Question: To determine whether the drug treatment reduces the mean number of wor

Question

Question: To determine whether the drug treatment reduces the mean number of worms, which t-test should be used in the case? Paired or Unpaired t-test?

At each of 7 farms a sample of 10 worm-infected lambs was randomly divided into 2 groups. Five were injected with the drug and the remainder were left untreated. The two groups of 5 lambs were raised in separate paddocks at each farm (i.e. 2 paddocks per farm) for a 1-month period, slaughtered and the following mean worm counts per five lambs recorded: Mean worm counts per 5 lambs (rounded to whole numbers) Farm Drug-treated lambs 19 43 28 50 16 33 14 Untreated lambs 40 54 26 63 21 37 39

Explanation / Answer

First of all the data is a 'count data' and hence not a continuous one, whereas t-test( any kind ) is used for continuous data. One resort is to assume that the average count of worms is large enough for both the treated and untreated group and then apply Central Limit Theorem. As we have not counted the worms from the same lamb before and after treatment of drugs so there is no question of arising paired data. If anyone wants to perform a t-test one should do a unpaired t-test even if the lambs were in identical condition before any application to them.

Related Questions

Navigate

• Browse (All)
• Browse Q
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.