If you are a research methods instructor and are interested in learning more about my tests, please contact me at rottman@pitt.edu.

How to Prepare:

Many of the test questions will look similar to questions from the homework assignments on "video game violence" and "procrastination". I will use the same formats as many of the multiple choice questions for these
Carefully read the In Defense of Parenthood article and be prepared to answer questions about it. While reading it, I suggest thinking about how the article connects to concepts we have discussed in class.
In general, the homework and in-class exercises are good practice for the sorts of things that will be on the test. So you can go over these by removing the answers and testing yourself. However, there are some key differences:
- For the test, it is multiple choice, so I will only give you things that I think have clear answers, unlike some of the questions I have given you previously that were intended to make you grapple with less clear cases
- I will not ask you to do stuff in excel like you did for one week in class. However, the concepts (what does controlling for variables mean) still are relevant

Topics:

The test will broadly cover topics from the following topics. There may be other topics that we have covered that did not make it on this list, but this is the majority:

understanding Causal Structures and creating a causal structure from text
Observational Studies vs. Experiments
Identifying the type of a variable (Binary, Categorical/Nominal, Ordinal, Metric/Interval/Continuous)
False Positives and False Negatives
The factors that influence p-values and power:
- sample size
- effect size
  - for two groups: mean difference and variation within each group
  - for a continuous IV: slope and variation/noise
Deciding which statistical test to run based on the types of variables
Understanding when to include a variable in a regression and why or why not
Understanding the relationships between causal structures, scatterplots, and regression results for the following types of causal structures:
- Confounds/Common Causes
- Noise/Alternative Causes
- Alternative Effects
- Common Effects
- Mediators
- Moderators/Interactions

4 Main types of Validity and the sub-types of Construct Validity
- Internal Validity (which has to do with causality, note that this is not the same thing as internal reliability)
- External Validity
- Statistical Validity
- Construct Validity (and the following sub-types)
  - Subjective aspects of Validity
    - Face Validity
    - Content Validity
  - Objective
    - Criterion Validity
    - Convergent and Discriminant Validity
    - Reliability
      - Inter-Rater Reliability
      - Internal Reliability
      - Test-Retest Reliability
Common problems with survey questions and how to write better survey questions