Author Question: A researcher reports on a study conducted to determine if a new educational program has helped ... (Read 79 times)

penza · **on:** Jul 8, 2018

A researcher reports on a study conducted to determine if a new educational program has helped dialysis patients become more compliant with their fluid restrictions. The findings indicated that there was no difference.

The report did include that a power analysis was performed to determine if the sample size (n = 100) was adequate. The power level was .5. What should the reader conclude?
a. A sample of 100 is certainly adequate in a clinical study. The researcher should adjust the educational program.
b. Chances are high that a Type I error has occurred.
c. Findings of no difference are not surprising; it is difficult to make an impact on this population of patients.
d. There is a high likelihood that the sample size was not adequate, and the study should be replicated using more subjects.

Question 2

Which of the following types of studies would need the largest sample size?

a. Case studies
b. Descriptive studies
c. Experimental studies
d. Quasi-experimental studies

kusterl · **Reply #1 on:** Jul 8, 2018

Answer to Question 1

ANS: D
A power level needs to reach .8 to be acceptable, and this one had only .5. There is a good chance that a Type II error occurred, and the study should be replicated.
A broad estimate is that samples in nursing studies should average more than 200 subjects to achieve enough power to detect differences. A power level needs to reach .8 to be acceptable, and this one had only .5. There is a good chance that a Type II error occurred, and the study should be replicated.
Chances are high that a Type II error has occurred. A power level needs to reach .8 to be acceptable, and this one had only .5. There is a good chance that a Type II error occurred, and the study should be replicated.
Without additional information, this conclusion is way overstated. A power level needs to reach .8 to be acceptable, and this one had only .5. There is a good chance that a Type II error occurred, and the study should be replicated.

Answer to Question 2

ANS: B
Descriptive studies often require very large samples. Multiple variables may be examined, and extraneous variables are likely to affect subject response(s) to the variables under study.
Case studies are, by definition, composed of a very small sample, possibly even just one individual.
Experimental studies can have fewer subjects than a descriptive or correlational study.
Quasi-experimental studies can have fewer subjects than a descriptive or correlational study