Designs of two variables: map, dot plot, line chart, table

The New York Times found evidence that the richest segments of New Yorkers, presumably those with second or multiple homes, have exited the Big Apple during the early months of the pandemic. The article (link) is amply assisted by a variety of data graphics.

The first few charts represent different attempts to express the headline message. Their appearance in the same article allows us to assess the relative merits of different chart forms.

First up is the always-popular map.

The advantage of a map is its ease of comprehension. We can immediately see which neighborhoods experienced the greater exoduses. Clearly, Manhattan has cleared out a lot more than outer boroughs.

The limitation of the map is also in view. With the color gradient dedicated to the proportions of residents gone on May 1st, there isn't room to express which neighborhoods are richer. We have to rely on outside knowledge to make the correlation ourselves.

The second attempt is a dot plot.

We may have to take a moment to digest the horizontal axis. It's not time moving left to right but income percentiles. The poorest neighborhoods are to the left and the richest to the right. I'm assuming that these percentiles describe the distribution of median incomes in neighborhoods. Typically, when we see income percentiles, they are based on households, regardless of neighborhoods. (The former are equal-sized segments, unlike the latter.)

This data graphic has the reverse features of the map. It does a great job correlating the drop in proportion of residents at home with the income distribution but it does not convey any spatial information. The message is clear: The residents in the top 10% of New York neighborhoods are much more likely to have left town.

In the following chart, I attempted a different labeling of both axes. It cuts out the need for readers to reverse being home to not being home, and 90th percentile to top 10%.

The third attempt to convey the income--exit relationship is the most successful in my mind. This is a line chart, with time on the horizontal axis.

The addition of lines relegates the dots to the background. The lines show the trend more clearly. If directly translated from the dot plot, this line chart should have 100 lines, one for each percentile. However, the closeness of the top two lines suggests that no meaningful difference in behavior exists between the 20th and 80th percentiles. This can be conveyed to readers through a short note. Instead of displaying all 100 percentiles, the line chart selectively includes only the 99th , 95th, 90th, 80th and 20th percentiles. This is a design choice that adds by subtraction.

Along the time axis, the line chart provides more granularity than either the map or the dot plot. The exit occurred roughly over the last two weeks of March and the first week of April. The start coincided with New York's stay-at-home advisory.

This third chart is a statistical graphic. It does not bring out the raw data but features aggregated and smoothed data designed to reveal a key message.

I encourage you to also study the annotated table later in the article. It shows the power of a well-designed table.

[P.S. 6/4/2020. On the book blog, I have just published a post about the underlying surveillance data for this type of analysis.]

 

 

Consumption patterns during the pandemic

The impact of Covid-19 on the economy is sharp and sudden, which makes for some dramatic data visualization. I enjoy reading the set of charts showing consumer spending in different categories in the U.S., courtesy of Visual Capitalist.

The designer did a nice job cleaning up the data and building a sequential story line. The spending are grouped by categories such as restaurants and travel, and then sub-categories such as fast food and fine dining.

Spending is presented as year-on-year change, smoothed.

Here is the chart for the General Commerce category:

The visual design is clean and efficient. Even too sparse because one has to keep returning to the top to decipher the key events labelled 1, 2, 3, 4. Also, to find out that the percentages express year-on-year change, the reader must scroll to the bottom, and locate a footnote.

As you move down the page, you will surely make a stop at the Food Delivery category, noting that the routine is broken.

I've featured this device - an element of surprise - before. Remember this Quartz chart that depicts drinking around the world (link).

The rule for small multiples is to keep the visual design identical but vary the data from chart to chart. Here, the exceptional data force the vertical axis to extend tremendously.

This chart contains a slight oversight - the red line should be labeled "Takeout" because food delivery is the label for the larger category.

Another surprise is in store for us in the Travel category.

I kept staring at the Cruise line, and how it kept dipping below -100 percent. That seems impossible mathematically - unless these cardholders are receiving more refunds than are making new bookings. Not only must the entire sum of 2019 bookings be wiped out, but the records must also show credits issued to these credit (or debit) cards. It's curious that the same situation did not befall the airlines. I think many readers would have liked to see some text discussing this pattern.

