Thursday, November 27, 2014

Nan, Man, Yan



It doesn't rhyme perfectly as the title suggests. I mean Nan the Royal palace, then Mandalay, then Yangon and by that, Yan will strike a discordant note in your ears.

When I reread U Ponnya's poem Jasmine, the first line of the second stanza,
I see the Royal Palace, the palace grounds cut off by the moat, and the environs where city dwellers live and go about their business. And I get the sweet scent of jasmine from the royal garden spreading over onto the abodes of the commoners. I don't know if jasmine is tabooed on the other side of the moat. But by any reason, I won't think so, and sabe (jasmine) is not thazin (Bulbophyllum auricomum), the ethereal queen of flowers, poised on topmost branches of the tallest trees, and reserved for the Royalty. If the uncouth hands of the commoners were allowed to plant and tend their jasmines, the scent from them will mingle over the moat with those of the Royal jasmines and this is the sight and smell roused in me by U Ponnya's poem. And so I do not feel like seeing the last part of the first line cited simply as sabe getting plentiful because it is in season in the royal garden.


I haven't stayed any time long enough in Mandalay to get a bit familiar with its geography, especially the way its streets are referenced. Nevertheless, I heard that the pya (streets) there now are invaded by imported bare, fair thighs and emulated by some stocky, not-so-fair ones of locals that were used to be covered up by sarongs hung down over the feet, so as to give crafty display of their golden anklets.

Now that the palace has been rebuilt, I guess the public has access to the palace grounds which used to be garrisoned and closed to the public. I've no idea, how open or close this place is but I felt it should be as accessible as the pyramids of Egypt or the ruins of Chiang Mai in Thailand (disclaimer: never been to either one). At least I am glad.

In Yangon, we used to live in a place called Fa-Sa-Pa-La Sanpya kwet-thit or AFPFL (Anti-Fascist People's Freedom League) model quarter. Kwet-thit means new-quarter and places with that name usually were vacant land used by people without permission, I guess, most of them moving in from rural areas looking for better life. However, since our kwet-thit bears the adjective Sanpya meaning model, the government must have at least some involvement in it. Before this AFPFL quarter, I vaguely recall living in a place called Kya-gu kwet-thit or tiger's-crossing quarter, it was somewhere around the location of the previous Ministry of Construction on the Shwe Dagon Pagoda Road.

The Fa-Sa-Pa-La model quarter was bounded on the west by one of a few thoroughfares from north to the downtown Yangon passing over across the main rail-line. It was called the Signal-Pagoda Road and later as Alan-pya-paya Lan and I don't know if the name is still used. On the east was the Upper Phayre Street (now Ah-tet Pansodan Street). I am not sure about the roads bounding the quarter on the north and south. I think, on the south it was Victoria Avenue, and on the north it could be Zoo Road. The quarter was laid out neatly with one main road going from east to west through the middle and bisecting ten lanes running north to south. It happened that the whole quarter was burnt down, accidentally, or may be otherwise as it was the popular opinion in those days. I was about 11 or 12. Afterwards the government claimed the land and erected some public housing. After an intervening year or so after the fire, our family got settled in a public housing site in East Yangon partly as compensation. After 58 years there we have to move again, now temporarily. But that's another story.

When we were living in Fa-Sa-Pa-La, there on the north and across the road was the Botanic Garden of the Rangoon Horticultural Society. There was some building that housed the Society at the far end of the garden to the east and it was right on the Upper-Phayre Street. Now a hotel is there. The rest of the garden has now been occupied by commercial game centers and restaurants and the zoo adjacent to the garden on the far west part has taken up a large slice of the garden.

Well, I missed this nature park with different trees and wild life right in the middle of metro Yangon. I remember particularly the clumps of wamin (king bamboo) with dark green bulging internodes and stately golden bamboo clumps. The southern border was lined with thick walls of sein-na-ban bushes bearing yellow, white, orange, red, and pink flowers and purple berries. Close to the south-east corner, there was an avenue lined with kabwi (casuarinas), along which actor Kyaw Swe walked in the sad climax of the memorable movie "Mhway" based on the story by the writer Khin Hnin Yu.


