Big Data, Plainly Spoken (aka Numbers Rule Your World)

How to act like a data scientist 12: say no to the daily grind

One of the graphical obsessions of the Covid-19 era is the daily tracking chart, showing the  counts of total or new cases , deaths, hospitalizations and so on. Later, people started talking about rates, and some charts show daily rates, e.g. of case fatality, or positive tests. For example, this chart in OurWorldinData:

The recent rising tide of Covid-19 cases in the U.S. conceals the bedrock of bad news: based on daily tracking charts like this, I have heard various people claim that while infections have grown, the death rate has been dropping. This trend is a statistical artifact because deaths lag infections.

On these daily tracking charts, the case fatality rate for any given day is the total number of reported deaths divide by the total number of reported cases. Among those reported cases are new cases, people who just tested positive. It usually takes one week to a month before a patient either recovers or dies. Any new cases, say reported within the last week, are extremely unlikely to contribute to death today but they will affect the fatality count in the coming weeks. So when an epidemic is spinning out of control, and infections surge, the proportion of reported cases that are less than a week old grows rapidly - these patients don't affect the number of reported deaths (yet), and so the case fatality rate as defined drops.

The solution to this problem is cohort adjustment. If its takes up to a month to know if a patient may die, then we track all the new infections today for the next month and at the end of the month, compute what proportion have died. In this definition, every infected person has equal amount of observation time.

One consequence of cohort adjustment is you must have patience. You reject the daily grind, the temptation to make bad "real time" decisions.

***

If you have run an A/B test, you most likely have fallen prey to the same fallacy. If you run a Facebook or Google ad, you typically set a time window, say a week. This means that the audience may see your ad on Monday, the first day of the campaign, or on Sunday. Most testing platforms show you a daily tracking chart. For any given day, they show you a "sales rate". That's the total number of sales from the beginning of the campaign up to that day divided by the total number of ad impressions.

A viral tweet sends a lot of traffic to the site suddenly driving up ad impressions but interest quickly dies down. Such traffic often takes a few days to a week to convert into sales. So the daily tracking chart shows traffic surging on the day of the viral tweet but the sales rate plunging! This doesn't mean these prospects are less likely to buy. It just means you need to wait for at least a week to give them enough time to make a decision.

When audiences see the ad on different days but they are measured on the same ending date, the daily sales rate is affected by the "age" distribution. If you apply a cohort adjustment, then the ending date will depend on the day of the ad impression, and vary by person. If your software does not let you do this, it's time to shop around!

How to act like a data scientist 11: do not assume random

Credit: ep_jhu/Flickr

 

You might find it curious, that someone who pointed out how testing drives the case numbers from my very first Covid-19 post (here), and who called for broad, non-targeted testing in this Wired column also from early in the pandemic (also here), has been silent on testing statistics. Specifically, I have not - until now - commented on the media's two favorite testing metrics: the number of tests per capita, and the "positivity" ratio, i.e. the proportion of tests that came back positive.

The reason? I quote from my post earlier in the week:

In the short history of the Covid pandemic, people started with case statistics. Then, they claimed that death statistics are less manipulated than case statistics, when they learned about testing. Then, they claimed that testing statistics are less manipulated, until they realized governments determined who got tested. Some governments counted tests shipped, not test results. Then, they claimed hospitalization statistics are less manipulated, until they learned that hospitals sent sick patients to nursing homes.

Commentators citing and interpreting these statistics behave as if they aren't manipulated. The President already contradicted them on that count. Manipulation is both intentional and unintentional. (I use unintentional intentionally to include intended unintentional.)

In my Wired column, I pointed out that one huge concern with non-random testing is it produces biased data that prevent us from understanding the pandemic. What makes this worse is when analysts speak about the statistics as if they came from randomized testing.

The positivity rate going up only tells you that of the people who were recently tested, more people have tested positive. If people who got tested were a random selection of the entire population, then yes, it indicates that the coronavirus would have spread more widely. A higher proportion of the population would have been infected.

