Chapter 6 Review Flashcards
Quantitative and Qualitative measurement approach
- Timing
- Data
- Data-concept-measurement linkages
Timing
Quantitative researchers convert measures in the design stage
Qualitative researchers do this later after some data is gathered
Data
Quantitative researchers develop techniques that produce numeric data
Qualitative researchers don’t try to convert data into numbers
Data-concept-measurement linkages
Quantitative researchers develop measurement techniques that bridge concepts and data
- Moving from the concepts, the researcher develops ways to measure that concept
Qualitative researcher use the data to develop their concepts
- Moving from the data, the researcher develops definitions of concepts
Operationalization
This links a conceptual definition to a specific set of measurement techniques or procedures
Conceptualization
The researcher develops concepts as they work through data collection and analysis
Develops concepts, defines them, and considered relationship amount them
Qualitative operationalization and conceptualization
Conceptualization
Research develops concepts as they work through data collection and analysis
Develops concepts, defines them, and considers relationship among them
Often, concepts are derived from data (eg. field notes, interviews)
Quantitative Conceptualization and operationalization
Conceptual hypothesis - A type of hypothesis in which the researcher expresses variables in abstract
Empirical Hypothesis - A type of hypothesis in which the researcher expresses variables in specific terms and expresses the association amount the measured indicators of observable empirical evidence
Reliability
If we use the same technique to study the same phenomenon repeatedly we should see the same or at least similar results
Exmaple: Measures should be consistent across different researchers
Validity
This refers to the degree to which a measure reflects the concept in use
Does a measure really reflect the concept its supposed to
Reliability and Validity in Quantitative Research
- Clearly conceptualize constructs
- Increase the level of measurement
- Use multiple indicators of a variable
- Use pretexts, pilot studies and replication
Clearly conceptualize construct
Reliability increases when a single construct or sub-dimension of a construct is measured
Increase the level of measurment
Indicators at a higher or more precise levels of measurements are more likely to be reliable than less precise measures because the latter pick up less deteailed information
Ex. Income - measure in 10,000 rather than high medium or low
Use multiple indicators of a variable
Use of two or more of the same construct are better than one
Eg. To examine “concern for the environment” attitude measures and behaviour measures
Use of pretexts, pilot studies and replication
Protest or trial a measure first
Involves developing a preliminary version and using it before hypothesis testing
Reliability in Qualitative Research
Reliability
- Some qualitative researchers criticize the idea of reliability
Questioned because the social world evolves, changes over time
Researchers relationship to what they studied and the relationship with people they studied
Validity in Qualitative Research
Authenticity - Means giving a fair, honest and balanced account of social life from the viewpoint of the insider of those who participated in the research
They use servers techniques to convey the insiders view to others
Credibility - Are findings consistent with reality
Transferability - Are findings applicable beyond the immediate setting
Types of measurment in validity
- Face Validity
- Content Validity
- Criterion validity
Face Validity
It is a judgement by the scientific community that the indicator measures the construct
How much a person earns per year would be a good indicator of income
Content Validity
It addresses the question “is the full content of a definition represented in a measure”
Example: feminism commitment to full equality between men and women at work and at home
We would need indicators for both gender equality at work and gender equality at home to fully represent the measure
Criterion Validity
This uses some standard that is already judged as valid to indicate a construct accurately. Value is assessed by comparing the measure to the standard
2 subtypes
- Concurrent Validity
- Must be associated with a pre-existing indicator that is judged to be valid (it had face validity) - Predictive Validity
- An indicator predicts future events that are logically related to a construct
eg. If a student gets a high LSAT score then they will do well in law school
Levels of Measurement
Nominal Measures
Ordinal Measures
Interval Measures
Ratio Measures
Nominal measures
They indicate only that there is a difference among categories
eg. Marital status, married, single
The attributes CANNOT be ranked
Ordinal Measures
They indicate a difference plus the categories can be ordered or ranked
Eg. Letters, grades
Interval Measures
Identifies differences amount variable attributes, and ranks, and measures distance between categories
Distance between adjacent attributes is the same
No true zero
Eg. Temperature
Ratio Measures
The most precise level of measurement for which variable attributes can be ranked in order, the distance between attributes is precisely measured and has an absolute zero
Eg. Income, years of education, age
Qualities of a good measure
Scales and Indices
They are often used in social research to measure degree or intensity
Both scales and indices produce ordinal or interval-level measures
Scales and indices condense or compress information that is gathered
Eg. Instead of using 8 separate items we compress these into an index and get one measure
Scales is an ordinal, interval, or ratio measure of a variable expressed as a numerical score
Used to:
1. Help conceptualize and operationalize. Scaled show the fit between a set of indicators and a single construct
- Scaling produces quantitative measures can be used with other variables to test hypotheses