Term
| key to increasing reliability |
|
Definition
- providing consistent measures in comparable situations
- when two respondents are in the same situation, they should answer the same question in the same way
|
|
|
Term
| Good question properties: increasing reliability |
|
Definition
- research side of question and answer process is entirely scripted so questions as written fully prepare a respondent to answer questions
- questions mean the same thing to every respondent
- kinds of answers that constitute an appropriat response to the question are communicated consistently to all respondents
|
|
|
Term
| define double barreled question |
|
Definition
| ask two questions at once (better to ask two separate questions) |
|
|
Term
|
Definition
- avoid double barreled questions
- avoid implicit negatives
- ex) restrict, forbid, restrain, outlow, oppose
- ask questions with hidden questions
- questions that tend to suggest an answer
- you MUST use neutral terms, phrases
- avoid loaded terms (ex: fair, honest)
- ambiguous concepts, abbreviations
|
|
|
Term
| Effects of framing a question |
|
Definition
- different, but logically equivalent wods or phrases (ex: 95% employment or 5% unemployment)
- occurs when frames cast same critical information in positive or negative light can cause individuals to have different preferences
|
|
|
Term
| measurement is carried out in what ways |
|
Definition
- nominal
- ordinal
- interval data
- ratio data
|
|
|
Term
| types of question: based on responses |
|
Definition
- closed- a list acceptable responses is provided to the respondent
- open- acceptable responses are not exactly provided to the respondent
|
|
|
Term
|
Definition
- yes/no/dont know
- MC- must include all possible categories
- likert scales
- ex: good, very good, etc.
- good attitude questions
- open ended
- don't know- when should you include this
|
|
|
Term
| why responses may not be accurate (aka why it may not be valid) |
|
Definition
- they do not understand the question
- they dont know the answer
- they cannot recall it, although they do know it
- they do not want to report the answer in the interview context
|
|
|
Term
| reasons for invalide responses: lack of knowledge and how to avoid |
|
Definition
- ID and interview household member best informed
- use data collection procedures to permit the respondent to consult with other househould members
- eliminate proxy respondents
|
|
|
Term
| what can make invalid: social desirability and how to avoid |
|
Definition
- minimize sense of judgement
- maximize importance of accuracy
- use self administered data collection
- assure confidentiality and anonymity
|
|
|
Term
| increasing validity of subjective questions |
|
Definition
- better to have more categories than fewer
- ask multiple questions with different question forms that measure the same subjective state
|
|
|
Term
| different types of validity |
|
Definition
- face validity- cursory review of survey itens by untrained judges (casual, not soft)
- content validity- subjective measure of how appropriate the items seem to a set of reviewers who have some knowledge of the subject matter (still very qualitiative)
- criterion related validity- measure of how well one instrument stacks up against another instrument or predictor (more concrete, and instrument usually put against gold standard measure)
- can do concurrently or predictively
- construct validity- most valuable and most difficult measure of validity
- measure of meaningfulness of scale or instrument when in practical use
|
|
|
Term
|
Definition
- convergent- implies several different methods for obtaining the same information about a given trait or concept produce similar results
- divergent/discriminant- ability of measure to estimate the underlying truth in a given area
- must be shown not to correltate to closely with different but similar concepts
|
|
|
