Eric’s Enlightenment for Monday, May 25, 2015

1. A plant called thale cress (Arabidopsis thaliana) can detect sounds that are made by caterpillars that feed on its leaves.  In response, it mounts a defense by producing glucosinolates and anthocyanins – cool research by Heidi Appel and Reginald Croft!
2. Economists offer 10 pieces of data-driven advice for university graduates about succeeding in today’s job market.
3. Very nice and in-depth interview with Claudia Goldin on labour economics and education, especially in terms of differences between men and women.
4. I was very sad to learn of the deaths of John Nash and Alicia Lopez-Harrison de Lardé.  Here is a nice obituary by Benjamin Morris, with examples of non-cooperative games and Nash equilibria from soccer, football, basketball and rock-paper-scissors.

Eric’s Enlightenment for Friday, May 22, 2015

1. John Urschel (academically published mathematician and NFL football player) uses logistic regression, expected value and variance to anticipate that the new farther distance for the extra-point conversion will not reduce its use in the NFL.
2. John Ioannidis is widely known for his 2005 paper “Why most published research findings are false“.  In 2014, he wrote another paper on the same topic called “How to Make More Published Research True“.
3. Yoshitaka Fujii holds the record for the number of retractions of academic publications for a single author: 183 papers, or “roughly 7 percent of all retracted papers between 1980 and 2011”.
4. The chemistry of why bread stales, and how to slow retrogradation.

Eric’s Enlightenment for Thursday, May 21, 2015 – A Special Edition on the Mental Health of Chemistry Graduate Students

Today, combining

• my passion for chemistry,
• my experienced knowledge of university culture in North America,
• and my deep concern for mental health issues,

The Chemical Statistician will feature a collection of writing about the struggles that graduate students in chemistry face during their studies, and how those struggles affect their mental health.  This is a special edition of Eric’s Enlightenment.

1. Chemjobber began a dialogue with Vinylogous about mental health and graduate studies in chemistry in 2013.  It started with this blog post as Part 1, containing reflections of Chemjobber’s own experience and thoughts on general issues on this subject.
2. In Part 2 of their dialogue, Vinylogous responds to Chemjobber with a very detailed post on his conjectures of why graduate studies in chemistry is so hard on a student’s mental health.
3. In Part 3 of their dialogue, Chemjobber responds to some of Vinylogous’ main points and addresses possible solutions to mental health challenges for chemistry graduate students.  He/She also begins to answer the question “Is a graduate degree in chemistry worth the sacrifice?”.
4. In Part 4 of their dialogue, Vinylogous examines some alternative issues in this subject, including possible benefits of chemistry graduate studies for mental health, how some research supervisors aggravate mental health problems, and differences between sub-fields of chemistry.
5. Finally, in Part 5, Chemjobber concludes this discussion by trying to answer some of the key questions that this dialogue generated and summarizes some of the key points that they learned.
6. I am surprised that I never learned about this sad story during my studies as a chemistry student: Jason Altom was an accomplished and well-liked doctoral student in chemistry at Harvard University, yet he committed suicide at age 26, citing excessive pressure from abusive research advisers, including his supervisor, Nobel Laureate Elias Corey.  Notably, his suicide notes contained policy recommendations on how academic departments can better protect their students.

The dialogue between Chemjobber and Vinylogous was very productive, with many other chemistry bloggers adding valuable perspectives in their own blog posts.  I highly encourage you to read those articles, too.

I also highly recommend you to read the comments in all 5 blog posts – they add great diversity to the perspectives and experiences about this complicated topic.

Here are some key quotations that I gathered from these articles:

Chemjobber – in Part 1 of the dialogue with Vinylogous.

After weeks and weeks of long hours and frustration in the lab in either my 2nd or 3rd year of graduate school, I remember walking into my apartment bathroom, smashing the mirror with my fist and sitting on the edge of the bathtub. I seem to recall yelling at the top of my lungs “What am I going to do!?!?” about whatever reaction sequence of my total synthesis that simply was not going anywhere.

I can easily say that was one of the darkest periods of my time in graduate school. I am not sure if I was depressed — I’m a synthetic chemist, not a clinical psychologist. Close to ten years later, it’s mostly an unpleasant memory, with little recall of the details that set me off. But I can remember sitting on that bathtub edge, the deep despair of a project that wasn’t going well and the feeling that my entire life was an utter failure. Now, of course, I don’t feel that way at all. I can leave my work at work (mostly, anyway), and my self-worth is not entirely reliant on the yield of my last reaction. But there was a lot of pain in between then and now.

