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.

Discussion Points:

1. Why Use Propensity Score Matching?

   • To control for confounding variables

   • To simulate some of the characteristics of randomized controlled trials (RCTs)

   • To improve causal inference in observational data

2. Steps to Perform PSM in SPSS:

   • Step 1: Install SPSS Plugins

     • You may need to install the SPSS Python Essentials or use an extension like the PS Matching plugin (available via SPSS Amos or the IBM extension hub).

     • Alternatively, use R plugins if needed.

   • Step 2: Estimate Propensity Scores

     • Go to: Analyze → Regression → Binary Logistic

     • Set the treatment (binary) as the dependent variable

     • Enter covariates (potential confounders) as independent variables

     • Save the predicted probabilities (Propensity Scores)

   • Step 3: Perform Matching

     • Matching is not directly available in basic SPSS—options include:

       • Use SPSS Custom Dialog: "PS Matching"

       • Use syntax with Python plugins

       • Export data and use R with MatchIt package

   • Step 4: Check Balance

     • Compare covariates between matched groups

     • Standardized mean differences (SMD), t-tests, chi-square tests

   • Step 5: Outcome Analysis

     • Analyze matched pairs with appropriate tests (paired t-tests, conditional logistic regression, etc.)

3. Resources and Tutorials:

   • IBM Extension Hub - PS Matching Dialog

   • SPSS PSM Syntax examples

   • YouTube video tutorials

   • R packages (MatchIt, PSAgraphics) for more advanced matching options

4. Common Pitfalls to Avoid:

   • Poor overlap of propensity scores between groups

   • Unbalanced covariates after matching

   • Small matched sample sizes