Research Methods II: Introduction to Social Psychology PDF

RESEARCH METHODS II INTRODUCTION TO SOCIAL PSYCHOLOGY https://osf.io/qk6dc THE REPLICATION CRISIS The way we do research and report findings has increased the rate of false positives. This is a major cause of replication failures. QRPs: P-ha...

RESEARCH METHODS II INTRODUCTION TO SOCIAL PSYCHOLOGY https://osf.io/qk6dc THE REPLICATION CRISIS The way we do research and report findings has increased the rate of false positives. This is a major cause of replication failures. QRPs: P-hacking Cherry-picking data HARKing Others… https://xkcd.com/882/ KEY RESEARCH IN SOCIAL PSYCHOLOGY Stanford Robber’s Social Prison Cave Priming Experiment Experiment Experiments STANFORD PRISON EXPERIMENT (PHILIP ZIMBARDO) Psychological effects of perceived power Is the brutality amongst American prison guards due to the sadistic personalities of guards or to the setting – that strips prisoners of their humanity? Results showed that guards (ordinary university students randomly assigned to condition) became very brutal https://www.youtube.com/watch?v=yUZpB57PfHs PROBLEM Behaviour of both prisoners and guards was heavily “directed” Guards told how to behave Prisoners instructed to act submissively “Prison warden” was also an experimenter who periodically escalated conflicts that arose Ultimately, conclusions do not hold https://www.newyorker.com/science/maria-konnikova/the-real-lesson-of-the-stanford- prison-experiment KEY RESEARCH IN SOCIAL PSYCHOLOGY Stanford Robber’s Social Prison Cave Priming Experiment Experiment Experiments  “Realistic Conflict Theory” 11-13 year old boys in a “summer camp” Assigned to 2 “houses” ROBBER’S CAVE Scarce resources (MUZAFER SHERIF) Fostered competition Houses driven to enmity Finally, with common purpose, groups reunite PROBLEM Robber’s cave wasn’t Sherif’s only experiment In his first experiment, he allowed the houses to become acquainted before the competition phase. Because it didn’t appear to support the theory it was aborted. Robber’s cave was carefully orchestrated to show the desired effect. https://www.theguardian.com/science/2018/apr/16/a-real-life-lord-of-the-flies-the-troubling-legacy-of-the-robbers- cave-experiment KEY RESEARCH IN SOCIAL PSYCHOLOGY Stanford Robber’s Social Prison Cave Priming Experiment Experiment Experiments IMPLICIT SOCIAL PRIMING “Automaticity Effects” Concept called to mind with a subtle prompt Words or images Activity or experience Participant then behaves in a way that is consistent with concept https://www.youtube.com/watch?v=5g4_v4JStOU PROBLEM Experimenters are often aware of task conditions They might change their OWN behaviour Either on purpose or by accident That leads to changes in participant behaviour that are consistent with conclusions Results do not speak to theories – but instead to experimental differences EXPERIMENTER EFFECTS IV: Implicit social power prime High power (“general”, “command”, “boss”) Low power (“soldier”, “obey”, “worker”) DV: Abstract thinking (number of choices) Confounding Variable: Experimenter belief KEY RESEARCH IN SOCIAL PSYCHOLOGY Stanford Robber’s Social Prison Cave Priming Experiment Experiment Experiments HOW TO ENHANCE RESEARCH OUTCOMES Use Simmons, Nelson, & Simonsohn’s (2012) 21-word solution “We report how we determined our sample size, all data exclusions (if any), all manipulations, and all measures in the study” Fully disclose study materials and procedures Some scholarly journals have recently initiated “open practice” requirements for authors Pre-register empirical studies (e.g., on the Open Science Framework) Describe the purpose of the study, specific hypotheses, all variables, what materials will be used, how data will be analyzed, etc. in advance and post online for others to see Post the raw data and results after the study is completed REDUCE P-HACKING p-hacking: “Data fishing expedition” Try different types of analyses until p-value is driven below.05 p-curve analysis attempts to detect presence of p-hacking and publication bias/file drawer problem (Simonsohn, Nelson, & Simmons, 2014) “p-curve” is the distribution of significant p-values in a body of research p-curve for p-hacked data will be left-skewed (tail on left side) More p-values around.04 or.05 than.01 or.02 p-curve for non-p-hacked data will be right-skewed (tail on right side) More p-values around.01 or.02 than.04 or.05 Don’t pass off exploratory research as confirmatory REDUCE HARKING Hypothesizing After Results are Known Look at the data first and then create a post-hoc hypothesis; present it as if it were developed a priori State your hypotheses before collecting data Make your hypotheses as clear as possible in your research proposals Should be directional, comprehensive yet parsimonious HYPOTHESIS EXAMPLES Bad hypothesis: “There will be a significant difference in collectivism between Americans and Indians” (non-directional) Good hypothesis: “Indians will be significantly higher in interdependence than Americans” Even better: build a process model in your hypotheses “Indians will be significantly higher in interdependence than Americans and, in turn, will demonstrate a more holistic cognitive style”  Simmons et al. (2011) suggest at least N = 20 per group, but this isn’t large enough Within-subject designs rather than between-subjects designs More power for the same number of participants Cohen’s effect size guidelines: IMPROVE Small effect sizes (most of what we study in social psychology) can only STATISTICAL be accurately detected with high statistical power POWER  Take-Away Point Replication should be just as important as innovation in science Registered replication reports (results are registered in advance) TAKE-AWAY Science is a continual process of updating what we know, self-correcting as we go along POINTS & Scientific Method: innovation; if doesn’t replicate, figure out why; more innovation; GOOD NEWS replicate Good News… Psychology has changed in the last few years. Open science becoming the norm. Social media instrumental.