Variables & Analyses
A researcher wants to know if studying causes confidence, which then causes higher exam scores. What kind of variable is studying?
Independent Variable
The type of sampling where every member of the population does not have an equal chance of being selected.
Nonprobability Sampling
Methods that aim to test association claims by measuring all variables.
Correlational Methods
The variable that is used to put participants into pairs in a paired subjects design.
Matching Variable
The type of third variable that contextualizes the effect of the independent variable on the dependent variable.
Moderator
This kind of variable is held constant across conditions to rule it out as a potential confound.
Control Variable
A possible solution to questionable research practices where planned hypotheses, design, and analyses are documented before starting the research.
Preregistration
A type of qualitative interview where questions are read word for word with no follow-up, requiring the least experience and professional judgement.
Structured Interview
The between-subjects experimental design where the dependent variable is measured only once, after the manipulation of the independent variable.
Posttest-Only
If a factorial experiment has two factors, one with two levels and one with three levels, what is the factorial notation?
2x3
A researcher conducts a study with one independent variable with three conditions and a between-subjects design. What statistical analysis should they use?
One-way ANOVA
This is the probability of detecting a significant effect in your sample if there is truly an effect in the population.
Statistical Power
A type of qualitative coding where the researcher creates themes by reading through all of the data and identifying subjects that come up repeatedly.
Data-Driven Coding
The method of putting participants in conditions in an experiment that turns systematic variability into unsystematic variability.
Random Assignment
This type of mediation (indicated by a non-significant pathway c') means that even when we remove the effect of the mediator, the independent variable still causes the dependent variable - but the effect is weaker.
All participants in a study watch both a happy video and a sad video. They rate their mood after each video on a scale from 0-100. What statistical analysis should be used to compare the mood after the happy video to the mood after the sad video?
Dependent/Paired Samples t-Test
A probability sampling technique where the researcher selects specific demographic categories and then randomly selects individuals from each category.
Stratified Random Sampling
A threat to internal validity in quasi-experimental designs where an external event occurs for everyone in the study at the same time as the manipulation.
History Threat
Paired subjects designs use the same statistical analyses as this type of design.
Within-Subjects Designs
This pathway tests the effect of the mediator on the dependent variable when the independent variable is controlled.
Repeated Measures ANOVA
A questionable research practice where researchers analyze their data many different ways to look for significant effects, increasing the odds of any significance being a fluke.
p-Hacking
A type of quasi-experimental design that compares several measurements of a group before and several after the manipulation.
Time-Series Design
This major limitation of between-subjects designs is reduced, but not eliminated, in paired subjects designs.
Nonequivalent Groups Problem
This kind of effect is present when the effect of one factor is especially strong at a certain level of the other factor.