Comments on economics, mystery fiction, drama, and art.

Wednesday, July 31, 2013

MOOCs and Me

Here's a link to a short essay on my first experience with a massive, open, on-line course (a MOOC).

Monday, July 15, 2013

Reflections on the Trayvon Martin Case

This has next to nothing to do with what I usually write about here, but I wanted to get down what I have been thinking about the events and the trial and the aftermath.

Let me say, first of all, that had I been on the jury, I would have voted for conviction.  But that I would almost certainly never have made it onto the jury had I been a potential member of it.  Had I been called as a potential juror, and answered questions honestly during voir dire, I would have been excused.  Because I would have said I was unable to approach the case impartially--I believed then and I believe now that George Zimmerman is guilty of unlawfully taking the life of another person.

But I understand the verdict, intellectually, and, I guess, legally.  Florida's incarnation of self-defense ("stand your ground") says, among other things:
A person who is not engaged in an unlawful activity and who is attacked in any other place where he or she has a right to be has no duty to retreat and has the right to stand his or her ground and meet force with force, including deadly force if he or she reasonably believes it is necessary to do so to prevent death or great bodily harm to himself or herself or another or to prevent the commission of a forcible felony.
In this case, the only "eyewitnesses" were Martin and Zimmerman--and only one of them is around to describe the events.  And Zimmerman chose (wisely, I think, from the point of view of his defense) not to testify, so all that is available is what he told the police and to which they testified.  (Had Zimmerman testified, the prosecution, on cross-examination, could have asked, for example, why he chose to disregard the instructions/suggestions of the police operator not to follow Martin, not to confront Martin, to wait for the police to arrive.)

Replicate these events.  Two people, no other eyewitnesses.  One is dead, the other has some superficial wounds, but claims to have been attacked and responded with what he believed to be necessary (and lethal) force.  But suppose our survivor in fact provoked the confrontation, knowing they were unobserved.  What we have is a legal justification for what is, in fact, murder.

I do not believe that Trayvon Martin initiated the encounter.  I believe Zimmerman did.  I have no evidence of this, of course, because I was not there.  But it's a belief that is consistent with what is reported of Zimmerman's activities (including the call he made to the police).  I don't believe that Zimmerman went out that night intending to kill someone.  I do believe he was willing to interpret innocent behavior as somethng else, based on the appearance (apparel, ethnicity, etc.) of someone he observed doing something he thought was strange.  I believe that instead of waiting for the police, he provoked an encounter, even that he may have been coming off the worse for that encounter, and responded by killing an innocent person.

And doing so in a way which made it all but impossible for him to be convicted.

Update, 16 July 2013:  And read this:

Monday, July 08, 2013

An Assessment of a MOOC

I'm a member of Indiana University's "teaching academy" (the Faculty Colloquium on Excellence in Teaching), and I was asked to do a presentation on MOOCs at this year's gathering in May.  As part of the background for that, I enrolled in a MOOC—“Generating the Wealth of Nations”—through Coursera, taught by an economist at the University of Melbourne.  (My intention had been to avoid economics courses, as I did not want to be constantly critiquing the choice of material and approach; however, the availability of courses in the time frame I needed led me to this one.)  The course has finished, and here are some data from it, from a end-of-course email from the instructor:

“As we reach the end of the course, I thought you might be interested to know some statistics on participation and outcomes.  At the end of last week, out of the 28,922 enrollments in Generating the Wealth of Nations, 12,197 people had actually participated in the course at some stage, and in the last week 1,935 had been active.  There had been 164,946 viewings of individual lecture videos.  On the Discussion Forum there were 423 threads, 2370 posts and 1413 comments.  About 700 of you completed the first assessment (average grade was 2.86), and about 500 the second assessment (average grade 3.09).”

NOTE:  To complete the course, a participant had to complete all three assessments with a “satisfactory” rating (2 or 3 on a 1—low—to 3—high—scale).  Assuming that all of the people who completed the second assessment also completed the first one, and that everyone who completed the first two assessments also completed the third, we have about 500 completions out of (take your pick) 28,922 initial registrants of 12,197 actual participants.  That’s somewhere between 1.7% and 4.1% (and about 25% of those active during the last week). 