As a boy, I knew vaguely that there were military buildings on the Signal Pagoda hill across the Signal Pagoda Road. I remember that this hill was commonly called gaw-ya kone, meaning the hill of English soldiers, I think. We used to have a piece of quite spacious and slightly raised land with shady trees, mostly kok-ko between the first lane of our quarter and the Signal Pagoda Road.  Right across this space and on their side of the road there was what I thought the World War II memorial pillar. Now it is no more. In those days they didn't have a fence or a wall on the side of the hill that bordered this road. So we can climb up the hill to reach the pagoda on top. I might have gone up once or twice and on one occasion I remember vividly tasting zee fruits (jujube) still green on the tree and it tasted really bitter. It was so calm up there and I saw no one. 

Now that the seat of Government has moved to Nay-pyi-taw and the military complex on Signal Pagoda hill has been vacated, people who know this place could have dreams of seeing this place reserved entirely for the public.

If it were the reality, then from the top of the hill you would enjoy the breathtaking view of Yangon river itself on the south; Hlaing river to the west as it flows into Yangon river together with the Twentay canal in the south; Pazundaung creek in the east meeting Bago river and flowing into Yangon river in the south-east from where they flow out to the sea; there in the distance you would see the Kyaik-khauk pagoda and Thanlyin Town. Towards the north you will get an unobstructed view of our great Shwe Dagon.

Imagine that there will be old people, young people, and children as you'd seen in Union Day celebrations; beautiful people, beautiful dressess, beautiful languages and accents those of which you were not used to seeing or hearing. There will also be our own urbanites as well as foreigners.

Let's think about trees, flowers, green grass and scents of sabe, gangaw, saga, khayay, and padauk. Anyone could sit and enjoy the views all around, forgetting time, forgetting everything.
                                                              
Now, if nice and beautiful hotels spring up all around the hill, and if you were allowed to enjoy this hill in some restricted space cordoned off by walls of the hotels, how would you feel?
For me, I would feel like walking along the narrow pavement between the walls of the Traders Hotel and the Bogyoke Aung Sun Street as I do now, at times doubting if I am living here on our land.


Tuesday, November 25, 2014

Meza, variations in a theme



In 1988 December I was lucky to pick up a hardcover publication of the Burma Research Society of 1960 entitled Fiftieth Anniversary Publication No. 2 which contained selections of articles from the Journal of the Burma Research Society (JBRS) from a sidewalk bookshop around Pansodan Street. Like many others in my small collection it was unread, though I must have skimmed it when it was freshly bought.

Some weeks before I was looking for information on ancient mathematics in Myanmar and looked into that publication again, though I knew I could expect very little because the inside-cover page said it is confined to (History and Literature). It contained many interesting articles, such as, the ancient Pyu, economic life of the early Burman, Shin Uttamagyaw and his Tawla, translations of Letwethondhara's Meza Ratu, among others.

So, I have a thought. Why not introduce the different versions of translation of the Meza ratu so that young people keen on learning English would be inspired to approach the ratu from both sides—Myanmar and English, and get something out of it, working only from the materials presented. A more ambitious hope is like expecting some of them to go to the source and go deeper, wider, and in the process got convinced that an independent institution like BRS should exist, and get something started, however modest.

At the very least, U Khin Zaw's article provided a rare chance to peek into the minds of the leading intellectuals of those days, including himself, as each tried individually to merge his poetic soul with the soul of Letwethondhara and bring forth the ratu in a language different from the poet's own. The results appeared collected in an article A New Translation of Let-We-Thon-Dhara's Famous Ratu by U Khin Zaw in Vol. XXI, part-iii, pages 129-36 of BJRS.

His article carried a sub-heading: With a critical study of the previous translation of Professor Pe Maung Tin, Dr. Ba Han and Mr. Luce. In his introductory note he summarized the different styles, and the approach he used:

"... a translation into 'rhythmic prose' by Dr. Ba Han. ...
Of Prof. Pe Maung Tin's translation it is to be noted that he aimed as he says at making them 'simple and literal ... without any attempt at literary effect.' Mr Luce's translations are attractive literary experiments. He is imitating in English the internal rhyming of the Burmese prosody.
    The work of translation presented some interesting problems ... I present the work of the earlier translators and add my own efforts with some comments especially on points where I differ from my predecessors."

