Saturday, June 20, 2015

Voter turnout


You told me you are a would-be electoral analyst. Then, where's your version of the road map fellow traveler? For a nation with a voting experience that could be counted by fingers on one hand, ye budding electoral analysts may have to burn a good deal of midnight oil to try to start at a galloping pace. Too bad!


Now, you may want to start with something modest. How about voter turnout?


How large was the number of eligible voters that actually come out to vote, the voter turnout, usually shown as a percentage, may indicate the level of health of a democracy. This must have been mostly correct when all democracies were young and starting to take root all over the world. But has the democracy turn sour for some bulwarks of democracy of our world?


The editorial on November 11 of 2014 of International New York Times claimed "The Worst Voter Turnout in 78 Years" for U.S.


The abysmally low turnout in last week’s midterm elections — the lowest in more than seven decades — was bad for Democrats, but it was even worse for democracy. In 43 states, less than half the eligible population bothered to vote, and no state broke 60 percent.


In the three largest states — California, Texas and New York — less than a third of the eligible population voted. New York’s turnout was a shameful 28.8 percent, the fourth-lowest in the country, despite three statewide races (including the governor) and 27 House races.


Over all, the national turnout was 36.3 percent; only the 1942 federal election had a lower participation rate at 33.9 percent.


Surprising! How could that be, you may wonder. And there are lessons to be taken to heart, especially for a fledging democracy.


... The reasons are apathy, anger and frustration at the relentlessly negative tone of the campaigns.


To me that sounds like a confirmation that voters are no fools! Some of us may be uneducated or lack self-confidence or lack the lingo to argue, but we are no fools.


Republicans ran a single-theme campaign of pure opposition to President Obama, and Democrats were too afraid of the backlash to put forward plans to revive the economy or to point out significant achievements of the last six years. Neither party gave voters an affirmative reason to show up at the polls.


Here, we don't have enough information to say if affirmative reasons are lacking or if our people, at least a sizable portion of them, are disinterested or downright ignorant. Nevertheless, we need not be overly worried whether we are going to be the last in the turnout race in the ASEAN as the following graph shows:


Yet we'll need to get twenty percent or more of our eligible people to vote in order to be in the league of the highest voter turnout achievers in our region.


For this graph I have summed up all the counts of votes for all parliamentary elections reported and divided it by the sum of reported registered voters to get the average vt. Similarly for average vapt. In between, care is needed in interpreting the International IDEA data, especially with the voting age population reported. Its Code book for Voter Turnout data, 2014, explained about voting age population:


Voting age population refers to the total number of potential voters of voting age in a given country. The VAP figure is a rough estimate and it does not take into account the fact that there might be people who are above voting age but still not enfranchised due to legal or systemic barriers. The estimate gives an approximate figure of the number of eligible people but it does not reflect the exact number. VAP might differ largely from the number of registered people since registration in many countries is voluntary or not accurate.


In comparison to the ASEAN voter turnout, here's the voter turnout for selected developed countries:


In its explanation of why a voter turnout website is needed, IDEA explained that participation, whether through the institutions of civil society, political parties, or the act of voting, is always seen as an essential prerequisite of any stable democracy. Yet, it is not so simple in real life:


Beyond the widespread belief that participation in political life is a positive thing, there is little agreement on what constitutes a good or democratic level of voter turnout. Does high turnout indicate great enthusiasm for the political process or reflect compulsion, sometimes subtle, other times overt, that a government places on its citizens to vote? Does low turnout indicate a weak political system, or merely reflect a widespread contentment among the people with the system as it is? If the statistics that International IDEA has gathered indicate anything, it is that high voter turnout does not necessarily mean that a multi-party democracy is stable.


What is interesting here is that a low turnout may be reflecting a support for the current political system, or else a weak democracy! Obviously, our interpretations have to go beyond mere numbers.


