Race,Ethnicity,and Structural Variations in Youth Risk of Arrest

Abstract

Missing from the considerable body of literature on disproportionate minority contact is an examination of the factors that influence risk of juvenile arrest. Using the National Longitudinal Survey of Youth 1997, the author examines racial/ethnic disparities in youth arrest, net of self-reported delinquency. Drawing from research using a minority threat perspective, this study examines whether disparities are exacerbated by macro levels of the relative size of the minority population and minority economic inequality. The results indicate Black youth have a higher risk of arrest than White youth in all contextual climates, but this disparity is magnified in predominantly non-Black communities. Differences between Hispanic and White youths’ risk of arrest did not reach statistical significance or vary across communities. The findings failed to yield support for the threat perspective but strongly supported the benign neglect thesis. Implications for theory and future research are discussed.

Keywords

race and ethnicity arrest juvenile justice disproportionate minority contact racial threat benign neglect hierarchical linear modeling

Despite being only 17% of the U.S. juvenile population ages 10 through 17, in 2012, Black youth comprised 32% of all arrests of juveniles (Puzzanchera, 2013; Snyder & Mulako-Wangota, 2015), 36% of cases handled in juvenile courts (Sickmund, Sladky, & Kang, 2014), and 40% of youth in residential placement (Sickmund, Sladky, Kang, & Puzzanchera, 2013). Over the past three decades, studies of juvenile justice system processing have shown that controlling for relevant legal factors, youth of color experience more severe treatment in the juvenile justice system, relative to their White counterparts (for reviews, see Bishop, 2005; Bishop, Leiber, & Johnson, 2010; Engen, Steen, & Bridges, 2002; Leiber, 2002, 2003; Leiber & Peck, 2013; Paternoster & Iovanni, 1989; Pope & Feyerherm, 1990a, 1990b; Pope, Lovell, & Hsia, 2002). Race significantly predicts the arrest decision (Tapia, 2010, 2011), intake (Leiber & Fox, 2005; Leiber & Johnson, 2008; Leiber, Johnson, Fox, & Lacks, 2007; Leiber & Mack, 2003), pretrial release (Rodriguez, 2007, 2010), diversion (Leiber & Johnson, 2008; Leiber & Mack, 2003; Rodriguez, 2010), petition (Leiber & Mack, 2003; Rodriguez, 2010), adjudication (Leiber et al., 2007), and disposition (Rodriguez, 2010). This recent research clearly supports the conclusion that race plays an important role in the processing of youth.

Despite the accumulation of empirical and theoretical research documenting the effects of extralegal variables on police behavior and juvenile justice processing, the overwhelming majority of studies include only Black and White youth. Only a handful has examined the impact of ethnicity on arrest and court processing (see, for example, Cicourel, 1967; Dannefer & Schutt, 1982; DeJong & Jackson, 1998; Huizinga, Thornberry, Knight, & Lovegrove, 2007; Kempf-Leonard & Sontheimer, 1995; Leiber, Peck, & Rodriguez, 2013; MacEachern & Bauzer, 1967; Maupin & Bond-Maupin, 1999; Tapia, 2010, 2011; Terry, 1967). The few studies that do include Hispanic youth have shown mixed findings. Early researchers found no evidence of bias against Hispanic youth (Cicourel, 1967; MacEachern & Bauzer, 1967; Terry, 1967). More recently, researchers have found evidence that Hispanic youth have a higher risk of justice system contact than their White counterparts but have disagreed about whether the level of bias against Hispanic youth exceeds (Kempf-Leonard & Sontheimer, 1995) or falls below the level of bias against Black youth (Dannefer & Schutt, 1982; DeJong & Jackson, 1998; Huizinga et al., 2007; Maupin & Bond-Maupin, 1999; Tapia, 2010, 2011).

In addition to the effects of race and ethnicity, research has begun to examine the macro-level characteristics that influence variation in juvenile court processing (Armstrong & Rodriguez, 2005; DeJong & Jackson, 1998; Feld, 1991, 1995; Freiburger & Jordan, 2011; Leiber et al., 2013; Rodriguez, 2007, 2010; Sampson & Laub, 1993; Tittle & Curran, 1988). Recent studies that have incorporated structural indicators have shown that court processing outcomes are influenced directly by macro-level urbanism, geographic region, population composition, underclass poverty, racial economic inequality, residential mobility, structural disadvantage, and the percentage of female-headed households (Armstrong & Rodriguez, 2005; DeJong & Jackson, 1998; Freiburger & Jordan, 2011; Leiber et al., 2013; Rodriguez, 2010; Sampson & Laub, 1993).

Despite repeated calls for research to consider racial/ethnic disparities within the context of broader racialized social structures (see, for example, Peterson, 2012), few studies have examined how contextual climate may indirectly influence the effects of race and ethnicity on likelihood of contact with the justice system. Focusing on different decision points in particular court jurisdictions, these studies have shown mixed results. A small group of studies demonstrate the interactive effects of racial ethnic group and contextual climate on the juvenile justice decisions. For example, using data from Maricopa County in Arizona, Rodriguez (2007) found that higher levels of county unemployment and poverty reduced the probability of preadjudication detention for Latino/a youth and increased the probability of detention for White youth. A similar study of the preadjudication detention decision using 9 years of data on criminal incidents perpetrated by juveniles in one Southeastern state showed that the differential treatment of Black youth was magnified in communities with a small relative Black population, higher ratios of Black to White socioeconomic inequality, and higher levels of economic competition (Thomas et al., 2013). Focusing on the petition decision, Freiburger and Jordan (2011) used offense-level data from all youth processed for misdemeanor and felony offenses in the juvenile court system in West Virginia during the year 2005 and found that offenses were more likely to be petitioned to juvenile court in counties with higher rates of poverty and lower proportions of female-headed households. Cross-level findings indicated that the effects of racial group and county-level poverty significantly interacted such that Black youth had the highest likelihood of petition in counties with greater levels of poverty.

Other studies have discounted the importance of community context in understanding racial and ethnic disparities in the justice system. For instance, Rodriguez (2013) showed that all youth were more likely to be detained in zip codes characterized by greater levels of structural disadvantage. However, cross-level interaction effects between racial ethnic group and structural disadvantage failed to show significance. Similarly, a recent analysis of data from 37 juvenile court jurisdictions from three states failed to show that minority population composition or economic inequality affected the social control of minority youth at the intake, adjudication, and detention decision points (Leiber et al., 2013). Hayes-Smith and Hayes-Smith (2009) examined the effects of race and community context on the decision to withhold adjudication in drug cases in the Florida juvenile justice system, a practice that is typically used to divert youth. They showed that, although Black youth were less likely to have their adjudications withheld, this disparity was not affected by the relative size of the Black population, racial economic inequality, concentrated disadvantage, nor the crime rate.

