Experimental Research Design
The outline, plan, or strategy used to answer the research question
2 purposes of "research design"
1. Control for unwanted variation
2. Suggests how data will be statistically analyzed
one-group pretest-posttest design
Design in which a treatment condition is interjected between a pretest and posttest of the DV
one-group posttest only design
Administration of a posttest to a single group of participants after they have been given an experimental treatment condition
Why does a "one-group posttest only design" not allow causal conclusions?
1. No pretest or control group
2. Almost all threats to internal validity apply
How is the "one-group pretest-posttest design" an improvement over the "one-group posttest only design"?
By adding a pretest to measure the DV before the treatment is introduced
>>allows to see a change!!
Which variables (i.e., THRIMMS) remain uncontrolled in a "one-group pretest-posttest design"?
posttest only design with nonequivalent groups
Design in which the performance of an experimental group is compared with that of a nonequivalent control group at the posttest
Which major threat to internal validity remains uncontrolled in "posttest only design with nonequivalent groups"?
NO RANDOM ASSIGNMENT!!
>>may confound selection w/ treatment effect
strong experimental research designs
-W/in participants = counterbalance & randomization
posttest-only control group design
Administration of a posttest to 2+ randomly assigned groups of participants that receive the different levels of the IV
How does a "posttest-only control group design" control for threats to internal validity (i.e., THRIMMS)?
By including a randomized control group
A control technique to equate groups of participants
>>accomplished by ensuring that every member has an equal chance of being assigned to any group
What are the weaknesses of the "posttest-only control-group design"?
-Does not guarantee equivalence of groups, particularly with SMALL SAMPLE SIZE
-No pretest to assess equivalence
pretest-posttest control-group design
Administration of a posttest to 2+ randomly assigned groups of participants after the groups have been pretested & administered the different levels of the IV
How does "pretest-posttest control-group design" control for threats to internal validity (i.e., THRIMMS)?
The internal validity of both designs are threatened only if one of the threats acts differently
>>i.e., if the effect of the threat occurs on one group but not the other
What are the advantages for including a pretest for the "pretest-posttest control-group design"?
1. Determine if ceiling effect (scores too high) has occurred
2. Allows use of analysis of covariance to statistically control for pretest differences
3. Allows researcher to asses the change in DV from pretest to post-test
4. Assess random assignment
5. Assess the effects of additional variables that may interact with IV
What is the major weakness of the "Pretest-post-test control-group design"?
May not GENERALIZE to situations with no pretest
How is threat to internal validity of "w/in participants study" ruled out?
Counterbalancing necessary to eliminate linear sequencing effects
>>vary the order
within participants post-test only design
All participants receive all conditions, and a post-test is administered after each condition is administered
2 advantages of "within participants post-test only design"
1. Increased sensitivity b/c effects of individual differences are controlled
2. Fewer research participants needed
2 disadvantages of "within participants post-test only design"
1. Difficult for participants
2. Potential problem for differential carryover effects
Which threat to internal validity do researchers automatically have to be concerned with when conducting a within-participants study?
A design that includes looking at ALL possible combos of 2+ IVs
A treatment combination of 2+ IVs
The effect of one IV
Mean of each row or column
When the effect of one IV depends on the level of another IV
In a factorial table, what indicates an interaction?
Indicated by different patterns among cell means
In a graph, what indicates an interaction?
When displayed graphically an interaction yields NON-PARALLEL LINES
mixed model factorial design
Uses a combination of w/in-participants & b/w participants
3 advantages of factorial designs
**1. 1+ IV allows for more precise hypotheses
>>can test for interaction
2. Control of extraneous variables by including as an IV
3. Ability to determine interactive effect of 2+ IVs
3 disadvantages of factorial designs
1. Requires more participants as # of IVs and cells decrease
2. Using 2+ IVs may be logistically cumbersome
>>present practical problems, which make it difficult to run study
3. Higher-order interactions are difficult to interpret