Must Headlines Deceive?

A February 23, 2024, article in Diverse headlined “Growing Number of College Grads Earn Less Than the Typical High School Graduate” was based on a report from the HEA Group – a headline mirrored in other national media (including CBS MoneyWatch, Forbes, and the New York Times).  That report examined the median earnings of graduates from 3,887 US “higher ed institutions” and found that 74% of these graduated students who earned more than the estimated earnings of a typical high school graduate (26% did not – 1022 institutions). 

Unfortunately, the HEA report was fatally flawed and misleading, a typical example of research that starts with the headline and then produces an “analysis” to support it. It is also an example where the media fails to independently assess the accuracy of claims, instead accepting the words of Washington insiders (the author of the HEA report is a former administrative official in charge of the College Scorecard data on which the report was based).Dr. Dale Lehman

There are many problems with the HEA report but the most serious is that it fails to mention that only 1,657 of the 3,887 institutions are actually bachelor’s degree institutions (less than half). 1,363 are Certificate institutions (mostly beauty “colleges”) and 867 are community colleges.  Of the remaining 1,657 true four-year colleges, only 84 had graduates earning less than the typical high school graduate 10 years after graduation – 5% rather than the 26% cited in the study.

But the problems continue. These 84 colleges contain 31 that are in Puerto Rico, where median high school graduate wages are 43% of the rest of the country (and all have median graduate earnings greater than those of high school graduates in Puerto Rico), 21 are religious institutions (e.g., bible colleges, Yeshivas, etc.),  and 7 specialize in arts or music, leaving at most 25 true four-year “colleges” for which earnings are below those of the typical high school graduate (1.5% rather than the reported 26%).  The headline could have read: “Virtually All Colleges Have Median Earnings 10 years After Graduation that Are Greater Than Typical High School Graduate Earnings.”

Admittedly, that is a low bar – we should expect college graduate earnings to do better than that.  But why inflate the number to generate a headline and why is the media so willing to run with it without doing any independent analysis?  The data is available (the HEA Group did provide that) and not difficult to examine – which I did.

I eliminated the for-profit colleges, community colleges, beauty schools, arts and music colleges, schools in Puerto Rico, and explicit narrowly focused religious schools (e.g. seminaries), most of which exhibit relatively low graduate earnings, to obtain a more accurate picture of how college and high school graduate earnings compare.  I also corrected the typical high school graduate earnings used in the HEA report ($32,000 which appears to be earnings of those without a high school degree – the correct high school graduate median earnings are $39,700).   Notably, a number of these are HBCUs or Tribal colleges or universities.  9% of the remaining 1418 colleges have graduate earnings lower than the median earnings of a high school graduate (7% if weighted by the number of graduates – the low earnings schools are disproportionately small).  That is still too high for the significant investment of time and money required to get a college degree.  But rather than the shocking headline “over one quarter, the true picture is that most college graduates earn more after receiving their degree.

The College Scorecard data contains many interesting discoveries, as Michael Itzkowitz the author of the HEA report surely knows (since he was the director of it).  There are significant differences in earnings depending on field of study – and these are as significant as the differences between colleges.  The colleges with low graduate earnings are in particular segments of the higher education industry – often with missions that eschew the motivation for future earnings.  Much of the difference in graduate earnings across colleges can be attributed to the type of students they accept, including test scores, high school GPA, family income, gender, ethnic, and age characteristics.  Once these are taken into account, the remaining differences between colleges is relatively small.  The importance of each of these factors is interesting and important and not at all analyzed in the HEA report.  The College Scorecard data is an excellent opportunity for college applicants to learn how to analyze data and make their own meaningful discoveries – skills that will last them a lifetime. 

But the HEA report was designed to persuade rather than inform.  There are many problems with the value and cost of a college education, but the data contains far more nuanced and interesting stories than the headline produced by the HEA Group.  This was an opportunity to educate the public, but one lost due to the desire for headlines and political points.  It is disconcerting that the media just ran with that headline with no analysis of their own.  The motives of Mr. Itzkowitz and the media are unknown to me. My contact attempts (with the exception of Diverse) were met with silence.  The headline remains. Time to move on, I guess.

Dr. Dale Lehman is a professor of Business and Public Policy at Alaska Pacific University