[Experiment design] A friend of yours only had 3 hours of sleep before an exam a

Question

[Experiment design]

A friend of yours only had 3 hours of sleep before an exam and earned 75 out of 100 points. Another friend, who had 8 hours of sleep, earned an 85/100. You are curious if sleep time has an effect on exam performance.

A. Would you choose a descriptive or explanatory research methods to answer this research question and why?

B. What would you use for independent variable(s) and what would be your dependent variable?

C. What analysis method would you use to analyze your data (assume you know the exam grades of 15 students who had 3 hours of sleep and 15 students who had 8 hours of sleep), and why?

D. How would you categorize the levels of measurement for your IV(s) and DV?

E. What experimental assumptions may be critical to application of the method you identified in Part (c)?

Explanation / Answer

Part (A) : We would choose explanatory research method to answer this question. Because descriptive research method is used when one is willing to increase his/her knowledge about a particular subject matter, whereas explanatory research method is used at one advanced step in the overall research when we already have gathered sufficient knowledge on a particular subject matter and now we are interested in providing reasons as to why the gathered knowledge is in the way it is. In this question also; we already have gathered data of 2 friends and now we desire to explain the reasoning behind the observed data pattern.

Part (B) : Independent Variable is Number of hours of sleep before the exam. Dependent Variable is the total points earned in the exam. Because we wish to determine how the points earned in exam is affected by number of sleep hours.

Part (C) : We will use Hypothesis testing using 2-sample t-test. Because we are given two different samples of students of size 15 each. And we will use the sample mean as the test statistic, which can be computed from the given exam grade points of individual students; and we know that for N=15 < 30, sampling distribution of difference in sample means follows t-distribution. Therefore, 2 sample t-test is the right choice and not z-test. For z-test you need N>30.

Part (D) : Level of measurement for IV (i.e. number of sleep hours) is Ratio type , And for DV (i.e. exam points) is also Ratio type. Because in each variables' observation data, the absolute zero is defined and there is well-defined distance function in each.

Related Questions

Navigate

• Browse (All)
• Browse #
• Subjects

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