<p>The Tech (MIT's campus newspaper) did a survey of students about their work-life balance and the amount of pressure they feel they are under. These are the sorts of questions that prospective students ask us students and alums pretty frequently around these parts, so I thought the survey results would be of interest. (Also, I have a deep and abiding love for data. Occupational hazard.)</p>
<p>One thing I find particularly fascinating is that, on average, everybody at MIT thinks he or she spends less time doing work than the average MIT student, and that only about 30% of MIT students think they are academically above average. Most people consider themselves highly stressed, but also pretty happy.</p>
<p>Thanks for sharing, Mollie. These results are extremely well presented and, yes, very interesting.</p>
<p>I am surprised that more seniors responded that they are academically above average in every category. I would tend to think that it be the other way around-- pre-frosh/freshman think they are above average until they start at MIT and realize that everyone else is as smart or smarter. However, I am not sure if that is significant. Any way to get stdev on these results? I haven’t looked deeply enough into it to see if that’s available. There are what appear to be some awesome linear trends here. It would be even nicer if these graphs were presented with error bars for significance to see whether those trends hold.</p>
<p>Although the results are definitely interesting the survey methodology was less than ideal so you should exercise considerable caution when looking at results. The survey gathered responses by sending out emails </p>
<p>
.</p>
<p>Notably, it starts off with leading questions which both skew the pool of respondents and prime those who do respond. My understanding is that these are both pretty serious survey design errors that have the potential to significantly skew results [you’d be surprised how much question wording can affect results].</p>
<p>It’s especially fascinating to see the gaps in pressure felt by female and male students…
9& of female students felt their accomplishments were impressive compared to an average MIT student’s. </p>
<p>I also found the data on majors particularly math majors (course 18) fascinating. Math majors spent the third lowest time working on homework every week [behind business and general humanities majors]. This is despite of presumably spending less time in class as advanced math classes only meet 3 hours per week and not having labs. I also think math UROPs tend to be relatively rare too. Math majors are also the most likely to consider themselves above average academically at MIT (56%) although they are only somewhat more likely to consider themselves above average among math majors (35%). The 21% gap between above average at MIT and above average in their major is also the largest.</p>
<p>
</p>
<p>I think the survey is somewhat confusing in this regard but from how I interpreted this is barely true. Undergrads reported doing 26.22 hours of homework per week. Undergrads estimated the average undergrad does 28.93 hours of homework per week. This is close enough that it’s possible the discrepancy is because the population that filled out the survey differs from the entire student body. However, grad students though undergraduates did about 38 hours of homework per week which gave raised the average estimate of hours undergraduates spent on homework per week to 33.60.</p>
<p>Another observation is that freshmen spend about 6 hours less per week on homework than upperclassmen.</p>
I actually feel like it makes sense to me – after all, GPAs rise linearly as students progress through MIT, so it doesn’t surprise me that self-perception of academic competence rises also. And very few people fail as many tests as seniors as they did as freshmen.</p>
<p>But I would also dearly love to see some statistical parameters on the data, even if, as UMTMYP student says, it’s probably subject to some survey design flaws.</p>
<p>
I was looking at that question via the dorm cross-tabs, and the trend is there for every dorm. It’s subtle, no doubt, and I’m not staking my life savings on it being true. But that’s the direction of the trend in every dorm.</p>
<p>I had the same problem with the survey, as well as with the hours of homework and number of credits. ESPECIALLY the number of credits. For instance, The Tech tries to make a point that students “think” other students are taking more credits, when actually other students take, on average, something like .6 credits less than expected (or some other negligent number). This to me barely has any statistical significance, and as you said, the email itself was very leading. I filled it out while being not stressed at all, but I feel like I may have been an outlier. It would be interesting to see an actual distribution of the responses to get a better picture of the “sample” that was surveyed.</p>
<p>Where did you find the data on number of credits? That could potentially provide a useful check on the accuracy of the survey as the registar presumably has complete and accurate data on the number of credits taken by students.</p>
<p>^I read it in The Tech. And yes, but unfortunately they didn’t check it. They only asked us to tell them how many credits we are taking (relying on honesty, I guess). Plus that is more efficient given the small sample size…</p>
<p>Although this thread has been inactive for about two weeks I thought this was important enough to revive it. Earlier in this thread, I voiced skepticism over the number of self-reported hours of work. I was browsing through MIT’s senior surveys and found data that supported my earlier skepticism. <a href=“MIT Institutional Research”>MIT Institutional Research; includes data on “During the fall term of your senior year, about how many hours did you spend during a typical week doing the following?” on page 19. The responses for “studying or doing research outside of class” are [in hours]
0 3.3%
1-5 6.3%
6-10 15.1%
11-15 18.1%
16-20 18.7%
21-25 14.3%
26-30 8.8%</p>
<br>
<br>
<p>This is hard to square with seniors doing 28.47 hours of homework per week as reported by the Tech. There are a number of possible conclusions that ca be drawn from this but none of them lead to me giving much credence to the Tech survey.</p>
<p>For one, studying and doing research is not the same as doing homework. For another, I have no idea how many hours a week I actually spend doing work (does bashing my head against my desk between p-set problems count? what about the time I spend stuck, when I’m not really doing work, just trying to figure out what I’m doing? what about the time I spent an hour stuck on a single sentence of an essay?). I guesstimated on the survey, and I assume everyone else guesstimated as well.</p>
<p>Yes those are all good points but my larger point is that we should be generally skeptical of self-reported time estimates and drawing sweeping conclusions from them.</p>
<p>It’s not perfect, but it’s better than nothing (which is what was there before). If you were to redo this study, what specifically would you change?</p>
<p>I think you just have to acknowledge that it’s a potentially serious problem and answers might vary considerably on how it is asked [some of the other surveys included lots of other activities]. Standardizing the surveys would also be an improvement instead of asking slightly different questions each time. I still consider that kind of data interesting but I don’t put a whole lot of faith in it and I wouldn’t draw sweeping conclusions about stress at MIT from it.</p>
But that’s the direction of the trend in every dorm.</p>