Which analysis should I use?

Compare two groups (e.g., male vs female)

→ Independent t-test

Tests whether two group means are statistically different

Compare before & after (e.g., pre/post training)

→ Paired t-test

Tests whether the same subjects changed over time

Compare against a standard (e.g., mean > 3)

→ One-sample t-test

Tests if sample mean differs from a reference value

Compare three or more groups

→ One-way ANOVA

Compares means across multiple groups at once

Two factors at once (e.g., gender × age)

→ Two-way ANOVA

Tests main effects and interaction of two factors

Same subjects measured 3+ times

→ Repeated Measures ANOVA

Tests change over multiple time points

Non-normal data, two groups

→ Mann-Whitney U

Non-parametric alternative to the t-test

Non-normal data, three+ groups

→ Kruskal-Wallis

Non-parametric alternative to ANOVA

Do two variables move together?

→ Pearson Correlation

Measures linear relationship strength

Ranked or non-normal data relationship

→ Spearman Correlation

Rank-based relationship measure

Which factors influence an outcome?

→ Multiple Regression

Identifies predictors and their effect sizes

Outcome is yes/no

→ Logistic Regression

Predicts probability of a binary outcome

A→B→C indirect pathway

→ Mediation

Tests indirect effects through a mediator

Effect changes depending on a condition

→ Moderation

Tests whether an effect varies by condition

How do survey items group together?

→ Exploratory Factor Analysis (EFA)

Discovers hidden factor structure

Are items measuring consistently?

→ Reliability Analysis

Checks internal consistency of a scale

Does my hypothesized structure fit?

→ Confirmatory Factor Analysis (CFA)

Tests a pre-specified factor model

Full causal model with latent variables

→ Structural Equation Modeling (SEM)

Tests measurement + structural models together

Group similar respondents into types

→ Cluster Analysis

Automatically finds natural groupings

What distinguishes the groups?

→ Discriminant Analysis

Identifies key variables that separate groups

Test causal flow between variables

→ Path Analysis

Tests sequential causal relationships

How many responded to each option?

→ Frequency Analysis

Counts and percentages per category

Multiple-choice question tallying

→ Multiple Response Frequency

Tallies items where multiple selections were allowed

Mean, SD, min, max overview

→ Descriptive Statistics

Summarizes basic properties of numeric variables

Are two categorical variables related?

→ Cross-tabulation

Tests association between two categorical variables

Build prediction rules by conditions

→ Decision Tree

Creates intuitive tree-based prediction rules

Forecast trends over time

→ Time Series Analysis

Analyzes and forecasts temporal patterns

Find key words in open-ended responses

→ TF-IDF

Extracts important words automatically

Classify positive/negative/neutral

→ Sentiment Analysis

Auto-detects emotional tone of text

Discover hidden themes in many texts

→ Topic Modeling

Automatically groups texts by theme