Tag Archives: education

“Personal anecdote trumps data”

3 Sep

What “works” in education? Educational stakeholders and international development workers alike strive to answer this question and, ideally, apply their answers through real-world programming that is contextually suited to the specific populations the programmes seek to benefit. The NYT ‘Science Times’ special section on “Learning What Works” highlights attempts made by the Institute of Educational Sciences to discern what truly “works” through the use of randomized control trials (RCTs). To date, 175 trials have been supported by the institute, testing the effectiveness of instructional supplies, curricula, and computerized learning platforms.

I spent most of my graduate studies at NYU learning of the burgeoning value of RCTs in the field of educational development, with some professors incorporating RCTs into their own research during my time there. Coming across this piece in the Learning What Works series illustrates the importance of utilizing these assessment methodologies:

Without well-designed trials, such assessments are largely guesswork. “It’s as if the medical profession worried about the administration of hospitals and patient insurance but paid no attention to the treatments that doctors gave their patients,” the institute’s first director, Grover J. Whitehurst, now of the Brookings Institution, wrote in 2012.

But the “what works” approach has another hurdle to clear: Most educators, including principals and superintendents and curriculum supervisors, do not know the data exist, much less what they mean.

A survey by the Office of Management and Budget found that just 42 percent of school districts had heard of the clearinghouse. And there is no equivalent of an F.D.A. to approve programs for marketing, or health insurance companies to refuse to pay for treatments that do not work.

Unfortunately, this type of research is costly – not only in terms of funding and staffing required to carry out the data collection, but RCTs take a significant amount of time, and as implied by the name, also require a level of control that may not be available in the case of education in emergency situations. This is a challenge faced by ministries of education and education stakeholders alike. However, without proper assessment of programs, the likelihood of education programming doing more harm than good inevitably increases. Regardless of where your heart is and what your intentions may be, some ideas are simply bad ideas. When we ignore the data coming out of the field, there can be no hope for educational development anywhere in the world, in any context. Even the best laid plans and ideas can flop if not paired with sufficient assessments and evaluations (a new book critiquing Jeff Sachs’ Millennium Villages Project highlights exactly that). Research truly is the key to sustainable interventions, and I am very fortunate that my graduate studies and my professors at NYU were able to place such a strong emphasis on that fact.

Don’t worry field, we’re bringing our skills to the table.


Liberia students all fail university admission exam

26 Aug

Liberia students all fail university admission exam

I don’t have any substantive commentary to add to this headline: it is so bizarre as to be worth a repost, and it allows me to actually mention education (for once), but beyond that, I really have no idea what to make of this. Usually I’d associate absolute numbers like this with the sort of brazenly doctored statistics of authoritarianism – zero crime, perfect literacy, entire electorates, etc. – but I can’t see why anyone would contrive this. If anything, this seems a moment ripe for a little less transparency: surely someone could have bribed or flattered their way to a passing grade?