|
Activity Number:
|
343
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Social Statistics Section
|
| Abstract - #308397 |
|
Title:
|
Statistical Application in Law: Empirical Investigation on Suspected Drunk Drivers' Data in New Jersey
|
|
Author(s):
|
Ayako Kambara*+
|
|
Companies:
|
|
|
Address:
|
2640 Route 70 Bldg1A, Manasquan, NJ, 08736,
|
|
Keywords:
|
Statistical Consulting ; Law ; Drunk Driving ; New Jersey
|
|
Abstract:
|
The purpose of this study is to test empirically to what extent the newly introduced alcohol detection instrument is too difficult to blow. Linear probability and binary logit models are estimated using suspected drunk drivers' data in New Jersey. It is known that the Breath Alcohol Concentration (BrAC) is an increasing function of the air volume submitted by the defendant. Without any functions or instructions to alert defendant to stop blowing, the BrAC can keep elevating. Further, the instrument does not set the breath sample criteria based on defendant's biological factors so equality can be an issue. People can falsely be convicted by the instrument's reported result when the instrument is designed too difficult to blow and when the breath sample criteria is set without considering individuals' biological characteristics.
|
