Sunday, February 19, 2006

Latest Evidence on Gender & Travel

Virtually all studies on the subject over the past few decades show strong evidence that women, as a group, persistently drive differently than men. Historically, their licensing rates were much lower, their commutes were shorter, and they took more trips overall. One explanation was their disproportionate burden of household-centered responsibilities. All in all this argued for treating gender as a substantive, if understudied, issue in travel behavior research. The 2005 Sandi Rosenbloom-edited TRB report of a 2004 TRB symposium offers a good cross section of current research on related issues.

It includes a paper by Rachel Gossen and Chuck Purvis using travel diary data from the San Francisco region from 1990 to 2000, indicating that gender differences -- at least for certain subgroups -- may be converging or even reversing. However, there are problems with their data (the survey method changed, etc.) and, even if there weren't, one wonders how well the Bay Area reflects national trends.

I am looking at the American Housing Survey. These individual-level data cover 1985-2003 for all U.S. metropolitan areas, in some cases tracking the same people over time.

The bottom line is that, so far as I can tell, while there are signs of convergence, the gender gap in commuting isn't going away any time soon.

Two important dimensions are age and family structure. As the 1st chart at the top of this post shows, females in every age group reported shorter commutes (in miles) than males in 1997. This gap rose with age, raising the questions of how much is gender vs age cohort.

The 2nd chart shows how this changed over the period. Commutes rose for all age groups and both sexes, but rose most for the youngest men and then for women in their 30s, 40s and 50s.
The age dimension might track participation in the labor force but that is complicated. Many women enter the labor force with adulthood, leave for parenting, and then reenter. How that pattern might reflect on those left driving to work in any particular age cohort is worth looking into further.

Among those working, family structure probably matters. Again, in these data, women have shorter commutes in every family type (including single, no children) at any point in time. In families with children, the gap between men and women is greatest. That is, men are more likely to drive farther to work if they have children, and vice versa for women.

The last chart shows how these changed in recent years, nationwide. Women's commutes are lengthening most, and men's least, in families with children, whether married or single. These are the signs of convergence but the differences remain large in absolute terms.
Differences by race, income, etc., matter, not always as expected. There are lots of kinds of people and part of the challenge is making comparisons, and tracking trends, that get at the most insightful factors, while controlling for the critical controls. I hope to finish writing this up soon. In the meantime, comments and suggestions are welcome.

No comments:


My blog has moved!

You should be automatically redirected in 6 seconds. If not, visit
and update your bookmarks.