What is Logistic Regression?

Logistic regression is a statistical method used when the outcome (dependent) variable is binary — for example, disease vs. no disease or success vs. failure. It helps estimate the probability of an event occurring, based on one or more independent variables. Example: predicting whether a patient develops hypertension based on BMI and age.

2. Setting Up Logistic Regression in SPSS

In SPSS, you can run logistic regression by going to:
Analyze → Regression → Binary Logistic
Select your dependent variable (must be coded 0 and 1) and independent variables (categorical or continuous). Example: Outcome: Diabetes (1 = Yes, 0 = No), Predictors: BMI, Smoking Status.

3. Interpreting the Output: Odds Ratio (Exp(B))

The key result is the Exp(B) column in SPSS, which gives the odds ratio. An odds ratio >1 means increased odds of the event; <1 means decreased odds. Example: An odds ratio of 2.0 for smoking means smokers are twice as likely to develop the disease compared to non-smokers.

4. Assessing Model Fit: The Hosmer-Lemeshow Test

Always check the Hosmer-Lemeshow goodness-of-fit test to evaluate how well your model fits the data. A p-value >0.05 suggests good fit. Example: A p-value of 0.21 means your model fits the data well.

5. Reporting Results for Publication

When writing your paper, report the odds ratios, 95% confidence intervals, and p-values clearly. Example: “Smoking was associated with higher odds of hypertension (OR = 2.0, 95% CI: 1.5–2.7, p < 0.001).” This format is suitable for most medical and scientific journals.