Chapter 4: Data and Nature of Measurement Flashcards
The four properties of the abstract number system are:
- Identity, 2. Magnitude 3. Equal Intervals 4. Absolute zero
Nominal Scales
no place in numerical system. Only give identityi.e. different categories as in sex or occupation
Ordinal Scales
identity and magnitude: there is an order to the data, but it is still categorical e.g. educational level
Interval Scale
identity and magnitude and equal intervals.Interval scale data is called score data, since the value represents a score e.g. temperature: (note: zero degrees is not equal to zero temperature)
Ratio Scales
identity and magnitude and equal intervals and absolute zero. Also called score data (e.g. distance, length, weight)
response-set bias
a measurement error,basically the tendency to respond in a different was across situations. An example is the social desirability, which is the tendency to give a socially accepted answer, rather than a truthful one
Operationalization
the process of making a variable measurable; an operationalized construct is thus a measurable version of a theoretical construct, but taken from an abstract level to an empirical one
convergent validity
different kinds of measurement that give the same result, called multilevel-approach
three factors that determine the quality of a measure
Reliability, Validity and Objectivity
Reliability
consistency of a measure
inter-rater reliability
level of agreement between different raters
Test-retest reliability
measured by making participants re-take a test
internal consistency
extent to which different parts of a test show the same results
effective range
extent to which an instrument is appropriate for the participants
scale-attenuation
occurs when the reach of a scale is limited. e.g. ceiling effect and floor effect (important)