RMS Racial Politics

Omi and Winant (2017): Racial Formation

• Race is not biological. It is a social construction that is created, changed, and sometimes destroyed through historical and political struggle, a process they call racial formation. Racial projects link ideas about race (representation) to how resources are distributed (structure), and racialization is the process of giving a group a racial meaning.

• Theoretical / historical

• Race vs. ethnicity, panethnicity

Hutchings and Valentino (2004), Racial Determinants of Politics

• Literature review of survey and experimental evidence

• Group centrism, symbolic racism

• Reviews how racial attitudes shape American politics. Overt "old-fashioned" racism has declined, but subtler racial resentment still shapes vote choice and policy opinion, and campaign messages can activate racial considerations (explicitly or through coded cues).

Lee (2004): Identity-to-Politics Link

• critiques the common assumption that sharing a demographic or racial/ethnic label automatically translates into shared political interests and collective action

• Peer-review article

• Panethnicity, instrumentalism

Ozawa v. United States (1922)

• Takao Ozawa, a Japanese immigrant, sought naturalization under a law limiting it to "free white persons." The Court ruled that "white" meant Caucasian, so Ozawa, as Japanese, was ineligible.

• legal case

• Race as legal construction

Thind v. U.S. (1923)

• Bundle of sticks, race construction

• Bhagat Singh Thind, an Asian Indian, argued he was Caucasian and therefore white. The Court rejected this, saying "white" should follow the common understanding of ordinary people, not scientific classification.

Bertrand and Mullainathan (2004): CV Experiment

• They sent out fictitious resumes in response to real job ads, randomly assigning names that sounded white (e.g., Emily, Greg) or Black (e.g., Lakisha, Jamal). White-sounding names got about 50% more callbacks, and better resumes helped white applicants more than Black applicants.

• field experiment

• Bias, internal validity

Adida, Laitin and Valfort (2010): Ethnic Names Experiment

• Ethnicity vs. race, bias

• Tests discrimination against Muslims in France by sending job applications from candidates with French-Christian or Muslim names (Senegalese origin). Muslim-named candidates were called back less often, pointing to a barrier to integration tied to religion/ethnicity rather than skin color alone.

• field experiment

Terkildsen (1993)

• Survey

• Examines how white voters evaluate Black candidates depending on the candidate's skin tone (light vs. dark) and the voter's level of prejudice. Prejudiced white voters evaluated dark-skinned Black candidates more negatively. It is an experiment.

Michelle Alexander (2012): The New Jim Crow

• Argues that mass incarceration, driven by the War on Drugs, works as a modern racial caste system. Drug use is similar across races, but enforcement and sentencing fall hardest on Black Americans. Once labeled a felon, a person faces legal discrimination in jobs, housing, voting, and jury service.

Hajnal et al. (2017), Voter IDs

• Observational, large-N survey data

• Finds that strict voter ID laws reduce turnout among racial and ethnic minorities and widen the participation gap, using large national survey data. It is observational, and the findings have been debated by other researchers.

Wedeen (2010): Ethnography

• A methods piece defending ethnographic and interpretive work in political science. It argues that long immersion in a setting reveals meanings and practices that surveys and statistics miss.

• Qualitative research

King, Keohane and Verba (KKV, 1994): Unified Logic of Inference

• Argues that qualitative and quantitative research share the same logic of inference. Both should aim for valid descriptive and causal inference, with clear theory, careful measurement, and attention to bias and uncertainty.

• fundamental problem of causal inference = how can we observe the counterfactual?

Shively (2013): Research Classification

• A research methods text that sorts types of research (empirical vs. normative, positive, engineering, theory-oriented) and explains the basics of the research process, hypotheses, and variables

Orfield and Eaton (1996): Racial Integration Measurement

• Segregation in schools is tightly linked to socioeconomic segregation

• has increased for Black/Latino students since 1880s

• uses data and exposure index to compare regions, finds strong correlation btween % poor and % minority in school

• Black / Latinx 12x more likely to attend extreme poverty schools (higher dropout)

• not soley about race, more about access to resources

Perez (2015): Implicit Bias

• Explains implicit attitudes, which are automatic evaluations people may not be aware of, and how they are measured (for example, the Implicit Association Test). Discusses how they differ from explicit, self-reported attitudes and whether they matter for political behavior.

