Premium
A Procedure for Editing Survey Data
Author(s) -
Hocking R. R.,
Huddleston H. F.,
Hunt H. H.
Publication year - 1974
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2346993
Subject(s) - computer science , survey data collection , information retrieval , statistics , mathematics
Summary The questionnaires for a survey can be divided into T groups based on the patterns of incompleteness which occurred for the p content items. For each group, the sample mean vector and sample covariance matrix are computed for those items which are recorded. An indicator matrix is used to describe the pattern of incompleteness in each group. A computer program, ESTMAT, is then used to combine this information to yield maximum‐likelihood estimates of the mean vector and covariance matrix under the assumption of multivariate normality. As an option, individual questionnaires may then be edited for any incompleteness using multiple‐regression techniques for the missing items. Two examples from agricultural surveys are presented to illustrate the technique and are examined in some detail with respect to the gain in information achieved by using ESTMAT as opposed to other, commonly used procedures.