You will come across many different data types in Six Sigma. Likewise, the type of data you have will dictate what you can do and the tools you can use.

Discrete Data

• Best at showing whether or not we have a defective product or service.
• “Pass/fail” is better for failure analysis: (however, failure analysis is opposite to the philosophy of Six Sigma. Preventing defects, not trying to figure out what went wrong later.)
• Counted data is discrete because it uses whole numbers.
• Examples:
  • # of dimples on a golf ball.
  • # of people in a stadium.
  • 80 / 100 to discrete – it is out of a finite set.

Discrete Data Notes:

Operational definition- When would you start your watch for a room service delivery and stop the stopwatch?

If you do this for a project, each team member will have a different set of numbers because we do not have a precise operational definition.

If the data is different, then the data will get attacked; resistance.

Qualitative Data

Qualitative data refers to non-numerical data that is based on characteristics or qualities. In other words, it describes the quality of things. Qualitative is descriptive or observations and uses words. This type of data is often subjective and cannot be easily quantified.

• An example of qualitative data is color because it cannot be described as a number.

While qualitative data is often used in social sciences and humanities, it can also be used in fields where subjective information is relevant, such as market research or product design.

Attribute Data

• This type of data can be anything that is classified as either/or
• Furthermore, very binary
• Pass/fail, go/no-go, good/bad.
• Ex. paint chips per unit, percent of defective units in a lot, audit points.
• Also see: Attribute Charts

Continuous Data / Variable Data

• Anything that can be measured continuously.
• It can always be divided into smaller increments.
• Exists on a continuum.
• Preferred over Discrete.
• Use continuous data where possible because it tells us the magnitude of the issue.
• Gives control over the process & provides enough discrimination.
• Examples:
  • Length (inches, half an inch, tenths of an inch …)
  • Weight
  • Temperature
  • Time
  • Anything you can measure: torque, tension, length, volume.

Teaching Discrete & Continuous Data

Imagine you have a young child who says that he is sick. As a parent, you first touch their head to see if they feel warm; that is an example of collecting discrete data.

If it feels like he has a fever, you’ll likely use a thermometer to take his temperature–another type of data collection. You need to know the magnitude of the fever because that will determine the course of action; 105 – ER, 101 – TYLENOL. In this situation, the temperature reading is continuous data that exists on a continuum.

Location data

Six Sigma is a methodology for process improvement that emphasizes the use of statistical analysis to identify and eliminate defects in a process. Location data can be used in Six Sigma to help identify patterns and trends that may be affecting the quality of a process. One way that location data can be used in Six Sigma is by creating process maps or charts. Charts that use locational data are called “measles charts,” and you could record it on a measles diagram.

For example: Determining the root cause of paint flaws that are taking place on a car production line.

Furthermore, location data can be a valuable tool in Six Sigma for identifying patterns and trends that may be affecting the quality of a process and for developing strategies to eliminate defects and improve process performance.

Converting Types of Data

On one hand, it can be difficult to translate after-the-fact attribute (go/no go) data to a variable. But in most cases, you can find a way to convert attributes to a variable during measuring. For example, how far out of tolerance a product is. Of course, this measurement can be easily assigned a variable.

It is always easy to convert variable data to attributes if you have a standard. Example: Water is too cold to swim at less than 75 degrees. No go is < 75. Then put all of the data set that is less than 75 to “no go,” and all above “go.”

Types of Data Videos

Six Sigma Black Belt Certification Types of Data Questions:

Question: One trait of attributes data is that it is always: (Taken from ASQ sample Black Belt exam.)

(A) continuous
(B) discrete
(C) expensive to collect
(D) read from a scale of measurement

Answer:

B: Discrete. Attribute data is discrete as it takes the form of a binary condition – go/don’t go or pass/fail.  Continuous data cannot be described as an either/or statement, so (a) is not the correct answer. Attribute data certainly does not have to be expensive to collect. And since it is an either/or value, attribute data is not read from a scale of measurement.