Much has been written about millennials and their travel choices. The common story line in the popular press is that millennials are driving less, owning fewer cars, and/or not getting their driver’s license. Within this discussion, there has been a thread that millennials are also bicycling more (along with vehicle sharing, transit, and walking). But, is this really the case?  Using survey data from the 50 largest urban areas in the US, I examined how often millennials do bike compared to older generations. The answers are in a post on the TREC Blog.

You’ll see there that the millennials are not necessarily biking more than GenXers, and that there were not many other differences between the two generations. But, one thing did stand out. The bicycling gender gap disappears for millennials (see below).

Source: Analysis of 2015 National Community and Transportation Preference Survey, more details on survey here

After doing the analysis and writing the post, that figure above kept nagging at me. (It was also highlighted in one of the tweets about my original post, so it clearly stood out to others as well.) On the one hand, I was happy to see the gender gap disappear. But, did it disappear because millennial men are not biking as much as older men, or is there a real difference between millennial women and other generations of women? I spent a little time digging deeper into my data. I’m not sure if I have a clear answer yet, but here’s what I have found. If you have ideas, I would like to hear them.

I first looked to see if there were differences in utilitarian vs. recreational cycling. Millennial men and women were equally likely to bike for transportation or only for recreation. The gender gap is significant for biking for both transportation and for recreation only for the other generations. In addition, rates of biking for transportation are about the same for the men by generation (14%, 14%, 12%), but go down for women (12%, 7%, 4%). On the other hand, biking only for exercise spikes among GenX men. So, two things appear to contribute to the lack of a gender gap among millennials: millennial women biking more for transportation than the older generations and millennial men biking less for recreation only compared to GenX.

Note: Number of respondents in the Silent/Greatest Generation that responded to these questions too small to draw firm conclusion, so not included in graph.

I’m most interested in understanding why millennial women are cycling more, and about the same as millennial men, so I dug into that a bit more. I first checked a measure of attitudes, since we know that’s usually a major predictor of bicycling. The survey simply asked for agreement with the statement “I like riding a bike.” Across all generations, women were less likely to agree with that statement, though the gap is larger in the two oldest generations.

I then looked at the other questions in the survey related to bicycling, comparing the genders by generation and looking for differences, particularly where there were gender differences in older generations but not within millennials. A few things emerged.

Millennial men and women were equally likely to say that a reason for not biking is that “the places I need to go are too far to bike.” In the two oldest generations, women are more likely to say that distance is a barrier. Whether this is an indicator of younger women living in different environments than older women, or differences in perceptions of distance and how far they are willing to ride is not clear. It’s likely a combination of factors.

Similarly, having too few bike lanes or trails was an equal barrier for millennial women and men (as well as GenXers). The gender differences were significant for Baby Boomers.

Women were more likely than men to not feel safe because of traffic for all but the oldest generation. The lack of a difference among the silent/greatest generation is not a finding to focus on; it is likely due to the sample. The question was only asked of people physically able to ride a bike, which shrunk the numbers in that generation to under 200, and less likely to be a representative 200.  Rather, notice the higher share of women Baby Boomers that do not feel safe because of traffic.

Another difference was the effect of weather. This was a bigger barrier for women than men for GenX and Baby Boomers

Health was a barrier for bicycling for a very small share of the respondents, but this is where the gender gap gets more pronounced in the older generations. Note that this question was only asked of people who could physically ride a bicycle, so it may be capturing a range of health issues.

Finally, I looked at comfort level biking on different types of infrastructure. These are questions I have used to categorize people into the ‘four types of cyclists” developed by Roger Geller. You can read more about that work here. The average levels of comfort for men and women for the three youngest generations are shown below. (Note that I left quiet residential street off the graph to simplify things. Also, the survey questions used descriptions of the infrastructure, not these labels.) There are two main sets of differences. First, millennial women are just as comfortable riding in most environments as millennial men, except on busy streets without bike lanes or only striped bike lanes. On the other hand, there is a gender difference in every environment for GenX and Baby Boomers.  Second, you may notice that for every scenario, comfort levels of women decline significantly with the generations, most noticeably in the major street situations. That only happens for men for the major street without a bike lane and with a striped bike lane. In other words, men are often equally comfortable across the generations, but older women are less comfortable in all situations. Taken together, these differences indicate that millennial women are more comfortable biking in different environments than older women and are similar to millennial men in many cases.

So, what does this all mean? As with the answers to most questions about people’s travel behavior, there’s no one explanation. With respect to women, the following factors may contribute to the lack of a gender gap among millennials:

• a smaller gender gap in attitudes (liking biking)
• better environments (more reasonable biking distances and not lacking bike infrastructure)
• higher comfort levels biking in most environments compared to older women
• fewer health concerns and greater tolerance of poor weather

The big question that I can’t answer from these data is whether these differences will persist as millennial women get older. Will their comfort level remain the same as they age? There is evidence that women are more risk averse than men (here, here), which helps explain their desire more for separated infrastructure. I need to spend some time exploring the research on how this changes with age.  My convenience sample of one (me) suggests that there may be a correlation. When I was of millennial age, I rode my bike all over west Los Angeles, often without bike infrastructure. I doubt I would do so today, nearly 30 years older, assuming the infrastructure hasn’t changed. But, I’m not a millennial today, so the comparison may not hold true.

In addition, there are several hypotheses that I could not test, given the limited survey questions. One involves household responsibilities. Much has been written about women doing more of the household maintenance-related travel, e.g taking kids to school and soccer practice, older parents to medical appointments, grocery shopping, etc. These types of trips are less feasible on a bicycle and have been hypothesized as a reason for lower rates of cycling among women. Younger women are less likely to have some of these responsibilities (e.g. fewer have children) and responsibilities may be more equally shared with a partner in the household. But again, will there be a cohort effect, with these patterns persisting as millennials age, or will behavior change as millennial women age, so that they resemble the older generations of today? Where are the longitudinal data when you need them?