ProcessMA Resource




What is a Measure?

A measure describes the dimension, quantity, capacity, performance, or characteristic of a process or population. We need and use measures to quantify where we are, where we want to be and how far we need to go.

Data Type

Data is whatever measurements we make or take. There are two basic types of data, continuous and attribute data. It is important to understand the type of data you have because the method of analysis is different.

Continuous data - Continuous data is described as anything that can be measured on a continuum or scale, it can take a value from negative to positive infinity and between any two values, there could always be another. Examples of continuous data are like length, height, weight, time, revenue, cost, etc.

Attribute Data - Attribute data (also known as discrete data) can take only certain pre-determined values. The main types of attribute data are:

  • Boolean (e.g. Yes/No, Pass/Fail)
  • Ordered and unordered categories (e.g. Survey response)
  • Limited and unlimited count (e.g. number of errors per document, number of defective documents)

In most circumstances, continuous data is superior if they are readily available. Continuous data yields more information and sensitivity. Unlike attribute data, continuous data do not treat all defects the same way.

Converting Between Data Types

Continuous data can be easily converted into attribute data by grouping data into meaningful categories. To convert attribute data into continuous data would be trickier. For example, for boolean data (Yes/No), a pseudo continuous data can be achieved with the use of percentage (i.e. % of Yes), provided the sample size is large enough. For ordered categories, a pseudo continuous data can be achieved by using a finer and wider scale (e.g. increasing the scale from 1-7 to 1-100).

Types of Measures

There are generally two types of business measures, namely effectiveness and efficiency measures, and both are equally important. A dashboard should have a good balance of both types of measures.

Effectiveness Measures - It measures the degree to which customer needs and requirements are met. Examples of effectiveness measures are such as timeliness, accuracy to specifications, yield, etc.

Efficiency Measures - It measures how well resources are used in the process of meeting customer requirements. Examples of effectiveness measures are such as cost per transaction, cycle-time, amount of rework, etc.

Principles of a Good Measure

With the aid of information technology, businesses can sometimes be inundated with too many measures. The following are some guidelines for good measures:

  • The measure must be important to the customer
  • The measure must be easy to understand (clear definition)
  • The measure is sensitive to the right variables (drivers)
  • The measure promotes appropriate analysis and action
  • Data needed must be easy to collect