Factorial Design
There are four conditions in this familiar design
What is a 2x2 factorial design
In this design, the researcher has 1) manipulated the IV, and 2) there are equivalent groups (random assignment)
What is a true experiment
The extent to which the results from my experiment can generalize to another group of people, another time, another place, etc.
What is external validity
When the response options cannot be ordered in any meaningful way. An example of this is gender.
What is the nominal level of measurement
When all members of the population are equally likely to be chosen as part of the sample.
What is random selection
A crossed design where IV1 has 3 levels and IV2 has 4 levels
What is a 3X4 factorial design
In this design, all variables are measured, there is no random assignment, and we can only talk about prediction (not causation)
What is a correaltional study design
The therapist in a study of therapy effectiveness on substance abuse where the experimental group undergo therapy with a general psychodynamic approach administered by Dr. Elektra, while the other group participated in therapy with a general cognitive approach administered by Dr. Ponder.
What is a confounding variable
The Cronbach's alpha level we need to say our measure had "good" reliability
What is alpha = .70
An example of this is asking participants to write a few sentences about their experience at the writing center.
What is an unstructured response format
An example is, in the swimsuit/sweater study, the marginal mean of math test performance is higher for the sweater condition than for the swimsuit condition (regardless of gender).
What is a main effect
An extra DV researchers insert towards the end of a study to help determine how well an experimental manipulation worked
What is a manipulation check
The researcher's definition of a variable at an abstract level.
What is the conceptual definition of a variable
When the responses on a scale are correlated between the first time respondents complete the scale and the second time respondents complete the scale.
What is test-retest relibility
When I don't think it matters if ask "how satisfied are you with your job?" before or after "how satisfied are you with your life?," I have forgotten about this.
What is the importance of question placement
Compare these to see if you have an interaction effect
What are the simple effects
When two variables are correlated but not because one causes the other, but because a some other variable causes both.
What is the third variable problem, a confounding variable, or systematic error
When an experimental study's results can be explained by something other than the IV causing the DV.
What is low internal validity
When a measured construct is not too strongly correlated with a related, but different, construct.
What is good discriminant validity
An example is, "Do you feel welcomed by staff or other youth at the center?"
What is a double-barreled question
The effect of one IV is different across levels of the other IV
What is an interaction effect (or interaction mantra)
An example is the participant's angry or happy behavior in the "suproxin" study, as recorded by research assistants looking on via a two-way mirror.
What is an observational or behavioral measure
A statistic we can calculate that gives us information about the strength and direction of a relationship between two variables.
What is the correlation coefficient, r
In order for a measure to be valid, it must also be this.
What is reliable
This is what I am going to do with the responses on my three questions assessing DV#1 when I get my data back.
What is calculate an average for each participant