Researchers Validate First Attitude Scale on Mind and Intelligence Games in Primary Schools

Researchers in Türkiye have developed and validated a psychometric instrument that measures primary school class teachers’ attitudes toward the educational use of mind and intelligence games, according to a study published in Discover Education on 26 October 2026. The survey-based study, carried out with 723 class teachers working in various provinces of Türkiye during the 2024–2025 academic year, produced a 20-item scale with three subdimensions: Use Value, Motivation for Use, and Use Resistance. The authors report that the final instrument demonstrated strong validity and reliability evidence, filling a gap they identified in the measurement literature.

Mind and intelligence games, which include games such as chess and tangram, have attracted growing attention in educational settings. Previous research cited in the study links these games to gains in problem solving, decision making, critical thinking, reasoning, communication and research skills, visual perception, hand–eye coordination, and three-dimensional and analytical thinking. Studies referenced by the authors also report that tangram develops geometric thinking, mental rotation and spatial perception, and that using these games in mathematics, science and social studies lessons is associated with higher motivation, better retention, fewer discipline issues and improved academic achievement. However, the authors note that no validated scale existed that specifically measured primary school teachers’ attitudes toward using such games in their teaching.

The authors ground the study in attitude theory. According to Eagly and Chaiken’s three-component model, an attitude is a psychological disposition combining cognitive beliefs, affective responses and behavioral intentions, while the theory of planned behavior emphasizes perceived behavioral control in translating attitudes into action. Teachers may value games and feel motivated to use them, yet encounter obstacles such as lack of materials and physical space, time constraints, curriculum incompatibility and classroom dynamics that produce pedagogical resistance. Accurately diagnosing this multidimensional situation, the authors argue, requires a theory-based psychometric instrument rather than a measure of general approval.

To build the instrument, the research team conducted a literature review and interviews with 15 class teachers, generating a pool of 42 items formatted as a 5-point Likert scale ranging from Strongly Disagree to Strongly Agree. Expert opinion was then sought from a professor specializing in scientific research methods and data analysis, a professor specializing in psychometrics and attitude scale development, an associate professor specializing in mind and intelligence games, an associate professor specializing in classroom education, and a Turkish language expert who examined grammar, spelling and clarity. Twenty class teachers additionally reviewed the items for linguistic clarity. After revisions, the draft was finalized at 40 items. Ethical approval was obtained from the Scientific Research Ethics Committee of the Faculty of Social and Humanities Sciences at Inönü University, and data were collected online via Google Forms.

The study employed a survey design with convenience sampling. The 723 participating class teachers were divided into two independent groups: 424 teachers provided data for exploratory factor analysis (EFA), and a different 299 teachers provided data for confirmatory factor analysis (CFA). Sampling adequacy for factor analysis was high, with a Kaiser–Meyer–Olkin value of 0.962 and a significant Bartlett’s test of sphericity, indicating the data were suitable for the planned analyses. No univariate outliers were found, as all standardized Z scores fell within the range of −3 to +3.

The EFA, conducted using principal axis factoring with direct Oblimin rotation, identified three factors with eigenvalues of one or higher, supported by parallel analysis and a scree plot, which together explained 71.493% of the total variance. The first factor accounted for 48.50% of the variance, the second 14.61% and the third 8.37%. Item loadings after rotation exceeded the 0.40 threshold, ranging from 0.546 to 0.933 in the first subdimension, 0.783 to 0.902 in the second, and 0.547 to 0.924 in the third. The item reduction process was systematic: eight items were removed for low or nonsignificant loadings, three for cross-loading, four for low communalities below 0.30, and five for conceptual overlap, leaving 20 items with at least three items per factor. Based on item content, the researchers named the factors Use Value, Motivation for Use, and Use Resistance.

