About Me

When not at work with students, I spend my time in my room either reading, calculating something using pen and paper, or using a computer. I read almost anything: from the pornographic to the profound, although my main interests are mathematics and physics. "When I get a little money I buy books; and if any is left I buy food and clothes." -Erasmus

Saturday, January 22, 2011

Blog Comments and Vaccination

I've been reading Orac's blog, and I've been entertaining myself (as well as procrastinating) by reading the comment thread. It's a lot of fun because it's uncensored, and you get to see how perfectly nasty (in a good way) some people can be. It's a case of schadenfreude-- it's enjoyable reading how trolls get nuked by other readers; it's also nice to see that the comments include links to evidence (articles in peer-reviewed journals).

His blog seems to be the kind that draws a lot of fire from the anti-science mainstream in the U. S.; it's terribly disappointing to see how bad science education is in the U. S.  I recall A Private Universe, a video where Harvard students (arguably considered the brightest in the United States) were asked about the origin of the seasons, and they failed to answer such a basic question correctly.

I probably wasted half of the afternoon just reading the comments, and following up on the cited papers. I also learned, for example, about the vaccination rates at various kindergarten schools in California. One of the statistics that intrigued me is the connection between vaccination rates and the rates for PBE's (or personal belief exemptions, where a parent has to sign an affidavit asking that his or her child be exempted from vaccination).

One of the worries of  the pro-vaccine camp (of which I am a member) is the reduction of herd immunity. Vaccines are never 100%  effective. There is still some chance that you can get infected if you get exposed to the pathogen, and the vaccine didn't take. More vaccinated people reduces the likelihood of being exposed to the pathogen, so getting vaccinated is also a good thing for your neighbor.

On the other hand, if you live in a region where most people aren't vaccinated, you are more likely to catch the disease compared to those living in areas of high vaccination rates. So it makes sense to look for pockets where many people are not vaccinated, because they would be most at risk. So I looked at the vaccination rate data in Californian kindergartens.

One of the stats on the California Department of Public Health is the PBE rate; high PBE rates mean parents are opting out of immunization. The recommended immunization rates for herd immunity is greater than 90 percent, and I see schools where more than 50 percent of parents opt out. This means not enough immunizations for herd immunity, and it's no surprise when outbreaks do occur at these places.

Another statistic I noticed was the how often Waldorf schools get on the list of kindergartens with PBE's greater than 50 percent. Coincidence? I immediately thought of outbreaks at such schools, and unhappily found out that I was right to suspect that outbreaks would occur at these schools. In fact, they have occurred.

I then read news items on Waldorf schools, and found that there's actually a journal article Eurosurveillance, Volume 15, Issue 26, 01 July 2010 , in which an outbreak of measles is associated with a Waldorf school. I also saw news items on outbreaks involving other Waldorf schools. A search on google using the keywords: outbreaks Waldorf school (without parentheses) will give you a host of links.

Waldorf schools, by the way, are expensive private schools, and it's downright alarming to find supposedly well-educated well-off parents not getting their kids immunized. It's almost as if they're waiting for an outbreak. By then, of course, it would be too late.

Peer Instruction and the Need To Walk Around

(I had too much work to post anything this week, and I still have a pile of papers to mark. So here, at last, is where my meandering mind has brought me:)

I use a modified version of Mazur's Peer-Instruction in teaching physics. The idea behind Peer-Instruction is that the traditional lecture-- where the professor writes on the board and talks, with little or no interaction with students-- is an ineffective way of getting our students to learn. A better way is to get our students to read the book, and then use class time for deepening our students' understanding.

During class-hours, I give short summaries of the material and then ask questions using a multiple choice format, where the possible answers are labeled from 0 to 5. These questions, called ConcepTests by Eric Mazur, are intended to get students to talk to each other. To make them effective, the choices have to be chosen so that the wrong answers will also look attractive to students.

When I display a question, I expect all of my students to think silently on their own, without consulting their notes, books or their neighbors. I usually give them around two minutes, and if the question is extra hard, I would ask students who are still thinking to raise their hands. If there are a goodly number, I give an extension of a minute or two.

When time is up, I ask my students to close their eyes and show me their answers. They do this by placing a hand on their forehead and using fingers to indicate their choice. A closed fist stands for zero, a finger for one, and so on. A good ConcepTest gets the class evenly divided; I count the number of students who give each possible answer and I tell my students how the answers are distributed. I then ask my students to open their eyes, show their answers to their neighbors, and then try to convince their neighbors. 

