Race and ethnicity: categories, identities and comparable counts
A category count can change because people change, the observed population changes or the coding rule changes. Reconstruct all three before interpreting a trend.
Course topic: Race and Ethnicity
Study route and exercise: University Reading List editorial project. How we curate · Sources checked:
On this page
Scope and responsibility
A and B below are invented self-declared cultural affiliations, not biological populations or real ethnic groups. The packet does not measure prejudice, discrimination or the causes of a real group’s circumstances.
Choose your starting point
Begin with cultural assumptions
Use Kennedy to examine cultural framing and the difference between individual experience and institutional patterns.
Clarify concepts
Use Kivisto/Croll for race/ethnicity, prejudice/discrimination and inclusion.
Study historical formation
Use Omi/Winant after introductory sociology to examine the US-specific relation between racial meanings and institutions.
Compare the three books
Beyond Race: Cultural Influences on Human Social Life
Vera Kennedy · Edition: 1 · 2018
Free cultural foundation
- Choose it when
- You want an accessible route into cultural assumptions and social relations.
- Use another route when
- You need a complete account of all racial histories or an advanced racial-formation argument.
- Selected reading
- Use the named culture, identity and intergroup-relations discussions with the classification packet.
- Limits and trade-offs
- A broad cultural perspective should not turn structural processes into individual attitudes alone.
Try before choosing
Separate the people’s declared sets from the institution’s single-code rule.
Check your work: P5’s two affiliations exist in the packet even when only A is recorded.
Race and Ethnicity: The Basics
Peter Kivisto; Paul R. Croll · 2012
Conceptual introduction
- Choose it when
- You need clear distinctions before interpreting evidence.
- Use another route when
- You need one detailed national history or an advanced theory monograph.
- Selected reading
- Read chapters 1–3 on definitions, prejudice/discrimination and intersecting inequalities.
- Limits and trade-offs
- Different concepts require different evidence: counts alone do not identify discriminatory treatment.
Try before choosing
Explain what the table can show about coding and what it cannot show about prejudice.
Check your work: An unequal or changing count is not by itself a measured motive or mechanism.
Racial Formation in the United States
Michael Omi; Howard Winant · Edition: 3 · 2014
Racial-formation theory
- Choose it when
- You have race/ethnicity foundations and US historical context.
- Use another route when
- You need a first definition or a universal template for every society.
- Selected reading
- Read the selected third edition’s discussions of racial formation and racial projects, alongside its historical framing.
- Limits and trade-offs
- The US case must not be silently generalized to all contexts or current conditions.
Try before choosing
Ask who sets the priority code and what the code makes administratively visible.
Check your work: Identify a classification mechanism without attributing an unrecorded intention or treating identities as fixed essences.
Check your preparation
Must categories be mutually exclusive?
Only if the rule makes them so. A multiple-affiliation table may count the same person in A and B.
Which denominator is appropriate?
People for a person-membership percentage; selections for a share of all selections. Name the estimand before dividing.
Does a count show a biological group?
No. Here categories are declared affiliations interpreted through explicit coding rules; no biological inference is licensed.
A six-step study plan
Steps completed: 0 / 6
These are your own study marks, not an assessment. They are not sent to this site.
-
Record the definitions
Selected reading: Concept introduction: chapter 1.
What to do: Write affiliation, code, person and selection as separate fields.
Check your work: Do not collapse identity into one administrative code.
-
Reproduce both maps
Selected reading: Kennedy with packet.
What to do: Assign all six people under each period’s rule.
Check your work: P5 changes representation, not declared affiliation.
-
Audit totals
Selected reading: Definitions and evidence.
What to do: Compute person counts, selections and memberships.
Check your work: Seven selections come from six people.
-
Check comparability
Selected reading: Racial-formation lens.
What to do: Identify who changed the rule and which trend claim it affects.
Check your work: Do not infer that the category itself grew in a comparable measure.
