simplystats · sandipan · Oct 21, 2016 · Oct 21, 2016 · Oct 26, 2016 · Oct 26, 2016
diff --git a/_Rmd_files/2016-12-13-leekgroup-plots.Rmd b/_Rmd_files/2016-12-13-leekgroup-plots.Rmd
@@ -0,0 +1,40 @@
+---
+title: "Leek group guide to making plots"
+output: html_document
+---
+
+I have written a few guides for people in academics including:
+
+* [How to write your first paper](https://github.com/jtleek/firstpaper)
+* [How to review a paper](https://github.com/jtleek/reviews)
+* [How to share data](https://github.com/jtleek/datasharing)
+* [How to write an R package](https://github.com/jtleek/rpackages)
+* [How to read academic papers](https://github.com/jtleek/readingpapers)
+
+The purpose of these guides has partially been for other people to use them outside of my research group. But the main driver has been having a set of tutorials that can be a sort of "onboarding" for new members of my research group. 
+
+Recently we had to work collectively on a project where multiple members were each sending in plots and I realized that they looked very different in aesthetic, color scheme, and organization. The result is that it was pretty hard to put the figures together in a paper. It also means that when we use each other's slides in talks there is no coherent pattern to what a plot will look like. 
+
+Other organizations - like [fivethirtyeight](http://fivethirtyeight.com/) have a consistent look and feel to their graphics. They do this (I imagine) largely as a defense mechanism - they have to produce plots every day! But I think that it also adds to the professionalism of the data analysis products they produce.
+
+I realized I would like my research group to have a similar type of professionalism to our plots since we regularly produce data products and have to illustrate scientific data. 
+
+This is a guide for how plots should be made in the Leek group. I hope it will evolve over time as members of the group weigh in on their opinions. There is a corresponding 
+
+*[Leek group plotting R package](link TBD) 
+
+that you can use to make plots like ours if you want to with both ggplot2 and base R plotting parameters set up. 
+
+## Expository versus exploratory graphs
+
+If you are analyzing data you make plots all of the time. This is part of the interactive data analysis workflow. When exploring data you should not spend time on how the plots look. They should be ugly and fast so you can quickly explore a data set. This guide does not apply to exploratory plots. 
+
+Expository plots are plots that we intend to distribute as part of a paper, blog post, or other communication of our results. Expository plots differ from exploratory plots because they are intended to communicate information to someone who is not you. The key principles behind Leek group expository plots are:
+
+(1) They communicate the answer to a specific scientific question
+(2) Each plot answers a single scientific question
+(3) Each plot will have a figure caption describing the key story in the plot
+(4) The figure and legend are sufficient to communicate a scientific message without the surrounding paper text. 
+(5) They have a consistent color theme, point type, and font. 
+
+Point (4) is directly related to the Leek group [guide to writing the first paper](https://github.com/jtleek/firstpaper)
diff --git a/_images/2017-04-06/IMG_7075.jpg b/_images/2017-04-06/IMG_7075.jpg
diff --git a/_images/2017-04-06/IMG_7076.jpg b/_images/2017-04-06/IMG_7076.jpg
diff --git a/_images/2017-04-06/cambio-en-matricula.png b/_images/2017-04-06/cambio-en-matricula.png
diff --git a/_images/2017-04-06/costo.png b/_images/2017-04-06/costo.png
diff --git a/_images/2017-04-06/matricula.png b/_images/2017-04-06/matricula.png
diff --git a/_images/2017-05-04/haircuts.png b/_images/2017-05-04/haircuts.png
diff --git a/_images/2x2-table-results.png b/_images/2x2-table-results.png
diff --git a/_images/2x2-table.png b/_images/2x2-table.png
diff --git a/_images/Flowchart-full.png b/_images/Flowchart-full.png
