Exploratory Data Analysis in Six Sigma

Overview/Description
In the Analyze stage of the Six Sigma DMAIC process, you closely examine the output variable (known as y) and its possible causes or input variables (known as x's) collected in the Measure stage to get a deeper understanding of their relationships. The goal of this analysis is to narrow down the many possible x's identified earlier during the Measure stage, to a few probable ones. This analysis is generally conducted through the use of two important toolsets: exploratory data analysis and hypothesis testing. Methods and tools used in these broad toolsets help to identify a few probable root causes impacting process performance and the Six Sigma project goal. This course introduces some key exploratory data analysis tools used in Six Sigma such as multi-vari studies, correlation, and regression models. The course takes you through the multi-vari analysis to identify positional, cyclical, and temporal variations and how to apply an effective sampling plan for conducting this analysis. It also explains the correlation coefficient, its statistical significance, and how it is different from causation. In addition, the course helps you interpret the linear regression equation and explores how you can use it to model relationships for prediction and estimation of data. This course is aligned with the ASQ Certified Six Sigma Green Belt certification exam and is designed to assist learners as part of their exam preparation.

Target Audience
Candidates seeking Six Sigma Green Belt certification; quality professionals, engineers, production managers, and frontline supervisors; process owners and champions charged with the responsibility of improving quality and processes at the organizational or departmental level

Expected Duration (hours)
1.5

Lesson Objectives

Exploratory Data Analysis in Six Sigma

recognize how multi-vari analysis helps a Six Sigma team during the Analyze stage

match the steps in a multi-vari analysis to their descriptions

match variation types with their graphical examples

sequence examples of the steps in creating a sampling plan for a multi-vari analysis

interpret scatter diagrams to determine the correlation between variables

match the correlation coefficient value with the scatter diagram that best illustrates that value

identify reasons why Six Sigma teams should determine the statistical significance of a correlation coefficient

match statements about correlation and causation to examples of their respective characteristics

predict a variable value for a given regression model and other variable value

recognize the elements of the simple linear least-squares regression equation in a given example