Qualitative Cluster Analysis

1. The Nature of Data



By Christopher Miller, Ph.D., NPDP

The NPD Discovery Stage is exploratory, conjecture driven, creatively interpreted and above all, fact based. How do you structure unstructured, undisciplined and generally unruly data and convert it into meaningful results?

It is not uncommon for organizations to be confronted with data from multiple studies done over a period of time. The creation and internal understanding of this diverse research can represent a substantial investment of both time and money. Data sources can be radically different, coming in from technology, market research, strategy and others. These diverse data sets all have value.

In discovery we have choices:
  • Allow the data to wash over our collective minds, see what sticks (too hot)
  • Restrict ourselves to solid reproducible information sources (too cold)
  • Ask a consultant to summarize what we think we know (too distant)
  • Create a new charter-targeted set of information (too expensive)
  • Use a proven process (just right)

Remember NOIR (Nominal, Ordinal, Interval and Ration)? Discovery data comes in all forms, but when cutting across data sources it is, most assuredly, nominal. The best way to deal with it? Standardize your nugget data. Generally, this can be done with bits and pieces of fact based information from the conclusions of multiple studies/sources.

Information Nugget Examples –

“The Hispanic population dominates in old Mexico; the borders of 1750 survive today.”

“The need to drill deeper. The aquifers of the south western United States subside at an average rate of X meters per decade.”

“The World Health Organization lobbies the UN and other global entities to make access to potable water a basic human right.”

I like to use a format:
What, So What, Now What
  • What is the nugget of information, quote or data point, opinion, picture
  • What does it matter, implications of the nugget of information
  • What could it mean for our charter, application ideas, conversational points

To equalize the meaning in each data point, you must start by structuring your data in a uniform manner. If this format, or another, becomes a habit across studies, the ned for massive review may be avoided.

In the next Innovative Issues I will discuss the issue of Bias vs. Intuition and the Nurturing of Informed Intuition. Publication January 2017 is scheduled. Chris would enjoy hearing about your synthesis experience, the good, the bad and the ugly… and as always volunteer readers are welcome.

No Comments Yet.

Leave a comment