1.1.2. Intervention studies
Observational studies can only show a correlation between computer ownership or use and student outcome. Other studies use designs
that can provide stronger evidence of causation. Malamud and Pop-Eleches (2011) use an interrupted time series design: the study uses data
from a 2008 Romanian program on vouchers to low income families with children to buy computers. Therewas a significant reduction of the
Math, Romanian, and English test scores for the students that received the voucher; the effect size was from 0.2 to 0.5, that is, the group
that received computers had test scores from 0.2 to 0.5 times the standard deviation lower that similar students that did not receive the
voucher. The study also found some positive effects on cognitive skills and computer skills.
A random intervention design was used by Cristia, Ibarraran, Cueto, Santiago, and Severín (2012) in the 2009 evaluation of the “One
laptop per child” program in rural Peru. Random schools were given laptops for all their students. There was evidence of increase computer
use and some evidence of improved