Similar to other post-communist countries, the education system of the Czech Republic is moving away from the concept of segregation towards inclusive education (IE). In 2016 the Amendment to the Education Act came into force (Education Act No. 561/2004), which significantly transformed the organizational, financial and content aspects of IE provided in Czech primary schools. While the rights of pupils with special educational needs to IE in common primary schools were significantly strengthened, a number of substantial changes in education were implemented, with teachers taking on most of the burden.

The attitudes of teachers towards IE along with teacher beliefs regarding their own efficacy to implement inclusive practices in the education of diverse students greatly influence the success of IE (Jordan, 2018). In the international context, a number of studies have shown these two constructs as very closely related (Saloviita, 2020). The professional literature points to the importance of validity and reliability in identifying these constructs, for which validated tools have been designed and implemented (Li & Cheung, 2021). In the Czech research context, various questionnaires as well as self-constructed scales have been used without strict specifications with regard to their theoretical foundations and psychometric properties. It may be said that the results from these Czech instruments do not fully reflect the range of issues related to teacher attitudes and self-efficacy in relation to IE. In the international context, standardized validated instruments have been regularly used to measure teacher attitudes and self-efficacy in relation to IE: in the measurement of attitudes the instruments “Attitudes Towards Inclusion” (AIS) as well as “Intention to Teach in Inclusive Classroom” (ITICS) (Sharma & Jacobs, 2016) are typically used, with the tool “Teacher Efficacy for Inclusive Practices” (TEIP) (Sharma et al., 2012) commonly used to determine self-efficacy.

From an epistemological point of view, these three instruments reflect IE in a more holistic way, with an emphasis on socio-cultural perspectives regarding diversity as opposed to, for example, the approach of scales (e.g. Teachers’ Sense of Efficacy Scale) that reflect IE through an emphasis on a medical paradigm (Oliver, 1990). The items of the three instruments under study in the present analysis are formulated in the form of statements. From a psychometric point of view, the original English version of the 8-item AIS questionnaire aims to identify attitudes reflecting beliefs (BI) and feelings (FI) regarding IE. In contrast, the 7-item ITICS questionnaire is aimed at identifying the teacher’s intention to implement inclusive measures, i.e. the tool determines attitudes towards curriculum change strengthening IE (Cch) and the intention to cooperate with experts (Cons). The 18-item TEIP questionnaire focuses on the assessment of the teacher’s own efficacy in relation to the implementation of procedures associated with IE, specifically in the area of: cooperation with experts (EC), management of disruptive student behavior (EMB), and implementation of IE (EII).

The three instruments, especially the TEIP, have been translated into several languages. The psychometric properties of the questionnaires have also been analyzed in a number of validation studies across individual countries and regions. Most of these studies have provided empirical evidence of the validity and reliability of questionnaires, demonstrating that the instruments work well in different socio-cultural contexts. The psychometric properties of the Czech version of the AIS, ITICS and TEIP, however, remain unknown.

In order to dependably and precisely apply in the Czech environment the above-mentioned scales in the identification of the attitudes and self-efficacy of primary school teachers in relation to IE, the aim of this study is to present information on (1) construct validity (factor structure), (2) reliability, and (3) measurement invariance of the Czech versions of the instruments.

In total, 1434 teachers from 140 Czech common primary schools participated in this quantitative research study (88% women; about ¼ of teachers aged 41–50). A two-step quota sampling was chosen which may be considered a quasi-representative sample, i.e. the statistical compliance not ensured between the base and the sample selected. Out of 14 regions, a respective number of primary schools was proportionally selected from each region in the Czech Republic. The research sample included only common primary schools (not including special schools) with different socio-demographic characteristics. Each primary school had a set number of teachers selected (always 5 teachers from a lower primary and 5 teachers from an upper primary school). The translation of the questionnaires from English to Czech and their use in this research was subject to the consent of the main author of the questionnaires (U. Sharma). The authors of the original version of the AIS declared a two-factor structure of the questionnaire. The BI factor consists of four items, as does the FI factor; a high reliability α>0.8 was identified for both scales. According to the authors, ITICS is also made up of two factors (Chc with four items, Cons with three); both scales reached α<0.8. In contrast, the original version of the TEIP consists of 3 factors, each with 6 items (EII, EMB and EC reached values of α>0.8). The authors of the original versions of the questionnaires do not provide information regarding the measurement invariance of the instruments. The tested versions of the instruments in our own research were identical to the original versions, featuring the same number of items, including their relations to individual factors, the response format, and an identical calculation of the raw score, all of which were maintained. The factor structure of the Czech versions of the questionnaires was verified using confirmatory factor analysis (CFA) through the maximum likelihood method as well as reliability through McDonald's omega (ω). Measurement invariance testing was performed using a multi-group confirmatory factor analysis (MGCFA) method to compare the model among the sample of teachers in terms of their age. The data analysis was conducted using SPSS Amos (ver. 28). The questionnaires were administered to the primary school participants in person by trained interviewers. This quantitative research study was conducted in 2022. All ethical aspects of the research were in compliance with the standards of the American Psychological Association.

