Research Skills Flashcards
Tests for Ordinal data (or that don’t meet requirements for ratio/interval) for both independent/related T-tests and ANOVA. NON-PARAMETRIC basically.
Mann-Whitney U
Wilcoxon Signed Ranks
Friedman’s ANOVA
Kruskal-Wallis
Threats to scientific validity
I. Internal Validity: how well the study is conducted. (history, attrition, selection/sampling, maturation, learning effect, regression)
II. External Validity: how well is the study applicable in “real-life” (outside of the lab) a.k.a ecological validity.
III. Statistical Validity: right statistical test is done, power is not limited, and effect size is accurate.
IV. Construct Validity: we are measuring the right thing in our experiment.
V. Expectancy Effects: interaction between experimenter and participant has consequences. (Hawthorne Effect: altering your behaviours because you’re beign observed)
What is protocol analysis ?
It is analysis of person’s thinking aloud.
What is discourse analysis?
Analysis of a conversation basically. Emphasies the contextual meaning of language, social constructionism epistemology.
Most popular quasi-experimental design
Nonequivalent control groups, participants not randomly assigned. Results can be both interpretable and uninterpretable
Mixed factorial design with one non-manipulated variable.
A type of nonequivalent control group where there is more than one IV.
Cross-sectional studies ; Longitudinal Studies ; Cross-sequential Studies.
Cross-sectional studies - study of people at different ages all at the same time (confound with Cohort Effects)
Longitudinal Studies - study of the same people over a long period of time (confound with Secular Trends - a change that is taking place in the general population over time)
Cross-sequential Studies - a mix of both longitudinal and cross-sectional studies - testing two or more cohorts at two or more times.
Changing criterion design, what is it
Introduces successively more stringent criteria for reinforcement to see if behaviours changes with the criteria. Like reading 1 page a day, then 2 pages a day the next time and so on…