Wikipedia gives considerable details on the complexities of voter turnout, its implications, and causes of low or high turnouts in its voter turnout page among which is a model for the voting decision of an individual shown below.


  • P  the probability that an individual's vote will affect the outcome of an election,
  • B  the perceived benefit that would be received if that person's favored political party or candidate were elected,
  • D  represents any social or personal gratification an individual gets from voting, and
  • C  the time, effort, and financial cost involved in voting, is


PB + D > C.


It just says that the individual will weigh his/her chance of influencing the outcome of the election together with the benefit likely to receive if the party/candidate were elected plus the perceived social/personal gratification against the cost involved to decide whether to vote or not.  And practically, "Since P is virtually zero in most elections, PB is also near zero, and D is thus the most important element in motivating people to vote."


D originally stood for Democracy or civic duty. The modern understanding of D is said to have five major forms of personal gratification: (i) complying with the social obligation to vote; (ii) affirming one's allegiance to the political system; (iii) affirming a partisan preference (also known as expressive voting, or voting for a candidate to express support, not to achieve any outcome); (iv) affirming one's importance to the political system; and, (v) for those who find politics interesting and entertaining, researching and making a decision.


In analyzing voter turnout, you may have to inspect all these potential factors that may exert influence (source: Wikipedia):


Socio-economic factors. More significant in low turnout countries. A person with better education is more likely to vote. Education is considered to be the key factor in developing an individual's habit for voting.  Income and class are also important. Ethnicity, race and gender may influence turnout in many nations, but considered to be not important in Western democracies today. Young people are less likely to vote than the elderly.


Hereditary factors. May be important according to some studies. But the idea is controversial.


Differences between elections. Different elections have different turnouts and the importance of an election may be the key factor in getting a higher turnout.


International differences. Tends to be lower in the United States, Asia and Latin America than most of Europe, Canada and Oceania. The differences between nations tend to be greater than those between classes, ethnic groups, or regions within nations. There are some surprising facts. Nations with better-educated populaces do not have higher turnouts. There are two main causes of these international differences—culture and institutions.


Cultural factors. Civic duty is the main factor. It takes time to develop and require such social conditions as:
  • trust in government;
  • degree of partisanship among the population;
  • interest in politics, and
  • belief in the efficacy of voting.


Institutional factors. Characteristics of any of these factors: (i) Voter registration, (ii) compulsory voting, (iii) salience, the perceived effect that an individual vote will have on how the country is run, (iv) proportionality, that is whether proportional representation or plurality system(simple plurality/first-past-the-post/single-choice voting), (v) ease of voting, and (vi) voter fatigue, could have significant effect on voter turnout.


Now that the preliminary voter list has been released for Myanmar's 2015 election, the first action you may have been thinking about is assessing its quality with your good intention of helping to improve it. I've no idea where you'll get the data, though. For me, I couldn't find any online resource for it by the Election Commission as yet. All the same I bet you've already pored over the publication Voter List Pilot Project: Findings and Recommendations by People's Alliance for Credible Elections, Myanmar, of July 2014. If there are other relevant publications, do share it. And keep up the good work friend!



Tuesday, June 16, 2015

Two identical triangles, but one with hole


I couldn't believe my eyes. Couldn't see how's that possible. Thought for a while. Gave up and constructed them with Excel:



What I saw was this post from Quora Digest, one of 16 answers under: What are some iconic mathematical images?

In one of my posts I recalled seeing a picture of the famous Russian mathematician Andrei Kolmogorov teaching primary class in a Time-Life paperback I once had. Below is the picture appearing at the top of the Quora answer: Pál ErdÅ‘s explaining a problem involving some sequence of integers and a squarefree number n to a 10 year-old Terence Tao at the University of Adelaide in 1985. Couldn't help being nostalgic imagining myself in the boy's place.


















Well, some are luckier than others. But not Ramanujan. And he didn't need any tutoring in math; but badly needed advice on how to take care of his health!




















