You have /5 articles left.

Every year, students register later than they did the year before.

That’s not just an impression. An impression would be that every year they get younger. I can talk myself through the reasoning to realize that no, they’re not getting younger; I’m just getting older. It happens.

Registering later, though, is real. I know that because we keep statistics on comparative enrollment levels from year to year, based on the number of days until classes start. For example, 10 days before classes start, maybe we’re down 4 percent. If we’re still down 4 percent on the first day of classes, that doesn’t mean that nobody registered for 10 days; it just means that the same number of people registered in those 10 days as did in the same 10 days of the previous year.

The statistics aren’t completely perfect, for various reasons. Days before classes start and days of the week don’t align from year to year, given normal shifts in the calendar. “Drop for nonpayment” days can move a little, which can cause blips in either direction for a day or two. A snow day can throw things off, if it meant that registrations were delayed by a day. But if you follow the reports daily -- and I do -- they tend to follow consistent patterns.

Over the last several years, the consistent pattern has been that we’re down 2x for weeks, then cut the deficit to x in the last run-up to the start of classes. When I ask people why they think that is, I usually get a shrug and some variation on “our students register late” or “students procrastinate,” but those don’t actually answer the question. The numbers are relative to last year. For the gap to close late must mean that more students are registering later than last year. This has happened year after year after year. As with “human nature” arguments generally, it can’t address change. Human nature doesn’t change 4 percent per year.

So I’m looking for reasons that a larger proportion of students registers later every year.

The intuitive answer, which is probably at least partially correct, is money. After a certain date, they have to pay on the spot to register, unless they have financial aid. That creates an incentive to wait. As tuition goes up, the incentive gets a little bit stronger every year.

I suspect there’s something to that, but tuition doesn’t go up that much. That can’t be the whole issue.

Given how many jobs students have, there may be an effect from greater sensitivity to shifts in working hours. Entry-level jobs often have unsteady hours, so they may be waiting for the dust to settle -- or for time to run out -- before picking classes. We know that more students are attending part-time, which is partly a function of the job market, so that may be a factor.

Given enrollment declines generally, there may be some effect from a relative lack of urgency. When the most popular time slots and professors went quickly, students had a reason to register early. When plenty of good seats are still available at the end, though, there’s less perceived risk in waiting.

The question matters because late registration hurts both the college and the students. It hurts the college to the extent that it makes planning more difficult. Decisions about which sections to cut are harder to make when more students wait until the bitter end. It hurts students to the extent that they lose the chance to plan their transportation, job hours and family obligations. We know from the literature that the last students in are usually the first ones out.

Given a fiscal cushion, I’d love to move the enrollment cutoff to a couple of weeks before classes start. But we really don’t have that option, and even if we did, it would be a hard sell culturally. When demand was higher, it almost would have been redundant anyway.

Two questions for my wise and worldly readers: Are you seeing a similar pattern where you are? And if you have, have you figured out what’s behind it and how best to deal with it?