PS295 Study Guide - Midterm Guide: Latin Square, Internal Validity, Demand Characteristics
Experimental Research
3 essential properties:
1) Varying the IV
-IV must have 2 or more levels which are the experimental conditions (vary quantitatively or
qualitatively)
-types include (1) environmental (2) instructional (3) invasive
2) Assignment:
-researcher must assign participants to IV conditions in a way that ensures their initial equivalence
-pilot tests ensure levels of IV are different enough to be detected
Simple random assignment:
-every participant as equal probability of being placed in any condition
Blocked simple random assignment:
-between subjects design, different participants experience different conditions
-participants randomly divided in blocks then randomly assigned to conditions
-keeps sample sizes equal and ensures equal number of participants in each condition
Matched random assignment:
-uses pretest and participants are ranked on relevant measure from lowest to highest and then
matched by being put into clusters/blocks of size k, highest matched w highest etc
Randomized groups design:
-between subjects design, different participants assigned to each condition
Repeated measures design:
-within subjects design, each participant measured more than once/serve in all conditions of
experiment
-eliminates need for random assignment/each participant serves as their own control
-more powerful, requires fewer participants but susceptible to order and carryover effects (effect
persists after condition ends/when new IV is introduced, results due to lingering effects)
Order effects:
Practice effects: performance improves bc DV is completed several times
Fatigue effects: becoming tired/bored as experiment progresses
Sensitization: beginning to realize what hypothesis is/responding differently after a few rounds
Counterbalancing:
-involves presenting levels of IV in different orders to different participants
-requires a lot of participants to complete every single possible order of events
Latin square design/incomplete counterbalancing:
-each condition appears once at each ordinal position (1st, 2nd, 3rd) and each condition precedes and
follows every other condition once
-first row follows the formula, 1, 2, N, 3, N-1, 4, N-2... where N is the number of conditions
Document Summary
Iv must have 2 or more levels which are the experimental conditions (vary quantitatively or: varying the iv qualitatively) Types include (1) environmental (2) instructional (3) invasive: assignment: Researcher must assign participants to iv conditions in a way that ensures their initial equivalence. Pilot tests ensure levels of iv are different enough to be detected. Every participant as equal probability of being placed in any condition. Between subjects design, different participants experience different conditions. Participants randomly divided in blocks then randomly assigned to conditions. Keeps sample sizes equal and ensures equal number of participants in each condition. Uses pretest and participants are ranked on relevant measure from lowest to highest and then matched by being put into clusters/blocks of size k, highest matched w highest etc. Between subjects design, different participants assigned to each condition. Within subjects design, each participant measured more than once/serve in all conditions of experiment. Eliminates need for random assignment/each participant serves as their own control.