This is not a given when the sample of people tested is non-random. Another reason why positivity rate goes up is more infected people are getting tested than before. (This is a broader statement than saying more people are getting infected.) Randomized testing almost never happens on its own - it must be organized. If the sample were truly random, you'd expect that someone who is not infected is equally likely to present themselves for testing as someone who is infected. Are you kidding?

Greater spread is one of several reasons for positivity rate to go up. Unless the analyst has examined the other reasons and found them lacking, then the conclusion is based on assuming greater spread, and assuming the other reasons do not apply. It is a tautology.

So, anyone saying positivity rate going up implies Covid-19 is spreading is doing some or all of: (a) assuming that Covid-19 is spreading, (b) assuming that the tested population is randomly selected, (c) assuming that the tested population has always been randomly selected [demonstrably wrong], (d) assuming that uninfected people are comprising a growing proportion of those being tested, (e) assuming testing data are comprehensive and accurate, (f) assuming no one is manipulating the testing data.

***

Regarding manipulation, we have indirect evidence, such as the U.S. President musing about slowing down testing. We also know that the UK government counted "tests shipped" rather than "tests results returned". This is like automakers counting "sell through" rather than "sell out". Sell through tallies include cars sitting in lots of dealers, and is part and parcel of the "channel stuffing" strategy to inflate short-term revenues. Sell out happens when drivers take ownership of the cars. Recently, the UK government stopped publishing testing numbers. In the U.S., I previously reported that California's testing numbers were fishy. I'm suspicious of testing data in the U.S. given the variety of public and private entities conducting tests (while being appreciative of those who put in effort to gather such data.)

The fact that few governments have implemented random testing, even to a small subpopulation, makes it clear that politics is interfering with the public health imperative. The side effect is suboptimal policies misled by low-quality data.

Recall what I said here: don't blame low-quality data for misguided policies when politicians have the wherewithal to improve the data quality.

***

For data scientists, this is yet another instance of garbage in garbage out. In an age in which the provenance of data is frequently unclear, it's worth our time to investigate and interrogate the data - and never assume random when we simply don't know.

 

 

 

What about the dead people?

Business media blasted this headline today: "The Treasury sent $1.4 billion stimulus money to 1.1 million dead people." (For example, this BusinessInsider article).

This sent me looking for U.S. death statistics, and to this CDC webpage.

Annually about 3 million Americans die for all kinds of reasons. For a rough calculation, I assume deaths are evenly spread out over the year so about 250,000 per month. The GAO study covered April and May, two months. About 0.5 million people died during those two months. Since the Treasury can't know who will die in advance, these payments will eventually show up as paid to the dead.

Death records are not immediately available to authorities. The CDC webpage that had full mortality statistics for 2017 was published only in mid 2019. If reported deaths are lagged by 4 to 5 months, in other words, if the Treasury used data from even 6 months ago, this delay accounts for the entirety of the payments that went to "dead" people. It is inevitable that some small percent of stimulus money will be paid to the dead.

There is fraud involved too, certainly, but probably a minority of the headlined amount. (Even though BusinessInsider called this a "significant portion" at the start of the article for clickbait, by the end of the article, you learn that the $1.4 billion is 0.5% of the total payments, which they unironically describe as "a slither".)

***

That's potential fraud by the little people. What about fraud by businesses? For example, did anyone look at how many companies took out PPP loans that subsequently declare bankruptcy? This is just one of many possible ways to milk stimulus money.

***

Here is another thing I learned while reading those death statistics. Out of the roughly 3 million who died in 2017, 56,000 died from "influenza and pneumonia". The U.S. death toll from Covid-19 today already exceeds 122,000, more than double that number. The Covid-19 count doesn't include deaths from influenza and pneumonia this year. Without counting any deaths between now and end of year, Covid-19 is already the 6th leading cause of death (referencing the 2017 ranking). So we are facing a public health crisis.

