Logistic Regression: Description, Examples, and Comparisons

Citation

Morgan, S. Philip & Teachman, Jay D. (1988). Logistic Regression: Description, Examples, and Comparisons. Journal of Marriage and the Family, 50(4), 929-936.

Abstract

Family studies have seen a dramatic increase in the use of statistical tools for the analysis of nominal-level variables. The general class of such models is referred to as log-linear models. When one nominal-level variable is considered to be dependent on a set of predictor variables (which may be nominal, ordinal, or interval), log-linear models are also known as logit models or logistic-regression models (Knoke and Burke, 1980).' In recent years, logistic regression has been used to study topics as diverse as marital formation and dissolution (Abdelrahman and Morgan, 1987; Heaton, Albrecht, and Martin, 1985; Rank, 1987; Speare and Goldscheider, 1987), contraceptive use (Bean, Williams, Opitz, Burr, and Trent, 1987; Studer and Thornton, 1987), poverty (Smith and Zick, 1986), premarital sexual experience (Newcomer and Udry, 1987), premarital pregnancy (Robbins, Kaplan, and Martin, 1985; Yamaguchi and Kandel, 1987) and spouse abuse (Kalmuss and Seltzer, 1986). Despite its growing popularity, there still exists confusion about the nature and proper use of logistic regression in family studies. We present a nontechnical discussion of logistic regression, with illustrations and comparisons to better-known procedures such as percentaging tables and OLS regression

URL

http://dx.doi.org/10.2307/352104

Reference Type

Journal Article

Year Published

1988

Journal Title

Journal of Marriage and the Family

Author(s)

Morgan, S. Philip
Teachman, Jay D.