One method to examine disproportionate minority contact (DMC) with the justice system that has been neglected in research is the analysis of longitudinal individual-level data. Traditional studies of disproportionate youth contact with the justice system have relied on official data from youth who have already made contact with the justice system. Similarly, because the research literature is predominated by studies that rely on official records and focus on particular court jurisdictions, they are unable to control for variations in youth participation in delinquent behavior and lack geographic generalization. Missing in the extant literature is an analysis of the extent to which race and ethnicity interact with broader social structures to influence youth risk of arrest, net of delinquency. The objective of the present study is to contribute to the research by focusing on the structural conditions that may mitigate or magnify racial/ethnic disparities in juvenile risk of arrest. A nationally representative, longitudinal sample is used to examine the impact of county-level population composition and racial/ethnic economic inequality on White, Black, and Hispanic risk of juvenile arrest, net of self-reported delinquency. The results, which can be generalized to the national population of adolescents, have implications for assessing the validity of existing theoretical explanations of DMC and unmasking larger issues of social welfare and inequality.

Theoretical Perspectives

Much of the research examining the influence of racialized social structures on social control is based on the racial/minority threat thesis (Blalock, 1967) or the symbolic threat thesis (Tittle & Curran, 1988). The racial threat perspective suggests that Whites perceive racial minorities, particularly African Americans, as potential competition in the labor market and political sphere. Blalock (1967) argued that minority groups will face increased levels of social control when they threaten the dominant social order. This perspective suggests that as the relative size of the minority population increases, members of the dominant community will perceive the increased presence of minorities as a threat to the status quo of the dominant group. The dominant group will perceive the encroachment of minorities as a direct threat to White economic, political, and social hegemony. This perceived threat to the status quo motivates Whites to increase efforts to control the threat through the mobilization of resources, concerted discrimination, residential segregation, and formal control mechanisms against minorities. Blalock further suggested that there is a curvilinear association between minority population size and the use of formal social control mechanisms, with a decelerating slope in communities where the size of the minority population becomes or is already large. Thus, efforts at social control continue until the minority population size reaches a critical mass.

The symbolic threat thesis attributes the increased social control of minority groups to the social-psychological reactions of justice system personnel. Tittle and Curran (1988) argued that the group threat and traditional conflict approach fail to capture the essence of the process of justice system decision making. They note, “It is hard to imagine that adult Whites actually fear that racial minorities or youth will overthrow them politically or submerge them economically” (p. 53). Instead, Nonwhite youth symbolize “resentment-provoking or fear-provoking qualities like aggressiveness, sexuality, and absence of personal discipline” (Tittle & Curran, 1988, p. 52). Their findings suggest that discriminatory processing occurs in contexts where the minority group constitutes a highly visible symbolic threat, and disparities in social control are contingent on community-level perceptions of threat. Certain community structural conditions, such as deeply entrenched racial/ethnic inequalities and amplified perceptions of group differences, lead to social-psychological reactions to minorities by justice system decision makers. They are “perceived and stereotyped by decision-makers as either more dangerous and/or drug offenders who fail to abide by middle-class standards, and/or society” (Leiber et al., 2007, p. 472). These perceptions and social-psychological reactions to Nonwhite youth are then expressed through disparities in social control.

The symbolic threat thesis has been further refined by Sampson and Laub (1993). Their structural conflict model integrates the symbolic threat thesis with theoretical dimensions of structural inequality. They argue that to the extent that racial minorities represent a visible symbol of threat to the dominant group, community-levels of inequality and poverty should explain variation in juvenile court processing. Sampson and Laub discovered that contextual levels of poverty and racial inequality were significantly related to juvenile justice processing, particularly for preadjudication detention and adjudicated residential placement. Furthermore, the effects of structural indicators were magnified for Black youth. These findings are consistent with the argument that underclass Black males will be viewed as a threatening group by those in power and subjected to increased social control.

The literature examining formal social control using a threat perspective has found mixed or inconsistent results. Studies using the racial threat thesis to examine formal social control have typically operationalized racial threat with indicators of the relative size of the Black population (i.e., percent Black). Such studies have shown that the relative size of the minority population affects police strength (Stults & Baumer, 2007), arrest rates (Liska & Chamlin, 1984; Ousey & Lee, 2008; Stolzenberg, D’Alessio, & Eitle, 2004; Stucky, 2012), and prison populations (Greenberg & West, 2001). Until recently, few studies have included measures of economic competition (i.e., White-to-Black unemployment ratios). For instance, Thomas, Moak, and Walker (2013) found that juvenile preadjudication detention decisions are affected by county levels of racial economic threat but not the relative size of the Black population. Leiber et al. (2013) applied the racial threat thesis to both Black and Hispanic groups and similarly found that juvenile court outcomes are affected by racial and ethnic economic threat but not the relative size of the Black or Hispanic population. Other studies have found statistically nonsignificant effects of the relative size of the Black population and racial economic inequality on Black arrest rates (Parker, Stults, & Rice, 2005) and the processing of youth drug cases (Hayes-Smith & Hayes-Smith, 2009).

The Current Study

Although the threat perspective has been used to examine racial disparities in aggregate arrest rates (e.g., Eitle, D’Alessio, & Stolzenberg, 2002; Stucky, 2012) and processing in the juvenile (e.g., Leiber et al., 2013; Thomas et al., 2013) and adult (e.g., Caravelis, Chiricos, & Bales, 2011) justice systems, no study to date has examined the effects of racialized social structures on risk of juvenile arrest. Yet, to learn about disproportionate processing within the justice system, one must understand which youth are likely to come in contact with the system and what characteristics predict such contact. The current study extends recent efforts to examine the effects of context on youth justice system processing and addresses the lack of a focus on the racialized social structures that may partially explain disproportionate levels of youth penetration into the justice system. Drawing from research that has examined juvenile justice system processing using a threat theoretical perspective, this study examines the contextual climates that may exacerbate racial/ethnic disparities in risk of juvenile arrest. The use of a nationally representative sample in a data set that includes detailed information on youth delinquency and contact with the justice system allows the examination of individual and contextual effects, net of reported delinquent behavior. Given the mixed and inconsistent results of most studies examining the effects of community characteristics on juvenile justice outcomes using a threat perspective, there is need for research that examines the utility of this framework using recent longitudinal individual-level data.

