PingalaPingalaOpen app

Why does doing many separate t-tests on the same groups raise the type I error rate?

Free Applied Statistics in Healthcare practice question on Parametric tests for comparing groups for BSN and Post-RN nursing students. Choose your answer, then open the answer and the explanation.

  1. AEach extra test adds another chance of a false positive
  2. BThe sample size in each group becomes smaller each time
  3. CThe variance within each group is underestimated each time
  4. DThe power of each individual test falls with every comparison
Show answer and explanation

✓ Correct answer A. Each extra test adds another chance of a false positive

Explanation

The type I error rate (alpha) is the chance of wrongly rejecting a true null hypothesis. If alpha is 0.05, each separate test carries a 5% risk of a false positive, and these risks accumulate as more tests are run on the same data. So with several comparisons the overall (family-wise) chance of at least one false positive becomes much larger than 0.05, which is why ANOVA or a correction such as Bonferroni is used instead of repeated t-tests.

Why each option is right or wrong

  • ✓A. Each extra test adds another chance of a false positive — Every additional comparison carries its own 5% chance of a false positive, and these accumulate into a family-wise error rate above alpha.
  • ✗B. The sample size in each group becomes smaller each time — The group sizes stay fixed; the number of comparisons does not change the number of subjects in each arm.
  • ✗C. The variance within each group is underestimated each time — Repeated testing does not change the spread of the data; the within-group variance is estimated from the same observations.
  • ✗D. The power of each individual test falls with every comparison — Power is not steadily lost by adding comparisons; the real problem is the growing chance of a false positive.
Number of independent testsChance of at least one false positive (alpha = 0.05)
15%
3about 14%
5about 23%
10about 40%
EXAM HINT Each test is a lottery ticket for a false positive — more tickets, more chances to 'win' wrongly.
اردو وضاحت
Type I error کیا ہے؟
Type I error کا مطلب ہے کہ null hypothesis سچ ہونے کے باوجود ہم اسے غلطی سے رد کر دیں۔ عام طور پر alpha 0.05 رکھا جاتا ہے، یعنی ہر ٹیسٹ میں 5 فیصد امکان ہوتا ہے کہ ہمیں جھوٹا positive مل جائے۔
بار بار t-test کرنے سے مسئلہ کیوں بڑھتا ہے؟
جب ایک ہی ڈیٹا پر کئی t-tests کیے جائیں تو ہر ٹیسٹ اپنے ساتھ 5 فیصد غلطی کا موقع لے کر آتا ہے، اور یہ مواقع جمع ہوتے جاتے ہیں۔ اس لیے مجموعی طور پر کم از کم ایک جھوٹا positive ملنے کا امکان 0.05 سے کافی زیادہ ہو جاتا ہے، اسی لیے ANOVA یا Bonferroni correction استعمال کی جاتی ہے۔

Key terms

  • family-wise error rate — The overall chance of at least one false positive across a whole set of tests. (کئی ٹیسٹوں کو ملا کر کم از کم ایک جھوٹا positive ملنے کا مجموعی امکان۔)

✓ Checked by Pingala Judge · Updated 13 Nov 2026

Practise 10 more Applied Statistics in Healthcare questions free →

Adaptive MCQs with explanations in English and Urdu, for BSN and Post-RN nursing students.

More Applied Statistics in Healthcare MCQs

All Applied Statistics in Healthcare MCQs · All subjects