0  0

The graph *did* account for non-response. Note the asterisk in the bottom right that says “27.39% of students responded that they did not know”.

Also I think their choice of using different sized ranges is somewhat justified. The difference in lifestyle (perhaps the larger implicit question the survey was addressing) can increase greatly with a 20k jump from 60k to 80k a year, but not so much from 160k to 180k, and less yet say, from $1,060,000 to $1,080,000.

Breaking up the groups the way they did – even if another 50k sized bin could have ended up with a larger share of the respondents – still makes some sense because a single group for 100k-150k is more meaningful in terms of the larger implied question – lifestyle and ability to afford a WashU education – than a 50-100k group (especially since tuition is something of a fixed sum, unless one’s financial aid greatly mitigates that). Unless that is, there was a large subcluster say, at 100k-120k. That might have justified a 100k-120k of its own, but the surveyors couldn’t have foreseen that.

Certainly methodological issues such as student information of their parents’ income is a concern, but that is true of any survey. Given the legal issues of trying to gather data from the registrar’s office or the parents themselves, this current method is understandable.

Objective definitions for “upper-middle class” or “middle-class” don’t really matter here because it’s all about the subjective perceptions of the students anyway. Their intent was to see how students felt they matched up with their own conceptions of privilege. Assuming that most people don’t have distorted subjective perceptions – viewing “upper-middle class” as a squalid existence or “working class” as luxurious, their concept is valid.

Though I’m with you on the 180k not falling into the bracket surrounding the 50% point for responses…I’m curious how that worked out.

0  0

It works both ways. One of my University College students told me she avoided the day school courses because of frustration dealing with “naive privileged rich kids.” They may have something to teach us too. Why not have more cross-listing of day school courses with University College, for a more diverse classroom experience?

0  0

Keep up the good work!

0  0

0  0

The next time you do a statistical survey and try to draw any conclusions from it, be much more careful with the survey design. You should not have different sized ranges unless you provide a very good reason for doing so. The fact that there are more people in the 100k-150k range is massively unhelpful considering all of the lower income brackets are at most 20k apart. In fact, if you add up the values in the equivalent 50k-100k range, you get a much greater value than any other bracket. Not to mention, you did not account for many significant sampling errors, including:
-Failure to account for cost of living in different parts of the country. (If my parents make 100k in San Francisco, that isn’t worth as much as it is here.)
-From the graph shown, the median could not possibly be outside of the 100k-150k range.
-Your percentage values don’t even add up to 100%! If students respond that they don’t know, you shouldn’t consider them in that part of the sample. Try recalculating those percentages based on how many people actually responded.
-There is no reason to assume that students provide an accurate response. I have no idea how much exactly my parents make every year, but I do know it fluctuates greatly.
-What definitions did your group use for distinguishing the classes? If I self-identify as middle class and you say I’m upper class, what is the line of demarcation that you use to say I’m wrong? Is that line consistent with nation-wide economic definitions?

I hope someone from StudLife can help put these objections out there. For a cover story based on statistics, there is a shocking and appalling lack of understanding of the accurate representation of statistical data.

0  0

0  0

Chase Sackett
WU/FUSED Co-Chair

0  0

The mean (average) is going to be skewed badly by the small handful of unrepentant zillionaires here.

0  0

0  0