Vocab CH 4 Flashcards
Three horned dilemma
Three kinds: precision, situation generalizability, people generalizability. The dilemma lies in the fact that any attempt to maximize one of these qualities through a particular design will result in the reduction of the other two.
Dilemmatic approach
viewing it as a series of negotiations and interlocking choices among multiple research designs
Four designs most commonly used by applied social psychologists:
- true experiments
- correlational studies
- quasi-experiments
- surveys
True experiment
• has two components that allow it to test a causal relationship between two or more variables
o manipulation
o random assignment
Manipulation
occurs when the experimenter systematically varies the level of one or more variables holding constant other variables that may have an effect
Independent variable
manipulated variables
Dependent variable
outcome variable
Random assignment
assigning groups on an arbitrary basis
Confounds
• an additional variable that systematically varies with the independent variable
- person confound
- procedural confounds
Person confound
when the presence of an individual difference influences the outcome variable
Procedural confound
when the experimenter unwillingly varies two or more variables at once
Noise
refers to variables that influence the dependent variable, but unlike confounds, they are evenly distributed across conditions
Interactions
indicate that the effect than one or more independent variables have on the dependent variable depends on an additional independent variable
2 x 2 factorial design
the investigator is interests in studying the independent and combined effects of two independent variables each with two levels
Boundary conditions
conditions that demarcate when a theory does or does not apply
Mundane realism
that the physical setting of the study is similar to the real-world setting in which the phenomenon occurs
Experimental/Psychological realism
maximized not by asking id a study ‘looks’ realistic on observable dimensions, but rather by asking if it “feels: realistic on psychological dimensions
Correlational research
investigates the relationship between different measured variables, typically with an interest in determining how these variables interrelate in naturally occurring situations
The correlation coefficient
o ranges from -1 to 1
o absolute numerical value indicating magnitude of relation between the variables
A positive correlation
indicates that as the value of one variables increases, so does the other
A negative correlation
indicates that as the value of one variable increases, the other variable decreases
issue of reversed causality
there are multiple ways to interpret a significant correlation between two variables. The direction of the relationship is not clear.
third variable problem
additional variable is responsible for the observed relation
covariates
allow experimenters statistically to remove the influence of variables
longitudinal designs
the variables of interest are measured at two or more time period and then the correlations between these variables are analyzed between and across time
Quasi-experimental design
researcher only has partial control over the IV, because random assignment is impossible: participants are assigned to groups based on some other naturally occurring criterion
Person-by-treatment quasi-experiment
one of the Independent variables is manipulates ( treatment variables) and one is measured ( person variables)
Survey research
the process of collecting information from a sample of people who have been systematically selected to represent a larger population
Random sampling
equally likely chance of being selected
Cluster sampling
alternative to random sampling
Sampling error
the likely discrepancy between the results obtained in a representative sample and the results one would have obtained if everyone in the population of interest had been studied