If the class is evenly divided, the convince-your-neighbor phase should get students to shift from the incorrect answer to the correct answer. Students can be very good at convincing other students, so I let them do that. The variety of opinions makes for some interesting conversations. 

 As a bonus, I can immediately tell if the class has a difficulty with a particular topic. I can slow down, change tack, or try other approaches. 

I do have to walk around while my students talk to each other for a number of reasons. I need to make sure that they are actually talking about the question and not using the time for something else. Other times, there will be neighbors who have different answers who don't talk to each other. I then try to get them to convince their neighbor by asking questions. 

Sometimes there will be groups of students who happen to be right, and when such a cluster occurs, I have to go to that cluster and play Devil's advocate. This means I have to do a lot of thinking and reading to figure out how a student, who finds other choices attractive, thinks. It's  hard work-- I have to read the Physics Education Research literature to learn why students find some wrong answers attractive, in order to generate good ConcepTests.  A good ConcepTest is a question that stimulates discussion; it can't be too easy, because no one will talk; neither can it be too hard because there would not be enough correct people to help other people find the correct reasoning.  It seems chaotic, so other people might dislike the method, especially if their idea of a good class is a silent class. 

Many of my students seem to like the format; others see a lot of work because the method requires that they read before coming to class. (You can't please everybody.) One of the happier advantages of Peer Instruction is that it keeps the class awake. If I ask a question, it's easy to see who's asleep-- they don't display their answers when I ask them to. When that happens, I walk to the person and stand beside him or her. It's better than a caffeine shot. 

A confession: the traditional lecture turns out to be easier to prepare for than a class using Peer Instruction. So if you see me giving a traditional lecture, the odds are good that I wasn't able to prepare!

Sunday, January 16, 2011

addenda: you might be an altie if...

Another post by Orac. Some of the symptoms that go with being an anti-science believer of "alternative medicine" and will never admit that conventional medicine may sometimes be effective.

The Pharma Shill Gambit

An entertaining post I found-- talks about the ad hominem arguments used by people who believe in alternative medicine whenever some guy (such as yours truly) asks for evidence (clinical trials, refereed research, etc.).

Saturday, January 15, 2011

How to Lie with Statistics

I've recently read a paper from the journal Contraception, a journal published by Elsevier. The paper, Contraception 83 (2011) 82–87Trends in the use of contraceptive methods and voluntary interruption of pregnancy in the Spanish population during 1997–2007  by Duenas et al, is being used by people against the Reproductive Health Bill. The paper has noted that from 1997 to 2007, the Spanish population's abortion rate (coyly termed early voluntary interruption of pregnancy) increased from 1997 to 2007, at the same time that the contraception rate increased. Anti-R. H. bill writers have used this paper to argue that encouraging contraceptive use is the same as encouraging abortions, which is nonsense, as I see it.

There are a number of reasons why I disagree (after all, although I think that people should use contraception or go on cold turkey, it doesn't necessarily mean that I would make adoption freely available, so there is a counter-example!), but what makes this a case of lying with statistics is the citation of numbers without even mentioning the changing demographics of the Spanish population. The paper, if read, contains a more nuanced interpretation precisely because the numbers, as presented, do not make a strong case for the premise that more contraception means more abortion.

Who is most likely to use abortion? A good candidate would be a poor, young, uneducated unwed pregnant woman who does not wish to take on the burden of caring for a child. If we want to know if encouraging contraceptive use within this segment causes increased use of abortion, we should focus on data within this segment only, and not on the population as a whole.

This group, by the way, is probably among that part of the population that is least likely to use contraception.  If this group has increased in the population, at the same time that more of the other parts of the population are adopting effective birth control methods, then we should be able to see an increase in both the incidence of birth control use as well as abortion.

Demographic data from Spain does support an increase in  that part of the population. Immigrants, for example, (likely to be poor and uneducated) have increased from a mere 2 percent of the population (at the start of the study) to 10 percent of the population (at the end of the study). The authors of the study probably did not expect such large demographic changes; and worse, they do not have the numbers of poor, unwed, pregnant women (the portion at risk). I'm still reading the paper to see if I can tease apart these numbers from the data. For now though, the cautious conclusion-- the data is insufficient to support the premise that contraceptive use increases abortion use-- holds.

An analogy: cheating on an exam versus studying for the exam. Suppose you have a group of 1000 students, with 10 of them deciding to cheat, 490 of them using ineffective study methods, and the rest (500) using effective study methods. For simplicity, let us assume that these groups are mutually exclusive. We will now have the following percentages: 50% for effective study methods, 49% using ineffective study methods, and 1 percent cheating.