In the following, I have taken only the third stanza of the ratu wholly and give the different versions of its translation, but give only the first few lines of each of the translations.

The third stanza in full in Myanmar language:



Prof. Pe Maung Tin
                                                                                             
How glorious is the Shwe-gu-taw where the Meza people bow their heads in supplication with faithful hearts on the occasion of the time-honoured banyan watering feast! The mountain vapour rising from the Meza Valley darkens the appearance of the whole mountain and gradually floats along, forming a link between one range and another, and the duskiness of it rises even to the sky.  ...

Dr. Ba Han

The residents of Meza cling to the hoary custom and hold the Nyaungye carnival. A motley crew in reverent awe and ardent faith pray before the Golden Cave that lends endless charms to the clouds kissed height rising from the heart of Meza Glen. The self-same mountain is shrouded in gloom and it exhalations wreathe themselves round every range. ...

Mr. G H Luce

At the Nyaung-Ye festival
Me-za people, all devout,
Dully fall in prayer and pour
Water to the Banyan tree,
From the valley, see the shrine
Rises even to the sky.
Meza! thy pagoda old
Of the Golden Cave is there,
Fair to look on evermore.
Dimly glimmering o'er the mount,
Steady hang night's vapours yet;
Soon, as a billowing net, are stirred,
Surging fretful as a flood,
Huddle upward, tower, and crowd;
...

U Khin Zaw

Obedient to the honoured custom's behest, Meza folk foregather for the Banyan-watering Festival and pray at the Shwe-gu-daw, and my eyes see that it is good. But when from Meza's chasms rise the congregated might of mists, shrouding summit, mountain, ridge and hill, annihilating sky, then falls the rainless rain ravelled mist drip-drop like rain. ...



Jasmine and Khayay buds




Two, three days ago, I was looking for a satirical verse by U Ponnya on the web. I think it was in his collection of Myittasa and he called it "A group of men despised by Lord Buddha". Instead, I found the simple and beautiful translation of his little gem "Jasmine" by Reverend
Friedrich V. Lustig, Buddhist Archbishop of Lavita at


Jasmine

When jasmine
Is not in bloom
In palace gardens
Courtiers must
Bedeck themselves
With Kan flowers

When jasmine
Is in full bloom
In Mandalay's
Palace gardens
Courtiers never
Adorn themselves
With Kan flowers

Then Kan flowers
Find favor only
Among villagers


I like to imagine we Myanmars, old and young alike will know and love this beautiful poem about khan, a rustic flower and a poor, fleeting substitute for the regal jasmine. If so, they will also love the poem by Min-thu-wun about khayay and roses. I have forgotten the name of the poem, but I have the lines in my heart, so I just put it "Rose" and tried my words. If they are passable, let them be my tribute to Bagyi Wun.


Rose


Way back from rice fields
I'd plucked khayay buds;
Given to her 'cause she'd love
Wearing 'em in her hair.


Towards the morning,
My most graceful, when I saw,
Roses in her hair,
Giving an aura o' status,
Oh, so fair.



And my tribute to Pho Ponnya:

Jasmine


'cause jasmine is not in season
In the royal garden, be pampered,
Courtiers console themselves
Wearing stringed khan flowers.


When Mandalay palace, environs,
Pervade with fragrance o' jasmine
Thence, khan flower slips;


Fleeting glory;
Henceforth 'tis rustic legacy;
'tis a flower, fancied by villagers.




Friday, November 21, 2014

Blunder of experts and numeracy of our bosses



I had for some time getting a "brainwave" to write a post like this. Ramanujan, the strange mathematical genius, said he got his ideas revealed by 'Lord Krishna" in his dreams. No one ever thought of questioning the source of his genius, because the pudding has been eaten and only after many generations later that his discoveries have been properly understood and rigorously proved. Still some remains and this is not surprising because when Ramanujan sent his theorems to England to Hardy, and others in the spring of 1913, Hardy, a well-known mathematician of his time, could only say: I had never seen anything in the least like them before. A single look at them is enough to show that they could only be written by a mathematician of the highest class. They must be true because, if they were not true, no one would have the imagination to invent them. However, one mathematician, M. J. M. Hill of University College London, commented that Ramanujan's papers were riddled with holes and said that although Ramanujan had a taste for mathematics, and some ability, he lacked the educational background and foundation needed to be accepted by mathematicians.  A monumental blunder indeed.

