Definitions
|
|
Reliability
All definitions
|
Reliability refers to "the consistency of a measure either across different testings (external) or within itself (internal)" Coolican, 2004, p. 187).
"External reliability" means that the test or instrument" consistently produces the same reading on the same amount" and that it is stable across time (Coolican, 2004, p. 188). Spector (1992) refers to this same idea as "test/retest reliability," meaning that "a scale yields consistent measurement over time" (Spector, 1992, p. 6). "Internal reliability" or "internal consistency" means that the test or instrument is consistent within itself. In other words, the individual items that represent a specific construct on a test will all be answered similarly by the same person if the test has high internal reliability. Spector describes "internal-consistency reliability" as "a measurable property of items that implies that they measure the same construct" (Spector, 1992, p. 30). If a scale or survey has high internal consistency, then "multiple items designed to measure the same construct," will intercorrelate with one another" (Spector, 1992, p. 6). Cronbach's alpha, also known as "the coefficient alpha is a measure of internal consistency" (Cronbach, 1951). Related terms References Coolican, H. (2004). Questionnaires: Using psychological scales and tests in research. Chapter 7 in Research Methods and Statistics in Psychology. London: Hodder & Stoughton. Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297-334. Spector, P. (1992). Summated rating scale construction: An introduction. Newbury Park, London, New Dehli: Sage. Read a practice exam essay in response to the question:
"It is often said that traditional research methods in the field of work and well-being have limited our understanding. Evaluate this position with reference to theory and research." Exam essay practice answer |
|
|
|