Chapters 5 and 6 Flashcards
Independent Variable
variable that you manipulate to see effect on DV
Dependent Variable
outcome behavior; measured outcome of experiment
Extraneous Variables
Variables that can’t be controlled
Confounding variables
Extraneous variables that covaries with the independent variable that could provide alternative explanations for results
Manipulated variable
Researchers place participants into conditions
Subject variable
Participants are in condition due to pre-existing condition or self-selected into
Statistical conclusion validity
Extent to which a researcher uses stats properly and draws appropriate conclusions from statistical analysis
Validity
How will a test actually measures what is supposed to measure; accuracy
Construct validity
Adequacy of operational definition
External validity
Generalizability; degree to which research findings generate beyond specific context of experiment being conducted.
Ex) other populations
Internal validity
Methodologically sound; Able to show changes in independent variable causes changes in dependent variable
As Internal validity increases external validity decreases
As external validity increases internal validity decreases
True
Threats to internal validity
Any conn found such as history or maturation
History
Some event occurs between pre-and post test that causes large changes unrelated to treatment progression
Maturation
Producing developmental changes
Regression to the mean
Extreme scores by chance on first measurement move closer to the average when measured a second time
Testing & instrumentation
If pretest and posttest are the same pretest could affect post if you try a different measurement for a post test you may not be measuring the same thing
Selection
Different selective processes for experiment and control group
Selection by maturation
group matures at different times
Selection by history
One event happens to one group but not the other
Attrition
People who don’t come back; people who remain might be different than the larger group
Between subject design
Different participants in each condition
When to use between subject design
Subject variables. Being in one condition negates possibility of being in the other condition
How to create equivalent groups for between subject design
- Random equal chance assignment: participants have an equal chance of being placed in a condition
- Matching/random assignment: pre-test on variable Interest, organize scores in order, matched gets assigned number
Within subject design (repeated measures )
Same participants in each condition
Advantages of within subject design
Fewer participants equals greater power because same subject variables are repeated
Disadvantages of within subject design
- increases length of time participants in the study (fatigue)
- Order or sequence: once participant finishes one part of the study it could affect the other part of the study
Progressive effects
Order doesn’t matter
Ex) fatigue 
Carryover effects
Having condition a over condition b is different than having condition b then condition a
 Counterbalancing
Designed to control sequence effects; multiple people do the conditions in different orders
Complete counterbalance
All possible orders are given
Partial counterbalance
Each condition occurs in each position an equal amount of times
Reverse counterbalance
Condition given in one order than the reverse
Issues with counterbalancing
Hard to illuminate fatigue completely; assumes progressive effect and not order effect
Cross Sectional study
Between subjects design; compares people at different developmental points and measures each group once
Cohort effect
Provides alternate explanations for differences in between groups
Longitudinal study
Within subjects design; group measured multiple times over developmental period 
Cohort sequential design
One group measured in certain range another group measured where first group left off
experimenter bias
the experimenter might do something to sway the participant to act a certain way
Demand characteristic
if participant finds out hypothesis they may act a certain way to confirm it
Eval. apprehension
participants want to have good eval. so they attempt to behave in a way to get good eval
social desirability bias
people want to be socially acceptable
