Saturday, May 02, 2015

Houston - 5

I think several of our postings have started with a Sunday experience.  This one does, too. 

Last Sunday (4-26-15), we attended the Bellaire United Methodist Church.  Our first time there, and it turned out to be a serendipitous choice.  As we entered the sanctuary we saw a large group of nicely dressed young people milling around in the choir loft.  It was the Texas A and M (Blogger doesn't like an ampersand) Century Singers, a choral group of about 80 students, on tour.  Great voices, great harmony, great selection of songs: Deep River, Down to the River to Pray, Amazing Grace (with congregation participation), and an Old Irish Blessing. This link is to a previous choir edition's rendition of Amazing Grace.  The closing congregational hymn was Because He Lives, I Can Face Tomorrow.  One tomorrow at a time, with faith, is the way Susie and I face tomorrow's treatments and upcoming surgery.

Friendly church members, too.  We were early and a lady came over and visited with us.  She had NM connections -- a ranch, I believe, near Silver City, in SW NM, and family in Lovington (SE NM).  She pointed out that the next evening there would be a clogging group at church putting on a show and she invited us - this event is done at the church periodically and is very popular.  That got my attention; I went and I believe it was the highlight, so far, of my extra-curricular activity (Susie, many of you know, is a dedicated FAN of Dancing With The Stars, so that's #1 on her Monday evening agenda.)  (Actually, upon further reflection, the cloggers are my number 2 so far; number 1 is still the Nolan Ryan Center in Alvin.) 

At the meet and greet part of the service Susie talked to a lady two rows in front of us and soon learned that she had a son who lives in Albuquerque.  Mentioning Albuquerque is always a good conversation starter.

Before I get into the clogging, one comment on the sermon.  The preacher mentioned that Thomas Edison tried more than 6000 filaments before he found the one he ultimately used in his patented light bulb.  This was 1880.  My takeaway from that fact was that it was too bad Edison didn't have the benefit of having a statistician on his staff who understood statistical design of experiments (which got its boost in the 1920s and 1930s from the research of Sir Ronald Fisher.)

(The Wikipedia author on experimental design cites a controlled experiment in 1747 by James Lind that found that scurvy could be cured by citrus fruit.  Sad to say, but I was not aware of that important fact, and here I am the author of a soon to be published book on Statistical Experimental Design - more on this later, so you can see why my mind wondered in this direction during the sermon.)

Back to clogging.  The group that performed is the Texas Lovin' Cloggers.  They've been a group for 28 years, have won many competitions, and traveled internationally.  Here's a  youtube link. 

One of the best things about clogging is the music.  This group danced to a variety of recorded music, including bluegrass, country, Dixieland, pop, ... .  The you tube link is to their performance to the bluegrass song, Shady Grove.  They did that number at this show and also a dance to The Foggy Mountain Breakdown.  One interesting number was all percussion - stomping and clapping, no music.  They said that at competitions the judges sit with their backs to the dancers so they're judging the precision of the choreography by what they hear, not what the stompers looked like.

Here's one picture I got.


It's the music, I tell you.  And the choreography.  Not the leg show.  This is a family blog.

I'd guess there were 250 or so in attendance.  Before I cropped this photo there was a much larger sea of white between me and the stage.

The group took a couple of breaks to change costumes and had a local vocalist perform in those periods.  She was good - did a little bit of Patsy Cline and other music - a very good rendition of Harper Valley PTA.  Ditto for Unchained Melody.

After the second break the cloggers, men and women, came out for their final number all dressed as seasoned women, with gray bun hairdos and walkers.  I think the song was If You Could See Me Now, ... .  Hilarious, really brought the house down.  As I said, a very fun evening.  And I got home in time to see some undeserving couple get booted off of Dancing With The Stars.  Now, if someone would only create Clogging With The Stars.


Sunday, after church, we made our near-weekly pilgrimage to Cracker Barrel.  Susie had been having phone trouble the previous few days - the people she was talking to couldn't hear her.  ("I hear you.  Do you hear me?  Huh?  I'll try again. ... ")  That's not good, so after lunch we found a Verizon store to have the phone checked out.  They found out that the only solution was to upgrade to a newer model, which we did.  Think we should have gotten a second opinion?  Actually, the lady seemed sincere and helpful and we were due for an upgrade. 

