Abstract
Criminology can easily be characterized by its investigation of change. We need to understand the conditions that facilitate the change in order to inform policy makers on how to reduce crime or improve social welfare. Yet, much of the published research in our field relies on cross-sectional data. As most criminological research questions are inherently dynamic, criminologists have more recently adopted the methods of analyzing changes over time. This chapter introduces a set of methodological choices to estimate the effects of changes in an independent variable on a dependent variable. It begins by outlining several methodological options when the scholar has repeated the measures of a single unit. Then, several other options for a scenario in which the data include many units that are repeatedly measured, are discussed. The chapter concludes with a discussion designed to guide the readers’ methodological decisions while analyzing dynamic data.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
If the model includes a lagged dependent variable, the Durbin–Watson statistic will be biased toward no autocorrelation. Turn to Greene (2008: 646) for an adjustment to the Durbin–Watson statistic for the lagged dependent variable. The other two tests produce unbiased estimates regardless of whether the lagged dependent variable is included in the model.
- 2.
Autocorrelation can also be problematic in panel data. See Greene (2008) for more detail.
- 3.
This approached used by Cantor and Land (1985) and the others has been criticized by Greenberg (2001). Greenberg explains that nonstationary crime rates are more likely explained by variables that also follow nonstationary processes (unemployment is stationary). Greenberg (2001) shows that while crime and divorce are each nonstationary, when combined they follow a stationary process, suggesting that they are cointegrated (Banerjee et al. 1993). By using error correction models that account for cointegration instead of relying on first- or second-differencing, long term effects can be estimated. See Britt (2001) for continued discussion in this debate.
- 4.
Most standard statistical packages run ARIMA models.
- 5.
- 6.
Some of correlations in the ACF can still be large in a stationary series; however, they will follow a pattern consistent with an autoregressive or moving average process.
- 7.
For simplicity, the ARIMA(1,0,0) is written as if the series needed no differencing. The text could also refer a differenced ARIMA(1,d,0) process.
- 8.
Most standard statistical packages will produce ACF and PACF tables. For example, the Stata command corrgram will produce a table the autocorrelations, partial autocorrelations, and the Portmantequ (Q) statistics for time series data. The commands AC and PAC produce a graphical version of the ACF and PACF values.
- 9.
See Cook and Campbell (1979: 247–250) for a more thorough discussion of mixed models.
- 10.
By repeated substitution, autoregressive processes can take a moving average form (Greene 2008).
- 11.
He finds that Granger causality is unidirectional from money stock to income, dismissing arguments that changes in income lead to changes in money stock.
- 12.
This system works better if the variables are stationary. This can be easily determined using the augmented Dickey–Fuller test, which is found in most statistical packages. If the data are nonstationary, it is usually recommended that the analyst estimate Vector Error Correction Models (Stata Press 2005; Greenberg 2001). However, others suggest that there is still value in estimating VAR if the short-term dynamics are the main focus of the analysis (Brandt and Williams 2007).
- 13.
This notation is borrowed from Brandt and Williams (2007).
- 14.
Stata has a post estimation command vargranger that performs pairwise Granger causality tests after running VAR (Stata Press 2005).
- 15.
This decomposition is done by first inverting the VAR to its Vector Moving Average representation (VMA) and then decomposing the error covariance matrix. This method is sensitive to the ordering of the variables and all orderings should be considered (Brandt and Williams 2007). The Stata post estimation command fevd (forecast-error variance decompositions) calculates these values.
- 16.
This method also relies on inverting the VAR to its VMA representation and then decomposing the error variance matrix. Once again, all orderings should be considered (Brandt and Williams 2007). The Stata post estimation command irf (impulse-response functions) produce these graphs.
- 17.
The notation for all equations is borrowed from Greene (2008).
- 18.
We could easily add another term, { T} ′ t γ, to represent those variables that change over time, but are stable across units.
- 19.
Note that we could also add a time specific constant that captures variation that is constant across units, but varies over time.
- 20.
Exceptions are found in the more recent population heterogeneity versus state dependency literature, which will be discussed below.