diff --git a/_images/Flowchart-partial.png b/_images/Flowchart-partial.png
diff --git a/_images/Flowchart.png b/_images/Flowchart.png
diff --git a/_images/ai-album.png b/_images/ai-album.png
diff --git a/_images/alexa-ai.png b/_images/alexa-ai.png
diff --git a/_images/cartoon-phone-photos.png b/_images/cartoon-phone-photos.png
diff --git a/_images/chromebook2.jpg b/_images/chromebook2.jpg
diff --git a/_images/heights-with-outlier.png b/_images/heights-with-outlier.png
diff --git a/_images/images-to-numbers.png b/_images/images-to-numbers.png
diff --git a/_images/importance-not-size.jpg b/_images/importance-not-size.jpg
diff --git a/_images/jeff-color-names.png b/_images/jeff-color-names.png
diff --git a/_images/jeff-rgb.png b/_images/jeff-rgb.png
diff --git a/_images/jeff-smile-dots.png b/_images/jeff-smile-dots.png
diff --git a/_images/jeff-smile-lines.png b/_images/jeff-smile-lines.png
diff --git a/_images/jeff-smile.png b/_images/jeff-smile.png
diff --git a/_images/jeff.jpg b/_images/jeff.jpg
diff --git a/_images/labels-to-numbers.png b/_images/labels-to-numbers.png
diff --git a/_images/many-workflows.png b/_images/many-workflows.png
diff --git a/_images/movie-ai.png b/_images/movie-ai.png
diff --git a/_images/notajftweet.png b/_images/notajftweet.png
diff --git a/_images/papr.png b/_images/papr.png
diff --git a/_images/pisa-2015-math-v-others.png b/_images/pisa-2015-math-v-others.png
diff --git a/_images/pisa-2015-scatter.png b/_images/pisa-2015-scatter.png
diff --git a/_images/silver3.png b/_images/silver3.png
diff --git a/_images/timeline-ai.png b/_images/timeline-ai.png
diff --git a/_images/us-election-2016-538-prediction.png b/_images/us-election-2016-538-prediction.png
diff --git a/_images/us-election-2016-538-v-upshot.png b/_images/us-election-2016-538-v-upshot.png
diff --git a/_images/ux1.png b/_images/ux1.png
diff --git a/_images/ux2.png b/_images/ux2.png
diff --git a/_images/workflow.png b/_images/workflow.png
diff --git a/_posts/2011-12-03-reverse-scooping.md b/_posts/2011-12-03-reverse-scooping.md
@@ -18,4 +18,4 @@ tags:
   - advice
   - Rant
 ---
-I would like to define a new term: r_everse scooping_ is when someone publishes your idea after you, and doesn&#8217;t cite you. It has happened to me a few times. What does one do? I usually send a polite message to the authors with a link to my related paper(s). These emails are usually ignored, but not always. Most times I don&#8217;t think it is malicious though. In fact, I almost reverse scooped a colleague recently.  People arrive at the same idea a few months (or years) later and there is just too much literature to keep track-off. And remember the culprit authors were not the only ones that missed your paper, the referees and associate editor missed it as well. One thing I have learned is that if you want to claim an idea, try to include it in the title or abstract as very few papers get read cover-to-cover.
+I would like to define a new term: _reverse scooping_ is when someone publishes your idea after you, and doesn&#8217;t cite you. It has happened to me a few times. What does one do? I usually send a polite message to the authors with a link to my related paper(s). These emails are usually ignored, but not always. Most times I don&#8217;t think it is malicious though. In fact, I almost reverse scooped a colleague recently.  People arrive at the same idea a few months (or years) later and there is just too much literature to keep track-off. And remember the culprit authors were not the only ones that missed your paper, the referees and associate editor missed it as well. One thing I have learned is that if you want to claim an idea, try to include it in the title or abstract as very few papers get read cover-to-cover.
diff --git a/_posts/2012-11-07-nate-silver-does-it-again-will-pundits-finally-accept.md b/_posts/2012-11-07-nate-silver-does-it-again-will-pundits-finally-accept.md
@@ -25,6 +25,6 @@ While the pundits were claiming the race was a &#8220;dead heat&#8221;, the day
 
