SPC and Global Warming Update

Quick Links

SPC for Excel Software

Visit our home page

SPC Training

SPC Consulting

Ordering Information

Thanks so much for reading our publication. We hope you find it informative and useful. Happy charting and may the data always support your position.


Dr. Bill McNeese
BPI Consulting, LLC

View Bill McNeese

Connect with Us

Comments (4)

  • Scott HindleMarch 7, 2017 Reply

    Bill,Good, useful article – thanks – that should put the global warming argument to bed unless, that is, people don’t trust the data (which I guess forms part of the global warming sceptic’s argument). If we say that the Shewhart “model” is based on a prediction within 3-sigma process limits, and not exact values, I don’t agree that the predicted next value is 0.746 for 2016. Isn’t the prediction simply that no change in the system is expected to give a value between 0.509 and 0.976 ?Using the formal detection rules, is the 2016 value a detected process change (using Rule 1 in the software)? If the upper limit for 2016 is 0.976 and the observed 2016 deviation value is 0.98, shouldn’t it be a detectable signal of change, albeit just a fraction over the upper limit? Your Figure 3 looks visually as though the 2016 value is just inside the upper limit, not a fraction outside it.When you write “And look at the last 4 years” in the last paragraph before the Summary, I take it you’ve taken the 2016 observation to be a signal and you have a licence to dig deeper, e.g. looking at the four most recent years. With a change detected, you have evidence that your 1997-2015 “model” no longer applies to the new 2016 value. In short, I take it that the temperatures have increased year on year, you’ve proven this, but now the increase is different and even more pronounced. Do you agree?Thanks, Scott.

    • billMarch 7, 2017 Reply

       Thanks  Scott.  You are correct on the interpreation.  Control chart says it is between 0.509 and 0.976.  I just use the center line as the predicted value.  People don't like having that big of range!  Yes 0.98 is above 0.976 so it should be a signal.  When you run the data through the software rounding happens last.  That is why the last chart has the point below the limits.   Probably should have changed the calculations to match it.  The UCL is really just a little above 0.98, but it is right there.  The last four years appear to be forming a new pattern for sure.  I agree with that.

  • Allan PatersonMarch 11, 2017 Reply

    Great work as usual. Would love to promote your articles on Twitter but there is no link to do so?Am I missing something or is there not one?

    • billMarch 11, 2017 Reply

      Thanks Allan.  I don't have a Twitter account, but will look into it.

Leave a Reply

Your email address will not be published. Required fields are marked *