Beyond that.  The course was basically a talking-head-over-Power-Point-slides.  Of its type, the lectures were well-prepared, well-sourced and documented, and the slides were generally good.  The course had an extensive set of suggested readings (I downloaded all of the suggested readings that were available on-line—108 of them—and there were many more that were not available, and actually read about half of them).  The assessment protocol was to complete three written assignments, which were peer-assessed and "graded" on a 1 (low) to 3 (high) scale.  To receive a certificate of completion, you had to complete all three assessments with a score of 2 or 3.  While I was (very) skeptical of the peer assessment component of the course, my experience with is was at least satisfactory.  The (anonymous) participants who scored my (anonymous to them) work seemed to have read them carefully, applied the rubrics provided reasonably well, and even made some cogent comments.*

I should say something about the discussion forums.  They were not moderated, although they were monitored (occasionally either the instructor or one of the course assistants would make an appearance).  I did not read all 2,370 posts, but I did participate fairly actively.  In general, although a high percentage of the posts (about half, maybe more) were not based on any evidence presented by the poster, but rather on unsupported assertion and opinion, the tone was generally civil, and several of the posters were well-informed and making an effort to provide information and interpretation.  I thought that, under the circumstances, this was a creditable performance by the participants.

Overall, for a free, non-credit course, I would judge my experience to have been generally positive.
*An example of one of the assessments I wrote:
1. a. Data collected and ratio calculated                 3
b. Description of whether/how it changed         3
2. a. Data on 3 indicators
b. Reasonable explanation of why these                       3
Reasonable explanation of which is most valuable      3
3. a. Is the description of the change in GDP relative to US
accurate                                                                                                3
b. Is it interpreted using catch-up/convergence/structural change?     3
Is the information to make this assessment correctly described?        3
c. If there is not an increase, is there an understanding of
convergence and structural change?                                                    3
Are barriers plausible and is evidence of one of these
barriers presented?                                                                               3
4. a. Is there an understanding and application of institutions/
culture/geography/contact with success (only 1 needed)?                  3
b. Evidence to support the argument                                                   3
Nice presentation of the relative real GDP per capita data.
I’m not necessarily sure I’d describe Italy’s income distribution as “more just;” clearly, “more equal” is an accurate, and less value-loaded term.  Before we can evaluate the meaning of this, we need to have a little more on why (and to what extent) a more equal distribution is better.  The life expectancy data is important and nicely interpreted.  Also, the data on expenditure on innovation seems quite important.  It might be useful to provide a little more information about what that measure means, though.
Is Graph A.1 Italian RGDP pc relative to the US or something else?  Could part of the immediate post-WW2 growth be catch-up?  Good mention of the Marshall Plan.  The shift out of agriculture is also important, so it would be nice to have seen the extent of it.
The importance of the EEC is nice to see mentioned.  How about the Euro, though?  One way to think about the impact of the EEC would be to look at how Italian trade with the other members of the EEC changed in the post-WW2 period.  Were exports to and  imports from the EEC nations growing?  In what other ways did the EEC lead to increased integration between Italy and the rest of Europe?
Overall: 3

Friday, July 05, 2013

Generating the Wealth of Nations 36: Population Density and Economic Growth

For a couple of the things we did during the course, and that I looked at because I was interested, I was looking at data that seemed to suggest that countries with larger initial population densities had higher subsequent growth rates.  There are a number of good reasons to suspect that this might be true; in countries with higher densities, the costs of providing some publich services (like education and health care) and the costs of building and maintaining a transportation system, and the costs of recruiting workers, will all probably be lower.  So I decided to look at it systemattically.  I took all the countries in the Madison Project for which I had both 1950 and 2007 (I stopped before the Great Global Recession) real GDP per captia data and for which I could find both 1950 and 2007 population data and area data.  I then calculated the population density in 1950 and the total percentage growth in real GDP per capita from 1950 to 2007.  The I plotted the data, and I got this:

(Click to enlarge.)

The red line is a linear trend line (as calculated by Excel), which suggests that countries with higher population densities in 1950 also grew faster.  (The correlation coefficient between density and growth is 30.5%, and it's significant at the 1% level, for those of you deeply into statistics.)  Now this does not prove that density causes growth, but it's surely suggestive that it might.

(This is probably my last post related to the Generating the Wealth of Nations course; I hope some of this has been of interest.)