It is widely known among quality and process improvement practitioners that the lack of a clearly defined scope or charter is perhaps the leading cause for projects not getting started or completed on time and within budget. What are other causes? From my experience, the No. 2 cause for restarting process improvement projects is poor data. Without verifying the integrity of the data, project results can be meaningless.

In Donald J. Wheeler’s “Probability Models Don’t Generate Your Data” from his column “Thinking About Data Analysis” (Quality Digest, March 2009), Wheeler stresses, that the primary question of data analysis is, and always has been, “Are these data reasonably homogeneous, or do they contain evidence of a lack of homogeneity?”

The advice by Wheeler, a respected quality professional, is of course, statistically sound advice. This article expands on this central question by offering 10 additional key questions that can help to provide an even more complete story when analyzing data. The table in figure 1 summaries these 10 questions with an interpretation for each.

 …