Statistical Test

Statistical Test: hypothesis testing

Statistically significant relationship? or difference between two or more groups.

Test statistics: a number that describes how much the relationship between variables in the test.
p-value (probability value): extreme-> infer a statistically significant relationship.

Assumption

1. Independence of observations.
2. Homogeneity of variance.
3. Normality of data

 

Regression Test

- cause and effect relationships. 
- estimate the effect of one or more continuous variables on another variable.

Comparison Test

- Difference among group means.
- T-test for comparing the means of precisely two groups.
- One sample t-test: Compare sample and population.
- Two sample t-tests (Independent T-test): Compare two independent groups.
- Paired T-Test: After the experiment, something different?
- Anova tests the means of more than two groups.
- Two-way ANOVA: Interaction effect. (한 처치변수의 변화가 결과변수에 미치는 영향이 다른 처치변수의 수준에 따라 달라지는가?)

Correlation Test

- Check whether variables are related without hypothesizing a cause-and-effect relationship.
- Pearson's r