The evaluation design was a comparison of two intervention corridors with two comparison corridors at multiple time periods including pre-project baseline, during each of the media-enforcement waves, and after completion of the project. The same types of data were collected for each of the four project corridors.
The TACT program included an extensive evaluation component conducted both by the WTSC and by Dunlap and Associates, Inc. of Stamford, Connecticut under a task order agreement from NHTSA. This section first discusses the evaluation approach and design and then presents the specific methods and results for each of the evaluation measures.
As described earlier, the TACT program consisted of enforcement and education interventions applied along two road corridors— Bellingham and Lacey/Olympia. In programs of this type, the objective is to alter driver behavior by conveying information on the correct way to perform and by creating general deterrence of illegal actions through a heightened fear of an enforcement sanction. The effectiveness of the effort will depend to some degree on the extent to which the intended audiences receive the message and perceive an increased risk of enforcement. It will also depend on whether the information is understood and recalled at the time when the correct, legal behaviors must be performed—in this case when the driver interacts with a semi truck on the highway. Program success will also be related to the ability of the driver to actually carry out the advice. For example, can drivers adequately judge that they have allowed sufficient distance after passing a semi truck before they pull back in?
In order to evaluate TACT fully and fairly, it was necessary to measure its effects at various points in the intervention process. Discussions earlier in this report covered the extent to which information materials were distributed and the number of citations written by law enforcement. Those can be thought of as the “input” to the program. The next measurement point was the determination of the extent to which these inputs were actually received by the intended audience. This was measured by a survey of exposure, knowledge, and self-reported behavior.
The determination of whether driver behavior changed and violations around semi trucks declined was accomplished using observations of vehicle interactions with semi trucks on the highway. Also, it was of interest to determine if the residual violations—those that still occurred after the TACT program—had changed in nature or severity. It was certainly possible that exposure to the TACT messages and the fear of a ticket prompted drivers to behave better even though their actions still constituted a violation of the law. This was assessed through ratings of behaviors observed before and after the TACT intervention.
Thus, the approach was to measure to the extent possible the effects of the TACT enforcement and media program through the actual behavior of the intended motorists on the road. In this manner, it was possible to document how the intervention processes performed as the project unfolded.
The evaluation approach used a design that measured before and after shifts in key measures of effectiveness at the two intervention locations— Bellingham and Lacey/Olympia—and contrasted them to similarly derived measures collected at two untreated “comparison” sites—Kelso and Spokane. This is a relatively powerful evaluation design because it assesses pre to post changes at the treated sites while simultaneously determining if those changes might possibly have occurred naturally without the TACT intervention at the comparison sites.
The presentation of evaluation results below follows the steps in the intervention process. First, a public awareness survey determined whether the intended audience was exposed to the TACT program and recalled its content. This is followed by the analysis of the violation rates before and after the TACT intervention. Then, the results of the assessment of the change in the nature of the violations themselves are presented. Finally, an additional survey to examine the possible spillover of TACT countermeasures into the Kelso comparison site is discussed. For each evaluation step, the data collection methods are presented first followed by a summary of the most meaningful results.