Suppose that next year, you get another sample of 1000 students. Of this sample, 60 decide to cheat, 340 use ineffective study methods, the rest (600) have adopted effective study methods. Then we will obtain the following percentages: 60 percent use effective study methods, 34 percent use ineffective study methods, while 6 percent cheat. Again assume that these segments are mutually exclusive.

We now have an amusing result: a 10 percent increase of people who study effectively, and at the same time, a 5 percent rise in the incidence of cheating.  Ergo, encouraging the use of effective study methods increases the incidence of cheating. (There are other ways of calculating the numbers in a way to make a more dramatic point. We could say instead, the fraction of people who cheat has increased six times, at the same time that more people use effective study methods.)

I've recently read Darrell Huff's How to Lie with Statistics, and I was greatly entertained. It's a good book for people who want to learn how to detect statistical shenanigans; J. Michael Steele, a Wharton professor, has written a review of the book, so I will not talk much about it. I recommend it because it contains lots of tales of how statistics can be misused, especially when there is actually less than meets the eye.

Wednesday, January 12, 2011

AM radio can be hazardous to your health

I thought that I had convinced my Mom that the doctor she believed in was a charlatan; I was wrong.

Last night, I got home to find an empty house. I waited until 9 pm until my Mom got back. When I asked where she went,  I learned that she had just visited Dr. Navarro, and that she has decided not to take her medication. We got into an argument-- we spent a lot on the hospital confinement, and we're continuing to spend money on her medication, and all of it goes down the drain if she decides to quit her medication. (This charlatan, by the way, advocates an all-beef and egg diet for cardiovascular disease and hypertension.)

She has already seen the evidence that this charlatan's degrees are fake-- the Oxford College of Applied Science is either nonexistent or a diploma mill, the Committee for Basic Research does not exist in Oxford, etc. If the guy is likely to be lying about his degrees and experience (he probably doesn't know that a good search engine can unearth the non-existence of the places he claims to be associated with!), then he should not be trusted.

She argues that other doctors (a Doctor Sy, in particular) on AM all say the same thing, and that I should open my mind. I counter of course, that a perfectly open mind believes in everything-- one needs to choose what to believe, and due diligence requires that we double-check.

And so I visted Dr. Gary Sy's website.

Among the things I've seen being promoted on Dr. Sy's website is GeroVital, a drug that is banned by the F. D. A. I've also looked at PubMed, and among the things I've seen are articles where use of GeroVital led to acute renal failure, studies where it was shown that GeroVital is not more effective than a placebo. So it looks like negligence on the part of Dr. Sy, or a lack of bullshit detectors on his part. (This is a generous assessment.)

You can look at the other entries within the  health and beauty services part of the website. and judge for yourself. The radio appearances, just like Dr. Navarro's, seem to be a way of attracting business.

Getting back to Doctor Navarro: she claims that he has healed a lot of people, and cites testimonials.

I tried to explain  about selection bias, and why I choose to disregard testimonials as medical evidence. Scientific medicine performs clinical trials instead of relying on testimonials. A clinical trial goes like this: Suppose that a particular drug is tested on a group of people; suppose that a control group is given a placebo, while an equivalent group is given the actual drug.

A percentage of the control group will get well in spite of having no medication. To be able to accept the drug, a significantly larger percentage of the treated patients should get well compared to the control group.

Why are testimonials unreliable? First, in contrast to a clinical trial, no comparison is made between a control group and the treated group. You can expect that testimonials will only be comprised of people who got well, and this will include people who will get well even without the treatment.

Second, the people who did not get well, or who died due to the treatment will not give testimonials. This means testimonials are biased in favor of the suggested treatment-- we do not see the unsuccessful cases. Even if the fraction of the people who got well in the treated cases is much smaller than the people who got well untreated (which may imply that the treatment may actually worsen the condition), we only see the "successes."

I think that doctors on AM radio use radio exposure (television coverage is better) it as a means of promoting something they want to sell. It means they may have hidden agenda, and every listener is well advised to remember Caveat emptor. I've said it before, but it's worth reiterating. There ought to be a sign on AM radios: listening to Doctors on AM radio may be hazardous to your health.

Sunday, January 9, 2011

Teetotaling, almost

I used to drink beer with friends; there's nothing like conversation that is smoothened by the flow of alcohol. We used to call the drinking sessions our research meetings; we liked to talk about physics and math questions that we found interesting. Sometimes we talked about the progress of our research.

