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
What is informed consent
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
No more risk than what someone could reasonably be expected to experience in their day-to-day lives
What is minimal risk
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
What is an operational definition
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
The principles of respect, beneficence, and justice.
What are the Belmont Principles
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
This is what we tell participants our study is about in an attempt to camouflage the true purpose of our study.
What is a cover story
The effect of one IV is different across levels of the other IV
What is an interaction effect (or interaction mantra)
An example is whether a kid takes one piece of candy or a handful in the self-awareness (halloween) study.
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
When a researchers has no information about the identity of participants, no contact information, and has never seen them in person.
What is an anonymous study