Method

Participants

The current study used data collected from participants enrolled in the National Longitudinal Survey of Youth (NLSY97). The NLSY97 cohort comprises two independently selected probability samples: a self-weighting “cross-sectional” sample and a supplemental oversample of Black and/or Hispanic respondents. Each sample was collected using a multistage stratified cluster probability sampling design. After weighting, the NLSY97 cohort is representative of the U.S. population who were aged 12 to 16 years on December 31, 1996, and resided in the United States when the survey began in 1997. Interviews with respondents have been conducted on a yearly basis. Although the NLSY97 was launched to enable researchers to examine life-course school and labor force transitions, the NLSY97 also collects extensive information on respondents’ personal characteristics, delinquent and criminal behaviors, and contact with the justice system. The NLSY97’s breadth and longitudinal design make it uniquely capable of examining individual and contextual determinants of youth justice system contact.

The initial wave of NLSY97 respondents included 8,984 youth aged 12 to 16. Age-specific arrest rates (e.g., Bureau of Justice Statistics, 2014) and scholars (e.g., Farrington, 1986) examining the age–crime relationship have demonstrated that youth probability of arrest and subsequent justice system contact begins increasing sharply around age 12 or 13. Therefore, it was assumed that the inception of risk of arrest begins at age 12, and the longitudinal investigation was limited to youth who were aged 12 and 13 during the first round of data collection. Retrospective interview data were used to track respondents until their 18th birthdays.

County-level data from the U.S. Census of Population and Housing, Summary Tape File 3 were appended to the individual-level NLSY data using the confidential NLSY97 Geocode CD, which contains information on respondent migration patterns. At each annual interview, the NLSY97 collects information about each respondent’s residence at the time of the interview and since the previous interview date. Respondents who had moved since the previous interview date report the date of each move and the county and state of each residence. During the period of longitudinal observation (i.e., youths’ 12th to 18th birthdays), respondents reported residing in all 50 states, the District of Columbia, and a total of 560 unique counties.

Of the 2,878 age-eligible NLSY97 youth, two youth arrested as juveniles were excluded from the analyses because they lacked valid information on the timing of the arrest. This resulted in a sample size of 2,876 youth. Table 1 displays descriptive information for the sample. Slightly more than half of the sample is male and, after weighting, racial/ethnic group proportions are consistent with the U.S. Census population estimates (U.S. Census Bureau, 2004).

Table 1:

Descriptive Statistics (N = 2,876)

	%	95% CI
Sociodemographic characteristics
Age 12 in Round One	41.11	[39.03, 43.18]
Age 13 in Round One	58.89	[56.82, 60.97]
Male	51.53	[49.42, 53.64]
Race/ethnicity
Black	15.28	[12.49, 18.06]
Hispanic	12.71	[9.86, 15.56]
White	72.01	[68.07, 75.96]

Note. Estimates are weighted. The survey design was taken into account using Taylor series method. CI = confidence interval.

Research Variables

Individual-Level Variables

A dichotomous outcome variable indicates whether youth were arrested as juveniles (1 = arrested, 0 = not arrested). Longitudinal data that provide information on the month and year the arrest occurred allowed measurement of the timing of arrest (conceptualized as youths’ age in months at the time of their first juvenile arrest).

Because the current study relies on self-report measures of arrest, the validity of the findings may be affected by respondent recall and/or underreporting (Thornberry & Krohn, 2000). Little is known about the validity of self-report data on the timing of arrests; however, Morris and Slocum (2010) found that retrospective self-report data on the prevalence and frequency of arrest yielded accurate measures, and self-report data on the timing of arrest yielded accurate measures for the timing of recent arrests. The NLSY97’s collection of information on the occurrence and timing of youth arrest during each annual interview increases the likelihood that the data provide accurate estimates. Despite the limitations associated with the self-report method, it remains one of the most valid and reliable methods of measuring delinquency and justice system contacts (Thornberry & Krohn, 2000).

Sex is a dichotomous independent variable, with female as the reference group. Racial/ethnic group consists of dummy variables for White, Black, and Hispanic youth, with White as the reference group.¹

Delinquency is captured with five variables measured annually indicating youth frequency of participation in behaviors in the last year: (a) vandalism is the number of times youth purposively destroyed property, (b) theft is the number of times youth stole something worth US$50 or more, (c) assault is the number of times youth attacked or assaulted others, (d) other property crime is the number of times youth committed property crimes other than vandalism, and (e) drug sales is the number of times youth sold drugs. Youth reporting 50 or more delinquent acts were censored at 50. In the multivariate analyses, vandalism, theft, assault, other property time, and drug sales are treated as time-varying covariates with values varying according to annual reports of participation in delinquent behavior.

County-Level Variables

This study includes four contextual measures consistent with prior research. Percent Black is the proportion of Black residents in each county. Racial economic inequality is measured using the ratio of Black to White unemployment among those in the civilian labor force; higher values indicate greater levels of Black unemployment compared with White unemployment. To examine the influence of ethnic threat on youth risk of arrest, two additional measures were constructed similar to the racial threat measures. Percent Hispanic indicates the proportion of Hispanic residents in each county, and ethnic economic inequality indicates the ratio of Hispanic to White unemployment rate. As with the racial economic inequality measure, higher values of ethnic economic inequality indicate greater levels of Hispanic unemployment relative to White unemployment.

Statistical Method

Event history modeling techniques allow the study of “length of time until the occurrence of some event” (Hox, 2010, p. 159). Defined as a “qualitative change in state” (DeMaris, 2004, p. 383), the event modeled in the proposed research is a first arrest. The use of event history methods has a distinct advantage for the study of youth risk (hazard) of arrest: its ability to deal with right-censoring of the data, which takes place when individuals have not experienced the event (an arrest) by the end of the study period (Singer & Willett, 2003). In these cases, the researchers have no way of determining if or when the individuals will experience the event. To understand which characteristics predict youth contact with the justice system, data from both censored and uncensored cases must be simultaneously incorporated in the analysis. Censored cases provide important information, especially about the probability that youth will avoid arrest.

