By Steve Kelman
I needed to vote early this morning, because I’m teaching and then immediately leaving for Washington. I arrived at my small polling station in Concord, Mass., at 6:45 am, fifteen minutes before the polls opened, and there were already about 20 people in line ahead of me. I was able to vote quickly – done by 7:10 a.m. – and when I left the polling place, there were maybe 40 people waiting in line. (My younger daughter emailed me a little later in the morning that she had been waiting in line for an hour.) However, this is of course modest compared with the stories of people waiting 6 to 8 hours in line to vote early in Florida, for example. And I have already seen Facebook posts this morning about people waiting an hour in line or more.
The lines to vote are an interesting commentary – and critique – of two common and influential ideas from social science research and theory.
One is the idea of free riding. There is actually an academic literature about why it is irrational for any individual to vote. It is highly unlikely that an individual vote will determine the outcome of an election – indeed, even in the famous 2000 Florida case, there were several hundred votes separating the candidates. So, it has been noted, no individual should vote, but instead count on everyone else to vote. The paradox, of course, is that if everyone thinks this way, nobody will vote at all. We do know that people do vote, so the challenge for social science theorists has been to think about why people vote even though, from a self-interest perspective – voting has costs, especially with these long lines, and there is only an infinitesimal chance that an individual’s vote will elect their favored candidate.
The answer has to be some version of people feeling a moral obligation to vote and people feeling proud as individuals to participate in democracy. There is a larger point here. The theory that people will not vote because the self-interested costs outweigh the self-interested benefits reflects an impoverished view of what makes people tick – ignoring the role of moral obligations and attachments to a larger community. So these lines on election day actually tell us something about who we are as human beings.
The second idea that is dealt something of a blow by these voting lines is the psychological idea of habituation – put simply, the idea that everything gets old after a while. On this view, it is one thing when people in new democracies wait in line for hours to vote (I remember the footage of long lines when South Africa had its first universal suffrage election in the nineties), but after a while it loses novelty and hence attractiveness. Needless to say, Americans have been at democracy for a long time, so habituation certainly has had time to settle in. But there may be certain things we value on an ongoing basis, such that we continue to appreciate them even after a long time, sort of like a long-standing love as opposed to a first blush of attraction.
And this also may be something we want to pass on to our children: I overheard at the train station on the way to work a woman telling other commuters about how there was a first-time voter in line while she was waiting, and everybody gave her a big cheer.
Having said all this, it is of course also true that many Americans do not vote, that many feel cynical and jaded, that many are both self-interested and habituated. Today, though, let’s concentrate on those exceptions to social science theory.
Posted on Nov 06, 2012 at 10:45 AM4 comments
I have expanded this year the material on managing teams for my introductory management course for Kennedy School master’s students. I introduced two classes on team management two years ago, expanded it last year to three and now have raised it to five.
This is partly because the subject has a mixture of good available in-class experiential exercises and a rich scholarly literature, a nice mixture for a course. But it is mainly because teams, often cross-functional teams, are becoming more and more important for making increasingly complex organizational decisions or producing organizational products and services.
We started the five classes with an in-class exercise called Desert Survival, set up by a DVD describing a group of travelers whose light plane crashes in the Sonora Desert in Arizona and who must figure out the relative importance of various of the items they have available at the crash site for surviving in the desert. Students first solve the problem individuals and are then divided into teams of five to make the rankings as a group. Individual student and team answers are then compared with the prioritization provided by an expert on surviving in the desert.
It’s a rich exercise, but let me give four headlines.
In every one of the teams in my class, the teams did better as a group solving the problem than the average individual on the team did. Those who believe that “a camel is a horse designed by a committee” are, at least sometimes, incorrect: most individuals in the class did worse on their own than as part of a group.
Second, when I had students answer questions individually, the average of their scores was lower than the same students answering the questions in groups. So teams beat out the “wisdom of crowds,” which has been getting a lot of attention recently – the deliberative process of the team produced better results than mere averaging of the variety of individual results.
Third, greater initial opinion diversity among the individual team members produced better team results. (We measured this by plotting the standard deviation of scores of the individual members of each team – a measure of opinion diversity – against team results.) Getting access to diverse opinions and knowledge is the central reason to make decisions as a team, and, consistent with this prediction, greater diversity of opinion improved team results.
Finally, while 100 percent of my student teams did better than the average individual in the team, this is true for a still large but noticeably smaller 81 percent of student classes that have done this exercise in the past. Since this exercise has been done almost exclusively at business schools, one way of thinking about these results – something I find appealing, though these results hardly prove it – is that public-service oriented students at the Kennedy School do better at a cooperative exercise such as working in a team than more competitive, individualistic students at business schools. Interesting if true.
Posted on Nov 01, 2012 at 10:20 AM1 comments
There are few words more praised in discussions about government than the word “accountability.” Everybody is for it, nobody thinks we have enough of it. Public discussions about using performance measurement in government are commonly expressed as efforts to increase accountability; for example, the school testing movement is typically referred to as “school accountability .”
I think we need to be clear, however, about a tension between the “accountability” that everyone rushes to support and the underlying effort to improve government performance, which is supposedly one – and presumably the most important – purpose of the exercise in the first place. A recent report by Matthew Weigelt on fcw.com, discussing a panel discussion on the topic of stretch goals at George Washington University, illustrates one aspect of the tension.
Fundamentally, the tension comes from the fact that the idea of accountability – where one is trying to perform well in order to please external actors – has a fundamentally punitive meaning. The popular phrase, “Where’s the accountability” is actually just a (slightly) nice way to say, “Who are we going to fire or send to jail?”
So agency managers naturally react in several ways when, say, performance measurement is associated with “accountability.”
First, they try to keep it as far away from themselves as they can – isolating “performance measurement” into a staff function consisting of writing reports for the Office of Management and Budget or the Government Accountability Office, not a line activity used by actual managers in charge of delivering the organization’s performance.
Second, feeling no ownership of the measures, they have no compunction to game the system, staving off the external accountability without actually improving performance. A number of years ago, for example, the IBM Center for the Business of Government published a report by William & Mary professor Richard Gilmour on how federal agencies succeeded in moving their old PART scores on performance management from “red” to “green.” The report was extraordinarily depressing because it outlined a series of ways this occurred – none of which included actually improving the program’s performance.
Weigelt’s article discusses a third tension in the accountability perspective. There is overwhelming research evidence that setting concrete goals improves an individual’s performance compared to just telling the individual to “do their best.” The evidence is also strong that, within reason, challenging “stretch goals” motivate better than less-challenging ones. But as the article noted: “Fears of congressional appropriators or an inspector general knocking at the door can persuade some federal managers to stay in their comfort zone, to set goals they’re sure they can reach.” He went on to quote Patricia McGinnis, a professor at the Trachtenberg School of Public Policy and Public Administration at the George Washington University: “They think somebody in Congress or somebody in [the Government Accountability Office] or somebody who has an oversight responsibility is going to punish them if they don’t meet a stretch goal.”
The problem, though, is that when people set easy goals because of fear of accountability, an opportunity for greater performance improvement through setting a stretch goal is lost.
Frankly, there is too much of a hallelujah chorus around the word “accountability” and not enough appreciation for a delicate balancing act needed around this concept, so as to avoid a situation where calls for accountability actually worsen government performance, rather than improving it.
Posted on Oct 24, 2012 at 10:06 AM3 comments