Ch 4 Flashcards
Validity: (Is it true?) 4 types
- construct validity: does this measurement accurately represent the variable?
- internal validity: Can we draw an accurate conclusion about cause and effect?
- external validity: are these findings accurate to other populations and settings?
- statistical validity: is the conclusion we drew from analyzing our study accurate?
construct validity
does this measurement accurately represent the variable?
internal validity
Can we draw an accurate conclusion about cause and effect?
ext. validity
are these findings accurate to other populations and settings?
statistical validity
is the conclusion we drew from analyzing our study accurate?
Variables
- variable= it varies (on two or more levels)
operational defintion
how will you measure or manipulate the variable (is our operational definition good?)
construct validity=
how well does my operational def and measurement match reality
Relationships between variables
positive relationship:
- if a goes up then b goes up
negative relationship
- if a goes up then b goes down
curvilear relationship
- a and b both go up (to a certain point) then b goes down
- depicted by anything other than a straight line
No relationship
- no connection between a and b
- correlation coefficient:
a numerical index of the strength of relationship between 2 variables
- the closer a numer is to 1 (+,-) the stronger the relationship, and the closer a numer is to 0 the weaker a relationship
Reduction of Uncertainty
Uncertainty= random variablity
- people and events are not completely predictable
- we can reduce uncertainty if we can identify related variables
Nonexperiemental vs. experimental methods
- nonexperimental method: observe variables of interest as they occur naturally
- experimental methods: directly manipulate and control one variable and then observe the response on another variable
Nonexperimental method
- correlation method is a very common form
- third variable 6 (extraneous variable’)= any variable that isn’t one of the two you are trying to study the relationship between
- confounding variable= an extraneous variable that “moves with” one of the variables of interest
Experimental method
– dependent variable
- experimental control
- - independent variable
- field experiment
- -internal validity
- external validity
- randomization
same variables cannot/should not be manipulated
- participant variables can only be measured, not manipulated
experimental control
holding all extraneous variables constant