***

Now, let me put on a data analyst's hat, and describe some thoughts that raced through my head as I read these charts.

Data analysis is hard, especially if you want to convey the meaning of the data.

The charts clearly illustrate the trends but what do the data reveal? The designer adds commentary on each chart. But most of these comments count as "story time." They contain speculation on what might be causing the trend but there isn't additional data or analyses to support the storyline. In the General Commerce category, the 50 to 100 percent jump in all subcategories around late March is attributed to people stockpiling "non-perishable food, hand sanitizer, and toilet paper". That might be true but this interpretation isn't supported by credit or debit card data because those companies do not have details about what consumers purchased, only the total amount charged to the cards. It's a lot more work to solidify these conclusions.

A lot of data do not mean complete or unbiased data.

The data platform provided data on 5 million consumers. We don't know if these 5 million consumers are representative of the 300+ million people in the U.S. Some basic demographic or geographic analysis can help establish the validity. Strictly speaking, I think they have data on 5 million card accounts, not unique individuals. Most Americans use more than one credit or debit cards. It's not likely the data vendor have a full picture of an individual's or a family's spending.

It's also unclear how much of consumer spending is captured in this dataset. Credit and debit cards are only one form of payment.

Data quality tends to get worse.

One thing that drives data analyst nuts. The spending categories are becoming blurrier. In the last decade or so, big business has come to dominate the American economy. Big business, with bipartisan support, has grown by (a) absorbing little guys, and (b) eliminating boundaries between industry sectors. Around me, there is a Walgreens, several Duane Reades, and a RiteAid. They currently have the same owner, and increasingly offer the same selection. In the meantime, Walmart (big box), CVS (pharmacy), Costco (wholesale), etc. all won regulatory relief to carry groceries, fresh foods, toiletries, etc. So, while CVS or Walgreens is classified as a pharmacy, it's not clear that what proportion of the spending there is for medicines. As big business grows, these categories become less and less meaningful.

Hope and reality in one Georgia chart

Over the weekend, Georgia's State Health Department agitated a lot of people when it published the following chart:

(This might have appeared a week ago as the last date on the chart is May 9 and the title refers to "past 15 days".)

They could have avoided the embarrassment if they had read my article at DataJournalism.com (link). In that article, I lay out a set of the "unspoken conventions," things that visual designers are, or should be, doing more or less in their sleep. Under the section titled "Order", I explain the following two "rules":

• Place values in the natural order when it is available
• Retain the same order across all plots in a panel of charts

In the chart above, the natural order for the horizontal (time) axis is time running left to right. The order chosen by the designer  is roughly but not precisely decreasing height of the tallest column in each daily group. Many observers suggested that the columns were arranged to give the appearance of cases dropping over time.

Within each day, the counties are ordered in decreasing number of new cases. The title of the chart reads "number of cases over time" which sounds like cumulative cases but it's not. The "lead" changed hands so many times over the 15 days, meaning the data sequence was extremely noisy, which would be unlikely for cumulative cases. There are thousands of cases in each of these counties by May. Switching the order of the columns within each daily group defeats the purpose of placing these groups side-by-side.

Responding to the bad press, the department changed the chart design for this week's version:

This chart now conforms to the two spoken rules described above. The time axis runs left to right, and within each group of columns, the order of the counties is maintained.

The chart is still very noisy, with no apparent message.

***

Next, I'd like to draw your attention to a Data issue. Notice that the 15-day window has shifted. This revised chart runs from May 2 to May 16, which is this past Saturday. The previous chart ran from Apr 26 to May 9. 

Here's the data for May 8 and 9 placed side by side.

There is a clear time lag of reporting cases in the State of Georgia. This chart should always exclude the last few days. The case counts keep going up until it stabilizes. The same mistake occurs in the revised chart - the last two days appear as if new cases have dwindled toward zero when in fact, it reflects a lag in reporting.

The disconnect between the Question being posed and the quality of the Data available dooms this visualization. It is not possible to provide a reliable assessment of the "past 15 days" when during perhaps half of that period, the cases are under-counted.

