Chpt 8: Experimental Design (PSY302) Flashcards
Def: Confounding variable
a variable that varies along with the IV.
Confounding occurs when the effects of the IV and an uncontrolled variable are intertwined.
It cannot be determined which variable is responsible for the observed effect on the DV.
Good experimental design requires eliminating all possible confounding variables that could result in alternative explanations
When results can ________ be attributed to the IV, the experiment is said to have ____________ validity.
confidently, internal
Explain: Posttest Only Design
Obtain 2 equivalent groups of participants.
Manipulate the IV
Measure the effect of the IV on the DV
Procedures used have to eliminate any potential selection diffs: the ppl selected to be in the conditions cannot be different in any systematic way.
Def: selection diffs
systematic diffs between the participants assigned to each group - has to be avoided.
Def: Pretest-posttest design
a pretest is given before the experimental manipulation
A pretest is usually not necessary if participants are randomly assigned & the sample is sufficiently large.
Makes it possible to ascertain that the groups were equivalent at the beginning of the experiment.
Pros for using a Pretest-posttest design
The pretest enables the researcher to tell whether the experimental corpus were equivalent to begin with
Change can be incorporated into the analysis of group diffs & of individuals
The effects of attrition or mortality - the participants’ dropout factor in experiments - can be assessed.
Useful whenever there is a possibility that participants will drop out of the experiment which is most likely to occur over a long period of time.
Cons for using a Pretest-posttest design
It his time-consuming & awkward to administer
It can sensitize participants to what is benign studied, tho this can be mitigated by disguising the pretest.
Def: Solomon 4-group design
the experimental & control groups are studied with & without a pretest
Half the participants receive only the posttest & the other half of the participants receive only the pretest and the posttest
If there’s no impact of the pretest, the posttest scores will be the same in the 2 control groups & in the 2 experimental groups.
It’s also possible to assess the impact of the pretest _________ with the combo of both.
If one does it this way, participants receive only the __________ and the other half receive both.
directly, posttest
Def: independent groups design
participants are randomly assigned to the various conditions so that each participates in only a single group
Also known as between-subjects design bc comparisons are made between diff groups of participants
Def: random-assignment
the decision abt which group each participant is assigned to is completely random & not controlled by the researcher.
Prevents systematic biases
Yields equivalent groups
Removes participant diffs as an explanation for results
Def: Repeated measures design
all participants are in all conditions
Also called within-subjects design; comparisons are made within the same group of participants.
Pros for using a Repeated Measures design
Fewer participants needed
Extremely sensitive to statistical diffs
Cons for using a Repeated Measures Design
Order effect
Practice effect
Fatigue effect
Carryover effect
Def: Order effect
the order of presenting the treatments affects the DV.
Def: Practice effect
performance improves due to the practice gained from prev tasks
Def: Fatigue effect
performance deteriorates bc of fatigue, boredom, or distraction.
Def: Carryover effect
the effects of the 1st treatment carries over to influence the response to the 2nd treatment.
Def: Counterbalancing
a method of controlling for order effect by either including all orders of treatment presentation or randomly determining the order for each subject
Def: Latin square
a limited set of order constructed to ensure that 1) each condition appears at each ordinal position and 2) each condition precedes & follows each condition one time.
Researchers also need to ____________ determine the time interval between presentation of treatments & possible ___________ between them.
A rest period might counteract the _________ effect
If treatments involve administration of a drug or procedures that produce emotional changes, a long time ___________ is often necessary
carefully, activities, fatigue, interval
Some real-world situations are comparable to _____________ groups designs, while others more closely resemble repeated measures designs.
independent
Experiments that produce relatively permanent ___________ in ppl cannot use repeated measures designs.
changes
Def: matched Pairs design
participants are matched based on a participant variable; the matching variable will be either the DV or something that is sternly related to the DV
The members of each pair are then randomly assigned to the diff conditions of the experiment.
This ensures the groups are equivalent, which is particularly important with small sample sizes
Note that bc matching procedure can be costly & time-consuming, matched pairs is not a commonly used experimental design
What’s the 1st step to a Matched Paris design
First step is to get a measure of the matching variable from each of the participants.
Matched Pairs Design ensures that the groups are ___________ prior to into of the IV manipulation
equivalency
