Behavioral Research: Statistical Methods
CG3.402True / False (with reasoning)
Exposes shallow understanding. Always include the reason.
A p-value is the probability that H₀ is true.
Pearson r = 0 implies the two variables are independent.
The CLT requires the population to be Normal.
A measurement can be valid without being reliable.
A Bayes Factor can quantify evidence for the null hypothesis.
Adding more predictors always increases R².
Power = 1 − α.
Chi-square tests assume expected cell counts ≥ 5.
A significant F in one-way ANOVA tells us which specific groups differ.
Bonferroni correction is more conservative than Benjamini-Hochberg.
Logistic regression assumes residuals are normally distributed.
PCA components are uncorrelated by construction.
The median is robust to outliers; the mean is not.
A 95% confidence interval has a 95% probability of containing the parameter.
Welch's t-test does not assume equal variances.