Term
|
Definition
| Assignment of numbers to some phenomenon for analysis (usually go by categories) |
|
|
Term
|
Definition
| Statement that describes how a concept will be measured |
|
|
Term
|
Definition
| Variable/set of observations that results from applying the operation definition (Indicator=concept+error) |
|
|
Term
|
Definition
| Several indicators used to measure a single concept (triangulation- "home in" on a concept) |
|
|
Term
|
Definition
| Accurately measures the concept its intended to measure |
|
|
Term
|
Definition
| If indicators of a concept produce similar results, the indicators have convergent validity |
|
|
Term
|
Definition
| Whether the indicator allows the concept to be distinguished from other similar yet different concepts |
|
|
Term
|
Definition
| A measure is valid if it taps what the research analyst thinks it measures |
|
|
Term
|
Definition
| An indicator has face validity when the researcher accepts the indicator as valid |
|
|
Term
|
Definition
| When an indicator has consensual validity when numerous researchers accept the indicator as valid |
|
|
Term
|
Definition
| Validity established when an indicator correlates strongly with other accepted valid indicators |
|
|
Term
|
Definition
| An indicator that correctly predicts a specified outcome |
|
|
Term
|
Definition
| Consistently assigns the same # to some phenomenon that hasn't changed |
|
|
Term
|
Definition
| An indicator is reliable if it consistently assigns the same numbers to similar phenomenon (Two main threats to measurement reliability are subjectivity and lack of precision) |
|
|
Term
|
Definition
| Relies on the judgement of the measurer or of a respondent in a survey |
|
|
Term
|
Definition
| Threat to the reliability of measurement arising from the use of small samples of measurement scales lacking sufficient gradations |
|
|
Term
| Parallel Forms Reliability |
|
Definition
| The correlation between responses obtained on two sets of items measured for reliability (creation of alternative forms that are parallel and administered at the same time) |
|
|
Term
|
Definition
| Measure of reliability in which the set of items intended to measure a given concept is divided into two parts (Cronbach's alpha is based on this method; usually done with even and odd number questions) |
|
|
Term
|
Definition
| A technique used to assess the degree of consistency among individuals who are applying a measurement scheme to collect and code data |
|
|
Term
|
Definition
| Based on judgement of one or more persons |
|
|
Term
|
Definition
| Based on a report/docs and don't require any judgement on the researcher's part |
|
|
Term
|
Definition
| Circumvent the altering of behavior being assessed (avoidance of the Hawthorne event where the measuring of a phenomenon can alter the behavior that is being assessed) |
|
|
Term
|
Definition
| The precision inherent in the measurement of different types of measurement |
|
|
Term
|
Definition
| Highly precise level of measurement based on a unit or interval accepted as the common standard |
|
|
Term
|
Definition
| Level of measurement at which its possible to say that one object (or event/phenomenon) has more or less of a given characteristics than another but not how much or less of it (ex. satisfaction questions vs. weight) |
|
|
Term
|
Definition
| Shows how an item is either the same or different with no exact trait (interval level) or more/less of a characteristic (ordinal level) |
|
|
