Measuring teachers' perceptions about differentiated instruction: The DI-Quest instrument and model

Within a democratic and multicultural society, diversity is a reality, and differences between students are a fact which teachers have to deal with on a daily basis. Differentiated instruction aims to meet these differences in learning in order to provide all students with the best possible learning opportunities. However, to date no validated instruments exist to measure teachers' perceptions of differentiated instruction and their related classroom practices. This study, therefore, examined the factor structure and reliability of the Differentiated Instruction Questionnaire, called the DI-Quest instrument. A list of 87 items was constructed, building on existing prevalent theoretical models of Differentiated Instruction (e.g. Tomlinson, 2014; Hall, 2002). An exploratory and confirmatory factor analysis was undertaken to investigate the factor structure of the questionnaire. As a result, five factors emerged: two factors related to the teachers' philosophy of differentiated instruction (the teachers' mindset and their ethical compass), two factors referred to the practical principles that teachers apply to differentiate (flexible grouping and output = input) and the last factor (differentiated instruction) covered the self-reported extent to which teachers differentiated their instruction related to three types of differences in learning (students' interests, readiness and learning profile). As a result, the DI-Quest instrument entailed 31 items with a five-factor structure indicating a good fit (CFI = 0.919; TLI = 0.911; RMSEA = 0.041 [0.037-0.044 - 90% confidence interval, p(0.05) = 1.000]; SRMR = 0.048; χ2 = 5888.338, df = 465, p = 0.000). In addition, assuming theoretical relatedness between the factors, the validation of a DI-Quest model was empirically validated. We compared the model fit for two models by investigating which model had a lower BIC and AIC value and by comparing their chi square values. The best-fitting DI-Quest model showed four factors (teachers' mindset, ethical compass, flexible grouping and output = input as dependent variables) functioning as significant predictors of the fifth factor (the self-reported adoption of Differentiated Instruction, which served as an independent variable). Moreover, this paper also discusses the psychometric properties of the DI-Quest instrument and the implications of the model for schools, educators and researchers.