PSY309 CHP3
To evaluate how well a study supports a frequency claim, you need to focus on evaluating which of the following validities?
Construct validity and external validity
Which of the following validities is NOT examined in all research claims?
Internal validity
When examining the statistical validity of a frequency claim, one should look for
Margin of error estimate
What is an association/corellation claim How can it be distinguished
a variable is likely to be associated with another must have at least 2 variables variables measured not manipulated
What are conceptual variables/constructs? Provide an example
abstract concepts such as shyness/intelligence
What are the 3 types of claims FAC
association causal frequency
Why don't researchers attempt to achieve all 4 of the validities at once
based on the research and goals it sometimes may be impossible to achieve a specific validity also it may not be necessary
What is a type 2 error
conclude that there is no association when there may be one
What are the four types of validities
construct external statical internal
To evaluate how well a study supports an association claim, you need to focus on evaluating which of the following validities?
construct statistical external
To evaluate how well a study supports a causal claim, you need to focus on evaluating which of the following validities?
construct statistical external internal- very important
What are the 3 requirements for a causal claim? TIC
covariance- cause variable and outcome variable must be correlated-cannot have a 0 correlation temporal precedence-must show the causal variable came first internal validity/third variable rulet- must show no other explanations exist for the relationship/ change in a variable
Which variable is the measured variable
dependent variable
What is a frequency claim How can it be distinguished
describes a rate or degree of a variable, typically uses percentages or ratio freq. claims only focus on 1 variable variables are always measured, never manipulated 79% think re-gifting is acceptable
How is statistical validity assessed as it relates to frequency claims
extent to which a study's statistical conclusions are accurate
What is a type 1 error
false positive results show that there is an association but there is no association
How is construct validity assessed as it relates to frequency claims
how well a conceptual variable is operationalized/ how well researchers measure variable
How is external validity assessed as it relates to frequency claims
how well can the results be generalized to the rest of the population
What are causal claims?
one variable causes a change in the other 2 variables ex- music lessons enhance IQ
What are the three types of associations
positive association/correlation negative association/correlation zero correlation
What type of assignment can be used to increase internal validity
random assignment
What does internal validity assess
relationship between variables and determining whether a confounding variable is present
Why does the process of manipulating one variable and measuring another help with causal claims
temporal precedence- manipulating the causal variable ensures that it comes first ensure internal validity
What does external validity assess
the extent to which results from a study can generalize to a larger population
What does statistical validity assess
the strength of an effect and its statistical significance- the prob. that results could have been obtained by chance if there is no effect
What does it mean to operationalize variables?
turn a concept into a measured/manipulated variable example operationalize shyness as a set of structured questions used by a trained therapist to diagnose shy/not shy or weight gain- by weighing
What two errors can statistical validity identify in association claims
type 1 error type 2 error
What is a measured variable? Provide an example
variables whose levels are observed and recorded example-IQ, height gender
What kind of study can satisfy all 3 criteria for casual claims
well designed experiment where one variable is manipulated and the other is measured
What does construct validity assess
how well variables in a study are measured or manipulated
Which variable is the manipulated variable
independent variable