The number of deaths is the mortality rate multiplied by the number of infected. The mortality rate is being revised down because of undetected cases. Any revision is guesswork since if we really knew who's undetected, those people would have been placed in the confirmed category!

The researchers who bump up the number of cases to claim a lower mortality rate should realize that when you bump up the number of cases, you're enlarging the base of the infected. If the death rate is as low as the flu, then the number of infected must be much higher to attain double the number of deaths.

If you take the adjusted mortality rate and the unadjusted infection rate, the numbers don't work out. You undercount deaths. That's because the inclusion of undetected cases simultaneously affects mortality rate as well as infection rate. It's been argued that if Covid-19's death rate is just like the flu, we should treat it as nothing special. The problem is the rate of infection, which leads to more than double the number of deaths, even at the same rate of death. Thus, we need policies and behaviors that clamp down the infection rate.

 

 

Coronavirus has not gone away, neither have these Covid fallacies

It's a good time to be reminded of the fallacies that continue to pollute global dialogue about the pandemic.

 

Comparing Covid fatality rate to seasonal flu fatality rate

A few research groups are pursuing the hypothesis that Covid-19 is just a type of seasonal flu. These groups hope to show that the "true" fatality rate of Covid-19 is on a par with the seasonal flu. They have pushed for antibody testing, under the theory that many people have been infected but did not get symptoms. The much heralded study that claimed prevalence of over 15 percent is the study by the NY State Department of Health, much trumpeted by Governor Cuomo (about which no one involved either scientifically or politically has the guts to publish an official report. What are they hiding?)

This project was doomed from the start. It should be clear by now that data collected during any epidemic are subject to significant measurement errors, which are real scientific challenges, and exacerbated by political interference - perhaps expressing the desire to look good. But you already know this - and I'm not retreading this point today.

Even if we assume that the infection fatality rate of Covid-19 has been accurately measured, it is still inappropriate to compare that to the fatality rate of seasonal flu.

The fatality rate for seasonal flu is conditioned by widespread flu vaccination (at least in the U.S. where everyone is urged to take the flu shot) while there is currently no vaccine for SARS-CoV-2. This fact alone dooms the project. One would have to go back to the period before the flu vaccine was invented to find the right comparable.

What's more, how accurate is the flu's fatality rate? It also faces measurement challenges because the flu is endemic. Most flu infections are not reported, detected or confirmed. Undercounting is as much an issue with flu as it is with Covid-19. [Epis reading this, please correct me if the fatality rate of seasonal flu published by the CDC are not counts of actual confirmed cases but estimations based on modeling that imputes undetected cases.]

 

Hoping for herd immunity as well as flu-like fatality

Those who believe Covid is a type of flu also tend to advocate the strategy of "herd immunity." If enough people are infected, the remaining population will be naturally protected. I'm only interested in the analytical methods so I don't judge anyone on their beliefs. It's still possible Covid-19 becomes an endemic disease like the flu. What doesn't work is believe both the flu hypothesis and the herd immunity strategy.

Herd immunity requires immunity: someone who was previously infected develop antibodies that confer immunity to the virus upon recovery.

Most of the epi models adopt the SIR structure, for which immunity is built in. The R compartment consists of "recovered" (and dead) people. Recovered people and dead people are treated the same mathematically because neither can spread the disease to anyone else. That's only true if infected people who recover are not infectious.

And yet, no expert has been willing to say definitively that having antibodies for SARS-CoV-2 makes one immune to Covid-19.

Further, you'll learn that epidemiologists do not use SIR models for seasonal flu. Instead, the appropriate model is SIS. The SIS structure has no R compartment; anyone who recovers is susceptible again. That of course is our experience with the flu. Having caught it does not mean you won't catch it again - one might even catch the flu multiple times within one season.

If Covid-19 is indeed just like the flu, then immunity is irrelevant. Herd immunity is even more pointless.