Our most adventurous meal out was to a Vietnamese restaurant.  Susie once lived in Bellaire, a suburb on the SW side of Houston, and we had previously driven west on Bellaire Ave. until we ran out of town.  This is now a heavily Asian area, Chinatown in effect.  Also a big area for jewelry.  We went out there shopping for a couple of rings that would hold Susie's wedding ring on (perilously loose now because of her weight-loss).  There were too many Chinese restaurants to choose from, so we picked a Vietnamese one because it had a lot of cars around it. The spices are different, but we had a good lunch...(Susie says, "Well, an interesting lunch!")

One other adventure.  For several days I had been getting a lot of feedback from my left hearing aid - feedback as in a screech, not feedback as in, "That's a nice looking hearing aid, sir.  Do you like it?"  I called my audiologist and the suggested source of the problem was wax buildup. I tried some drops to try to break up the blockage, but no help there.  Simultaneously, I was having trouble keeping the thing in my ear - the little rubber dome that's supposed to hold it in my ear canal, wasn't.  My ear was getting sore from my efforts to jam the dome in there.  So, at Susie's insistence, I googled up an audiologist who handles my brand and we went to his shop (disguised as a house tightly surrounded by trees and apartment buildings, with a sign out front that wasn't very noticeable; after two fruitless drive-bys I had to call for directions and the receptionist stood out in front and waved us in). 

What the audiologist found was that I had a rubber dome stuck in my ear canal!  How I managed not to notice that at some point when I removed my hearing aid there was no dome on it, I don't know.  Probably happened when I was doing my monthly dome and filter replacement.  My guess is that while doing this I must have thought the old dome had fallen on the floor, so I put a new one on, not realizing that the missing dome was in my ear!  Efforts to get the new dome to seat properly were just jamming the wayward dome deeper.  Sorry, it's taking me so long to tell this story, but at least I left out the ear hair part.  All is well now.  The Houston audiologist easily extracted the misplaced dome.

Back to Statistics.  Many of you know that I've been working on a statistical textbook for quite a while, particularly the last year.  I sent the completed manuscript to my publisher late last year and the copy editor(s) went to work transferring it to book format.  Soon after we got here they sent me a list of Author Queries: questions to me about references and various textual issues.  Mostly cosmetic issues, not content.  I had been worried that the copy editors might try to convert my intentionally conversational prose to standard, dry textbookese, but they didn't.  So, it reads like this blog.  At any rate, there were no major issues and I sent my responses back in mid-March. Then, a couple of weeks ago I got page proofs to review.  I did a fair amount of wordsmithing (every time I read something I've written, like this blog, I tend to make changes) and corrections, and also found a couple of places where a figure or a table needed to be moved.  I sent the amended proofs back the week before last (incidentally, I'm working with copy editors in Singapore, which is 13 hours ahead of us, so we don't have instantaneous correspondence back and forth).  I also created an Index, much aided by being able to use the Find function to find all uses of selected terms in a document.  Some time soon I will get the final page proofs and a final chance to make minor changes - can't disturb the pagination. 

You can now go to Amazon and search on Easterling and find the publisher's blurb - and pre-order a copy.  The blurb says the book will be published in August.  (Added: I'll make it easier: you can click here.)  Funny thing: first time I went to the Amazon site, the cost of the book was $76.  Now it's $91.  Sorry about that.  Shop soon before the price goes up again.

The cover cartoon, shown below, is my depiction of statistical analysis.  Its meaning, explained on the back cover, is that indistinct, disorganized INFORMATION is dripping from a data cloud.  The purpose of statistical analysis is to capture and distill that precipitation and pour out clear, crisp, INFORMATION.  The process works best with some cloud-seeding aforehand, as indicated.  The distillation requires statistical methods and subject-matter smarts to get the most out the INFORMATION.  Whaddayathink?  Now, do you like statistics?