• Review

Tarman and Sears (2005): Symbolic Racism

• Defends the concept and measurement of symbolic racism, a modern prejudice that mixes anti-Black feeling with beliefs in individualism and hard work. They address critics who say it just measures conservatism rather than racial attitudes.

• ex: The perception that Black Americans' lower socio-economic standing is due to a lack of effort, discipline, or work ethic rather than structural factors

• Survey measurement debate

ethnic primordialism

• The view that ethnic identity is ancient, natural, and given by birth (blood ties, deep attachments). It treats ethnic conflict as rooted in unchangeable differences.

• Example: The idea that groups are "naturally" divided is the view Omi and Winant argue against; useful as the contrast case

Ge et al. (2020): Uber/Lyft Experiment

• Field experiment on ride-hailing in which riders with African American-sounding names faced longer waits and more cancellations than riders with white-sounding names. It shows discrimination in an app-based service where drivers can see a name.

Rolasky 2022 racial wealth gap

• Johnson became president after Lincoln’s assasination (turncoat)

• America's freed slaves were never fully compensated for centuries of exploitation and oppression.

• didn’t have same ability to accumulate wealth, historically disadvantaged

• Descriptive statistics

Wilkes (2015): Differences in Trust among Racial Minorities

• examines how racial and ethnic minority communities process trust in government and institutions differently than white Americans, showing that racial gaps vary significantly based on political context, partisanship, and social structure

• Survey analysis

Luca, Endelman, Svirsky (2016): AirBnB Experiment

• The researchers created guest accounts that differed only by name (distinctively African American vs. white) and sent booking requests to Airbnb hosts. Guests with African American-sounding names were accepted about 16% less often.

• field experiment

Brader, Valentino & Suhay (2008): bias in the electorate

• Studies how group cues in news stories (such as the ethnicity of immigrants shown) affect public opinion on immigration. The key claim is that cues showing Latino immigrants raised anxiety, which in turn increased opposition. It is a survey experiment.

• The researchers manipulated real-world news style articles focusing on the societal costs of immigration. They varied two main variables:

1. The economic/social impact frame: Highlighting severe negative impacts versus positive impacts.

2. The stigmatized group cue: Using a photo and typical name of either a Latino immigrant or a European (Russian) immigrant

Where do the concepts of race and ethnicity come from?

• Omi and Winant (2017) argue race is created through racial formation, a historical process where racial categories are made, changed, and destroyed through racial projects

• Race formed in US through socio-historical process that made categories form (initially just white/non-white, then free-colored persons /slaves, eventually Asian category introduced in late 1800s)

What is race? Ethnicity? What are the differences between the two?

• Race = biologically-defined social construction (based on physical characteristic like skin color, hard to move in between (sticky)

• Ethnicity = more culturally specific (Indian, Nigerian), and is a reflection of where you’re from / your ancestry. Is also the result of group formation process based on culture and descent (less obvious/sticky) - Omi & Winant

• ex: Ozawa / Thind - had to redefine what it meant to be white because neither “made the cut”

What makes race / ethnicity politically salient?

• Because race is a more prevalent decider then we think in who is elected, who gets represented, and who gets what opportunities in society.

• Emily / Greg CV experiment (Black sounding resumes less likely to get a callback, affects opportunities and resources)

• Incarceration / legal system targets people of color (Alexander) - systemically reproduces the effect of Jim Crow laws

What is racial bias? How is it different from racism?

• a systematic tendency to favor or disfavor people because of their group. It can be conscious (explicit) or unconscious (implicit), and it operates at the level of an individual's attitudes and behavior.