-
Hold the rule fixed
Selected reading: Follow-up.
What to do: Recount the complete entering/leaving population.
Check your work: A composition change is distinct from a continuing person changing affiliation.
-
Write a bounded interpretation
Selected reading: All three routes.
What to do: Report the measure, rule and population with a limit.
Check your work: Do not invent prejudice, hierarchy or motives from these counts alone.
Worked exercise
The same six people produce a growing category
Complete fictional declared sets, unchanged across periods: P1={A}, P2={A}, P3={B}, P4={B}, P5={A,B}, P6={none}. Period 1 assigns one code: A if A is present, otherwise B if B is present, otherwise O. Period 2 records every declared affiliation, using O only for none. The institution alone changes this coding rule; no person enters, leaves or changes declaration. Count both periods and assess “B membership grew”.
Assumptions
The six records are the entire population for this exercise. O means no listed affiliation, not a third biological or cultural group. A/B are fictional and overlap is permitted. No outcomes, attitudes or real demographic facts are supplied.
- Period 1 codes are A,A,B,B,A,O, so A=3, B=2, O=1 and six mutually exclusive codes. P5’s B affiliation is suppressed by the priority rule.
- Period 2 has A=3, B=3, O=1: seven selections but six people because P5 appears twice. B membership among people is 3/6=50%; its share of selections is 3/7, a different quantity.
- The old B code rate 2/6 and new B membership rate 3/6 do not use the same rule. Their difference records a classification change, not a new affiliation among these unchanged people. Sum of person-membership percentages can exceed 100% because categories overlap.
Answer and interpretation
B’s recorded count rises 2→3 while all declared affiliations and people stay fixed. A comparable population-growth conclusion is not supported. The full mapping explains both the extra selection and the visibility change.
Test the result
Reconstruct P5 explicitly; do not divide B by seven and call it the proportion of people. Neither code rule reveals anyone’s motive or biological essence.
Change an assumption
Now hold the rule fixed and change the population
Use the period-2 multiple-affiliation rule for both comparison points. Initially use the original six. At the later point P2 leaves and new P7={B} enters. Complete later population: P1={A}, P3={B}, P4={B}, P5={A,B}, P6={none}, P7={B}. Continuing people keep their declarations.
- Initially under this common rule A=3, B=3, O=1 among six people. Later A=2, B=4, O=1, again seven selections among six people.
- B person-membership rate changes from 3/6 to 4/6, an increase of 1/6, or about 16.67 percentage points, under a fixed definition and denominator size.
- The change is composition: P2 left and P7 entered. None of the five continuing people changed affiliation. Calling it an individual conversion from A to B would contradict the supplied person-level histories.
Answer and interpretation
A comparable B membership-rate increase is now present in this toy population, attributable here to the supplied entry/exit composition. It does not show changes among continuing people, a real migration mechanism or discrimination.
Test the result
Use person identifiers as well as totals. Constant population size does not mean the same people; multiple selections still do not create a seventh person.
Common mistakes
Code becomes identity
An institutional category can omit a declared affiliation.
Selections become people
Overlapping counts need a stated denominator and estimand.
Composition becomes conversion
Track continuing, entering and leaving people before explaining a trend.
Before moving on
- Choose readings by cultural framing, conceptual clarity and historically located theory.
- Reproduce a classification map and distinguish memberships from selections.
- Separate rule changes, population composition and person-level changes.
Sources and edition checks
Sources checked:
- Kennedy: official author-attributed open text
Kennedy’s open cultural text is a starting route, with individual credits and licence notices to respect.
- Kivisto/Croll: publisher first edition
The 2012 first edition distinguishes definitions, prejudice/discrimination and intersecting inequalities.
- Omi/Winant: selected third edition
The selected 2014 third edition is explicitly US historical analysis; its classification lens is not evidence of a particular motive in the invented institution.
These are proposed study activities, not official course assignments or a claim that every book has been read cover to cover.