Anyway, my brainwave arose from more humble sources: tea-shops, gossips, passer-bys, books (old days), the Web (today), and what meager experience I have got.

There was an expatriate who came as the expert to supervise a survey project for one UN agency in Myanmar that has been contracted out to my friend. When fieldwork had been completed and the work reached the analysis stage, this expert said that using weights in analysis is unnecessary. However, my friend didn't take the expert's advice and did the analysis using the appropriate weights. Whereby, the expert did a few tabulations of the data herself, compared the results and remarked they were not really different.

My friend didn't take the expert's advice because the sample for the survey had been designed as a two-staged PPS (probability proportional to size) design and it is standard practice to use the design-weight and non-response weight in analysis. In my opinion the expert was either (i) being ignorant of survey data analysis, or (ii) knew so much of it to choose model based inference over the customary design based inference. If it be the latter case, I don't know how it could be done in that given situation, except perhaps for some selected aspects.

Some years before that, I had the opportunity to do a survey for the same agency for the same type of survey. The sample design was essentially the same as that of my friend's. I had no problem with the two experts assigned to the task of looking after the project. The fieldwork went well. At the editing stage, the preliminary analysis showed that: a significant number of persons still in school and no longer in school who has completed primary school were in the category not literate. It was implausible, so I was about to change them into the literate category when a senior field staff chanced to visit me and explained me the reasons that they were in fact not literate as reported. So I was saved from making a blunder. Sometimes, an "expert" needs to be helped by not so experts to set things right.

When I had submitted the draft report, at least four or five experts ranging from local to the regional office and their headquarters sent back their multipage comments. They were mostly on the wording in the report and on implications in interpretation. In one instant they mistakenly and hastily assume that I was talking about an interval estimate (where data point as well as confidence limits of an indicator is given) when I was in realty giving only a point estimate. I had to explain at great length that I was talking about the range of individual point estimates for different cross-classes, and cite similar way of expressing the idea by Macro International the prestigious firm that runs the DHS (Demographic and Health Survey). 

I've heard of misgivings due to pigeonholing before. When I was in the Pacific, a regional adviser told me that the SIAP (Statistical Institute for Asia and Pacific) ran a course somewhere with the same stereotyped content as everywhere, suitable only to an audience with much less sophisticated knowledge of statistics than those of the participants in that particular country and particular course.

A real surprising case I knew was the case of a big survey where the sample design was stratified multistage with the first stage units drawn with PPS with townships as domains. However at analysis, the first stage weight used was uniformly 3 which mean that the sampling at first stage was assumed to be SRS (simple random sampling). This clearly is not justified by any consideration. Additionally, their data entry design was poor and after data entry process has been completed, the experts could not consolidate their data for analysis. So, what they did was took the data back to their country, worked there, and somehow delivered the results. Lesson: our people need numeracy. More specifically, they need statistical literacy or some basic capacity at least to sense if the experts and advisers are doing things right. For example they should be aware of and honor the principle that confidentiality of data is important for national interest as well as for individual respondents and should not have allowed the micro-data to be taken out of the country. On the other hand, I was surprised with the apparent lack of control of the international agency that provided the technical assistance and was supposed to oversee the technical process.

In my opinion, before we expect our people or our bosses to act responsibly, we need to remove a stumbling block and it is systemic. I don't know if this still operates, though—in our days the public service had been governed by three unwritten laws: (i) thou shalt not take initiative, (ii) thou shalt not meddle, then (iii) thou shalt be rewarded with immunity; better known as ma-loke, ma-shoke, ma-pyote in our language.



Wednesday, November 19, 2014

Masala and Meat, and Professional Integrity



Years ago, an ex-ambassador turned consultant for UNDP from a developed country remarked that I was been too critical of a number of survey reports in Myanmar. Well, my language might had been crude by diplomatic standards, but for statistical reports I believe more meat is better than more masala (spices) as my older friend, an architect, used to say. Think about a cv (coefficient of variation) of more than 300% reported for an indicator in a major nation-wide statistical survey. That means masala is three times, say, the weight of the meat. What kind of curry is that?

