Standardized Sobriety Test Battery Performance in Laboratory and Field Settings

297 drinking volunteers with blood alcohol concentrations from 0 to 0.18% were tested in a laboratory by ten police officers using a standardized battery of psychophysical tests. The officers correctly classified 81% of volunteers based on whether their BAC was above or below 0.10%, with officer estimates differing by 0.03% on average from actual BAC. A limited field evaluation involved 3128 data forms from drivers stopped over three months, showing officers increased arrest rates and improved BAC estimation after training.

Use Cases

• Modeling the accuracy of psychophysical tests for BAC classification based on the reported 81% correct classification rate.
• Analyzing interrater and test-retest reliability for standardized procedures based on reported coefficients ranging from 0.60 to 0.80.
• Evaluating the impact of officer training on field arrest rates based on the described increase post-training.
• Studying the correlation between officer-estimated and actual BAC levels based on the reported 0.03% average difference.

Strengths

• Laboratory evaluation involved 297 volunteers with a wide BAC range (0 to 0.18%).
• Reported interrater and test-retest reliabilities for the test battery ranged from 0.60 to 0.80.
• Field evaluation scale included 3128 data forms collected over a three-month period.

Limitations

• Row count and column-level documentation for the underlying data are unknown.
• The field evaluation is described as 'limited', and the data may reflect specific operational biases.
• Last update date and data freshness are unknown.

Provenance

Source

V. Tharp via paperswithcode.

Collection Method

Standardized administration and scoring of a sobriety test battery in laboratory and field settings.

Time Range

Field data collected over a three-month period (specific dates unknown).