Featured Post

Fix, Don’t Discard MCAS/PARCC

This fall I had one on one conversations with many of our state's leaders and experts on the misplaced opposition to testing in gen...

Thursday, May 15, 2014

Student joke responses skew some education data

Students who jokingly respond to surveys with wildly inaccurate answers can seriously skew results, with unfortunate public policy consequences, according to a study published today in “Educational Researcher,” the peer-reviewed journal of the American Educational Research Association.

Author Joseph P. Robinson-Cimpian found that “mischievous responders” can have a big effect. For instance, a recent survey on adolescent drug use found a huge disparity in cocaine usage between heterosexual students and those identifying as lesbian, gay, bisexual or transgender — but Robinson-Cimpian found that the gulf was wildly inflate because some students made up phony answers on multiple questions.
Robinson-Cimpian writes that students inclined to joke around often give incorrect answers on multiple questions, picking responses that they perceive as far out of the mainstream. So, for instance, they might report that they are gay, blind, deaf, use cocaine weekly and have multiple children — all answers that should show up relatively rarely on the surveys and are unlikely to appear in combination all that often.
Such surveys must be weeded out, Robinson-Cimpian argues, or they can skew perceptions and push public policy in unwarranted directions. He warns that the issue of “jokester youths” can also perpetuate negative stereotypes by inflating the degree to which marginalized groups such as gay and lesbian students engage in risky behavior such as drug use. His article proposes a four-step analysis that researchers can apply to student surveys to identify responses that are most likely false.


“If we want sound research and policy, we need to have sound data,” he said.

No comments:

Post a Comment