An evaluation of the narrowing gender gap in DUI arrests

https://doi.org/10.1016/j.aap.2011.02.017Get rights and content

Abstract

Although males account for the vast majority of those convicted of driving under the influence of alcohol and/or other drugs (DUI), female DUI convictions have increased over the past two decades. In this study, we examined the ratio of males-to-females who were court-mandated between the years 1992 and 2008 to attend the Mississippi Alcohol Safety Education Program (MASEP), a DUI intervention program in Mississippi. The data for this study came from MASEP records; the Behavioral Risk Factor Surveillance System (BRFSS); the Uniform Crime Reports (UCR); the Treatment Episode Data Set (TEDS); the National Household Travel Survey (NHTS); and National Highway Traffic Safety Administration (NHTSA), an agency within the US Department of Transportation. Augmented Dickey–Fuller (ADF) tests were used to assess the nature (i.e., convergence, divergence, or stability) of this trend and to identify predictors. The results showed that, over the 17-year period, the gender gap in DUI convictions, self-reported history of prior arrest, official drug arrests, and substance abuse treatment admissions has narrowed considerably. Results from the autoregressive integrated moving average (ARIMA) models show that three factors account for increases in the proportion of women mandated to attend MASEP: self-reported arrest prior to the DUI conviction, female admissions to substance abuse treatment, and annual miles driven. Changes in both women's behavior and law enforcement practices have increased female exposure to DUI arrests and narrowed the gender gap in DUI convictions.

Highlights

► This study examines the increasing proportion of women court-mandated to a DUI intervention program in Mississippi between the years 1992 and 2008. ► Results indicate that the increase is due to arrest history prior to the DUI conviction, female admissions to substance abuse treatment, and annual miles driven by women. ► We conclude that changes in both women's behavior and law enforcement practices have increased female exposure to DUI arrests and narrowed the gender gap in DUI convictions.

Introduction

Men account for the vast majority of arrests for driving under the influence of alcohol and/or other drugs (DUI). However, national arrest rates reveal an upward trend in the proportion of women arrested for DUI. The rate of women arrested for DUI has almost doubled since 1980. Over the past ten years, DUI arrests among men have decreased by 7.5%, while female DUI arrests have increased by 28.8% (Federal Bureau of Investigation, 2007). Since the increase in female DUI arrests was first noted (Popkin, 1989), there has been speculation regarding the factors associated with this phenomenon (Wells-Parker et al., 1996). One explanation is that changes in mechanisms of social control (i.e., laws and law enforcement practices) are responsible for this trend. An alternative explanation is that changes in women's alcohol use and driving behavior account for the narrowing gender gap. However, neither of these hypotheses has been empirically tested in multivariate models. The present study utilizes state-level data to examine the impact of social control mechanisms and women's behavior on trends in the gender ratio of individuals court-mandated to a DUI intervention program in Mississippi.

Changes in women's drinking and driving behaviors have been the object of examination, and studies have reached very different conclusions about the reasons for women's increasing involvement in fatal car crashes and in DUI arrests (Bergdahl, 1999, Schwartz and Rookey, 2008). Bergdahl (1999) examined Fatal Accident Reporting Systems (FARS) data from 1982 through 1991 to determine whether the convergence of men's and women's drinking and driving behaviors has led to the increase in fatal car crashes involving women. She found that although the number of male and female drunk drivers has decreased over this time period, women decreased their drinking and driving at a higher rate and were less likely to have tested positively for alcohol compared to men. These findings led the author to conclude that the increase in crash fatalities was due to an increase in women's driving, rather than to an increase in women's drinking and driving behavior. Bergdahl (1999) argued that societal changes such as increases in education and employment of women, rising divorce rates, and an increase in the proportion of female-headed households have increased the annual number of miles driven by women and thus increased female traffic deaths. Since 1975 the number of women licensed to drive and the annual miles driven have increased at a greater rate for women than men (Insurance Institute for Highway Safety, 2001). This increase in number of miles driven by women may also explain the increase in the proportion of female DUI arrests.

While there is evidence of convergence in men's and women's driving behaviors, Schwartz and Rookey (2008) did not think this accounted for the convergence in the number of DUI arrests between the sexes. DUI arrest statistics, self-reports of drunk driving, and BAC levels of drivers in fatal crashes were used to examine changes in drunk-driving patterns between 1980 and 2004 (Schwartz and Rookey, 2008). The authors found that while the number of DUI arrests has increased among women, there has been no systematic change in the number of alcohol-related traffic fatalities or self-reported drunken driving behavior. The lack of a positive correlation between DUI arrests, the number of alcohol-related traffic fatalities, and drunk-driving behavior led the authors to conclude that legislative measures and enhanced DUI detection, not women's behavior, account for the increase in female DUI arrests.