Due to the nested structure of the data (youth nested within counties), and because timing of arrest is measured discretely (i.e., month and year of arrest), I employed the multilevel extension of discrete time interval-censored hazard model (DeMaris, 2004; Hox, 2010). The general approach to analyzing discrete time survival data is to model the hazard as a linear function of the covariates and then transform the linear function of the hazard using the appropriate link function. The logit link function is generally used when data are interval-censored (DeMaris, 2004), and the multilevel extension can be obtained by introducing a normally distributed random effect into the linear regression equation (Hox, 2010).

Steps in the Data Analysis

All analyses were conducted using the SAS software, which has the capacity to generate accurate parameter estimates, standard errors, and tests of significance for complex sample designs. Collinearity diagnostics were performed prior to analysis. All continuous predictor variables were grand mean centered and standardized to allow the model intercepts to be interpreted as the expected outcome for the average youth.

The initial stage of data analysis involved generating descriptive statistics for the sample. The distribution of youth arrest was examined in 6-month intervals during the observation status. Point estimates of the percentage of youth arrested at each age assume censored cases are missing at random (MAR). Next, interval estimates for each age were calculated using the methodology presented by Brame, Turner, Paternoster, and Bushway (2011). The lower bound for each age was calculated assuming no censored cases had been arrested. This was estimated by dividing the number of youth arrested by that age by the total sample size. The upper bound for each age was calculated assuming all censored cases had been arrested. This was obtained by adding the number of youth arrested by that age to the number of missing cases and dividing the sum by the total sample size. Finally, prevalence of juvenile arrest was examined for each racial/ethnic group. Point estimates by race and ethnicity were calculated assuming censored cases are MAR.² Descriptive summary statistics are presented using the data weighted to be representative of the population. The survey design was taken into account using Taylor series method.

For the multivariate analysis of hazard of arrest, the data were rearranged into a person-period format with one observation for each month from the inception of risk (12 years of age) until censoring, first arrest, or the end of the observation period (18 years of age).³ Another advantage of event history analysis is its ability to incorporate into the analysis predictor variables whose values vary over time (i.e., time-varying covariates; Allison, 1984). In this study, youth participation in delinquent behavior varied as they aged, and youths’ contextual climates varied as they moved⁴ residences. Thus, the delinquency indicators and contextual data were treated as time-varying covariates, and sex and racial/ethnic group were treated as fixed covariates.

The effects of individual and contextual variables on youth risk of arrest during adolescence were estimated using multilevel modeling techniques in the SAS GLIMMIX procedure, which has the capacity to fit two-level generalized linear mixed models (GLMM), integrated over random effects (Gharibvand & Liu, 2009; SAS Institute, 2008). The analysis strategy following Hox’s (2010) bottom-up approach simplifies each model by adding each type of parameter one step at a time, excluding nonsignificant effects. Because of the interest in parameter effects rather than point estimates, multivariate analysis uses unweighted data (DuMouchel & Duncan, 1983; Olsen, 2009; Winship & Radbill, 1994).

Missing Data

Only modest amounts of data are missing in the NLSY97. For each indicator of delinquency, between 3% and 4% of youth were missing data. In such cases, I imputed missing values with the mean (Osborne, 2013). In addition, 2% of youth were missing one or more months of migration information. Missing person-period months were excluded from the analysis. This resulted in the exclusion of 382 person-period observations (0.2% of the analysis sample). Among youth missing migration data, most (55.2%) were only missing 1 month of migration information. Similarly, months corresponding to youth living out of the country were also excluded from the analysis (248 person-months from 23 youth, <0.2% of the analysis sample). The final sample size was 185,769 person-months from 2,832 youth.

Results

Prevalence of Youth Arrest

Table 2 summarizes youth prevalence of arrest and censoring during the observation period. As Table 2 shows, 18.41% of youth in the sample were arrested as juveniles. This is consistent with prior analysis of the NLSY97 data on the cumulative prevalence of arrest (Brame et al., 2011). Brame et al. (2011) analyzed the self-weighting “cross sectional” sample of the NLSY and found the MAR rate to be 17.8% by age 18.

Table 2:

Summary of Arrest Status: Cumulative Prevalence of Arrest to Age 18 (N = 2,876)

Arrested					Not Arrested	Censored
Age	Frequency	Cumulative Percentage	Cumulative Percentage		Frequency	Frequency
Age	Frequency	Cumulative Percentage	Lower Bound	Upper Bound	Frequency	Frequency
<12	44	1.40	1.40	1.40	2,832	0
12	61	1.76	1.76	1.83	2,813	2
12.5	98	2.92	2.91	3.29	2,768	10
13	128	3.88	3.85	4.43	2,732	16
13.5	157	4.93	4.89	5.68	2,698	21
14	195	6.23	6.18	7.07	2,657	24
14.5	251	8.09	8.02	8.91	2,601	24
15	293	9.45	9.36	10.35	2,557	26
15.5	356	11.77	11.63	12.79	2,490	30
16	402	13.21	13.01	14.53	2,436	38
16.5	444	14.88	14.62	16.42	2,387	45
17	489	16.55	16.18	18.41	2,330	57
17.5	540	18.41	17.93	20.54	2,270	66

Note. Cumulative percentage shows the percentage of youth in the sample who have been arrested by that age. Percentage estimates assume censored cases are missing at random (MAR). At each age, the lower bound presents arrest prevalence assuming no censored cases have been arrested, and the upper bound presents the arrest prevalence assuming all censored cases have been arrested. The survey design was taken into account using Taylor series method.

Before testing the impact of contextual characteristics on racial/ethnic disparities in arrest, it was of interest to examine whether prevalence of arrest at age 18 was associated with race or ethnicity. Table 3 provides a comparison of percent arrested by age 18 and racial/ethnic group, assuming censored cases are MAR. Consistent with prior research, Black youth reported a significantly higher risk of arrest than White and Hispanic youth. Over one quarter of all Black youth reported that they had been arrested before their 18th birthday (25.23%). Differences between White and Hispanic arrest rates did not reach statistical significance. Hispanic youth reported a cumulative prevalence rate of 18.05%, and White youth reported a cumulative prevalence rate of 16.95%.

Table 3:

Prevalence of Juvenile Arrest Among Noncensored White, Black, and Hispanic Youth (n = 2,810)

	White (n = 1,490)		Black (n = 709)		Hispanic (n = 611)		Test Statistic
	%	SE	%	SE	%	SE	Test Statistic
Arrested as a Juvenile	16.95	1.02	25.23	2.04	18.05	1.79	8.13***

Note. Estimates are weighted. Survey design taken into account using Taylor series method. Comparisons were done with design-based F test.

p < .05. **p < .01. ***p < .001.

