Tuesday, April 19, 2011

Testing and the value of education

The idea that educational degrees are just a screening tool has harmed education.  Education clearly produces useful skills like socialization, reading, mathematics, responsibility, communication skills, knowledge of facts, etc.  These are all costly to acquire and social-learning environment (classrooms) reduce the costs to most individual of acquiring them.  The recent vogue for testing is partly an effort to measure how productive schools are.  Unfortunately, the evidence indicates that standardized tests are almost useless at measuring useful skills.  For example, Heckman's research indicates that the GED is almost useless as a screening tool for measuring skills that are useful in a workplace.  Similarly, MBA programs are extremely expensive and we already have an exam to measure MBA-readiness: the GMAT.  But no companies pay a premium for potential employees who have high GMAT scores, so they indicate that they think that the GMAT is useless as a signal of skills. 
ETS has major-field tests in many disciplines, but no graduate schools use them for admission instead of an undergraduate degree .  Nor do companies use them in place of a bachelor's degree.  This indicates that they are nearly useless for employment or grad-school admission in the absence of being accompanied with a bachelor's degree.
In some fields (like math?) an exam might get close to testing actual competence, but perhaps even here it does not accomplish the same thing as an educational degree. 
The big standardized testing companies, American Council on Education (GED), ETS, ACT, and the College Board, could make millions if they could come up with a test to replace the value of a college degree.  For-profit colleges have a similar incentive.  The fact that nobody has come up with one is an indication of the value of standardized testing for measuring useful skills.