Spectrum | Feb. 15 3:00 pm EST
Campus Culture

Alone: A Columbia culture? A study of group habits in Ferris

File Photo

At Columbia, I often feel a strong sense of isolation. My friends and I are frequently too busy to spend time together, and acquaintances easily fall out of touch. I’d bet that in such a stressful school environment, others experience the same feeling. 

It seems Ferris is always rather empty in the middle but packed at the single-seating bar by the walls. My friends and I used to joke, “Oh look, the ‘antisocial’ section is in demand!” Over time, I began to get more curious. Do we actually tend to sit alone more in Ferris? Could this say something about student culture at Columbia?

I decided to observe this behavior one day in Ferris.

What I did: On Wednesday, Feb. 5th, and once an hour from 9 a.m. to 4 p.m., I counted up the number of people sitting alone and together. I selected a random sampling to approach and collected additional data, such as major and year (see more at the end). Only counts from the lower floor are included in this study. There were 98 students selected for the random survey, and 440 students were counted overall.

What I found: At first it seems there are about as many people sitting together as alone. That’s cool, but it doesn’t explain everything.


No significant difference—statistics explained below

Mother Columbia often forces us, for lack of space, to be right next to people even though we don’t know them. What if people were sitting in groups but not interacting, only because they were unable to find seating alone? When looking at how filled each section of seating was, I found significant differences.

seating by percentageStatistics explained below

This new counting might suggest that at least some of the group seating observed before could be due to spillover from the single-seating section. This makes sense, then, that our earlier numbers may have been misrepresentative. (The reverse, however, is not also true: there were only two instances where a two-person group was observed in the single-seating section—all others were alone.)

I also found some correlations between college major and whether or not you sit alone.

Check out an interactive visualization of majors most likely to sit alonestatic visualization shown here:

Screen Shot 2014-02-12 at 11.49.31 PMScreen Shot 2014-02-15 at 4.41.43 PM

We found that major is a significant predictor (p = .014) in whether a person is alone or together. Year was not a significant factor (p = .45).

What could cause this? Group dynamics studies in psychological literature offer many explanations, but which ones make sense at Columbia? While one group study shows that sitting alone is predicted by social anxietyothers show that overbearing parents  or lack of motivation to socialize can cause such behavior.

I spoke to Scarlett Tohme, CC ’14 and Floyd St Bernard-Springer, CC ’14—among others—while conducting the survey, and they agreed with the “lack of motivation” explanation.

“By senior year,” Springer said, “I’m bored of small talk. I just don’t care enough to chat with people I’m probably never going to speak to again.”

With a class size averaging 1500 students—well in the small-to-medium range of colleges—it seems strange that here at Columbia we should have the perception that we will not bump into each other again.

Approaching others is difficult. Just starting conversations with random students to run the survey was challenging, even though I consistently received pleasant responses. But sometimes, even the most basic interactions help us live happier, be less stressed and even learn better.

Although this is certainly interesting to think about, these results are certainly not conclusive—they are the results of one day’s worth of study. The purpose of this study was to illustratively stimulate discussion rather than to report a conclusive result. Share your thoughts in the comments section!

Fancy talk: Section-seating data is significantly different at α = .01. An F-test was run on linear regressions for each line. Sectional counts were divided by the 32 seats total in the antisocial section and 72 seats in the main section. Major results were obtained by a permutation test. Credit to my brother, Lucas Spangher, for helping me run the tests. 




  1. Anonymous • February 15, 2014 at 3:20 pm • Reply

    Awesome stuff! Really interesting.

    VA:F [1.9.22_1171]
    Rating: +1 (from 7 votes)
  2. Anonymous • February 15, 2014 at 3:44 pm • Reply

    Spend more time on campus. Be friendly and meet people

    VA:F [1.9.22_1171]
    Rating: +4 (from 4 votes)
  3. Specsucks • February 15, 2014 at 4:10 pm • Reply

    what the fuck is this pseudoscience bullshit? are those biology majors all non premeds? How did you even have enough people from each major? Why is the data smoothed?

    What would be more interesting would be to look at the groups and see what percentage of those were people who came together vs. joined after they saw people they knew

    VA:F [1.9.22_1171]
    Rating: -3 (from 17 votes)
    • Alex Spangher STAFF • February 15, 2014 at 4:31 pm • Reply

      Majors were self-reported. All students following the pre-med track were considered pre-med. Biology majors not following the pre-med track were considered biology majors.

      VN:F [1.9.22_1171]
      Rating: +5 (from 7 votes)
      • Specsucks • February 15, 2014 at 4:33 pm •

        and about how many of the 98 were each of those 10 majors? Were they all only those 10 majors?

        VA:F [1.9.22_1171]
        Rating: -1 (from 5 votes)
      • dude • February 15, 2014 at 4:51 pm •

        You don’t need enormous sample sizes for a permutation test

        VA:F [1.9.22_1171]
        Rating: +3 (from 5 votes)
      • dude • February 15, 2014 at 5:15 pm •

        but to generalize about an entire major from ~10 kids you found in ferris is…

        most kids who eat in ferris are first-years that haven’t even declared their majors, no?

        and then you results of “Year was not a significant factor” directly contradicts the only 2 people you quote, who said that by senior year they weren’t motivated anymore…

        VA:F [1.9.22_1171]
        Rating: +6 (from 8 votes)
      • well • February 15, 2014 at 5:32 pm •

        they do acknowledge that at the end:

        “Although this is certainly interesting to think about, these results are certainly not conclusive—they are the results of one day’s worth of study. The purpose of this study was to illustratively stimulate discussion rather than to report a conclusive result.”

        it sorta sounds like you’re a pre-med and you’re bitter.

        VA:F [1.9.22_1171]
        Rating: +2 (from 4 votes)
      • dude • February 15, 2014 at 7:01 pm •

        well it looks like it’s doing its job at stimulating discussion…

        I still fail to see how major would be a determining factor in why people sit together or don’t. Certainly everyone that sat together wasn’t of the same major and then there are very opposite “results” for similar majors (dance/music, bio/pre-med).

        VA:F [1.9.22_1171]
        Rating: +1 (from 3 votes)
  4. CC'16 • February 15, 2014 at 8:32 pm • Reply

    I’ve made several friends due to talking to people in the dining halls. Often, it just takes something little like a smile or a “hi.” Regardless of whether you eventually become friends with these acquaintances or not, it’s definitely worth it to get to know a bit about the awesome people on this campus! And for those who say they can’t be bothered, that’s your loss.

    VA:F [1.9.22_1171]
    Rating: +2 (from 2 votes)
  5. BC'14 • February 16, 2014 at 12:18 am • Reply

    I wonder how this compares to dining halls in other schools

    VA:F [1.9.22_1171]
    Rating: +2 (from 2 votes)
  6. yogi bear • February 16, 2014 at 12:19 am • Reply

    I wonder how this compares to dining halls in other schools

    VA:F [1.9.22_1171]
    Rating: +1 (from 1 vote)
  7. Chris Morris • February 19, 2014 at 10:52 pm • Reply

    Great Article, there is something to that polisci iscolation thing.

    VA:F [1.9.22_1171]
    Rating: 0 (from 0 votes)

Leave a Comment

Be nice. Don't use HTML tags. And consider reading our full comment policy.