Survey Measurement Design and Evaluation Using Reliability Theory.

by M. Susan Marquis, Kent H. Marquis

Purchase Print Copy

 FormatList Price Price
Add to Cart Paperback72 pages $25.00 $20.00 20% Web Discount

Considers the general classes of measurement error effects recognized by contemporary reliability theory, the consequences of these effects for some descriptive and inferential statistics, and the traditional methodologies used to assess the errors in survey data. Traditional reliability estimation strategies rest on stringent assumptions. This report treats extensions of basic reliability designs to test or correct for violations of these assumptions, including the three-trial model to estimate the effects of true change; variation in the length of time to reinterview to assess test-retest conditioning; multiple trait and multiple item, multiple trial strategies to assess method or condition effects; criterion reliability to estimate net bias; and the degree of covariance between error and the true attribute. 72 pp. Bibliog.

This report is part of the RAND Corporation report series. The report was a product of the RAND Corporation from 1948 to 1993 that represented the principal publication documenting and transmitting RAND's major research findings and final research.

The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.