Quantitative and Qualitative Measurement Flashcards
How do quantitative researchers measure data?
- Measurement occurs prior to data collection (deduction)
- Measurements and data are numbers
- Measurements link concepts to data
How do qualitative researchers measure data?
- Measurement occurs during data collection (induction)
- Measurements and data are abstract
- Concepts are developed based on measurements
What is conceptualization?
Process of developing clear conceptual definitions for abstract ideas
What is quantitative conceptual definition?
A careful, systematic definition of a construct that is explicitly written to clarify one’s thinking
What is quantitative operationalization?
The process of moving from the conceptual definition of a concept to a set of measures that allow a researcher to observe it empirically
What is quantitative operational definition?
The definition of a variable in terms of the specific activities to measure or indicate it with empirical evidence
What is a conceptual hypothesis?
- Researcher expresses variables in abstract, conceptual terms, and expresses relationship among variables in theoretical way
- Abstract, causal relationship between 2 concepts
What is an empirical hypothesis?
- Researcher expresses variables in specific terms and expresses the association among the measured indicators of empirical evidence
- Degree of association between indicators
What is the concrete empirical world?
If the operational indicators of variables are logically linked to a concept, they will capture what happens in the empirical social world and relate it to the conceptual level
What is qualitative conceptualization?
Process of forming theoretical definitions as the researcher organizes (CODES) data
What is qualitative operationalization?
- Describing how specific observations about data contributed to working ideas that are the basis of conceptual definitions
- Not pre-planned; after-the-fact description
- Occurs BEFORE conceptualization
What is reliability?
- Dependability/consistency of the measure of a variable
- Measurement will repeat same results under same conditions
- Necessary for validity (for quantitative) and easier to achieve
How do you increase reliability?
- Clear conceptualization to eliminate “noise”
- Each measure should indicate only 1 concept - Use a precise level of measurement
- To pick up detailed information - Use multiple indicators to measure a variable
- Several measures are less likely to have the same systematic error - Use pilot tests or pre-tests
- You can even search literature to replicate measures other researchers have used
What is validity?
- The ability to generalize findings outside a study, the quality of measurement, and the proper use of procedure
- A term meaning “truth” that can be applied to the logical tightness of experimental design
What are the 4 types of validity?
- Internal validity: ability of experimenters to strengthen a causal explanation’s logical rigour by eliminating potential alternative explanations for an association
- External validity: ability to generalize from experimental research to settings or people that differ from the specific conditions of the study
- Statistical validity: achieved when the appropriate statistical procedure is selected and the assumptions of the procedure are fully met
- Measurement validity: how well an empirical indicator fits with the conceptual definition of the concept
What are the 3 types of measurement validity?
- Face validity: indicator “makes sense” as a measure of a construct in the judgement of the scientific community
- Content validity: measure represents all aspects of the conceptual definition of a construct
- Criterion validity: relies on independent, outside verification; if 2 measures are for a similar concept, they should yield similar results
What are the 2 types of criterion validity?
- Concurrent validity: relies on pre-existing and already accepted measure to verify the indicator of a concept
- Predictive validity: relies on the occurrence of a future event or behaviour that is logically consistent to verify the indicator of a concept
- Eg. if LSAT has high predictive validity, then students who get high LSAT scores will do well in law score
What is authenticity?
- Giving a fair, honest, and balanced account of social life from the viewpoint of someone who lives it everyday
- We resist reliability because things change between individuals and over time ⟶ “dependability”
- We always want to have validity (tight fight between understanding of social world and what is occurring in it)
What are levels of measurement?
- A system that organizes the information in the measurement of variables into 4 general levels
- A wide range of powerful statistical procedures are available for higher levels of measurement, but not for lower
What are discrete variables and continuous variables?
Discrete variables: variables in which the attributes can be measured only with a limited number of separate categories (eg. marital status)
Continuous variables: variables measured on a continuum in which an infinite number of finer graduations between attributes is possible (eg. age, crime rate)
What are the levels of measurement for discrete variables?
- Nominal measures: the lowest, least precise level of measurement for which there is only a difference in type among the categories of a variable
- Eg. religion ⟶ Christian, Jewish, Muslim - Ordinal measures: identify difference among categories of a variable and allows the categories to be ranked
- Eg. letter grades ⟶ A, B, C, D, F
What are the levels of measurement for continuous variables?
- Interval measures: identifies differences among variable attributes, ranks, and categories, and that measures distance between categories, but there is NO true zero
- Eg. Fahrenheit or Celsius ⟶ 5, 45, 90
- A rise in temperature from 30 to 60 is not a doubling of temperature because 0 is not the absence of heat - Ratio measures: most precise level of measurement for which variable attributes can be ranked, distance between attributes is precisely measured, and an absolute zero exists
- Eg. money income ⟶ $10, $100, $500
- There is a true zero ($0) which makes it possible to state relations in terms of proportions
When constructing a measure, you should make sure your attributes are ___ and ___.
- Mutually exclusive: person’s responses fit into only one category
- Exhaustive: measure should provide a category for all possible responses
What are proxys?
Represent intangible concepts that cannot be measured (eg. GDP for quality of life)
What is an index?
- The summing of many separate measures of a construct or variable into a composite measure (single numerical score)
- Make sure every item in index has face validity
- Each part of the conecpt should be measured with at least 1 indicator (content validity)
- Often using ordinal or interval measures
What is a scale and what are they used for? What level of measurement are scales most often at?
- A type of quantitative data measure often used in survey research
- Captures the intensity, direction, level, or potency of a variable construct along a continuum
- Common in situations where a researcher wants to measure how an individual thinks about something
- Most are at the ordinal level of measurement
- Show fit (consistency) between set of indicators and a single construct
- Produces quantitative measures and can be used with other variables to test hypotheses
What is standardization/norming?
- The procedure to statistically adjust measures to permit making an honest comparison by giving a common base to measures of different units
- Eg. comparing crime rates between cities when different cities have different numbers of people
What are the components of trustworthiness?
- Authenticity
- Dependability
- Credibility ⟶ does this sound right? (internal validity)
- Transferability ⟶ the extent to which our findings are applicable beyond our immediate research setting (external validity)
An appropriate level of measurement for a variable depends on __ and __.
How a concept is defined
The type of indicator or measurement that a researcher uses
Can you measure everything?
- Every social phenomenon can be measured
- May have to use proxys ⟶ represent intangible concepts that cannot be measured directly (eg. GDP for quality of life)
What is the quantitative process?
- For BOTH independent and dependent variable:
- Conceptualization
- Operationalization - The LINK between independent and dependent variables
- Conceptual hypothesis
- Empirical hypothesis
- Concrete empirical world
How does external validity differ from generalization of results?
EXTERNAL VALIDITY = “Is my sample representative enough so that the results can apply to the population”
Not to be confused with generalizability of results (the phenomena is applicable to other cultures/populations)
What are attributes?
The categories of a variable
Eg. if variable = marital status, attributes = married, divorced, never married