PSYC 351 Research Test Questions with Correct Answers Latest Update

PSYC 351 Research Test Questions with Correct Answers Latest Update what is science - Answers slowly accumulate evidence in support (or refuting) theories -impossible to prove a theory (evidence just becomes overwhelming with more studies) what is replication - Answers for evidence to be considered evidence multiple independent labs must run the same experiment and get similar results over and over -if they cant replicate we question if it is real initial spark of the replication crisis - Answers -feeling the future article by Bem published in a highly prestigious journal -9 experiments found evidence for "psi" (pre-cognition: predicting the future) -randomly shown erotic pictures vs not -did better than chance but no one was convinced it was a real thing replicability project - Answers tried to replicate 100 studies, only 36% were replicated how do researchers publish false studies without committing outright fraud? - Answers -HARKing (Hypo after results are known): when you dont find supporting evidence for your actual hypo, you craft a new one to make the results you do find while making it sound like that was what you meant all along -p-hacking: (P-values = numbers used to make decisions about whether our hypothesis was supported) low p-values = good THUS people massage them to get the value a little lower ex) cut peoples results or values to get the value they really wanted add more data if it doesnt methods of research - Answers -ingredients: Institutional pressures "Publish and pressures" # of papers and where theyre published matters -flashy and significant effects needed to publish publish papers that have the more scientific impact (publication bias) if your manipulation finds no effect you will likely not get published -lots of ways to analyze the data garden of forking paths, p-hacking, intentional + unintentional there are always different ways to analyze a data set = subjective in how each person analyzes it -all contributing to less replicable science human error research replication in other domains - Answers -drug studies: 20-25% replicated -cancer treatment studies: 11% replicated -many others: chem, bio, physics, med etc solutions to incorrect research practices - Answers -establish best practices to avoid p-hacking (stats + methods) = reregistered reports, pre registration of hypotheses -revisiting established effects + support for replicating what we thought of as real things -developing better sense of what is likely to be a real effect (based on effect sizes) anchor points to compare to other effect sizes -"many labs" replication projects multiple labs conducting the same study + comparing results -open data, open code, methods = transparency what does the replication crisis illustrate - Answers -science is working as intended weaker effects, not real effect, not replicated = dropped over time = science is self correcting whats the problem with replication - Answers replication work isnt exciting or making an impact on science THUS not publishable -researchers are expected to find new things to push knowledge further academics wont get hired for replication work solution = CREP what are the 4 ways we know things - Answers 1) intuition: Gut responses based on what we know about the world (experiences)