The tested factor structure of the Czech AIS corresponds to the original two-factor solution. Values of indices/criteria for the accepted model: CFI=0.987; TLI=0.978; RMSEA=0.059; SRMR=0.021. The standardized factor loadings of the items for the respective factors were acceptably high. The reliability of the partial factors was satisfactory, and the questionnaire as a whole also showed acceptable values: ωAIS=0.90 (ωBI=0.89; ωFI=0.81). For the ITICS, the two-factor model fit the data worse (CFI=0.961; TLI=0.932; RMSEA=0.066; SRMR=0.032; AIC=119; BIC=204) than the subsequently tested model with one general factor “ITICS” (CFI=0.962; TLI=0.938; RMSEA=0.063; SRMR=0.032; AIC=117; BIC=197). For the model with a two-factor solution, a relatively low reliability of the scales was also found (ωChc=0.63 and ωCons=0.61). Therefore, a more efficient one-factor model with one general factor consisting of 7 items was adopted to express the overall intentions of teachers with regard to teaching in an inclusive class (ωITICS=0.75). The factor structure of the Czech TEIP questionnaire corresponds to the original version: values of indices/criteria for the final model were CFI=0.954; TLI=0.945; RMSEA=0.054; SRMR=0.038; BIC=0.965). Reliability of partial factors: ωEII=0.80; ωEC=0.84; ωEMB=0.87. Due to the high correlation of the factors (r>0.70), the second-order model was tested (CFI=0.954; TLI=0.946; RMSEA=0.053; SRMR=0.037; BIC=0.958), which was better to the first-order model. The analyzes thus captured the hierarchical factor structure of the Czech TEIP, i.e. three specific factors representing one general higher-order factor (ωTEIP=0.92). The results of MGCFA invariance tests for the final accepted models from CFA were satisfactory. For the AIS and TEIP, full strict measurement invariance was achieved for the compared groups of teachers in terms of age. A partially scalar invariant model was confirmed for ITICS. The analyzes show that the Czech versions of AIS, ITICS and TEIP are valid and reliable questionnaires that can be recommended for further use in research and diagnostic practice.

Czech Republic. Act No. 561/2004 Collection of Law, on Pre-school, Basic, Secondary, Tertiary Professional and Other Education (the Education Act). Available from: http://www.msmt.cz/vzdelavani/skolstvi-v-cr/act-no-561-2004-coll-of-24-september-2004-on-pre-school Jordan, A. (2018). The Supporting Effective Teaching Project: 1. Factors influencing student success in inclusive elementary classrooms. Exceptionality Education International, 28(3), 10–27. Li, K., & Cheung, R. (2021). Pre-service teachers’ self-efficacy in implementing inclusive education in Hong Kong: The roles of attitudes, sentiments, and concerns. International Journal of Disability, Development and Education, 68(2), 259–269. Oliver, M. (1990). The Politics of Disablement: A Sociological Approach. Basingstoke, Hampshire: Macmillan. Saloviita, T. (2020). Attitudes of Teachers Towards Inclusive Education in Finland. Scandinavian Journal of Educational Research, 64(2), 270–282. Sharma, U., & Jacobs, K. (2016). Predicting in-service educators’ intentions to teach in inclusive classrooms in India and Australia. Teaching and Teacher Education, 55(3), 13–23. Sharma, U., Loreman, T., & Forlin, Ch. (2012). Measuring Teacher Efficacy to Implement Inclusive Practices. Journal of Research in Special Educational Needs, 12(1), 12–21.