A researcher measures the reaction times (in milliseconds) of two independent groups of participants under different sleep-deprivation conditions. Although the reaction time data is measured at the interval level, a Shapiro-Wilk test reveals that the distribution of reaction times is significantly skewed (p<0.05p < 0.05p<0.05).
Which of the following describes the correct classification of the statistical test that the researcher should use to analyse the difference between the groups, and the primary justification for this choice?
A parametric test, because the independent group design allows the researcher to assume homogeneity of variance.
A non-parametric test, because the reaction time data violates the assumption of a normally distributed population.
A parametric test, because reaction times are measured at the interval level and are therefore inherently robust to skewness.
A non-parametric test, because reaction time measurements represent ordinal data that cannot be analysed using parametric methods.
