Improvement Insights Blog
Posts tagged "Data"
Here’s three New Year’s Resolutions to accelerate your quality improvements: 1) Raw Data Diet, 2) Quality Tools and 3) Worst First. Here’s why these will change your future and your company’s:
“Hi, I’m Jay Arthur. It is the New Year and so it’s time for some New Year’s resolutions. Here’s some resolutions I’d like you to consider.
“First, go on a raw data diet. That means you need to know everything about each individual defect, mistake and error, and then you can summarize that. If you start from summarized data you don’t know where the raw data is, and so most of the time you can’t actually figure out what to fix without the raw data.
Continue Reading "2021 New Year’s Resolutions for Quality"
Most companies are drowning in data, so you don’t need to Define and Measure anything new. Take DM out of DMAIC to accelerate results. Here’s why:
“Hi, I’m Jay Arthur, author of “Lean Six Sigma for Hospitals” and QI Macros [software].
“In the whole Six Sigma world, the DMAIC process, one of the things I’ve noticed is every company I’ve ever gone to… ever… has so much data they’re just… they’re drowning in their own data, and they don’t know how to analyze what they’ve got. And yet in DMAIC, we teach people to Define and then figure out something to Measure.
Continue Reading "Take DM Out Of DMAIC"
People often get stuck before the “Analyze” phase of DMAIC because of ugly data.
You can’t analyze data until it is in the right format for analysis.
In this data about medications that might affect patient falls, we have to:
- Transpose the data
- Use Excel’s Text-to-column feature to split the meds into separate cells
- Use QI Macros UnStack Columns tool to get all of the meds into a single column
- Use QI Macros Data Mining Wizard to create a PivotTable and Pareto chart of medications involved in patient falls
You can watch the video below:
Continue Reading "It’s Hard to Do Data Analysis on Ugly Data… Part 2"
I have found that people often get stuck before the Analyze phase of DMAIC because of ugly data. You can’t analyze data until it is in the right format for analysis.
In this example of patient falls data, we have to:
- Transpose the data
- Use PivotTables to summarize the data
- Use a c Chart to plot the data.
Continue Reading "It’s Hard to Do Data Analysis on Ugly Data"
This Sunday’s Dilbert hit the nail on the head:
Actually, you don’t even have to type. I’ve seen Six Sigma gurus use Minitab generate fake data to demonstrate a concept during a training. The fake data is statistically in control and indistiguishable from real data. Excel will let you generate random data.
This is what scares me about what we’re teaching students in business school. It’s too easy to create fake data. Isn’t this a symptom of what caused Wall Street to melt down?
Everyone I talk to in healthcare says that the medical mistakes and errors data are systematically underreported by a factor of 2-to-4.
Continue Reading "All Business Data is Intentionally Misleading"
Recently at a conference, I saw a noted Six Sigma practitioner use Minitab to generate data for analysis.
In that moment it struck me: colleges have been training students to manufacture “real-looking” data for class assignments. All you have to do is enter some parameters and out pops data that varies with whatever distribution you assign it.
Isn’t this what got us into trouble in this economy: creating data that justified our behavior and inflated results?
I’d like you to consider that seemingly innocuous behaviors like this can lead to catastrophic results: market crashes, inflated bonuses and general stupidity.
If teachers want to generate data for students, that’s probably a good idea.
Continue Reading "Making Up Data"