Should I use a t-test or ANOVA?

If you are comparing two groups, use a t-test. For three or more groups, one-way ANOVA should be used. Both tests assume normality.

Should I choose a parametric or non-parametric test?

If your data are normally distributed and the sample is large enough (n ≥ 30), choose parametric tests (t-test, ANOVA). If normality is not met or the sample is small, non-parametric alternatives (Mann-Whitney U, Kruskal-Wallis, Wilcoxon) give more reliable results.

Which correlation test should I use?

Use the Pearson correlation coefficient if the data are normally distributed; use Spearman's rank correlation for ordinal data or when normality is violated. Kendall's tau is suitable for small samples or many tied ranks.

Which test is used for categorical data?

The chi-square test of independence is used to test the relationship between two categorical variables. If cell frequencies are small (expected < 5), Fisher's exact test should be used.

Which test is used in a pre-test/post-test design?

If the same group was measured twice, use the paired-samples t-test. If the normality assumption is not met, use the Wilcoxon signed-rank test. With three or more measurement points, use repeated measures ANOVA or the Friedman test.

Which analysis is used for scale reliability?

Cronbach's alpha is calculated to measure the internal consistency of a questionnaire or scale. α ≥ 0.70 indicates acceptable, α ≥ 0.80 good and α ≥ 0.90 excellent reliability.

Which analysis is used to predict a continuous dependent variable?

Use simple linear regression with one independent variable and multiple linear regression with several. If the dependent variable is binary (0/1), logistic regression should be used.

Which Statistical Test Should I Use?

Q: Which test is used in a pre-test/post-test design?

If the same group was measured twice, use the paired-samples t-test. If the normality assumption is not met, use the Wilcoxon signed-rank test. With three or more measurement points, use repeated measures ANOVA or the Friedman test.

Q: Which analysis is used for scale reliability?

Cronbach's alpha is calculated to measure the internal consistency of a questionnaire or scale. α ≥ 0.70 indicates acceptable, α ≥ 0.80 good and α ≥ 0.90 excellent reliability.

Q: Which analysis is used to predict a continuous dependent variable?

Use simple linear regression with one independent variable and multiple linear regression with several. If the dependent variable is binary (0/1), logistic regression should be used.

Answer a few questions and find the right test for your research in seconds.

Step 1 —

Quick Guide to the Tests

You can run the test recommended by the decision tree on Analizus for free.

t-Test

Used to compare two group means. The independent-samples t-test compares different groups; the paired-samples t-test compares two measurements from the same group. It assumes normality.

Calculate on Analizus →

One-Way ANOVA

Compares the means of three or more independent groups. After a significant F, Tukey or Bonferroni post-hoc tests identify which groups differ.

Calculate on Analizus →

Correlation Analysis

Measures the direction and strength of the relationship between two variables. Pearson is parametric (normal distribution); Spearman and Kendall are non-parametric (ordinal data) correlation tests.

Calculate on Analizus →

Linear Regression

Predicts a continuous dependent variable from one or more independent variables. R², β coefficients and a VIF multicollinearity check are reported.

Calculate on Analizus →

Chi-Square Test

Tests the relationship between two categorical variables. When an expected cell frequency is below 5, Fisher's exact test is applied automatically.

Calculate on Analizus →

Exploratory Factor Analysis

Uncovers the latent factors underlying many variables. For scale validity, KMO (≥ .60), Bartlett's test (p < .05) and the factor loading matrix are examined.

Calculate on Analizus →

Wilcoxon Test

A non-parametric test comparing two dependent measurements. It is preferred when the normality assumption of the paired t-test is not met. Ideal for pre-test/post-test designs.

Calculate on Analizus →

Cronbach's Alpha

Measures the internal consistency of a questionnaire or scale. α ≥ 0.70 is acceptable, α ≥ 0.80 good. It is reported as evidence of scale reliability in theses and articles.

Calculate on Analizus →

Which Statistical Test Should I Use?

Quick Guide to the Tests

t-Test

One-Way ANOVA

Correlation Analysis

Linear Regression

Chi-Square Test

Exploratory Factor Analysis

Wilcoxon Test

Cronbach's Alpha

Share Your Analysis in the Forum