***

This graphical distortion due to "immature" data has become very commonplace in Covid-19 graphics. It's similar to placing partial-year data next to full-year results, without calling out the partial data.

The following post from the ancient past (2005!) about a New York Times graphic shows that calling out this data problem does not actually solve it. It's a less-bad kind of thing.

The coronavirus data present more headaches for graphic designers than the financial statistics. Because of accounting regulations, we know that only the current quarter's data are immature. For Covid-19 reporting, the numbers are being adjusted for days and weeks.

Practically all immature counts are under-estimates. Over time, more cases are reported. Thus, any plots over time - if unadjusted - paint a misleading picture of declining counts. The effect of the reporting lag is predictable, having a larger impact as we run from left to right in time. Thus, even if the most recent data show a downward trend, it can eventually mean anything: down, flat or up. This is not random noise though - we know for certain of the downward bias; we just don't know the magnitude of the distortion for a while.

Another issue that concerns coronavirus reporting but not financial reporting is inconsistent standards across counties. Within a business, if one were to break out statistics by county, the analysts would naturally apply the same counting rules. For Covid-19 data, each county follows its own set of rules, not just  how to count things but also how to conduct testing, and so on.

Finally, with the politics of re-opening, I find it hard to trust the data. Reported cases are human-driven data - by changing the number of tests, by testing different mixes of people, by delaying reporting, by timing the revision of older data, by explicit manipulation, ...., the numbers can be tortured into any shape. That's why it is extremely important that the bean-counters are civil servants, and that politicians are kept away. In the current political environment, that separation between politics and statistics has been breached.

***

Why do we have low-quality data? Human decisions, frequently political decisions, adulterate the data. Epidemiologists are then forced to use the bad data, because that's what they have. Bad data lead to bad predictions and bad decisions, or if the scientists account for the low quality, predictions with high levels of uncertainty. Then, the politicians complain that predictions are wrong, or too wide-ranging to be useful. If they really cared about those predictions, they could start by being more transparent about reporting and more proactive at discovering and removing bad accounting practices. The fact that they aren't focused on improving the data gives the game away. Here's a recent post on the politics of data.

 

The missing 100 million: how the pandemic reveals the fallacy of not in labor force

Last Friday, the U.S. published the long-feared employment situation report. It should come as no surprise to anyone since U.S. businesses were quick to lay off employees since much of the economy was shut down to abate the spread of the coronavirus.

I've been following employment statistics for a while. Chapter 6 of Numbersense (link) addresses the statistical aspects of how the unemployment rate is computed. The title of the chapter is "Are they new jobs when no one can apply?" What you learn is that the final number being published starts off as survey tallies, which then undergo a variety of statistical adjustments.

One such adjustment - which ought to be controversial - results in the disappearance of 100 million Americans. I mean, that they are invisible to the Bureau of Labor Statistics (BLS), considered neither employed nor unemployed. You don't hear about them because the media report the "headline" unemployment rate, which excludes these people. They are officially designated "not in the labor force". I'll come back to this topic later in the post.

***

Last year, I used a pair of charts to visualize the unemployment statistics. I have updated the charts to include all of 2019 and 2020 up to April, the just released numbers.

The first chart shows the trend in the official unemployment rate ("U3") from 1990 to present. It's color-coded so that the periods of high unemployment are red, and the periods of low unemployment are blue. This color code will come in handy for the next chart.

The time series is smoothed. However, I had to exclude the April 2020 outlier from the smoother.

The next plot, a scatter plot, highlights two of the more debatable definitions used by the BLS. On the horizontal axis, I plot the proportion of employed people who have part-time jobs. People only need to have worked one hour in a month to be counted as employed. On the vertical axis, I plot the proportion of the population who are labeled "not in labor force". These are people who are not employed and not counted in the unemployment rate.

The value of data visualization is its ability to reveal insights about the data. I'm happy to report that this design succeeded.

