6: Methodological Control in Experimental Research Flashcards
Why are between-subjects designs used
If the independent variable is a subject variable
used for studies where participating in one level makes it impossible to partake in another level (deception)
Prime advantages and disadvantages of between-subjects designs
Advantage: each subject enters the study fresh with a naïve perspective
Disadvantage: requires a lot of people
differences in conditions might be due to the independent variables but also might be due to differences between the individuals in different groups
Between subjects designs need
equivalent groups
How can equivalent groups be made
random assignment (including blocked random assignment) and matching
Describe matching
participants are grouped together on some subject variable such as their characteristic level of anxiety and then distributed randomly to the different groups in the experiment
What three conditions must be met for matching to take place
Must have good reason to believe the matching variable will have a predictable effect on the outcome of the study
Must believe it to be correlated with the dependent variable - figure this out from previous research
Must be a reasonable way of measuring or identifying participants on the matching variable
small number of subjects are available
Describe a within subjects design
Also called repeated-measures designs
Each participant is exposed to every level of the independent variable
Advantages of within subjects designs
Fewer participants are needed
Eliminates possibility that differences between levels of the IV could be due to individual differences
Issues with within subject designs
Ordered/sequence effects
- progressive effects
-carry-over effects
Describe progressive effects
performance changes steadily from trial to trial
Describe carry-over effects
Some sequences might produce effects different from those of other sequences
what is the typical way to control for order effects
to use more than one sequence: A strategy known as counterbalancing
Describe complete counterbalancing
(Used for testing once per condition)
Every possible sequence is used at least once
X!
X: number of conditions !: factorial
Describe partial counterbalancing
(Used for testing once per condition)
A subset of the total number of orders used
Take a random sample of orders from the complete set of possible orders, or by randomizing the order of conditions for each subject
Can use the Latin Square strategy
Important for the number of subjects to be equal to or a multiple of the number of rows in a square
Why might we test more than once per condition in a within subjects design
to get more reliable measures
Describe block randomization for testing more than once per condition
The basic rule is that every condition must occur once before any condition can be repeated
Within each block, the order of conditions is randomized
Eliminates being able to guess what is coming next (unlike in reverse counterbalancing)
Describe reverse counterbalancing
(testing more than once per condition)
Experimenter presents the conditions in one order and then presents them again in the reverse order
how to control for experimenter bias
double blind procedure
training and protocols to follow as to not lead participants to act a certain way
What is the Hawthorne Effect
when behaviour is affected by the knowledge that one is in an experiment and important to the studies success
want to be the ‘good subject’
evaluation apprehension - want to be evaluated positively and behave a certain way
what is the concern with demand characerisitcs
that they will create participant bias as they may reveal aspects of the study that reveal what the hypothesis is
how to control for participant bias
manipulation check
deception or placebo control group
field research