• Racism: broader than bias. It includes beliefs about racial superiority, but also institutions, laws, and structures that produce racial inequality, whether or not any individual holds animus.

• Alexander (2012) New Jim Crow - mass incarceration of colored people is how bias (ex: overpolicing black/brown neighborhoods) is reproduced into systems (incarceration system)

What is implicit bias? How is it measured?

• Perez (2015) covers implicit attitudes, which are automatic associations absorbed from socialization (media, family, neighborhood, cultural stereotypes) and measured with tools like the Implicit Association Test (IAT).

• IAT test = ppl sort words using paired categories (black/bad, white/good, vice versa) and faster sorting for one pairing signals association

• People can sincerely reject prejudice on a survey and still show automatic association

How does the criminal justice system discriminate against people of color?

• Arrests (higher number of police stops, stop and frisk policies - only need reasonable suspicion, overoccupation of black neighborhoods)

• Prosecution (plea bargains - serve time to get lesser sentences, public defenders who are overworked/underpaid, jury selection tends to be predominantly white due to “elimination of bias”

• Sentencing (Blacks get 10% longer sentences, more likely to get mandatory minimums, more likely to get death sentence for looking black - Eberhardt et al., and covert discrimination - implicit bias of judge)

• Parole (Employment discrimination, stickines of felon label, (social exile and stigma), housing discrimination)

How does the criminal justice system perpetuate patterns of discrimination?

• Old Jim Crow Laws = “separate but equal”, served to segregate / disenfranchise people

• New Jim Crow = produces this same outcome but subtly, felons have similar rights to those under Jim Crow, and people of color are major targets of the criminal justice system

• Employment discrimination, stickines of felon label, (social exile and stigma), housing discrimination 

Panethnicity

• Label that groups multiple distinct ethnic or national groups into a single, broader identity. (ex: Asian - American)

• Ozawa/Thind, where the legal category of "Asian" excluded people from citizenship regardless of their specific origin

Bundle of Sticks

• A way of understanding ethnicity as a bundle of separate attributes (language, religion, region, skin color, ancestry) rather than one fixed trait. Different "sticks" can be emphasized or de-emphasized depending on context

• ex: In Thind, the Court had several "sticks" to choose from (scientific ancestry, skin color, common understanding) and picked whichever excluded him.

Ethnic primordialism

• The view that ethnic identity is ancient, natural, and given by birth (blood ties, deep attachments). It treats ethnic conflict as rooted in unchangeable differences.

• Example: The idea that groups are "naturally" divided is the view Omi and Winant argue against (argues that groups are result of racial formation is response to historical/political situations)

Ethnic instrumentalism

• The view that ethnic identity is a tool people and elites use strategically to gain resources and power, so it can be mobilized or downplayed depending on benefit.

• ex: Lee brings up how identity to politics link is the assumption that ppl who ethnic backgrounds will share political interests, so a politician could ethnically “hop” within a racial category to win more votes

Bias

• A systematic favoring or disfavoring of people based on group membership. It can be conscious or unconscious.

• Example: Bertrand and Mullainathan (2004): identical resumes with white-sounding names received about 50% more callbacks than ones with Black-sounding names

Explicit vs. Implicit cognition

• Explicit cognition is conscious, deliberate, controllable thinking (you know you are doing it); implicit cognition is automatic and outside awareness.

• Perez (2015): people consciously reject prejudice but still show automatic associations on the IAT

• example of implicit: depending on how you were raised, you may cross the street when you see a black person and not even realize

Explicit vs. implicit attitudes

• Explicit attitudes are evaluations you consciously hold and can report. They are what a survey captures. Example: telling a pollster “I don’t have any prejudice.”

• Implicit attitudes are automatic good/bad evaluations activated without intent or awareness. Example: the faster pairing of Black/bad than Black/good on the IAT.

Group Centrism

• Judging politics through the lens of group membership, such as preferring or favoring one's own group and evaluating policies by how they affect "us" versus “them."