Sadly, the last few times I did drink beer (and other drinks with higher alcohol percentages), even though it was not to excess, I ended up feeling unwell. I'm not sure what it is in beer that doesn't agree with my stomach (am I developing a beer allergy?), but because it kept on happening when I drank beer, I decided to give up beer. No such effects from wine.

When I was an undergrad, I didn't drink wine because I didn't like the taste. Whenever my Dad brought out a bottle, I sampled it, but couldn't make myself drink more than half a glass. I learned the reason later; my Dad preferred dry wine, while I preferred sweet wine. I discovered the difference when I sampled Sangria, and noticed that it was sweet. A little more reading and sampling led me to sweet wines.

While reading about wine, I chanced upon the notion of alcohol belts. It seems that Europe can be divided into three regions based on the most popularly consumed alcoholic drink. The highest latitudes would be the vodka belt, the next highest would be the beer belt, while the lowest latitudes (or the ones nearest the equator) would be the wine belt. The reason such a belt exists is probably climate:  vodka is made from potatoes, beer from grain and cereals, and wine from grapes. The ease of access to the ingredients is affected by latitude-- no grapes in the vodka belt, for example.  

My resolution this year: drink more wine, instead of beer or other spirits. I've read about the health benefits, and Novellino isn't expensive. It helps that there's a nearby shop where I can buy it at a discount.  I've limited myself to sweet wine (I don't understand the European desire to drink something dryer), and only to a glass right after dinner.

Friday, January 7, 2011

The Syllabus and the Exam

We'll be holding the first long exam of the modern physics course I'm teaching this coming Monday. To help our students prepare, we hand out, at the beginning of the term, a detailed, day-by-day syllabus. The syllabus contains the following information: for every meeting, we list (1) the sections in Young and Freedman to be read before coming to class, (2) a list of things that the student should be able to do after that meeting, (3) a list of suggested exercises and problems. 

An example from our syllabus:


Give an operational definition of an event and describe how coordinates are assigned to an event.

Define the spacetime  interval and differentiate it from the Euclidean notion of distance.

Use coordinates to calculate the spacetime interval using the records of an inertial frame.

Use invariance of the interval to relate spacetime separations in one inertial frame to spacetime separations in another inertial frame.

Events and Measurements

Invariance of the Interval

Read Sections 4 to 5 of Spacetime Physics
Exercises from Spacetime Physics: 1, 2, 3

The left-hand corner gives the meeting number, the middle gives the list of behavioral objectives, and the last column gives the major concepts to be covered during the meeting. The entry just below is the list of behavioral objectives gives the reading assignment (to be read before coming to class!), and suggested problems for solution.

Aside from the syllabus, we also hand out a "long problem set". The long problem set is a collection of 45 item-multiple choice questions, sorted by meeting number. We tell our students that after the discussion in class, they should be able to solve the items in the long problem set that correspond to the class meeting. Although we hope that our students try the questions immediately after the discussion (so that they won't be left behind), we collect the answers on a weekly basis so that if they do cram, they cram more often. For example, if, during the week, we've finished meetings 20, 21, 22, at the end of the week we collect the items that fall under those meeting numbers. 

When we make exam questions, we do it on a per-meeting basis, to make sure that the material is well-sampled. We also try to make sure that the exams reflect the objectives. The syllabus, if faithfully followed, lets students know what to expect from us. As a bonus, it makes the question construction process easier.

When I first joined the course group, the syllabus wasn't like this. The topics were grouped by hour(s), and there were entries in which the topics were grouped together for three hours or more. I suggested reworking the syllabus into hourly segments because I wanted my students to read before coming to class (since I used Eric Mazur's Peer Instruction; see also Eric Mazur's website), and I wanted to make the reading assignment manageable. The older syllabus had no reading assignments and suggested exercises; this, I think, made it harder for students to prepare for class. 

I was lucky that I had the course group leader as an accomplice. With his help, we were able to push forward a reformed syllabus. The ideas behind the detailed syllabus I owe to my thesis adviser; he was in turn, influenced by colleagues in the College of Education (Professor Adriano, in particular). 

On reflection, what dismays me is having to redo the syllabus all over again. Long before I was an instructor, my thesis adviser was involved with the construction of the same syllabus. The changes they introduced then got lost-- because the instructors who replaced them were unaware of why the syllabus was designed that way. I suspect the effort is Sisyphean; whatever changes we make today will probably be undone by an ignorant someone a few years down the line. But I persist, because the effort is worth it.