1/57
Looks like no tags are added yet.
Name | Mastery | Learn | Test | Matching | Spaced | Call with Kai |
|---|
No analytics yet
Send a link to your students to track their progress
Independent Variable (IV)
Variable manipulated or controlled by the researcher (cause)
Dependent Variable (DV)
Variable measured to see the effect (outcome)
How to identify IV
Ask "What did they change?"
How to identify DV
Ask "What did they measure?"
Control Variables
Variables kept constant to prevent affecting results
Levels of IV
Specific variations of the independent variable
Manipulated Variable
Variable the researcher directly controls
Between-subjects design
Another name for independent-groups design
Independent-groups design
Different participants in each condition
Within-groups design
Same participants in all conditions
Matched-groups design
Participants matched on traits before assignment
Posttest-only design
Measures DV only after manipulation
Pretest/Posttest design
Measures DV before and after manipulation
Repeated-measures design
Participants measured after each condition
Concurrent-measures design
Participants exposed to all conditions at once
Double-blind study
Neither participants nor researchers know group assignments
Random Assignment Purpose
Makes groups similar at the start
Internal Validity
Ensures only IV caused change in DV
Random Selection
Choosing participants from population for generalization
Selection Effects
Groups differ at the start
Carryover Effects
Previous condition affects later condition
Practice Effects
Improvement from repeating a task
Demand Characteristics
Participants guess purpose and change behavior
Counterbalancing
Changing order of conditions to avoid order effects
Design Confound
Second variable that varies with IV
Order Effects
Results affected by order of conditions
Why use within-groups
Fewer participants and same people in all conditions
Within-groups threat
Order effects
Types of order effects
Practice, fatigue, carryover, testing
Missing element in one-group pretest/posttest
A comparison group
Attrition
Participants dropping out
Fix for attrition
Remove extreme pretest scores
Null Results Reasons
Weak manipulation or too much variability
Ceiling Effect
Scores cluster at high end (too easy)
Floor Effect
Scores cluster at low end (too hard)
Unsystematic Variability
Error variance or noise
Reduce Measurement Error
Use reliable tools and larger sample
Double-blind alternative
Masked design or blind observer
Placebo Study Purpose
Separate real effect from expectations
Situation Noise
External distractions affecting results
Increase Precision
Stronger manipulation, control, larger sample
Pretest/Posttest vs One-group
Includes comparison group vs none
Factorial Design
Uses 2 or more independent variables
Factor
Another name for IV in factorial design
Interaction Effect
Effect of one IV depends on another
Main Effect
Overall effect of one IV
Number of Main Effects
One for each IV
Interactions vs Main Effects
Interactions can exist without main effects
Number of Conditions
Multiply levels of IVs
Marginal Means
Averages across levels of other IV
Noise (factorial)
Variability within groups
External Validity
Generalization to real world
Purpose of External Validity
Ensure results apply beyond lab
Advantage of Factorial Designs
Can test interactions and multiple variables