Final Report

 

Examination of DWI
Conviction Rate Procedures

Download PDF

Ralph K. Jones
Connie H. Wiliszowski
John H. Lacey

 

August 1999

 

Prepared for:

U.S. Department of Transportation
National Highway Traffic Safety Administration
Washington, D.C. 20590

 

Contract Number DTNH22-97-D-35018

 

Mid-America Research Institute, Inc. of New England
Winchester, Massachusetts

 

Technical Report Documentation Page

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

 

 

TABLE OF CONTENTS

FIGURES

TABLES

EXECUTIVE SUMMARY

INTRODUCTION

SCOPE AND APPROACH

ORGANIZATION OF THE REPORT

SITE DESCRIPTIONS AND ANALYSES

CALIFORNIA- RIVERSIDE COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

COLORADO - PITKIN COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

GEORGIA - ROCKDALE COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

NEBRASKA- OMAHA

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

City Prosecutor’s Office

Nebraska Governor’s Highway Safety Office

Calculated Conviction Rates

NEW JERSEY - MONMOUTH COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

NEW JERSEY - OCEAN COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

OHIO - CIRCLEVILLE AND PICKAWAY COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Circleville Municipal Court

Calculated Conviction Rates

TEXAS - TRAVIS COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Texas Department Of Transportation

Texas Commission on Alcohol and Drug Abuse

Calculated Conviction Rates

VERMONT - RUTLAND COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Governor’s Highway Safety Representative (GHSR).

Vermont Center for Justice Research.

Calculated Conviction Rates

VERMONT - WASHINGTON COUNTY

Description

DWI Enforcement System

Prosecution Patterns and Beliefs

Quoted Conviction Rates

Calculated Conviction Rates

COMPARISON OF CONVICTION RATES

QUOTED CONVICTION RATES

CALCULATED CONVICTION RATES

DISCUSSION OF CONVICTION RATES

CONCLUSIONS AND RECOMMENDATIONS

CONCLUSIONS

RECOMMENDATIONS

REFERENCES

 

 

FIGURES

 

Figure 2-1: Pitkin County, Colorado DUI Systems Flowchart

Figure 2-2: Rockdale County DUI Systems Flowchart

Figure 2-3: DWI Arrests In Omaha, 1992-1997

Figure 2-4: Omaha DWI Systems Flowchart

Figure 2-5: NJ Municipal Court DWI Systems Flowchart

Figure 2-6: Circleville OMVI Systems Flowchart

Figure 2-7: DWI Arrests In Texas, 1985-1990

Figure 2-8: DWI Convictions as a Percentage DWI Dispositions in Texas, R_C - 1996 - 1998

Figure 2-9: Vermont DWI Systems Flowchart - Charges Disposed

Figure 2-10: DWI Convictions in Vermont, 1981-1994

 

 

TABLES

Table 2-1: Project Sites

Table 2-2: Sex and Race Distribution of DWI Arrestees In Riverside County, California, 1995

Table 2-3: Dispositions of DUI Related Arrests by Pitkin County Sheriff’s Office During the First Six Months of 1995 by Reporting Date

Table 2-4: DWI Conviction Rates in the State Court of Rockdale County, Georgia, 1994 - 1996

Table 2-5: Circleville/Pickaway County Project Data

Table 2-6: Quoted Conviction Rates, R_C, FY 1996, for Travis County, Texas

Table 2-7: State of Vermont - Criminal DWI* Charges and Convictions, 1990 -1994

Table 3-1: Calculated Conviction Rates in Study Sites

 

 

EXECUTIVE SUMMARY

This is the final report of a study of methods used to calculate conviction rates for DWI offenses. Conviction rate is important as one measure of the performance of the adjudication subsystem of the Traffic Law System in a given jurisdiction. Conviction rates can also be used to measure the performance of the components of the adjudicative subsystem.

The objective of the project was to examine various conviction rate formulas, but was not concerned with the success or lack thereof of convicting persons charged with DWI offenses in the participating jurisdictions. We commend individuals in those sites who agreed to participate in the project and who provided data for our study, general system information, and views on the subject.

Specific objectives of the project were to:

·         examine methods of calculating conviction rates by using data from ten sites;

·         determine the implications (pro and con) of each of these methods; and

·         develop methods for estimating conviction rates within, and across, jurisdictions to provide objective, accurate, and consistent information.

To do this, we developed:

·         a description of the kinds of conviction rates constraints reported by the project sites along with their methods of calculating system-wide DWI conviction rates where applicable; and

·         methods for determining or estimating system-wide conviction rates, along with the justification for the use of the methods.

We found through our discussions with practitioners in the field that the common public perception of a DWI conviction rate is the percentage of persons who are arrested for DWI who are actually convicted of that offense. However, we also found that in practice the DWI conviction rates which are reported are seldom that simple number, but rather the conviction rate of some subset of DWI arrests, or that convictions for other offenses are included in the calculations. Given that finding, it is evident that there are any number of types of calculations which are represented as DWI conviction rates. In other words, what is reported as the DWI conviction rate in one jurisdiction may be an entirely different calculation than what is reported by another jurisdiction as its DWI conviction rate. For that matter, agencies within a jurisdiction may have similar differences.

Thus, when comparisons are made between jurisdictions on this important issue, it may well be like comparing apples to oranges. A jurisdiction which is receiving plaudits that reports only the results of cases which come to court, while many other cases are falling by the wayside, may well have a true conviction rate much lower than a jurisdiction which accurately reports a conviction rate that on the surface appears to be lower.

The major conclusions of the study are:

·         There is great diversity among and within jurisdictions in what is reported as their DWI conviction rate.

·         Many jurisdictions have difficulty relating DWI convictions to arrests made within a specific time frame.

·         In many jurisdictions, failure to appear (FTA) in court, is a frequent problem with DWI offenders.

·         Many agencies and jurisdictions choose to neither calculate nor publicize their DWI conviction rates.

·         There is a quite natural tendency for agencies within jurisdictions to calculate and report that form of conviction rate which places them in the most favorable light.

We recommend that:

·         When requesting DWI conviction rates, specify both the numerator and denominator, and request the raw numbers and the time period that the data are from, in addition to the calculated result.

·         NHTSA should consider the true conviction rate to be the number of DWI convictions resulting from and divided by the number of DWI arrests in a given time frame. FTAs clearly would not be counted as convictions because these cases have not been adjudicated. Plea bargains might be counted as convictions if a jurisdiction has more than one level of DWI offense and it is common practice to reduce the charge to the lower level. If a jurisdiction includes plea bargains, it should be explained clearly that the conviction rate includes these alcohol-related, plea bargained offenses.

·         NHTSA should consider encouraging jurisdictions and agencies to keep detailed records on the disposition of DWI arrests so that meaningful and consistent DWI conviction rates may be developed, published and compared. Statewide case tracking systems using information from a uniform traffic citation should be encouraged as a means for maintaining such records.

We believe the lack of credible field calculations of DWI conviction rates and the inconsistent methods of calculations should remain under careful scrutiny. Furthermore, we believe that agencies such as NHTSA should publicly define how DWI conviction rates should be calculated. Only then can progress be made in identifying and correcting system deficiencies.

 

 

INTRODUCTION

 

This is the final report of a study of methods used in several jurisdictions to calculate conviction rates for DWI offenses⁽¹⁾. It is designed to help answer such seemingly simple questions as, “given the number of drivers arrested for DWI, how many are convicted of a DWI offense?” What may appear to be an easy, straightforward question may be, in fact, impossible to answer across jurisdictions and populations.

In this study Conviction rate is important as one measure of the performance of the adjudication subsystem of the Traffic Law System in a given jurisdiction. The adjudication subsystem has the responsibility of determining whether a person apprehended by a law enforcement officer for violating a traffic law actually violated that law. It follows that, in a given jurisdiction, the larger the number of arrested drivers who are convicted (either by pleading or by conviction in a trial court), the higher the performance of the adjudication subsystem as a whole. (This rule, of course, is subject to the constraint that fundamental fairness is provided in all adjudicative proceedings.)

Conviction rates can also be used to measure the performance of the components of the adjudicative subsystem. For example, a prosecutorial agency may divide the number of DWI convictions by the number of cases charged, while a court administrator might divide the number of DWI convictions by the number of DWI case dispositions. Other factors, such as failures to appear, pre-trial diversion programs, etc. can influence the overall rate depending on whether they are included in the calculation. But here we are concerned with the performance of the system as a whole.

In addition to the inconsistency of the factors discussed above, adjudication and case-tracking time periods could vary and must be taken into account. There are potential problems with record keeping of multiple enforcement agencies policing the same jurisdiction and with a determination of the level of offense, for example, misdemeanor or felony, which can sometimes change when records are checked or if a death occurs after a DWI-related crash. Also, several courts sometimes handle DWI cases within a jurisdiction.

 

OBJECTIVES

The general objective of this project was to study various methods used across the United States to calculate conviction rates for DWI offenses. The National Highway Traffic Safety Administration (NHTSA) receives many requests for information regarding conviction rates of DWI offenders. However, there is no standard method of calculating a conviction rate.

Specific objectives of the project were:

·         to examine different ways of calculating conviction rates by using data from ten sites;

·         to determine the implications (pro and con) of each of these methods; and

·         to develop methods for estimating conviction rates within, and across, jurisdictions to provide objective, accurate, and consistent information.

This report seeks to accomplish these objectives by providing:

·         a description of the kinds of conviction rates constraints reported by the project sites along with their methods of calculating system-wide DWI conviction rates; and

·         methods for determining or estimating system-wide conviction rates, along with the justification for the use of the methods.

SCOPE AND APPROACH

This project involved several tasks. First, lists of prospective sites along with names of contacts and telephone numbers were compiled. The respective NHTSA regional offices were notified of the project and several responded with possible jurisdictions within their area that might be interested in participating as project sites. Sites were not selected based on the efficiency of their judicial and adjudication systems, nor on any belief that a conviction rate was “high” or “low,” but only had to be able to supply us with the necessary data and, whenever possible, methods used to calculate DWI conviction rates. However, we did try for diversity among the sites regarding such factors as geographic location and population size.

Telephone interviews were conducted with appropriate site contacts to obtain information about conviction rate formulas, sources of data, and interest in project participation. Once general information was received from a site, and the prospect of cooperation and interest seemed high, a site visit was made by project staff to arrange to receive appropriate data. In some areas, it was not necessary to visit the site to obtain all of the necessary information and records. Ultimately, ten sites were selected for study.

A very generalized data collection plan was written early in the project year. However, specific details had to be worked out as data arrived from the project sites. As expected, the data varied greatly from site to site as to:

·         availability (e.g. was there a tracking system which could provide data),

·         condition (e.g. accurate and up-to-date),

·         sources (e.g. was the same data reported differently from different agencies, also state versus local), and

·         medium (e.g. what types of computer systems and software were in use and whether some data was only available in paper form).

