Most everyone recognizes that talking on a cell phone while driving is a bad idea. It's distracting and so makes driving more dangerous.

It might seem that talking on a hands-free device would solve the problem, but it has been known for a while that's not so (e.g., Treffner & Barrett, 2004). Most of the distracting effect of mobile phone use is not due to looking away from the road to dial, nor the cost of holding the phone.

Talking on a phone is more demanding than talking to someone in the car: the audio signal is degraded, you have no visual information to rely on, your conversational partner doesn’t stop talking if the road situation becomes complicated, you may try to imagine being with the other person, and so on. (It is true, however, that driving is somewhat worse with hand=held than hands-free devices; Backer-Grøndahl & Sagberg, 2011.)

So what's the solution? Presumably, not answering your phone.

Now new data show that even that carries a cost.

A recent study (Holland & Rathod, 2013) had 27 young drivers use a highly realistic driving simulator, shown at right. Seven hazardous situations developed during the simulation (e.g., a pedestrian crossing the road, a car pulling out). Subjects had provided their mobile number to the experimenters, were aware that it might ring during the simulation, and that they should not answer it. The subject's cell phone rang during three of the hazards.

Seven dependent measures were collected, including crossings of the center line, collisions with other cars, and so on.

Ignoring the ringing mobile phone carried a cost. When their phone rang just prior to a hazard, subjects were more likely to hit pedestrians, to exceed the speed limit, and to cross the center line.

It is notable that these effects were more pronounced in subjects who said they usually answered their phone while driving than in subjects who said they did not. Hence, a significant contributor to the distraction may be the mental effort required to inhibit a habitual response.

What's the solution? The safest practice for drivers may be to turn the phone off, not just ignore it.

References


Backer-Grøndahl, A., & Sagberg, F. (2011). Driving and telephoning: Relative accident risk when using hand-held and hands-free mobile phones. Safety Science, 49, 324-330.

Holland, C., & Rathod, V. (2013). Influence of personal mobile phone ringing and usual intention to answer on driver error. Accident Analysis & Prevention, 50, 793-800.

Treffner, P. J., & Barrett, R. (2004). Hands-free mobile phone speech while driving degrades coordination and control. Transportation Research Part F: Traffic Psychology and Behaviour, 7, 229-246.

My Facebook feed today has lots of links to this article. The upshot: a new Pew study showing that Americans think that US 15 year olds rank "near the bottom" on international science tests, whereas the truth is that they "rank in the middle among developed countries."

I guess "the middle" covers a lot of terrain, but the way I look at the data, this assertion doesn't hold.

The international comparison in question is the 2009 PISA. Here are the rankings. (Click for larger image)

Most everyone would agree that it's not appropriate to compare scores of US kids to those of poorer countries with little infrastructure and funding to support education.

That's why the article specifies the ranking of the US among "developed countries," and by the author's reckoning, kids from 12 developed countries scored better, and kids from 9 developed countries scored worse. That would put US kids at the 41st percentile.

The US is ranked 30th on the list. Just eyeballing it, it's hard to see how 17 of the countries scoring better could be considered "not developed."

On measures of "developed" status would be the International Monetary Fund's definition of "advanced economies" which includes: Australia, Austria, Belgium, Canada, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hong Kong, Iceland, Ireland, Israel, Italy, Japan, Luxembourg, Malta, Netherlands, New Zealand, Norway, Portugal, San Marino, Singapore, Slovakia, Slovenia, South Korea, Spain, Sweden, Switzerland, Taiwan, United Kingdom, United States (Click image for larger image

By this definition of "advanced" US kids are 23rd out of 32 countries, or the 28th percentile.

It's true that "near the bottom" is too grim an assessment. But I can't see a way to put the 2009 PISA data together such that American kids are scoring about average.

I think of two very broad education reform camps. One calls into question the basic arrangement of institutions involved in U.S. education, arguing that the contradictory priorities in the system almost guarantee mediocrity. The solution, therefore, cannot be a nibbling around the edges of reform, but wholesale change: for some reformers, that means a market solution with greater parental choice, often coupled with more stringent human resources policies. For others the solution is a complete change—via technology—of the way we think of “learning.”

The second group of reformers argues that the system of education institutions is mostly fine, and that factors external to the system are responsible for our woes (which are, in any case, exaggerated). Some point to social and economic factors, others to the incoherence in curriculum (cf the Common Core), and others to the very reform measures (especially standardized tests used to evaluate schools and teachers) instituted by the other group of reformers.

