NHTSA DUI Driving Clues

Why read this page?

* Learn what the police are looking for when they see you driving at night
* Learn how innocent driving behaviors might actually be used to stop you for suspicion of DUI
* Avoid putting yourself into a situation in which the police have reason to detain you

Most materials and research used by government lawyers and police officers in training for DUI enforcement, and in actually enforcing West Virginia’s DUI laws, have some connection to the National Highway Traffic Safety Administration, or NHTSA. One of the key areas of DUI detection studied by police recruits during their training at the West Virginia State Police Academy is making observations of driving behavior that may indicate the driver is impaired by, or under the influence of, alcohol. This is identified by NHTSA as Phase I of the DUI detection process, also known as “vehicle in motion.” After reviewing the information below and the actual driving clues, we are confident that you’ll understand why we encourage you to be cautious whenever you go out for a drive.

NHTSA researchers studied over 100 driving characteristics of people who were later arrested for DUI offenses and found to have blood alcohol concentrations greater than the .08 limit that all states currently follow. The data was developed by large numbers of experienced DUI enforcement officers from around the country who made over 12,000 traffic stops and then reported the reasons their attention was first drawn to the vehicles they ultimately stopped. According to the 2004 edition of the NHTSA DWI Detection and Standardized Field Sobriety Testing manual used by police agencies across the United States, this research led to establishment of varying levels of statistical probability associated with the twenty-four most common driving patterns displayed by drivers who were charged with DUI offenses. But, there are some serious questions as to the reliability of the data and the conclusions drawn from it when one stops and questions the findings. For example, consider the following issues:

* Were the officers who generated the raw data told they were to look for driving patterns that indicated alcohol impairment? If so, were they therefore already inclined to patrol in areas where DUI was a known problem instead of working in all patrol areas to gain a greater statistical cross-section, or did they make more stops for things they would not have otherwise made stops for since they knew they were participating in a data collection exercise? (This is called as the Hawthorne Effect in the world of psychology)

* Did officers devote an equal amount of time to daytime traffic violations, or did their data come entirely from nighttime patrols when drinking activity is far more probable?

* Did officers even report statistics for people stopped but not charged with DUI? If they did not report non-DUI offenders, then the statistical relevance of the findings is greatly reduced because drivers who did the same things as those actually drinking were not included in the final analysis even though their driving was the same as the actually impaired drivers.

So, what is in a statistic? Pretty much whatever the researcher wants to have in the statistic if he or she is not focused on honest, objective results, or if he or she is careless in analyzing the data. While statistics can offer relevant and useful information, caution must be exercised because of the potential for abuse by the researcher, particularly when there is an agenda associated with outcome, such as strengthening opposition to occurrences of DUI. Click here if you are interested in learning more about how statistics can be improperly manipulated, on purpose or through carelessness.

When looking at the levels of probability associated with each of the twenty-four clues linked above in the second paragraph, note the generally wide range of probability, and also note how low many of the probabilities are. This overbroad approach to DUI enforcement in which officers are encouraged to make as many stops as possible to make a DUI arrest, often for driving patterns that have no independent relationship to use of alcohol, creates a ripe opportunity for abuse by unethical or overeager officers. By casting a wide enough net, they will certainly arrest drunk drivers, but they will also make unlawful stops on numerous innocent drivers. While no one can logically dispute the importance of preventing impaired people from driving, it is also a fair conclusion that we cannot authorize unreasonable violations of our privacy as a consequence of the objective. But, that is exactly what these low and widespread levels of probability permit and what our police are taught in training academies across the country.

So, the next time you head out for a drive, just remember, you may be doing something that tells a police officer you are under the influence - even if you haven’t had a drink.