Multivariate Analyses of Youth Hazard of Arrest

This section presents the results of fitting discrete time hazard models for youth risk of arrest. Table 4 shows the parameter estimates for the models that include youth individual characteristics, contextual characteristics, and cross-level interactions between youth characteristics and contextual environment.⁵ To facilitate interpretation, I subtracted a centering constant of one from the time variable so the intercept is interpretable as the value of the logit hazard during the first observation period (i.e., when the time variable equals zero). The baseline null model (results not shown) indicates youths’ estimated hazard of arrest during the month following their 12th birthdays is approximately 0.001 (e^−6.707). Expectedly, as time increases, youth risk of arrest increases. Beginning at age 12, each additional year increased youth odds of experiencing their first arrest increase by approximately 48.6% (100[e^{(0.036×12) + (−.003×12)} − 1]).

Table 4:

Estimating the Hazard of Youth Arrest (n = 2,825)

	Model 1	Model 2	Model 3	Model 4
Fixed part
Intercept	−7.239***	−7.276***	−7.277***	−7.250***
Time (in months)	0.035***	0.037***	0.036***	0.035***
Time²	<0.001*	<0.001**	<0.001*	<0.001*
Male	0.630***	0.632***	0.642***	0.633***
Black^a	0.410***	0.484***	0.423**	0.634***
Hispanic^a	0.155	0.236	0.258	0.256
Delinquency
Vandalism^b	0.046*	0.045*	0.053*	0.049*
Assault^b	0.047*	0.048*	0.043*	0.047*
Drug sales^b	0.166***	0.165***	0.166***	0.165***
Fixed county characteristics
% Black^b		−0.047	−0.039	0.064
Racial economic inequality^b		−0.104	−0.101	0.006
% Hispanic^b		−0.043	−0.040	−0.063
Ethnic economic inequality^b		0.380	0.367	0.004
Cross-level interactions
Black × % Black^b				−0.252*
Black × Racial economic inequality^b				−0.157
Black × % Hispanic^b				−0.014
Black × Ethnic economic inequality^b				−0.094
Random part
Intercept	0.008	0.017	0.004	0.003
Black slope			0.401*	0.493
Deviance	6,662.9	6,657.5	6,653.2	6,646.8
AIC	6,682.9	6,685.5	6,683.2	6,684.8

Note. Indicators of theft and participation on other property crimes were nonsignificant and were therefore excluded from the model for the sake of parsimony. This did not substantively affect the results. AIC = Akaike’s information criterion.

Reference category is Non-Hispanic White.

Variable is standardized.

p < .05. **p < .01. ***p < .001.

Model 1 in Table 4 includes individual-level explanatory variables added as fixed effects to determine their contribution to explaining youth risk of arrest. Indicators of theft and participation in other property crimes were nonsignificant in all analyses and were therefore excluded from each model for the sake of parsimony. This did not substantively affect the results. Model 2 adds the block of county-level fixed effects to examine the direct effects of racial/ethnic population composition and economic inequality. Model 3 adds a random coefficient for Black to examine if the effect of youth racial group varied significantly at the county level. The random coefficient for Hispanic was nonsignificant and was therefore excluded from the analysis. Finally, cross-level interaction effect model was generated to determine whether the significant variance component for Black racial group was explained by variation in county climate.

Looking at Model 4, the effects of time (measured in months), gender, racial group, and self-reported vandalism, assault, and drug sales were significantly related to youth hazard of arrest. During each month of observation, the odds of experiencing an arrest are nearly twice as high for males as for females (e^0.633). Youth who reported greater levels of self-reported delinquency are more likely to experience their first arrest. In particular, one standard deviation increases in vandalism and assault both increase the odds of experiencing an arrest by 5% (100 × [e^0.049 − 1] and 100 × [e^0.047 − 1], respectively). A one standard deviation increase in drug sales increases the odds of experiencing an arrest by 18% (100 × [e^0.165 − 1]).

The significant interactions between racial/ethnic group and county-level Black population concentration indicate that the magnitude of the bias against minority youth depends on the racial composition of the population. Looking at the significant interaction for race, although Black youth have a higher risk of arrest than their counterparts in all contextual climates, racial disparities are magnified in counties with a low concentration of Black residents. This is contrary to the racial threat hypothesis, which predicts higher levels of social control in areas with large visible minority populations. The Black × Percent Black interaction in Table 4 shows the hazard of arrest for Black youth is approximately 47% higher than for their White counterparts (100 × [e^{0.634 + (1 × −252)} − 1]) in counties with a large concentration of Black residents (one standard deviation above the mean). In areas with an average concentration of Black residents, the odds of experiencing an arrest are approximately 89% greater for Black youth in comparison with White youth (100 × [e^0.634 − 1]). Correspondingly, in counties with a small concentration of Black residents (one standard deviation below the mean), the hazard of arrest for Black youth is approximately 143% higher than for their White counterparts (100 × [e^{0.634 + (−1 × −0.252)} − 1]).

Discussion

The current study examined the impact of racial/ethnic group on youth risk of arrest, controlling for self-reported delinquency and structural factors indicating racial/ethnic threat. Despite the large body of research examining the effects of race on justice system processing, researchers have paid little attention to exploring the contextual environments associated with disparities in justice system processing. Even fewer studies have examined the indirect effects of community characteristics and disentangled the effects of race and ethnicity (for exceptions, see Leiber et al., 2013; Rodriguez, 2007, 2013). Using a nationally representative and longitudinal sample of youth, this research focuses on how the effects of race and ethnicity differentially affect youth risk of arrest depending on contextual environment, net of participation in delinquent behavior.

The bivariate results indicate that while approximately 18% of all youth are arrested before their 18th birthdays, the prevalence of arrest differs by race and ethnicity. More than one quarter (25.23%) of Black youth are arrested as juveniles, compared with 18.05% of Hispanic youth and 16.95% of White youth.

