Evaluation of Psychometric Properties of Tests Used in Educational Research

Saved in:
Bibliographic Details
Title: Evaluation of Psychometric Properties of Tests Used in Educational Research
Language: English
Authors: Satyendra Nath Chakrabartty (ORCID 0000-0002-7687-5044)
Source: International Journal of Psychology and Educational Studies. 2026 13(1):28-39.
Availability: International Journal of Psychology and Educational Studies. Sakarya University, Faculty of Education, Department of Educational Sciences, Sakarya, Turkey. e-mail: ijpesjournal@gmail.com; Web site: https://ijpes.com/index.php/ijpes/index
Peer Reviewed: Y
Page Count: 12
Publication Date: 2026
Document Type: Journal Articles
Reports - Evaluative
Descriptors: Evaluation Methods, Psychometrics, Multiple Choice Tests, Likert Scales, Test Validity, Test Reliability, Educational Research, Scoring Formulas, Error of Measurement, Test Theory
ISSN: 2148-9378
Abstract: It is imperative to assess psychometric parameters of tests and items by methodologically sound approaches and finding relationships among the psychometric qualities. The paper describes methods of computations of psychometric parameters of MCQ test and items under classical test theory, finds relationships among theoretically defined test reliability with factorial validity (FV) and separately with discriminating and difficulty values of test and items considering the entire data. Finding r_(tt(theoretical)) from a single administration offers significant benefits like computation of test error variance; testingH_0: r_(tt(theoretical))=1; finding battery reliability, estimation of true scores and its confidence interval. Relationships of r_(tt(theoretical)) derived with FV of standardized scores ([left white lenticular bracket]FV[right white lenticular bracket]_(Z-scores)), test discriminating value ([left white lenticular bracket]Disc[right white lenticular bracket]_T), Cronbach alpha in terms of item discriminating value [left white lenticular bracket](Disc[right white lenticular bracket]_i) and [left white lenticular bracket]Disc[right white lenticular bracket]_T, Item reliability as function of [left white lenticular bracket]Diff[right white lenticular bracket]_i and [left white lenticular bracket]Disc[right white lenticular bracket]_T, alpha using PCA results ([alpha]_PCA) with [left white lenticular bracket]FV[right white lenticular bracket]_(Z-scores). The point of inflection of the negatively slopped reliability-discriminating value curve gives [left white lenticular bracket]Disc[right white lenticular bracket]_T for which test reliability is optimal; Common point of [left white lenticular bracket]Disc[right white lenticular bracket]_i and [left white lenticular bracket]Diff[right white lenticular bracket]_i curves could be used in item deletions to improve test reliability.
Abstractor: As Provided
Entry Date: 2026
Accession Number: EJ1496157
Database: ERIC
Description
Abstract:It is imperative to assess psychometric parameters of tests and items by methodologically sound approaches and finding relationships among the psychometric qualities. The paper describes methods of computations of psychometric parameters of MCQ test and items under classical test theory, finds relationships among theoretically defined test reliability with factorial validity (FV) and separately with discriminating and difficulty values of test and items considering the entire data. Finding r_(tt(theoretical)) from a single administration offers significant benefits like computation of test error variance; testingH_0: r_(tt(theoretical))=1; finding battery reliability, estimation of true scores and its confidence interval. Relationships of r_(tt(theoretical)) derived with FV of standardized scores ([left white lenticular bracket]FV[right white lenticular bracket]_(Z-scores)), test discriminating value ([left white lenticular bracket]Disc[right white lenticular bracket]_T), Cronbach alpha in terms of item discriminating value [left white lenticular bracket](Disc[right white lenticular bracket]_i) and [left white lenticular bracket]Disc[right white lenticular bracket]_T, Item reliability as function of [left white lenticular bracket]Diff[right white lenticular bracket]_i and [left white lenticular bracket]Disc[right white lenticular bracket]_T, alpha using PCA results ([alpha]_PCA) with [left white lenticular bracket]FV[right white lenticular bracket]_(Z-scores). The point of inflection of the negatively slopped reliability-discriminating value curve gives [left white lenticular bracket]Disc[right white lenticular bracket]_T for which test reliability is optimal; Common point of [left white lenticular bracket]Disc[right white lenticular bracket]_i and [left white lenticular bracket]Diff[right white lenticular bracket]_i curves could be used in item deletions to improve test reliability.
ISSN:2148-9378