Skip to main content

Advertisement

SpringerLink
Log in

The Age Structure-Crime Rate Relationship: Solving a Long-Standing Puzzle

  • Original Paper
  • Published:
Journal of Quantitative Criminology Aims and scope Submit manuscript

Abstract

Objectives

Develop the concept of differential institutional engagement and test its ability to explain discrepant findings regarding the relationship between the age structure and homicide rates across ecological studies of crime. We hypothesize that differential degrees of institutional engagement—youths with ties to mainstream social institutions such as school, work or the military on one end of the spectrum and youths without such bonds on the other end—account for the direction of the relationship between homicide rates and age structure (high crime prone ages, such as 15–29).

Methods

Cross sectional, Ordinary Least Squares regression analyses using robust standard errors are conducted using large samples of cities characterized by varying degrees of youths’ differential institutional engagement for the years 1980, 1990 and 2000. The concept is operationalized with the percent of the population enrolled in college and the percent of 16–19 year olds who are simultaneously not enrolled in school, not in the labor market (not in the labor force or unemployed), and not in the military.

Results

Consistent and invariant results emerged. Positive effects of age structure on homicide rates are found in cities that have high percentages of disengaged youth and negative effects are found among cities characterized with high percentages of youth participating in mainstream social institutions.

Conclusions

This conceptualization of differential institutional engagement explains the discrepant findings in prior studies, and the findings demonstrate the influence of these contextual effects and the nature of the age structure-crime relationship.

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

Notes

  1. Although we do not engage in the debates about the individual-level variations in the age-crime relationship, these studies represent a substantively crucial aspect of this covariate of crime. For an extensive discussion, see Piquero et al. (2003) and Piquero (2008). Nevertheless, scholars engaging in this debate would find the implications of the research herein relevant.

  2. Although we choose to focus on control-based theoretical developments, other theoretical perspectives may be applied to the logic underlying differential institutional engagement.

  3. Published volumes for the latter two decennial periods were referenced to determine when missing homicide data for cities actually represented zero homicides. Furthermore, a few cases were retained by ensuring that at least two years of data comprised the homicide rate by retrieving data from the five year period circa each decennial period.

  4. Cronbach’s alpha scores for the economic deprivation index were .63, .62 and .58 for 1980, 1990 and 2000, respectively. Alpha scores were greater than .8 with median family income omitted.

  5. One model in Table 2 (for 1990) exhibited a VIF approaching 7, and two models in Table 4 (for years 1990 and 2000) produced a VIF of 5.6—all three associated with the economic deprivation index which is highly correlated with the unemployment rate (approximately .8). Nevertheless, parameter estimates for all three regressors, economic deprivation, Gini and unemployment rate, were statistically significant in the posited direction in those models. To ensure the reliability of these findings, the models were also estimated after excluding the unemployment rate and the results were compared with the original model specification. There were few substantive differences between the models and particularly none with regard to the percent young variable. No other VIF values exceeded 5.

  6. Additional analyses were conducted to determine whether including percent black in the model affected the regression results. Principle components analyses indicated that percent black loaded with the economic deprivation index, therefore an alternative deprivation index was created that also included percent black. Few parameter estimates from regression analysis of this model specification were substantively different from those that excluded percent black from the model. In particular, there were no substantive differences with regard to the age structure measure. Results from these analyses are available upon request.

  7. The parameter estimates presented in the tables were estimated using Stata 11.0’s regression command with the “robust” option. These models were also estimated with the “vce (cluster)” option specifying a state identifier which allows for intragroup correlation using the state within which the cities are located as the group. The results from this estimation technique showed no substantive differences in estimates for the age structure coefficient. Furthermore, the 1980 models were also estimated using negative binomial regression techniques because there were a substantial number of zero homicide cases in that year and because tests for over dispersion in Poisson regression analyses indicated the assumptions of equal mean and variance were not met. The results of these analyses differed little from the regression results reported in the tables and none produced substantive differences for the age structure measure. Results of these supplemental regression analyses are available upon request.

  8. The mean economic deprivation index values for 1990 and 2000 are .24 and .25, respectively when considering those cities for which data are available for the 1980 measures (in comparison with 1980’s mean of .33). This difference is largely due to missing homicide rate data for smaller cities in 1980. For example, when restricting the sample to those cases for which data are available in 1980 the component values are only slightly higher than those shown in this table for 1990 and 2000.

