Which term describes a non-parametric test that compares the variability of scores in two groups?

Which term describes a non-parametric test that compares the variability of scores in two groups?

• A. Monte Carlo method
• B. Moses extreme reactions
• C. Multicollinearity
• D. Multivariate

Answer: (B)

The question is asking about a non-parametric method used to compare how spread out scores are between two independent groups. The Moses extreme reactions test is designed exactly for this purpose: it assesses whether two samples have different variability without assuming that the data are normally distributed. It does this by focusing on extreme observations and how often they occur in each group, which provides a sense of which group has more dispersion. Because it targets variability rather than central tendency, it’s appropriate when distributions are skewed or when outliers make parametric tests unreliable. The other options don’t fit this goal. The Monte Carlo method is a general computational simulation approach, not a test of dispersion. Multicollinearity is a problem in regression where predictors are highly correlated, not a test about variability between groups. Multivariate refers to analyses involving more than one dependent variable, not specifically to comparing the variability of two groups.

The question is asking about a non-parametric method used to compare how spread out scores are between two independent groups. The Moses extreme reactions test is designed exactly for this purpose: it assesses whether two samples have different variability without assuming that the data are normally distributed. It does this by focusing on extreme observations and how often they occur in each group, which provides a sense of which group has more dispersion. Because it targets variability rather than central tendency, it’s appropriate when distributions are skewed or when outliers make parametric tests unreliable.

The other options don’t fit this goal. The Monte Carlo method is a general computational simulation approach, not a test of dispersion. Multicollinearity is a problem in regression where predictors are highly correlated, not a test about variability between groups. Multivariate refers to analyses involving more than one dependent variable, not specifically to comparing the variability of two groups.