STUDENTSTATSLAB
SPSS Thesis Quick Reference Pack
Decision Tree • Assumptions • Menu Paths • Output Reading • Reporting
KEEP THIS NEXT TO SPSS
Use these pages while planning an analysis, navigating SPSS, checking assumptions and writing your results.
Lamprou Maritina
Statistics • SPSS • Research Data Analysis
, STUDENTSTATSLAB | SPSS THESIS QUICK REFERENCE
1. Statistical Test Decision Tree
Start with the research question and outcome variable. Then follow the structure below.
STEP 1 - WHAT TYPE OF OUTCOME?
Continuous: a numeric score such as Exam_Score or Study_Hours_Week.
Categorical: categories such as Passed = Yes/No.
Scale reliability: several questionnaire items intended to represent one construct.
Decision map
If your question looks like this... Ask yourself... Common starting analysis
Compare a continuous outcome between 2 Are the groups independent? Independent-samples t-test
separate groups
Compare a continuous outcome across 3+ Are groups independent? One-way ANOVA
separate groups
Compare the same people at 2 time points Are observations paired? Paired-samples t-test
Test association between 2 categorical variables Are observations independent? Chi-square
Examine association between 2 continuous Is a linear association sensible? Pearson correlation
variables
Model a continuous outcome from one or more Are regression assumptions Linear regression
predictors reasonable?
Assess consistency of several items Do items plausibly measure one Cronbach's alpha
construct?
REMEMBER
This is a starting map, not an automatic test selector. Study design, measurement level, assumptions and the
exact research question still matter.
Page 2
SPSS Thesis Quick Reference Pack
Decision Tree • Assumptions • Menu Paths • Output Reading • Reporting
KEEP THIS NEXT TO SPSS
Use these pages while planning an analysis, navigating SPSS, checking assumptions and writing your results.
Lamprou Maritina
Statistics • SPSS • Research Data Analysis
, STUDENTSTATSLAB | SPSS THESIS QUICK REFERENCE
1. Statistical Test Decision Tree
Start with the research question and outcome variable. Then follow the structure below.
STEP 1 - WHAT TYPE OF OUTCOME?
Continuous: a numeric score such as Exam_Score or Study_Hours_Week.
Categorical: categories such as Passed = Yes/No.
Scale reliability: several questionnaire items intended to represent one construct.
Decision map
If your question looks like this... Ask yourself... Common starting analysis
Compare a continuous outcome between 2 Are the groups independent? Independent-samples t-test
separate groups
Compare a continuous outcome across 3+ Are groups independent? One-way ANOVA
separate groups
Compare the same people at 2 time points Are observations paired? Paired-samples t-test
Test association between 2 categorical variables Are observations independent? Chi-square
Examine association between 2 continuous Is a linear association sensible? Pearson correlation
variables
Model a continuous outcome from one or more Are regression assumptions Linear regression
predictors reasonable?
Assess consistency of several items Do items plausibly measure one Cronbach's alpha
construct?
REMEMBER
This is a starting map, not an automatic test selector. Study design, measurement level, assumptions and the
exact research question still matter.
Page 2
