Confidence intervals for averages should be calculated only on variables that are considered metric.

Question

Quizplus · Accepted Answer

Confidence intervals for averages, also known as mean confidence intervals, are typically calculated only on variables that are considered metric or numerical in nature. This is because the computations of the confidence interval involve assumptions about the normality of the population distribution and the properties of the sample mean. Categorical variables, on the other hand, do not have a natural ordering or meaningful units of measurement, and thus are not suitable for mean-based confidence intervals. For categorical variables, other types of confidence intervals, such as proportion or odds-ratio based intervals, may be more appropriate.

Confidence Intervals for Averages Should Be Calculated Only on Variables