Schaffer Online Library of Drug Policy Sign the Resolution
Contents | Feedback | Search
DRCNet Home
| Join DRCNet
DRCNet Library | Schaffer Library | Drugs and Driving 


Drinking-Driving and Speeding Profiles from the Police Reports

Marie-Chantal Jayet

INRETS-Dera, 2 avenue du Général Malleret Joinville, 94114 Arcueil Cadex, France


The paper is based on the data of a sample survey carried out at the regional level of a French "Département" on the motoring offences daily detected by the national police forces (Gendarmerie, Police), the objective being to enlighten the profiles of "offending driving" behaviours ordinary controlled and sentenced. Another broader objective dealing with law enforcement and regulation process is afterwards to rely these profiles to those resulting from two other parallel surveys carried out on the same local level (offences observed on road and reported in police accident records).

The two main hazardous behaviours detected by the police forces will be described and compared according to a list of varied criteria: the range and the frequency of drunk-driving and speeding amongst the variety of road offences reported by the police forces; their variations according to police forces, type of sentencing, type of road network, time of day and of week, type of vehicle, drivers' gender, age and occupation.

The drunk-driving and speeding profiles (multi-factorial analysis) issuing from Police screenings data and their comparison enlighten that, except the cluster of young male car drivers travelling by night on week-end, speeding and drinking are representing two different and characteristic types of offending driving. They also provide a global evaluation on both police strategies of law enforcement and target groups of preventive policy.


Amongst the variety of risk behaviours involved in traffic accidents, the driving speed being a risk factor as important as the alcoholic driver condition, the regulation of both driving behaviours and their respective enforcement have been often reinforced in France since the seventies. Most of the traffic safety approaches have considered each of these behaviours independently up until now. The present paper proposes to study the speeding and drinking-driving profiles issued from police reports required for prosecuting and sanctioning driving offences. The offending profiles are framed by information filled out when the Police records an offence committed against driving rules in force : the offence reports take over indeed some of the main space-time and socio-demographic variables of traffic accident reports. Provided they are collected, these police data enable a statistic work which is not available as yet to be carried out, in particular : who is controlled by whom? for what offence? on what road network? at what times of day and on what days in the week? with what sort of lawsuit? Moreover, they make possible the description of driving behaviour constituting a statutory offence on the basis of risk analysis categories.

The drinking-driving and speeding offence data here studied, come from two surveys carried out on Police reports at the local level of a French "département" and concerning the offences daily detected and reported by the police forces in charge of traffic control. The paper consists in comparing the characteristics of the two main road hazardous behaviours such as reported by the police forces and with reference to the space-time (roads, hours, days) and socio-demographic (age, gender) categories that are describing alcohol and speed risk contributions. The purpose is to look into and to see to what extent the design of both traffic safety regulations may encounter, or not, the same law implementation issues either on the ground of police control put into effect or on the characteristic ground of offending behaviours.


Road traffic in France is under the supervision of two main police bodies responsible for national security : the Gendarmerie Nationale, a rural police force reporting to the Ministry of the Armed Forces and Defence, and the Police Nationale, a urban police force run by the Ministry of the Interior. The "Gendarmerie" rural network includes a lot of small and medium-sized towns, while the Police urban network correspond to towns of 10 000 inhabitants and more. The police forces in the "département" where the surveys have been carried out, count a large number of police units broken down as follows : the "Gendarmerie" counts 8 Companies of 64 territorial brigades and 1 corps of 4 mobile brigades specialised in road traffic control, the Police counts 27 police stations.

The local surveys were conducted in two waves, each one corresponding to a specific police body (the Gendarmerie in 1992 and the Police in 1993). The survey data were gathered from two samples of police units each one being representative of every police body and of its control activity in the "département" (16 local brigades + 1 mobile brigade for the Gendarmerie sample, and 4 Police stations). The survey data cover 1,738 offences reported on the network under surveillance by the Gendarmerie over three months (April, September, October), and 2,146 offences reported on the network under Police surveillance (June, September, October). They were collected after the event on the basis of the police reports and concerned non-accident and non-parking offences detected and reported by the police forces in the course of their traffic control and road safety duties. The offences are described by space-time (A and B roads, rural or urban area, day and hour) and socio-demographic (gender, age, socio-professional category, vehicle, domicile) variables.