 **Update****: **Congratulations also to Sam Wang (<a href="http://election.princeton.edu/" target="_blank">Princeton Election Consortium</a>) and Simon Jackman (<a href="http://www.huffingtonpost.com/simon-jackman/pollster-predictions_b_2081013.html" target="_blank">pollster</a>) that also called the election perfectly. And thanks to the pollsters that provided the unbiased (on average) data used by all these folks. Data analysts won &#8220;experts&#8221; lost.
 
-**Update 2**: New plot with data from <a href="http://www.foxnews.com/politics/elections/2012-election-results/" target="_blank">here</a>. Old graph <a href="http://rafalab.jhsph.edu/simplystats/silver.png" target="_blank">here</a>.
+~~**Update 2**: New plot with data from <a href="http://www.foxnews.com/politics/elections/2012-election-results/" target="_blank">here</a>. Old graph <a href="http://rafalab.jhsph.edu/simplystats/silver.png" target="_blank">here</a>.~~
 
-<a href="http://rafalab.jhsph.edu/simplystats/silver2.png" target="_blank"><img height="500" src="http://rafalab.jhsph.edu/simplystats/silver2.png" width="500" /></a>
+![Observed versus predicted](https://raw.githubusercontent.com/simplystats/simplystats.github.io/master/_images/silver3.png)
diff --git a/...n-i-find-the-frequentists-versus-bayesians-debate-completely-inconsequential.md b/...n-i-find-the-frequentists-versus-bayesians-debate-completely-inconsequential.md
@@ -35,36 +35,34 @@ In a recent New York Times [article](http://www.nytimes.com/2014/09/30/science/t
 
 
 
-Because the real story (or non-story) is way too boring to sell newspapers, the author resorted to a sensationalist narrative that went something like this:  "Evil and/or stupid frequentists were ready to let a fisherman die; the persecuted Bayesian heroes saved him." This piece adds to the growing number of writings blaming frequentist statistics for the so-called reproducibility crisis in science. If there is something Roger, [Jeff](http://simplystatistics.org/2013/11/26/statistical-zealots/) <a>and </a>[I](http://simplystatistics.org/2013/08/01/the-roc-curves-of-science/) <a>agree on is that this debate is </a>[not constructive](http://noahpinionblog.blogspot.com/2013/01/bayesian-vs-frequentist-is-there-any.html). As </a>[Rob Kass](http://arxiv.org/pdf/1106.2895v2.pdf) <a>suggests it's time to move on to pragmatism. Here I follow up Jeff's <a href="http://simplystatistics.org/2014/09/30/you-think-p-values-are-bad-i-say-show-me-the-data/">recent post</a> by sharing related thoughts brought about by two decades of practicing applied statistics and hope it helps put this unhelpful debate to rest.</p> 
+Because the real story (or non-story) is way too boring to sell newspapers, the author resorted to a sensationalist narrative that went something like this:  "Evil and/or stupid frequentists were ready to let a fisherman die; the persecuted Bayesian heroes saved him." This piece adds to the growing number of writings blaming frequentist statistics for the so-called reproducibility crisis in science. If there is something Roger, [Jeff](http://simplystatistics.org/2013/11/26/statistical-zealots/) and [I](http://simplystatistics.org/2013/08/01/the-roc-curves-of-science/) agree on is that this debate is [not constructive](http://noahpinionblog.blogspot.com/2013/01/bayesian-vs-frequentist-is-there-any.html). As [Rob Kass](http://arxiv.org/pdf/1106.2895v2.pdf) suggests it's time to move on to pragmatism. Here I follow up Jeff's [recent post](http://simplystatistics.org/2014/09/30/you-think-p-values-are-bad-i-say-show-me-the-data/) by sharing related thoughts brought about by two decades of practicing applied statistics and hope it helps put this unhelpful debate to rest. 
+
 
-<p>
   Applied statisticians help answer questions with data. How should I design a roulette so my casino makes $? Does this fertilizer increase crop yield? Does streptomycin cure pulmonary tuberculosis? Does smoking cause cancer? What movie would would this user enjoy? Which baseball player should the Red Sox give a contract to? Should this patient receive chemotherapy? Our involvement typically means analyzing data and designing experiments. To do this we use a variety of techniques that have been successfully applied in the past and that we have mathematically shown to have desirable properties. Some of these tools are frequentist, some of them are Bayesian, some could be argued to be both, and some don't even use probability. The Casino will do just fine with frequentist statistics, while the baseball team might want to apply a Bayesian approach to avoid overpaying for players that have simply been lucky.
-</p>
 
