Measurement Scales

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 to your lean six Sigma green belt. I'm Katherine McKeever, and today we're gonna go over measurements.
So measurement scales are really interesting conundrum for the lean six Sigma practitioner because this is really one of the areas where you get to show your understanding of your organization and your discretion.
So when we're talking about measurement skills, remember you have your data collection plan? We did way back in a while ago. In the course
you have developed your sampling, and at this point, you should have some data and we're going to start our We're going to begin our graphical analysis. We're going to start looking at our data. This is where measurement skills become very relevant.
So with that, measurement skills were like the three little bears of data. What this means is that if you remember the story of the three little Bears and Goldilocks doing her breaking an injury number, the first bed that she laid in was too hard. It didn't work for her. The second bed that she laid in was too soft, still didn't work for her.
The last bed that she Laden was just right.
So what we? Why this is relevant to our lean six Sigma practitioners is because it is up to you to find the balance so we can have some broad generalities. But if your measurement scales do not tell the story, clearly you run the risk of making poor decisions.
Um, so if you think about it, we've done our data collection plan. We put a lot of time and effort into it.
We did our sampling. We've got a great we've got a great chunk of data to work from. That's a lot of work to invest in getting confused in the story that the data tells. So that's why I really wanted to specifically call out measurement scales using your discretion.
Really? The key Take away from this. It is perfectly fine
to play around with your data until you find of the right mix hard, soft, just right. Whatever works for your group.
So, if you remember I told you we were gonna look at our U. S. Presidents hype, data, birth and duration in this module, we're looking at their height. So what we're looking at here is every single US president by their height,
and we're going to say that I created these graphs because I want to figure out how high to set the chairs in the Oval Office. And for those of you who are very short like I am, you understand that swinging your legs does not make you feel like a professional. And for those of you who are very toll,
I don't know because I've never been cursed with being told in my life.
that being said, as I look at this graph, I have no idea how tall I need to set these chairs because I really have no good sense off. What is our U. S. Presidents height? So this is going to be too detailed
on the actual specifics on this graph
are that we are looking at one bar for one president.
So, conversely, here are all 45 of our U. S. Presidents. I can tell you from looking at this that our U. S. Presidents are somewhere between 64 76 inches toll still also not detailed enough to give me a sense
off what I'm looking for and what are my best course of action.
So, at the beginning of the course, we talk or the module. We talked about how you could make bad decisions. If you're good, data doesn't tell a good story right now. I still do not have what we call actionable insights or actionable data. So data that tells me this is a good idea. We should do this.
if we look at our U. S. Presidents Heights in two inch chunks, we will see that I will probably do best if I set my chairs for gentlemen who are between 70 and 74 inches
toll that will give me the majority of our U. S. Presidents.
You will see that if I want to be more inclusive, 68 inches would be helpful. That'll give us that mental middle section. You will notice there is a little bit of a skew towards taller gentleman. So if you want to read into your insights, we can say that the U. S. Tends to like to elect Toller people.
a couple of other things you will notice is you can see a little bit of your standard deviation where we're going to say our average is around 72 we're going to say that our standard deviations are about two inches. So between one standard deviation of the average we have 70
and 64.
We can see that that would be eyeball Onley between 50 and 60% of our total counts. Eso will remember we talked about using distributions. Early on, we said that data tells a story.
This is the type of story that we're looking to tell if I wanted to set a chair height for a US president. I probably be OK if I set for 72 inches
because I'm going to get the majority of the president's. So this is actionable data compared to the last two graphs which showed the exact same data set. But one was too specific and one was not specific enough.
So with that, remember that we want our data to tell a story. We wanted to be something that we can look at it and have a good sense off what might be the intervention. So going back to thinking about what is a process where we says why equals f of X, where are why is going to be our action?
So we're gonna look at our data. This is going to be our xar inputs
and we're going to say Okay, if I do this, I will see these results. So for the case of my presidential chairs, if I set my chair at 72 inches, I will see relatively happy presidents
again, always with the data telling the story. I particularly love this because there's specific graph
because it reminds me how very important it is to make sure that we're representative in our scale. A couple of other things that I'm just going to say the pet peeve of mine as you start doing more graphical analysis, which the good a chunk of what you're gonna do is a green belt
is using that data to tell the story. So get very comfortable with graphical analysis.
We tended to use hissed a grams and variations of hissed a grams because they pave the road for our distributions. But there are plenty of other ways to display. Your data, however, is most appropriate for your audience. But with that,
do not change your scale. This is my number one pet peeve with graphical analysis. You look along the x axis or the y axis. You see that we're really tight together. And then, Oh, as we get further away, we spread out. Changing your scale is really easy way funny numbers.
When you start looking at things
online or you start looking at some data that's reported, make sure to pay very, very close attention to your measurement skills because it is the easiest way to tell a different story than perhaps what the data is telling you. So my pet peeve. If you're gonna pick a scale of two inches for each president, keep
two inches for each president.
So with that today we went over measurement scales. We covered it. The fact that you are able to use your discretion and your organizational knowledge so that you are reporting stories that are most actionable for your audience and your organization,
and we're going to take that information and go right into our distributions
so I will see you in our next module from normal distributions.
Up Next