Analyses of the data was a continuous process. As data arrived from sites, records were checked and in some cases, sampling procedures had to be used to extract a certain number of files. The numbers were sometimes entered into the conviction rate formulas, and formulas from similar sites, and the results were compared to those obtained by using the current conviction rate formula at that site.

ORGANIZATION OF THE REPORT

This report contains five chapters and an appendix. In Chapter 2, we present a site-by-site description and analysis which includes both calculated conviction rates and conviction rates quoted by staff who process DWI cases. This is followed by a comparison of conviction rates across sites (Chapter 3). Chapter 4 presents our conclusions and recommendations, and Chapter 5 lists the references.

 

 

SITE DESCRIPTIONS AND ANALYSES

 

As indicated in the prior chapter, this project was concerned with methods for calculating the conviction rate of a jurisdiction’s overall adjudication system. The exact way of determining such a rate is simply to count the number of those found guilty (either as a result of a plea or a trial) as a percentage of a specified group of those arrested for DWI. The “specified group of those arrested” is most meaningfully defined as those arrested in a given period, say one year. Thus, the formula for the overall conviction rate, R, is:

R = 100 H (Number of DWI convictions of those arrested in a given period / Number of those arrested in the given period)

To compute R, one must start with the denominator. If the jurisdiction is a municipality served by just one law enforcement agency, then counts of DWI arrests by officers of that agency must be available for the period of interest. If all arrestees are adjudicated in just one court, then the dispositions of all these arrests must be available. Note that, because it takes time to process a case, not all of the dispositions occur in the same year of the arrests, but may be stretched out over a period of two or more years. Thus, one can never know the conviction rate for the current year or even the conviction rate for the prior year.

The situation becomes more complicated for jurisdictions with multiple police agencies and multiple courts. In general, a multiple-agency case tracking system (preferably, managed at the state level) is needed for such jurisdictions. In a word, the formula for the overall conviction rate is simple, but getting the data is not.

Jurisdictions without the requisite systems in place to compute an exact conviction rate (in our experience, most jurisdictions) have to resort to estimates. Two possible ways of estimating the conviction rate are:

R_EST1 = Number of convictions in a given period / Average annual number of arrests over a given period preceding the period for which the convictions were calculated

R_EST2 = Number of convictions in a given time period / Number of arrests in the same time period

Clearly, the first of these two estimates would be preferred in most instances. For example, if a jurisdiction’s court(s) had 3,000 convictions in 1998 and its police agencies arrested an average of 2,000 drivers per year over the past three years, then its estimated conviction rate R_EST1 would be 67%. The second estimate should only be used as a last resort where there is little year-to-year variation in arrests and convictions, because the convictions would be for a different time period than the arrests, and therefore would be a rough estimate. In one jurisdiction, we were able to calculate both the exact rate and the first estimated rate (R_EST1) above.

Finally, six jurisdictions only had data for computing the conviction rate for the court subsystem of the overall system, that is, a rate based on total dispositions. Note that this rate measures the performance of only a part of the court subsystem, since it does not account for defendants who fail to appear or, in some jurisdictions, whose cases are dismissed. We present this rate as matter of interest, since it was often quoted as a jurisdiction’s conviction rate. The formula for the court subsystem (designated R_C) is simply:

R_C = Number of guilty dispositions in a given period / Number of dispositions in a given period

This chapter presents the results of our attempts to calculate these rates in each of the following 10 jurisdictions (Table 2-1). Also presented in this chapter are detailed case studies of the jurisdictions’ adjudication systems and circumstances pertinent to determining the conviction rates.

Table 2-1: Project Sites

* BAC is the acronym for blood and breath alcohol concentration levels expressed as grams per deciliter.

**Texas currently has a BAC legal limit of .08 but the limit was .10 at the time of this study.

CALIFORNIA - RIVERSIDE COUNTY

Description

Riverside County, California borders the counties of Los Angeles, San Diego, Orange and San Bernardino and is the fourth largest county in the State of California. Between 1980 and 1990, the population rose 76%, according to the U.S. Census Bureau, making it the fastest growing county in California. The population rose another 21% from 1990 to 1998 (and almost 3% from 1997 to 1998). The population of Riverside County in 1998 was 1,478,800 with 86% living in urban areas and 14% in rural areas. Per capita income was reported by the Census Bureau at $14,510. The city of Riverside is the county seat and also the largest urban area within Riverside County.

DWI Enforcement System

DWI enforcement is handled by the California Highway Patrol (CHP), the sheriff’s department and the Riverside Police Department. Reportedly, the bulk of DWI arrests are handled by CHP. The sheriff’s department handles most issues surrounding warrants. The BAC per se level in California is .08.

Prosecution Patterns and Beliefs

DWI is a criminal offense in California handled by municipal and superior courts. Municipal traffic courts handle non-injury, first, second and third offense DWI cases which are considered misdemeanors. In these cases, the person accused signs a ticket at the scene or receives an order to appear at arraignment. Apparently, a significant percentage of persons charged with traffic offenses fail to appear at the court arraignment, although it is unclear how many of these are DWI related charges. Superior court handles all injury-related DWI cases and those involving four or more offenses. In these cases, the persons charged are to be arrested and held pending arraignment. However, due to overcrowded detention facilities, these individuals are often released. After their first court appearance for arraignment where the charges against the individual are read along with their status (for example, bailed out or cited out), these individuals are scheduled for a preliminary hearing before proceeding to court.

Quoted Conviction Rates

According to the Riverside District Attorney’s Office, DWI conviction rates are not calculated with any regularity, rather they rely on the state reports. However, they were one of the first agencies to question statistical methods used to calculate DWI conviction rates, and the apparent lack of uniformity among calculations. They suspected problems might stem from various reporting strategies by law enforcement agencies and how arrests are entered into databases used by state agencies as part of the statistical equations. While staff in the Riverside District Attorney’s Office were questioning how DWI conviction rates were calculated in the State of California, they contacted NHTSA to question if there was an accepted method of calculation. This inquiry was, in part, the impetus for this project.

Calculated Conviction Rates

We used the state-level data provided by the California DMV. DWI arrests during the calendar year 1995 were tracked through December 1997, permitting an accurate estimation of R. Out of the 9,190 arrests, 5,870 ultimately were convicted resulting in a conviction rate of:

R= 5,870 / 9,190 = 0.639 = 63.9%

The data also permitted the calculation of a few characteristics of the DWIs. Table 2-2 shows the sex and race distribution of the arrestees, indicating that 88% were males. About 90% were either white or Hispanic, with the Hispanic group being slightly the larger of the two. About 6% were African American, and Asian and American Indian were only about one-half percent each.

The mean age of the arrestees was 34.3 years, and the mean age of those convicted of DWI was essentially the same, 34.6 years. The arrestee group had a slightly higher percentage of younger drivers (age < 21 years) than did the offender group (6.8% versus 5.7%).

 

COLORADO - PITKIN COUNTY

Description

Pitkin County, Colorado is located in the Rocky Mountains near the center of the state. The largest urban area in the county is the town of Aspen which is also the county seat. The U.S. Bureau of the Census reported a 1998 population of 13,423. According to the 1990 U.S. Census Data, roughly 60% of county residents at that time resided in rural areas. Per capita income in 1993 was reported as $39,481, but had fallen to $26,755 in 1998.

DWI Enforcement System

The Pitkin County Sheriff’s Office (PCSO) has provided an informal coordinating function for DUI enforcement activities for Pitkin County law enforcement agencies for the past several years. This has largely grown out of a grant application and administration function as the cooperating agencies have periodically made application to the Colorado Department of Transportation for funding for enhanced DUI enforcement. Enforcement of DUI laws are considered an important element of patrol activities by the police agencies in the county and regular, publicized saturation patrol DUI enforcement efforts are mounted. The Sheriff’s Deputy who serves as the countywide DUI enforcement coordinator also keeps detailed information on arrests and dispositions, which are used for local purposes as well as for documentation to be included in grant applications for anti-DUI activities. Note that the Sheriff’s Department keeps DUI arrest and disposition records for the local enforcement agencies (Aspen Police Department, Snowmass Village Police Department, Basalt Police Department, Carbondale Police Department as well as their own alcohol-related arrests) but not the Colorado State Police operating within Pitkin County. The discussion of DUI conviction rates here is confined to PCSO DUI cases. But similar data are maintained for these other agencies.

Prosecution Patterns and Beliefs

DUI cases are adjudicated in the 9^th Judicial District which serves Pitkin and two other counties. DUIs (with the exception of offenses such as vehicular homicide while DUI) are misdemeanors in Colorado. The Pitkin County Court hears PCSO DUI cases at the Pitkin County Court House in Aspen. Cases are prosecuted by prosecutors from the 9^th Judicial District Attorney’s Office. In Colorado, the DUI offense is a per se offense at the .10 BAC level. Though clearly impaired drivers may be convicted of DUI at levels lower than .10, as a practical matter, most are not. However, there is a lesser included offense of Driving While Ability Impaired (DWAI) with a presumptive level of .05 and above. Unless extreme impairment is evident, if a DUI arrestee is tested and has a BAC between .05 and.10, the charge is immediately amended to be DWAI and they are tried for that offense. For the purposes of conviction rate calculations, local authorities consider a DWAI conviction to be a conviction as charged for this category of case. However, a DWAI conviction for a DUI charge is counted as a conviction for a lesser included offense. In practice, many non-crash involved first DUI offenders with BACs below .18 are allowed to plead guilty to DWAI rather than go to trial. By statute, DUI or DWAI may not be reduced to a non-alcohol related offense unless the District Attorney states in open court that he/she does not have a prima facie case. Such instances are extremely rare in Pitkin County as are instances of dismissal. Another rare occurrence is to offer deferred judgment and sentence where the case is dismissed if terms of a probation are served successfully.

The adjudication process in Pitkin County is outlined below.

Figure 2-1: Pitkin County, Colorado DUI Systems Flowchart

 

Quoted Conviction Rates

 

As indicated earlier, the Pitkin County Sheriff’s Office, with the assistance of the District Attorney’s Office, compiles and reports detailed conviction rates for DUI cases. These reports indicate the current status of all cases charged in a specific time frame. The example presented here is for arrests made by PCSO during the first six months of 1995. We present below Table 2-3 with three columns. It represents information from two reports prepared by the PCSO, the first in August 1995 and the second in August 1997.

 

Table 2-3: Dispositions of DUI Related Arrests by Pitkin County Sheriff’s Office During the First Six Months of 1995 by Reporting Date

One can see by examination of the table above that cases are tracked using date of arrest as the defining element of eligible cases and that the reporting is dynamic in nature in that the reported dispositions were updated for the two illustrative reporting dates of August 1995 and August 1997.