I had always loved this picture of the prince of the autodidacts and genius. But only now could I see he's doing math in his head while sitting for the photo! Thanks Quora User who posted it.


Then I was thrilled to see a sample of what Everiste Galois has scribbled on the night before his fateful duel the next day. He was my mathematical hero, not because I know his great contributions: group theory and Galois theory, but because he was a brilliant mathematician who died so young and foolishly at 20.






































Quora, the source for these posts was co-founded by two former Facebook employees, Adam D'Angelo and Charlie Cheever. It's a nice question-answer site. I've subscribed to the Quora Digest, and regularly received the digest through email. Many people, including my son has half-forgotten the email (old fashioned?), since they are happy working with social media. So it's refreshing to receive social media through email; a little twist.

Friday, June 12, 2015

Econometrics for the Masses, Blind Boy, and Courage


It must be in the early eighties. I was out of work and trying to find luck offering my help to some very qualified and energetic public health professionals. In a seminar or a meeting of some sort on management information systems, I boldly (or foolishly, depending your point of view) presented a note on GIS for the masses. That was based on some free software which was quite an innovation that time. The expatriate GIS professional who presided surely must have been annoyed though she went on unhindered to develop the system with her pet commercial GIS. Things have changed. Now the die hard supporters of commercial GIS software may have nothing better than to (reluctantly) acknowledge the power of free and open-source GIS software like Grass or QGIS.

So, if not by expertise, then by motivation and by intent I should be allowed to speak about econometrics for the masses. Well, I have tried the like of it before with my posts "An Unclaimed CD on Psychometrics with R or Intro to Anything with R" and "Blind leading the 20/20".

Because, we are woefully like the blind boy I remember from my school-boy days:

"O SAY what is that thing call'd Light,

  Which I must ne'er enjoy;

What are the blessings of the sight,

  O tell your poor blind boy!"


Any light could have been light enough and you wouldn't expect me to have much expertise to write such as those! And I'm feeling like Courage the Cowardly Dog.


Though, with a bit of soul searching, I realized mine were no better than the idle tales of my wanderings after trying to find my way in unfamiliar places at dusk. Anything goes. With that, I owe an apology to a friend of a friend—who is working in Naypyitaw and who happened to be looking for some econometrics software—if he were to find this post too shallow. I got the inspiration for this post from him.

Econometrics is "the unified study of economic models, mathematical statistics, and economic data. ... Econometric theory concerns the development of tools and methods, and the study of the properties of econometric methods. Applied econometrics is a term describing the development of quantitative economic models and the application of econometric methods to these models using economic data." (Econometrics, Hanson, 2000, revised 2015). In plain English it is "The Economist's Approach to Statistical Analysis" as explained in Econometrics for Dummies.

Groping through search results on the web for econometrics, I noticed a lot of introductory material mentioning Stata as their software for making econometric analyses. The reason is the same as when people casually talk about R vs. SPSS, R vs. SAS, or R vs. Stata. They have in mind the way these other software allow making statistical analysis by using drop-down menus, while R basically does not. Ease of use vs. steep learning curve, if you like.

Karem Tuzcuoglo, a PhD candidate in economics at Columbia explains:

"One-Click" Programs ((almost) no coding required, results obtained by one click)
STATA: Most of the econ undergrad programs use STATA. It is the best program (even at the PhD level) if you want to estimate panel data (i.e., where the data hava both cross sectional and time series dimension. Typical examples are surveys and international trade data sets).
Eviews: Less famous than Stata, but provides much better time series analysis. If you don't want to do time series forget about Eviews.
SPSS: I don't have much information about it. But I can tell that it's not widely used.

"Semi-Coding" Programs
SAS: It used to be a big deal 10-20 years ago. Right now not as famous as before - though there are some companies that still strictly prefer using SAS.
R: Maybe the most popular program nowadays. First of all it's free! R network and R packages (pre-written algorithms by others) are getting larger and larger. Actually, R can be listed in the next section as well because one can definitely code everything in R. However, the fact that there are so many ready-to-use packages in R makes it also Semi-Coding program if one wants to.