Previously, we learned that (a) part-timers as a proportion of employment tend to increase during periods of worsening unemployment (red dots moving right) while decreasing during periods of improving employment (blue dots moving left); and (b) despite the overall unemployment rate being about the same in 2007 and 2017, the employment situation was vastly different in the sense that the labor force has shrunk significantly during the recession and never returned to normal. These two insights are still found at the bottom right corner of the chart. The 2019 situation did not differ much from 2018.

What is the effect of the current Covid-19 pandemic?

On both dimensions, we have broken records since 1990. The proportion of people designated not in labor force was already the worst in three decades before the pandemic, and now it has almost reached 40 percent of the population!

Remember these people are invisible to the media, neither employed nor unemployed. Back in February 2020, with unemployment rate at around 4 percent, it's absolutely not the case that 96 pecent of the employment-age population was employed. The number of employed Americans was just under 160 million. The population 16 years and older at the time was 260 million.

Who are these 100 million people? BLS says all but 2 million of these are people who "do not want a job". Some of them are retired. There are about 50 million Americans above 65 years old although 25 percent of them are still in the labor force, so only 38 million are "not in labor force," according to this Census report.

It would seem like the majority of these people don't want to work, are not paid enough to work, etc. Since part-time workers are counted as employed, with as little as one working hour per month, these are not the gig workers, not Uber/Lyft drivers, and not college students who has work-study or part-time jobs.

This category has long been suspect, and what happened in April isn't going to help build its case. There is no reason why the "not in labor force" group should spike immediately as a result of the pandemic. It's not plausible to argue that people who lost their jobs in the last few weeks suddenly turned into people who "do not want a job". I think this spike is solid evidence that the unemployed have been hiding inside the not in labor force number.

The unemployment rate has under-reported unemployment because many of the unemployed have been taken out of the labor force based on BLS criteria. The recovery of jobs since the Great Recession is partially nullified since the jump in "not in labor force" never returned to the prior level.

***

The other dimension, part-time employment, also showed a striking divergence from the past behavior. Typically, when the unemployment rate deteriorates, the proportion of employed people who have part-time jobs increases. However, in the current situation, not only is that not happening, but the proportion of part-timers plunged to a level not seen in the last 30 years.

This suggests that employers are getting rid of their part-time work force first.

 

 

How Covid-19 deaths sneaked into Florida's statistics

Like many others, some Floridians are questioning their state's Covid statistics. It's clear there are numerous "degrees of freedom" for politicians to manipulate the numbers. What's not clear is who's influencing these decisions. Are they public-health experts, donors, voters, or whom?

A Twitter follower sent in the following chart, embedded in an informative article in Sun-Sentinel:

I like the visual design. It's clean, and conveys a moderately complex concept effectively. The reader may not immediately get what metrics are being plotted but the idea that the blue line should operate within the gray area.. until it doesn't is easily grasped. The range is technically an uncertainty band.

The metric is the proportion of total deaths (all causes) that are attributed to pneumonia and flu. Typical influenza deaths are found in that category. This chart investigates whether there were excess (unexplained) P&F deaths. The gray band measures the variability in the proportions of past years. When the blue line operates inside the band, the metric is normal. When it pierces the upper band, which happened here around week 25, a rare event has occurred.

The concern on Twitter was about the horizontal axis. Those integer labels can be confusing. The designer places a "how to read this" message in a footnote, explaining that week 1 is the first week of a typical flu season (which corresponds to late September 2019). This nugget of information helps a lot. We can see that the flu season peaks around week 20, and by the spring, it should be waning. Not so in 2020.

It's hard to escape the conclusion that deaths from Covid-19 are hiding inside the statistics of Pneumonia & Flu. As a statistician, I want to tell you Statistics Don't Lie! You can hide the data along one dimension, but they show up elsewhere. Misclassifying the deaths does buy someone some time. It takes a few weeks to compile all-cause mortality data (gasp, the CDC said mortality records are only 75 percent accurate after 8 weeks!)

The other small problem with the chart is the labeling. Neither axis has labels. The data label that shows up when you click on the line might be a default from the software that can't be turned off. It shows the two numbers being plotted without labels.

