Attribute Charts

Video Activity
Join over 3 million cybersecurity professionals advancing their career
Sign up with
Required fields are marked with an *

Already have an account? Sign In »

9 hours 53 minutes
Video Transcription
Hi, guys. Welcome back and Catherine McKeever. And this is your lean six Sigma green belt. So today we're going to wrap up the last of the greenbelt control charts on, and we're gonna go over attributes charts. So I want you to be able to understand when we use an attributes her and how you read it.
So attribute charts are completely different than any of the other control charts that we've talked about and kind of their own little entity. And if you think about the way that I've described statistical process control
because these are for discrete data and Onley discrete data, there are two types of attributes charts we generally don't recommend. You use statistical process control for attribute data until you're
a little bit higher level practitioner with a little bit more in depth understanding of the, um,
nuances and the statistics that relate to discrete data. Because discrete data is kind of a tricky entity in and of itself, because you really can't do statistics on it. But we do want to use our discrete data. So if you have interval or orginal data, if you think back to,
um your yellow Bell interval data is going to be something that's reported like a ratio. Orginal data is going to be like your customer satisfaction score.
We do not have a control chart for categorical data. So if you are calculating how many how many men and how many women you have, which is entirely assigned by categories or T shirt colors that have no relationship to each other? You cannot use statistical process control
discrete data you can. There are two main control charts
from discrete data. There the sea chart in the U chart. You read them exactly the same. The future only has two options. It's a buying Mary. It's a yes, no. So things like the binomial distribution that we talked about would be appropriate there. But reading is the same.
So this is what you look at when you're looking at attributes charts. So you still have your time measure on your X axes and on your y axes, you're going tohave. The values. So remember previously in our expert are in our expert m are you were looking at a spectrum
in this particular chart you're looking at absolute about, or you're looking at values for customer satisfaction. score.
So a couple of things that you're going to notice we still have an upper and a lower control limit. Thes. They're going to be from your voice of the business. They could be from your voice of the customer potentially as well. But these one specifically voice of the business, you're also going to see that you have a central control
a little bit different. Where is because we don't want to call it the, um the average because you really can't. You really can't calculate averages on things like how satisfied you are.
You're going to centralize it. So we're going to say right here is smack dab in the middle and you can still see that you have some variation.
So if you were to try and do some root cause analysis and use this to drive decisions, you would want to replicate what happened in measure for where you're up there on five at your upper control limit, or potentially do more root cause analysis in what exactly happened on Measure 14
where you're below your lower control limit.
Like, he said, these were provided by the customer by the voice of the business and really, what you're looking for is how do your patterns play out so you can see towards the end. You are kind of hanging out around your central line, which is what we're going to call our average line, except for one data point, which would be,
what we would consider almost like a special cause because you are outside of your upper or lower control limit. So with that attribute, cut charts can be very helpful if you want to do things like trend analysis. So let's see. Okay from points 1 to 4, you can see we had an upward swing, but then something happened and we had a downward swing.
So that gives you an idea of where you can identify best practices or where you need to implement solutions if you can work backwards and identify it. Like I said, these aren't exactly control charts
because there isn't really statistics behind it. On when you start looking at your you and your sea charts, you're going to be using your binomial distributions on, so you really want to be able to understand how to calculate it. But from a basic attributes church like what you are looking at right here.
You're going tohave. Whatever your values are, either your orginal where your interval on your y axes,
Um, time measurements on your X axes, and you're going to plot your data. So on measure number one, you got 2.3 on your customer satisfaction measure number four, you had a solid five. So that's how you would use these for kind of what is more common data
for greenbelts. So I think that I had mentioned that statistical process control tens to live in the land of the Black Belt, and that has a lot to do with the types of data that's collected attribute. Data tends to be more common when we're looking at greenbelt projects. So I wanted you to have a sense of
how you can plot the data and show the performance over time.
So with that, this is our last control chart. You know how to use it. You know that this is gonna be your discrete data. So your customer satisfaction scores, You're not going to do the screen categorical data. So your complaint types, even though you will build a history graham for them
on and you do know how to read these. So you are still going to be looking for
data, data patterns and trend analysis because this is a time study which would lead us into our next module, where we're going to go over the shoe heart rules and data patterns and trend analysis. So I will see you guys there.
Up Next