Legislators acted to reduce impaired driving in response to evidence that even small amounts of alcohol can impair driving and contribute to fatal crashes (Hingson and Heeren, 1999). States lowered the legal intoxication limit from a BAC of .10 to .08 (National Institute on Alcohol Abuse and Alcoholism, 2007), which Schwartz and Rookey (2008) argue widened the net used to catch impaired drivers. More women than men are expected to be affected by the change in public policy because women can become legally intoxicated after consuming fewer drinks than men (Nolen-Hoeksema, 2004) and can achieve higher blood alcohol concentrations even when adjustments are made for body weight (Mumenthaler et al., 1999). Furthermore, public sentiment against drunk drivers has led to increases in DUI detection efforts and, in turn, more DUI arrests. Law enforcement agencies have enhanced DUI detection efforts through improved officer training, updated BAC detection technology, and increased use of sobriety checkpoints and dedicated DUI patrols (Scott et al., 2006). Schwartz and Rookey (2008) argue that these changes may have increased women's representation in official arrest statistics “without any underlying change in women's actual drunk-driving behavior” (p. 646).

Bergdahl (1999) and Schwartz and Rookey (2008) make compelling arguments about the factors responsible for the increase in female DUI arrests (e.g., vehicle miles driven and changes in state laws or law enforcement practices), but these factors were not empirically tested. Both studies use data from the Fatal Accident Reporting System (FARS), a national census of all fatal motor vehicle crashes occurring on public roads since 1975 (National Highway Traffic Safety Administration, 2009a). Limitations of these studies are their use of fatal crash data and their use of BAC as the indicator of driving impairment.

Reliance on FARS to draw conclusions about DUI arrests is problematic. FARS data is not representative of the impaired driving population as a whole. First, FARS only captures the BACs of those individuals involved in deadly traffic accidents (i.e., individuals who are over the legal limit but are not killed are not included in the FARS database). For example, in Great Britain there were 460 deaths caused by drivers over the legal alcohol limit in 2007, and an estimated 14,000 alcohol-related traffic accidents in which the occupant(s) were injured (Institute of Alcohol Studies, 2010). Second, research shows that those with extremely high BACs (.15 or greater) are more likely to be involved in a crash than those with lower BACs and are 20 times more likely to be involved in a fatal crash than a sober driver (National Highway Traffic Safety Administration, 2008b). As a result, those with high BACs are overrepresented, while those with lower BACs are underrepresented. This leads to another problem with using FARS data. Women who drive under the influence of alcohol are still less likely than men to be involved in fatal accidents involving alcohol. In 2008, 25% of males who died in an accident had a BAC over .08. For women, 13% died in an accident in which the driver had a BAC over .08 (National Highway Traffic Safety Administration, 2008a). As a result, women remain underrepresented in the FARS data.

Third, both studies of impaired driving among women relied on BAC as an indicator of alcohol abuse. NHTSA collects BAC from drivers involved in fatal crashes. The quality of the BAC data has improved over time, but still leaves much to be desired. In 2002, the FARS annual report indicated that no objective BAC level was recorded for more than half of the drivers involved in fatal car crashes (Hedlund et al., 2004). One reason that information on BAC is not available is use of other drugs. After alcohol, marijuana was the most frequently detected drug among drivers in fatal accidents, and the rate of positive drug tests increased significantly from between 1991 and 2001 (i.e., from 25% to 35%, respectively) (Schwilke et al., 2006). Neither Bergdahl (1999) nor Schwartz and Rookey (2008) considered the impact drugged driving may have on the number of DUI arrests. In 2008, 11% of first-time DUI offenders in Mississippi were arrested for drug impaired driving. NHTSA has comparable findings from their 2007 National Roadside Survey, which indicated that more than 16% of weekend, nighttime drivers tested positive for illegal, prescription, or over-the-counter medicine (National Highway Traffic Safety Administration, 2009c).

Finally, there is the issue of whether or not gender differences in alcohol use are converging. Studies consistently find that alcohol consumption is considerably higher among men than women, but the gender gap is narrowing in some European countries (Wilsnack et al., 2000, Zilberman et al., 2003, Holmila and Raitasalo, 2005). The variation in the gender ratio by country suggests that changes in cultural or social structural factors may have influenced women's drinking behavior. In the United States, there is evidence that younger women are drinking more. The percent of females initiating alcohol use during adolescence now equals that of males (Brady and Randall, 1999). Furthermore, gender differences in alcohol use and heavy drinking among high school and college students has narrowed since 1975 (Johnson et al., 2009). With earlier onset of drinking and heavier drinking among females, the gender gap in alcohol dependence is also closing among young adults (Greenfield, 2001). Although young adults ages 21–35 have the highest rate of alcohol-impaired driving (National Highway Traffic Safety Administration, 2006), it is unclear whether the drinking patterns of young women has influenced DUI arrest rates among women of all ages.