- 21.
Their method differs from that used by Apel and colleagues (2008). Here they incorporated IV estimated with a generalized-method of moments framework.
- 22.
Multilevel modeling can be accomplished using a variety of statistical software packages such as Stata, SAS and SPSS. The packages HLM, MLwinN, aML, and WINBUGS were explicitly designed to estimate multilevel models.
References
Apel R, Bushway SD, Paternoster R, Brame R, Sweeten G (2008) Using state child labor laws to identify the causal effect of youth employment on deviant behavior and academic achievement. J Quant Criminol 24(4):337–362
Banerjee A, Dolado JJ, Galbraith JW, Hendry DF (1993) Co-integration, error correction, and the econometric analysis of non-stationary data. Oxford University Press, Oxford
Brandt PT, Williams JT (2007) Multiple time series models. Sage Publications, Inc., Thousand Oaks, CA
Britt CL (2001) Testing theory and the analysis of time series data. J Quant Criminol 17(4):343–357
Britt CL (1997) Reconsidering the unemployment and crime relationship: variation by age group and historical period. J Quant Criminol 13(4):405–428
Britt CL (1994) Crime and unemployment among youths in the United States, 1958–1990: a time series analysis. Am J Econ Sociol 53(1):99–109
Bryk AS, Raudenbush SW (1992) Hierarchical linear models. Sage Publications, Inc., Newbury Park, CA
Bushway S, Brame R, Paternoster R (1999) Assessing stability and change in criminal offending: a comparison of random effects, semiparametric, and fixed effects modeling strategies. J Quant Criminol 15(1):23–61
Cantor D, Land KC (1985) Unemployment and crime rates in the post-World War II United States: a theoretical and empirical analysis. Am Sociol Rev 50:317–332
Chamlin MB, Cochran JK (1998) Causality, economic conditions, and burglary. Criminology 36(2):425–440
Cochran JK, Chamlin MB, Seth M (1994) Deterrence or brutalization? An impact assessment of Oklahoma’s return to capital punishment. Criminology 32(1):107–134
Cohn EG, Rotton J (1997) Assault as a function of time and temperature: a moderator-variable time-series analysis. J Pers Soc Psychol 72(6):1322–1334
Cook TD, Campbell DT (1979) Quasi-experimentation design & analysis issues for field settings. Houghton Mifflin Company, Boston
D’Alessio SJ, Stolzenberg L (1995) The impact of sentencing guidelines on jail incarceration in Minnesota. Criminology 33(2):283–302
Devine JA, Sheley JF, Smith MD (1988) Macroeconomic and social-control-policy influences on crime rate changes, 1948–1985. Am Sociol Rev 53:407–420
Dugan L (2009) Series hazard modeling: an extension to the cox proportional hazard model to estimate intervention effects on recurrent events for a single unit. Unpublished manuscript. The University of Maryland
Dugan L, Huang J, LaFree G, McCauley C (2009) The Armenian secret army for the liberation of Armenia and the justice commandos of the Armenian genocide. Dynamics of Asymmetric Conflict (forthcoming)
Dugan L, LaFree G, Piquero A (2005) Testing a rational choice model of airline hijackings. Criminology 43(4): 1031–1066
Dugan L, Nagin D, Rosenfeld R (2003) Exposure reduction or retaliation? The effects of domestic violence resources on intimate partner homicide. Law Soc Rev 27(1):169–198
Dugan L, Nagin D, Rosenfeld R (1999) Explaining the decline in intimate partner homicide: the effects of changing domesticity, women’s status, and domestic violence resources. Homicide Stud 3(3):187–214
Durbin J (1954) Errors in variables. Rev Int Stat Inst 22:23–32
Enders W, Sandler T (1993) The effectiveness of antiterrorism policies: a vector-autoregression-intervention analysis. Am Polit Sci Rev 87(4):829–844
Freemen J, Houser D, Kellstedt PM, Williams JT (1998) Long-memoried processes, unit roots, and causal inference in political science. Am J Pol Sci 42(4):1289–1327
