Question

Consider the follow-up study data from the Evans County (Georgia) Heart Disease Study described by Cassels (1971). The data pertain to a cohort of 609 healthy white males between the ages of 40 and 76 who were residents of Evans County in 1960. The cohort was followed for 7 years, after which new cases of coronary heart disease (CHD) were identified. The level of circulating catecholamine (CAT) is the exposure variable of interest (defined as 1 = high and 0 = normal). The purpose of the study was to assess the putative association between CAT (as measured in 1960) and the subsequent 7-year incidence of CHD controlling for other potential confounders. The data are summarized below:

All other variables of interest for this example include:

AGE: age (in years);
CHL: cholesterol level (mg/dl);
SMK: smoking status (1 = ever smoked or 0 = never smoked);
ECG: electrocardiogram abnormality (1 = yes or 0 = no);
DBP: diastolic blood pressure (mm Hg);
SBP: systolic blood pressure (mm Hg);
HPT: high blood pressure defined as having SBP ≥ 160 or DBP ≥ 95 (1 = yes or 0 = no)

What type of analysis is this?

Select one:

a. ANCOVA

b. Linear regression

c. Logistic regression

d. One-way ANOVA

e. Two-way ANOVA

Answer 1

The analysis given is the Logistic regression.

The reason for using logistic regression for this problem is that the values of the dependent are represented by "1" and "0" , not cardinal numbers.