Your cart is currently empty!
M09.02.11 Statistical test
Selecting an appropriate statistical test is essential for meaningful data analysis. This choice hinges on the type and scale of the variables in the study, as well as the research question.
Overview
Statistical tests analyze relationships between variables and assess differences across groups. The type of variable (nominal, ordinal, interval, or ratio) and the number of groups are key factors in choosing the correct test. Here, we focus on common tests in biostatistics.
Types of Statistical Tests
Each statistical test serves a specific purpose, depending on the variable scale and research needs.
| Test Name | Variable Types | Application |
|---|---|---|
| Pearson Correlation | Interval | Measures the linear relationship between two variables. |
| Chi-square | Nominal | Tests for association between categorical variables. Supports any number of groups. |
| t-test | Interval, Nominal | Compares means between two groups. |
| One-way ANOVA | Interval, Nominal | Tests for mean differences across two or more groups. |
| Matched Pairs t-test | Interval, Nominal | Compares means between two related groups, often pre- and post-test. |
| Repeated Measures ANOVA | Interval, Nominal | Tests for differences across multiple time points within the same group. Suitable for linked data. |
Note:
- ANOVA stands for Analysis of Variance, which is used to compare group means and assess the variance within groups relative to the variance between groups.
Key Points to Remember
- Variable Type: The nature of your variables (e.g., nominal, ordinal, interval) determines the test you should select.
- Number of Groups: The t-test is limited to two groups, while ANOVA can handle two or more.
- Linked Data: Tests like matched pairs t-test and repeated measures ANOVA are specifically designed for linked or repeated data.
References
- Bowers, D. (2019). Medical Statistics from Scratch: An Introduction for Health Professionals. Wiley.
- Sullivan, L. M. (2011). Essentials of Biostatistics in Public Health. Jones & Bartlett Learning.