Goldkamp JS, Vilcica ER (2008) Targeted enforcement and adverse system side effects: the generation of fugitives in Philadelphia. Criminology 46(2):371–409
Granger CWJ (1969) Investigating causal relations by economic models and cross-spectral methods. Econometrica 37(3):424–438
Greenberg DF (2001) Time series analysis of crime rates. J Quant Criminol 17(4):291–327
Greene WH (2008) Econometric analysis, 6th edn. Pearson Prentice Hall, Upper Saddle River, NJ
Grubesic TH, Mack EA (2008) Spatio-temporal interaction of urban crime. J Quant Criminol 24(3):285–306
Hansen BE (2001) The new econometrics of structural change: dating breaks in U.S. labor productivity. J Econ Perspect 15(4):117–128
Hausman J (1978) Specification tests in econometrics. Econometrica 46:1251–1271
Hausman JA, Taylor WE (1981) Panel data and unobservable individual effects. Econometrica 49(6):1377–1398
Jacobs D, Helms RE (1996) Toward a political model of incarceration: a time-series examination of multiple explanations for prison admission rates. Am J Sociol 102(2):323–357
Johnson SD, Bernasco W, Bowers KJ, Elffers H, Ratcliffe J, Rengert G, Townsley M (2007) Space-time patterns of risk: a cross national assessment of residential burglary victimization. J Quant Criminol 23(3):201–219
Kaminski RJ, Marvell TB (2002) A comparison of changes in police and general homicides: 1930–1998. Criminology 40(1):171–190
LaFree G (2005) Evidence for elite convergence in cross-national homicide victimization trends, 1956–2000. Sociol Q 45:191–211
LaFree G (1999) Declining violent crime rates in the 1990s: predicting crime booms and busts. Annu Rev Sociol 25(1):145–68
LaFree G, Drass KA (2002) Counting crime booms among nations: evidence for homicide victimization rates, 1956 to 1998. Criminology 40(4):769–800
LaFree G, Drass KA (1996) The effect of changes in intraracial income inequality and educational attainment on changes in arrest rates for African Americans and whites. Am Sociol Rev 61(3):614–634
LaFree G, Dugan L, Korte R (2009) Is counter terrorism counterproductive? Northern Ireland 1969–1992. Criminology 47(1):501–530
Levitt SD (2001) Alternative strategies for identifying the link between unemployment and crime. J Quant Criminol 17(4):377–390
Loftin C, Heumann M, McDowall D (1983) Mandatory sentencing and firearms violence: evaluating an alternative to gun control. Law Soc Rev 17(2):287–318
Maddala GS (1987) Limited dependent variable models using panel data. J Hum Resour 22(3):307–338
Marsh LC, Cormier DR (2002) Spline regression models. Sage Publications, Inc., Thousand Oaks, CA
McDowall D (2002) Tests of nonlinear dynamics in U.S. homicide time series, and their implications. Criminology 40(3):711–735
McDowall D, McCleary R, Meidinger EE, Hay RA (1980) Interrupted time series analysis. Sage Publications, Inc., Thousand Oaks, CA
Messner SF, Deane GD, Anselin L, Pearson-Nelson B (2005) Locating the vanguard in rising and falling homicide rates across U. S. cities. Criminology 43:661–696
Nagin DS (2009) Group-based modeling: an overview. In: Piquero AR, Weisburd D (eds) Handbook of quantitative criminology. Springer, New York
Nagin DS (2005) Group-based modeling of development. Harvard University Press, Cambridge, MA
Nagin DS, Farrington DP (1992a) The stability of criminal potential from childhood to adulthood. Criminology 30(2):235–260
Nagin DS, Farrington DP (1992b) The onset of persistence of offending. Criminology 30(4) 501–523
Nagin D, Land KC (1993) Age, criminal careers and population heterogeneity: specification and estimation of nonparametric, mixed Poisson model. Criminology 31(4):581–606
Nagin D, Paternoster R (2000) Population heterogeneity and state dependence: state of the evidence and directions for future research. J Quant Criminol 16(2):117–144
