open science Flashcards
(14 cards)
what was the A study : methods
N = 100
Task with 20 minute duration
36 trials :
Difference types of image : positive, negative, romantic, erotic
PPTS asked to click on curtain when they feel it has a picture behind it
what was the results for the A study?
Hit rate significantly above change for erotic images
53.1%, t(99) = 2.51, p= .01
Hit not significantly different from change for other image types
what is statistical power?
probability of seeing a true positive
what is alpha?
the highest acceptable risk of a false positive (typically 5%)
what is publication bias?
this is where researchers biased towards results which support their theories
what is the file drawer problem?
this is where non significant results are not published and it is harder to publish non significant replications
what is harking?
hypothesising after results are known
what are the 3 ways to analyse a given data set?
different statistical test
different variables
different rules for excluding outlets
what is p hacking?
a way to cheat with stats by performing the analysis in a different way to get p < .05 only reporting sig result so there’s false positives
what is a multiverse analysis?
run many possible analysis and see how many get a significant result
how to solve the crisis?
transparency :
- open materials
- open data
- preregistered
what is open materials?
Share the materials eg
instructions, program and stimuli so others can replicate
what is open data?
share the raw data so other researchers can perform analysis and see how other variables affect the results
what is pre registration
plan the study in advance including
- materials
- planned analyses
prevents p hacking and harking
