Newsletter – January, 2016

On My Mind
Analytics is the collection and analysis of qualitative and quantitative data for decision making. All is well unless you are the decision maker and the datasets at your disposal contain a lot of errors. Some have suggested that messy data is almost inevitable in a big data context—messy means you have errors. This is troublesome in the digital age because of the propagation of errors. One data collection or data entry error—whether made by a human or a faulty sensor—can potentially spread as it moves from one log book/spreadsheet/system to another. Years ago an acquaintance of mine used various spellings of his name (e.g., John Doe, J. Doe, John Q. Doe) to better understand how information about him was being sold. Needless to say the information trails were quite interesting. Fast forward to today in the digital age and you can imagine how a piece of information like a misspelled name can spread. It is important to assure data accuracy when you have control over the data collection and data entry processes. If you don’t have control, then you should probably assume you have data errors and evaluate the risks inherent in your decision making process.

Interesting Stories
What good is poor eyesight? It might be a benefit in some situations. Sir Ronald Fisher was one of the greatest statisticians of all time and he would make my “Mount Rushmore” of statisticians. He unfortunately suffered from poor eyesight for much of his life (search for Wikipedia Sir Ronald Fisher). I remember a professor telling me once that Fisher had an amazing ability to visualize problems—and solutions—in geometrical terms because of his poor eyesight. This made it challenging for “equation-driven” statisticians because they could not “see” what he was talking about—they needed a theorem and its proof to be convinced which Fisher didn’t always provide. We should remember that—for whatever the reasons—people learn and communicate differently.

Company News
Strategic Improvement Systems can now officially be referred to as SIS for short. The 2016 schedule for public seminars and courses sponsored by SIS will be finalized soon – here is a sneak peek at the offerings: Yellow Belt for Continuous Improvement Champions (will be held February 22, 2016); Japanese-Style Hoshin Kanri; Innovation Tools for Change Agents; Statistical Methods for Continuous Improvement; and Quality, Analytics, and Big Data. Also, three courses in the Lean Six Sigma Series will be offered: Continuous Improvement Green Belt; Continuous Improvement Black Belt; and Continuous Improvement Master Black Belt.

Intriguing Reads
The 2015 book titled, Social Big Data Mining by Hiroshi Ishikawa of Tokyo Metropolitan University, nicely integrates big data and social media in a practical way. The diagrams—which include color plates in a special section at the end of the book—are very useful. I especially enjoyed the chapter on “Hypotheses in the Era of Big Data.”