Intro: Welcome to the E6S-Methods podcast with Jacob and Aaron, your source for expert training, coaching, consulting, and leadership in Lean, Six Sigma, and continuous improvement methods. In this episode number 53, “Rule out the Ruler – Part 3,” we review the different data types, and how that plays into analyzing your measurement or decision systems. Here we go.

***Rule out the Ruler Part 3– Measurement Data Types***

Objection 1: What’s the point? What does this have to do with Measurement Systems?

Counter 1: Data comes in many forms, choosing the correct statistical tool requires knowledge of the data type. Different measurement system analyses exist for different data types.

Types of Measurement Data

I Continuous Variable – Easily subdivided into smaller units (that make sense). Natural, physical measures. Separates data into Continuous Equal Intervals (Infinite in-between values possible, infinitesimal)

a. Time, Temperature, Pressure, Volume (fluid), Concentration (chemical), Distance, Mass/Weight, Velocity, Energy, Currency, Voltage, Current (electrical)

II Discrete:

a. Discrete Count - Separates data into Discrete Equal Intervals (No in-between values possible), in order, with no value below “0” possible. If enough discrete intervals (>10), may be treated as continuous for statistical evaluation

i. Discrete machine settings (pin settings)

1. Coarse control, no “in-betweens”

ii. Count of Days, Containers, Defects, Complaints

1. May also be represented as % or ratio, but still discrete data (technically)

b. Discrete Attribute Ordinal - Separates data into multiple groups in order (sometimes expressed as a number. Though not split into equal intervals, sometimes treated as nominal or continuous (given enough intervals), to assist in statistical analysis.

i. Likert & Rubrics –

1. (Poor/Fair/Neutral Good/Excellent), (Small/Medium/Large)

2. Rubrics - FMEA scaling, Student essays grading system, Performance Reviews

ii. Customer or Supplier Classification - (A/B/C), (Tier 1/Tier 2/Tier 3). Presumed to have an “order” in this example, May also be considered “Discrete Nominal.”

c. Discrete Attribute Nominal - Separates data into multiple groups with no order. Categorical Classification

i. Call type (Order/Complaint/Follow up/Other)

ii. Defect Type (Overweight/Scratch/Dent/Off-color/Broken)

iii. Direction (North/South/East/West)

iv. Color (Red/Yellow/Blue/Green)

d. Discrete Attribute Binomial - Separates data into 2 groups with No Order

i. (Pass/Fail), (Go/NoGo), (Good/Bad), (Stay/Go), (Make/Buy)

Outro: Thanks for listening to episode 53 of the E6S-Methods Podcast. Stay tuned for episode number 54, where we go through the Attribute Agreement Analysis. How “right” are we, and do we agree, statistically? If you would like to be a guest on podcast, contact us through our website. Follow us on twitter @e6sindustries or join a discussion on LinkedIn. Subscribe to past and future episodes on iTunes or stream us live on-demand with Stitcher Radio. Find outlines and graphics for all shows and more at www.E6S-Methods.com. “Journey Through Success”