Psych 230 CHPT 3

variable

needs 2+ values

constant

something that can vary but only has 1 level in the study

ex: everyone being canadian in a study

measured variable

levels are simply observed and recorded

manipulated variable

researcher controls this

• ex: researcher gives some participants 10mg and the other 20mg

• some variables cannot be manipulated; age, IQ, and else is all measured even IF you randomly assign based on those factors

constructs/conceptual variables

A variable of interest, stated at an abstract, or conversational, level

ex: satisfaction with life

operational definitions OR operational variables

the WAY a concept is measured/manipulated in a study

operationalize

to turn a conceptual definition of a variable into a specific measured variable or a manipulated variable

• ex: research design

conceptual defintion

actual theoretical definition of a construct (article titles)

claim

argument someone is trying to make

frequency claims

always % or fraction of a single variable

association claims

one level of a variable is associated with another variable

correlational study

testing two variables and how they connect together

causal claims

just notes a relationship between two variables and argues that one variable is responsible for changing the other or ISN’T responsible for changing the other

• use more exciting verbs than boring ones

the phrase “is more/less likely to” an association claim verb or causal claim verb?

association

validity

appropriateness of a conclusion/decision

must be reasonable, accurate, and justifiable

construct validity

how well a conceptual variable is operationalized (or measured/manipulated)

• to establish, researchers must measure each variable reliably and see if they have consistent scores on repeated tests

generalizability

how well settings in the study represent other settings in other contexts

external validity

how well results represent individuals OUTSIDE of the study

statistical validity/statistical conclusion

extent to which a study’s numerical estimates are reasonable, precise, and replicable

point estimate

1 estimate of a population value

• ex: 39% of teenagers went driving

confidence interval

interval w/ higher and lower range to capture the population value for a point estimate

• ie, gives a range of values (home repairs can cost bt/wn 10k to 12k)

margin of error of the estimate

statistic providing range of values that has a high probability of containing true population value

internal validity

study’s method ensures there are no other alternative variables for change in B

• 3rd criteria for establishing causal claim

covariance

extent to which two variables are observed to go together

• ex: study’s results show that A is associated w/ B

• 1st criteria to establish causation

temporal precedence

study’s method ENSURES A comes before B

• 2nd criteria to establish causation

independent variable

manipulated variable

dependent variable

measured variable