L3 - Variance, effects and experiment design Flashcards
(21 cards)
What is construct validity?
How accurately your measures asses what they are supposed to (you would have “poor” or “good” construct validity).
What is treatment variance?
Variance from the mean for each of our different conditions.
What is a confounding variable?
An extra variable that can (like the IV) have an effect on the DV and therefore mask or skew the results.
What is error variance?
Difference in results due to other variables/factors influencing the participants that the researcher does not examine (caused by confounding and extraneous variables).
what is the ‘Ceiling effect’?
The task is too easy meaning that all participants results are clustered around the “best” possible score.
what is ‘floor effect’?
The task is too hard meaning all the participants results are clustered around the “worst” possible score.
What are order effects?
When the order in which the task is presented to the participant affects the results.
What are history effects?
Events that happen in the environment that change the conditions of a study, affecting its outcome.
What are maturation effects?
Any biological or psychological process within an individual that systematically varies over time, independent of specific external events e.g. hunger, aging, mood.
What are testing effects?
When scores on the post-test are influenced by simple exposure to the pre-test (when prior observation affects later observation).
What are practice effects?
When the repetition of a task means the participant improves due to this and can then effect results.
What are fatigue effects?
When participants get bored/tired during an experiment which impacts the effort they put in and can subsequently impact the results.
What is a pretest-posttest design?
Where the participants are studied before the the IV is manipulated and again observed after the intervention.
What are demand characteristics?
Cues that might indicate to a participant the true purpose of the study.
What are 2 of the ways you can rule out/minimise order effects?
Counterbalancing and the Latin-square design.
What is counterbalancing?
When you have half of the participants complete a task in one order and the other half in a different order.
What is the Latin-square design?
When you have 4 different orders participants complete the tasks in, this is determined by a 4x4 column and row data set.
What is a matched design?
Assigning participants to different groups within the experiment based on specific variables that are significant to the test e.g. age, gender etc.
What is a passive control group?
The group of participants that don’t partake in anything or are made to do meaningless tasks.
What is an active control group?
A group of participants who could reasonably assume they’re doing something relevant to the experiment.
What is attrition?
the dropout rate of an experiment.
