Criteria for Good Measurement
Lecture 18 details the core criteria for assessing the goodness of measures in scientific research: validity, reliability, and practicality. Validity ensures that an instrument measures what it purports to measure. The lecture explores content validity, criterion-related validity (concurrent and predictive validity), and construct validity (convergent and discriminant validity). Reliability evaluates the consistency and stability of a measurement tool across time and items. The lecture discusses test-retest reliability, parallel-form reliability, and internal consistency measures including split-half reliability and Cronbach's alpha coefficient. Practicality assesses operational feasibility, cost efficiency, ease of administration, and interpretability. Students learn how measurement error—both systematic bias and random error—affects empirical findings in university exams. Non-probability sampling strategies, such as convenience sampling, purposive judgment sampling, quota sampling, and snowball sampling, are critically evaluated. Experimental research designs are analyzed with focus on internal validity control, external validity generalization, and manipulation of independent variables. True experimental designs, quasi-experimental designs, and pre-experimental arrangements are contrasted regarding treatment randomization and confounding control. Students explore non-reactive research methods, examining unobtrusive measures, physical traces, content analysis, and secondary archival data utilization. Field research and observational studies are investigated to capture naturalistic human behavior, participant observation dynamics, and qualitative field notes. Historical comparative research methodologies illustrate how archival primary sources, oral histories, and comparative historical analyses generate deep contextual understanding. Focus group discussions and qualitative interviewing techniques are evaluated for exploratory depth, group synergy, and moderator facilitation dynamics. Data transformation workflows, including categorical coding, data editing, missing value imputation, and scale index construction, are systematically outlined. Quantitative data presentation guidelines detail the construction of one-way frequency tables, bivariate contingency tables, and graphical displays. Statistical inference concepts illustrate how hypothesis testing, significance levels, p-values, and statistical power validate empirical research findings. Academic report writing standards cover the structural organization of research proposals, executive summaries, methodology sections, and empirical discussions. Scholarly citation standards, including American Psychological Association (APA), Chicago, and Harvard referencing formats, are demonstrated for academic integrity.
No questions available yet.
There are currently no examination questions published for Criteria for Good Measurement. Explore other curriculum units or build a custom quiz with the practice builder.