The purpose of the current study is to identify factors associated with changes in the female-to-male ratio of MASEP participants from 1992 to 2008, which we do in the following two ways:

  • (1)

    We examine labor force participation, alcohol use in the previous month, admissions to substance abuse treatment, licensed drivers, and annual miles driven to test the hypothesis that changes in women's behavior increase their exposure to DUI arrest.

  • (2)

    We examine the effects of social control mechanisms (e.g., time since the BAC law changed, self-reported prior arrest, and official drug arrests) on trends in female DUI convictions.

Section snippets

Measures and data sources

Potential risk factors for DUI arrests were drawn from longitudinal data sources including MASEP records; the Behavioral Risk Factor Surveillance System (BRFSS) (Centers for Disease Control and Prevention, 2009); Treatment Episode Data Set (TEDS-A) (Substance Abuse and Mental Health Services Administration, 2008); the Uniform Crime Reports (UCR) (Federal Bureau of Investigation, 2007); the National Household Travel Survey (NHTS); and the National Highway Traffic Safety Administration (NHTSA).

Results

Descriptive statistics (Table 1) indicate that the percentage of females convicted of DUI and mandated to participate in the MASEP program increased significantly between 1992 and 2008 (i.e., from 13% to almost 19%, respectively (χ2 = 82.97, p < .001)). There has also been a statistically significant increase over the past 16 years in the percentage of women who reported having been arrested prior to their participation in MASEP (χ2 = 72.75, p < .001), and are currently employed (labor force

Discussion

The present study expands upon previously conducted research that examined the narrowing gender gap in DUI arrests (Bergdahl, 1999, Schwartz, 2008, Schwartz and Rookey, 2008) by using multivariate analysis techniques to formally test factors thought to influence risk for DUI arrest. Our results indicate that changes in both women's behavior and mechanisms of social control have narrowed the gender gap in DUI arrests. Over the 17-year study period, the overall number of vehicle miles traveled

References (54)

  • Federal Bureau of Investigation

    2007 Crime in the United States: Table 33: Ten-year Arrest Trends by Sex, 1998–2007

    (2007)
  • Federal Highway Administration
    (1990)
  • Federal Highway Administration

    National Household Travel Survey

    (1995)
  • Federal Highway Administration

    National Household Travel Survey

    (2001)
  • Federal Highway Administration

    National Household Travel Survey

    (2009)
  • Federal Highway Administration

    Quick Find: Drivers and Driver Licensing

    (2010)
  • L. Gould et al.

    DWI: An isolated incident or a continuous pattern of criminal activity

  • C.A. Green

    Gender and use of substance abuse treatment services

    Alcohol Research & Health

    (2006)
  • Greenfield, S.F., 2001. Women and alcohol use disorders. AAAP News Summer,...
  • J. Hedlund et al.

    State Laws and Practices for BAC Testing and Reporting Drivers Involved in Fatal Crashes

    (2004)
  • Hill, R.C., Campbell, R.C., 2009. Using SAS for...
  • R.W. Hingson et al.

    Preventing impaired driving

    Alcohol Research & Health

    (1999)
  • M. Holmila et al.

    Gender differences in drinking: why do they still exist?

    Addiction

    (2005)
  • Institute of Alcohol Studies

    Alcohol and Accidents

    (2010)
  • Insurance Institute for Highway Safety

    Women drivers aren’t riskier; they’re in more crashes these days because they’re driving more

    Status Report

    (2001)
  • L.D. Johnson et al.

    Monitoring the Future: National Survey Results on Drug Use, 1975–2008

    (2009)
  • W.D. Marelich et al.

    Gender differences in the control of alcohol-impaired driving in California

    Journal of Studies on Alcohol

    (2000)
  • Cited by (25)

    • Trends in female and male drunken driving prevalence over thirty years: Triangulating diverse sources of evidence (1985–2015)

      2018, Addictive Behaviors
      Citation Excerpt :

      Although men continue to far outnumber women in problematic alcohol-related consequences like DUI, some are concerned by data from the early 2000s finding “increasing participation of women in alcohol-related crashes” and have begun to “contemplate the need for a revision of extant [gendered] paradigms” (Vaca, Romano and Fell 2014: 1486). Simple drunken driving arrest statistics might amplify such concerns, with female arrests tripling between 1980 and the early 2000s (Robertson, Liew, & Gardner, 2011). However, there is need to critically analyze these concerns using multiple forms of contemporary data.

    • Predicting self-reported drink driving among middle-aged women

      2017, Accident Analysis and Prevention
      Citation Excerpt :

      While men are still more likely to drink drive overall, these trends mean that drink driving is increasingly becoming a female problem. Several proposed explanations for women’s increasing participation in drink driving has been put forward, including increased road exposure, alcohol consumption, and risk taking behavior as well as changing attitudes toward women’s alcohol use in general (Beuret et al., 2014; Robertson et al., 2011; Romano et al., 2008). However, research aimed at understanding drink driving from a uniquely female perspective is still in its infancy.

    View all citing articles on Scopus
    View full text