Midterm 2 Review Flashcards
Define and explain the differences between construct validity, internal validity, and external validity
Construct validity: the extent to which the measurement or manipulation of a variable accurately represents the theoretical variable (construct being studied)
Internal validity: the accuracy of conclusions drawn about cause and effect
External validity: the extent to which a study’s findings can accurately be generalized to other populations and settings
Explain what it means to operationalize a variable
To operationalize a variable, you are defining it terms of the specific method used to measure or manipulate it
Ex. Bowling skill: a person’s avg. bowling score over the past 20 games
Describe positive linear, negative linear, curvilinear, and no relationship patterns between variables, using examples
+: as x goes up y goes up
-: as y goes down x goes down
curvilinear: there’s a curve
no relationship: no relationship
What is the purpose of a correlation coefficient? What info does it contain?
A correlation coefficiant shows how strongly variables are related to one another
Explain the key difference between no experimental and experimental research methods,
No experimental: observations
Experimental: direct manipulation and control of variables
In an experiment, what is the purpose of randomization?
Randomization ensures extraneous variables are just as likely to affect one group as another
Describe a possible experiment identifying the independent and dependent variables
Experiment: study whether exercise reduces anxiety
Independent variable: one group being manipulated (have them exercise everyday for week)
Dependent variable: one group not doing anything
measure level of anxiety at the end of the week
What are the strengths are weaknesses of laboratory experiments, compared to field expierments?
Lab
+ easily able to manipulate variables
+strong internal validity
- can limit external validity
Field
+ independent variable is investigate in natural setting
- loss of ability to directly control many aspects
Explain the idea of an “ex post facto” study, distuingishing it from a true experiment
In an ‘ex post facto’ study, groups are formed on the basis of actual differences rather than through randomization
a true experiment has completely random groups
Discuss the idea of a ‘true score’ and measurement error. Use words “variability”, “validity” and reliability” to help explain
A “true score” refers to the theoretical, perfect underlying value of a variable that would be obtained if a measurement could be taken without any error, representing the true underlying characteristic of an individual or phenomenon, while “measurement error” is the difference between the observed score (what is measured) and the true score, essentially the degree of inaccuracy in a measurement due to various factors that can introduce variability into the data, impacting the reliability and validity of the measurement.
What is reliability? Give examples of low and high reliability.
Reliability refers to the consistency or stability of a measure
low reliability: range of scores is big
high reliability: range is small
Explain how to read a Pearson product- moment coefficiant, and explain what this number can mean in the context of reliability
can range from 0.00 to 1.00 (+ or -) the stronger the relationship
0.00 tells us there’s no relationship
+/- tells the direction of the relaitonship
How do you conduct tests of test-retest alt forms, or split-half reliability? What would the results tell you?
Test-retest is done by measuring the same individual at the 2 points in time
Alt forms is done by administering 2 different forms of the same test to the same individual at 2 points in time
Split-Half is the correlation of the total score on the other half 2 halves are created by randomly dividing the items in 2 parts
What is Chronbach’s alpha? How is it calculated and what does it mean?
Chronbach’s alpha is a reliability indicator based on internal consistency; provides avg. of all possible split-half coefficiants
- To calculate scores on each item are correlated with scores on every other item; large # of correlation coefficients are produced
Define interrator reliability and offer an example?
Interrator reliability is the extent to which raters agree in their observations
What is construct validity?
Construct validity refers to the adequacy of the operational def of variables
- to what extent does the o.d. reflect the true theoretical meaning of the variable
Define and give examples of face validity, content validity, predictive validity, concurrent validity, and discriminant validity
Discriminant validity: when measure is not related to variables with which it should be
Face validity- the evidence for validity is that the measure seems to measure what it’s supposed to
content validity- based on comparing the content of the measure with the universe of content that defines the construct ex. measure of depression would have content that links to each symptom that defines the depression construct
Predictive vvalidity- research shows scores do predict behavior it is intended to predict
concurrent validity- research examines relationship between measurement and criterion behavior at the same true
What does it mean for a measure to be reactive? How can reactivity be reduced?
A measure is reactive if awareness of being measured changes an individual behavior
can be reduced by allowing time for individuals to become used to the observer’s presence or recording equiment