Vinylogous – in Part 2 of the dialogue with Chemjobber.

At one point during my previous degree, when I was doing research, taking classes, and teaching, my advisor told me frankly that my productivity needed to increase. It needed to double. At that point I already felt that I was at my absolutely limit in what I could accomplish in a week. At that point, I had nowhere near enough data for a paper and barely enough for a mediocre conference poster. Weekends had been given up, as had hobbies. When I mentioned to my advisor the many demands on my time, his response was short: “Sometimes you need to prioritize what’s important to you.” (The subtext: stop caring about class and teaching and hobbies). It was an existential moment. I managed somehow to increase my productivity and my efficiency, and within a year or so I had three first-author manuscripts. I defended my M.S. and graduated, moving to another (higher tier) school for a Ph.D. But I left with a pre-conditioned bitterness towards graduate work.

Vancouver SAS User Group Meeting – Wednesday, May 27, 2015

Are you a user of SAS software in Metro Vancouver? If so, join us at the next Vancouver SAS User Group (VanSUG) meeting at the BC Cancer Research Centre (675 West 10th Ave) on Wednesday, May 27! This will be a great opportunity to learn from other SAS users and network with other analytics professionals. This meeting is free, and a scrumptious breakfast will be served. Hope to see you there!

Visit our web site to register for the meeting.  Here is the agenda:

8:30 – 9:00am: Registration

9:00 – 9:20am: Update from SAS (Matt Joyce, SAS Canada)

9:20 – 9:50am: Competing Risk Survival Analysis Using PHREG in SAS 9.4 (Lovedeep Gondara, BC Cancer Agency)

9:50 – 10:20am: Switching to Enterprise Guide: Pros and Cons (Guillaume Colley, BC Centre for Excellence in HIV/AIDS)

10:20 – 10:45am: Networking Break

10:45 – 11:05am: Basics of Proc Tabulate (Anky Lai, BC Cancer Agency)

11:05 – 11:35am: Modeling Interactions and the Use of the CONTRAST Statement for Post-Fitting Comparisons (Huiru Dong, BC Centre for Excellence in HIV/AIDS)

11:35 – 12:15pm: Automating Your Metadata Reporting (Darryl Prebble, Prebble Consulting Inc.)

12:15 – 12:30pm: Evaluations, Prize Draw & Closing Remarks

Eric’s Enlightenment for Wednesday, May 20, 2015

1. A common but bad criticism of basketball analytics is that statistics cannot capture the effect of teamwork when assessing the value of a player.  Dan Rosenbaum wrote a great article on how adjusted plus/minus accomplishes this goal.
2. Citing Dan’s work above, Neil Paine used adjusted plus/minus (APM) to show why Jason Collins was one of the top defensive centres in the NBA and the most underrated player of the last 15 years of his career.  When Neil mentions regularized APM (RAPM) in the third-to-last paragraph, he calls it a Bayesian version of APM.  Most statisticians are more familiar with the term ridge regression, which is one type of regression that penalizes the inclusion of too many redundant predictors.  Make sure to check out that great plot of actual RAPM vs. expected PER at the bottom of the article.
3. In a 33-page article that was published on 2015-05-14 in Physical Review Letters, only the first 9 pages describes the research done for the article; the other 24 pages were used to list its 5,514 authors – setting a record for the largest known number of authors for a single research article.  Hyperauthorship is common in physics, but not – apparently – in biology.  (Hat Tip: Tyler Cowen)
4. Brandon Findlay explains why methanol/water mixtures make great cooling baths.  He wrote a very thorough follow-up blog post on how to make them, and he includes photos to aid the demonstration.

Eric’s Enlightenment for Tuesday, May 19, 2015

1. Melanie Bailey leads a team of scientists in developing a fingerprint test for cocaine use.  It “is based on surface mass spectrometry.  It desorbs molecules from fingerprints and detects not only cocaine but also its two metabolites, benzoylecgonine and methylecgonine, showing that the drug has been ingested, rather than only touched”.
2. 7 non-anglophone countries where anglophone university students can earn post-secondary degrees in English for free.
3. Paul Romer criticizes economic growth theorists for committing “mathiness” – the use of words and symbols that “leaves ample room for slippage between statements in natural versus formal language and between statements with theoretical as opposed to empirical content”.  He supplements this paper with a nice blog post, and he responds to Noah Smith and Brad DeLong in a follow-up blog post.
4. A randomized trial (n = 2538) of 4 different programs concludes that reward-based financial incentives work well in motivating smokers to quit smoking.  (Paying people to stop smoking works!)  Hat Tip: Alex Tabarrok.

