DOC PREVIEW
UW-Madison SOC 357 - Conceptualization Lecture Notes

This preview shows page 1-2-3-24-25-26 out of 26 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 26 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Conceptualization • 1. Definition • 2. Dimensions• 3. Indicators • 4. Comparison of Concept, Dimension, and Indicator• 5. ExampleDefinition• Conceptualization is the process of development and clarification of concepts. • In other words, clarifying one's concepts with words and examples and arriving at precise verbal definitions. • e.g., what is meant by education? • “Amount of knowledge and training acquired in school." Another Example• What do we mean by "social status?”– Wealth (millionaire)– Prestige (Harvard professor)– Power (military general)• These are called “dimensions” of social status.Dimensions• We classify different meanings into different groups. Such groups are called "dimensions." • A concept may have more than one dimension (e.g., as in case of social status). • At a practical level, we are usually more interested in dimensions than in concepts (which are more abstract, vague). Indicators • When a dimension is not directly observable, we use indicators. • For example, to measure power, we may use – (1) number of people under your supervision– (2) extent of your supervision (work‐related only, or sleep and food?)– (3) your annual budget– (4) amount of equipment under your controlComparison of Concept, Dimension, and Indicator• In practice, the three terms are often interchangeable (e.g., gender, race). • One difference is the level of abstraction:Concept Dimension IndicatorHighly abstract Abstract Concrete• One concept may have multiple dimensions; and one dimension may have multiple indicatorsA Related term: Variables• A variable is a statistical term, meaning a quantity that can take on different possible values. • Both dimension and indicator can be variables. • When a concept has only one dimension with one indicator, a concept is practically equivalent to a variable. SummaryConceptD2D3D1...I12...I11I13Example: SESSESWealth PowerPrestigeI32I31I33...Measurement • When it comes to measurement, we are talking about variables and indicators. • Definition of measurement: "the assignment of numbers or labels to units of analysis to represent variable categories." • Numbers mean different things under different circumstances. Types of Measurement• Nominal measurement• Ordinalmeasurement• Intervalmeasurement• RatiomeasurementNominal Measurement• Nominal measurement is a system in which cases are classified into two or more categorieson some variable.• Arbitrary numerical assignments. e.g., Race = 1 for white, 2 for black, and 3 for Asian.• Two criteria for classifications: – Exhaustiveness– Mutual exclusiveness Ordinal Measurement• In ordinal measurement, numbers indicate the ranking order on a dimension.• e.g., for a typical attitude question on surveys,Strongly Agree Somewhat Agree Neutral Somewhat Disagree Strongly Disagree 5 4 3 2 1 • There is no intrinsic scale – you only know relative rankings.• Should not do arithmetic (such as averaging).Interval Measurement• Interval measurement assumes equal distances or intervals between "numbers." • Numbers represent not only rankings but also “values.”• e.g., (70, 90, 80) is the same as (80, 80, 80) for the test component of your Soc.357 final grade.Ratio Measurement• If an interval variable has an absolute zero, it becomes a ratio variable. • E.g., weight, number of siblings, birth rate, etc.• Compare three temperatures: C, F, and K. A Comparison of the FourInformation Provided Nominal Ordinal Interval Ratio Classification X X X X Rank order X X X Equal intervals X X Relative value X More general More specificQuality of measurement• 1. PrecisionPrecision is the extent to which numerically‐detailed measurement remains meaningful (e.g., measuring annual income).• 2. AccuracyAccuracy is the extent to which the measuring instrument measures what it is intended to measure. (Commonly gauged by reliability and validity).Consequence of measurement error: descriptive versus explanatory studies• In explanatory studies, we are interested in relationships between two concepts, or two variables. When a concept is poorly defined, the relationship is still there.• e.g., gender and religiosity.• Say the research interest lies in the differentialbetween men and women. If we underestimate the absolute level by 30% for both, we still have an accurate measure of the relative value. Gender and Religiosity Example • i.e., explanatory analysis is robust to marginal changes of measurements. • Reason: overestimation or underestimation for allunits of analysis. • Beware of differentialunder(over)estimation.Gender True MeasuredMen 50 35 Women 100 70Methods for Assessing Reliability• Test‐retest methodAs in panel design, you measure the same quantity twice. Let us call the first measurement y1, the second y2. Reliability measure: correlation between y1and y2.• Low correlation means large random noise.Methods for Assessing Reliability (Continued)• Split‐half methodYou may calculate correlations among different items on the same survey instrument, assuming equivalence of different items. • High correlation: high reliabilityBias (Lack of Validity)• In regression context,yi= α + β xi+ εiεidoes not have a mean of zero. Rather, it could be that εi= 4+δi , where δiis well behaved (i.e., has a mean of zero). Thus, 4 is the bias. Methods for Assessing Validity• A. Face validitySee whether a measurement makes any sense to you and to others (also compare with similar independent measures – GSS)• B. Criterion‐related validityDoes this measure predict other measures that can be measured more objectively.• Use correlation orregression techniques.Example: Predictability of SAT score of College Performance• If SAT is a good measure of academic ability, why doesn’t it predict college performance well? Problems:• (1) selection bias; • (2) mediating factors between first measurement and criterion (e.g., instruction); • (3) measurement problems for the criterion (e.g., grade inflation, differential grading, etc.).Methods for Assessing Validity (Continued)• C. Content validityThe extent to which an empirical measurement reflects a specific domain of content. • D. Construct validityConstruct validity is concerned with the extent to which a particular measure relates to other measures in ways


View Full Document

UW-Madison SOC 357 - Conceptualization Lecture Notes

Documents in this Course
Syllabus

Syllabus

12 pages

Sampling

Sampling

35 pages

Class 7

Class 7

6 pages

Review

Review

3 pages

Load more
Download Conceptualization Lecture Notes
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Conceptualization Lecture Notes and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Conceptualization Lecture Notes 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?