***

Here is a re-working of the chart that tells the story:

The proportion of deaths attributed to P&F and Covid together is roughly double the upper end of what Florida should be seeing this time of the year (without Covid). Covid-19 accounts for half the gap. The other half are still being classified as P&F. However, I suspect CDC will adjust these numbers later to reflect the reality. (In making this chart, I also learned that Florida stopped including seasonal visitors in the death counts. This is egregious manipulation. If someone died while in Florida, they should be counted. I didn't investigate whether this counting rule applies only to Covid-19 deaths, or to deaths from all causes. If they had always done that, then I might give them a pass.)

On second thought, maybe not. The other egregious thing that appeared to have happened is that the Florida state health department unplugged their prior website (https://www.floridahealth.gov) so no one can cross-reference any prior documents. The only website I can access now for Florida state health is a Covid-specific site (https://floridahealthcovid19.gov).

There must be something juicy on the previous influenza page, no?

***

Lastly, when you look at my chart, please pretend that the last week is not on there. In all likelihood, the "drop" is fake because the mortality data have not been fully updated. My chart contains one more week than the Sun Sentinel chart. So you can see that the drastic decline shown on their chart turned up a big uptick on mine (next to last week).

This is a common mistake on many charts I see these days. Half-baked numbers are shown next to fully-baked ones.

Proportions and rates: we are no dupes

Reader Lucia G. sent me this chart, from Ars Technica's FAQ about the coronavirus:

She notices something wrong with the axis.

The designer took the advice not to make a dual axis, but didn't realize that the two metrics are not measured on the same scale even though both are expressed as percentages.

The blue bars, labeled "cases", is a distribution of cases by age group. The sum of the blue bars should be 100 percent.

The orange bars show fatality rates by age group. Each orange bar's rate is based on the number of cases in that age group. The sum of the orange bars will not add to 100 percent.

In general, the rates will have much lower values than the proportions. At least that should be the case for viruses that are not extremely fatal.

This is what the 80 and over section looks like.

It is true that fatality rate (orange) is particularly high for the elderly while this age group accounts for less than 5 percent of total cases (blue). However, the cases that are fatal, which inhabit the orange bar, must be a subset of the total cases for 80 and over, which are shown in the blue bar. Conceptually, the orange bar should be contained inside the blue bar. So, it's counter-intuitive that the blue bar is so much shorter than the orange bar.

The following chart fixes this issue. It reveals the structure of the data, Total cases are separated by age group, then within each age group, a proportion of the cases are fatal.

This chart also shows that most patients recover in every age group. (This is only approximately true as some of the cases may not have been discharged yet.)

***

This confusion of rates and proportions reminds me of something about exit polls I just wrote about the other day on the sister blog.

When the media make statements about trends in voter turnout rate in the primary elections, e.g. when they assert that youth turnout has not increased, their evidence is from exit polls, which can measure only the distribution of voters by age group. Exit polls do not and cannot measure the turnout rate, which is the proportion of registered (or eligible) voters in the specific age group who voted.

Like the coronavirus data, the scales of these two metrics are different even though they are both percentages: the turnout rate is typically a number between 30 and 70 percent, and summing the rates across all age groups will exceed 100 percent many times over. Summing the proportions of voters across all age groups should be 100 percent, and no more.

Changes in the proportion of voters aged 18-29 and changes in the turnout rate of people aged 18-29 are not the same thing. The former is affected by the turnout of all age groups while the latter is a clean metric affected only by 18 to 29-years-old.

Basically, ignore pundits who use exit polls to comment on turnout trends. No matter how many times they repeat their nonsense, proportions and rates are not to be confused. Which means, ignore comments on turnout trends because the only data they've got come from exit polls which don't measure rates.

 

P.S. Here is some further explanation of my chart, as a response to a question from Enrico B. on Twitter.

The chart can be thought of as two distributions, one for cases (gray) and one for deaths (red). Like this:

The side-by-side version removes the direct visualization of the fatality rate within each age group. To understand fatality rate requires someone to do math in their head. Readers can qualitatively assess that for the 80 and over, they accounted for 3 percent of cases but also about 21 percent of deaths. People aged 70 to 79 however accounted for 9 percent of cases but 30 percent of deaths, etc.

