Friday Nights in 2013 and 2014

August 15, 2015

My New Year's Resolution in 2013 was to write down at least one thing every day that made me smile. The happiness log that began as a piece of paper taped to the wall next to my bed has since blossomed into a beautiful series of Google spreadsheets that contains damn near everything — the good, the bad, and the ugly.




The data that I collect is far from structured or sophisticated. The easiest thing you can do with raw text is analyze word counts and frequencies, so let's start there. How much did I write in 2013 and 2014?

Average Words per Entry over Time

I was committed to minimalism for a pretty long time. At first, circumstance was to blame. Trying to form a new habit + handwriting entries = an unsurprising average of 14 words per day. I switched to electronic logs at the beginning of 2014 but continued to keep them short and sweet until I got the itch to elaborate in March. The busyness and excitement associated with graduation becomes evident as June approaches, and then BAM — word count skyrockets when I ship myself off to Europe in July. The volume of content tapers once I return stateside in September, until I experience a particularly traumatic breakup in November. Naturally, my instinct is to document everything in meticulous, painstaking detail, and the spike in the graph is the fruit of those efforts.

I never really had a plan for what to do with the rest of the data. I mostly wanted the logs for reference and memory-jogging, because my brain does pretty good when you feed it a little context. But it just so happens that OkCupid, the free online dating site that most of the people reading this are probably familiar with, has a profile question about typical Friday nights. The day I realized I could answer that question with data was a beautiful day indeed. This was in early 2014.




I isolated my Fridays and read through the entries a bunch of times to get a sense of the types of things I was doing. It turned out that each evening could be described pretty nicely by three criteria — who I was with, what I was doing, and where I was doing it. After a little more information architecture witchcraft and many long hours on Excel and Microsoft Paint, I managed to produce this pretty bugger (viewable in full size here):




I was particularly interested in how my Fridays would change after I was legally allowed to drink. I didn't care enough about drinking to, say, acquire a fake ID, but I was SO excited to be able to go to meetups at bars, 21+ concerts and events, etc. I wondered if there would be a change in who I hung out with, or where we went, or how much I imbibed, since drinking is such a huge part of culture here.

Before I turned 21, if I had any alcohol at all on a Friday, I was certainly binge drinking. After I turned 21, if I was out at a bar, I almost never had more than a drink or two. Being a zero-income college student whose closest friends were largely still under 21, I didn't see the point in going to bars with people I'd be hanging out with anyway and ordering a bunch of drinks that cost four times as much as they would at home. I continued to drink heavily at parties, however.

Another interesting observation about 2013 is that I never spent a Friday entirely alone. Even if you subtract the time I spent shut in my room on Google Hangouts with internet friends, I always had some form of face-to-face interaction with people beyond the friends I was living with. Though the same probably wasn't true for Saturdays or Sundays or Wednesdays, my Fridays appear to have been more outgoing than my perception would have pinned them to be.

When 2014 came to a close, I was really excited to see how it compared to 2013. I thought the who/what/where categorization was very effective, so I stuck with it again. This time, I employed some of my newfound programming skills to make the visualization a bit more interactive. That chart can be found here. But this is what it looks like static:


2014 was a year of significant change. I graduated from college, spent 3 months traveling (mostly alone), and started my first Big Kid job. I figured that my Friday nights would be drastically different in each of those phases, so I delineated them on the timeline. The only surprising thing about the subsections is how little changed about my life after I started working. I was finally living in the city proper, but the people I spent time with on Friday nights weren't drastically different.

Even with the tweaks I employed to make the Fridays 2014 chart easier to read, it still requires a bit of work to answer questions like who I ate the most dinners with, what I did when I was hanging out with my family, or where I was when I'd go out to dinner. Turns out heat maps are good for that sort of thing, even though they lose the ability to look at change over time.



How to read: choose a row on the y-axis. The darker the cells along the x-axis, the more often I did X on Friday nights when I was already doing Y. If we choose "homework" on the y-axis, we can see that I had a drink or two every night I stayed in to focus on school. But if you choose "drink (1-2)" on the y-axis, we see that I barely ever did homework on nights I was having a drink or two. Doing homework? Probably drinking. Drinking? Probably not doing homework. It's easy to get confused when making these distinctions, but the chart is not symmetrical, so the way it's read is important.

I walked most frequently when I was traveling, I drank the most when I was with friends of friends and people I met online, I drank frequently when I was with my significant other but never very much, I always saw my family when I was in my hometown, and I exclusively hung out with my high school friends in the city, even though many of them lived in my hometown at the time. Furthermore, I did the most diverse set of activities with my college friends and my family, whereas I only ever drank with colleagues, and much of the time I spent with high school friends was just bumming around.

I hope everyone realizes that despite the numbers and analyses and fancy graphs, every little inch of this is constructed with subjectivity and bias. The Friday charts by no means paint a full picture of a True Friday Night, just an overview of the people, places, and activities I chose to include. We can learn a lot by analyzing both the qualitative and quantitative changes in categories between 2013 and 2014.

Qualitative Differences in "Who"

In 2014, I made fewer distinctions between my friend groups in college, which was reflective of my desire to distance myself from school. On the other side of the same coin, I made more distinctions between my Chicago-based friend groups, who I hung out with more often.

I also created a distinction for romantic interest and significant other because non-monogamy became a much more important part of my life in 2014. I spent most of 2013 single, so there wouldn't have been much of a need to differentiate anyway.

Ever curious to investigate how my behavior changes around different groups of people, I also created a distinction between strangers (people who I didn't know and didn't expect to see again) and friends of friends (people who I didn't know but wasn't anonymous around).

Quantitative Differences in "Who"

I spent WAY more time hanging out with people I was dating, more time by myself, and less time with pretty much everyone else. My college friends felt the most drastic difference (of the 24 college-phase-Fridays, I spent nearly half off-campus and only 10 with college friends), followed only by my online friends, who I stopped doing Friday night Hangouts with altogether. My family and hometown friends took sizeable hits, too.

Qualitative Differences in "What"

In 2014, I created a category for "personal projects" that could include 2013 labels like "writing" and "crafting" and also things like working on this website or doing data analysis. I also created categories for homework, shopping, errands, walking, and phone calls/facetime - apparently I became more comfortable spending my Friday nights doing less wild and more calm/adult/responsible things.

Quantitative Differences in "What"

I binge drank significantly less in 2014 and casually drank significantly more. All in all, I was drinking on more unique nights, but probably less total alcohol. I also went out to eat much more often, and spent less Friday nights having unstructured quality time with people (which seemed to compose about half of what I did in 2013).

Qualitative Differences in "Where"

There isn't anything super profound here. In 2014, I decided to distinguish the regions of Chicago I was hanging out in. When I was still in school, if I trekked to the city, I was more likely to stick to the North side due to proximity. When I moved to the Northwest side, I was unsurprisingly more likely to spend time nearby, but I traveled North more often than I expected (including trips back to my college town).

Quantitative Differences in "Where"

I spent waaaay more Fridays traveling in 2014 (long-distance relationship + summer travels) and waaaay less Fridays in my college town (school year ended in June and I promptly moved away). Nothing surprising there.

As always, the fact that I create these graphs says more about my life than the graphs themselves, but I'm nevertheless excited for the next year of analysis.