  9. In cross-sectional analyses, age effects cannot be distinguished from cohort effects. Accordingly, the relationship between age structure and homicide rates cannot be disentangled from that of cohort and homicide rates in the present study.

  10. The analyses were replicated using two other model specifications to examine the notion that the West region of the US should be the focus of regional controls rather than the South. One set of nine models replaced South with West region dummy variable and the other set of nine models included three region dummies: West, South, and Northeast. Fifteen of the 18 models upheld the age-structure-crime relationship reported in these analyses. When substituting the West region for the South, one of the nine coefficients for age structure changed to not statistically significant—the engaged youth model for 2000 with 158 cases (p = .283). After including region dummy measures for South, West and Northeast in the models, two of the nine coefficients for age structure changed to not statistically significant—both engaged and disengaged youth models for 2000, one with 158 cases (p = .237) and one with 380 cases (p = .360). Furthermore, the West region dummy variable was negatively correlated with the homicide rate when statistically significant (13 of 18 models) as opposed to positively associated. Because of this and to maintain consistency with the majority of ecological studies of homicide which control for southern region, we report the findings for the model specifications using the South dummy variable for region.

  11. Blalock (1979: 163) states: “If one is interested in showing a theory to be correct, one will make significance tests only when results occur in the predicted direction. If they occur in the opposite direction, one need make no test since the data obviously do not support the theory anyway.” Accordingly, while we report the estimated t statistics for the b%15–29 coefficients for 1980, 1990, and 2000 in Table 2, we do not use them in a formal test.

  12. McCall et al. (2010) found that the percent young age structure covariate was not statistically significant for a city-level analysis of 1970 data. Because there are no data for measures of disengaged youths in 1970, the present analyses are restricted to 1980, 1990, and 2000.

  13. The institutional engagement mechanism may have broad applicability: Simon (2011) has linked recent cutbacks in youth services, and the mainstream institutional involvements that go therewith, in the United Kingdom with the youth riots that occurred in London and other UK cities in early August 2011. According to his account (Simon 2011: 40), “Youth clubs have already closed, youth workers have been sacked, and programs that in previous years have occupied urban youngsters in the long summer break are not running. As a result, many young people have ‘nothing to do.’”

  14. Persons outside the ages analyzed here are also expected to be influenced by differential institutional engagement; however, we focus on the youth population for two reasons. First, the empirical ambiguities we seek to clarify center on the inconsistent findings of the relationship between percent youth population and homicide rates. Second, prior research by Warr (1998, 2002) indicates that youth are more likely to be in closer physical and social proximity to other institutionally disengaged youth due to their life course stage.

References

  • Anderson E (1999) Code of the street: decency, violence, and the moral life of the inner city. W.W. Norton & Company, New York

    Google Scholar 

  • Blalock HM Jr (1979) Social statistics: revised, 2nd edn. McGraw-Hill, New York

    Google Scholar 

  • Brown E, Males M (2011) Does age or poverty level best predict criminal arrest and homicide rates? A preliminary investigation. West Policy J 8(1):1–30

    Google Scholar 

  • Bursik RJ Jr, Grasmick HG (1993) Neighborhoods and crime. Lexington, New York

    Google Scholar 

  • Cohen LE, Land KC (1987) Age structure and crime: symmetry versus asymmetry and the projection of crime rates through the 1990s. Am Sociol Rev 52:170–183

    Article  Google Scholar 

  • Crutchfield RD, Geerken MR, Gove WR (1982) Crime rates and social integration: the impact of metropolitan mobility. Criminology 20(3):467–478

    Article  Google Scholar 

  • DeFronzo J (1983) Economic assistance to impoverished Americans. Criminology 21(1):119–136

    Article  Google Scholar 

  • Fox JA, Piquero AR (2003) Deadly demographics: population characteristics and forecasting homicide trends. Crime Delinq 49(3):339–359

    Article  Google Scholar 

  • Greenberg DF (1985) Age, crime and social explanation. Am J Sociol 91(1):1–21

    Article  Google Scholar 

  • Hirschi T (1969) The causes of delinquency. University of California Press, Berkeley

    Google Scholar 

  • Hirschi T, Gottfredson M (1983) Age and the explanation of crime. Am J Sociol 89:552–584

    Article  Google Scholar 

  • Huff-Corizine L, Corzine J, Moore DC (1986) Southern exposure: deciphering the south’s influence on homicide rates. Soc Forces 64:906–924

    Google Scholar 

  • Land KC, McCall PL, Cohen LE (1990) Structural covariates of homicide rates: are there any invariances across time and social space? Am J Sociol 95:922–963

