Short Answer

A researcher argues that to maximize the calculated tt statistic, they should allocate a fixed sample size of N=100N = 100 unequally between two groups (e.g., n1=90n_1 = 90 and n2=10n_2 = 10) rather than keeping them equal (n1=50n_1 = 50, n2=50n_2 = 50), assuming the group means and standard deviations are equal and constant. Evaluate the validity of this researcher's argument using the mathematical structure of the independent-samples tt-test formula.

Question: A researcher argues that to maximize the calculated tt statistic, they should allocate a fixed sample size of N=100N = 100 unequally between two groups (e.g., n1=90n_1 = 90 and n2=10n_2 = 10) rather than keeping them equal (n1=50n_1 = 50, n2=50n_2 = 50), assuming the group means and standard deviations are equal and constant. Evaluate the validity of this researcher's argument using the mathematical structure of the independent-samples tt-test formula.

Sample answer: The researcher's argument is mathematically incorrect. In the denominator, dividing a variance by a very small sample size (like n2=10n_2 = 10) results in a much larger term than dividing it by a larger, equal sample size (like n1=50n_1 = 50, n2=50n_2 = 50), which increases the overall denominator. Because a larger denominator decreases the tt statistic, equal group sizes actually minimize the denominator and maximize the tt statistic.

Key points:

  • Evaluate the researcher's argument as incorrect or invalid.
  • Explain that dividing a variance by a very small sample size (e.g., n = 10) yields a large value for that term in the denominator.
  • Conclude that a larger denominator mathematically results in a smaller overall t statistic.

Rubric: Full credit requires evaluating the argument as incorrect, explaining that dividing variance by a smaller sample size (e.g., 10) dramatically increases that term in the denominator, and stating that a larger denominator mathematically reduces the final t statistic.

0

1

Updated 2026-05-27

Contributors are:

Who are from:

Tags

KPU

Research Methods in Psychology - 4th American Edition @ KPU

Related