ERIC Number: ED167566
Record Type: RIE
Publication Date: 1975
Reference Count: 0
A Survey Measurement Error Model for Repeated Binary Responses.
Folsom, Ralph E., Jr.
In large-scale surveys, it is no longer uncommon for repeated measurements to be obtained from respondents and analyses performed to gauge the magnitiude of nonsampling errors. This is particularly true for periodic surveys and longitudinal surveys where a very large investment in data collection is made. This technical note, aimed at the analysis of repeated survey measurements, emphasizes the misclassification error aspects of the Census Bureau model for zero-one variables. Section 2 of this note presents a specification of the Census Bureau model for zero-one variables which postulates a "true value" for each potential respondent and emphasizes the misclassification aspects of the error distribution. Section 3 explores the bias and variance implications of the measurement error model for sample proportions. Section 4 presents the expectations of several measures of inconsistency introduced for repeated measurement surveys by Hansen, Hurquitz, and Pritzker. Section 5 summarizes the results, concentrating on their practical implications regarding the analysis of repeated survey measurements. (Author/CTM)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Repeated Measures Design