 

Saying young people have a low risk of dying from Covid

This statement is literally true. What is missing is context. Young people have a low risk of dying, period. I'm sure if the overall death rate of flu is 0.1%, the rate for young people is much smaller than that. (Remember this post.) What matters is the excess risk over the normal risk of dying for each age group. Dave Spiegelhalter at Cambridge has done a lot of this work for the UK. Here is an example of the calculations involved. The bottom line is: having Covid-19 is like packing a full year's of risk into a few weeks, for all age groups.

That framing apparently confused some people so he re-wrote the headline later to say "Covid roughly doubles your risk of dying this year". That's because the normal risk of dying excludes Covid-19 and the risk of dying from Covid-19 is roughly equal (ages 20-49) or higher (50 and up) than the normal risk.

Thus, even for younger people, getting infected with Covid-19 doubles your chance of dying. It's one thing to say double the death rate is fine by me, it's another thing to say that the risk of Covid death is "low".

 

Claiming monitoring X could have predicted the outbreak

A slate of studies are making claims that sewage, parking lot volume, search engine traffic and so on are leading indicators, and could have alerted us to the outbreak before the first case was detected.

Studies that look for the virus in sewage are worth doing. They help epidemiologists nail down the spread of the virus at the early stage of the epidemic.

Such research goes off track when these investigators over-reach and claim predictive prowess. After the novel virus has been discovered, it is a straigthforward matter to find prior sewage samples, and test them for the presence of the virus. If we also know that the virus-infected sewage came from which buildings or households, we'd have learned a lot about how the virus spread. But this explanatory model does not predict.

It is utterly wrong to conclude that sewage monitoring is a leading indicator of viral spread. Dial time back to November before the first case was confirmed in Wuhan, China. We assume the local sewage company was testing the dirty stuff. But for what? At that time, no one knew anything about this coronavirus, there were no tests available, no virus samples to validate any tests. Is the lab supposed to test for every possible virus including all animal viruses that have non-zero chance of leaping to humans?

It's not just the sewage (link). It's also the parking lot images and search engine traffic (link). So, from here on out, if more people are searching for "diarrhea", we'll predict that a new virus is circulating in the population? Even if true, which virus is this?

This type of studies do not, and will never, become the basis of a predictive model. It is malpractice that scientific journals have allowed researchers to make such false conclusions.

 

Talking about metrics as if they aren't manipulated

All metrics are perverted. This is a theme in my second book, Numbersense (link), which starts with a chapter filled with every manner of statistical shenanighens practiced by deans of reputable colleges and universities. It is then followed by a chapter on rampant obesity in the U.S., which some medical researchers claimed was a result of "bad measurement". (This blog is a preview of that chapter.) Latter chapters focus on economic metrics like unemployment rate and inflation rate. You get the idea.

In the short history of the Covid pandemic, people started with case statistics. Then, they claimed that death statistics are less manipulated than case statistics, when they learned about testing. Then, they claimed that testing statistics are less manipulated, until they realized governments determined who got tested. Some governments counted tests shipped, not test results. Then, they claimed hospitalization statistics are less manipulated, until they learned that hospitals sent sick patients to nursing homes.

If you read Numbersense (link), you may recall some students with low GPAs or test scores are asked to spend their first year at other schools or part-time programs, so they disappear from the statistics submitted to U.S. News for the school rankings. Does it shock you that hospitals sent patients to nursing homes?

Only the naive takes a "see no evil" view. In the age of Big Data, taking such a view is simply irresponsible - as there is too much leeway to twist the data or analysis to whatever story one wants to spin.

 

A glimpse into surveillance data via the Harvard study of parking lots

Last week, I read the new study by Harvard researchers (link) who found “evidence” in parking lot images and search engine traffic that the novel coronavirus started circulating in Wuhan in August 2019. This study showcases the phenomenon of “story time”. Crucially, these datasets do not contain an outcome variable (i.e. the thing you're trying to explain). The connection between the signals (images and search traffic statistics) and the outcome (Covid cases) is tenuously established via coincidence.