Obviously, in the definition cv = standard error of an indicator divided by the indicator, I'm equating the numerator with masala and the denominator as meat. Thus this formula measures for each piece of meat in your curry how much masala goes with it. Now you could go on debating about the right (optimal as the experts would say) combination of masala and meat, but survey statisticians used a working rule, a rule of thumb, like cv of more than 20% is suspect. US Census case studies say: (i) High reliability: CVs less than 15%, (ii) Medium Reliability: CVs between 1530%  be careful (iii) Low Reliability: CVs over 30%  use with extreme caution.

But having found that your cv is 300% what would you do? Tick one:
       Sweep it underneath the carpet
       Report as it is

I dare say that in the old days the first one will be ticked automatically. But now you'll think. Besides, survey statisticians noted that the order of response categories is important. So the order of response items in the above question may have to be reverse to make it fair, I don't know.

Back to the question; in my personal opinion, you should report as it is, and note somewhere in the report about the possible cause(s), and caution the reader on proper use.

Anyway, in our case the report came out intact with the cv of 300+ percent as it was found means the institution, the professionals, and the officers responsible merits great respect for their professional integrity and the report for its statistical integrity. And I would say that it is the kind of thing that makes people trust a public institution for—sincerity, regardless of occasional shortcomings and some irregular performances. 

Tuesday, November 18, 2014

An Unclaimed CD on Psychometrics with R or Intro to Anything with R



We have a young relative who was a lecturer in Psychology at a university in a province in central Myanmar about a year ago. He was interested in learning data analysis and I promised to help him. Later he was lucky to be transferred to a university in Yangon and came to see me again a month ago. Then I urged him to get hold of a laptop, and a cell phone (he already has one) to serve as a mobile hotspot for internet connection. As a complement to this combination or alternately as a second rate option, I urged him to use the computer facilities and internet access available in his university.

He hasn't come back yet. May be he has changed his priorities. May be quantitative analysis is not that important for his work, or may be there are other reasons.

Meanwhile, I started assembling a very rudimentary kit consisting of core data analysis software (The R Statistical Environment software with a few additional packages for quantitative psychology) and some reference materials on Psychometrics, about R, some statistical text, and list of resources on the Web. I also wrote down my intention and rationale in a README file. Then I burned them onto a CD and this has not been claimed yet. The following is an excerpt from this file:

A few words of explanation

This is an attempt to expose Myanmar Psychometricians (who are not familiar with R) to the power of the R statistical environment. R is very powerful and completely free. We no longer need to rely on pirated copies of SPSS or SAS or STATA software to do complex statistical analyses.

R is said to have a steep learning curve. This is mostly a myth. I am not much more than a beginner in R, but I taught myself to do useful things with R without too much trouble!
Besides, there are a lot of introductory R courses and tutorials on the Internet. I have collected some material that I think would be relevant for Psychometrics. My collection may be too deficient for the psycho-professionals and academics, so I urge them to discard what they don't need and add what is useful.

My sole purpose is to stimulate awareness of a powerful set of statistical tools for the data analysts in various fields. If you are interested, I hope that my collection here would help induce you to go on and learn R by yourselves. Towards that end, you are free to share all the material on this CD. You will see that apart from tidbits on Psychometrics and R, I have included the R program and a few packages relating to Psychometrics.

As for myself, lack of opportunities has driven me to be an autodidact to some extent. I learnt a few things I know about personal computers, data analysis, and R mainly via this route. So, based on my experience and some whiffs of imagination I would recommend the following to get started with R.

(i) Get some idea of what R could do. There are lots of materials on the Web for this. Don't search for "R". Try "cran R" or "cran r".