 

Calculated Conviction Rates

We were provided the raw data used by the PCSO in calculating their DWI conviction rate for the 34 drivers arrested during the first six months of 1995. These individuals were tracked manually by PCSO staff to determine the outcomes of their cases as of August 1997. The conviction rate for those charged or plea-bargained of either DUI or DWAI was:

R = 32/34 = 94.1%

This is exactly the same as indicated in the preceding table under the column for disposition as of August 1997 (33% convicted of original charge plus 61% convicted of DWAI = 94%), and is the correct measure of the performance of the County’s adjudication subsystem in processing the 34 arrests. The remaining two arrests (5.9%) resulted in a failure to appear (FTA) in court as of the reporting date (August 1997). All of those who did appear were convicted of an alcohol-related offense, either DUI or DWAI. Conceivably, at some later date, the two FTAs could appear and, possibly, be convicted, resulting in a conviction rate of 100%.

 

GEORGIA - ROCKDALE COUNTY

Description

Rockdale, the county with the smallest area in Georgia, is southeast of Atlanta. The county encompasses small urban, suburban and rural areas. The county seat is in Conyers, Georgia. According to the Bureau of the Census, the population of Rockdale County has been increasing steadily from 36,600 in 1980 to a 1995 population of 64,500. Per capita personal income for the County in 1993 was $19,267.

DWI Enforcement System

In Georgia, the DWI offense is a per se offense at the .10 BAC level. DWI laws are enforced by three agencies, the Rockdale County Sheriff’s Department, the Conyers Police Department, and the Georgia State Patrol. DWI arrests by the Georgia State Patrol (GSP) and the Sheriff’s Department are adjudicated in the State Court of Rockdale County. The Rockdale County court is, as the name implies, classified as a Astate court,@ and has one judge who hears about 500 DWI cases a year. (See study by Jones, R.K.; and Lacey, J.H., 1998.) DWI arrests by the Conyers Police Department are adjudicated by the Conyers Municipal Court which also operates the City’s Traffic Violations Bureau. The Municipal Court has less than half the volume of DWI cases as the State Court.

DWI tickets are brought to the prosecuting clerk’s office. DUI defendants do not get a chance to come to court until prosecutors file the case. This allows prosecutors time to prepare the case before the speedy trial clock starts. The laws regarding speedy trials require a case to be tried within the remainder of the current court session or by the end of the next court session. Each court session is three months long, so a case going to trial will be completed within a maximum of 180 days. A Criminal Case Management System has been in place since 1987. This system tracks each case through the criminal justice system including jail. The State Court judge keeps statistics on every case, getting information on Apleaders@ just before the pronouncement of a guilty verdict and then using that information to structure sentences.

Prosecution Patterns and Beliefs

Police officers in Rockdale County believe the courts there impose tougher sentences than do other counties in Georgia. As with most other jurisdictions, few DUI cases go to trial; one officer reported that out of 800 DUI arrests, he only testified at four jury trials. A prosecutor reported that only four out of 62 DUI cases on the docket on a particular day would be tried. Officers, nevertheless, must be prepared to take all cases to trial. Preparation will result in more guilty pleas without any trial. A prosecutor reported that written, documented information is extremely important.

Pre-sentence investigations are not routinely performed, although the judge does question suspects at arraignment and at trial just before sentencing. The following diagram depicts the flow and possible outcomes of DUI cases handled in Rockdale County.

Figure 2-2:Rockdale County DUI Systems Flowchart

 

Quoted Conviction Rates

Conviction rates were not being calculated at the time of this report by the courts nor the law enforcement agencies operating within the county.

Calculated Conviction Rates

We were able to locate data for determining the dispositions in the State Court of Rockdale County of all the DWI arrests made by the Sheriff’s Department and the GSP in 1994, 1995, and 1996. The arrest date were obtained from the two enforcement agencies, and the dispositions were obtained from the Judge’s own database as maintained by court staff. The results are shown in Table 2-4 below.

The last column contains the conviction rate, R. The highest rate occurred in1994 (86%), with 1995 and 1996 having the same rate at 78%. The average rate for all three years was approximately 80%. The rates are approximate, since a few arrests from the Conyers Police Department went to the State Court rather than the Municipal Court, and a few arrests from the other two agencies went to the Municipal Court rather than the State Court.

Note that the numbers of convictions shown did not necessarily occur in the arrest year. For example, six persons out of those who were arrested in 1994 and eventually convicted, were not convicted until 1996. However, nearly all of those arrested in a given calendar year were convicted by the end of the following calendar year.

Table 2-4: DWI Conviction Rates in the State Court of Rockdale County, Georgia, 1994 - 1996

 

NEBRASKA - OMAHA

Description

Omaha is located in Douglas County, Nebraska and covers most of the county. Douglas County is located mid-center on the eastern border of Nebraska, directly west of Council Bluffs, Iowa. Major industries include insurance companies, telephone services, oil refining and agricultural related industries. According to U.S. Census Bureau data, the total resident population of Douglas County in 1995 was roughly 434,000 persons. The 1990 census data reported approximately 396,000 persons lived inside the Douglas County, Nebraska urbanized area (Omaha) and approximately 20,500 individuals lived outside the urbanized area and in rural areas. The Douglas County per capita 1993 income was reported as $23,322.

DWI Enforcement System

Various discussions were held during the course of this project with the City Prosecutor and staff about the county court system. The number of court filings (all cases) has increased dramatically during the past twenty years from 10,118 in 1977 to 61,000 in 1997. (Note: Until mid-1985, there were the following courts in Omaha: Municipal Court, County Court and District Court. On July 1, 1985, the Municipal and County Courts were merged and so currently there is County Court and District Court.) Individuals in the city prosecutor’s office had estimated 85% of the county population live within the city of Omaha, while the census data above indicates a figure closer to 95%. Due to this population placement within Douglas County, the Omaha police department (OPD), as opposed to multiple enforcement agencies, arrests the majority of DWI suspects. For this project, we concentrated on suspects arrested by OPD and adjudicated in County Court.

The number of arrests for DWI has ranged from about 3,300 to 4,000 per year during the six year period 1992-1997 (Figure 2-3). In Nebraska, the DWI offense is a per se offense at the .10 BAC level. When an officer makes an arrest, the resulting ticket number and information is sent to the Douglas County Retention Center where the data is entered into the criminal justice information system. The data center merges the information on the ticket with various automated items and the cover sheets are printed and sorted by date. The attorneys in the city prosecutor’s office then decide what charges to set. Reportedly they decline to prosecute only about one in every 200 cases. These few cases are captured in the system and classified as dismissed charges although the reasons for dismissal are not noted. The system codes are not finely defined.

Figure 2-3:DWI Arrests In Omaha, 1992 - 1997

Prosecution Patterns and Beliefs

Charges of DWI are considered by the City Prosecutor’s office to be the most important cases handled by the County Court and are actively prosecuted as has been the practice for many years. The City Prosecutor is proud of the fact that they actively pursue DWI offenders. The following diagram depicts the flow and possible outcomes of DWI cases.

Figure 2-4:Omaha DWI Systems Flowchart

The majority of DWI defendants plead guilty. There was a time when the city prosecutor’s office might have 50 trials set and have six trials Ain a bad month.@ Prosecutors estimated winning 90% of their trials for a long period of time and the number of trials declined to a fraction of the previous levels (the current rough estimate is 12 trials per year). But the number of requests for trials has not declined, in part, because a third offense is a class one misdemeanor which brings a mandatory jury trial. In those cases, a jury trial must be waived by the defendant; otherwise, a trial is automatic. So, the number of potential jury trials is still fairly high, but the number of actual trials is relatively low. There are 30-40 requests per month for third offense DWI cases.

There is another dynamic change that took place during the course of this project. The state legislature passed a law which makes a fourth offense DWI a felony. This means the office of the county attorney (not the city prosecutor) will be handling felony DWI offenders. The County has handled some pending cases in the past, but the majority of cases have been handled by the city prosecutor’s office. However, this change began in 1998 and does not affect this project or the 1997 data provided.

Batches of DWI files come in daily and the cases are divided by the prosecutors and set for court dates. Under the current system, the attorneys look at criminal history, which includes both a traffic printout and criminal history. This will show if the defendants had a DWI charge dismissed previously. The attorneys will also look for such things as “dummy names.” If the city prosecutor’s office finds a defendant with three or more prior DWI convictions, the case will now be sent to the county attorney to handle as a felony.

Quoted Conviction Rates

City Prosecutor’s Office. Conviction rates are calculated internally by the city prosecutor but are not used for the purpose of publication. When asked to quote a rate, such as was reported in the Omaha Morning World Herald article in 1997, “City’s DWI Conviction Rate was 98% in >96,” the following calculation is used:

Conviction Rate = R_C = Number of guilty dispositions / Number of all dispositions

There is a real concern about releasing a conviction rate to the media and to the public because of perceived misinterpretations of what the rate represents. This is magnified if FTAs are factored into the calculation because in areas with a large number of outstanding warrants (incidentally, not the case in Omaha), the rate dramatically decreases. The public and the offenders will most likely get the message that it is possible to avoid retribution if they can avoid the system.

In addition, theoretically, some prosecutors could be worried enough about conviction rates to get rid of cases on the front end during the charging process which they might lose to improve or maintain a high conviction rate.

Reportedly, there are probably several dozen cases a year calculated as lost cases which actually were prosecuted successfully for a more serious crime. For example, willful reckless and, occasionally, reckless are amended charges. Both of these are class 3 misdemeanors which carry considerably higher penalties. Sometimes offenders accept the more serious amended charges rather than the stigma of a DWI with a possible enhanceable future offense. The defendant pleas to the more serious charge and that’s the only fallback position the prosecutor’s office gives the offender. The city prosecutor’s office counts that as a lost case for a DWI conviction but it shows up on the county database as a conviction for the amended charge.

Another problem is the Adismissal docket@ where a batch of old cases is sent to the City Prosecutor’s office and staff are asked to get rid of the cases when the persons charged do not appear in court and a lengthy period of time has passed. They find this process incredulous, stating, “Let’s reward these people for not bothering to show up for trial. They’re not discharged for speedy trial. These are people with legitimate warrants.” The prosecutors appear and argue and sometimes win, allowing the cases to remain, but sometimes the cases are dismissed despite their arguments. The City Prosecutor considers FTAs as pending cases, although sometimes pending for a long period of time. AYou can’t count a pending case as a disposition.@ Pending cases are still viable.

After thinking about this project, reviewing a draft report and reflecting on the way local conviction rates had been calculated in the past, the Omaha City Prosecutor concluded they had been keeping track of cases that had been lost, whether declined, or dismissed or not guilty. They really have a Anon-conviction rate of 2%.@ FTAs were not counted in the equation. He now would likely recommend selecting a number of cases (e.g., beginning November 1^st, the next 100 cases will form the pool), write down the arrest numbers so these cases may be electronically retrieved, return after a one year period (or some specified period of time), find out the dispositions for those cases, decide on how (or if) to report FTAs, and calculate the conviction rate. This eliminates using arrests and convictions for different cases from different periods of time in the same formula.

