| Sumario: | This study investigates the reliability of teacher burnout measurements and proposes a method to correct reliability estimates for sample variability and attenuation using the statistical software R. Traditional internal consistency reliability coefficients, such as Cronbach's Alpha and McDonald's Omega, may be biased due to sampling error and attenuation. This paper introduces an R function to adjust these reliability estimates, using a teacher burnout scale from the Trends in International Mathematics and Science Study (TIMSS) in 2019 data using as a case study data from Saudi Arabia. The results demonstrate that the corrective procedures significantly improve reliability estimates, suggesting that uncorrected measures may underestimate the true reliability of teacher burnout assessments. These findings underscore the importance of considering measurement error in the evaluation of psychological constructs.
|
|---|