In his new book Improbable Scholars,  UC Berkeley professor of education David Kirp offers an unusually readable account of what improved schooling would look like if you’re in the second camp. His explicit mission: to show that educational excellence is possible with the system as it exists now, even in districts that face enormous challenges.  He makes a fair case, given the limitations of the method he employs.

Improbable Scholars follows in the tradition of numerous education books by recounting time that the author spent in a school or district. Kirp tells the story of Union City, NJ, a city like so many others in the US: it has a great manufacturing past (“Embroidery Capital of the United States”) but was unable to find a new economic identity when cheap imports undermined its industries. Now most of its residents live in poverty, and a large percentage are recent immigrants who speak little English.

But Union City schools are unlike most districts with this profile. Despite the demographics, Union City students score about average on state tests. Ninety percent graduate high school, and sixty percent go on to college.

How they do it is Kirp’s subject, and in one sense this book has the feel of many others.  The account is told through stories. We meet Alina Bossbaly, a local legend of a third-grade teacher who is able to connect even with the most difficult children, and to make them feel a part of the classroom community, a process that has come to be known as “Bossbaly-izing” children.

We meet long-time Union City Mayor Brian Stack, strong supporter of education, savvy politico in a tough political town, and point man in the procurement of funding for the new 180 million dollar high school.  

Kirp is an academic, not a journalist, so although he’s an able writer, you’re not in the hands of a professional storyteller or fact-finder. But what you get from Kirp is a deeper analysis, a better-than-even tradeoff in this case.

So what is Kirp’s conclusion? He offers a list of key factors that he says must be in place for a district to thrive:

• District leaders put the needs of students ahead of those of staff
• They invest in quality preschool
• They insist that a rigorous curriculum is consistently implemented
• They make extensive use of data to diagnose problems
• The engender a culture of respect among the staff
• They value stability and avoid drama—they make a plan and stick with it for the long haul
• They never stop planning and reviewing the results of their plans.

When a district posts a remarkable record, it’s natural to ask “how did they do it?” The obvious problem is you’re looking at a single district. Maybe the real key to Union City is the Mayor. Maybe it’s the fact that many of the students come from countries with a tradition of respect for authority.

Kirp makes a case that other unusually successful districts have the same set of factors in common. It’s no substitute for a quantitative analysis, but KIrp at least shows that he’s aware of the problem.

And to be clear, I read the book in this wise, as something like an ethnographic study. Books like this offer detail and texture that larger scale, more rigorous analyses lack. In so doing, they ought to be inspiration to more quantitatively oriented researchers for what they are missing and where to turn their sights next. 

When it comes to criticizing methods he thinks are ineffective, Kirp is less sure-footed. He dismisses the notion that the relationship between school funding and student achievement is uncertain by noting that such suggestions leave administrators “shaking their heads.” There is an extensive and complex literature on the impact of funding, and the proper conclusion is by no means as simple Kirp would like us to believe.

Likewise, I’m rankled by Kirp’s assertion that “If you’re a teacher or principal whose job is on the line and your ordered to accomplish what seems unattainable, cheating is a predictable response.” This sounds an awful lot like a tacit pass to cheating educators.

The section of Improbable Scholars devoted to “what doesn’t work” left a bad taste in my mouth because it comes at he end of the book, but it is a mere five pages long.

If you’re curious about one vision of successful education that more or less maintains the status quo and actually gets into some detail, Improbable Scholars is a good choice.

What would you say if a major corporation took out a full-page ad in the New York Times to advertise a message that you thought was important and mainly agreed with, only to find that the text of the message was rife with misspellings, grammatical errors, and misused words?

That's the feeling that I get from the new video all over my Facebook feed (and with over 7 million views in 4 days) titled "Dove Real Beauty Sketches." If you haven't seen it, here you go. (I summarize it below.)

In brief, a woman describes herself to a forensic sketch artist, who cannot see her. He draws her portrait based only on her description, and then draws her again based on the description of a stranger who just met her. The woman then sees both portraits and recognizes that she has been rather hard on herself in her self description. (The process is shown for several women.)

The associated website calls this "a social experiment." But it's a terrible example of experimentation.

We are invited to draw the conclusion that women see themselves as less attractive than others do. I don't know the self-perception literature well, but I'm pretty sure this conclusion is right. But this experiment is a terrible way to illustrate that.