The three subscales capture distinct facets of teachers’ attitudes. The Use Value subdimension contains 11 items reflecting the pedagogical value that mind and intelligence games provide for developing pupils’ multifaceted skills, including problem solving, decision making, attention, strategic thinking, creativity, vocabulary and socialization. The Motivation for Use subdimension includes 5 items describing teachers’ engagement when teaching subjects such as Turkish, mathematics, life studies, social studies and science through these games. The Use Resistance subdimension comprises 4 items addressing perceived barriers, such as the beliefs that dedicated spaces like workshops or spare classrooms are unnecessary, that institutional support is not required, and that tournaments should not be organized.

The CFA results supported the three-factor model, with fit indices reported as χ²/df = 2.349, NFI = 0.93, RFI = 0.92, CFI = 0.96, GFI = 0.89, AGFI = 0.85, RMSEA = 0.067 and SRMR = 0.043, with NNFI = 0.95, IFI = 0.96 and RMR = 0.020 indicating excellent fit. A unidimensional competing model fell well below acceptable thresholds, and a chi-square difference test showed the three-factor model fit significantly better (Δχ²(3) = 744.430, p < .001), empirically confirming the scale’s multidimensional structure. Three error covariances between item pairs within the same dimensions were specified on theoretical grounds. Convergent validity was supported by average variance extracted (AVE) values above 0.50 and composite reliability (CR) values above 0.70 for all three factors, with CR exceeding AVE in each case. Discriminant validity was supported because the maximum shared variance (MSV) and average shared variance (ASV) values were lower than the AVE values for every subfactor, and standardized regression coefficients were 0.53 or higher.

Reliability evidence was also strong. The authors report an overall Cronbach’s alpha of 0.91 for the EFA sample and 0.93 for the CFA sample, with the Use Value subdimension showing high internal consistency at α = 0.96. The authors caution that Cronbach’s alpha is structurally influenced by the number of items, but argue that the high coefficient does not indicate item redundancy, noting that the items represent different areas of cognitive achievement rather than synonym repetitions, that item–total correlations were balanced, and that no single item caused an artificial deviation in the alpha-if-item-deleted analysis. All item–total correlations exceeded 0.30, indicating high item discrimination. Independent-samples t-tests comparing the mean item scores of the upper and lower 27% groups were statistically significant for all items (p < .001), with Cohen’s d effect sizes ranging from 1.20 to 6.86, all exceeding the large-effect threshold of 0.80.

In discussing the findings, the authors connect the three-factor structure to the tripartite model of attitudes, in which cognitive, affective and behavioral components are distinguished, and note consistency with an earlier three-dimensional teacher scale by Kurupınar and Aydoğan. They highlight the Use Resistance dimension as a distinctive contribution. Whereas the negative attitude dimension in that earlier scale covered general affective reactions such as becoming bored quickly or lacking interest, the Use Resistance dimension identified in this study reflects operational, spatial and structural barriers within the primary school classroom ecosystem, such as curriculum intensity, time pressure, inadequate physical infrastructure and institutional workload. The authors also outline limitations: data were collected online on a voluntary basis through convenience sampling, limiting generalizability; the EFA and CFA samples differed markedly in professional seniority, although the three-factor structure held in the independent CFA group; the scale was developed in Turkish and tailored to Turkish educational culture; and alternative reliability analyses such as McDonald’s omega and test–retest reliability were not conducted. They recommend applying the scale to other educational levels and subject areas, investigating the items qualitatively, collecting data through probability-based sampling across socioeconomic strata, and adapting the instrument to other languages and cultures to test cross-cultural generalizability.

Subject of Research: Development and validation of an attitude scale measuring primary school teachers' attitudes toward the use of mind and intelligence games

Article Title: Development of an attitude scale regarding the use of mind and intelligence games in primary schools

Article References: Ceylan, A., Kasa Ayten, B., & Kıraç, Ş. (2026). Development of an attitude scale regarding the use of mind and intelligence games in primary schools. Discover Education, 5(1), Article 1226. https://doi.org/10.1007/s44217-026-02347-3

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02347-3

Keywords: attitude scale, mind and intelligence games, primary school teachers, scale development, exploratory factor analysis, confirmatory factor analysis, psychometrics, use resistance, teacher attitudes, educational measurement, Türkiye, Discover Education