Again, the City Prosecutor offered words of caution. In Omaha, he is proud that police and prosecutors work together to prosecute cases where the defendant is at or below the legal limit, when officers know the person was impaired and a danger to the public. “If you start becoming a slave to your conviction rate, the temptation is, if we’re not winning very many .10 cases, don’t fool with them so our conviction rate doesn’t suffer.” He warns, “...conviction rates can eventually trickle down to not only dictating filing decisions, but even arrest decisions.” “We really should be encouraging our prosecutors to trust our law enforcement officers.”

Nebraska Governor’s Highway Safety Office. The State of Nebraska calculates DWI conviction rates by county from year to year. However, they caution that this is a Araw calculation@ and should only be used to provide a general indication of conviction rate trends. The rates are calculated starting with convictions recorded in a calendar year on driver records files, divided by arrests in a calendar year, that is:

Conviction Rate = R_EST2 = Number of Convictions / Number of arrests

The Governor’s Highway Safety Office points out that, in some cases, the convictions in the numerator are for a different calendar year than the arrests in the denominator. In addition, the driver records are for drivers licensed in Nebraska, and so do not indicate out of state drivers. In Omaha, these numbers could be significant because the city is located on the border of Iowa. And, conversely, residents of Douglas County arrested in another county, or perhaps out of state, appear in the convictions group but not in the arrests made in Douglas County. There are other factors, such as FTAs are not accounted for, and the cases are combined to cover all enforcement agencies and courts within Douglas County (although, as stated above, the vast majority of DWI arrests are made by the Omaha Police Department and adjudicated in County Court by the city prosecutor’s office).

 

Calculated Conviction Rates

The data we used to calculate conviction rates consist of 4,089 records of DWI charges extracted from the city prosecutor’s files for 1997. The variables include the arrest number, charge(s), incident date, statute, defendant date of birth, driver license number, BAC level, disposition and the date of the disposition. The possible findings (dispositions) include: guilty, probation, suspended sentence, juvenile (transferred to juvenile court), dismissed (not prosecuted and those found not guilty) and other (mis-assigned). The database does not distinguish between cases lost, dismissed at time of trial and later refiled, or rarities such as if the defendant died, all of which appear as dismissed dispositions. The finding will show up as not guilty if a trial was held and the defendant was found not guilty. There was concern about the possibility of being unfairly tagged with numerous dismissals for other reasons which would inaccurately be labeled as cases that the prosecutors did not win.

As stated above, some of the system codes are not finely defined. For example, if a person is arrested for drunk driving and convicted of willful reckless driving, initial charges are recorded in one column, and in another column, there is the amended or disposition charge which would be different than the drunk driving charge. The problems are that there is an inability to differentiate between

1.) what was done initially at filing (see next paragraph) because of, for example, a low BAC or,

2.) what was amended at trial or found guilty by the court for a lessor included charge.

An additional problem is that the database always lists the first charge as the primary charge, which may not have been the case at the time of arrest. If there are three different charges, the prosecutor can change the order of the charges when setting the case which affects how the record shows up in the database. These data files show if a case is dismissed or an offender is convicted along the line, but do not show reduced charges. The city prosecutor’s office was uncertain how reduced charges were recorded.

We first calculated an estimate of the overall system conviction rate, R. To do this, we used data covering just the first six months of DWI arrests (January-June 1997). This allowed a at least a full year after arrest for case disposition, ensuring disposition of just about all cases for which the defendant appeared in court.

There were a total of 1,993 such cases, 1,774 (88.0%) of which resulted in a finding of guilty (by plea or by trial). The remaining 12% of the dispositions were composed of a not guilty finding (3.7%) and no finding (8.3%). The not guilty findings include cases for which the DWI charge was dismissed. The Ano finding@ cases apparently consist of those where the persons failed to appear in court at some point along the judicial process so that no disposition of these cases occurred. These cases are considered a good estimate of the failure to appear rate for Omaha.

The data allowed one to compute dispositions as a function of defendant age. There was no significant difference (p=0.50) in case disposition between young drivers (age < 21 years) and older drivers (age $21 years).

The estimated conviction rate R_EST1 was computed for the year 1997 using arrest data from 1995 and 1996 (see Figure 2-2), and conviction data from 1997. This rate was 3,514 / 3,800 = 92.5%, about 4.5 percentage points higher than R.

We also calculated the conviction rate based on total dispositions in a year. The year was 1997, and R_C turned out to be 3,514 / 3,665 = 0.959, indicating that about 96% of all DWI case dispositions were guilty findings.

Finally, as a matter of general interest, we calculated the mean time from arrest to disposition for Omaha DWI defendants. For defendants arrested in the first six months of 1997, an average of 2.93 months were required to dispose of a DWI case. Cases resulting in a finding of not guilty took slightly longer on the average than did cases resulting in a finding of guilty (3.16 months and 2.92 months, respectively), reflecting the higher likelihood of a trial for the not guilty group. This difference between means, though small, is statistically significant (p < 0.0001).

To summarize, our calculated conviction rates were:

R = 0.880 = 88.0%

R_EST1 = 0.925 = 92.5%

R_C = 0.959 =95.9%

 

NEW JERSEY - MONMOUTH COUNTY

Description

Monmouth County, New Jersey is located south of Newark. The U.S. Census Bureau reports a population of 603,434 up 1.1% from 1997. Roughly 90% of the population resides in urban areas. Per capita income in the County is listed at $20,565.

DWI Enforcement System

New Jersey State Police and various Municipal Police Departments handle most DWI arrests. In New Jersey, the DWI offense is a per se offense at the .10 BAC level. All law enforcement agencies use uniform traffic tickets (UTT) when making a DWI arrest. These traffic citations are all compiled by the statewide Automated Traffic/Complaint System. Police dispatchers have access to this system which, in turn, provides officers in the field with valuable information.

Prosecution Patterns and Beliefs

In New Jersey, a DWI offense is not considered a crime, but is labeled a Aquasi-crime@ by statute and is handled as a traffic matter by the municipal courts. There are 538 municipal courts in the state of New Jersey. A very small percentage (estimated at 1%) of DWI cases are handled in Superior Court and only when another serious crime, some indictable offense, has been committed in conjunction with the DWI offense, which is considered a non-indictable offense. These cases must be sent to the prosecutor’s office and adjudicated in Superior Court. The New Jersey DWI Systems Flowchart (Figure 2-5) appears on page 25.

All law enforcement agencies use UTTs when making a DWI arrest. These traffic citations are all compiled by a statewide Automated Traffic/Complaint System (ATS/ACS). This system has reduced case backlog and reduced time to disposition, which was a major concern before the ATS/ACS was implemented. All municipal courts have access to this system and depend on the system for administrative and statistical information. Conviction rates are not routinely calculated, but if there were reason to do so, the rate would most likely be calculated using numbers generated from the ATS/ACS.

Although a civil matter, DWI offenders can still receive jail time. There are no administrative per se or administrative license removal (ALR) laws in New Jersey. Judicial court proceedings are required to be quick which counters the need for ALR. In fact, it is a goal of the New Jersey Administrative Office of the Courts to have DWI cases adjudicated within 60 days. The DMV is notified only of convictions.

It is the practice of the New Jersey Court Municipal Services Division to not factor in those cases where the person has failed to appear in court into the conviction rate equation. Those cases are not considered closed and therefore have not reached any final disposition.

Quoted Conviction Rates

A prosecutor’s office in this county confirmed what we learned from the New Jersey Municipal Court Services Division that DWI conviction rates are not calculated locally in Monmouth County on a regular basis. If a need should arise to perform that calculation, the appropriate state agencies would be notified to provide either a rate, or the numbers necessary to allow a calculation to be performed.

It is the practice of the New Jersey Court Municipal Services Division to not factor in those cases where the person has failed to appear in court into the conviction rate equation. Those cases are not considered closed and therefore have not reached any final disposition. However, the agency states if asked how they would calculate a conviction rate, they would first determine the total number of cases that were disposed (guilty, not guilty, dismissed) and then divide that number into the total number of guilty judgments.

Calculated Conviction Rates

We received statewide data for New Jersey for the year 1995 in the form of 31,636 DWI adjudication records from the State of New Jersey Municipal Court Services. There were 2,930 records for Monmouth County. Each record represents a DWI charge for which there was a finding. Possible findings and their percentages for Monmouth County were:

1. Guilty 63.0%

2. Not Guilty 25.1%

3. Dismissed 10.4%

4. Merged 1.5%

The designation Amerged@ is given to charges stemming from incidents with multiple charges. These charges were merged with some other charge. Failure to appear was accounted for in the database only if the defendant eventually did appear. Records for defendants who had not appeared when the database was queried would not have been included in the output.

Thus, these data allowed us to calculate only the conviction rate for the court system, R_C = 63.0%.

NEW JERSEY - OCEAN COUNTY

Description

Ocean County is located on the New Jersey coastline opposite Burlington County which contains Camden. The U.S. Census Bureau reports a population of 489,819 up 1.5% from 1997. Roughly 80% of the population resides in urban areas. Per capita income in the County was listed in 1993 at $22,849.

DWI Enforcement System

New Jersey State Police and various Municipal Police Departments handle most DWI arrests. In New Jersey, the DWI offense is a per se offense at the .10 BAC level. All law enforcement agencies use uniform traffic tickets (UTT) when making a DWI arrest. These traffic citations are all compiled by the statewide Automated Traffic/Complaint System (ATS/ACS). Police dispatchers have access to this system which, in turn, provides officers in the field with valuable information.

Prosecution Patterns and Beliefs

In New Jersey, a DWI offense is not considered a crime, but is labeled a Aquasi-crime@ by statute and is handled as a traffic matter by the municipal courts. There are 538 municipal courts in the state of New Jersey.

Figure 2-5: NJ Municipal Court DWI Systems Flowchart

A very small percentage (estimated at 1%) of DWI cases are handled in Superior Court. This happens when another serious crime, some indictable offense, has been committed in conjunction with the DWI offense, which is considered a non-indictable offense. These cases must be sent to the prosecutor’s office and adjudicated in Superior Court. The flowchart in Figure 2-5 depicts the flow of DWI cases in Ocean County.

As previously stated, all law enforcement agencies use UTTs when making a DWI arrest. These traffic citations are all compiled by a statewide system, the ATS/ACS. This system has reduced case backlog and reduced time to disposition, which was a major concern before the ATS/ACS was implemented. All municipal courts have access to this system and depend on the system for administrative and statistical information.