(ii) Get more specific idea of what package(s) would suit your work. R has the base package and packages that do specialized statistical work. You could search on the Web like "Psychometrics with R", "GLM with R", "Factor analysis with R", etc. If there is a specific data analysis task in hand, you will not miss some R package(s) that would do it for you. The more organized way to find out which package suits which data analysis task is to look at the "Task Views". I'm giving a structural view of that below:

CRAN Task View: Psychometric Models and Methods
Maintainer:
Patrick Mair
Contact:
mair at fas.harvard.edu
Version:
2014-10-25
Psychometrics is concerned with theory and techniques of psychological measurement. Psychometricians have also worked collaboratively with those in the field of statistics and quantitative methods to develop improved ways to organize, analyze, and scale corresponding data. Since much functionality is already contained in base R and there is considerable overlap between tools for psychometry and tools described in other views, particularly in SocialSciences, we only give a brief overview of packages that are closely related to psychometric methodology.
Please let me know if I have omitted something of importance, or if a new package or function should be mentioned here.
Item Response Theory (IRT):
  • The eRm package fits extended Rasch models, i.e. the ordinary Rasch model for...
  • The package ltm also fits the simple RM. Additionally, functions for estimating...
  • TAM fits unidimensional and multidimensional item response models and also...
. . . . . .
  • WrightMap provides graphical tools for plotting item-person maps.
Correspondence Analysis (CA):
  • The package ca comprises two parts, one for simple correspondence analysis and ...
  • Simple and canonical CA are provided by the package anacor. It allows for ...
. . . . . .
  • SVD based multivariate exploratory methods such as PCA, CA, MCA (as well ...
Structural Equation Models, Factor Analysis, PCA:
  • Ordinary factor analysis (FA) is the package stats as function factanal(). Principal...
  • The sem package fits general (i.e., latent-variable) SEMs by FIML, and structural...
. . . . . .
  • The MplusAutomation package allows to automate latent variable model estimation and interpretation using Mplus.
Multidimensional Scaling (MDS):
  • The smacof package provides the following approaches of multidimensional scaling...
  • The PTAk package provides a multiway method to decompose a tensor (array) of ...
. . . . . .
  • The package MLDS allows for the computation of maximum likelihood ...
Classical Test Theory (CTT):
  • The CTT package can be used to perform a variety of tasks and analyses ...
  • Functions for correlation theory, meta-analysis (validity generalization), ...
. . . . . .
  • QME (not on CRAN) computes measures from generalizability theory.
Knowledge Structure Analysis:
  • DAKS provides functions and example datasets for the psychometric theory of ...
  • The kst package contains basic functionality to generate, handle, and manipulate ...
Other Related Packages:
  • The psychotools provides an infrastructure for psychometric modeling such as ...
  • Recursive partitioning based on psychometric models, employing the general MOB...
. . . . . .
  • The TestScorer package provides a GUI for entering test items and obtaining raw ...

CRAN packages:
... ...
Related links:
. . . . . .


In the "CRAN packages" listed above I've counted 119 packages to take care of your specific needs of Psychometric analysis. If you are interested, download the software and packages you want and march on ...

(iii) Before doing that, you need to know how to get to a "mirror" site. I have consistently used the Singapore mirror: http://cran.stat.nus.edu.sg/. Use it and once you are at that home page, use contents of left pane to navigate.

(iv) Get some exposure
I always liked titles like XXXXX for Dummies for myself and other beginners (no offence intended), but for R they don't have it free. Now try these. Look promising (haven't read them myself, though):

Do it yourself Introduction to R (2014)

Teach Yourself R (2008)

(v) On psychometric theory
Don't know anything. But this looks good—An introduction to psychometric theory with R available at: http://www.personality-project.org/r/book/ .
Author William Ravelle explains in his overview:

This page is devoted to teaching others about psychometric theory as well as R. It consists of chapters of an in progress text as well as various short courses on R.
The e-book is a work in progress. Chapters will appear sporadically. Parts of it are from the draft of a book being prepared for the Springer series on using R, other parts are just interesting tid-bits that would not be appropriate as chapters.
It is written in the hope that I can instill in a new generation of psychologists the love for quantitative methodology imparted to me by reading the popular and then later the scientific texts of Ray Cattell [Cattell, 1966b] and Hans Eysenck [Eysenck, 1964, Eysenck, 1953, Eysenck, 1965]. Those Penguin and Pelican paperbacks by Cattell and Eysenck were the first indications that I had that it was possible to study personality and psychology with a quantitative approach.

For orientation you may like to visit: http://www.personality-project.org/index.html

A lazy concluding remark: As of today the complete list of R Task Views is as shown in the table below. Look in the category of your choice; modify the text above appropriately and voila! you'll get an introduction tailored to your need.