9 hours 53 minutes
Welcome back, guys. I'm Captain MC Iver, and this is your lean six Sigma green Belt introduction to lean six Sigma statistics.
So in today's listen, we're going to go over our core lean six Sigma statistics. So at this point in the course, we have finished our data collection plans and our measurement system analysis. So if you were playing along with an imaginary, um,
project, you would at this point, have some data for us to play with, which is nice, because you will be sitting there like, What are we gonna do with us? Um,
until orient you back to your domestic model from Yellow Belt. So how we actually performer Lean Six Sigma projects we are right now. It's half dancing the line between measure and analyze. So you create your data collection system plans in your measurement phase, you'll get some of those preliminary and measurements
and you'll use some of your analysis tools that will go over in this chunk as we move towards analyze,
to develop, lose baseline measurements. So it's like a Choctaw were in measure, but we're going to step, are toe and analyze so that we make sure that we know that what we're gonna do is our baseline is how we actually want to report the data at the end of the project. So data collection done, you should have some data to work with.
We're chuch eyeing between measure and analyze.
So let's talk about statistics. So there are two fundamental types of statistics. There are descriptive statistics which do what it sounds like. They describe our data so they tend to give us roll up information or summary information.
They are foundational. You have to have good, descriptive statistics
before you can have inferential statistics, and you can work with these in a smaller sample size. And when I say
a smaller sample size, this is what we're going to be capturing for our baseline. Descriptive can intend to be population metrics. So all measures within a chunk if we have the ability to access that. So, for example, if we wanted to
look at client satisfaction scores, we could either look at 100% of them completed. Or maybe we want to do a sampling because there's,
you know, thousands of them that we don't want to look at. But descriptive tends to be where we feel more comfortable working with all elements in the data set. We use it a lot for base lining. It'll make sense when we get to the next one. As far as what that actually means. The second piece of
statistics is inferential, which is where we
wait for it. Make inferences where? So this is an idea where we get some generalized generalization or generalize ability about how our daughter performs. So what does the data seem to do? What is the likelihood? What is the probability you're if you hear the phrases like likelihood, Probability? What you were thinking is inferential
is tends to be predictive eso
how will likely are we able tohave This result our data distributions the those magical curves and pictures I keep telling you about are actually inferential tools. However, we have to have really solid descriptive statistics to build them, so they're very cyclical upon each other.
The last thing that's really important for inference Tral,
inferential and I know we remember we talked about it a little bit in yellow belt. We'll talk about it more. It's hypothesis testing or
rejecting or failing to reject the null hypothesis. This let lives an inferential land because what we're looking for here is statistical significance or the likelihood that what we're seeing is in fact due to our intervention or actions as compared to
the moon aligned and the stars saying, and we got the results that we were looking for So
two types of statistics, they are very interrelated with each other, you need one toe, have the other. But if you were to look at this in a linear way, you're gonna want to do descriptive and then inferential once you get your inferences, you're gonna want to reduce descriptive again so that you report on it
for me, there is an easier way to think about descriptive and inferential statistics. And when we switch over to business metrics, we're going to kind of switch our terminology a tiny bit and move away from statistician and more into manager. I think about inferential and descriptive as
leading and lagging measures.
Eso lagging measures are things that have already happened. They are done so in our little graph, it is weirdly growth. We can't predict yearly growth will.
definitively state yearly growth. We can predict inferential statistics. We can't definitively stay yearly growth until the year has passed. So for me lagging these air look back measures these are things that have already happened that we want to know
very, very common in financial reporting. You see them in quite a few organizations. When we start talking about dash boarding, um,
you're going to have to ask yourselves as a metrics person and is a data driven person because you're in the culture of, um,
Are you are your dashboards leading or lagging? So then the flip side to that is leading measures. Leading measures are things that can give you a sense of what is going toe happen. Inferential statistics are leading measures because we look at a small chunk off all of the possible abilities. And we say, Is this or isn't that's going to happen? Or
do we think this could happen?
So an example for a leading measure is if we go back to Yellow Belt and we think about my pizza joint, which is actually really badly run if you think about it throughout all of these courses, but you think about my pizza joint.
A lagging measure is how many pizzas did we sell in a specific shift. A leading to measure is how many people are in line outside the door. That one are fabulous Pizza That's gonna tell me how hard my kitchen is going to need to work. So if I popped my head out and I see my cash register person
because we didn't implement the the the perfect best case scenario where people just
psychically ordered in the pizza arrived. I look outside, I see my cash register and I see eight people behind them. I know reasonably, I'm probably going to get eight orders, which means I have to have my kitchen teed up to do eight sets of work.
So this is a leading measure. This is something that gives us a sense of what needs to come, Really Common leading measure are predictive hiring.
So if we see that we're having a growth of 15% or even more looking at you, Sai Buri. Um, what we're gonna want to do is we're gonna start progress are proactively hiring. So when that growth hits us, we're gonna have the people here ready to go,
and there's going to be no impact felt by our end user or customer.
So the difference between the lagging and leading descriptive lagging these air actuals leading inferential these are could should would keep those in mind as we start talking about them and will make more sense as the different tools kind of known together for you.
So today we went over an introduction of lean six Sigma statistics, and I want to call out specifically these Air Lane six segments statistics because there is so much knowledge in statistics
that the only thing that we are touching on in this course are the statistics that you're going to use every single day. So the ones that will be in your tool set as a green belt.
If you want to learn more about statistics, it's everywhere. You're welcome to learn it or paying me if you have any specific tools you'd like to learn about and our next, um, of course we're going to go over an introduction to descriptive statistics. So we're gonna start talking about this a little bit more. I will see you guys there