-<p>
-  It is also important to remember that good applied statisticians also *think*. They don't apply techniques blindly or religiously. If applied statisticians, regardless of their philosophical bent, are asked if the sun just exploded, they would not design an experiment as the one depicted in this popular XKCD cartoon.
-</p>
 
-<p>
+
+  It is also important to remember that good applied statisticians also **think**. They don't apply techniques blindly or religiously. If applied statisticians, regardless of their philosophical bent, are asked if the sun just exploded, they would not design an experiment as the one depicted in this popular XKCD cartoon.
+
+
+
   <a href="http://xkcd.com/1132/"><img class="aligncenter" src="http://imgs.xkcd.com/comics/frequentists_vs_bayesians.png" alt="" width="234" height="355" /></a>
-</p>
 
-<p>
+
+
   Only someone that does not know how to think like a statistician would act like the frequentists in the cartoon. Unfortunately we do have such people analyzing data. But their choice of technique is not the problem, it's their lack of critical thinking. However, even the most frequentist-appearing applied statistician understands Bayes rule and will adapt the Bayesian approach when appropriate. In the above XCKD example, any respectful applied statistician would not even bother examining the data (the dice roll), because they would assign a probability of 0 to the sun exploding (the empirical prior based on the fact that they are alive). However, superficial propositions arguing for wider adoption of Bayesian methods fail to realize that using these techniques in an actual data analysis project is very different from simply thinking like a Bayesian. To do this we have to represent our intuition or prior knowledge (or whatever you want to call it) with mathematical formulae. When theoretical Bayesians pick these priors, they mainly have mathematical/computational considerations in mind. In practice we can't afford this luxury: a bad prior will render the analysis useless regardless of its convenient mathematically properties.
-</p>
 
-<p>
-  Despite these challenges, applied statisticians regularly use Bayesian techniques successfully. In one of the fields I work in, Genomics, empirical Bayes techniques are widely used. In <a href="http://www.ncbi.nlm.nih.gov/pubmed/16646809">this</a> popular application of empirical Bayes we use data from all genes to improve the precision of estimates obtained for specific genes. However, the most widely used output of the software implementation is not a posterior probability. Instead, an empirical Bayes technique is used to improve the estimate of the standard error used in a good ol' fashioned t-test. This idea has changed the way thousands of Biologists search for differential expressed genes and is, in my opinion, one of the most important contributions of Statistics to Genomics. Is this approach frequentist? Bayesian? To this applied statistician it doesn't really matter.
-</p>
 
-<p>
-  For those arguing that simply switching to a Bayesian philosophy will improve the current state of affairs, let's consider the smoking and cancer example. Today there is wide agreement that smoking causes lung cancer. Without a clear deductive biochemical/physiological argument and without<br /> the possibility of a randomized trial, this connection was established with a series of observational studies. Most, if not all, of the associated data analyses were based on frequentist techniques. None of the reported confidence intervals on their own established the consensus. Instead, as usually happens in science, a long series of studies supporting this conclusion were needed. How exactly would this have been different with a strictly Bayesian approach? Would a single paper been enough? Would using priors helped given the "expert knowledge" at the time (see below)?
-</p>
 
