# 6.E: Research Design (Exercises)

## General Questions

### Q1

To be a scientific theory, the theory must be potentially ______________.

### Q2

What is the difference between a faith-based explanation and a scientific explanation?

### Q3

What does it mean for a theory to be parsimonious?

### Q4

Define reliability in terms of parallel forms.

### Q5

Define true score.

### Q6

What is the reliability if the true score variance is $$80$$ and the test score variance is $$100$$?

### Q7

What statistic relates to how close a score on one test will be to a score on a parallel form?

### Q8

What is the effect of test length on the reliability of a test?

### Q9

Distinguish between predictive validity and construct validity.

### Q10

What is the theoretical maximum correlation of a test with a criterion if the test has a reliability of $$0.81$$?

### Q11

An experiment solicits subjects to participate in a highly stressful experiment. What type of sampling bias is likely to occur?

### Q12

Give an example of survivorship bias not presented in this text.

### Q13

Distinguish "between-subject" variables from "within-subjects" variables.

### Q14

Of the variables "gender" and "trials," which is likely to be a between-subjects variable and which a within-subjects variable?

### Q15

Define interaction.

### Q16

What is counterbalancing used for?

### Q17

How does randomization deal with the problem of pre-existing differences between groups?

### Q18

Give an example of the "third variable problem" other than those in this text.

## Contributor

• Online Statistics Education: A Multimedia Course of Study (http://onlinestatbook.com/). Project Leader: David M. Lane, Rice University.