The background text is selected quotes that I think are pertinent - and entertaining.


Here's an excerpt from my Preface that tries to describe my goal.  For your bedtime reading:


I have a dream: that professionals in all areas – business; government; the physical, life, and social sciences; engineering; medicine, and others – will increasingly use statistical experimental design to better understand their worlds and to use that understanding to improve the products, processes, and programs they are responsible for.  To this end these professionals need to be inspired and taught, early, to conduct well-conceived and well-executed experiments and then properly extract, communicate, and act on information generated by the experiment.  This learning can and should happen at the undergraduate level – in a way that carries over into a student’s eventual career.  This text is aimed at fulfilling that goal. 

………………………………………………………………………............................................................

Successful experiments require subject-matter knowledge and passion, and the statistical tools to translate that knowledge and passion into useful information.  Archie Bunker, in the TV series, All in the Family, once told his son-in-law (approximately, and with typical inadvertent profundity), “Dont give me no stastistics (sic), Meathead.  I want facts!”  Statistical texts naturally focus on “stastistics:” heres how to calculate a regression line, a confidence interval, an analysis of variance table, etc.  For the professional in fields other than statistics, those methods are only a means to an end: revealing and understanding new facts pertinent to his or her area of interest.  This text strives to make the connection between facts and statistics.  Students should see from the beginning the connection between the statistics and the wider business or scientific context served by those statistics. 


To achieve this goal I tell stories, about experiments, and bring in appropriate analyses, graphical and mathematical, as needed to move the stories along.  I try to describe the situation that led to the experiment, what was learned, and what might happen after the experiment: “Fire the Quality Manager!  Give the worthy statistician a bonus!” Experimental results need to be communicated in clear and convincing ways so I emphasize graphical displays more than is often done in experimental design texts.

 …………………………………………………………………………….................................................................

I am convinced that personal and organizational progress, and even national and global progress, depend on how well we the people, individually and collectively, deal with data.  The statistical design of experiments and analysis of the resulting data can greatly enhance our ability to learn from data.  In George Boxs engagingly-illustrated formulation (Box and Friends 2006), scientific progress occurs when intelligent, interested people intervene, experimentally, in processes to bring about potentially interesting events and then use their intelligence and the experimental results to better understand and improve those processes.  My sincere hope is that this text will advance that cause.


So, I've been thinking more than usual about how we the people, and our representatives, deal with data.  Here are some ruminations.

"Correlation is not Causation."

This is a common Stat101 lesson, sometimes illustrated by data on stork sightings and human birth rates.

Here's a government example.

Statistics show, in various respects, that homeowners, overall, have a better quality of life than those who aren't.  E.g., income, health, crime, education, unemployment, marriage and divorce rates, .... .  There are various quality-of-life 'markers'  that show this correlation, or association, of life-quality with home-ownership.

Politicians of both parties interpreted that correlation as causation and decided that they could improve the population's overall quality of life by creating more homeowners.  Thus, began sub-prime lending.  Requirements were relaxed and lots of people got loans to buy houses and then couldn't make the payments, and more problems ensued.  The housing bubble burst.  I think we're approaching our seventh Recovery Summer from that fiasco. 

Lots of government and private actions have unintended consequences, many, I suspect, due to treating correlation as causation.  Glenn Reynolds, a U Tennessee law faculty member and blogger (his blog is called Instapundit and I often check it during the course of a day) has a Reynolds' Law that expresses this situation as:

“Subsidizing the markers of status doesn’t produce the character traits that result in that status; it undermines them.”

I'll leave it as an exercise for the reader to name other examples.


"Abolish numerical goals."

This statement is one of 14 rules for management, promulgated by the late W. E. Deming, a statistician and quality guru who became famous by helping Japan greatly improve the quality of their commercial products, after WWII. 

You might think this dictum is anti-statistical, since statisticians like to measure things, to collect data.  And if you measure something, surely you need a goal against which to measure your progress.  ("Don't call me Shirley," Airplane 1980.)  But Deming's point is more about management methods.  He wants data to be used to better understand and improve processes.  Management may not take time for that and may even think that by setting numerical goals they are empowering employees to bring their knowledge to bear on how the process works and thus meet the goal.  Moreover, some may think, setting "stretch goals" will really get the workers' attention and enthusiastic efforts and miracles of ingenuity will happen.

