We love research. Studies grounded in empirical research drive a lot of what we do and who we talk to for State of Opportunity reporting. Yet, if we take a step back, maybe we should periodically reflect on the actual practice of research, in addition to intent and outcomes.
In the documentary, American Winter, the filmmakers create a striking balance in their storytelling. On the one hand, they elicit empathy for the film's subjects through effective storytelling. But they also give the audience striking statistics about the dramatic increase in homelessness and poverty in the US [factsheet]. Those stats are often linked to data showing that having people in poverty and state-dependent is, in the long-run, not the most cost-effective way to help people move up the economic ladder.
Similarly, our stories here at State of Opportunity have asked whether pre-school is worth the cost. In the short run, the research shows, yes, it's far cheaper to pay for Head Start now than to pay the Corrections Corporation of America later.
Data can be used to prove whatever point one wants to make. And this Wall Street Journal article talks about the partisanship that enters into how research on pre-school is conducted.
Do we really want 300 kids to have high quality preschool while another 300 languish at home for the sake of an experiment? Probably not.
Conservative voices question the representativeness of small-scale studies, such as the much-heralded Perry Preschool study in Ypsilanti. However, researchers who advocate for early education note the ethical considerations of, for example, a large-scale study that provides a potentially beneficial treatment to one group, but not to another. Do we really want 300 kids to have high quality preschool while another 300 languish at home for the sake of an experiment? Probably not.
Long-term studies about pre-schools' effectiveness. Major, large-scale sampling. Academic research that is easily translatable to policy and practical applications. Rather than arguing over whose outcome and data are more biased, we should be looking for combined efforts to have better, more empirically grounded, ethical research.