Findings from the multivariate analyses show the effect of racial/ethnic group on risk of arrest, controlling for self-reported delinquency. Net of self-reported delinquency, Black youth have a higher hazard of juvenile arrest than their White counterparts. The results do not support the idea that Hispanic youth have a higher hazard of arrest than White youth. The findings also show that while the effect of youth race varies significantly across counties, the effect of Hispanic ethnicity does not. Furthermore, cross-level interaction effects suggest the magnitude of the effect of race depends on the relative size of the Black population. In particular, as the relative size of the Black population increases, racial disparities in hazard of arrest are mitigated. In other words, racial disparities in risk of arrest are most pronounced among youth residing in predominately non-Black communities. County levels of the relative size of the Hispanic population and White-to-Hispanic economic inequality did not explain county-level variation in the effect of Black racial group. The null findings related to youth ethnic group are particularly important given the dearth of research literature on the role of ethnicity on disparities in juvenile justice system contact and processing.

Taken together, these findings raise concerns about the utility of the racial and ethnic group threat perspective in explaining how social structure influences the social control of youth of color. The finding that the size of the Black population has a decelerating effect on the risk of arrest for Black youth is contrary to the racial threat perspective and consistent with the benign neglect hypothesis. The benign neglect hypothesis posits an inverse relationship between the relative size of the racial/ethnic minority population and measures of social control for that population (see, for example, Liska & Chamlin, 1984; Myer & Chamlin, 2011; Parker et al., 2005). A small number of studies examining racial disparities in arrest rates and juvenile detention decisions support the benign neglect perspective (e.g., Parker et al., 2005; Thomas et al., 2013). For instance, Stolzenberg et al. (2004) found an inverse relationship between the size of the Black population and the probability of a crime reported to the police resulting in an arrest. Other research finds an inverse relationship between the size of the Black population and social control; when the relative size of the Black population is larger, minority populations are perceived not to be a threat to majority interests, and formal social control against minorities is less frequent (Stucky, 2012). This relationship is said to result from two mechanisms (Myer & Chamlin, 2011). First, crime in communities of color is likely to be intraracial, and police and victims may view intraracial crime as a personal problem that does not require official intervention. Second, residents of communities of color may lack the sociopolitical capital necessary to legitimize their problems, resulting in a decreased level of crime control.

The conceptual model that frames this research extends existing models of disproportionate youth contact with the justice system that favor individual-level explanations for disparities in youth penetration into the justice system. The study’s use of a nationally representative sample drawn from the general youth population allowed the incorporation of data from youth who avoided justice system contact during adolescence. This allowed the empirical examination of what characteristics predict such contact, net of reported delinquent behavior, with broadly generalizable results. Further extending previous research on disproportionate contact with the justice system are the current study’s units of analysis. Traditional studies of disparities in juvenile justice examine decisions about youth as their unit of analysis. The current study’s use of a longitudinal data set covering youth’s entire period of development allows for the examination of youth as the units of analysis. Finally, the study’s use of event history analysis allowed the effects of predictors on youth risk of arrest to vary over time, as youth changed their behavior, moved residences, and as the characteristics of geographic areas changed. The results, overall, suggest disparities in youth contact with the justice system are partially explained by macro-structural environments. Specifically, this study found racial disparities in youth risk of arrest, which are magnified in predominately non-Black communities.

This study has several limitations. First, although the NLSY97 includes a host of variables relevant to youth delinquency and contact with the justice system, it lacks detailed information on police encounters with youth. Second, because the data analyzed are from a nationally representative sample of youth, the findings cannot be generalized to more severe offending populations. Although the NLSY97 has an adequately large sample size for the analysis of delinquency and contact with the justice system, which is a relatively rare occurrence, it is likely that youth who have committed severe offenses appear in the sample in very small numbers and are therefore not sufficiently represented in these data. However, evidence indicates racial disparities in justice system processing may be more pronounced when offenses are less serious (Piquero, 2008). Moreover, because the overwhelming majority of research examining DMC relies on official data within particular jurisdictions, the present study complements and extends previous research precisely because of its use of a nationally representative sample and broad generalizability.

Future research should investigate the impact of racial threat measures on racial inequity across successive stages of juvenile justice system processing. Previous research documents the importance of examining multiple stages of justice system processing when examining disproportionate penetration into the justice system. For example, racial/ethnic group differences may be more pronounced at some decision-making levels, and decisions made in earlier decision points may affect subsequent outcomes (Bishop, 2005; Bishop et al., 2010; Dannefer & Schutt, 1982). Small differentials may also accumulate over each decision level, “transforming a more or less heterogeneous racial arrest population into a homogeneous institutional Black population” (Liska & Tausig, 1979, p. 197). Some research (Dannefer & Schutt, 1982; Rodriguez, 2007, 2010) also provides evidence of a compensatory effect in the courts where judges may attempt to counteract differential treatment by the police. Dannefer and Schutt (1982), for example, demonstrated the presence of court-level corrections processes in urban areas, where Black youth received more favorable court dispositions than White youth, indicating that the effects of race on youth penetration into the justice system depend on the processing stage and social context. A clear priority for future research on DMC with the justice system is to examine how racialized social structures might affect racial and ethnic disparities across stages of the juvenile justice system.

Like most matters of race and ethnicity, disproportionate contact with the justice system is rooted in the social contexts into which youth are born, including racial residential segregation (Hoytt, Schiraldi, Smith, & Ziedenberg, 2003). Given that police jurisdictions and juvenile courts are organized at the local level, analysis that ignores structural variations across communities may be misleading (Feld, 1991). According to Johnson (2006), ignoring potential contextual effects in disparate treatment masks larger issues of social welfare and inequality and may obscure effective strategies for addressing disproportionate justice system contact. Without the commitment to the structural reform of inequalities, reducing racial and ethnic disparities is unlikely.

Footnotes

Acknowledgements

The author thanks Merry Morash, Miriam Northcutt Bohmert, the anonymous reviewers, and the editor for their helpful suggestions on the previous version of this article. The author would also like to thank the staff at the Bureau of Labor Statistics (BLS) for allowing access to the National Longitudinal Survey of Youth 1997 (NLSY97) Confidential Geocode Data.

This research was supported in part by an award from the National Institute of Justice (2012-IJ-CX-0018). The opinions, findings, and conclusions or recommendations expressed in this research are those of the author and do not necessarily reflect those of the Department of Justice.

Notes

Tia Stevens Andersen is an assistant professor in the Department of Criminology and Criminal Justice at the University of South Carolina. She received her PhD from Michigan State University in 2013. Her primary research agenda involves examining how gender, race, and structural inequalities influence juvenile justice system processing.