Eric’s Enlightenment for Friday, May 15, 2015

1. An infographic compares R and Python for statistics, data analysis, and data visualization – in a lot of detail!
2. Psychologist Brian Nosek tackles human biases in science – including motivated reasoning and confirmation bias – long but very worthwhile to read.
3. Scott Sumner’s wife documents her observations of Beijing during her current trip – very interesting comparisons of how normal life has changed rapidly over the past 10 years.
4. Is hot air or hot water more effective at melting a frozen pipe – a good answer based on heat capacity and heat resistivity ensues.

Eric’s Enlightenment for Thursday, May 14, 2015

1. Alcohol kills more people worldwide than HIV, AIDS, violence and tuberculosis combined.
2. Some crystals don’t recrystallize after heating and cooling, but form amorphous supercooled liquids.  Modifying the molecular structure of diketopyrrolopyrrole using shear forces can induce this type of behaviour.  Here is a video demonstration.  Here is the original paper.
3. How pyrex was born out of an accident in cooking spongecake 100 years ago.  (Hat Tip: Lauren Wolf)
4. Check out David Campbell’s graduate statistical computing course at SFU.  It dives into some cool topics in his research that are not always covered in statistical computing, like approximate Bayesian computation and many computational Bayesian methods.

Eric’s Enlightenment for Wednesday, May 13, 2015

1. James Trussel et al. used a Markov model to estimate the relative cost effectiveness of contraceptives in the United States from a payer’s perspective.  Did you know that 49% of the 6.4 million pregnancies each year in the United States are unintended?
2. Jason Furman (Barack Obama’s Chair of the White House Council of Economic Advisers) cites empirical research to find social programs that have produced measurable long-term benefits for children in the USA.  (My question for economists: How do you establish causality in such studies?)
3. Alex Tabarrok blogs on the growing movement to give money directly to poor people in developing countries, especially in using mobile phones to do so.  He also talks about the use of analytics to evaluate charities.
4. The organic chemistry and biochemistry of allergies (Hat Tip: Lauren Wolf).

Eric’s Enlightenment for Tuesday, May 12, 2015

1. A great list of public data sets on GitHub – most are free.
2. Is the 4% withdrawal rule still effective for determining how much you can spend to attain perpetual retirement?
3. Jeff Leek compiled a great list of awesome things that people did in statistics in 2014.  Here is his list for 2013.  (Hat Tip: Cici Chen and R-Bloggers)
4. A video demonstration of the triple point of tert-butyl alcohol.

Eric’s Enlightenment for Monday, May 11, 2015

1. Benjamin Morris used statistics to assess the value of Dennis Rodman as a rebounder and as a basketball player in general – and wrote one of the most epic series of blog posts in sports analytics.  Contrary to popular opinion, he eloquently argued why Rodman was a better rebounder than Wilt Chamberlain and Bill Russell.  In a digression in Part 1/4 (a), he used assist percentage to assess John Stockton’s greatness as a passer.
2. I enjoy reading David Sherrill’s notes on quantum and computational chemistry.
3. Read the first slide of this biostatistics lecture to learn how to calculate the concordance statistic (a.k.a. the C-statistic or the area under the receiver-operating characteristic (ROC) curve).
4. Here are all of the videos of David Zetland’s lectures for his course on natural resource economics at Simon Fraser University.

Eric’s Enlightenment for Friday, May 8, 2015

1. A nice set of tutorials on Microsoft Excel at OfficeTuts by Tomasz Decker.
2. “We had proved that an assertion was indeed true in all of the difficult cases, but it turned out to be false in the simple case. We never bothered to check.”  Are mistakes in academic mathematics being effectively identified and corrected?  Vladimir Voevodsky (2002 Fields Medalist) published a major theorem in 1990, but Carlos Simpson found an error with the theorem in 1998.  It wasn’t until 2013 that Voevodsky finally became convinced that his theorem was wrong.  This motivated him to develop “proof assistants” – computer programs that help to prove mathematical theorems.
3. Synthesizing artificial muscles from gold-plated onion skins
4. Andrew Gelman debriefs his presentation to Princeton’s economics department about unbiasedness and econometrics.

