Logistic Regression in SPSS: A Complete Guide

Explore how logistic regression uses binary outcomes and variable types—interval, ratio, ordinal, and nominal—and how to operationalise concepts and recode variables in SPSS.

Explore logistic regression by linking one predictor variable to multiple dependent outcomes, such as nicotine predicting cannabis and e-cigarette use, and screen time predicting anxiety, obesity, and poor sleep.

Logistic regression offers five reasons why it is useful in social science research. It explains odds ratios, effect size, R-squared, controls for other variables, and p values for hypothesis testing.

Explore the three main outputs of a logistic regression in SPSS: odds ratios, p values, and R-squared, and learn how each reflects effect, significance, and explained variance.

Explore how odds ratios from logistic regression in SPSS indicate whether each independent variable increases or decreases the odds of the dependent outcome, with percentage interpretations.

R-squared shows the proportion of variation in the dependent variable explained by the independent variables in your SPSS model. Higher values imply more variation explained, guiding model comparison for predictors.

Introduce the steps of the quantitative research process and show how logistic regression fits into them, including evaluating secondary data and operationalising concepts.

Identify the hypothesis as a prediction about the relationship between dependent and independent variables. Contrast it with the null hypothesis of no relationship.

Use p-values to test hypotheses in logistic regression with SPSS, rejecting the null when the p-value is under 0.05 to demonstrate statistical significance and assess relationships.

Craft clear, focused, and arguable research questions, distinguish them from hypotheses and null hypotheses, and plan regression analyses, including time-series and controlled factors, to study outcomes like adolescent sleep duration.

Assess the questionnaire content and variable coding to determine if the dataset supports logistic regression by recoding variables to binary, focusing on sleep duration and screen time.

Perform univariate analysis in SPSS during preliminary data exploration using frequency tables and central tendency measures like mean, median, and mode, with charts such as bar, pie, histogram, and line.

Review the quantitative research process, from literature review and theory to formulating hypotheses and null hypotheses, testing with p-values in SPSS, and preparing and cleaning data for logistic regression.

Explore exploratory analysis techniques in SPSS, learn to load data and navigate functions, and set up logistic regression using a screen time and sleep example.

Apply logistic regression in SPSS to examine how social media and smartphone use relate to sleep duration among American adolescents. Use survey data to test hypotheses and control factors.

See how SPSS converts text responses to numbers, build a codebook coding female as 1 and male as 0 in Excel, and view value labels in SPSS for logistic regression.

Learn to recode variables in SPSS into binary forms for logistic regression. The lecture demonstrates recoding sleep duration and screen time, plus creating frequency tables to interpret binary outcomes.

Explore cross tabulations and bivariate analysis in SPSS by linking screen time to a binary sleep variable, using row percentages and chi square to assess correlation.

Visualize the relationship between screen time and eight hours of sleep with a line chart of percentages, showing increased screen time relates to fewer people achieving eight hours of sleep.

Explore simple logistic regression with one binary dependent variable and one independent variable, using the recoded sleep variable and screen time, and prepare with contingency tables and a reference category.

Perform a simple logistic regression in SPSS with sleep as the binary outcome and screentime as a categorical predictor, using no screen time as the reference and interpreting the results.

In this SPSS logistic regression, interpret case processing, dependent coding, and the reference category; observe odds ratios, R-squared about 0.01, and significant effects for three or more hours.

Visualize odds ratios from logistic regression by creating a simple bar chart in Google Sheets or Excel, using a no screen time reference to interpret eight hours of sleep outcomes.

Summarize visualizations from a logistic regression, including a time series of eight hours or more sleep per night and screen time, and odds ratios with p-values and a Nagaki R-squared value.

Conclude that logistic regression reveals a significant negative link between screen time and eight hours of sleep in adolescents, with higher screen time lowering sleep odds; prep for multiple regression.

Explore how to control for other variables in multiple logistic regression by adding demographic factors, health behaviors, and other factors to isolate screen time's effect on sleep duration.

Find the control variables for logistic regression in SPSS by coding age and sex with reference categories, and review health behavior, electronic bullying, mood variables, and odds ratios.

Execute binary multiple logistic regression in SPSS (model 2) to predict eight hours of sleep from screen time, age, and sex, using reference categories and reporting odds ratios and p-values.

Explore how Model 4 in SPSS logistic regression links screen time, electronic bullying, and depressive feelings to eight hours of sleep, interpreting odds ratios and R-squared improvements.

Explore logistic regression in SPSS, showing how adding demographic factors and health behaviors changes screen time's effect on eight hours of sleep, with R-squared and p-values indicating significance.

Celebrate completing the course and gain confidence to perform independent simple or multiple logistic regression in SPSS, applying the quantitative research process.