References

Allison

P. D.

(1984). Event history analysis: Regression for longitudinal event data. Newbury Park, CA: SAGE.

Armstrong

G. S.

Rodriguez

(2005). Effects of individual and contextual characteristics on preadjudication detention of juvenile delinquents. Justice Quarterly, 22, 521-539.

Bishop

D. M.

(2005). The role of race and ethnicity in juvenile justice processing. In Hawkins

D. F.

Kempf-Leonard

(Eds.), Our children, their children: Confronting racial and ethnic differences in American juvenile justice (pp. 23-82). Chicago, IL: University of Chicago Press.

Bishop

D. M.

Leiber

M. J.

Johnson

(2010). Contexts of decision making in the juvenile justice system: An organizational approach to understanding minority overrepresentation. Youth Violence and Juvenile Justice, 8, 213-233.

Blalock

H. M.

(1967). Toward a theory of minority-group relations. New York, NY: Wiley.

Brame

Turner

M. G.

Paternoster

Bushway

S. D.

(2011). Cumulative prevalence of arrest from ages 8 to 23 in a national sample. Pediatrics, 129, 21-27.

Bureau of Justice Statistics. (2014). OJJDP statistical briefing book. Retrieved from http://ojjdp.gov/ojstatbb/crime/qa05301.asp

Caravelis

Chiricos

Bales

(2011). Static and dynamic indicators of minority threat in sentencing outcomes: A multi-level analysis. Journal of Quantitative Criminology, 27, 405-425.

Cicourel

A. V.

(1967). The social organization of juvenile justice. New York, NY: Wiley.

10.

Dannefer

Schutt

R. K.

(1982). Race and juvenile justice processing in court and police agencies. American Journal of Sociology, 87, 1113-1132.

11.

DeJong

Jackson

K. C.

(1998). Putting race into context: Race, juvenile justice processing, and urbanization. Justice Quarterly, 15, 487-504.

12.

DeMaris

(2004). Regression with social data: Modeling continuous and limited response variables. Hoboken, NJ: John Wiley.

13.

DuMouchel

W. H.

Duncan

G. J.

(1983). Using sample survey weights in multiple regression analysis of stratified sample. Journal of the American Statistical Association, 78, 535-543.

14.

Eitle

D’Alessio

S. J.

Stolzenberg

(2002). Racial threat and social control: A test of the political, economic, and threat of black crime hypotheses. Social Forces, 81, 557-576.

15.

Engen

R. L.

Steen

Bridges

G. S.

(2002). Racial disparities in the punishment of youth: A theoretical and empirical assessment of the literature. Social Problems, 49, 194-220.

16.

Farrington

D. P.

(1986). Age and crime. Crime and Justice: A Review of Research, 7, 189-250.

17.

Feld

B. C.

(1991). Justice by geography: Urban, suburban, and rural variations in juvenile justice administration. The Journal of Criminal Law and Criminology, 82, 156-210.

18.

Feld

B. C.

(1995). The social context of juvenile administration: Racial disparities in an urban juvenile court. In Kempf-Leonard

Pope

C. E.

Feyerherm

(Eds.), Minorities in juvenile justice (pp. 66-97). Thousand Oaks, CA: SAGE.

19.

Freiburger

T. L.

Jordan

K. L.

(2011). A multilevel analysis of race on the decision to petition a case in the juvenile court. Race and Justice, 1, 185-201.

20.

Gharibvand

Liu

(2009). Analysis of survival data with clustered events. Cary, NC: SAS Institute.

21.

Greenberg

D. F.

West

(2001). State prison populations and their growth, 1971-1991. Criminology, 39, 615-654.

22.

Hayes-Smith

(2009). Race, racial context, and withholding adjudication in drug cases: A multilevel examination of juvenile justice. Journal of Ethnicity in Criminal Justice, 7, 163-185.

23.

Hox

J. J.

(2010). Multivariate analysis: Techniques and applications (2nd ed.). New York, NY: Routledge.

24.

Hoytt

E. H.

Schiraldi

Smith

B. V.

Ziedenberg

(2003). Reducing racial disparities in juvenile detention: Eight pathways to juvenile detention reform. Baltimore, MD: Annie E. Casey Foundation.

25.

Huizinga

Thornberry

T. P.

Knight

Lovegrove

(2007). Disproportionate minority contact in the juvenile justice system: A study of differential minority arrest/referral to court in three cities (Unpublished report). Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

26.

Johnson

B. D.

(2006). The multilevel context of criminal sentencing: Integrating judge- and county-level influences. Criminology, 44, 259-298.

27.

Kempf-Leonard

Sontheimer

(1995). The role of race in juvenile justice in Pennsylvania. In Kempf-Leonard

Pope

C. E.

Feyerherm

(Eds.), Minorities in juvenile justice (pp. 98-127). Thousand Oaks, CA: SAGE.

28.

Leiber

M. J.

(2002). Disproportionate minority confinement (DMC) of youth: An analysis of state and federal efforts to address the issue. Crime & Delinquency, 48, 3-45.

29.

Leiber

M. J.

(2003). The contexts of juvenile justice decision making: When race matters. Albany: State University of New York Press.

30.

Leiber

M. J.

Fox

K. C.

(2005). Race and the impact of detention of juvenile justice decision-making. Crime & Delinquency, 51, 470-497.

31.

Leiber

M. J.

Johnson

J. D.

(2008). Being young and Black: What are their effects on juvenile justice decision making? Crime & Delinquency, 54, 560-581.

32.

Leiber

M. J.

Johnson

Fox

Lacks

(2007). Differentiating among racial/ethnic groups and its implications for understanding juvenile justice decision making. Journal of Criminal Justice, 35, 471-484.

33.

Leiber

M. J.

Mack

K. Y.

(2003). The individual and joint effects of race, gender, and family-status on juvenile justice decision-making. Journal of Research in Crime & Delinquency, 40, 34-70.

34.

Leiber

M. J.

Peck

J. H.

(2013). Race in juvenile justice and sentencing policy: An overview of research and policy recommendations. Law and Inequality, 31, 331-367.

35.

Leiber

M. J.

Peck

J. H.

Rodriguez

(in press). Minority threat and juvenile court outcomes. Crime & Delinquency. Advance online publication. doi:10.1177/0011128713495776

36.

Liska

A. E.

Chamlin

(1984). Social structure and crime control among macro-social units. American Journal of Sociology, 90, 383-395.

37.