• ex: Hutchings and Valentino: White Americans pushing back against welfare b/c they don’t want to support people of color (feel like they ask for too much)

the “New Jim Crow”

• The idea that the incarceration system in the U.S serves as a new form of Jim Crow Laws in the sense that they reproduce discrimination against those of color, as they are targeted by law enforcement, systemically disadvantaged in the legal system, and come out of incarceration with stripped rights, similarly to a colored person during the peak of Jim Crow (Alexander)

• Once labeled a felon, a person faces legal discrimination in jobs, housing, voting, and jury service

• War on drugs: Alexander (2012): drug use is similar across races, yet Black Americans are arrested and imprisoned for drug offenses at far higher rates.

symbolic racism

• a modern prejudice that mixes anti-Black feeling with beliefs in individualism and hard work - we no longer live in an explicitly racist enviornment (mostly) - but racism shows up symbolically through institutions (incarceration system, like judges / jury bias)

• It shows up as beliefs that Black people don't try hard enough or get undeserved advantages, rather than as open claims of inferiority.

Racial Wealth gap

• because of history, there is a significant gap between minority groups wealth

• Wealth passes between generations, so past disadvantages carry forward even after discriminatory laws end.

• redlining = a discriminatory practice in which banks, insurers, and other financial institutions systematically deny loans, mortgages, or services to specific neighborhoods based on their racial or ethnic composition rather than the creditworthiness of individual applicants (lecture)

conceptualization vs. operationalization

• conceptual = The process of defining and clarifying what an abstract concept or construct means in the context of a study (focus of the research, ex: defining success)

• operational = turning abstract conceptual definitions into concrete, empirical, and observable measures (ex: defining success via income)

4 types of variables

• Nominal: categories with no order (race, religion). Matters because you can only compare frequencies and groups, not compute averages.

• Ordinal: ordered categories without equal spacing (strongly disagree to strongly agree). Matters because you can rank but can't assume equal distances.

• Interval: numeric, equal spacing, no true zero (temperature in °F, many survey scales). Matters because you can compute means and differences.

• Ratio: numeric, equal spacing, true zero (income, wealth, number of callbacks). Matters because ratios are meaningful (twice the wealth).

inductive vs. deductive

• inductive = data - analysis - develop theory

• deductive = theory - analysis - hypothesis supported / rejected

types of research designs

• Experiment: researcher randomly assigns treatment. Pro: strong internal validity (causal claims). Con: may be artificial, so lower external validity.

• Observational study: researcher observes without assigning treatment. Pro: realistic, large samples. Con: weak on causation (confounders, selection).

• Case study / ethnography: in-depth study of one or few cases. Pro: rich detail, good for generating theory. Con: hard to generalize.

External, internal, and measurement validity

• internal = How accurately a study proves a cause-and-effect relationship without bias or alternative explanations. ex: Did the independent variable truly cause the change in the dependent variable?"

• external = How well study results generalize or apply to other people, settings, times, and real-world environments.

• measurement = How accurately a tool, test, or scale measures what it claims to measure

element of a hypothesis

• A good hypothesis is testable, states a relationship between an independent variable (IV, the cause) and a dependent variable (DV, the outcome), and specifies direction.

• Example: "White-sounding names (IV) increase the likelihood of a resume callback (DV)." Direction: positive effect for white-sounding names.

Types of research (Shively 2013, research classification)

• applied - aimed at solving problems

• recreational - pursued for its own sake

• empirical - discovers new facts, and nonempircal doesn’t

• normative theory = non-empirical/applied (what ought to happen)

• engineering research = empirical / applied

• positive theory = non-empirical / recreational

• theory-oriented = empirical / recreational

Qualitative vs. quantitative

• Quantitative: numerical data, statistical analysis, large samples (callback rates, survey regressions, IAT scores).

• Qualitative: words, observations, interpretation, smaller samples (ethnography, interviews).

• KKV (1994) argue both share a unified logic of inference, so both can make causal and descriptive claims if designed well. Wedeen (2010) defends ethnography and interpretive work as valuable in political science.