Two-Sample Tests Math Example 4

Follow the full solution, then compare it with the other examples linked below.

Example 4

hard
Two independent samples: Group 1 (n=20,xห‰=100,s=15n=20, \bar{x}=100, s=15), Group 2 (n=20,xห‰=95,s=20n=20, \bar{x}=95, s=20). Calculate the t-statistic and df for a Welch's t-test (unequal variances).

Solution

  1. 1
    SE: 15220+20220=22520+40020=11.25+20=31.25โ‰ˆ5.59\sqrt{\frac{15^2}{20} + \frac{20^2}{20}} = \sqrt{\frac{225}{20} + \frac{400}{20}} = \sqrt{11.25 + 20} = \sqrt{31.25} \approx 5.59
  2. 2
    t-statistic: t=100โˆ’955.59=55.59โ‰ˆ0.895t = \frac{100-95}{5.59} = \frac{5}{5.59} \approx 0.895
  3. 3
    Welch's df: df=(s12/n1+s22/n2)2(s12/n1)2n1โˆ’1+(s22/n2)2n2โˆ’1=(31.25)211.25219+20219โ‰ˆ976.66.67+21.05โ‰ˆ35.2df = \frac{(s_1^2/n_1 + s_2^2/n_2)^2}{\frac{(s_1^2/n_1)^2}{n_1-1}+\frac{(s_2^2/n_2)^2}{n_2-1}} = \frac{(31.25)^2}{\frac{11.25^2}{19}+\frac{20^2}{19}} \approx \frac{976.6}{6.67+21.05} \approx 35.2
  4. 4
    With dfโ‰ˆ35 and tโ‰ˆ0.895, p-value > 0.05; fail to reject Hโ‚€

Answer

tโ‰ˆ0.895t \approx 0.895, dfโ‰ˆ35. Fail to reject H0H_0. No significant difference between groups.
Welch's t-test is preferred when group variances are unequal (15 vs 20 here). Unlike pooled t-test, Welch's uses separate variances and adjusts df accordingly. The unequal variances make df slightly lower than the pooled approach.

About Two-Sample Tests

Hypothesis tests and confidence intervals for comparing parameters (means or proportions) of two independent populations. The two-sample t-test compares means; the two-proportion z-test compares proportions.

Learn more about Two-Sample Tests โ†’

More Two-Sample Tests Examples