Conceptualization
 | Translates abstract theory into testable
hypothesis |
| Reviewing definitions to formulate your
own definition |
Multidimensional
Concepts
| Bringing together several related concepts
|
Operationalization
| measuring the concepts |
Binge Drinking
(Wechsler et al., 1998)
The consumption of at
least five drinks in a row for men or four drinks in a row for women during the 2 weeks
before the students completed the questionnaire.
Forms of
Questions
| Single Questions |
| Question Sets |
| Scales and Indexes |
Open-ended Survey
Question
| Respondents answer in their own words |
| Useful when a topic has never been studied
before |
Closed-ended
Questions
| Easy to process with computers |
| Amenable to statistical analysis |
| Likely to reduce ambiguity |
| Risk obscuring what people really think |
| May not be exhaustive |
Idiosyncratic
variation
| Occurs when respondents answer questions
on the basis of individual reactions to particular words or ideas in the questions |
Index
| Useful in research because it combines
answers to several questions |
| Reducing idiosyncratic variation |
Unobtrusive
Measures, Eugene Webb
| physical trace evidence |
| archives |
| simple observation |
| contrived observation (hidden recording) |
Triangulation
| Use of more than one operation to measure
a variable
 | interviewing people |
| recording observations at the time of the
interview |
| having the people fill out a survey |
|
Nominal Level
| No mathematical interpretation |
| Variables must be mutually exclusive and
exhaustive |
| ex. occupation |
Quantitative
Levels
| Ordinal level |
| Interval level |
| Ratio level |
Ordinal Level
| Specify order
| less than and greater than |
| ex. occupational prestige |
|
Interval Level
| Fixed measurement units with no absolute
zero point |
| ex. Temperature |
Ratio Level
| Fixed measuring units and an absolute zero |
| ex. years (age) |
Dichotomies
| variable having only two values |
| male and female |
Measurement
Validity
| Face
| obviously pertains to the meaning of a
concept |
|
| Content
| covers full range of meaning |
|
| Criterion
| scores accurately compare to a validated
measure |
|
| Construct
| measure relates to a variety of other
measures as specified in theory |
|
Reliability
| When a measurement procedure yields
consistent scores when the phenomenon being measure is not changing, the measurement is
said to be reliable. |
Reliability
| Test-retest
| intraobserver or intrarater (by one
observer at different times) |
|
| Interitem
| internal consistency of a set of questions |
|
| Alternate-forms |
| Split-halves |
| Interobserver (by more than one observer) |
Measurement
Operations
| Available Data |
| Questions |
| Observations |
| Indirect Measures |
| Combined Measures |
The Inadequacy of
Self-Reports Regarding Socially Desirable Behavior: Observed versus Self-Reported Church
Attendance
High school seniors in a small town are
asked a series of questions to measure the frequency of their drinking and driving. The
students give consistent answers to the series of questions, but police records indicate
that drinking and driving is much more common than indicated by the students’
answers. This suggests that the measure of frequency of drinking and driving is reliable
but not valid!
URLs for Substance
Abuse Sites
|