Question

Critical thinking exercise in Triola nd Triola biostatistics assignment

"From Data to Decision" on page 565 of the textbook. perform the one-way analysis of variance (ANOVA) for the data

Placebo group: 77, 61, 66, 63, 81, 75, 66, 79, 66, 75, 48, 70

10mg Treatment group: 67, 48, 79, 67, 57, 71, 66, 85, 75, 77, 57, 45

20mg Treatment group: 72, 94, 57, 63, 69, 59, 64, 82, 34, 76, 59, 53

Based on results, does it appear that there is sufficient evidence to claim that the drug lowers pulse rate? Are there any serious problems with the design of the experiment? Given this was only involving males, can we say that results apply to females? Is there a better way in measuring the drug's effectiveness in lowering pulse rates? How would you characterize the overall validity of the experiment? Based on available results, should the drug be approved?   

Answer 1

solution:

Descriptive of Pulse Rate
	N	Mean	Std. Deviation	Std. Error	95% Confidence Interval for Mean
					Lower Bound	Upper Bound
Placebo	12	68.92	9.288	2.681	63.02	74.82
10 mg Treatment Group	12	66.17	12.379	3.574	58.30	74.03
20 mg Treatment Group	12	65.08	15.222	4.394	55.41	74.76
Total	36	66.72	12.279	2.047	62.57	70.88

ANOVA of Pulse Rate with three groups.
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	93.722	2	46.861	.298	.744
Within Groups	5183.500	33	157.076
Total	5277.222	35

1. Conclusion: Drug has no significant effect even with change in dose i.e. there is NO evidence to claim that drug lowers pulse rate.
2. Definitely there are serious problem in the design of experiments. In the design Local Control is one of the important Principle. The text does not reflect to what extent it is applied.
3. When the drug is to be recommended to the population in general, it is highly essential to incorporate the results from Female and then decision should be taken.
4. It is better to collect the data on pulse rate before the treatment is given (Pre Data) and the data on pulse rate after the treatment is given (Post Data). The difference in these two must be tested for its significance change.
5. In fact, it is better not to approve this drug in general.