    Article  Google Scholar 

  • Laub J, Sampson RJ (1993) Turning points in the life course: why change matters to the study of crime. Criminology 31:301–325

    Article  Google Scholar 

  • Laub JH, Sampson RJ (2003) Shared beginnings, divergent lives: delinquent boys to age 70. Harvard University Press, Cambridge, MA

    Google Scholar 

  • Laub J, Nagin D, Sampson RJ (1998) Trajectories of change in criminal offending: good marriages and the desistance process. Am Sociol Rev 63:225–238

    Article  Google Scholar 

  • Lee MT, Martinez R Jr (2002) Social disorganization revisited: mapping the recent immigration and black homicide relationship in northern Miami. Sociol Focus 35:363–380

    Article  Google Scholar 

  • Lee MT, Slack T (2008) Labor market conditions and violent crime across the metro-nonmetro divide. Soc Sci Res 37:753–768

    Article  Google Scholar 

  • Lee MT, Martinez R Jr, Rosenfeld R (2001) Does immigration increase homicide? Negative evidence from three border cities. Sociol Q 42(4):559–580

    Article  Google Scholar 

  • Loftin C, Hill RH (1974) Regional culture and homicide: an examination of the Gastil-Hackney thesis. Am Sociol Rev 39:714–724

    Article  Google Scholar 

  • Loftin C, Parker RN (1985) An errors-in-variable model of the effect of poverty on urban homicide rates. Criminology 23(2):269–285

    Article  Google Scholar 

  • Martinez R Jr (2002) Latino homicide: immigration, violence, and community. Routledge press, Taylor & Francis Group, New York

    Google Scholar 

  • Marvell TB, Moody CE (1991) Age structure and crime rates: the conflicting evidence. J Quant Criminol 7(3):237–273

    Article  Google Scholar 

  • Massey DS, Denton NA (1988) Suburbanization and segregation in United States metropolitan areas. Am Sociol Rev 94:592–626

    Article  Google Scholar 

  • Maxwell CD, Garner JH, Skogan WG (2011) Collective efficacy and criminal behavior in Chicago, 1995–2004. U.S. Department of Justice. Retrieved 16 Aug 2011. https://www.ncjrs.gov/pdffiles1/nij/grants/235154.pdf

  • McCall PL, Land KC, Parker KF (2010) An empirical assessment of what we know about structural covariates of homicide rates: a return to a classic 20 years later. Homicide Stud 14(3):219–243

    Article  Google Scholar 

  • McCall PL, Land KC, Parker KF (2011) Heterogeneity in the rise and decline of city-level homicide rates, 1976–2005: a latent trajectory analysis. Soc Sci Res 40:363–378

    Article  Google Scholar 

  • Messner SF (1983a) Regional and racial effects on the urban homicide rates: the subculture of violence revisted. Am J Sociol 88:997–1007

    Article  Google Scholar 

  • Messner SF (1983b) Regional differences in the economic correlates of the urban homicide rate: some evidence on the importance of cultural context. Criminology 21:477–488

    Article  Google Scholar 

  • Morenoff JD, Sampson RJ, Raudenbush SW (2001) Neighborhood inequality, collective efficacy, and the spatial dynamics of urban violence. Criminology 39(3):517–558

    Article  Google Scholar 

  • Nagin D, Land KC (1993) Age, criminal careers, and population heterogeneity: specification and estimation of a nonparametric, mixed poisson model. Criminology 31(3):327–362

    Article  Google Scholar 

  • Neilson AL, Martinez R Jr, Lee MT (2005) Alcohol, ethnicity, and violence: the role of alcohol availability and other community factors for group-specific non-lethal violence. Sociol Q 46:477–500

    Google Scholar 

  • Pampel FC, Gartner R (1995) Age structure, socio-political institutions, and national homicide rates. Eur Sociol Rev 11:243–260

    Google Scholar 

  • Parker RN (1989) Poverty, subculture of violence, and type of homicide. Soc Forces 67(4):983–1007

    Google Scholar 

  • Parker KF, McCall PL, Land KC (1999) Determining social structural predictors of homicide: unit of analysis and other methodological concerns. In: Smith MD, Zahn MA (eds) Homicide: a sourcebook of social research. Sage Publication, Thousand Oaks

    Google Scholar 

  • Paternoster R, Brame R, Mazerolle P, Piquero A (1998) Using the correct statistical test for the equality of regression coefficients. Criminology 36(4):859–866

