Suggestions - National Trend Interpretation Standard (s)?

Whether it suicide rates, crime, or GDP, the national media (and most everyone else) concludes significance by comparing two data points. To introduce the fact that there are generally accepted trend interpretation standards, should the quality community advocate for the application of one or more standard (s)?  A trend chart would be a "gateway" to control charts (trend charts with control limits). If yes, is there one or more standards that you would recommend?  If we had a link to the standards, we could reference the link when responding to misleading headlines.  Thoughts?

Comments

  • Matt SMatt S PQ Systems Employee
    edited August 2018
    "Should the quality community advocate for the application of one or more standard (s)?"  Yes.
    Like you, I am confused why the "national media" and others tend to only report "[____] went up" or "[____] went down."

    Sometimes this is done to fit their narrative, other times it is done for simplicity. I like your suggestion of a trend chart with a Median or Mean line as a starting point.

    While looking at various articles, I have come to the conclusion that there is no standard, and it would be nice if there was. I personally advocate a phrase I heard Donald Wheeler state, which I remember as something like: "More data is better, but less will do." In Wheeler's book, Making Sense of Data, he stated "... we often have to work with fewer values, and useful [control] limits may be computed with as few as five or six values."

    They key word in that sentence is useful. The more data one has, the more useful (confidence) one should have in what the data is telling the reader.
Sign In or Register to comment.