Log in

Racial Invariance or Asian Advantage: Comparing the Macro-Level Predictors of Violence Across Asian, White, and Black Populations

  • Published:
Race and Social Problems Aims and scope Submit manuscript

Abstract

Research shows that structural disadvantage is a key source of violent crime rates across racial/ethnic groups, a finding that has become more commonly known as “racial invariance.” However, this literature has focused primarily on white, black and Latino comparisons, with little attention to Asian populations. This omission is problematic considering that (1) Asians are the fastest growing minority group in the U.S. and (2) the sources of Asian crime could differ from those of white and black populations. Drawing on the racial invariance hypothesis, the current study uses 2010 city-level data to compare the structural predictors of violent crime arrest rates (homicide, robbery, rape, and aggravated assault) for white, black, and Asian populations. Findings reveal that disadvantage contributes to violence for all three racial/ethnic groups, but the magnitude of these effects and effects of other structural predictors differ. Findings from the current study offer implications for the racial invariance debate.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Subscribe and save

Springer+ Basic
EUR 32.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or Ebook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price includes VAT (Germany)

Instant access to the full article PDF.

Similar content being viewed by others

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. Data were obtained from the following publicly available sources: (1) the FBI Uniform Crime Reporting Program (UCR), (2) the 2010 U.S. Census, and (3) the Brown University American Communities Project.

Code Availability

The coding generated during and used during the current study are available from the corresponding author on reasonable request.

Notes

  1. Although the term Asian includes several nationalities each with their own distinct culture and language, the current study examines Asian populations as a singular group due to data limitations. This is a common practice in analyses of Asian populations but a noteworthy caveat (see Harrel (2009), Logan and Zhang (2013). Thus, we are unable to disaggregate findings and effects across Asian sub-groups and national origins. UCR arrest data also does not account for ethnicity. This limitation confounds racial comparisons drawn from the current analyses somewhat because of its inability to separate effects of ethnicity.

  2. The initial sample included 29,338 cities before a selection criteria was applied. Upon inspection of the data, most of these cities had small populations. We then applied the first filter (≥ 50,000) to protect against skew data from smaller cities. This produced a sample of 688 cities. Next, we looked at Asian arrest figures. Following previous research, a second filter was applied to capture cities with stable Asian arrest rates (≥ 1000 Asian residents). White and black arrest figures were stable and did not pose instability threats.

  3. Highly-aggregated study units are less able to capture local, neighborhood-level effects on crime across race/ethnicity (a point we return to in the Discussion). However, as Sampson (2013, p. 7) noted, “[t]he phenomenon of crime does not privilege any one type of place or ecological unit either.” In light of these points, we believe that cities offer a well-established and appropriate study unit for the current analysis and research questions.

  4. Harrell (2009) found that Asians reported violent crimes at a similar rate to whites and Hispanics (roughly 49%). However, Asians were less likely to report simple aggravated assault (37%) compared to whites (43%), blacks (49%), Hispanics (43%), and American Indians (58%). Asians reported robbery at a significantly higher rate than any other racial group.

  5. The isolation index is one of the most commonly used and well-established racial segregation measure in demographic research (Massey & Denton, 1988a; 1988b). An isolation index taps into “the extent to which minority members are exposed only to one another” (Massey & Denton, 1988b, p. 288). An isolation index suits the study’s focus because it captures the polarization of a racial group from the total population rather than the “evenness” of racial distributions across space as measured by the index of dissimilarity (Feldmeyer, 2010). However, it is susceptible to size. Smaller racial locales yield a smaller isolation index and vice versa. To protect against this limitation, we restrict the study sample to focus on large urban cities. Supplemental analyses using a dissimilarity index produced similar results.

  6. This finding was surprising because prior research has shown Asian education levels to be similar (or greater) than whites (Zeng & **e, 2004). Close inspection of the data reveal that the lower Asian education levels reflect cities with small Asian populations and low Asian graduation rates, which drives the mean Asian education level downward. In contrast, cities with the largest Asian populations and most established Asian communities had much higher Asian graduation rates. In light of this finding, we conducted supplemental analyses using different selection criteria (≥ 10,000 Asians) and the effects were substantially the same. Although we are unable to determine the precise source of lower Asian graduation rates in some cities, it may stem from different Asian immigration patterns across locales.

  7. SUR models protect against highly correlated coefficients that may produce biased results when comparing violent predictors across different race groups from the same sample of aggregated study units (see Feldmeyer, 2010; Ousey, 1999; Steffensmeier et al., 2010). However, the tradeoff is that SUR models in standard statistical packages do permit use of negative binomial estimates as well as the “suest” command (in STATA) to compare coefficients across groups.

References

Download references

Funding

Not applicable.

Author information

Authors and Affiliations

Authors

Contributions

Conceptual/theoretical development: DS and BF. Research/methodological design: DS and BF. Data collection/coding/analysis: DS. Manuscript drafting: DS. Critical analytical/writing revisions: DS and BF. Managing/supervising the project: DS and BF.

Corresponding author

Correspondence to Diana Sun.

Ethics declarations

Conflict of interest

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sun, D., Feldmeyer, B. Racial Invariance or Asian Advantage: Comparing the Macro-Level Predictors of Violence Across Asian, White, and Black Populations. Race Soc Probl 14, 114–130 (2022). https://doi.org/10.1007/s12552-021-09344-1

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12552-021-09344-1

Keywords

Navigation