Methods Flashcards
What is an experiment?
A scientific procedure aimed at testing a hypothesis. Entails Isolation of the IV from all other potential causes of DV either through lab manipulation or random assignment
Cook and Campbell (1979)
3 types of validity:
Internal- Confidence of drawing cause-effect conclusions
External- extent of phenomenon generalizing to other settings (cultures/people/time)
Construct- Correct identification of the nature of the IV, DV, and their underlying relationship
What is significance level alpha?
The probability of rejecting the null hypothesis when it is true
Difference between type I and type II errors?
Type I Error is when you report your findings as significant when they are not
Type II error is when a researcher concludes a ns effect when it truly is significant
What is mediation? Moderation? How are they tested?
Moderator (AKA interaction) either activate or inhibit a causal process. The causal effect is stronger for one group. Tested by regressing DV on the product of a predictor and moderation variable
Mediation - variation in x creates variation in M which in turn creates variation in Y
What is statistical significance?
refers to whether a result is likely due to chance or to your factor of interest
Wilson, Aronson, and Carlsmith (2010)
Handbook chapter on laboratory experiments
Advantages – most important advantage is randomization across conditions. It isolates IV and cancels out extraneous effects
Disadvantages - artificial/unrealistic settings
4 stages of lab experiments- setting the stage, constructing the IV, measuring DV, post-experimental followup
Judd and Kenny (2010)
Handbook chapter on analysis issues in SP in relation to mediation/moderation; MLM; missing data; invariance
Baron and Kenny (1986)
4 step approach to mediation
Hayes (2017)
Book on anything related to mediation/moderation/conditional process analysis (mediated moderation or moderation mediation)
O’Brien (1993)
focus groups can better inform words/phrases of questionnaire, assist in recruitment, inform hypotheses, constructs, and lend credibility to the researcher
Fazio and Olson (2003)
Covers implicit measures, response latency, IAT
Stein and Mankowski (2004)
Qualitative research consists of 4 steps: asking (identifying people to study), witnessing (listening/being present), interpreting, knowing (creating publicly available representations of knowledge)
Schwarz, Groves, and Schuman (1998)
survey methods/procedures and common issues of in-person, telephone, mail methods
Reis and Gosling (2010)
Handbook chapter on methods for conducting research outside the lab. discussed field experimentation, internet methods, diary methods, trace measures
Advantages of field experiments- maximize external reliability; observe phenomena naturally; Ps not aware they are being studied
Fine and Elsbach (2000)
Makes the argument that SP should revive efforts to build theory on qualitative dat in the field in conjunction with quantitative data in the lab. Qual data could further complement the full-cycle model and lead to theories that are simple, accurate, and generalizable.
Shadish and Cook (2009)
A guide to field experiments. 4 types of quasi-experimental designs:
randomized experiments
regression-discontinuity - uses cutoff score on a measure to assign to experimental/control group
time series - measures a single unit over consecutive times on the same outcome variable
nonrandomized w/ nonequivalent comparison group - approximate answers from randomized designs through matching on stable covariates
Greenwald, Poehlman, Uhlmann, and Banaji (2009)
Meta on predictive validity on IAT. IAT has good predictive validity. explicit measures better at predicting Bx unless socially sensitive. Implicit measures better at measuring Socially sensitive topics