As soon as the complaint summons is issued, an initial arraignment is set for the next available court date, which is typically within just a few days. The purpose of the arraignment is to make certain the accused individual has a copy of the complaint, understands the charges, is advised of the sanctions he or she would face if found guilty of the charges, and to determine if an attorney’s services are available and/or needed, or if the person wishes to request a public defender. The individual charged of a DWI offense may plead guilty at the initial arraignment. However, most people who enter a guilty plea do not do so at this point in the judicial process. Most guilty pleas are entered after the case is scheduled for trial disposition and after the defense attorneys have received whatever discovery materials they are entitled to (police report, breath test results, investigation reports, etc.). A high percentage of persons who are arrested for DWI do plead guilty based on New Jersey statutes that state a person can be proved guilty under:

·         per se which states that if the breath test has resulted in a BAC equal to or higher than .10, then that person legally is under the influence, which is all the State has to prove to convict; or

·         the statute dealing with BAC test refusal which states, if the breath test has been refused, there is an automatic license suspension and an inference that the person is guilty due to the refusal; or

·         other attendant circumstances that the person was impaired, that the use of alcohol deleteriously affected the person’s ability to operate a motor vehicle.

Although a civil matter, DWI offenders can still receive jail time. The DMV is notified only of convictions. There are no administrative per se or administrative license removal (ALR) laws in New Jersey. Judicial court proceedings are required to be swift which counters the need for ALR. In fact, it is a goal of the New Jersey Administrative Office of the Courts to have DWI cases adjudicated within 60 days. Although this is not always feasible, the Presiding Judges strive to encourage the Municipal Courts and Judges to handle cases within 60 days from date of filing to order of disposition.

Since 1994, when the New Jersey legislature approved funding for the specialized positions, Presiding Judges have provided expertise to Municipal Courts and Municipal Court Judges in administrative and procedural matters. The reasons for nominating Presiding Judges were to reduce case backlog and to expedite the hearing of serious traffic matters such as DWI by having Presiding Judges advise and train other Municipal Court Judges in efficient methods. Presiding Judges are appointed by the Chief Justice of the New Jersey Supreme Court with the advice and consent of the Assignment Judges. Once a Presiding Judge is appointed by the Chief Justice at the request of the Assignment Judge of a certain county, that person remains the Regional Presiding Judge until such time as there is a change in status based on another request made by the Assignment Judge to the Chief Justice. Presiding Judges meet regularly at the State Administrative Office of the Court where occasional supplemental training is provided to help them perform their functions, such as administrative training, backlog reduction, and substantive materials to help Judges better understand processes or trends. Training for Municipal Prosecutors is provided through the County Prosecutor’s office as designated by the State Attorney General’s Office.

The first time an individual fails to appear (FTA) in court, a failure to appear notice is forwarded to that defendant. If the person does not appear on the return date, after receiving the notice, usually a bench warrant is issued. The Municipal Court Judge has the discretion to issue a bench warrant immediately, even for a missed first appearance, if the Judge believes the magnitude of the charged offense is serious enough, such as drunk driving especially when a serious crash or fatality has occurred.

Quoted Conviction Rates

A Presiding Judge in New Jersey confirmed what we learned from the New Jersey Municipal Court Services Division and from a prosecutor’s office in Monmouth Count and that is, DWI conviction rates are not calculated locally in this county. If a need should arise to perform that calculation, the appropriate state agencies would be notified to provide either a rate, or the numbers necessary to allow a calculation to be performed. The Presiding Judge reported that the major concern has been with case disposition rates and rates pertaining to length of time to disposition. This information is provided by the ATS which is heavily relied upon to make sure cases are handled expeditiously. By Order of the New Jersey Chief Justice there is a mandate to move cases to disposition in 60 days which, as stated previously, is a goal of the New Jersey Administrative Office of the Courts.

It is the practice of the New Jersey Court Municipal Services Division to not factor in those cases where the person has failed to appear in court into the conviction rate equation. Those cases are not considered closed and therefore have not reached any final disposition. However, the agency states if asked how they would calculate a conviction rate, they would first determine the total number of cases that were disposed (guilty, not guilty, dismissed) and then divide that number into the total number of guilty judgments.

Calculated Conviction Rates

Again, we used data from the State of New Jersey Municipal Court Services for 1995. (See page 24 for discussion). There were 2,351 records for Ocean County. As with Monmouth County, each record represents a DWI charge for which there was a finding. Findings as a percentage of all charges were:

1. Guilty 68.7%

2. Not Guilty 29.1%

3. Dismissed 1.3%

4. Merged 0.9%

From the above, the conviction rate for the court system, R_C = 68.7%. Available data did not allow us to compute system conviction rate, R.

 

OHIO - CIRCLEVILLE AND PICKAWAY COUNTY

Description

Circleville is located in Pickaway County, Ohio, a rural area which lies directly south of Columbus (Franklin County). According to U.S. Census Bureau data, the population of Pickaway County was 52,500 in 1996 and has been growing steadily (1980 population of 43,700). In 1990, the Census Bureau reported a population of 48,300 with 11,700 residing in urban areas and 36,600 living in rural areas. Per capita income in 1993 for Pickaway County was $15,050. According to our site contacts, Circleville currently has a population of approximately 14,000 and population of the County is approximately 55,000.

Circleville is a town with major industries such as E.I. DuPont, PPG Industries, Thomson Consumer Electronics, and Owens Illinois. Two major thoroughfares, Interstate 71, and U.S. Route 23, bisect the County. Interstate 71 is the major thoroughfare between Cincinnati and Cleveland.

DWI Enforcement System

The Pickaway County Sheriff’s Department has retained computerized arrest records since 1992 which allows them to access predetermined forms from their database program. The Circleville Municipal Court tracks cases which are adjudicated in that court and provided us with data for this project. The Ohio Bureau of Motor Vehicles receives a copy of the original ticket and a copy from the court once disposition is made. Reportedly there is a court statistics reporting plan with a goal of obtaining uniform data from all the courts. But, as pointed out by the Municipal Court judge, courts do not report data in the same way and use of such data would be erroneous.

DWI cases, referred to as OMVIs (operating a motor vehicle under the influence) in Ohio, may be cited under state code or local ordinance and handled in the Municipal Court System, which is a state wide court system. Municipal Courts have jurisdiction by statute, which means the jurisdiction of each court depends on how that court was initiated by the state legislature. This particular Municipal Court covers the entire county. Of the 168 Municipal Courts in Ohio, about 25 of them are countywide.

There have not been any major law changes recently relating to OMVI offenses in Ohio. After arrest all people taken into custody appear in Municipal Court initially, and cases are sorted by the seriousness of the offense. Felony cases go to Common Pleas Court through grand jury indictment. The misdemeanor cases, which means all cases which have a potential penalty of less than one year in jail, remain in Municipal Court. (If the defendant is under 18 years of age, the case would be initiated in or transferred to Juvenile Court.)

The next step would be to accept a guilty plea or proceed to trial. Occasionally there are amendments to the charges, which would happen sometime after arrest and, of course, before disposition. Occasionally there are problems with individuals arrested who fail to appear at arraignment or trial. In these cases, warrants are issued for arrest and that information goes into the state’s law-enforcement automated data system. It varies by jurisdiction as to when or if those warrants are served. Certain sheriff departments have active warrants squads designated to make arrests specifically when the warrants are forwarded to their agency for processing. In the departments which do not have specific warrant execution details, arrest typically does not occur unless there is subsequent contact between a law-enforcement agency and the individual who has absconded.

Prosecution Patterns and Beliefs

By state statute, from time of arrest through adjudication, unless the defendant waives time, the case is to be tried and disposed of within 90 days. Figure 2-6 depicts the flow and possible outcomes of OMVI cases handled in Municipal Court. Although depicted in the diagram, plea bargaining resulting in charge reduction is not a common practice. The majority of persons arrested on OMVI charges plead guilty. The majority of cases which do proceed to trial are also found guilty. The approach of the Municipal Court in Pickaway County is to try to reduce OMVI recidivism by interposing strategic sentencing plans.

Figure 2-6: Circleville OMVI Systems Flowchart

 

Quoted Conviction Rates

Circleville Municipal Court. Conviction rates have been described as high in Pickaway County by the Sheriff’s Department, the state’s Governor’s Highway Safety Representative’s (GHSR) office and the Court. The presiding judge is known to be progressive and innovative when sentencing convicted offenders. Our contact at the Sheriff’s Department believes this has contributed to the high conviction rates. He also has said the judge has been very particular about the presentation of OMVI cases which makes it clear what is expected of the officers. There appears to be a good working relationship between the enforcement and adjudication organizations as well as the GHSR and regional NHTSA personnel. Strategic sentencing involves use of an evaluation test, targeted probation goals, intense supervision and polygraph-enforced compliance as well as distinct license plates for DUI offenders. According to our site contact, the number of DUIs and alcohol-related fatalities have been reduced significantly.

An actual OMVI conviction percentage rate is not routinely calculated in Circleville. Per conversations with site contacts, if asked, the computation for the year of concern would most likely be as follows:

R_C = Number of guilty findings / Number of all findings

Calculated Conviction Rates

The data we used to calculate conviction rates consist of 618 OMVI adjudication records obtained from the Circleville Municipal Court database. These records comprise OMVI cases disposed of in 1997 and early 1998. The violation dates range from 1995 through early 1998 as follows.

Table 2-5: Circleville/Parking County Project

Note that these records indicate individual charges, and account for more than one charge per incident.

The variables include ticket number, driver license number, date of birth, gender, zip code of residence, violation date, charge, DWI test type, DWI test results, plea, charge reduced (Y/N), plea finding and date of disposition. Pleas are coded into three categories: guilty, not guilty and no contest. Some sentencing codes are merged when there is more than one charge and there has been a guilty plea or finding. When individuals fail to appear, those cases are usually coded AF@ for bond forfeiture.

In the calculations, we used only those records containing data on the primary DWI offense. Thus, the results pertain to the outcomes of a single alleged DWI incident. The data file described above could only be used to calculate convictions as a percentage of total findings, which turned out to be:

R_C = 406 / 445 = 0.912 = 91.2%,

where the denominator is the total number of DWI findings in 1997 and the first quarter of 1998, and the numerator is the number either pleading guilty or found guilty at trial during the same period. Only four of the 406 guilty defendants (0.9% of the 445 charged) were found guilty at trial; the remaining 402 plead guilty at some point in the adjudication process.

Those pleading not guilty (24% of those charged) fared much better during adjudication. Their conviction rate was only 66.0%, with all but three of their not-guilty findings resulting from dismissals.

The sex and age distributions of the DWI defendants were available in the data file. Approximately 84% were males, and the mean age was 34.6 years. Only about 6% of the defendants were under the age of 21 years.

BAC data were available for 292 of the defendants. The mean BAC for this group was .16, with 92% over .10. Twenty-one percent of the defendants refused to take a BAC test, and their conviction rate was slightly lower than that of non-refusers (90.5% vs. 93.5%).

