The following email was sent to WSDOT and WSTC on June 29th discussing a few of the problems in WSDOT’s report to WSTC (and legislators) regarding the 6 month performance of the #405ETL. They acknowledged receipt and will be providing a collective response:
This email is in regard to the WSDOT presentation to WSTC regarding the I-405 Performance and Operations Update and the Annual Tolling Report given May 17, 2016.
video archive: http://www.tvw.org/watch/?eventID=2016051059 beginning at 35:23
Presentation materials: http://www.wstc.wa.gov/meetings/agendasminutes/agendas/2016/May17/default.htm
I reviewed the data presented in the 6 month I-405 Performance and Operations Update presented at the Bothell Library and again at the May 2016 WSTC meeting. I have some concerns about the information presented.
The absence of any performance data for midday hours, and nights, and weekends
Errors or misrepresentations in creating, collecting, interpreting, and reporting the survey
The use of youtube videos as a means to educate the public
Success criteria are not clearly defined and do not include important factors
First, Major changes were made to the tolling system when they were eliminated on nights and weekends. Yet there is absolutely no mention of any performance data during midday hours (between peaks), nighttime, and weekends. I would expect this to be relevant and essential in determining the performance of the I-405 corridor with tolls vs without tolls. There should be at least 5 comparisons made for the nighttime and weekend data:
1. Performance before construction began.
2. Performance for the year prior to opening the tolls
3. Performance with tolling operating 24×7
4. Performance after tolls were removed at nights and weekends
5. Performance weekdays from 9:00 AM to 3:00 PM compared to the same time on weekends without tolls.
This data is crucial to determine if tolls are actually reducing congestion or if the cited reduction in congestion is strictly a result of infrastructure improvements and other variables like gas prices, weather, economics, and unemployment. By comparison, this article WSDOT reported how the tolls caused congestion to increase on weekends: http://wsdotblog.blogspot.com/2015/11/i-405-express-toll-lanes-6-week-review.html
The omission of weekend/nighttime performance since tolls were eliminated leaves the public suspicious as to the motives of leadership and it is perpetuating a loss of trust.
The second concern is the method of soliciting responses in the survey and how the data was reported to WSTC and presented to the public. WSDOT sent an email to people who have a GoodToGo pass. (a copy of the email is appended.) In Statistical Science this is referred to as Sample Bias and is the opening topic presented in the book “How to Lie With Statistics” by Darrell Huff. This is not a valid method for getting a representative sample of the entire group of people served by I-405 or even of those people using the toll lanes. This is a tiny fraction of the total users of the highway and selectively avoids people who are likely to give negative responses. In the stop405tolls.org petition with over 30,000 signatures demanding legislature to remove the tolls entirely, many people have passionately commented that they adamantly refuse to buy a pass for the toll system. It also biases the survey by avoiding people who don’t typically access the Internet, such as the poor and the elderly. This is not a random survey.
Here are several examples of how the data was misrepresented:
– The following graph was reported on slide 8 of the Annual Tolling Report:
This graphic implies all income levels favor expanding HOT lanes equally. This is deceptive for several reasons:
It is taken exclusively from people who already pay to use the HOT lanes. As such, they already demonstrated some tolerance. (yet 1/3 of them do not agree to expanding)
It does not identify the actual number of people in each household income category.
The range of income for each category is not equal.
How does the income distribution of pass holders compare to the income distribution of the general population?
So the conclusions this graph attempts to make are feckless.
Senator McAuliffe worked to get an amendment to the transportation budget to provide a study of the demographics of users on I-405 including the income of people paying to use the HOT lanes and how often they used them. WSDOT collected part of this data in the survey, but did so poorly; and what they did collect, they did not report.
– The results of the survey were given as percentages, not total number of responses. Reporting percentages alone is misleading. Actual counts are crucial in validating the statistical relevance of each question. This holds true for all data given in the report. Representative Stanford is a professional if the field of statistics and might give a more expert opinion on this.
