Exceptional Children, 41(4), 277-291.

25.

Hair

J. F. J.

Black

W. C.

Babin

B. J.

Anderson

R. E.

(2010). Multivariate data analysis (7th ed.). Englewood Cliffs, NJ: Pearson Prentice Hall.

26.

Ham

L. S.

Zamboanga

B. L.

Meca

Blumenthal

Hardy

S. A.

Hurd

L. E.

(2017). Factor structure of the brief comprehensive effects of alcohol scale in adolescents. Assessment, 26(4), 604-618.

27.

Hill

Mammarella

I. C.

Devine

Caviola

Passolunghi

M. C.

Szűcs

(2016). Maths anxiety in primary and secondary school students: Gender differences, developmental changes and anxiety specificity. Learning and Individual Differences, 48, 45-53.

28.

Hirvonen

Yli-Kivistö

Putwain

D. W.

Ahonen

Kiuru

(2019). School-related stress among sixth-grade students - Associations with academic buoyancy and temperament. Learning and Individual Differences, 70, 100-108.

29.

Hopman

J. A. B.

Tick

N. T.

van der Ende

Wubbels

Verhulst

F. C.

Maras

Breeman

L. D.

van Lier

P. A. C.

(2018). Special education teachers’ relationships with students and self-efficacy moderate associations between classroom-level disruptive behaviors and emotional exhaustion. Teaching and Teacher Education, 75, 21-30.

30.

Hopman

J. A. B.

van Lier

P. A. C.

van der Ende

Struiksma

Wubbels

Verhulst

F. C.

, … & Tick

N. T.

(2018). Impact of the good behavior game on special education teachers. Teachers and Teaching, 24(4), 350-368.

31.

L. T.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1-55.

32.

Huang

F. L.

Reinke

W. M.

Thompson

Herman

K. C.

(2018). An investigation of the psychometric properties of the early identification system-student report. Journal of Psychoeducational Assessment, 37(4), 473-485.

33.

Hunter

L. M. L.

Blake

Simmons

Thompson

Derouin

(2019). Implementing a parent education program in the special care nursery. Journal of Pediatric Health Care, 33(2), 131-137.

34.

Karami

Mozaffari

Nourzadeh

(2019). Examining the psychometric features of the teacher’s sense of efficacy scale in the English-as-a-foreign-language teaching context. Current Psychology. Advance online publication.

35.

Klassen

R. M.

Bong

Usher

E. L.

Chong

W. H.

Huan

V. S.

Wong

I. Y. F.

Georgiou

(2009). Exploring the validity of a teachers’ self-efficacy scale in five countries. Contemporary Educational Psychology, 34(1), 67-76.

36.

Klassen

R. M.

Tze

V. M. C.

Betts

S. M.

Gordon

K. A.

(2011). Teacher efficacy research 1998-2009: Signs of progress or unfulfilled promise? Educational Psychology Review, 23(1), 21-43.

37.

Kleinsasser

R. C.

(2014). Teacher efficacy in teaching and teacher education. Teaching and Teacher Education, 44, 168-179.

38.

Kurth

J. A.

Born

Love

(2016). Ecobehavioral characteristics of self-contained high school classrooms for students with severe cognitive disability. Research and Practice for Persons with Severe Disabilities, 41(4), 227-243.

39.

Lavian

R. H.

(2012). The impact of organizational climate on burnout among homeroom teachers and special education teachers (full classes/individual pupils) in mainstream schools. Teachers and Teaching, 18(2), 233-247.

40.

Lengua

L. J.

Stormshak

E. A.

(2000). Gender, gender roles, and personality: Gender differences in the prediction of coping and psychological symptoms. Sex Roles, 43(11), 787-820.

41.

León

F. R.

Morales

Vértiz

Burga-León

(2017). Universality of gender differences in 10 aspects of personality: A study of younger and older adult Peruvians. Personality and Individual Differences, 112, 124-127.

42.

Leyser

Zeiger

Romi

(2011). Changes in self-efficacy of prospective special and general education teachers: Implication for inclusive education. International Journal of Disability, Development and Education, 58(3), 241-255.