Deming's point was that, in the absence of what he called "profound knowledge" of how a system or process works, the connection between process inputs and their outputs, numerical goals tend to lead to either distorting the system or distorting the data.  Which can make things worse.

Case in point: VA hospitals.  Long waiting times for veterans to get appointments has been long recognized as a serious problem.  A new Secretary for Veterans' Affairs came in and said to his hospital administrators: Your goal is to reduce the average waiting time to 20 days, say.  Oh, and your year-end bonuses depend on how well you do.  Upper management, in its own view, had taken bold steps.  Hope and Change, to borrow a phrase.  Now, it was up to the administrators and workers to make the dream come true.

Well, the result was distortions of systems and data.  Phony numbers were created; the actual data were hidden.  Bonuses were paid.   Veterans were not better served.

Another case: Educational testing.  When schools and teachers are scored based on student test scores, distortions happen.  Some Atlanta teachers and administrators recently received jail sentences for falsifying test scores.   A few years ago I read the book, Freakonomics.  One of the authors is an economist/statistician very proficient in analyzing large amounts of data and finding interesting patterns (thereby embodying my book cover).  One chapter was on Sumo wrestlers and Chicago teachers.  Noodling through the data he found evidence of cheating in both situations.

To lay the blame for cheating as "numerical goals" is not to condone cheating.  It should be a lesson for higher-ups that their methods can have unintended consequences (a la the Reynolds' Law) and they need to understand how systems and processes work before they even think about desired outcomes (or can recognize the difference between correlation and causation).  You may need a lot of (good) data to gain that understanding.  For example, what are the personal traits and practices that enable a person or family to buy a house.  Is it just privilege - e.g., inherited money - or is it something, or many things, else?  And can those characteristics, once identified, be more widely attained?  I don't think free college tuition will do it.

To tie both of these semi-statistical topics together, I'll note that government regulators told mortgage lenders that if you don't make enough of these high-risk loans (meet numerical goals), we'll punish you.


Enough on statistics and government.  Time for some family pictures:

Macy dressed to celebrate Chinese New Year.


 
Malia ready to compete.
 


Susie's Striders, NYC


The grand-twins on the train.


What I need here is a picture of grandson, Jason, in NC, age 15+, behind the wheel of a car.  Use your imagination.


Continuing my trek through Houston sports, I went to a soccer game last night (Friday): Houston Dynamo vs. FC Dallas.  The home team got waxed, 4-1. 


Given the Houston-Dallas rivalry in many contexts (the Houston Rockets had just beaten the Dallas Mavericks in a first-round NBA series) I expected a larger and rowdier crowd and a more emotional game than what transpired, but it was still fun.  A columnist in the paper a few days ago wrote about how Houston is better than Dallas in many ways, which I've mostly forgotten.  The stinging rebuke, though, was that Dallas can't stand on its own; it needs its hyphenated little buddy to help it be a real city, as in Dallas-Fort Worth.  Or, DFW airport, etc.  Hope my Texas readers see this.

I did the whole trip by Metrorail and walking.  Glad I did.  Noticed that close-in parking was $40.  I had about 15 min. walks on each end. 

One more item about the Astrodome.  A couple of weeks ago I ran across this nostalgic article:

http://m.mlb.com/news/article/119126222/richard-justice-astrodome-remembered-as-baseball-innovation

It's a good one.

Susie will finish her combined chemo and radiation treatments a week from Monday (May 11).  Then we drive home for some R&R.  We spent quite a bit of time this week checking out various furnished apartments to rent when we return in about two months.  Found one that's as nice as the one we're in and not as pricey.  Now, we're busy deciding what to take home and what to store here (at a storage locker in the area) that we'll need when we get back.  On Friday we'll meet with surgeon to find out the schedule for surgery. 

We'll be in touch.

Susie and Rob