– In spite of the fact that the survey only included people who are GoodToGo pass holders, only 53% indicated they are satisfied. This is a horrible customer satisfaction rating, but pales in comparison to what it would likely be if the survey also included the rest of drivers who don’t have a GoodToGo pass..
Third, I am concerned about the need for so many education materials needed and the reliance on youtube videos as a means for educating the public on how the express toll lanes work. In addition to publishing 7 youtube videos posted here:
WSDOT has this interactive map:
and this website with numerous lengthy webpages explaining how to use the toll lanes:
And there is a mobile app to find out what the toll price is right before you use them.
Seattle is a gateway city to the Pacific Rim. You might expect some local residents to put the effort into studying all these materials, but the majority of people don’t know they exist and the people who most need them are least likely to watch them. It is unreasonable to expect the millions of drivers in the Puget Sound region to review all of this. Even more so of infrequent visitors from around the state, around the country, Canada, and other countries. Consider that youtube is blocked in China. Not all people have handy access to the Internet, like low income and elderly people. This creates yet another barrier to widen the social inequity of the Express Toll Lanes. If it takes a series of videos, a set of webpages the length of a small book, and an interactive map just to understand them, then the system is inherently too complicated. With all the problems from distracted drivers, the fact that WSDOT created a mobile app for drivers to check toll prices just before entering the toll lanes is incredible.
Lastly, as the 6 month performance report of the I-405 Express Toll Lanes is reviewed, what are the criteria for determining if the tolls are a success or should be eliminated? In engineering this is referred to as acceptance testing. Does that criteria include public acceptance? Does it include a minimum level of service improvement for congestion across all lanes? Does the service level consider overall traffic or just the ETL? If it only considers the ETL and not overall highway performance, then why? Does the service level improvement measurement differentiate between the gains made by imposing a toll vs. the improvements made by increasing capacity? Since capacity was added at the same time the tolls were introduced (for part of the highway), the added capacity masks the fact that the tolls increased overall congestion. That would be revealed if the data regarding weekends and nights were published along with the data already shown by the section north of 522. The only 2 criteria for determining the degree of success or failure that I have seen are:
1. The toll lanes maintain 45 MPH for 90% of peak hour and
2. The lanes must not operate at a financial loss.
Yet these factors alone do not address the interests of the people.
In summary, the report shows a clear bias to yield positive results in support of the tolls. It avoids contradictory evidence, like the results of eliminating the tolls at night and on weekends and WSDOT used sample bias to influence the survey responses, then used reporting methods to obfuscate, prevaricate, mislead or omit facts entirely.
I am concerned about the impact that the errors and omissions in this report will have on any decisions made by WSTC and the legislature and the potentially harmful consequences of those decisions. WSTC and legislators are making policy decisions based on this data. Errors and misinterpretations in what WSDOT tells them could lead to making decisions that create a more deadly highway, increased congestion, wasted tax money, and heightened public frustration and mistrust. If bonds are sold, it could lock us into those bad decisions; not to mention it could cost elected officials their political office. Given the tenuous balance in both the House and Senate, this could change the party majority and affect policy on many other issues beyond transportation.
Your response to myself and my legislators regarding these issues is appreciated.
Invitation email for survey sent to GoodToGo Pass holders:
From: “Washington State Department of Transportation” <firstname.lastname@example.org>
Date: 03/30/2016 11:01 AM
Subject: I-405 Express Toll Lanes Customer Survey
Dear Good To Go! customer,
You are receiving this survey invitation because our records show that you have used the Interstate 405 express toll lanes. We hope you will consider taking a few minutes to answer our questions. Your anonymous answers will help us understand more about how people are using the express toll lanes.
If the link above doesn’t work please copy and paste this address into your browser: https://www.surveymonkey.com/r/JGVw17XJ8Rl1tJSp
If you have any questions or comments about the survey, please contact us at email@example.com or 206-464-1230.
This email was sent to firstname.lastname@example.org using GovDelivery, on behalf of: WSDOT
310 Maple Park Ave SE · Olympia, WA 98504
Awaiting a response…