What I did was to scale the distribution of deaths so that they can be compared to the cases. It's like fitting the red distribution inside the gray distribution. Within each age group, the proportion of red against the length of the bar is the fatality rate.

For every 100 cases regardless of age, 3 cases are for people aged 80 and over within which 0.5 are fatal (red).

So, the axis labels are correct. The values are proportions of total cases, although as the designer of the chart, I hope people are paying attention more to the proportion of red, as opposed to the units.

What might strike people as odd is that the biggest red bar does not appear against 80 and above. We might believe it's deadlier the older you are. That's because on an absolute scale, more people aged 70-79 died than those 80 and above. The absolute deaths is the product of the proportion of cases and the fatality rate. That's really a different story from the usual plot of fatality rates by age group. In those charts, we "control" for the prevalence of cases. If every age group were infected in the same frequency, then COVID-19 does kill more 80 and over.

 

 

 

The rule governing which variable to put on which axis, served a la mode

When making a scatter plot, the two variables should not be placed arbitrarily. There is a rule governing this: the outcome variable should be shown on the vertical axis (also called y-axis), and the explanatory variable on the horizontal (or x-) axis.

This chart from the archives of the Economist has this reversed:

The title of the accompanying article is "Ice Cream and IQ"...

In a Trifecta Checkup (link), it's a Type DV chart. It's preposterous to claim eating ice cream makes one smarter without more careful studies. The chart also carries the xyopia fallacy: by showing just two variables, readers are unwittingly led to explain differences in "IQ" using differences in per-capita ice-cream consumption when lots of other stronger variables will explain any gaps in IQ.

In this post, I put aside my objections to the analysis, and focus on the issue of assigning variables to axes. Notice that this chart reverses the convention: the outcome variable (IQ) is shown on the horizontal, and the explanatory variable (ice cream) is shown on the vertical.

Here is a reconstruction of the above chart, showing only the dots that were labeled with country names. I fitted a straight regression line instead of a curve. (I don't understand why the red line in the original chart bends upwards when the data for Japan, South Korea, Singapore and Hong Kong should be dragging it down.)

Note that the interpretation of the regression line raises eyebrows because the presumed causality is reversed. For each 50 points increase in PISA score (IQ), this line says to expect ice cream consumption to raise by about 1-2 liters per person per year. So higher IQ makes people eat more ice cream.

***

If the convention is respected, then the following scatter plot results:

The first thing to note is that the regression analysis is different here from that shown in the previous chart. The blue regression line is not equivalent to the black regression line from the previous chart. You cannot reverse the roles of the x and y variables in a regression analysis, and so neither should you reverse the roles of the x and y variables in a scatter plot.

The blue regression line can be interpreted as having two sections, roughly, for countries consuming more than or less than 6 liters of ice cream per person per year. In the less-ice-cream countries, the correlation between ice cream and IQ is stronger (I don't endorse the causal interpretation of this statement).

***

When you make a scatter plot, you have two variables for which you want to analyze their correlation. In most cases, you are exploring a cause-effect relationship.

Higher income households cares more on politics.
Less educated citizens are more likely to not register to vote.
Companies with more diverse workforce has better business performance.

Frequently, the reverse correlation does not admit a causal interpretation:

Caring more about politics does not make one richer.
Not registering to vote does not make one less educated.
Making more profits does not lead to more diversity in hiring.

In each of these examples, it's clear that one variable is the outcome, the other variable is the explanatory factor. Always put the outcome in the vertical axis, and the explanation in the horizontal axis.

The justification is scientific. If you are going to add a regression line (what Excel calls a "trendline"), you must follow this convention, otherwise, your regression analysis will yield the wrong result, with an absurd interpretation!

 

[PS. 11/3/2019: The comments below contain different theories that link the two variables, including theories that treat PISA score ("IQ") as the explanatory variable and ice cream consumption as the outcome. Also, I elaborated that the rule does not dictate which variable is the outcome - the designer effectively signals to the reader which variable is regarded as the outcome by placing it in the vertical axis.]