    Article  Google Scholar 

  • Phillips JA (2006) The relationship between age structure and homicide rates in the United States, 1970 to 1999. J Res Crime Delinq 43:230–260

    Article  Google Scholar 

  • Piquero AR (2008) Taking stock of developmental trajectories of criminal activity over the life course. In: Liberman AM (ed) The long view of crime: a synthesis of longitudinal research. Springer, New York

    Google Scholar 

  • Piquero AR, Farrington DP, Blumstein A (2003) The criminal career paradigm. In: Tonry M (ed) Crime and justice: a review of research, vol 30. University of Chicago Press, Chicago

    Google Scholar 

  • Pratt TC, Cullen FT (2005) Assessing macro-level predictors and theories of crime: a meta-analysis. In: Tonry M (ed) Crime and justice: a review of research, vol 32. University of Chicago Press, Chicago, pp 373–450

  • Sagi PC, Wellford CF (1968) Age composition and patterns of change in criminal statistics. J Criminol Law Criminol Police Sci 59(1):29–36

    Article  Google Scholar 

  • Sampson RJ (1988) Local friendship ties and community attachment in mass society: a multi-level systemic model. Am Sociol Rev 53:766–779

    Article  Google Scholar 

  • Sampson RJ, Groves WB (1989) Community structure and crime: testing social disorganization theory. Am J Sociol 94:774–802

    Article  Google Scholar 

  • Sampson RJ, Laub JH (1990) Crime and deviance over the life course: the salience of adult social bonds. Am Sociol Rev 55:609–627

    Article  Google Scholar 

  • Sampson RJ, Morenoff JD, Earls F (1999) Beyond social capital: spatial dynamics of collective efficacy for children. Am Sociol Rev 64:633–660

    Article  Google Scholar 

  • Sampson RJ, Raudenbush SW, Earls F (1997) Neighborhoods and violent crime: a multilevel study of collective efficacy. Science 277(5328):918–924

    Article  Google Scholar 

  • Shaw CR, McKay HD (1942) Juvenile delinquency in urban areas: a study of rates of delinquency in relation to differential characteristics of local communities in American cities. University of Chicago Press, Chicago

    Google Scholar 

  • Shihadeh ES, Thomas SA (2007) Institutional attachment and violence: the concentration of youth disengagement and serious crime. Paper presented at the 2007 annual meeting of the American Society of Criminology

  • Shryock HS, Siegel J et al (1976) The methods and materials of demography. U.S. Government Printing Office, Washington, DC

    Google Scholar 

  • Simon S (2011) A violent convulsion of kids on holiday from high school. Newsweek. 22 and 29 Aug 2011:37–41

  • Steffensmeier D, Harer MD (1987) Is the crime rate really falling? J Res Crime Delinq 24:23–48

    Article  Google Scholar 

  • Steffensmeier D, Harer MD (1991) Did crime rise or fall during the Reagan Presidency? The effects of an ‘aging’ U.S. population on the nation’s crime rate. J Res Crime Delinq 28:330–359

    Article  Google Scholar 

  • Warr M (1998) Life-course transitions and desistance from crime. Criminology 36:183–216

    Article  Google Scholar 

  • Warr M (2002) Companions in crime: the social aspects of criminal conduct. Cambridge University Press, Cambridge

    Book  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Sociology and Anthropology, North Carolina State University, Box 8107, Raleigh, NC, USA

    Patricia L. McCall & Cindy Brooks Dollar

  2. Department of Sociology and Center for Population Health and Aging, Duke University, Durham, NC, USA

    Kenneth C. Land

  3. Department of Sociology and Criminology, University of Delaware, Newark, DE, USA

    Karen F. Parker

Authors
  1. Patricia L. McCall
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Kenneth C. Land
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Cindy Brooks Dollar
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Karen F. Parker
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Patricia L. McCall.

Additional information

Paper presented at the American Society of Criminology meetings, San Francisco, 2010.

Appendices

Appendix 1: Data Definitions and Sources

The data source from which most covariates were collected is Sociometrics’ CDA from the 1970, 1980, and 1990 Census Extract Data (1998). The 2000 data (not available from CDA) were derived online from the US Bureau of the Census American Fact Finder’s 2000 STF3 detailed tables (Census); and a few measures for 1980 and 1990 were also obtained from the US Bureau of the Census. Other covariates were collected from the Minnesota Population Center’s NHGIS. The variables in these analyses were obtained from sources specified below. More specific information is available upon request from the authors.

