The process of measurement begins after a researcher has formulated a research question and determined the variables and units of analysis. The main concern here is to develop the clear definitions and to create measurement that will yield precise and accurate findings.
Measurement process (COM):
Conceptualization: is the process of taking a construct or concept and refining it by giving it a conceptual or theoretical definition. Eg. What is religious?
It involves carefully thinking, directly observing and consulting with others, reading others’ work and trying possible definitions.
Operationalization: the process of developing an operational definition for the construct. It involves to use indicator to measure a concept. Eg to use times to go to church or religious service as one of indicators to measure being religious.
In this process, a researcher links the world of ideas to observable reality.
Rule of correspondence or auxiliary theory, are the logical statements of how an indicator corresponds to an abstract construct.
For example, a rule of correspondence states that a person’s verbal agreement with a set of ten specific statements is evidence that the person holds strongly anti feminist beliefs and values.
For example, auxiliary theory suggests that the construct of alienation has four parts: family relations, work relations, relations with community and relations with friends.
Empirical measurement, using indicators in empirical world.
Reliability: means that information provided by indicators eg. a questionnaire does not vary as a result of characteristics of the indicator, instrument, or measurement device itself. It means that the repeated measurement under the same condition should be the same.
Three types of reliability:
o Stability reliability is reliability across time
o Representative reliability is reliability across sub populations or groups of people
o Equivalence reliability apples when researchers use multiple indicators (e.g., several items in a questionnaire all measure the same concept). If several different indicators measure the same concept, then a reliable measure should give the same result with all indicators.
How to improve the reliability:
1. Clearly conceptualize all concepts, reliability increase when a single concept or sub-dimension of a concept is measured. That shows the more precise a concept is defined, the less noise in measurement.
2. Increase the level of measurement, for example, 1 to 10 level categories is more reliable than high or low.
3. Use multiple indicators, it allows researchers to measure a wider range of content of a conceptual definition; one indicator may be imperfect, several indicators are less likely to have the same systematic error. Multiple indicators tend to be more stable than measures with one item.
4. Use pretest, pilot study and replication.
Validity is the degree of fit between a concept and indicators of it. It refers to how well the conceptual and operational definitions mesh with each other. The better the fit, the greater the measurement validity.
Four types of validity:
1. Face , in the judgment of others
2. Content, capture entire meaning
3. Criterion, agree with an external source
- concurrent, agrees with a preexisting study
- predictive, agrees with future event
4. Construct, multiple indicators are consistent.
- convergent, alike ones are similar
- discriminate, different ones differ
No comments:
Post a Comment