Research

Giving CC Students Home Computers Won't Set Them up for Greater Success

• By Dian Schaffhauser
• 02/12/18

Handing out computers to incoming community college students won't influence their success in higher education, maintaining jobs or lifting earnings. That's the bottom line for a study that examined data from several sources to understand the impact of providing home computers to students who don't have them. What it does show, however, is the possible presence of "positive selection bias" in related studies.

The research project pulled confidential data from California's employment office, the state's community college system and the National Student Clearinghouse on all participants in the study for seven years after the "random provision of computers." Results were published in a working paper by the National Bureau of Economic Research. The work was done by Robert Fairlie, a professor in the Department of Economics at the University of California at Santa Cruz, and Peter Bahr, an associate professor in the School of Education at the University of Michigan.

The study set out to understand the returns generated by a workforce with computer skills. In fall 2006, the research project handed out free computers to a random set of freshman community college students who were receiving financial aid. The researchers chose a community college because these students tend to live off campus, limiting their ready access to computer labs and other campus computing resources, "making personally owned computers potentially important for acquiring computer skills and knowledge." This group's eventual earnings and college completion rates were compared to a "control group" with comparable characteristics.

The results? No evidence that computer skills have a positive effect on earnings or employment; and no evidence that computers have a positive effect on college enrollment. In other words, "null effects."

This countered findings showing more positive results, the report noted. For example, using data from the U.S. Census Bureau Current Population Survey, the researchers found a "large, positive and statistically significant" correlation between quarterly earnings and home computer ownership; all other things being equal, computer owners had earnings over those three months that were "roughly $700 to $1,700 higher than non-computer owners." Likewise, using that same data, they found a positive relationship between computer ownership and college enrollment; computer owners had enrollment rates 12 to 16 points higher, on average, than non-owners.

So why the differences between the two sets of studies? The latter estimates were developed through "non-experimental" analysis of potential returns. As result, the report suggested, the various outcomes were probably overstated.

The lack of difference between the community college study group and the control group, on the other hand, may be a result of employers training their workers on the tech skills they need for their jobs, or other skills besides computing surfacing as important for employment. "Alternatively," the researchers wrote, "we cannot rule out the possibility that the experiment failed to detect an effect because it is difficult to perfectly alter the computer skills of workers."

The complete working paper is available for a small fee on the NBER website.