In a two-sample t-test, the dependent variable should be measured on which scale?

In a two-sample t-test, the dependent variable should be measured on which scale?

• A. Ratio (continuous)
• B. Ordinal
• C. Nominal
• D. Binary

Answer: (A)

The main requirement is that the variable you compare between groups supports meaningful arithmetic operations, since the two-sample t-test uses means and standard deviations. That means the dependent variable should be continuous and measured on a ratio (or at least interval) scale. A ratio scale provides a true zero and meaningful ratios between values, which underpins the calculation of differences in means and the associated variability the t-test relies on. If the outcome were ordinal, nominal, or binary, arithmetic like averaging or computing standard deviation wouldn’t be appropriate, and you’d typically use non-parametric tests or chi-square/logistic approaches instead.

The main requirement is that the variable you compare between groups supports meaningful arithmetic operations, since the two-sample t-test uses means and standard deviations. That means the dependent variable should be continuous and measured on a ratio (or at least interval) scale. A ratio scale provides a true zero and meaningful ratios between values, which underpins the calculation of differences in means and the associated variability the t-test relies on.

If the outcome were ordinal, nominal, or binary, arithmetic like averaging or computing standard deviation wouldn’t be appropriate, and you’d typically use non-parametric tests or chi-square/logistic approaches instead.