Liska

A. E.

Tausig

(1979). Theoretical interpretations of social class and racial differentials in legal decision-making for juveniles. The Sociological Quarterly, 20, 197-207.

38.

MacEachern

A. W.

Bauzer

(1967). Factors related to disposition in juvenile police contacts. In Klein

M. W.

Myerhoff

B. G.

(Eds.), Juvenile gangs in context (pp. 192-210). Englewood Cliffs, NJ: Prentice Hall.

39.

Maupin

J. R.

Bond-Maupin

L. J.

(1999). Detention decision-making in a predominately Hispanic region: Rural and non-rural differences. Juvenile & Family Court Journal, 50, 11-23.

40.

Morris

N. A.

Slocum

(2010). The validity of self-reported prevalence, frequency and timing of arrest: An evaluation of data collected using a life events calendar. Journal of Research in Crime & Delinquency, 47, 210-240.

41.

Myer

A. J.

Chamlin

M. B.

(2011). Aggregation bias and the benign neglect hypothesis. Journal of Crime & Justice, 34, 124-138.

42.

Olsen

(2009, July). Weighting NLS data. Paper presented at the General NLS Workshop, Columbus, OH.

43.

Osborne

J. W.

(2013). Best practices in data cleaning. Thousand Oaks, CA: SAGE.

44.

Ousey

G. C.

Lee

M. R.

(2008). Racial disparity in formal social control: An investigation of alternative explanations of arrest rate inequality. Journal of Research in Crime & Delinquency, 45, 322-355.

45.

Parker

K. F.

Stults

B. J.

Rice

S. K.

(2005). Racial threat, concentrated disadvantage and social control: Considering the macro-level sources of variation in arrests. Criminology, 43, 1111-1134.

46.

Paternoster

Iovanni

(1989). The labeling perspective and delinquency: An elaboration of the theory and an assessment of the evidence. Justice Quarterly, 6, 359-394.

47.

Peterson

R. D.

(2012). The central place of race in crime and justice—The American Society of Criminology’s 2011 Sutherland address. Criminology, 50, 303-328.

48.

Piquero

A. R.

(2008). Disproportionate minority contact. Future of Children, 18, 59-79.

49.

Pope

C. E.

Feyerherm

W. W.

(1990a). Minority status and juvenile justice processing: An assessment of the research literature (Part I). Criminal Justice Abstracts, 22, 327-335.

50.

Pope

C. E.

Feyerherm

W. W.

(1990b). Minority status and juvenile justice processing: An assessment of the research literature (Part II). Criminal Justice Abstracts, 22, 527-542.

51.

Pope

C. E.

Lovell

Hsia

H. M.

(2002). Disproportionate minority confinement: A review of the research literature from 1989 through 2001. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

52.

Puzzanchera

(2013). Juvenile arrests 2011. Washington, DC: Office of Juvenile Justice and Delinquency Prevention.

53.

Rodriguez

(2007). Juvenile court context and detention decisions: Reconsidering the role of race, ethnicity, and community characteristics in juvenile court processes. Justice Quarterly, 24, 629-656.

54.

Rodriguez

(2010). The cumulative effect of race and ethnicity in juvenile court outcomes and why preadjudication detention matters. Journal of Research in Crime & Delinquency, 47, 391-413.

55.

Rodriguez

(2013). Concentrated disadvantage and the incarceration of youth: Examining how context affects juvenile justice. Journal of Research in Crime & Delinquency, 50, 189-215.

56.

Sampson

R. J.

Laub

J. H.

(1993). Structural variations in juvenile court processing: Inequality, the underclass, and social control. Law & Society Review, 27, 285-311.

57.

SAS Institute. (2008). The GLIMMIX procedure: SAS/STAT user’s guide. Cary, NC: Author.

58.

Sickmund

Sladky

Kang

(2014). Easy access to juvenile court statistics: 1985-2011. Retrieved from http://www.ojjdp.gov/ojstatbb/ezajcs/

59.

Sickmund

Sladky

T. J.

Kang

Puzzanchera

(2013). Easy access to the census of juveniles in residential placement. Retrieved from http://www.ojjdp.gov/ojstatbb/ezacjrp/

60.

Singer

J. D.

Willett

J. B.

(2003). Applied longitudinal data analysis: Modeling change and event occurrence. Oxford, UK: Oxford University Press.

61.

Snyder

Mulako-Wangota

(2015). Arrest data analysis tool. Available from www.bjs.gov

62.

Stolzenberg

D’Alessio

S. J.

Eitle

(2004). A multilevel test of racial threat theory. Criminology, 42, 673-698.

63.

Stucky

T. D.

(2012). The conditional effects of race and politics on social control: Black violent crime arrests in large cities, 1970-1990. Journal of Research in Crime & Delinquency, 49, 3-30.

64.

Stults

B. J.

Baumer

E. P.

(2007). Racial context and police force size: Evaluating the empirical validity of the minority threat perspective. American Journal of Sociology, 113, 507-546.

65.

Tapia

(2010). Untangling race and class effects on juvenile arrests. Journal of Criminal Justice, 38, 255-265.

66.

Tapia

(2011). Gang membership and race as risk factors for juvenile arrest. Journal of Research in Crime & Delinquency, 48, 364-395.

67.

Terry

R. M.

(1967). Discrimination in the handling of juvenile offenders by social-control agencies. Journal of Research in Crime & Delinquency, 4, 218-230.

68.

Thomas

S. A.

Moak

S. C.

Walker

J. T.

(2013). The contingent effect of race in juvenile court detention decisions: The role of racial and symbolic threat. Race and Justice, 3, 239-265.

69.

Thornberry

T. P.

Krohn

M. D.

(2000). The self-report method for measuring delinquency and crime. In Duffee

Crutchfield

R. D.

Mastrofski

Mazerolle

McDowall

Ostrom

(Eds.), CJ 2000: Innovations in measurement and analysis (pp. 33-83). Washington, DC: National Institute of Justice.

70.

Tittle

C. R.

Curran

D. A.

(1988). Contingencies for dispositional disparities in juvenile justice. Social Forces, 67, 23-58.

71.

U.S. Census Bureau. (2004). National estimates by demographic characteristics—Single year of age, sex, race, and Hispanic origin. Washington, DC: Population Estimates Program, Population Division, U.S. Census Bureau.

72.

Winship

Radbill

(1994). Sampling weights and regression analysis. Sociological Methods & Research, 23, 230-257.