The police data of "drinking and driving" offences refer to the legal "Breath Alcohol Concentration" threshold of 0,4mg/1000 in force in France since 1983 and enforced by random testing or testing when drivers are stopped for another road offence (in fact, this last procedure is not systematically implemented). The "speeding" offences refer to the violation of the speed limits enforced according to the type of road, vehicle or driver. The speed limits in force in France are: a general speed limit of 50 km/h in built-up areas and local 45 km/h limits by municipal order; a general speed limit of 90 km/h on A and B roads with local 70 km/h limits; a general limit of 110 km/h on dual carriageways and 130 km/h on motor ways; maximum authorised speeds for heavy vehicle drivers on the open road and a maximum authorised speed for beginner drivers (90 km/h).


Speeding offences (23%) form with non-wearing seat belts (21%) and 21% of administrative-type offences (driving licence, insurance, vehicle registration papers), the group of road offences the most frequently reported by the police forces. Speeding is prevalent in Gendarmerie reports (29%) and on the third rank of offences reported by the Police (16%).

The proportion of drink driving over the legal Breath Alcohol Concentration (BrAC +) fetch up at 3.1% of the offences reported, with a higher score on rural road network (4% of the Gendarmerie sample) than on urban network (2% of the Police sample). Drinking and driving takes part of the group with the lowest frequencies of police reports (decreasing order : BrAC, solid central line, tyres, helmet), the medium group being made of offences committed against stop (traffic lights and stop signs) and lorries regulations.

Note that, while the intensity of drinking-driving control has grown with the years (the month average by the time of the surveys fetched up at 254 operations of random tests and 12,649 persons tested by breathalyser), the random test return is decreasing from year to year (1% of driver with an illegal BrAC in 1988 and 0.5% in 1993).

If one refers to the location of the offences, drinking-driving and speeding are mainly detected in built-up areas : drinking-driving is more often reported on urban network (58%) and on B roads (35%) especially on sections crossing built up areas (20%), while speeding is in the first place recorded on highroads (60%) and mainly on B (26%) and A (24%) road sections crossing towns (especially cities of less than 2,000 inhabitants, 32%), the urban ways coming in second position (40%).

As to the excess of Breath Alcohol Concentration (BrAC) or speed over the legal limits, the figures show firstly that, from the moment a BrAC is over 0.4mg, the offence is reported by police forces while the speeding offences began to be registered only when they exceed 10km/h over the legal limit (more often over 15km/h). The BrAC levels reported indicate that, even if the proportion of drivers offending the law is decreasing from year to year, the scores of illegal heavy BrAC levels stay high : BrAC, from 0.4mg to 1mg, represent 69 % of the drinking-driving offences, while BrAC, between 1mg and 1.5mg, then over 1.5mg, have the respective scores of 27% and of 5%. The greatest mass of speeding offences is located between 20km/h and 29km/h over the limit (68% in town and 36% in rural areas). However, one out five speeding offence involves a very great excess of speed (40km/h or more) mainly detected on road sections crossing cities of less than 5,000 inhabitants (51% A + 25% B sections), precisely where the highest ratios of fatal accidents are located in that Département.

Note that most of alcohol (88%) and speed (90%) offences concern car drivers. Furthermore, driving offenders living in the "département" are over-represented in BrAC offences reported (93% against 80% ; moreover with highest BrAC, 98%) and under-represented in speeding offences (70% particularly in the case of great exceed speeds 61%). These comparative figures seem to indicate that the local drivers are rather better informed about speed control occurrence (location and time) than about drinking-driving control implemented by the local forces of police.


The breakdown of both offences according to the daytime periods (Table 1) let see variations of frequencies between drinking-driving and speeding that are as dependant on the planning of police control enforcement (speeding) as on the time frequencies of offending (drinking-driving).

Table 1
Time Period Distributions (police reports)

  5-7h 7-11h 11h-14h 14-17h 17-21h 21-0h 0-5h total
BrAC + 0 4 2 17 24 28 25 100%
Speeding + 0 32 11 31 21 4 0 100%

When looking into the proportion of illegal BrAC by daytime period (ratio = BrAC/all offences), drinking-driving which represents 3% of the whole offending-driving reported by the police forces, increases severely during the hours of night (Table 2). On the contrary, the speeding offences representing 27% of all offences reported, are decreasing all day long until to be quite absent from late night reports, the bulk being located during the morning hours (Table 2).

