Better Data as a Remedy to Low U.S. Graduation Rates

by Thomas J. <>
Hochstettler - October 06, 2006

Imagine a refrigerator manufacturer whose product failed to keep food cold
roughly half the time, or an accountant who got the numbers right at only a
50 percent rate. Both, of course, would be out of business very quickly. 


Data on the nation's higher education system-a system whose "product" is far
more fundamental to our national future than any refrigerator-indicates that
our colleges and universities are graduating only 51 percent of our students
within five years of their entry. More startling yet, the Education Trust
finds that 50 U.S. colleges have a six-year graduation rate below 20


The comparison between a refrigerator and a college education is imperfect,
to say the least. But few would defend failure rates of this magnitude. That
is why I find it surprising, and certainly frustrating, that many in higher
education are opposing a proposed national database about student progress
and outcomes that I believe would do much to help us diagnose performance
shortcomings in higher education and, better yet, fix them.


The issue of a database has been the subject of public debate since its
proposal by the so-called Spellings Commission earlier this year.  As part
of a sweeping proposal to reform American higher education, the federal
commission is urging the establishment of a privacy-protected information
clearinghouse to track students as they pursue their varied paths through
the higher education system and into the workforce. The commission's
proposal, importantly, specifies that data on individual students will not
be matched with social security numbers or other personal identifiers, thus
ensuring that Uncle Sam does not learn what classes a student took or
dropped, or how well he or she performed in them. The basic concept is to
apply to our college and university system the same rigorous information
gathering and analysis that we expect of our faculties and students in the
scholarly work they undertake.


Students are falling through the cracks, and-surprising for an enterprise
such as ours-we have done very little to understand the reasons. Why do
students drop out? Where do they go when they do? What factors in primary
and secondary school, beyond GPAs, class rankings, and standardized test
scores, best predict their success or failure in college? Other than large
endowments and hyper-selectivity, what common ingredients are shared by
colleges that graduate students at rates better than 90 percent? What impact
does students' educational experience have on their success or failure after


We are at present ill-equipped to answer these questions. Without basic
information, both individual institutions and society lack the tools to
assess how the system is working, how it is failing, and how it might be


Much of my colleagues' criticism of the Spellings Commission database
centers on the perception that we can't trust the federal government with
such sensitive information.  Privacy concerns are understandable,
particularly in a time when politicians have demonstrated a willingness to
overlook basic American liberties to protect national security or make
political points, depending on one's perspective. Yet I believe the solution
to that problem is not to squander an opportunity to gather much-needed data
on our sector but, rather, to establish such checks and balances within our
political and judicial systems as will ensure that such information remains


Moreover, those objecting to the database have not, to this point, explained
how the Spellings Commission proposal represents a dramatic departure from
federal data collection projects that have been underway for decades and are
familiar to us all. Most notable among them is FAFSA-the Free Application
for Federal Student Aid-which requires aid applicants to provide their
Social Security and driver's license numbers as well as information on their
families' income and assets. 


Where some see the proposed database as a Big Brother peering over our and
our students' shoulders, I see a potential for a robust (and
privacy-protected) set of metrics that would yield essential data with
tremendous potential for advancing our individual institutions and for
identifying with greater precision those areas where our national education
policy needs to be strengthened. Where some see the specter of government
intrusion, I see the possibility of transforming our current separate data
reporting schemes into a streamlined system that is beneficial to students
and useful to faculty and administrators. Such a system could help us bring
much needed improvements to graduation rates, to be sure, and accomplish so
much more for our students and the country.


We can be sure that other countries are collecting such data and putting
them to good use-countries that have designs on matching, and surpassing
America as the home of the world's best higher education system. 


Having worked as a college administrator in Germany, I know that data on
student progress and outcomes are carefully tracked and studied in that
country. The 19 European countries party to the Bologna Declaration-among
them the Czech Republic, Sweden, and Turkey-are not only collecting such
student information, but also sharing it with each other as part of their
commitment to reciprocity in their university systems. Other would-be,
long-term rivals to American supremacy in higher education-China and India,
for example-are certain to be gathering such intelligence and putting it to
good use. The consequences for national security and economic
competitiveness are both obvious and profoundly important.


Could a database such as that proposed by the Spellings Commission help us
improve low graduation rates? Certainly yes-that, and much more. It is time
that those of us leading our colleges and universities stop imagining
ourselves above scrutiny and devote ourselves to the task of improving our
contribution to our students and the nation.



Cite This Article as: Teachers College Record, Date Published: October 06,
2006 <>  ID Number:
12775, Date Accessed: 10/9/2006 9:20:53 AM





To access the LRNASST-L archives or User Guide, or to change your
subscription options (including subscribe/unsubscribe), point your web browser to

To contact the LRNASST-L owner, email [log in to unmask]