1. ## Appropriate Statistical Analysis

Hello, I am getting ready to take comprehensive exams for my Master's Degree and have not had statistics in four years. I need some assistance in several areas to help me prepare for the exam. One area is dealing with determining which statistical analysis is appropriate for a scenario. These are a few that were given to me by others who have statistics in the past.

Scenario 1- A statistics professor want to examine the relationship between the number of hours spent studying statistics and performance on a midterm examiniation.

Scenario 2 - A psychologist tests a new therapy developed for reducing phobias by selecting a sample of individuals who have phobias about spiders. Before therapy he measures each subjects' fear response to a picture of a spider, and repeats the measurement after the therapy.

Scenario 3 - An education researcher examines differences in the effectiveness of four school reform projects.

Scenario 4 - A researcher investigates whether classical background music is more or less soothing to air traffic controllers than more modern background music. He plays a classical section to one group and a modern selection to another group.

I think scenario 3 is a anova and scenario two is a regression.

Any help in answering these questions is appreciated. I need to know I am on the right track.

Thank you

2. Hi. as I see it....

Senario one: If the hours spent are continuous, then use a regression model. If they are organised catergorically (o-5; 5-10 ...) assuming the test scores are normally distrubted use one-way ANOVA.

Senario two: I think this is a sign rank test (non) parametric - might need confirmation from another post though!

Senario three: One way anova if the measure of "effectiveness" is continuous and meets the assumptions for a parametric test.

senario four: the study is flawed, because how would you know if one group just happen to like modern music better than classical music? Both types would have to be played to both groups.

Phil

3. Scenario 2: T-test for dependent samples for k = 2 groups, or a 1-way repeated ANOVA for k > 2 groups

4. but will that apply if the fear responce is a rank? (say 1-5)

5. It can be either parametric or nonparametric, the question doesn't specify enough information. When in doubt, cover all the bases!

 Tweet

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts