Mid-Week vs. Whole Week Data

[VeloCityPlanner]

Hi there,

I am involved in establishing a bicycle and pedestrian count program from scratch. Earlier in my career, I was involved in collecting motor vehicle and transit ridership data. I recall that one aspect of this work was the strong preference for collecting and basing calculations on weekday traffic data, specifically Tuesday/Wednesday/Thursday.

In my research on bicycling and pedestrian count programs, it appears that guiding documents recommend collecting and using data gathered at any and all days of the week for extrapolations (such as in calculating AADB). Why is this? Wouldn’t it be best to base calculations on mid-week traffic data when it represents typical travel or commuting patterns, and exclude weekend data when many recreation-based trips are made? What are the pros and cons of basing calculations on mid-week vs. whole week? In my situation, we are using automated counters collecting continuous data, both permanent and mobile.

Can those involved in this discussion group please provide an explanation or suggest where I may find more information on this topic?

[Dozier, Emily]

We have only done manual counts, but we follow the methodology of the National Bicycle & Pedestrian Documentation Project, which recommends counting on a Tuesday, Wednesday or Thursday as well as a Saturday. See the Data Collection Instructions on their ‘Downloads’ page.

 

-Emily Dozier

Emily S. Dozier, AICP
Senior Planner
Poughkeepsie-Dutchess County Transportation Council
27 High Street, 2nd Floor
Poughkeepsie, NY 12601
Phone: (845) 486-3600  Fax: (845) 486-3610
Email: edo...@dutchessny.gov

www.dutchessny.gov

--
You received this message because you are subscribed to the Google Groups "Walk-Bike-Count" group.
To unsubscribe from this group and stop receiving emails from it, send an email to walk-bike-cou...@googlegroups.com.
To post to this group, send email to walk-bi...@googlegroups.com.
Visit this group at http://groups.google.com/group/walk-bike-count.
For more options, visit https://groups.google.com/d/optout.

[Krista Nordback]

This is a great question. If you can only count on one day Tuesday, Wednesday or Thursday are the best days, as you point out. They tend to have the most stable traffic patterns. However, if you can count a full week, this should generally result in higher AADB estimation accuracy since it will allow you to differentiate between sites with higher weekend counts than weekdays. Since you have mobile automated count equipment, I hope that counting for 7 days at each location isn't too difficult. 

So yes, weekends tend to be more variable, but if you apply temporal adjustment factors to a short term count assuming it's got low weekend, and find out later that the counts are highest on weekends, your error will be high. Make sense?

For a summary of some research on this topic see http://www.pdx.edu/ibpi/select-duration

Also, we provide general guidance on count programs based loosely on the FHWA's Traffic Monitoring Guide, Chapter 4, at http://www.pdx.edu/ibpi/count

As Emily points out, the National Bicycle and Pedestrian Documentation Project also provides guidance, but this is mostly for manual count programs.

-Krista

Krista Nordback, Ph.D., P.E.

Postdoctoral Research Associate

Transportation Research and Education Center (TREC)

Portland State University

nord...@pdx.edu

office:  503-725-2897

--