Any data series about Wuhan that shows an unusual spike within months before December 1, 2019, the date of the first confirmed case, can be used to establish the “true” date of arrival, by following the exact analytical plan used by these researchers. What elevates one such data series above another is the assumption that one is a better proxy for Covid cases than the other.

For argument’s sake, let’s say the search traffic trend for “feeling dizzy” spiked around September 2019. In spirit, this pattern is exactly what buttressed the Harvard study – the only difference between this and that is the assertion that search traffic for “diarrhea” is a proxy for Covid cases while that for “feeling dizzy” isn’t. But the only justification for using search traffic for “diarrhea” as a proxy for Covid cases is it fits the story. Is it the only thing that fits the story? If “feeling dizzy” were accepted, the researchers would have concluded that the arrival date was in September rather than August.

***

Even if these novel datasets can be trusted, the patterns themselves do not tell the story without accepting a host of unverified assumptions (see last week’s post for a list). In this post, I take a deeper look at how the trend of parking lot traffic was analyzed. The preprint provides a glimpse into the nature of surveillance data that are being touted as the "new oil".

Here is the key chart from the Harvard study:

The orange line is the trendline fitted to the data. The vertical axis is labeled “relative car volume,” which is an index. This means the raw data – counting cars in satellite images of the parking lots – have been converted into an index, where the value of 1.0 means the traffic on that day is in line with the average traffic on similar days. Typically, one would just compare Monday’s traffic to the historical Monday average, etc. These researchers decided to divide the week into weekdays, Saturday and Sunday so weirdly, Monday’s traffic is relative to the historical average of all weekdays. Is it really true that the parking lot traffic is indistinguishable on any weekday in these six Wuhan hospitals? I doubt it.

In any case, when the relative car volume has a value of 1.1, it means on that day, the parking lot volume is 10% above the usual volume on comparable days. Based on the chart above, on August 1, the traffic was right around 1.0, and on September 1, it was just under 1.1. So, I don't think this orange line provides strong support for the conclusion that the virus landed in Wuhan in August.

The trendline is a loess smoother, which means that each value is a weighted average of the raw data within a "local" time window centered at the date of the value. So, what these researchers produced is smoothing (a form of averaging) on top of indexing (over subgroup averages). The raw data have been smoothed twice. Smoothing data is useful for discovering long-term trends – it is not the proper tool to look for specific points in time.

To their credit, the researchers also show the individual data series of the six Wuhan hospitals for which satellite images were analyzed. Those background lines are very strange indeed. If I am interpreting these lines properly, then there were fewer than 15 data points (satellite images) per hospital during the full year between May 2018 and May 2019, only about 1.25 observations per month per hospital!

The full analysis period was from Jan 9 2018 to Apr 30 2020, with a total of 140 observations across 6 hospitals, according to the preprint. These images are clustered, with maximum frequency during April 2020 (well after the peak of the pandemic in Wuhan). With 6 hospitals, and 842 days within the analysis period, there should be up to 5,000 or so daily parking lot images. This analysis looked at 140 of these, with timing that was far from even. The irregular spacing of the data should make one very nervous about the smoothing analysis.

Another sign of trouble is the skinny confidence intervals (the gray band around the orange line) that supposedly convey the uncertainty of the trend estimate. One should expect the uncertainty to bulge when data are sparse since fewer observations means we are less sure. That's not the case here as the gray band pretty much exhibits the same amount of uncertainty throughout the analysis period in spite of the uneven spacing.

We should also expect the confidence intervals to widen where there are abrupt turns - unless this is counteracted by having more samples around those turns.

It's unclear what level of confidence is used in the gray band. Typically, one should show 95% intervals. In the context of loess smoothing, one should expect, roughly speaking, the gray band to encompass about 95% of the data within that local time window. When I look at the period around May 2019 for example, I just don't understand that confidence interval. It's most likely just plus or minus one standard error, which would be about 67% confidence.

