Interpreting the results

Mitchell H. Katz

doi:10.1017/CBO9780511974175.009

8 - Interpreting the results

Published online by Cambridge University Press: 01 April 2011

Mitchell H. Katz

Show author details

Mitchell H. Katz: Affiliation:
University of California, San Francisco

Book contents

Get access

Summary

What information will my multivariable analysis produce?

Multivariable techniques produce two major kinds of information: Information about how well the model (all the independent variables together) fit the data and information about the relationship of each of the independent variables to the outcome variable (with adjustment for all other independent variables in the analysis). In this chapter, we will review information that is routinely output from multivariable software programs. In the next chapter we will delve deeper into how well the assumptions of the models are fulfilled and how to improve the fit of the models by looking at supplementary techniques that you may request.

How do I assess how well my model fits the data?

Although there is some overlap, the methods for determining how well a model accounts for the outcome differ by type of multivariable analysis (Table 8.1). The methods for each model are discussed below.

A Multiple linear regression

We start the assessment of a multiple linear regression model by testing whether the independent variables predict the outcome better than assuming that everyone in the study had the mean value for the outcome. If knowing the values of the independent variables improves the fit more than would be expected by chance, then the value of F will be large. A large F value for a given sample size and a given number of variables in the model (which determines the degrees of freedom) will result in a small P value.

Information

Type: Chapter
Information: Multivariable Analysis
A Practical Guide for Clinicians and Public Health Researchers
, pp. 140 - 161

DOI: https://doi.org/10.1017/CBO9780511974175.009 [Opens in a new window]

Publisher: Cambridge University Press

Print publication year: 2011

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Book purchase

Temporarily unavailable

References

Glantz, S.A. and Slinker, B.K.Primer of Applied Regression & Analysis of Variance. 2nd edn. New York, NY: McGraw-Hill, 2001, pp. 294–302Google Scholar

Hosmer, D.W. and Lemeshow, S.Applied Logistic Regression. New York, NY: Wiley, 1989, pp. 8–18Google Scholar

Menard, S.Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage Publications, 1995, pp. 19–21Google Scholar

Harrell, F.E.Regression Modeling Strategies: With applications to linear models, logistic regression, and survival analysis. New York: Springer, 2001, p.231CrossRef Google Scholar

Gordon, H.S., Harper, D.L., and Rosenthal, G.E. “Racial variation in predicted and observed in-hospital death.” JAMA 276 (1996): 1639–44CrossRef Google Scholar PubMed

Goldman, L., Cook, E.F., Brand, D.A., et al. “A computer protocol to predict myocardial infarction in emergency department patients with chest pain.” N. Engl. J. Med. 318 (1988): 797–803CrossRef Google Scholar PubMed

Harrell, F.E., Lee, K.L., Matchar, D.B., et al. “Regression models for prognostic prediction: Advantages, problems, and suggested solutions.” Cancer Treat. Rep. 69 (1985): 1071–7Google Scholar PubMed

Hosmer, D.W. and Lemeshow, S.Applied Logistic Regression. 2nd edn. New York: John Wiley, pp. 305–8CrossRef

Hosmer, D.W. and Lemeshow, S.Applied Logistic Regression. 2nd edn. New York: John Wiley, pp. 280–7CrossRef

Colford, J.M., Tager, I.B., Hirozawa, A.M., et al. “Cryptosporidiosis among patients infected with human immunodeficiency virus: Factors related to symptomatic infection and survival.” Am. J. Epidemiol. 144 (1996): 807–16CrossRef Google Scholar

Hauck, W.W. and Donner, A. “Wald's test as applied to hypotheses in logit analysis.” J. Am. Stat. Assoc. 72 (1977): 851–3Google Scholar

,SOS-Kanto Study Group. “Cardiopulmonary resuscitation by bystanders with chest compression only (SOS-Kanto): An observational study.” Lancet 369 (2007): 920–6CrossRef Google Scholar

Rothman, K.J. “No adjustments are needed for multiple comparisons.” Epidemiology 1 (1990): 43–6CrossRef Google Scholar PubMed

Savitz, D.A. and Olshan, A.F. “Multiple comparisons and related issues in the interpretation of epidemiologic data.” Am. J. Epidemiol. 142 (1995): 904–8CrossRef Google Scholar PubMed

Thompson, J. “Re: ‘Multiple comparisons and related issues in the interpretation of epidemiologic data.’” Am. J. Epidemiol. 147 (1997): 801–6CrossRef Google Scholar

Goodman, S.N. “Multiple comparisons explained.” Am. J. Epidemiol. 147 (1997): 807–12CrossRef Google Scholar

Savitz, D.A. and Olshan, A.F. “Describing data requires no adjustment for multiple comparisons: A reply from Savitz and Olshan.” Am. J. Epidemiol. 147 (1997): 813–14CrossRef Google Scholar

Thompson, J.R. “A response to ‘Describing data requires no adjustment for multiple comparisons.’” Am. J. Epidemiol. 147 (1997): 815CrossRef Google Scholar

Accessibility standard: Unknown

Why this information is here

This section outlines the accessibility features of this content - including support for screen readers, full keyboard navigation and high-contrast display options. This may not be relevant for you.

Accessibility Information

Accessibility compliance for the PDF of this book is currently unknown and may be updated in the future.