-<p>
+  Despite these challenges, applied statisticians regularly use Bayesian techniques successfully. In one of the fields I work in, Genomics, empirical Bayes techniques are widely used. In [this](http://www.ncbi.nlm.nih.gov/pubmed/16646809) popular application of empirical Bayes we use data from all genes to improve the precision of estimates obtained for specific genes. However, the most widely used output of the software implementation is not a posterior probability. Instead, an empirical Bayes technique is used to improve the estimate of the standard error used in a good ol' fashioned t-test. This idea has changed the way thousands of Biologists search for differential expressed genes and is, in my opinion, one of the most important contributions of Statistics to Genomics. Is this approach frequentist? Bayesian? To this applied statistician it doesn't really matter.
+
+
+
+  For those arguing that simply switching to a Bayesian philosophy will improve the current state of affairs, let's consider the smoking and cancer example. Today there is wide agreement that smoking causes lung cancer. Without a clear deductive biochemical/physiological argument and without the possibility of a randomized trial, this connection was established with a series of observational studies. Most, if not all, of the associated data analyses were based on frequentist techniques. None of the reported confidence intervals on their own established the consensus. Instead, as usually happens in science, a long series of studies supporting this conclusion were needed. How exactly would this have been different with a strictly Bayesian approach? Would a single paper been enough? Would using priors helped given the "expert knowledge" at the time (see below)?
+
+
+
   <img src="http://cdn.saveourbones.com/wp-content/uploads/smoking_doctor.jpg" width="234" height="355" class="aligncenter" alt="" />
-</p>
 
-<p>
+
   And how would the Bayesian analysis performed by tabacco companies shape the debate? Ultimately, I think applied statisticians would have made an equally convincing case against smoking with Bayesian posteriors as opposed to frequentist confidence intervals. Going forward I hope applied statisticians continue to be free to use whatever techniques they see fit and that critical thinking about data continues to be what distinguishes us. Imposing Bayesian or frequentists philosophy on us would be a disaster.
-</p>
diff --git a/_posts/2014-11-04-538-election-forecasts-made-simple.md b/_posts/2014-11-04-538-election-forecasts-made-simple.md
@@ -21,10 +21,10 @@ categories:
 ---
 Nate Silver does a [great job](http://fivethirtyeight.com/features/how-the-fivethirtyeight-senate-forecast-model-works/) of explaining his forecast model to laypeople. However, as a statistician I've always wanted to know more details. After preparing a "<span class="s2"><a href="http://cs109.github.io/2014/pages/homework.html">predict the midterm elections</a>" </span>homework for my [<span class="s2">data science class</span>](http://cs109.github.io/2014) I have a better idea of what is going on.
 
-[Here](http://rafalab.jhsph.edu/simplystats/midterm2012.html) is my best attempt at explaining the ideas of 538 using formulas and data. And [here](http://rafalab.jhsph.edu/simplystats/midterm2012.Rmd) is the R markdown.
+[Here](http://simplystatistics.org/html/midterm2012.html) is my best attempt at explaining the ideas of 538 using formulas and data. ~~And [here](http://rafalab.jhsph.edu/simplystats/midterm2012.Rmd) is the R markdown.~~
 
 &nbsp;
 
 &nbsp;
 
-&nbsp;
+&nbsp;
-Original file line number
+Diff line change
@@ Expand Up / @@ -18,4 +18,4 @@ tags: @@
       - advice
       - Rant
     ---
-    I would like to define a new term: r_everse scooping_ is when someone publishes your idea after you, and doesn&#8217;t cite you. It has happened to me a few times. What does one do? I usually send a polite message to the authors with a link to my related paper(s). These emails are usually ignored, but not always. Most times I don&#8217;t think it is malicious though. In fact, I almost reverse scooped a colleague recently.  People arrive at the same idea a few months (or years) later and there is just too much literature to keep track-off. And remember the culprit authors were not the only ones that missed your paper, the referees and associate editor missed it as well. One thing I have learned is that if you want to claim an idea, try to include it in the title or abstract as very few papers get read cover-to-cover.
+    I would like to define a new term: _reverse scooping_ is when someone publishes your idea after you, and doesn&#8217;t cite you. It has happened to me a few times. What does one do? I usually send a polite message to the authors with a link to my related paper(s). These emails are usually ignored, but not always. Most times I don&#8217;t think it is malicious though. In fact, I almost reverse scooped a colleague recently.  People arrive at the same idea a few months (or years) later and there is just too much literature to keep track-off. And remember the culprit authors were not the only ones that missed your paper, the referees and associate editor missed it as well. One thing I have learned is that if you want to claim an idea, try to include it in the title or abstract as very few papers get read cover-to-cover.