The Meaningless Complexity of Data & Analytics Can Cause CRI
A Simple 10 Step Checklist For Avoiding It
Having clicked on this article, you may be wondering what is CRI? Good question. CRI stands for Cranial Rectal Impaction and it happens to the best of us when we attempt to understand new and complex data. If you work in the field of Data Science & Analytics, this is an inevitable and potentially daily reality. Here is a simple checklist for avoiding it.
1] Know The Size
It amazes me how many analysts jump headlong into a data set without any sizing what so ever. They promptly run a menagerie of filters and then get lost in the meaninglessness…
select count(*) from … is the easiest SQL code — use it.
2] Know The Process
If you don’t know where the data comes from, you are begging for CRI. This failure is probably more prevalent then the first. All data is created by a process. That process has gaps, problems, and inaccuracies. You MUST understand them!
3] Know The Definition, Ignore The Name
You know what assuming does? It leads to Cranial Rectal Impaction! Don’t be a big ass, know the definition of what you are looking at or for. Analysts often assume things based on a name. That doesn’t happen with well defined terms. Also recognize when the definition ought to have another name. People have a bad habit of creating meaningless complexity by renaming everything!
4] Know Your Denominator
Numbers generate fear in people. While it is a real phobia, it is likely a more terrible cognitive bias. Big numbers make people crazy. Broken ratios and percentages are just as bad. Analyst know thy denominator!
5] Always Have Benchmarks
If a report doesn’t provide benchmarks, don’t read it. But even if it does, be careful. Benchmarking a thing against itself has lead to a lot of cranial rectal impaction. I heard a story about a venture capitalist… or was it a virologist … who suffered CRI so severely… no wait it was an economist … oh, screw it — the punchline was jaws of life. Always have benchmarks! Preferably several.
6] Never Assume Causality
The analyst must have an ego, but if they get a giant head about them and start assuming false causalities … well, let’s just say avoid logical fallacy by never assuming causality period.
Correlation is more than enough in most cases.
7] Break It Down
It is call analysis, duh! Break it down. So much meaningless complexity comes from analysts trying to do too much in one giant step. You need to pull the numbers, the process, and population apart. IOW — histograms are your friend.
8] One Step At A Time
No one ever experienced a cranial rectal impaction slowly. It just doesn’t happen that way. They are always the result of high speed activity. Unless you are a bizarre level of masochist.
After you break it down, perform your analysis in one measured step at a time. Rushing to the end is the easiest way to find yourself in a royal jam… cranial jam? You get the idea.
9] Get A Second Opinion
Outside perspective, a second set of eyes — it all works to your advantage. I am not saying a double cranial rectal impaction has never occurred but typically two heads are better than one in this case. Just be sure the second opinion is a thoughtful and critical one… other wise you will just end up with double the pain.
10] The Sniff Test
The thing about cranial rectal impaction is that if you just stop for a second and take a deep breath, you will probably realize you are in the middle of it. Again, this most often occurs when analysts are working at high speeds or under extreme pressure. Slow down. Smart can take some time, but stupid most often happens in a damn hurry. And cranial rectal impaction will make you look really damn stupid!
Thanks for reading! Stay safe. More important, stay informed! #cranialrectalimpaction
