The Affliction of Low Achievement

It’s much easier to get approval for drugs that are marginally effective in, say, half the population than drugs that are very effective in a small fraction of patients.

That’s Harvard neurologist Peter Lansbury on the problem with medical research. Given my constant harping about the similar inefficiencies plaguing medical and education research, it’s no surprise I found the quote very telling. Replace “patients” with “students” and “drugs” with “curricula” or “educational technologies,” and the statement is just as true.


The Downside to Rational Argument

In a series of 4 experiments and 2 pilot studies, we demonstrated that when the generation of counterarguments was easy, negotiators who did not add arguments to their first offers achieved superior results compared with negotiators who used arguments to justify their first offer. We hypothesized and provided evidence that adding arguments to a first offer was likely to cause the responding party to search for counterarguments, and this, in turn, led him or her to present counteroffers that were further away from the first offer.

Perhaps there’s a logic to the “climate change is a myth, period” position after all. The full paper can be found in the latest issue of the Journal of Personality and Social Psychology.

There’s no Shame in Being 90% as Good

The internet’s paper du jour is a new Columbia University study that looks at online classes in community colleges. The study found that online students had an 82% chance of finishing a course whereas students in face-to-face classes had a 90% chance of finishing.

For the most part, these numbers have been cited to critique online learning and downplay it as a potential solution, and for the most part, this is a completely misguided reaction. The right response is “Only 10% worse? That’s not bad. Not bad at all.” In fact, there are at least four reasons why reformers and blended learning enthusiasts should be high-fiving each other over these results.

First, when you’re talking about a system that can teach 10, 100, or 1,000 times as many students for the same cost, a 10% decrease in quality is nothing. If you could vaccinate 100 times as many African children but have the vaccines be 10% less effective, I don’t think anybody would say that’s a bad thing.

Second, it’s only 2011. In the grand scheme of things we’ve barely scratched the surface with educational technology. Using this study to draw conclusions about the failure of online learning is like writing off mp3s in 1997 because of poor sound quality. The study also took place over five years, which means some of the data comes from whatever primitive online interfaces were being used from 2006-2008.

Third, as with most education research, the average performance isn’t as important as it seems. Even if on average fewer people complete online classes, there are countless students whose chances of completing a class go from 0% to 99% percent because of an online option. If there’s a new cancer treatment that does the same on average as existing treatments, but magically cures 20% of people far beyond anything that exists, nobody would abandon work on it because of its “on average” results.

Fourth, there is some selection bias built into the study. The researchers attempted to control for student backgrounds, but when somebody chooses to take an online class rather than a more time-consuming face-to-face class, it means they’re more likely to have other important commitments or less overall interest that students in conventional classes. The selection bias means that the observed downsides of online classes may not even exist.

Too often when it comes to research on things like educational technology and charter schools people want results that show a new option always better. We’re not there yet. But as long innovations can do the same for less, or do more for a certain subset of people, we’ll continue to make progress.

Don’t Bribe the Tea Party

The collectivism vs. individualism debate has advanced to new territory:

We found a significant effect of the degree of collectivism versus individualism present in a national culture on the propensity to offer bribes to international business partners. Furthermore, the effect was mediated by individuals’ sense of responsibility for their actions. Together, these results suggest that collectivism promotes bribery through lower perceived responsibility for one’s actions.

For those keeping score at home, Asian countries with collectivist cultures are destroying America when it come to education, but at least once their kids grow up to be successful entrepreneurs they’re more likely to be corrupt.  I think there’s a “winning the future” joke in their somewhere, and I’m pretty sure the U.S. is the punchline.

Blended Learning Will Also Improve Higher Education

Imagine a world where each day children study a few rigorously designed 4-8 week curriculum modules. The modules are all online and students would choose from a database containing hundreds of them.  Personal finance, the Cold War, graphic design, botany, labor economics, Hemingway, Dutch, paleontology, the Pythagorean Theorem, Javascript.  Certain crucial modules would be mandatory. When the students complete the prescribed final evaluation they get credit for passing the module (and a badge!)

Most benefits to this system have been widely discussed. Being able to choose what you learn makes school more of a voluntary activity and less of a chore. Students are less likely to waste valuable school time daydreaming through something they don’t care about. Providing more choice may give accelerated students an easy way to graduate early, or give students who are behind a way to catch up.

One thing that has been overlooked is that a true blended learning system would improve higher education by making it easier to efficiently match students to universities. A student light years ahead in computer science would be able to show a college exactly what he can do. A student who loves history can take numerous classes to show his dedication. The college admissions process would move beyond GPA and SAT scores.

Once students begin to differentiate themselves universities might attempt to differentiate themselves too. For example, because a high school student could take as many biology classes as they wanted, schools wanting to attract top biology students could actually target the top biology students instead of being forced to target the best overall students. With a concerted effort a middling state school could become the #1 place for undergraduate biology. The result is that similar departments at different universities would begin to differentiate themselves in qualitative and quantitative ways. Undergraduate education would take on one of the best qualities of graduate education.

How to Hack a Salary Negotiation

Two studies were conducted to examine the effects of implausible anchors on initial salary offers. Participants provided a salary offer to a candidate after receiving a relevant anchor and an implausible anchor. The results of Study 1 indicated that a high implausible anchor influenced salary offers, even in the presence of the relevant anchor. Study 2 examined whether a more extreme implausible anchor would also affect salary offers. The results indicated that both the high anchor and the extremely high anchor led to higher salary offers than did the control condition.

The author is Todd Thorsteinson and the paper can be found in the Journal of Applied Social Psychology.

Many studies have shown the existence of anchoring effects, but Thorsteinson found the effects exist in salary negotiations even when there is an appropriate anchor (e.g. when the “hirer” is told the job candidate’s previous salary was $29,000), and even when the anchor is utterly implausible (e.g. the candidate suggests a salary of $1 million.) Thorsteinson also gets bonus points for mentioning Scott Boras’ outlandish salary demands for Manny Ramirez as a real-world anchoring scenario.

Are We Capable of Setting Real Educational Priorities?

It’s taken as a given that any educational standards must demand high achievement in as many academic subjects as possible. But all subjects aren’t equal, and pretending they are leads to inefficiencies in the education system. For example, although elementary school children are old enough to comprehend the “nature of science” (NOS), many students graduate high school without a proper understanding of evidence, reasoning, and other crucial NOS concepts.

To fix this kind of problem there needs to be a drastic shift is how we perceive educational standards and priorities. People need to go on record about what is and isn’t absolutely crucial. We could start by acknowledging that it’s worth making sacrifices in the humanities in order to do better in literacy.

The big question is whether we reach a point where the politics of our education system will allow elementary schools (or even high schools) to set real, non-standard priorities. Certain Head Start programs are already experimenting with placing a heavier focus on literacy and math, and so far the results have been promising. It’s considered blasphemy to downplay the importance of history, literature, and science (yes, much K-8 science is heavy on memorization), but doing so would be worthwhile if it ensures every student knows how to read and reason.