Goodness-of-fit Test for a Logistic Regression Model Fitted using Survey Sample Data

Abstract

After a logistic regression model has been fitted, a global test of goodness of fit of the resulting model should be performed. A test that is commonly used to assess model fit is the Hosmer–Lemeshow test, which is available in Stata and most other statistical software programs. However, it is often of interest to fit a logistic regression model to sample survey data, such as data from the National Health Interview Survey or the National Health and Nutrition Examination Survey. Unfortunately, for such situations no goodness-of-fit testing procedures have been developed or implemented in available software. To address this problem, a Stata ado-command, svylogitgof, for estimating the F-adjusted mean residual test after svy: logit or svy: logistic estimation has been developed, and this paper describes its implementation.

Keywords

st0099 svylogitgof goodness of fit survey design svy logistic regression logit

References

Archer

K. J.

2001. Goodness-of-fit tests for logisitic regression models developed using data collected from a complex sampling design. Ph.D. thesis, Ohio State University.

Graubard

B. I.

, Korn

E. L.

, and Midthune

1997. Testing goodness-of-fit for logistic regression with survey data. In Proceedings of the Section on Survey Research Methods, Joint Statistical Meetings, 170–174. Alexandria, VA: American Statistical Association.

Hosmer

D. W.

Jr. , and Lemeshow

1980. Goodness-of-fit tests for the multiple logistic regression model. Communications in Statistics: Theory and Methods, Part A 9: 1043–1069.

Hosmer

D. W.

Jr. 2000. Applied Logistic Regression. 2nd ed. New York: Wiley.

National Center for Health Statistics. 1999. National Health Interview Survey: Research for the 1995–2004 redesign. Hyattsville, MD: Vital and Health Statistics.

Skinner

C. J.

, Holt

, and Smith

T. M. F.

1989. Analysis of Complex Surveys. New York: Wiley.

Thomas

D. R.

, and Rao

J. N. K.

1987. Small-sample comparisons of level and power for simple goodness-of-fit statistics under cluster sampling. Journal of the American Statistical Association 82: 630–636.