Nagin D, Paternoster R (1991) On the relationship of past and future participation in delinquency. Criminology 29(2):163–190
O’Brien RM (1999) Measuring the convergence/divergence of “serious crime” arrest rates for males and females: 1960–1995. J Quant Criminol 15(1):97–114
Ousey GC, Wilcox P, Brummel S (2008) Déjà vu all over again: investigating temporal continuity of adolescent victimization. J Quant Criminol 24(3):307–335
Paternoster R, Brame R (1997) Multiple routes to delinquence? A test of developmental and general theories of crime. Criminology 35(1):49–84
Paternoster R, Bushway S, Brame R, Apel R (2003) The effect of teenage employment on delinquency and problem behaviors. Soc Forces 82(1):297–335
Paternoster R, Dean CW, Piquero A, Mazerolle P, Brame R (1997) Generality, continuity, and change in offending. J Quant Criminol 13(3):231–265
Petras H, Masyn K (2009) General growth mixture analysis with antecedents and consequences of change. In: Piquero AR, Weisburd D (eds) Handbook of quantitative criminology. Springer, New York
Phillips JA, Greenberg DF (2008) A comparison of methods for analyzing criminological panel data. J Quant Criminol 24(1):51–72
Pridemore WA, Chamlin MB, Cochran JK (2007) An interrupted time-series analysis of Durkheim’s social deregulation thesis: the case of the Russian Federation. Justice Q 24(2):271–290
Pridemore WA, Chamlin MB, Trahan A (2008) A test of competing hypotheses about homicide following terrorist attacks: an interrupted time series analysis of September 11 and Oklahoma City. J Quant Criminol 24(4): 381–396
Raphael S, Winter-Ebmer R (2001) Identifying the effect of unemployment on crime. J Law Econ 44:259–283
Rosenfeld R, Fornango R, Rengifo AF (2007) The impact of order-maintenance policing on New York City homicide and robbery rates: 1988–2001. Criminology 45(2):355–383
Sherman LW, Gartin PR, Buerger ME (1989) Hot spots of predatory crime: routine activities and the criminology of place. Criminology 27(1):27–55
Sherman LW, Weisburd D (1995) General deterrent effects of police patrol in crime “hot spots”: A randomized, controlled trial. Justice Quarterly 12(4):625–648
Simpson SS, Bouffard LA, Garner J, Hickman L (2006) The influence of legal reform on the probability of arrest in domestic violence cases. Justice Q 23(3):297–316
Sims CA (1972) Money, income, and causality. Am Econ Rev 62(4):540–552
Smith MD, Devine JA, Sheley JF (1992) Crime and unemployment: effects across age and race categories. Sociol Perspect 35(4):551–572
Stata Press (2005) Stata times-series reference manual release 9. Stata Press, College Station, TX
Stolzenberg L, D’Alessio SJ (1994) Sentencing and unwarranted disparity: an empirical assessment of the long-term impact of sentencing guidelines in Minnesota. Criminology 32(2):301–310
Tracy P, Wolfgang ME, Figlio RM (1990) Delinquency careers in two birth cohorts. Plenum, New York
Weisburd D, Bushway S, Lum C, Yang S (2004) Trajectories of crime at places: a longitudinal study of street segments in the city of Seattle. Criminology 42(2):283–321
Witt R, Witte A (2000) Crime, prison, and female labor supply. J Quant Criminol 16(1):69–85
Wu D-M (1973) Alternative tests of independence between stochastic regressors and disturbances. Econometrica 41:733–750
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer Science+Business Media, LLC
About this chapter
Cite this chapter
Dugan, L. (2010). Estimating Effects over Time for Single and Multiple Units. In: Piquero, A., Weisburd, D. (eds) Handbook of Quantitative Criminology. Springer, New York, NY. https://doi.org/10.1007/978-0-387-77650-7_35
Download citation
DOI: https://doi.org/10.1007/978-0-387-77650-7_35
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-0-387-77649-1
Online ISBN: 978-0-387-77650-7
eBook Packages: Humanities, Social Sciences and LawSocial Sciences (R0)