Table 2
Time Period Ratios (police reports)

time periods
(% all offences)
BrAC (3%) / 1% 1% 2% 3% 15% 25%
Speeding + (27%) / 29% 24% 24% 22% 14% 2%

According to the days in the week, half the illegal BrAC is recorded on week-end (from 7h p.m., on Friday evening, to 5h a.m., on Monday morning), one out of four illegal BrAC being reported on Sunday, while three out of four exceeds of speed are reported on weekdays, especially on Monday (21%).

The day ratios of Table 3 emphasise on one part the Sunday and Friday weights for drinking-driving offending and on the other part the unexpected Monday score for speeding (Monday being not a very special day for traffic, risk or journey type densities). If one refers to the data from surveys conducted on the same area network and during the same periods (Peytavin, 1995), these ratios indicate that the time and day sequences of Police control phase the time-and-day probability of drinking-driving behaviours, and not at all the speeding one.

Table 3
Day Ratios (police reports)

(all offences)
BrAC + (3%) 3 1 2 3 4 3 6
Speeding + (27%) 31 22 23 20 23 24 20


The age and gender frequencies show that the drivers under 35 years of age are the more concerned by illegal BrAC (61%) and speed offences (54%) recorded by the police forces (Table 4), drinking-driving being quite an exclusive male offence (96%) affecting especially young men from 25 to 35 of age (42%), while speeding adds up the female offending contribution (26%) including every age class. However, the age class ratios show that the 25-35 and 45-55 years of age register the higher proportions of illegal BrAC and that the percentage of speed offences increases when the years of age increase.

Table 4
Age Proportions and Ratios (police reports)

(all offences)
< 25yrs
BrAC + 19 42 18 18 1
Speeding + 23 31 23 14 8
  ratios ratios ratios ratios ratios
BrAC + (3%) 2 4 3 5 1
Speeding + (25%) 20 24 29 32 33

If one refers to the values of excess, the trends differ : the younger the offenders are, the more the speed means are high and conversely, in other words : on average, the excess values decrease while the age of offenders increases, the mean varying from 32km/h (<25 years of age) to 25km/h (>55 years of age). The mean value of illegal BrAC by age class show a different trend : on average, the excess over the legal limit grows with age from 0.8mg (<25yrs) to reach its highest point between 35 and 45 years of age (1.06mg) and lightly decrease then to 0.91mg (45-55 years of age).


The findings of multi-factorial analysis give drinking-driving and speeding clusters that do not converge on demographic and time categories. The clusters emerging from police reports can be summarised as follows :

Nevertheless, beyond these differenciations between speeding and drinking-driving profiles, an in-depth analysis of great excesses (speeding over 35km/h and breath alcohol concentration over 0.8mg) stresses a common age relationship : the drivers over 35 years of age reveal a same propensity to great excesses and to frequent offending for both offences, while the younger drivers (under 35 years) associate frequent offences and few great excesses.


The space-time profiles of drinking-driving and speeding emerging from police reports pinpoint at first some differences of control implementation : the drinking-driving control reflects rather closely the main space-time categories of drinking-driving risk and behaviours, while the control of the excess of speed reveals itself rather independent of speeding trends. Moreover, as far as police detection practices are concerned, they seem to be strictly enforced for drinking-driving excess and to admit a part of tolerant attitude toward lower excesses of speed.

The drivers under 35 years of age are the more often reported for both offences, while, compared to the whole offending driving score of their age group, the drivers over 35 years of age reveal a greater propensity to speed and drink-driving offending and to great excesses.

The findings of multi-factorial analysis show clusters of speeding and drinking-driving that are widely differentiated on the basis of time and socio-demographics categories.


PEYTAVIN, J.F. Evaluation de l'action réglementaire, évolution des vitesses au niveau local, rapport DERA, 9502, INRETS, janvier 1995.

INRETS, ONISR, ORSR, Les conducteurs interpellés sous l'empire d'un état alcoolique dans la région Nord/Pas de Calais, décembre 1994.