If they are showing one standard error, then the usual two standard error intervals will be 20% on either side of the orange line. This means that any difference within 20% is not statistically meaningful. That fact pretty much wipes out the entire picture, since the peak of the orange line right around December 2019 was only slightly above 1.2 (20% above "normal"). Thus, the satellite image data are too noisy even to establish that the parking lot traffic volume was unusual in December 2019, let alone August 2019.

***

Surveillance data are touted as the new oil. They're being sold to businesses and governments as some kind of X-ray that reveals all. It's not just a privacy issue. The bigger problem is that such data frequently yield incorrect insights - and such insights may actively harm customers and citizens due to blind trust. In addition to asking questions about the ownership of such data, we need to start measuring the impact of such data, whether incorrect insights are harming people, and which subgroups are disproportionately hurt.

Working with poorer ingredients

Antonio forwarded this Guardian article to me, with the headline “coronavirus may have been in Wuhan in August, study suggests.” This headline is based on a preprint promoted by Harvard, available here.

The report introduced two pieces of evidence: (a) occupancy of major hospitals increasing and (b) web search volume for Covid-related keywords increasing. The data sources were respectively satellite images of parking lot traffic, and search trend data from the leading search engine in China. This type of data analysis is typical of “big data,” which I define using the OCCAM criteria (see here).  

I have come across this report on a different site, and planned to write about it – as the latest example of what I call “story time”. It’s a feeling I frequently experience in the midst of reading studies. As I start to doze off from the lullaby of the data presentation, the researchers have moved seamlessly into creative narration – they’ve called “story time.” After I wake up, I realize that the story is only minimally connected with the data.

One of the telltale ingredients of “story time” is proxies. The most important sentence in the preprint is this one: “Between 2018 and 2020, there was a general upward trend of increased hospital occupancy as measured by the parking lot volume proxy.” (my emphasis). Everyone should master the skill of “proxy unmasking”.

What must be true in order for parking lot volume to be an effective proxy for hospital occupancy? An incomplete list includes:

• People who are sick drive to one of those six hospitals in Wuhan.
• People who drive to these hospitals park their vehicles in the hospitals’ parking lots.
• People who park in these parking lots are patients, not staff, or random people say visiting places nearby.
• These hospitals have enough capacity in their parking lots.
• People who visit the hospital are themselves sick, not relatives or friends of someone who is sick.
• People who visit these hospitals will occupy hospital beds.
• People who are hospitalized keep their vehicles in the parking lots until they are discharged.
• People who are hospitalized have the novel coronavirus (as yet unknown pathogen).

Notice that the data from satellite images confer none of this information. They are simply assumed without comment. I sensed “story time” when I realized that most of these assumptions are contradicted by my own experience. Even if I believed them, I make a distinction between data and assumptions.

The same proxy unmasking applies to the web search data.

It's really a tough task to prove that one's featured story is the right story as opposed to a story. For example, here is the chart showing the trend in search volume for "diarrhea":

The story of the preprint focuses on the first and third vertical dashed lines. There is a small spike in the red line (diarrhear) around the time of the first line (August) and then a huge spike around the time of the third line (January 23, when the first case was identified). The blue line for "cough" is taken by these researchers as a "control," saying that its trend is explained by seasonal influenza. Given that cough is still a key symptom of Covid-19, one should expect the volume of "cough" searches to become elevated when both influenza and Covid-19 are affecting search volume. Further, the period to the right of the third vertical dashed line was the epidemic period - when more people were worried about community spread, and what can explain the dramatic drop in searches?

***

When the quality of data is lower, the standard of analysis must be set higher. A chef needs a bag of tricks when working with lower-quality ingredients. Lower-quality data cannot be an excuse for lower standards. Quite the contrary, "big data", OCCAM data demand more stringent benchmarks. Is this counterintuitive?

Know your data 27: IamNotHome.com

This week on Junkcharts, I featured the New York Times's analysis and visualization of how New Yorkers living in the richest sections of town are staying away from the Big Apple during the pandemic. For commentary on the graphics, see this post.

