1/213
Looks like no tags are added yet.
Question Wording
-if you can use someone else's calendar use that
-keep it neutral
-use language of research participants
-avoid unnecessary negatives
Use language of research participants
keep it applicable and familiar for research participants
Avoid unnecessary negatives
"do you agree or disagree that people who cannot read well should not be allowed to teach in public schools
Avoid double-barreled questions
"how well do you get along with your managers and coworkers"
Avoid leading and loaded questions
"most students think that virtual learning isnt working. Do you agree"
"Too little money is being spent on the poor v on welfare"
Levels of Measurement
Nominal, Ordinal, Interval, Ratio (NOIR)
Nominal measurement
a measure for which different scores represent different, but not ordered, categories (Categorical)
Ordinal measurement
a measure for which the scores represent ordered categories that are not necessarily equidistant from each other (Specific order)
Interval Measurement
a measure for which a one-unit difference in scores is the same throughout the range of the measure (no zero and evenly dispersed)
Ratio Measurement
a measure for which the scores possess the full mathematical properties of the numbers assigned (there is a zero)
Why do we care?
-Information content
-statistical tests
-ecological validity
Statistical tests
type of measurement tells us what stat test
Ecological validity
make it real world but observe phenomena
Response formats
-comparative scaling
-itemized
-graphic
-numerical
Comparative scaling
present participants with a set of stimuli and ask to compare
-salary vs promotion
-salary vs flexible
-flexible vs promotion
-Method of rank order
Itemized rating scales
-multiple choice questions with specific responses
-demographics
Graphic Rating Scales
-Visual analogue scales
-people indicate their responses periodically
-gender and major
-seen with children
Numerical Rating Scales
-Participants assign numerical values to their responses
anchors (numerical)
verbal labels used as anchors
Multi-Item Scales
-composed to 2 or more items in a rating scale format, each of which is designed to assess the same variable
-Likert
-Thurstone
-Guttman
-Semantic Differential
Likert Scales
-Set of statements about a person, thing, or construct
-agree/disagree with statements in a nominal scale
Misuse of a Likert scale
-write large # of items
-high/moderate/low
-large # of people
-Call it "numerical rating scale"
Thurstone Scale
-Attitude measurement
Process
-Create a large # of items
-20-30 judges into 11 groups
Final items
-must rep entire range of attitudes
-low tolerance
Participants
-view items
-check off those they agree with
Scoring
-means of favorability
Guttman Scales
-participants are presented with a set of ordered attitude items designed so that a participant will agree with all items after that point (and disagree with all items after that)
Semantic Differential
-Bipolar adjective points
-interpersonal impression
-nonabrasive ------abrasive
Response Biases
-Respond to items for reasons other than the responses being a reflection of the construct or behaviors being assessed by conflict in this item
-question bias
-person bias
Question Bias
-scale ambiguity
how much beer?
-Category anchoring
one beer, two beer,,,,
seeing what is appropriate
Estimation Biases
what days did you drink in the last 30 days
Person-Related Biases
-Social desirability
-Acquiescence: Agreeable
-Extremity
-Halo and leniency
Halo more positive attitude, leniency similar
Questionnaire design
-Demographic items location
-will order of items/survey affect results
-use of preexisting measures
-differences between scales
Other tips
-make it easy to follow
-bold/underline things
-response options easy to review
-check how survey looks when take it using a phone
-show survey progress bar
-syntax
Population
-Target population
-Study population
-Research Sample
Target population
everyone results apply too (all college students)
Study population
Everyone who meets operational definition and criteria
Research Sample
Who you get out of study that you collect
Sampling
-Probability
-nonprobability
-Simple
-stratified
Probability sampling
Every member of the study population has a known probability of being selected for the research sample
-not used as often because you do not know full frame