About 12% of the cases involved reduced charges, all the result of a guilty plea. We performed several analyses of court case processing time. The mean time over all defendants was slightly under two months (57.9 days). As expected, processing time was much higher for those pleading not guilty (96.6 days). About 82% of the cases were disposed in 90 days or less.

TEXAS - TRAVIS COUNTY

Description

Travis County, Texas is located in east, central Texas and includes the state capital of Austin. The U.S. Census Bureau reports a population in 1996 of 664,800, up from a population in 1990 of 576,400. The 1990 population was classified as 91% urban and 9% rural. The 1996 per capita income for the county was $21,127. The state capital, Austin, is located in Travis County, Texas.

DWI Enforcement System

Project staff visited Austin, Texas and met with staff at various agencies relative to the project: the Texas Department of Transportation (TXDOT), the Department of Public Safety (DPS) Traffic Law Enforcement Division, the Driver Information Center (DIC), the Austin Police Department (APD), Travis County Sheriff’s Department, the Travis County Prosecutor’s Office, and the Texas Alcohol and Beverage Commission (TABC). All of the contacts proved helpful in sharing information and appeared interested in the project.

The Department of Public Safety (DPS) is the Texas state police agency. The Department enforces traffic laws on Texas highways (federal, state, and county) through the Traffic Law Enforcement (TLE) Division. The Division includes specific services, the largest of which is the Highway Patrol. The TLE tracks DWI arrests, but not convictions. This holds true for the other enforcement agencies we talked with (APD, the Sheriff’s Department, and TABC). None of these enforcement agencies were aware of DWI case disposition rates.

Prosecution Patterns and Beliefs

There is a two-tier court system for handling DWI offenders. The courts which handle first-time offenders (unless, for example, there is a felony charge for a fatality) are the County Courts and the County Court of Law Courts. Multiple offenders and/or felony cases are handled in District Courts.

The Travis County Prosecutor’s Office has an information department which records case dispositions, but the Trial Chief was not comfortable with the accuracy of that information. He believed the DWI conviction rate would be in the 80% range if it were to be calculated.

It does not appear that the agencies we met with have tracked DWI conviction rates in the past.

Quoted Conviction Rates

Texas Department Of Transportation. TXDOT and DPS are two of the principal agencies in Texas which collect and track data pertinent to this project. TXDOT receives arrest data from DPS and the Office of Court Administration (OCA). TXDOT reported that the conviction rate for Texas for 9/1/95 through 9/31/96 could be quickly calculated at 63% for County Court based on the annual report for the fiscal year 1996 published by the Texas Judicial Council (Table 2-6). District Court activity reflected a 65% conviction rate. The formula used was:

Conviction Rate = R_C = Number of guilty dispositions / Number of all dispositions

As stated by persons at other project sites, there could be a public relations backlash if there are misinterpretations of what the rate represents.

 

Table 2-6: Quoted Convictions Rates, Rc, FY 1996, for Travis County, Texas

 

Texas Commission on Alcohol and Drug Abuse. The following information was taken from the report, ADWI Recidivism in Texas: 1985 through 1988” published by the Texas Commission on Alcohol and Drug Abuse.

Again, the conviction rate formula used was:

Conviction Rate = R_C = Number of guilty dispositions / Number of all dispositions

The number of arrests state wide for 1985-1990 were obtained, and as shown below in Figure 2-7, fluctuated during that time period. However, the arrests are reported as a total by calendar year and so we were unable to perform additional calculations. (Arrests from one year may not be adjudicated until the following year or later.)

Figure 2-7: DWI Arrests In Texas, 1985 - 1990

 

Calculated Conviction Rates

The Texas Judicial Council provided monthly court data on the dispositions of DWI cases during the period January 1996 through October 1998. Available data did allow a determination or an estimate of the overall system conviction rate. Only conviction as a percentage of all disposed cases (R_C) could be calculated. A total of 9,857 cases were disposed of during that period, an average of 290 per month or 3,479 per year. 6,917 of all dispositions were convictions, resulting in a conviction rate of:

R_C = 6,917 / 9,857 = 0.702 = 70.2%

Figure 2-8 shows how this rate varied with month over the 34-month period. The highest conviction rate during this period was 79.7%, and the lowest was 61.6%.

 

Figure 2-8: DWI Convictions as a Percentage DWI Dispositions in Texas, Rc - 1996 - 1998

VERMONT - RUTLAND COUNTY

Description

Rutland County, Vermont is located on the New York state border just north of Bennington County which borders Massachusetts and New York in the southwest corner of Vermont. The largest urban area in the county is the city of Rutland. The U.S. Bureau of the Census reported a 1995 population for Rutland County of 62,732. According to the 1990 U.S. Census Data, 71% of county residents at that time resided in rural areas. Per capita income in 1993 was reported as $18,588.

DWI Enforcement System

DWI cases are adjudicated in District Court where all cases (from DWI to murder to shoplifting) are filed, both misdemeanors and felony cases. There are 14 Districts in the state-wide court system. In Vermont, if there are two prior DWI convictions within 15 years, the third conviction is a felony with sentencing of up to five years in jail. If there is serious injury or death, the charge is automatically a felony even for a first offense. Appeals are heard in Supreme Court which is the court of last resort. The adjudication process in Vermont is outlined in the Figure below.

Figure 2-9: Vermont DWI Systems Flowchart - Charges Disposed

 

Prosecution Patterns and Beliefs

There are prosecutorial policy differences between the charges of DWI filed in Rutland County and those filed in Washington County, Vermont (the final site described next in this report). Typically, although not always, a single DWI charge is filed in Rutland County. This charge might be “under the influence” or “having an excess of .08 alcohol in your blood system while operating a motor vehicle@ also known as ADWI .08.” (As discussed in that section, both counts are typically charged in Washington County.) More often, the charge would be ADWI under the influence.@ ADWI .08” would be the charge when it would be more difficult for a jury to find a defendant under the influence beyond a reasonable doubt, for example, in instances where the individual can not perform physical dexterity tests as required by field sobriety test requirements due to physical limitations or handicaps, but a breath test revealed an alcohol concentration or BAC of >.08.

Charge reduction policy is apparently the same in both Rutland and Washington Counties. Prosecutors go forward to trial with substantially strong cases that they believe they can prove beyond a reasonable doubt. But if there are substantial problems such as problems with the evidence, then based on experience, prosecutors may reduce the charges to Athe closest thing we can bargain for@ which is usually reckless or careless driving.

In Vermont, if a blood or breath test is taken more than two hours after motor vehicle operation, the BAC must be related back to operation time. But if the chemist can relate the BAC back at the time of the driving to >.08, then that fact is enough to convict and it is not necessary to have the arresting officer testify as to the definition of impairment when attempting to prove an Aunder the influence@ charge.

The assistant state’s attorney we talked with reported that he is not aware of conviction rates being calculated routinely in their office. If this is done, it would not be for publication, but most likely for administration purposes.

One area of repeated frustration to prosecutors has been the problem of getting HGN (horizontal gaze nystagmus) results into trials. Vermont case law allows the use of the HGN test, but as with all scientifically or medically related evidence, it must meet the rules of evidence under the evidence codes. There have been several decisions which say there must be a trained officer who has performed the HGN test correctly and there must be an expert witness to testify to the validity of HGN and its relevance to proving impairment. Finding expert witnesses has proven to be difficult, if not impossible, in Rutland County (and reportedly in other Vermont Counties as well), and there are no funds to pay those expert witnesses, should they be found. All of this is especially frustrating because a leading training academy in the state urges police officers to conduct the HGN evaluation first, in general because they believe it is the most validated, most standardized test. “That’s all well and good but it assumes we have a...(chance) to get” this into court and in front of a jury which has not been possible.

Quoted Conviction Rates

No conviction rates were reported from our contacts in Rutland County. However, some conviction rate information for the state as a whole was obtained and is presented below.

Governor’s Highway Safety Representative (GHSR). The GHSR office in Vermont reported that all of the raw data statewide is sent either to The Vermont Center for Justice Research or the Vermont Incident Based Reporting System (VIBRS), a state police agency records system which is gradually evolving into a statewide records network. This was echoed by the Management of Information Systems (MIS) Director for the statewide Court Administrator who reported all raw data is sent monthly to the Vermont Center for Justice Research as well as the DMV for driver histories, and the criminal information center for criminal histories. The Court Administrator’s office does not routinely calculate conviction rates. It appears most agencies rely on the Vermont Center for Justice Research for conviction rate calculations and information dissemination.

Discussions were held with the Director of the Vermont Center for Justice Research at Norwich University in Montpelier, Vermont. The Center operates a statewide system which collects prosecution and adjudication data through disposition and sanctions from District Courts. The Director of the Center suggested choosing two sites in Vermont because DWI offenders are charged differently in Rutland County than in Washington County. In addition, Rutland County was recommended as a site because the Rutland Police Department has been connected to the VIBRS records system since 1993.

Vermont Center for Justice Research. We were provided with several reports based on 1995 (and earlier) data which reported that there were slightly higher dismissal rates and lower conviction rates since the BAC level for DWI was reduced from .10 to .08 in 1991. However, statewide, they reported in 1995 that the likelihood of conviction for the DWI charges brought forward remained extremely high with the rate based on prosecuted charges. Roughly 12.1% of DWI charges were reportedly changed or dismissed during the years 1980-1994. Conviction rates as calculated by The Vermont Center for Justice Research are presented in Table 2-7 for the years 1990 - 1994. The table calculates conviction rate as:

Conviction Rate = Total number of convictions / Total number of prosecuted charges

 

Table 2-7: State of Vermont - Criminal DWI* Charges and Convictions, 1990 - 1994

* Charges are for DWI, DWI 2nd or subsequent, DWI death resulting and DWI Refusal.
** Based on prosecuted charges Source: Vermont Center for Justice Research, 5/95

 

The Vermont Center for Justice Research reports that, statewide, DWI convictions rose in the mid-1980s, but have remained fairly constant at about 3,000 per year since that time as indicated by the graph in Figure 2-10 below.

Figure 2-10: DWI Convictions in Vermont, 1981-1994

Calculated Conviction Rates

Each District Court records and maintains data for all criminal motor vehicle offenses. The Vermont Court Administrator’s Office (CAO) coordinates the electronic data collection from each county’s District Court in the state. The CAO forwards this information to the Vermont Department of Motor Vehicles (motor vehicle crime and citation data) and the Vermont Center for Justice Research for statistical analyses and reporting. (The Vermont Center for Justice Research supplied the data for this project.)

According to a traffic records system audit performed in 1995 by the Vermont Center for Justice Research, criminal motor vehicle offenses are maintained by the CAO in a complex relational database containing approximately 200 variables. Most statistical analysis done by the CAO is descriptive in nature and provides summary information such as the number of cases filed and disposed in each District Court by month.

