Term
| features of between subjects design |
|
Definition
-each level of IV contains a different subject group
-equivalent groups
-necessary when the IV is a subject variable OR Ss need to be naive
eg eyewitness memory study |
|
|
Term
| advantages of between subject design |
|
Definition
| subjects remain naive to conditions |
|
|
Term
| disadvantages of between subjects design |
|
Definition
-more Ss needed
-expensive
-differences could be due to the IV, chance, or group differences |
|
|
Term
| disadvantages of between subjects design |
|
Definition
-more Ss needed
-expensive
-differences could be due to the IV, chance, or group differences |
|
|
Term
| methods for creating equivalent groups (3) |
|
Definition
| 1. random assignment 2. block randomization (ensures each condition of the study has a subject randomly assigned to it before any condition has a subject assigned to it again) 3. matching |
|
|
Term
| random assignment features |
|
Definition
-equal chance of being placed in any group of a study
-equally spreads out potential confounds |
|
|
Term
| within subjects design block randomization features** |
|
Definition
-ensures each condition is expressed before repeating -each block contains N groups, order in each block randomized, then assign Ps randomly to each block -eg CBA, BCA, ABC, ACB
(equal numbers in each group) |
|
|
Term
|
Definition
-deliberate control over potential confound
-Ps grouped together on a trait, then distributed randomly to different conditions
USED ONLY WHEN
-small number of S
-matching V correlates with DV (predictable effect on study outcome)
-measuring matching V is feasible |
|
|
Term
| features of within subjects (repeated measures) design |
|
Definition
| -each participant is exposed to every level of IV |
|
|
Term
| advantages of within subjects design |
|
Definition
-fewer Ps required (good for small pops)
-eliminates individual differences from analysis (effect is only due to chance + systematic diffs) |
|
|
Term
| disadvantages of within subjects design |
|
Definition
| -sequence effects (experience influences performance) |
|
|
Term
| types of sequence effects (2) |
|
Definition
1. progressive effect
2. carryover effect |
|
|
Term
|
Definition
performance changes steadily from trial to trial
-practice/fatigue effect |
|
|
Term
| def: carryover effect + eg |
|
Definition
systematic changes in performance occur as a result of performing one sequence of conditions rather than a different sequence
-eg hard task --> easy task = frustration, decreased performance
easy task --> hard task = confidence, increased performance |
|
|
Term
| types of counterbalancing if testing once per condition (2) |
|
Definition
1. complete counterbalancing
2. partial counterbalancing |
|
|
Term
| complete counterbalancing |
|
Definition
-every possible sequence used at least once (N!),
eg (AB)(BA) |
|
|
Term
|
Definition
-random sample from all possible sequences
eg latin square
-every condition of the study occurs equally often in every sequential position
-each condition precedes and follows every other condition only once |
|
|
Term
| types of counterbalancing if testing more than once per condition (2) |
|
Definition
1. reverse counterbalancing
2. block randomization |
|
|
Term
|
Definition
-within subjects - present conditions in one order, and then again in reverse order
eg ABBA |
|
|
Term
| block randomization in within subjs |
|
Definition
| - assignment technique when Ps are tested in each condition more than once, they experience each condition once before experiencing any condition again |
|
|
Term
| cross sectional design adv/disadv |
|
Definition
+ less time
- cohort effects (environments in which they were raised) |
|
|
Term
| longitudinal design adv/disadv |
|
Definition
+ no cohort effects
- attrition |
|
|
Term
| cohort sequential design features |
|
Definition
-combines longitudinal and cross sectional
-group selected and retested, new cohorts selected and retested every few years as well |
|
|
Term
| experimenter bias + how to control |
|
Definition
-experimenter expectations influence subjects beh
-control by: automating, double blind |
|
|
Term
| participant bias + how to control |
|
Definition
-demand characteristics - cues give away the purpose of the study, Ps try to be "good subjects"
-control by: effective deception (placebo)
-manipulation checks
-field research |
|
|
