DISPERSION ANALYSIS

STEP 1
Write down the effect to be investigated and draw the "backbone" arrow to it.  In our example the effect is "low teaching evaluations."

STEP 2
Identify all the broad areas, that is root causes, of enquiry in which the causes of the effect being onvestigated may lie -- skills, communication, experience, gender, recruiting, technology, environment, students, etc.  The root causes are the underlying processes and system properties that allowed the contributing factors to culimnate in a harmful event. 

STEP 3
This step requires the greatest amount of work and imagination because it requires you and your team to write all the detailed possible causes in each of the broad areas (i.e., root cuases) of enquiry.  Each cause identified should be fully explored for further, more specific causes which, in turn, contribute to them.  For each cause you ask, "Why does this dispersion (cause) occur?"  This diagram helps us outlining the reasons for any variability, or dispersion. Unlike cause enumeration (see below) where smaller causes that are considered insignificant are still listed, in dispersion analysis, causes that don't fit the selected major causes are not listed.  In other words, sometimes small causes are not isolated or observed.  Consequently, it is possible that some root causes will not be identified in dispersion analysis.

CAUSE ENUMERATION

This is not so much a different type of diagram but a different method of constructing a diagram.  Instead of building up a chart gradually (starting with the "backbone," deciding on root causes, then adding more and more branches), you postpone drawing the chart and simply list all the possible causes first.  Then draw the chart in ordr to relate the causes to each other.  This method has the advantage that the list of possible causes will be more comprehensive because the process has a more free-form nature.  The disadvantage is that it is more difficult to draw the diagram from this list rather than from scratch.

Sometime it may be very difficult to determine the root causes to be included in the diagram.  If that is the case, after we have determined the characteristic or effect we are examining, we follow these steps:
• Use brainstorming to create a list of all the possible causes. The list will contain a mixture of primary, secondary and tertiary (or root causes, middle sized bones and small bones) causes.
• Sort the list by grouping causes that are related.
• Identify or name each major grouping and make your cause-and-effect diagram.  (Thus cause enumeration facilitates the identification of root causes because all conceivable causes are listed.)
• Machine, Manpower, Material, Measurement, Method and Environment are frequently used major causes that can apply to many processes.

PRODUCTION or PROCESS CALSSIFICATION TYPE

When cause-and-effect diagrams are constructed for process analysis, the emphasis is on listing along the "backbone" of the diagram the causes in the sequence in which the operations are actually conducted. The advantage of this diagram is that, since it follows the sequence of the production process, it is easy to assemble and understand.  The disadvantage is that similar causes may appear again and again, and causes due to a combination of more than one factor are difficult to illustrate.

The following diagram is taken from: Kauro Ishikawa. 1982. Guide to Quality Control. Asian Productivity Organzization: 150.