The raw data we received in conjunction with this project for Rutland County, Vermont consists of 825 records for calendar years 1995 and 1996 of DWI charges extracted from the state database maintained by The Vermont Center for Justice Research. The 825 records include those for multiple charges flowing from the same incident. The variables include the docket number, defendant date of birth, gender, date of offense, original statute charge, offense level (felony, misdemeanor), plea, final charge statute, final plea, court disposition, disposition date and originating police department. The possible findings (dispositions) include: convicted, probable without verdict, not guilty/insanity, acquitted, dismissed, civil procedure, off calendar, guilty/mental, transferred to juvenile court, mistrial, nolle prosequi, other extradited, not disposed by court and missing/unknown.

The data provided permitted us to calculate R₂, but not R₁,. Our analyses were concerned with the dispositions of the primary DWI charge (775 records), and included the dispositions of cases for which the original charge was reduced, for example, from DWI-second offense to DWI-first offense.

A disposition of guilty for all classes of DWI (not including refusal to take a BAC test) as a percentage of all dispositions was calculated as:

R_C = 626 / 775 = 0.808 = 80.8%

Counting refusals as a DWI would increase this rate to 87.4%.

A disposition of guilty for all classes of DWI (including refusals) or for the reduced charge of reckless driving as a percentage of all dispositions was calculated as:

R_C’ = 726 / 775 = 0.937 = 93.7%

VERMONT - WASHINGTON COUNTY

Description

Washington County, Vermont is located in the north central region and includes the state capital of Montpelier. The U.S. Bureau of the Census reported a 1995 population of 56,367. According to the 1990 U.S. Census Data, 68% of county residents at that time resided in rural areas. Per capita income in 1993 was reported as $19,616.

DWI Enforcement System

DWI cases in Vermont are adjudicated as described on page 37 for Rutland County.

Prosecution Patterns and Beliefs

Charges of DWI are filed differently in both Washington and Bennington Counties in Vermont from other counties statewide. Typically two counts are filed even though it is possible for the defendant to be found guilty of only one charge. The two charges are “under the influence” and the second alternative charge is “having an excess of .08 alcohol in your blood system while operating a motor vehicle.” So whenever there is a breath test, the case is filed with both counts. For every case where these two counts of DWI are filed, if there is a guilty plea or finding, one charge must be dropped. Commonly called “DWI under the influence” and “DWI .08” the jury is instructed that the defendant, if found guilty, may only be convicted of one charge only. The other charge is dismissed by the jury, by the court, or by prosecutor if there is a change of plea. Consequently, the statistics may appear deflated depending on the reporting method in the sense that it could appear they have lost approximately 50 percent of their trial cases.

While the method of filing two charges may, at first, be more confusing for the jury, the advantages are quickly apparent. For example, with hard core alcoholics who do not exhibit observable symptoms while driving under the influence, that is, their speech is okay, balance is okay because they are professional drinkers, but they have a high BAC, the prosecutor can ask the jury to merely look at the BAC level. In other words, the jury does not have to convict on the DWI under the influence charge if the observable symptoms are not there, but they instead can convict on the DWI .08 charge because it is against the law to drive with a certain amount of alcohol in your system. It gives the jury two scenarios to consider, either of which they may use to convict the defendant.

Our site contact thinks double charging has given prosecutors a huge advantage, along with the fact it is a crime to refuse a breath test which reportedly has made a tremendous difference, especially in the case of a fatal crash. If the driver refuses a breath test after a crash resulting in a fatality, and if the state proves that police had grounds to believe the driver was drinking and involved in a fatal crash, and refused to submit to a breath test, that refusal, in itself, carries a 15 year licensing penalty. AI can prove someone refused a test far easier than I can prove someone was drunk.@ This law has made a tremendous difference.

Charge reductions usually occur when the state attorney has difficulty proving the charges. If a DWI charge is reduced, the resulting charges are usually careless driving or reckless driving. Reckless driving carries a 2 year (maximum) licensing sanction and careless driving is a one year sanction (maximum penalty).

DWI conviction rates are not calculated or tracked in Washington County. They do receive information on convictions from the Court Administrator’s Office (CAO) and the Vermont Center for Justice Research.

The state’s attorney’s office reports that local and state enforcement agencies have some very good officers who gather very good evidence. It was estimated that fifty percent of DWI arrests are handled by state police and 50 percent are handled by the four or five local city police departments. The sheriff’s department typically does not handle any DWI arrests. The Washington County office of the state’s attorney works with the police to educate them that the breath test results are not “the end all for DWIs.” For example, a motorist calls in to report a suspected DWI who is driving “all over the highway” and the dispatcher neglects to get the name and telephone number of the person reporting the offense. The officers need to report in their affidavit who called in the complaint. The person who reported the DWI is typically a Adynamite witness.@ Some officers don’t understand the rules of evidence or hearsay, they don’t understand that they cannot testify as to what the missing motorist said. (Dispatchers have now been instructed to always record the name and phone number of individuals who reported suspected impaired drivers.)

In Vermont, if a blood test is taken more than two hours after motor vehicle operation, the BAC must be related back to operation time. This is difficult to calculate if the person’s drinking patterns are not known. And if the driver refuses to talk, it is difficult because the time frame of the first drink, last drink, etc. is not known.

As in other jurisdictions, relatively few DWI cases proceed to trial. And there are not a lot of cases where either the evidence of a person’s guilt is very strong, or vice versa, where there is no chance of successful prosecution. What usually happens is the good, solid cases do not go to trial; instead, the suspects plead guilty. And the really weak cases are either reduced, or thrown out or are not amended. The remaining few cases proceed to trial.

On tracking DWI conviction rates among jurisdictions, the state’s attorney offered words of caution:

“...And of course you have to keep in mind the prosecutor, you must watch the statistics in general, since a prosecutor can control his statistics. What is he filing? I could have a 100 percent conviction rates in everything if I only file the cream of the crop. And the DWI cases which are >dogs,= don’t file them as DWI charges, file them as careless driving. If my job performance is measured by statistics, of the next 50 DWIs, I might only file the 30 which I am certain to win. I look like a superstar.”

While other prosecutors who are filing all cases including difficult cases, trying to see what they can successfully prosecute, are going to have lower conviction rates. It might be the prosecutors are incompetent; it might also be that they are very competent, but are filing very tough cases. “We are probably the only kind of people who can control our own stats. Most people can’t....” “That is the first caviate, what (cases) are they filing?@ “Or the way they have played the game with the stats perhaps.” In some counties stats are calculated in the clerk’s office by counting docket numbers and the funding is often dependent upon how many cases are filed. The bigger counties get more deputies. AIf you want to bump up your stats, you can do that by filing multiple cases instead of one which gives the appearance of a larger case load. So let’s assume I have a DWI against somebody and the facts support careless driving as well. If I chose to file that new charge on a separate docket number, it would look like two people I had charged that day. Instead of filing it as a second count of the original docket number. And then it will only show up as one kind of case. We probably file 1,600 cases a year. If I played the docket number game, I could get up to 2,800.@

The court clerks send their stats from closed and pending cases to the Court Administrator’s office. The Court Administrator tracks how efficient the judge is, how many cases are back logged, how many felonies or misdemeanors, etc. are handled by each District Court.

Reportedly, Washington County does not have a problem with FTAs; 96-98% of the people who are cited show up in court typically because they are local. AThey show because they know they will be arrested if they don’t and they don’t have the money or inclination to flea anywhere. In a small state like Vermont, the cops know who you are and where to find you if you don’t show.@

Quoted Conviction Rates

No conviction rates were reported from our contacts in Washington County. However, some conviction rate information for the state as a whole was obtained and is presented under the section on Rutland County, Vermont, beginning on page 39.

Calculated Conviction Rates

Each District Court records and maintains data for all criminal motor vehicle offenses. The Vermont Court Administrator’s Office (CAO) coordinates the electronic data collection from each county’s District Court in the state. The CAO forwards this information to the Vermont Department of Motor Vehicles (motor vehicle crime and citation data) and the Vermont Center for Justice Research for statistical analyses and reporting. (The Vermont Center for Justice Research supplied the data for this project.)

According to a traffic records system audit performed in 1995 by the Vermont Center for Justice Research, criminal motor vehicle offenses are maintained by the CAO in a complex relational database containing approximately 200 variables. Most statistical analysis done by the CAO is descriptive in nature and provides summary information such as the number of cases filed and disposed in each District Court by month.

The raw data consists of 1,522 records for calendar years 1994, 1995 and 1996 of DWI charges extracted from the state database maintained by The Vermont Center for Justice Research. The 1,552 records include those for multiple charges flowing from the same incident. The variables include the docket number, defendant date of birth, gender, date of offense, original statute charge, offense level (felony, misdemeanor), plea, final charge statute, final plea, court disposition, disposition date and originating police department. The possible findings (dispositions) include: convicted, probable without verdict, not guilty/insanity, acquitted, dismissed, civil procedure, off calendar, guilty/mental, transferred to juvenile court, mistrial, nolle prosequi, other extradited, not disposed by court and missing/unknown.

The data provided permitted us to calculate R_C, but not R. Our analyses were concerned with the dispositions of the primary DWI charge (843 records), and included the dispositions of cases for which the original charge was reduced, for example, from DWI-second offense to DWI-first offense.

A disposition of guilty for all classes of DWI as a percentage of all dispositions was calculated as:

R_C = 672 / 843 = 0.797 = 79.7%

A disposition of guilty for all classes of DWI or for the reduced charge of reckless driving as a percentage of all dispositions was calculated as:

R_C’ = 817 / 843 = 0.969 = 95.9%

 

COMPARISON OF CONVICTION RATES

In this chapter we summarize and compare the conviction rates developed in the prior chapter for the ten study sites. Both quoted rates and calculated rates are discussed.

QUOTED CONVICTION RATES

When asked for their DWI conviction rate, most of the sites (seven out of ten) quoted it as a percentage of total DWI dispositions in some period, usually a year. As noted earlier, such a rate serves as a measure of the performance of the court subsystem, but does not measure the performance of the adjudication system as a whole. In addition, a state agency quoted the conviction rate for the study site in its state as the ratio of convictions in one calendar year to arrests in that year. In this case, not all of the reported convictions would be for arrests in that year, but rather some from previous years.

Three sites used case tracking data in their quoted conviction rate, providing a conviction rate that measured the percentage of a cohort of DWI arrestees that ultimately were convicted of DWI. This rate is equivalent to the overall adjudication system conviction rate.

The type of conviction rate quoted was largely a function of the type of agency doing the quoting and the data available to that agency. Police agencies were more likely to quote rates based on arrests, while prosecutors and court administrator were more likely to quote rates based on total dispositions. Several jurisdictions pointed out the sensitivity of conviction rate information which may be misinterpreted and/or used for political purposes. For this reason there is a general reluctance to quote conviction rates based on arrests which are inherently lower than rates based on total dispositions. This is especially true in jurisdictions with high failure-to-appear rates where the lower conviction rates may not be due to prosecutorial failures and in fact are beyond the control of the prosecutor.