Data Definitions

Homicide rate: (Number of murder and non-negligent manslaughter offenses/total resident population) × 100,000. Source: FBI and/or Fox 2008, Victim file.

Population size: Number of total resident population. Source: CDA, Census.

Population per square mile: (Total population/land area in square miles). Source: CDA, Census.

Population structure: (z-score of Population size + z-score of Population per square mile).

Economic Deprivation: (factor-score − weighted z-scores of following three measures).

Percentage of families with children that are female headed families: (Female single parent households with children/(Married couple families with children + Male single parent households with children + Female single parent households with children)) × 100. Source: CDA, Census P046.

Median family income (in 2000 constant dollars): Source: Census (1980, 2000), NHGIS (1990).

Percentage of families living below the official poverty level: Source: Census (1980, 2000), NHGIS (1990).

Gini index of income concentration for families: For 1980, 17 category family income distribution used; for 1990, 25 category family income distribution used; and for 2000, 16 category family income distribution used to compute the Gini Index of Income Concentration:

Gi = (∑XiYi+1) − (∑Xi+1Yi) where Xi and Yi are respective cumulative percentage distributions; (Shryock et al. 1976:98–100). Source: CDA, Census.

Percentage divorced males: (Number divorced males/number males 16 years old and over) × 100. Source: NHGIS.

Percentage of the population 15–29 years of age: (Number of 15–29 year olds/total resident population) × 100. Source: NHGIS.

Unemployment rate: (Number employed in civilian labor force/number in civilian labor force) × 100. Source: CDA, Census.

Ratio of White to Black per capita income: (Per capita income for Whites/Per capita income for Blacks). Source: NHGIS.

Dissimilarity index: ∑[ti|pi − P|/2TP(1 − P)], computed using Stata’s “seg …, d” command. Source: CDA, Census; Massey and Denton (1988:284).

Male/Female sex ratio: (Males age 16–34/Females ages 16–34) × 100. Source: CDA, Census.

Heterogeneity index: (1 − (Proportion of population non-Latino Whites + Proportion non-Latino Blacks + Proportion Latinos)). Source: CDA, Census.

Percent foreign born: (Foreign born population/total resident population) × 100. Source: CDA, Census.

Percent enrolled in college: (College enrollment, private and public/Total resident population) × 100. Source: NHGIS (1980 & 1990), Census (2000).

Disengaged youth: ((High school grad, not in labor force, ages 16 to 19 + High school grad, Unemployed, ages 16 to 19 + Non-high school grad, not in labor force, ages 16 to 19 + Non-high school grad, Unemployed, ages 16–19)/Population ages 16–19) × 100. Source: NHGIS (1980 & 1990), Census (2000).

South region: Dummy variable for southern geographic location as defined by US Census bureau. Source: Census.

Sources

Fox, J. A. Uniform Crime Reports [United States]: Supplementary Homicide Reports With Multiple Imputation, Cumulative Files 1976–2005 [Computer file]. ICPSR22161-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2008-09-18. doi:10.3886/ICPSR22161.

Minnesota Population Center. (2004). National Historical Geographic Information System: Pre-release Version 0.1. Minneapolis, MN: University of Minnesota, http://www.nhgis.org.

Sociometrics. (1998). Contextual Data Archive from the 1970, 1980, and 1990 Census Extract Data (1998), National Opinion Research Center, University of Chicago, Chicago, IL.

US Department of Justice, Federal Bureau of Investigation. (Various years). Uniform Crime Reports or Crime in the United States. Washington, D.C.: Government Printing Office.

US Bureau of the Census. (Various years). Census of Population. Vols. 1 and 2. Characteristics of the Population. Washington, D.C.: Government Printing Office. (State volumes).

Appendix 2

See Table 5.

Table 5 Descriptive statistics for sample restricted to cities with greater than the mean disengaged youth and to cities with less than the mean percentage of population enrolled in college
Full size table

Appendix 3

See Table 6.

Table 6 Descriptive statistics for sample restricted to cities with greater than the mean percentage of population enrolled in college and to cities with less than the mean disengaged youth
Full size table

Rights and permissions

Reprints and permissions

About this article

Cite this article

McCall, P.L., Land, K.C., Dollar, C.B. et al. The Age Structure-Crime Rate Relationship: Solving a Long-Standing Puzzle. J Quant Criminol 29, 167–190 (2013). https://doi.org/10.1007/s10940-012-9175-9

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10940-012-9175-9

Keywords

Search

Navigation