43.

McDonald

R. P.

(1999). Test theory: A unified treatment. Mahwah, NJ: Lawrence Erlbaum.

44.

McFarland

D. J.

(2019). Factor-analytic evidence for the complexity of the Delis–Kaplan executive function system (D-KEFS). Assessment. Advance online publication.

45.

Marsh

H. W.

Morin

A. J. S.

Parker

P. D.

Kaur

(2014). Exploratory structural equation modeling: An integration of the best features of exploratory and confirmatory factor analysis. Annual Review of Clinical Psychology, 10, 85-110.

46.

Martínez Ortega

Gomà-i-Freixanet

Valero

(2017). Psychometric properties and normative data of the Zuckerman–Kuhlman personality questionnaire in a psychiatric outpatient sample. Journal of Personality Assessment, 99(2), 219-224.

47.

Meijer

C. J. W.

Foster

S. F.

(1988). The effect of teacher self-efficacy on referral chance. The Journal of Special Education, 22(3), 378-385.

48.

Minghui

Lei

Xiaomeng

Potmesilc

(2018). Teacher efficacy, work engagement, and social support among Chinese special education school teachers. Front Psychol, 9, 648.

49.

Negy

Snyder

D. K.

Diáz-Loving

(2004). A cross-national comparison of Mexican and Mexican American couples using the marital satisfaction inventory-Revised (Spanish). Assessment, 11(1), 49-56.

50.

Newman

D. A.

(2014). Missing data. Organizational Research Methods, 17(4), 372-411.

51.

Nie

Lau

Liau

(2012). The teacher efficacy scale: A reliability and validity study. The Asia-Pacific Education Researcher, 21(2), 414-421.

52.

Nwoke

M. B.

Onuigbo

E. N.

Odo

V. O.

(2017). Social support, self-efficacy and gender as predictors of reported stress among inpatient caregivers. The Social Science Journal, 54(2), 115-119.

53.

Orri

Rouquette

Pingault

J.-B.

Barry

Herba

Côté

S. M.

Berthoz

(2018). Longitudinal and sex measurement invariance of the affective neuroscience personality scales. Assessment, 25(5), 653-666.

54.

Pagliaccio

Luking

K. R.

Anokhin

A. P.

Gotlib

I. H.

Hayden

E. P.

Olino

T. M.

, … & Barch

D. M.

(2016). Revising the BIS/BAS scale to study development: Measurement invariance and normative effects of age and sex from childhood through adulthood. Psychological Assessment, 28(4), 429-442.

55.

Paukert

A. L.

Pettit

J. W.

Kunik

M. E.

Wilson

Novy

D. M.

Rhoades

H. M.

, … & Stanley

M. A.

(2010). The roles of social support and self-efficacy in physical health’s impact on depressive and anxiety symptoms in older adults. Journal of Clinical Psychology in Medical Settings, 17(4), 387-400.

56.

Pihl

Grytnes

Andersen

L. P. S.

(2018). Violence prevention in special education schools - an integrated practice? Research in Developmental Disabilities, 77, 87-97.

57.

Podell

D. M.

Soodak

L. C.

(1993). Teacher efficacy and bias in special education referrals. The Journal of Educational Research, 86(4), 247-253.

58.

Rasmussen

C. A.

Hogh

Andersen

L. P.

(2013). Threats and physical violence in the workplace. Journal of Interpersonal Violence, 28(13), 2749-2769.

59.

Ruan

Nie

Hong

Monobe

Zheng

Kambara

You

(2015). Cross-cultural validation of teachers’ sense of efficacy scale in three Asian countries. Journal of Psychoeducational Assessment, 33(8), 769-779.

60.

Ruppar

A. L.

Neeper

L. S.

Dalsen

(2016). Special education teachers’ perceptions of preparedness to teach students with severe disabilities. Research and Practice for Persons with Severe Disabilities, 41(4), 273-286.

61.

Sarıçam

Sakız

(2014). Burnout and teacher self-efficacy among teachers working in special education institutions in Turkey. Educational Studies, 40(4), 423-437.

62.

