I was listening to a presentation being given by a timekeeping vendor. Within the demonstration, a report was run providing a detail of when people decided to take sick days, and the report stated that the large majority of sick days were taken on Mondays and Fridays. Our brilliant analyst noted that there was a high degree of employees in almost any company that abused their sick days to create long weekends. His system would help the company identify these people and fix the problem.
Intrigued, I decided to look into my own timesheets (I’m a consultant, so I have a long history of records). Turns out, I pretty much took 80% of my sick days on Mondays and Fridays. Considering I’m a person who nobody actually cares if I take a couple hours here and there (I do work 3000 hours a year after all), I’m not anywhere close to lying about sick time. These are not days that I’m “sick” but days that I’m actually SICK.
I began to wonder why I tend to get sick during weekends and on vacations. A friend gave me some good insight. High performing people have bodies that tend to know when there is downtime coming and can force illnesses into those times to create rest. Now I have no idea if this is true, but the proportion of time I’ve been sick during vacations versus work time is completely off kilter.
We use analytics to create truths (ok, trends) from which we extrapolate “facts.” Unfortunately these facts coming from quantitative data are about as good as anything you get with statistics – you can create a version of truth for everything. You can find a correlation for your best employees and show how they abuse the sick time system to create 5 long weekends a year. (I just took Friday and Monday – I had a 104 degree fever). There might be people who went to Las Vegas. And I guess it was true… I was laying in bed fairly drugged up.
10 responses to “How to Lie With Analytics”
How to Lie With Analytics I was listening to a presentation being given by a timekeeping vendor. Within the demonstration, a report was run providing a detail of when people decided to take sick days, and the report stated that the large majority of sick days … [
Culture at Microsoft & Google Open-Source Journalism: It’s a Lot Tougher Than You Think I Learned About the Internet Yesterday Inside an Indian Call Center: “Spread the Wealth” (Hat Tip ..Doc) How to Lie With Analytics
As the old saying goes:
“There are lies, damned lies, and statistics!”
Another plausible explanation is that it may be a conscious effort on the part of the sick employee to minimize overall productivity loss.
People work when they’re sick all the time. Management may espouse that workers shouldn’t work when they’re sick, but the reality is that there is often an expectation that productivity shouldn’t slip because of it.
With the advent of alternative work schedules, often there is a smaller impact in taking a sick day on a Monday or Friday. So it could be that when the sick employee gets to a Monday or Friday and is still sick, s/he decides to finally take a sick day in an attempt to recuperate. Sort of a work/health balance…
One can only imagine the impact on morale if a dedicated employee takes this approach, only to be later admonished and accused of abusing sick time.
Gary – I think you’re right. If I’m going to take a day, Friday is usually the slowest day and the best day to be sick. It’s true that I’ll usually work through the others if I can. However, ever since seeing that demo of the timekeeping system, I now feel as though taking a Friday is an obvious and wrong thing to do.
I guess the point is that no matter what the metric or analysis is, we should all be digging a bit deeper or at least know what the assumptions are that we make before jumping to absolute conclusions.
-Dubs
Every good supervisor in the company probably knows who’s abusing the sick day policy. Don’t buy the fancy system to identify things we ought to know. Put part of what you save into training supervisors, since they have huge impact on productivity, morale, retention, et al. Take the rest and put it into analytics that give you something you don’t know.
Interesting comment e-mailed to me by “Jim B”:
I also have 15 years of daily activity logs (both in education, government and consulting).
After reading your observation “on days absent for sick”: Yesterday, I went through 8 of the 15 years of logs and reviewed my days of SICKNESS. [My assumption is the previous 7 years is no different.]
GUESS WHAT? 25% of all my days Sick (really sick) were on Fridays and 25% of all my days Sick (really sick) were on Mondays, w/ Tuesday @ 20%. The least sick day was Wednesday with the remainder of the times sick falling on Sat and Sunday over the last 8 years.
My spouse who is a school director (last 30 years) also has the same Sick pattern. Most days sick are Friday and Monday because of the Sat/Sun recovery time not impacting work and work loads.
I would guess that this is a normal sickness pattern of most individuals in most companies.
My guess is that companies selling time punch clocks have to use this metric as a sales tool to help prospective companies buy the concept that “identifying when people are sick” can help that company to reduce those sick days significantly. The paradigm should be to find out what is causing the people to become sick, not when they become sick! With the construction (and constrictions) of the normal work week MON-FRI, I doubt that will be changing much.
As I have always been said, “Liars can figure and figures can lie.” The smart manager is the one who can decipher the figures and the liars from the metrics and the product.
I want to pick up on two comments.
“Most days sick are Friday and Monday because of the Sat/Sun recovery time not impacting work and work loads.”
I’m convinced that most people would be more productive and more rested with a 4/10 plan such as that used by many police agencies. The workweek is four 10 hour days followed by three days off. When I’ve looked at that in police agencies, there was no increase in either complaints or accidents on the longer shifts, compared to eight hours. Productivity stayed the same with the same “tailing” pattern as shift change time approached.
Officers loved the plan. As one put it to me “There’s a day to get over the week, a day to do errands and a day to rest.” Seems like 4/10 is more do-able in businesses now that we’re in a more connected and 24/7 staffing environment.
The other comment was “The smart manager is the one who can decipher the figures and the liars from the metrics and the product.”
That’s true, but I don’t think it goes far enough. The smart manager is the one close enough to the people who work for him/her that there’s no need for statistics to know who the slackers are.
[…] of days off sick I was intrigued to come across a July post by SystematicHR on sickness metrics, How To Lie With Analytics.As we all know, a large majority of sick days are taken on Mondays and Fridays. It’s easy to think […]
[…] How to Lie With Analytics […]