Eric’s Enlightenment for Thursday, May 7, 2015

1. “Across the United States, more than half a million kids are poisoned by lead each year…” – much of it due to lead being used in building houses.
2. At a global business consulting company where 80-hour work weeks are normal, 31% of men and 11% of women examined in Erin Reid’s study “managed to achieve the benefits of a more moderate work schedule without explicitly asking for it“.  Surprisingly, they “received performance reviews that were as strong as their hyper-ambitious colleagues. For people who were good at faking it, there was no real damage done by their lighter workloads”.
3. Tom Short’s very useful reference card of major and commonly used functions in R – only 4 pages!  In case that first link ever becomes severed, here is the PDF file.
4. I offer my deepest condolences to Sheryl Sandberg for the recent death of her husband, Dave Goldberg.  I normally post only technical information on this blog, but Katie Hafner’s letter to Sheryl deserves much sharing, and it may help others who endure a similar tragedy.  It is very well written and full of practical and philosophical advice from her own experience in suddenly losing her first husband at an early age.

Eric’s Enlightenment for Wednesday, May 6, 2015

1. Moldova has mysteriously lost one-eighth of its GDP, possibly to fraudulent loans.
2. Kai Brothers was diagnosed with HIV in 1989, but did not show any symptoms for 25 years.  Does he have a natural defense against HIV?  Now that he is starting to show symptoms, should he start taking anti-retroviral drugs and deny scientists the chance to look for that natural defense in his blood?
3. Use the VVALUE function in SAS to convert formatted values of a variable into new values of a character variable.
4. Alex Reinhart diligently compiled and explained a list of major “egregious statistical fallacies regularly committed in the name of science”.  Check them out on his web site and in his book entitled “Statistics Done Wrong“.  I highly recommend reading the section entitled “The p value and the base rate fallacy“.
   

Eric’s Enlightenment for Tuesday, May 5, 2015

1. The inherent flaws of defining and estimating job vacancy rates – a commentary by Philip Cross, a former chief economic analyst at Statistics Canada.
2. Adding to my previous entry about CRISPR, here is Douglas Mortlock’s in-depth discussion of the problems in Jiang et al.’s study.  Note that his entire blog is devoted to CRISPR.
3. Robin Hanson’s proposal to evaluate teachers and students using linear regression while controlling for related variables.
4. A video on the health benefits of avocados from a chemical perspective – including the best way to cut an avocado and how to slow the browning of a guacamole dip.

Eric’s Enlightenment for Monday, May 4, 2015

In light of John Oliver’s recent video about standardized testing on “Last Week Tonight”, this edition of “Eric’s Enlightenment” will highlight some statistical discourse on standardized testing and measuring the impact of teachers.

1. Raj Chetty, John Friedman and Jonah Rockoff published 2 papers in the American Economic Review on the long-term impact of teachers on a student’s development.  This web site contains links to the 2 papers plus a lot of supplementary information, including an executive summary, the STATA code, and a video presentation.
2. The American Statistical Association released a statement on value-added models for educational assessment.  Chetty et al. responded to this statement.
3. Moshe Adler criticized Chetty et al.’s study.  Chetty et al. responded to the initial criticism, and Adler responded to their rebuttal.  All 3 statements can be found here.
4. Audrey Amrein-Beardsley compiled 86 academic articles on value-added models to assess the impact of teachers on student development.  Note that this list is sorted alphabetically.
5. Diane Ravitch posts some commentary along with Margarita Pivovarova et al.’s full criticism of Chetty et al.’s study.

Eric’s Enlightenment for Friday, May 1, 2015

1. PROC GLIMMIX Contrasted with Other SAS Statistical Procedures for Regression (including GENMOD, MIXED, NLMIXED, LOGISTIC and CATMOD).
2. Lee-Ping Wang et al. recently developed the nanoreactor, “a computer model that can not only determine all the possible products of the Urey-Miller experiment, but also detail all the possible chemical reactions that lead to their formation”.  What an exciting development!  It “incorporates physics and machine learning to discover all the possible ways that your chemicals might react, and that might include reactions or mechanisms we’ve never seen before”.  Here is the original paper.
3. A Quora thread on the best examples of the Law of Unintended Consequences
4. In a 2-minute video, Alex Tabarrok argues why software patents should be eliminated.