DEVELOPING RELIABLE AND VALID PSYCHOLOGICAL TESTS: A COMPREHENSIVE FRAMEWORK FOR MODERN ASSESSMENT PRACTICE

Abstract

The present study came up and confirmed a dependable and valid psychological test using a basic quantitative and descriptive research design. It entailed the synthesis of items, testing on the specialists and statistical validation of items to guarantee excellent methodological rigor and usefulness. The sample size was 150 participants whose professional and educational backgrounds were varied to allow generalisation of the results. The instrument was designed on five-point Likert scale, and the data were analysed to obtain descriptive statistics, reliability and validity. The computation of reliability was done with the use of Cronbach's alpha and split-half method, whereas validity was examined as content and criterion validity. The internal consistency was strong based on the Cronbach's alpha (0.89) and a constant split-half reliability coefficient of r = 0.85. The CVI (0.91) indicated that there was an excellent expert consensus, and criterion validity was also determined to have a significant positive correlation with a scale that was already standardized (r = 0.76, p < 0.01). The outcomes of these results support the claim that the test is a psychometrically sound and useful tool. The research results in the conclusion that underdeveloped psychological test may be effectively developed using well-organized, transparent, and easy to access quantitative techniques.