PDA

View Full Version : Survey Of Duke Frosh Touching On Athletics



hallcity
11-30-2018, 05:04 PM
The Chronicle has done a survey of the freshman class (https://www.dukechronicle.com/article/2018/10/class-of-2022-survey-duke).

Here's a few questions related to sports:

Number of games watched (I assume Duke games but it's not clear what kind of games or when)
0 44%
1-10 40%
10-20 9%
20-30 3%
30+ 4%

Likeliness to tent
Definitely not 6%
Might or might not 38%
Probably yes 28%
Definitely yes 11%

And I'll include this one since there has been discussion about whether race/ethnicity affects interest in college athletics
White 44%
Asian 39%
Black-African American 6%
Hispanic 5%
Native American 1%
Other 6%

By the way, I'm happy to see that Duke appears to not be following the lead of Harvard and some other schools which have policies that have the effect of limiting the number of Asian-American students admitted. If Asian-American kids are less interested in college athletics, so be it. Duke needs to admit the best.

I wonder whether an undergraduate's interest in college athletics predicts that same person's propensity as an alum to donate to their alma mater. I've heard that former student athletes donate at a higher rate than non-athletes. What about those who just attended football and basketball games as students? Are they more likely to give? I'd guess so but I don't know.

Bluedog
11-30-2018, 05:12 PM
1. There is already another thread on this article.
2. The basketball watching refers to how many games students watched BEFORE they arrived at Duke
3. The race/ethnicity breakdown is the survey participants, not the demographics of the Duke student body. These should theoretically match relatively closely in a well defined study, but they don't as this was an unscientific survey.

BLPOG
11-30-2018, 05:16 PM
1. There is already another thread on this article.
2. The basketball watching refers to how many games students watched BEFORE they arrived at Duke
3. The race/ethnicity breakdown is the survey participants, not the demographics of the Duke student body. These should theoretically match relatively closely in a well defined study, but they don't as this was an unscientific survey.

Additionally, the total number of participants in the survey is small in absolute terms and as a fraction of the class. The truth is that the error bars on these questions are large, as are the selection effects in terms of participants. I hate to sound so cynical, but the survey is pretty much garbage.