• The artist should be blind to condition. He knows when he's basing the drawing on the description of the subject vs. the stranger, and so could unconsciously bias the result
• The descriptions are not based on perception, they are based on memory. If you want to claim that it's about how women see themselves, not how they remember themselves, then each person should do their best to describe the woman based on the same photograph
• A the end the sketch artist tells each woman the source of each sketch. What would have happened if he had asked her to say which looks more like her, and to say which she thought was based on her description? If women's perception is really distorted, then the woman should see sketch based on her description as being more like her. An alternative hypothesis is that women more or less know what they look like, but talk about themselves in negative terms.
  
• The foregoing point raises another issue: social conformity. If the result is not due to perception but to people conforming to social norms, the difference in the sketches might be due to the women's reluctance to seem vain in their self-descriptions, and to the stranger feeling that he or she ought to describe the woman nicely.
  

How important are these criticisms to the overall message of the video? Not very. The point of the video is that women shouldn't be so hard on themselves in judging their looks. It's a good message.

That's why I draw the analogy to grammar, punctuation, and spelling in a written message. If Dove had published a print ad full of grammatical and spelling errors, I expect someone would have called them out on it.

Dove presents this as an experiment, but it's a terrible experiment. It would not have been hard to do a video making the same point with a better experiment. Any graduate student of social psychology could have improved this ten-fold.

I would have given the video 9/10 (subtracting one point for scientific sloppiness) if not for the statement made here in the video:

Well, I'd prefer a different message. Rather than "It couldn't be more critical to your happiness" and "be grateful for your natural beauty" I'd prefer a message amounting to "what you look like matters less than you think."

But I can't expect everything from a company selling beauty products. 8 out of 10, Dove.

A great deal has been written about the impact of retrieval practice on memory. That's because the effect is sizable, it has been replicated many times (Agarwal, Bain & Chamberlain, 2012) and it seems to lead not just to better memory but deeper memory that supports transfer (e.g., McDaniel et al, 2013; Rohrer et al, 2010). 

("Retrieval practice" is less catchy than the initial name--testing effect. It was renamed both to emphasize that it doesn't matter whether you try to remember for the sake of a test or some other reason and because "testing effect" led some observers to throw up their hands and say "do we really need more tests?")

Now researchers (Szpunar, Khan, & Schacter, 2013) have reported testing as a potentially powerful ally in online learning. College students frequently report difficulty in maintaining attention during lectures, and that problem seems to be exacerbated when the lecture occurs on video.

In this experiment subjects were asked to learn from a 21 minute video lecture on statistics. They were also told that the lecture would be divided in 4 parts, separated by a break. During the break they would perform math problems for a minute, and then would either do more math problems for two more minutes ("untested group"), they would be quizzed for two minutes on the material they had just learned ("tested group"), or they would review by seeing questions with the answers provided ("restudy group.")

Subjects were told that whether or not they were quizzed would be randomly determined for each segment; in fact, the same thing happened for an individual subject after each segment except that each was tested after the fourth segment.

So note that all subjects had reason to think that they might be tested at any time.

There were a few interesting findings. First, tested students took more notes than other students, and reported that their minds wandered less during the lecture.

The reduction in mind-wandering and/or increase in note-taking paid off--the tested subjects outperformed the restudy and the untested subjects when they were quizzed on the fourth, final segment.

The researchers added another clever measure. There was a final test on all the material, and they asked subjects how anxious they felt about it. Perhaps the frequent testing made learning rather nerve wracking. In fact, the opposite result was observed: tested students were less anxious about the final test. (And in fact performed better: tested = 90%, restudy = 76%, nontested = 68%).

We shouldn't get out in front of this result. This was just a 21 minute lecture, and it's possible that the benefit to attention of testing will wash out under conditions that more closely resemble an on-line course (i.e., longer lectures delivered a few time each week.) Still, it's a promising start of an answer to a difficult problem.

References

Agarwal, P. K., Bain, P. M., & Chamberlain, R. W. (2012). The value of applied research: Retrieval practice improves classroom learning and recommendations from a teacher, a principal, and a scientist. Educational Psychology Review, 24,  437-448.

McDaniel, M. A., Thomas, R. C., Agarwal, P. K., McDermott, K. B., & Roediger, H. L. (2013). Quizzing in middle-school science: Successful transfer performance on classroom exams. Applied Cognitive Psychology. Published online Feb. 25

Rohrer, D., Taylor, K., & Sholar, B. (2010). Tests enhance the transfer of learning. Journal of Experimental Psychology. Learning, Memory, and Cognition, 36, 233-239.

Szpunar, K. K., Khan, N. &, & Schacter, D. L. (2013). Interpolated memory tests reduce mind wandering and improve learning of online lectures. Proceedings of the National Academy of Sciences, published online April 1, 2013 doi:10.1073/pnas.122176411