"Pure-Coding" Programs
MATLAB: The most famous program among (high level) econometricians. Many applied economics have been done by Matlab. A lot of researchers put their Matlab codes online. It has a good Econometrics package - one still needs to code though.
PYTHON: It's more powerful and faster than Matlab. However, it's a very new language; it's still developing. 
C++: If one wants to do hardcore coding, then C++ is the ultimate program. It's extremely fast in terms of computation (once, my simulation took 25 hours in Matlab, whereas C++ ran the code in 3-4 hours).
FORTRAN: Professors above 55+ age will know this program. It's (almost) not used anymore - though we should show some respect to the Father of Coding Programs!

BONUS: There are several other programming languages of course. If you are in UK (especially in Oxford), you will end up using a program called Ox., which is an optimized program for matrix algebra and, thus, for econometrics.Gretl is an extremely easy to use - but less to offer- program. 

Among all of the options, I would suggest you to learn R regardless of whether you want to work in academia or in industry (more and more companies begin using R by the way). But if you want to stay away from coding, then go for STATA.
    
A good collection of resources is the special volume on Econometrics in R, vol. 27, July 2008, by the Journal of Statistical Software, which is available for download. As usual, the first site to consult relating to R packages would be the CRAN tasks views, and in this case it is the Econometrics Task View available on any CRAN mirror site. Here the description of packages are organized into eight topics: (i) Basic linear regression, (ii) Microeconometrics, (iii) Instrumental variables, (iv) Panel data models, (v) Further regression models, (vi) Time series data and models, (vii) Data sets, and (viii) Miscellaneous. As of now, there are a total of 117 R packages on this task view.

At the end of the econometric task view page, there is a list of useful links. Among them the link for (i) A Brief Guide to R for Beginners in Econometrics took me to the University of Stockholm website and I was lost. Looking for free lunch I followed the link, (ii) Book: Applied Econometrics with R (Kleiber and Zeileis) and found that I've to buy the book. I was so frustrated, I missed the offer for free downloads of some pages of the book at the bottom of the page. Then, (iii) Book: Hands-On Intermediate Econometrics Using R (Vinod) another book you have to buy, allows you to download the foreword and chapter-1.
Luckily and thanks to Dave Giles' blog Econometrics Beat, I was led to the downloadable link to An Introduction to Programming Econometrics With R by Bruno Rodrigues of the University of Strasbourg.

Again I was lucky to find on Giles' post the comment by Achim Zeileis, one of the authors of Applied Econometrics with R (AER) pointing out the free resources accompanying their book: "The book itself is not free (published by Springer) but the first two chapters can be downloaded for free from the publisher and we have free presentation slides/scripts for the whole book plus some extras. Everything is collected online at: http://eeecon.uibk.ac.at/~zeileis/teaching/AER/".
I was aware of, for some time, that one staple reference for intro to econometrics with R has been the Econometrics with R by Grant Farnsworth of 2008.

Finally, I managed to find the A Brief Guide to R for Beginners in Econometrics by Mahmood Arai of Stockholm University at http://homes.chass.utoronto.ca/~jmaheu/4050/R_intro.pdf.

At last, relieved of my self-proclaimed duty to say something about Econometrics for the Masses ..., I could go back to enjoy the Blind Boy. Ever a blind boy, we could still be defiant. The last stanza reads:

"Then let not what I cannot have

  My cheer of mind destroy;

Whilst thus I sing, I am a king,

  Although a poor blind boy."


And yet, while at the bottom of our heart we knew where we rightfully belong, we shouldn't stop groping for light even when we are in total darkness, shouldn't we? For me, I would love to have this third stanza in place of the last:

"My day or night myself I make

  Whene'er I sleep or play.

And could I ever keep awake

  With me 'twere always day."