On December 8, I dissected Kevin Carey’s attempt to use frequency distributions to draw pointed conclusions about the state of affairs in higher education. Last week, I came across an article from the Chronicle that builds up the value of college degrees simply because 90% of recent grads responded positively to a survey that asked if college was “worth it.”
With more people then ever possessing college degrees, one would think that we would have the most educated populace in history. Yet why is it so easy to find examples of misusing statistical data?
According to this “young-alumni satisfaction” poll conducted by The American Council on Education:
Respondents expressed broad satisfaction with their undergraduate experience. Eighty-nine percent of respondents said their college experience had been worth it, and 85 percent said their education had prepared them at least adequately well for the jobs they now held.
About 80 percent of alumni said they would attend the same undergraduate institution if given the chance, though the number was higher at four-year than at two-year colleges.
Two years ago when looking at similar data from ACE, Ashley Thorne noted that there is a possible social desirability bias that may confound such findings. Hence, these satisfaction statistics are not the end of the research agenda; they are the beginning. These results make for good exploratory data to design more probing questions. For example, if merely “adequate” preparation is considered acceptable for higher education stakeholders, it would be interesting to see if the survey respondents see a more cost-effective way of achieving this adequate preparation.
Thus, these data points are way too general to have any policy implications. Nevertheless, a few paragraphs later, the Chronicle article has this quote and follow-up:
"For something that takes this much time and this much money, it still draws a nearly unanimous declaration of its value," said the council's president, Molly C. Broad.
Officials said the poll provided a strong argument against government officials who were thinking about cutting support for higher education.
Higher education is not justified for all because young alumni replied in a poll that college was “worth it.” The supposed wage premium for higher education has crowded the thinking of young adults and their parents to the extent that other viable options like trade school are not even a career option (e.g. see commenter ASW’s thoughtful reply to my essay last week titled “Undoing College”). In addition, I wonder if graduates with six figures of debt and no job prospects think college was worth it. Where are the correlations between satisfaction and student loan debt levels?
As a further attempt to validate the argument for the “value” of higher education, the article then presents this whopper:
"The average elected official would kill to have such a positive rating as we have in higher education," said Kevin P. Reilly, president of the University of Wisconsin system, one of the institutions that took part in the survey.
This statement has more truth if the intent is to discuss the murderous tendencies of elected officials. The Amazon Kindle has a 4.5 star average rating from over 8,000 respondents. Is that satisfaction rating a justification for the government to provide Kindles for everyone?
The more pressing issue at the moment is whether the average reader can see the misuse of statistics that forms the crux of this article that is the subject of this essay. If the higher education establishment likes poll data here’s a telling statistic – American students score in the middle of the pack in mathematical ability, yet they score near the top in their self-esteem about their math skills. It’s no wonder the American Idol tryouts are so popular.
While the college degree is not completely worthless, its current value is certainly a topic for heated debate. But, in order to have this debate, I urge the public not to roll over when the word “study” or “percent” makes an appearance.