The Times's article is important for another reason. It represents a type of analysis that "big data" surveillance technology enables. The key data sources are cell phones which track users everywhere they go (typically through installed apps, although cell phone providers and vendors also have the data). In the past, to know if people are home, one would conduct small surveys across all neighborhoods.

An essential difference between surveys and surveillance is consent. Anyone who answers a survey knowingly hands over the data to the data collector - traditionally, the data collector does not inquire about the survey-taker's identity (this red line may be eroding). For cell phones and mobile apps, the typical user is not aware their locations are being captured and sold. The user however most likely "consented" via one of two ways: (a) concealed language buried in terms and conditions, or (b) blackmail refusing service unless consent is granted.

The report compared data from three sources, which reveals two things. There is an entire cottage industry of data vendors who are trading our location data, and these datasets are not identical. The differences may arise from the method of surveillance, such as which apps are tracking us, and from the method of inference. The question of whether someone is home or not is not merely the question of one's current longitude and latitude. Based on location data over time, the app vendor assigns each person a home location. There is no exact way to do this. For example, we can set a time window and compute the most frequently occurring location. Or, we can limit the analysis to the wee hours, when you'd most likely to be home. There are all kinds of possibilities for errors, such as extended vacations, or night shifts. Despite the dragnet intention of the surveillance industry, much of the data aren't as precise as advertised. Errors are tolerable when aggregate statistics are the end product, but as you probably know, vendors tout the granularity of their data.

The report clearly stated that these vendors anonymized the data. Anonymized data is not the same as anonymous data. Please don't let the prevailing narrative mislead you. Anonymized implies the data collector stores your personal details, and then selects who gets to see them. Anonymous is not collecting any personal details, meaning no one, not even the data collector, have them. For the type of "big data" analysis performed by the New York Times, the analyst only needs to know that someone lives in a certain neighborhood - not the name and personal details of each person. For most applications, there is no need to know people's personal information.

Not only is personal data not needed, the analytical results shown by the Times remain unchanged if the dataset consists of a random sample of, say, 10,000, rather than millions. That's because understandable insights are aggregated statistics. The best chart of the Times article plots the average percentage of time at home for top 1%, 5%, 10%, 20% and bottom 80% segments. These lines are generally 5 to 10 percent apart. The visual story does not change with a much smaller dataset.

Personal location data can easily be used to harm us. If you're a protestor, the harm is self-evident. If burglars get a hold of such data, they can pick your pocket while you're out. Years ago, someone put up a website that was called, if I remember correctly, iamnothome.com, by scraping messages left by Facebook users that inferred they were out and about. That site was immediately shut down by authorities. Location data vendors are providing the same information, now with more and better sources, and yet have not faced scrutiny. Add in forced consent, and harm turns into self-harm.

Harm frequently arises from "story time" based on the data. I've mostly used the term "story time" to describe academics who publish reports that first present a lot of data, and then a story revealed by the data, except if you think twice, you realize the data were the lullaby to feed you the story while you're dozing off! Recent reports said some colleges have purchased location data of students. Administrators can then compile detailed records of which students spend how much time in bars, dance clubs, libraries, etc. The harm comes when the administrators (sometimes abetted by the app providers) label certain students as "likely to fail" and send them early intervention notices. This is story time! The percentage of time spent in bars and clubs does not define a failing student; it may be a reasonable interpretation but the causal linkage is not supported by the data.

***

Recently, Andrew featured another instance of story time, in which an influential psychologist found that "time trends in religious attendance correlate with time trends in homicide rates in low-IQ countries but not in high-IQ countries." Some iffy data analysis was offered to make that claim. But the conclusion was "The prescriptive values of highly educated groups (such as secularism, but also libertarianism, criminal justice reform, and unrestricted sociosexuality, among others) may work for groups that are highly cognitively sophisticated and self-controlled, but they may be injurious to groups with lower self-control and cognitive ability.” If you're interested in politics and psychology, Andrew has a lengthy discussion of the article here.