Validity
What is internal validity?
The ability to show that manipulation of the independent variable actually causes the dependent variable to change
What is a spurious causation?
It seems like X and Y are related causally but the relationship is actually caused by another variable Z
Bonus: Give an example of a spurious relationship
In a normal distribution, what percentage of observations are within 2 standard deviations of the mean?
95%
Bonus: What does this number have to do with a confidence interval?
What is the difference between a measure and a concept?
A concept is a broader phenomenon, while a measure is a specific empirical thing that you can collect direct data on
Type of sampling where after the first round, the researcher will ask the participant with help identifying potential other participants
Snowball Sample
What is external validity?
The ability to generalize research findings from one study to another.
Bonus: Give an example of a study that could have low external validity
Define the independent and dependent variable.
Independent - variable that is manipulated by the researcher to explore its effect on the dependent variable
What does the N of a study refer to?
How large the sample size is
What is the difference between a model and a theory?
A model is a way of simplifying a phenomenon that makes it easier to research, a theory is a broad framework for understanding a class of phenomena (usually well tested and established)
Bonus: Give an example of each
Style of sampling where groups are identified as sampling units and then only groups containing those units are sampled
Cluster Sampling
Bonus question: What might be an example of cluster sampling?
Convergent Validity
Does the measure correlate highly with other, previously established measures of the same concept? Does the measure have a clear relationship with other independent and dependent variables in ways that match our theoretical expectations?
Bonus: Give an example of how we could use convergent validity
What are the four components of a good hypothesis?
Empirical statement, generality, plausibility, specificity
Bonus: Give an example of a good hypothesis
Why is the mean often not the most useful statistic for understanding an issue?
Number of ways that means can be easily biased by outliers
Bonus: Give an example where the mean would not be a helpful statistic to use
Why does political science not have "laws" in the way the natural sciences do?
People respond very differently in different situations in contrast with molecules which respond the same under conditions. It is also much more difficult to isolate a particular causal variable
What is the difference between a stratified random sample and a random sample?
Elements of a stratified sample are divided into groups and elements in each stratum are sampled in proportion to representation in the total population.
Bonus: What is an example of something you would study where a stratified sample would be necessary?
What is the difference between validity and reliability?
Validity is the extent to which your study is free of non-random error, reliability is the extent to which an individual measure is free of random error (ie does your scale show the same results when you use it multiple times)
How, generally, do political scientists attempt to prove causality in terms of a relationship between X and Y?
Articulate a mechanism for why X would cause Y
Show that X and Y are associated
Show that the order is right (X precedes Y)
Control for spurious variables (Z)
Give an example of a phenomenon that would be very difficult to measure quantitatively.
Dennis's Discretion :)
What is constructionism?
Bonus: Give an example of a social construct and explain why it is socially constructed.
Why, from a statistical perspective, is having a large sample better than a smaller one?
Larger samples are more likely to be representative of the population, outliers are less impactful on your findings
Does the SAT have a high degree of validity and reliability? Explain for each
Dennis's Discretion
Political Scientists have often noted that on days when people buy more ice cream, there are higher levels of crime. One could be tempted to argue that this is a causal relationship. Why is this a spurious relationship?
Dennis's Discretion :)
Give an example of a phenomenon that you would expect to be roughly normally distributed and explain why you have that expectation.
Dennis's Discretion :)
Explain the difference between ordinal, nominal, interval, and ratio measures
Ordinal - Variables assigned to categories but where some measures are considered higher than others
Nominal - Values assigned only represent different categories for that variable
Interval - Same as above but the size of the gaps between variable categories has meaning
Ratio - Scale including relationships between variables where the intervals and order matter, and the scale has a meaningful zero
Bonus: Give an example of each
Imagine you need to sample to get a good idea of how people feel about COVID vaccines. What are sampling techniques you could use to get a reasonable population estimate?
Variety of correct answers, Dennis's discretion on points