SSR Exam 2 Flashcards
confirmation bias
searching for confirmation of existing ideas
hindsight bias
after seeing the results, you convince yourself that you had predicted this in advance
perseverance bias
disproved ideas still affect our reasoning
motivated skepticism
looking more critically at unexpected results than at finding predicted results
bias blind spot
we think of ourselves that we are not bias and we think that other people are way more bias than we are
direct replication
same experiment, different sample. good for filtering out random findings
conceptual replication
different experiment, different sample, same phenomenon (concept)
paradigma-driven
expand immediate replication with targeted procedural changes
texas sharpshooter fallacy /
draw target after you shoot / normative neglect
polynomial contrast
this tests for trends in the data and in its most basic form it looks for a linear trend
orthogenal
when contrasts are independent
quadratic trend
where there is a curve in the line
cubic trend
where there are two changes in the direction of the trend
quartic trend
this trend has three changers in direction (you need at least 5 categories of the independent variable)
post hoc tests consist of
pairwise comparisons that are designed to compare all different combinations of the treatment groups
assumption of homogeneity of regression slopes
means the the relationship between the covariate and outcome variable should be similar at different levels of the predictor variables
assumption of sphericity
is about assuming that the relationship between scores in pairs of treatment conditions is similar (the level of dependence between means is roughly equal)
commpound symmetry
which holds true when both the variances across conditions are equal and the covariances between pairs of conditions are equal
mauchly’s test
assesses the hypothesis that the variances of the differences between conditions are equal
lower bound estimate of sphericity
in which if there are five conditions the lower limit of green house geisser estimate will be 0.25
mixed design
when a design includes some independent variable that were measured using different entities and others that used repeated measures§
reference class/collective
the hypothetical infinite set of tosses (or events more generally)
alpha is
the probability (long-run relative frequency) of a type 1 error when the null hypothesis is true
sensitivity can be determined in 3 ways
power, confidence intervals and finding an effect significantly different form another reference one