Bicycling Survey Information

Why have a survey directory? Most of this site is based on my experiences and my analysis of those experiences. However, I live in just one small corner of this world, so I hardly know everything. I have learned to be suspicious of commonly accepted opinion because it often isn't based on factual information but just on hearsay and speculation. Statistics and surveys provide an opportunity to look at the facts objectively, provided the information was gathered correctly. In addition, I think that a monthly survey will add a bit of fun to the site.

My first exposure to this kind of information came from reading a book called How to Lie with Statistics, way back in high school. Of course, by that time, I was already aware of Mark Twain's statement that there were three ways to avoid telling the truth, "Lies, damn lies, and statistics." In college, I saw a good example in a weekly student survey explaining what students at our school really wanted. Besides the biased questions and a biased selection of participants, the poll takers also queried few students, in one case six, indicated by the percentages being multiples of 16.

If someone wants to control the results, it is very easy to do so. One way is by writing questions in such a way as to change anyone's opinion; for instance, "Do you see anything wrong in destroying a few dozen human cells?" and "Do you believe we should kill babies?" could both be asking whether you think abortion should be legal or not. Vague, confusing, and misleading questions tend to have the same effect, although not as strongly or predictably. Results can also be controlled by selecting the participants, for instance, handing out copies of a survey on gun control at a rifle show. Finally, the results can really be skewered in the presentation at the end, which is what How to Tell Lies with Statistics focused upon. A good survey avoids these faults as much as possible and is honest about them when they occur.

I think there is a real need for some good bicycling surveys. I have studied most of the published surveys, and I find major problems with some of them. In addition, many questions have just not been asked on any survey.

While it is ideal for a survey to represent accurately the entire US population, even such a careful survey can produce inaccurate results (for instance the CPSC survey on Bicycle Use and Hazard Patterns). On the other hand a very small survey, with just a couple of hundred non-random participants, can provide new information, provided the right questions are asked. For instance, at the University of Alabama in 1991, I surveyed a couple of hundred students about some stereotypical sex differences between men and women, and I discovered that they just weren't true. A huge survey could have established no more.

As this site receives over 12,000 visitors each month, I was hoping that I could get a few thousand replies within a month's time. However, since only a couple of hundred cyclists are taking these surveys, we have to look at these surveys basically as polls rather than as statistical studies. This is wise anyway, because even if every visitor took the survey, we would still not have a balanced population of cyclists; just those cyclists who use the internet and visit my web site. Nonetheless, these polls can be quite valuable with only a few hundred answers because they are objective methods of discovering behavior and opinions, even though they can not pretend to represent the entire population.

Survey Results from KKBP

The September 2001 Bicycle Safety Survey The results of a 20-question bicycling survey run for 30 days with 231 cyclists responding. Questions concerned amount of bicycling, kind of bicycling, traffic behavior, traffic attitudes, and injuries.

The October 2001 Bicycle Political Survey  Questions concerning bicycles and politics with two hundred cyclists responding, dealing with helmet laws, police behavior, local government support, what cyclists need from government, and whether environmental protection still makes sense.

The November 2001 Bicycle Touring Survey  A numeric investigation into what touring is all about. Included in this survey: location of the trip, trip length, the number touring, amount spent per day, how gear was carried, injuries and equipment failures on the trip, and where nights were spent.

The December 2001 Health Survey  An investigation into the exercise, diet, and illnesses of cyclists. Still in the draft version.

Survey Results and Statistics from Other Sources

How Many Bicycle Commuters Are There in the USA? -- Data from US Census  The supplimental data sheet, sent as part of the 2000 census, asks a question about transportation to work that can be used to derive the number of bicycle commuters in the US. Unfortunately, due to the question asked and the method of computing the data, the figure is greatly understated.