CALCULATED CONVICTION RATES

The data provided by the study sites enabled us to calculate several conviction rates. The results are summarized in Table 3-1 below. Four of the ten sites had data for calculating the overall adjudication system rate (R), and seven had data for calculating the rate for the court subsystem (R_C). Data from two of the sites enabled us to determine an alternate court subsystem rate calculated by including conviction on a reduced charge (R_CN). Estimates of R (R_EST1 and R_EST2) were possible in two sites, both of which also had data for calculating R correctly (see Table 3-1).

The sites with data for determining R were quite diverse, ranging from Riverside, California, for which data were provided by the State DMV’s DWI tracking system to Pitkin County, Colorado, which had data on the dispositions of the arrests by the Pitkin County Sheriff’s Office. Another site (Rockdale County, Georgia) gave us access to arrest data from two enforcement agencies and from the court the related adjudication information, and a city prosecutor at a fourth site (Omaha, Nebraska) paved the way for us to receive their records from the County Information System as well as arrest information from the city police department.

The objective of the project was to examine various conviction rate formulas, but was not concerned with the success or lack thereof of convicting persons charged with DWI offenses in the participating jurisdictions. Individuals in those sites who agreed to participate in the project should be commended for providing data for the study, general system information, and views on the subject.

Table 3-1: Calculated Conviction Rates in Study Sites

As one can see from the table above, there is a variety of conviction rates within overall rates and court subsystem rates and meaningful interpretation of those rates requires an understanding of details such as the definition of a conviction, how the calendar is treated and how failures to appear are treated. These issues are discussed further in the following section.

 

DISCUSSION OF CONVICTION RATES

Jurisdictions which track DWI data use a variety of methods to calculate conviction rates due to differences in state laws (pre-trial diversion), enforcement practices, adjudication practices (plea bargains), schedules (speedy trial laws) and extenuating circumstances such as arrestees who fail to appear (FTAs) to answer charges of DWI offenses. Even within a jurisdiction, different calculations may be used depending upon the needs of the agency tracking such information. The police officers might want to know how many individuals that they arrest on charges of DWI are then convicted. Prosecuting attorneys might be interested in the number of cases that are not prosecuted successfully. Therefore, an enforcement agency may divide convictions by the number of arrests to arrive at a conviction rate, while the office of a prosecuting attorney may divide convictions by the number of cases entering the specific court. If a state agency were tracking conviction rates for the same jurisdiction, that agency might be following all cases within that jurisdiction based on driver license numbers and residency addresses, which would allow for arrests and/or prosecution of residents of that jurisdiction outside of that area. Also, not all cases may end up in one court. Additionally some agencies may consider only convictions for the originally charged offense as a conviction in their calculations, while others may consider lesser included offenses to be DWI convictions.

And all of these interests may change if a news agency asks for a DWI conviction rate to quote to the public. There is always a consider of possible misinterpretations of what a rate represents. This threat is very real, because a rate which includes FTAs and those pled down to a lessor offense would be lower and might send a message either that anti-DWI adjudication is not a high priority in the area or that it is possible to avoid retribution if an offender just drops out of the system. Additionally, several persons interviewed for this project indicated if prosecutors are judged solely on a conviction rate on cases tried, then those conviction rates will increase. However, rates will increase for the wrong reasons when prosecutors refuse to try Aborderline@ or difficult cases because losing those cases would lower their conviction rate. This disconcerting insight highlights the need to review the entire process system and not solely rely on one final number as a means to judge system performance and efficiency.

In addition to varying interests in defining what components a DWI conviction rate should be based on, there are also a number of other confounding organizational factors. These include the number of law enforcement agencies with overlapping jurisdictions, and perhaps numerous courts which may handle DWI related cases within a jurisdiction. Additionally, data may be reported and contained in multiple databases and systems in multiple formats. Then consider the possible use of different periods of time to calculate rates. Plus, the fact that it appears that many of the larger data systems actually begin with case disposition. Therefore, they are historical and based only on cases which have entered the judicial process and in some systems only cases which have reached some conclusion, but may not include cases where the persons arrested failed to appear or cases that have been pending for long periods of time. All of this illustrates how a variety of conviction rates could be calculated for the same jurisdiction.

And the most feared reasons for not calculating a complete enforcement and adjudication, system-wide conviction rate even if data are available to do so are:

·         that an all inclusive DWI conviction rate includes components beyond the control of those separate organizations contributing data, and

·         this method tends to provide a lower overall conviction rate than one which calculates a rate based on one segment of the legal process.

This lower rate is naturally somewhat disturbing, at least initially. But it is also reason for gathering information throughout the entire system to pinpoint weaker areas which could then be revised to strengthen the overall system performance. Also, calculating conviction rates on a regular basis, as pointed out by one GHSR’s office, allows interested parties to follow trends which can provide valuable information on how the overall system is operating over an extended period of time. We would have thought that a state-wide, complex system would be necessary to provide the data required to calculate an overall, system-wide DWI conviction rate. However at our smallest site, the Pitkin County Sheriff’s Department proved that, with cooperation from all of the organizations involved with DWI offenses, some initial planning and dedication, an effective system can be set up in any jurisdiction.

For an accurate DWI conviction rate to be calculated, it is necessary for the following data elements to be available: arrest date, specific offense code(s) charged, arraignment and/or pre-trial dates and outcomes where applicable, pleas entered or bargained for, trial dates and outcomes, and sanctions imposed and outcome of the cases, including dates and conviction offense(s). Also at any point in the system, if a person fails to appear and if a warrant is issued, that information should also be captured. Ideally, the system would track each case from arrest forward instead of historically from disposition backward. Then all cases would be included and not just those that have entered the court system. This information will allow the following conviction rate to be calculated:

R = 100 H (Number of DWI convictions of those arrested in a given period / Number of those arrested in the given period)

We believe this calculation provides the most accurate assessment of a DWI conviction rate.

Whenever a conviction rate is requested from any jurisdiction, a full description of what that rate includes and how it was calculated is imperative to understanding the true nature of that rate. In addition, merely imposing a mandate for higher conviction rates would provide a threat to those who apprehend and try DWI offenders which might encourage a relaxing of policy regarding DWI arrests and a lowering of prosecutorial standards. This could result in the more difficult, borderline cases, or those where individuals refuse a breath test, not being arrested and/or prosecuted which would, in turn, send the wrong message to the public and to DWI offenders, particularly recidivists, that this illegal behavior is tolerated. The results could very well have serious repercussions to public safety on our roadways.

Instead, all jurisdictions and agencies should be encouraged, and even rewarded in some way, for constructing a reporting system which provides a true picture of the entire enforcement and adjudication process through a uniform DWI conviction rate.

CONCLUSIONS AND RECOMMENDATIONS

We found through our discussions with practitioners in the field that the common public perception of a DWI conviction rate is the percentage of persons who are arrested for DWI who are actually convicted of that offense. However, we also found that in practice the conviction rates which are reported are seldom that simple number, but rather the conviction rate of some subset of DWI arrests or that convictions for other offenses are included in the calculations. Given that finding, it is evident that there are any number of types of calculations which are represented as DWI conviction rates. In other words, what is reported as the DWI conviction rate in one jurisdiction may be an entirely different calculation than what is reported by another jurisdiction as its DWI conviction rate. For that matter, agencies within a jurisdiction may have similar differences.

Thus, when comparisons are made between jurisdictions on this important issue, it may well be like comparing apples to oranges. A jurisdiction which is receiving plaudits that reports only the results of cases which come to court while many other cases are falling by the wayside, may well have a true conviction rate much lower than a jurisdiction which accurately reports a conviction rate that on the surface appears to be lower.

CONCLUSIONS

Major conclusions are:

·         There is great diversity among and within jurisdictions in what is reported as their DWI conviction rate.

·         Many jurisdictions have difficulty relating DWI convictions to arrests made within a specific time frame.

·         In many jurisdictions, failure to appear in court, is a frequent problem with DWI offenders.

·         Many agencies and jurisdictions choose to neither calculate nor publicize their DWI conviction rates.

·         There is a quite natural tendency for agencies within jurisdictions to calculate and report that form of conviction rate which places them in the most favorable light.

 

RECOMMENDATIONS

We recommend that:

·         When requesting DWI conviction rates, specify both the numerator and denominator, and request the raw numbers and the time period that the data are from, in addition to the calculated result.

·         NHTSA should consider the true conviction rate to be the number of DWI convictions resulting from and divided by the number of DWI arrests in a given time frame. A FTA clearly would not be counted as a DWI conviction in a judicial proceeding. Plea bargains might be counted as convictions if a jurisdiction has more than one level of DWI offense and it is common practice to reduce the charge to the lower level. If a jurisdiction includes plea bargains, it should be explained clearly that the conviction rate includes these alcohol-related, plea bargained offenses.

·         NHTSA should consider encouraging jurisdictions and agencies to keep detailed records on the disposition of DWI arrests so that meaningful and consistent DWI conviction rates may be developed, published and compared. Statewide case tracking systems using information from a uniform traffic citation should be encouraged as a means for maintaining such records.

We believe the lack of credible field calculations of DWI conviction rates and the inconsistent methods of calculations should remain under careful scrutiny. Furthermore, we believe NHTSA should publicly define how DWI conviction rates should be calculated. Only then can progress be made in identifying and correcting system deficiencies.

REFERENCES

Capital Consulting Corporation. 1997. Driving While Intoxicated Tracking Systems. Volume 2: State Tracking System Descriptions. Washington, DC: National Highway Traffic Safety Administration.

 

Clements, W. 1995. Vermont Traffic Records System Audit. The Vermont Center for Justice Research. Montpelier, Vermont.

Jones, R.K. and Lacey, J.H. 1998. Evaluation of an Individualized Sanctioning Program for DWI offenders. Washington, DC: National Highway Traffic Safety Administration.

 

Joscelyn, K.B. and Jones, R.K. 1981. Managing the traffic crash risk. The University of Michigan Highway Safety Research Institute report no. UM-HSRI-81-51. Ann Arbor, MI: The University of Michigan Highway Safety Research Institute.

 

1. In this report, the term “DWI” is used generically to describe driving with an illegally high blood alcohol concentration (BAC). Other terms that are used by some jurisdictions include DUI (driving under the influence) and OMVI (operating a motor vehicle while intoxicated), among others. Further, we are defining a conviction to be that obtained in a judicial proceeding on a criminal charge of DWI, not to include a “conviction” obtained by an administrative agency and resulting in action against a driver’s license.

2. Crime in Texas, Uniform Crime Reporting, Texas Department of Public Safety, 1985-1990.