Exam 3 Flashcards
Why conduct studies with more than 2 groups?
can answer more questions with multiple group design
ANOVA
Analysis of Variance
Evaluates the influence of our IV on our DV
compare the ratio of between-group variability to within-group variability
To achieve a significant “error of variability” (F)
maximize between-group variability and minimize within-group variability
Factorial Design
Study with more than one Independent Variable
When a factorial study involves both between and within-subject variability we call it a
“Mixed” Factorial design
If all IVs have correlated groups then the design is called a
within-subject factorial design
If all IVs have independent groups then the design is called
between-subject factorial design
What is the main effect?
the effect of one IV in a factorial design
When thinking about the main effect, you can ask
“Is there a significant difference between the two (or more) levels of my variable of interest
Interactions
the joint effect of multiple IVs (where the effect of one depends on the other)
When thinking about interactions, you can ask
“Is there a difference between the two (or more) levels of my variable - the answer depends on the other variable
When there are more IVs you have
more interactions
How are single-case experimental designs different from case studies
Case studies are purely observational; single-case experimental designs involve the manipulation of a variable for a group of one
When might we use single-case experimental designs?
Sample of one
When you can assume generalizability
Rare population
Time-consuming or extensive research
Operational Definition
specific descriptions of the procedure
In a single-case experimental design, the study is ____ manipulated
actively
Why is baseline so important in single-case designs?
the critical comparison between the level of symptoms score reported or observing Phase A and the level of scores during Phase B
A-B-A Design
A (baseline) - B (posttest) - A (removal of IV)
If change in Dv is actually due to the IV, then there should be a reversal; more support for cause and effect
A-B-A-B
A (baseline) - B (posttest) - A (return to baseline) - B (posttest)
Quasi-Experimental designs are used when
we are interested in variables that you can’t assign people to (such as gender or age)
Ethical concerns; Social phenomena
Evaluating a program that was begun before you decided to examine it
Expense, time, or monitoring difficulties
Types of Quasi-Experimental Designs Include
Non-equivalent group designs
Interrupted time-series designs
In non-equivalent group designs, you create a
comparison group; not a control group since there is no randomization you cannot control for certain factors
Quasi-experimental designs tend to have problems with _____ validity
internal
In interrupted time-series designs
single groups are measured repeatedly expansion of A-B designs