Statistical significance explainer, and Instagram's experiment to hide Likes

There are some statistical concepts that all data visualization practitioners should know about, and the concept of statistical significance is one of them.

It's a hard concept to grasp because it requires one to think beyond the data that are collected. The abstract thinking is necessary since we typically want to make general statements - while using the collected data as evidence.

My new video in the Data Science: The Missing Pieces series explains statistical significance. To be precise, it explains NOT statistically significant. When something is not significant, it causes all sorts of anxieties, panics, half-measures, re-examinations, and havoc. Much of the time, the result is confusion and misinterpretation.

The video addresses a recent news item - Instagram's experiment to hide the Like count. See for example this article. After running this experiment, Instagram's analysts will look for statistical significance. If the result is NOT significant, what does it mean?

Check out the video for more.

 

***
Placed here to serve the machine:

Announcement: Advancing your data skills, Fall 2019

Interrupting the flow of dataviz with the following announcement.

If you're looking to shore up your data skills, modernize your skill set, or know someone looking for hands-on, high-touch instruction in Machine Learning, R, Cloud Computing, Data Quality, Digital Analytics,  A/B Testing and Financial Analysis, Principal Analytics Prep is offering evening classes this Fall. Click here to learn about our courses. 

Our instructors are industry veterans with 10+ years of practical industry experience. And class size is capped to 10, ensuring a high-touch learning environment.

 

Choosing between individuals and aggregates

Friend/reader Thomas B. alerted me to this paper that describes some of the key chart forms used by cancer researchers.

It strikes me that many of the "new" charts plot granular data at the individual level. This heatmap showing gene expressions show one column per patient:

This so-called swimmer plot shows one bar per patient:

This spider plot shows the progression of individual patients over time. Key events are marked with symbols.

These chart forms are distinguished from other ones that plot aggregated statistics: statistical averages, medians, subgroup averages, and so on.

One obvious limitation of such charts is their lack of scalability. The number of patients, the variability of the metric, and the timing of trends all drive up the amount of messiness.

I am left wondering what Question is being addressed by these plots. If we are concerned about treatment of an individual patient, then showing each line by itself would be clearer. If we are interested in the average trends of patients, then a chart that plots the overall average, or subgroup averages would be more accurate. If the interpretation of the individual's trend requires comparing with similar patients, then showing that individual's line against the subgroup average would be preferred.

When shown these charts of individual lines, readers are tempted to play the statistician - without using appropriate tools! Readers draw aggregate conclusions, performing the aggregation in their heads.

The authors of the paper note: "Spider plots only provide good visual qualitative assessment but do not allow for formal statistical inference." I agree with the second part. The first part is a fallacy - if the visual qualitative assessment is good enough, then no formal inference is necessary! The same argument is often made when people say they don't need advanced analysis because their simple analysis is "directionally accurate". When is something "directionally inaccurate"? How would one know?

Reference: Chia, Gedye, et. al., "Current and Evolving Methods to Visualize Biological Data in Cancer Research", JNCI, 2016, 108(8). (link)

***

Meteoreologists, whom I featured in the previous post, also have their own spider-like chart for hurricanes. They call it a spaghetti map:

Compare this to the "cone of uncertainty" map that was featured in the prior post:

These two charts build upon the same dataset. The cone map, as we discussed, shows the range of probable paths of the storm center, based on all simulations of all acceptable models for projection. The spaghetti map shows selected individual simulations. Each line is the most likely trajectory of the storm center as predicted by a single simulation from a single model.

The problem is that each predictive model type has its own historical accuracy (known as "skill"), and so the lines embody different levels of importance. Further, it's not immediately clear if all possible lines are drawn so any reader making conclusions of, say, the envelope containing x percent of these lines is likely to be fooled. Eyeballing the "cone" that contains x percent of the lines is not trivial either. We tend to naturally drift toward aggregate statistical conclusions without the benefit of appropriate tools.

Plots of individuals should be used to address the specific problem of assessing individuals.