Sass

D. A.

(2011). Testing measurement invariance and comparing latent factor means within a confirmatory factor analysis framework. Journal of Psychoeducational Assessment, 29(4), 347-363.

63.

Shillingford

Karlin

(2014). Preservice teachers’ self efficacy and knowledge of emotional and behavioural disorders. Emotional and Behavioural Difficulties, 19(2), 176-194.

64.

Stempien

L. R.

Loeb

R. C.

(2002). Differences in job satisfaction between general education and special education teachers. Remedial and Special Education, 23(5), 258-267.

65.

Strand

P. S.

Gotch

C. M.

French

B. F.

Beaver

J. L.

(2017). Factor structure and invariance of an adolescent risks and needs assessment. Assessment, ■■■, 950480091.

66.

Strand

P. S.

Gotch

C. M.

French

B. F.

Beaver

J. L.

(2019). Factor structure and invariance of an adolescent risks and needs assessment. Assessment, 26(6), 1105-1116.

67.

Taylor

S. E.

(2011). Social support: A review. In Friedman

H. S.

(Ed.), Oxford handbook of health psychology (pp. 189-214). New York, NY: Oxford University Press.

68.

Troesch

L. M.

Bauer

C. E.

(2017). Second career teachers: Job satisfaction, job stress, and the role of self-efficacy. Teaching and Teacher Education, 67, 389-398.

69.

Tschannen-Moran

Hoy

A. W.

(2001). Teacher efficacy: Capturing an elusive construct. Teaching and Teacher Education, 17(7), 783-805.

70.

Tschannen-Moran

Hoy

A. W.

Hoy

W. K.

(1998). Teacher efficacy: Its meaning and measure. Review of Educational Research, 68(2), 202-248.

71.

Tsigilis

Koustelios

Grammatikopoulos

(2010). Psychometric properties of the teachers’ sense of efficacy scale within the Greek educational context. Journal of Psychoeducational Assessment, 28(2), 153-162.

72.

Valdivia-Salas

Martín-Albo

Zaldivar

Lombas

A. S.

Jiménez

T. I.

(2017). Spanish validation of the avidance and fusion questionnaire for youth (AFQ-Y). Assessment, 24(7), 919-931.

73.

Viel-Ruma

Houchins

Jolivette

Benson

(2010). Efficacy beliefs of special educators: The relationships among collective efficacy, teacher self-efficacy, and job satisfaction. Teacher Education and Special Education: The Journal of the Teacher Education Division of the Council for Exceptional Children, 33(3), 225-233.

74.

Wang

Wan

Huang

Kong

(2017). Psychometric properties of multi-dimensional scale of perceived social support in Chinese parents of children with cerebral palsy. Frontiers in Psychology, 8, 2020.

75.

Wisniewski

Gargiulo

R. M.

(1997). Occupational stress and burnout among special educators: A review of the literature. The Journal of Special Education, 31(3), 325-346.

76.

Wong

Rodríguez

Quevedo

Fernández de Cossío

Borges

Reyes

, … & Alvarez

(2012). Questionnaire of executive function for dancers. Assessment, 19(3), 383-387.

77.

C. Y.

(2002). Evaluating cutoff criteria of model fit indices for latent variables models with binary and continuous outcomes (Unpublished doctoral dissertation). Los Angeles: University of California.

78.

Zee

Koomen

H. M. Y.

(2016). Teacher self-efficacy and its effects on classroom processes, student academic adjustment, and teacher well-being. Review of Educational Research, 86(4), 981-1015.

79.

Zhang

Wang

Stegall

Losinki

Katsiyannis

(2017). The construction and initial validation of the student teachers’ efficacy scale for teaching students with disabilities. Remedial and Special Education, 39(1), 39-52.

80.

Zimet

G. D.

Dahlem

N. W.

Zimet

S. G.

Farley

G. K.

(1988). The multidimensional scale of perceived social support. Journal of Personality Assessment, 52(1), 30-41.

81.

Zucoloto

M. L.

Maroco

Campos

J. A. D. B.

(2014). Psychometric properties of the oral health impact profile and new methodological approach. Journal of Dental Research, 93(7), 645-650.