Paradigm Shifts

What happens, really, when a "paradigm" shifts? How can you benefit from knowing in advance which way a paradigm is moving? A paradigm has all the characteristics of a "bell shaped curve" - that is, not everyone agrees to THE PARADIGM - so you have people think at the extremes with most in the middle (note in the chart, the x axis could be opinion or time). So, when the AVERAGE view of a paradigm moves to a new position (see second curve), you have a PARADIGM SHIFT. If your view is on the side of the first curve that is approaching the NEXT paradigm, you will be "anticipating" the shift. By watching which way the first paradigm moves, you might be able to see the early warning signals of the new paradigm.

 

   Two views - S shaped curve or bell shaped curve

The same data can be graphically presented in two very different ways - looking at it both ways can help your understanding of the underlying principles or trends. The curve on the right is the familiar bell shaped curve. The one on the left is the also familiar S shaped curve. These are just two ways of presenting the same data (the left is the integrated form and the right is the differentiated form). It is useful to use one form or the other to make your point or to understand how trends change.

 

  Incremental changes over time can equal big changes

Some trends last a long time (e.g., world population growth) but others are often made up of a series of smaller trends (e.g., type of aircraft engines). When these smaller trends are tied together they give the impression of one "long term trend". Each of these smaller trends is like an S shaped curve - it starts slowly, grows quickly, and then dies out. But, another product or viewpoint is ready to take over and continue the trend. Over time the accumulated changes can be very large. For the casual observer, the overall trend may seem like one smooth curve, but it is really a series of smaller trends that come and go.

 

  Discontinuities and uncertainties

Any trend-like representation is subject to a perturbation that is unlikely to be accounted for in advance. The perturbation may cause a temporary disruption in the trend or affect the rate of change or direction of the trend, or it can be large and alter the trend so it is no longer recognizable as the original trend. The big jump about half way along the x axis was due to a "perturbation". Normally these perturbations are unexpected, and they make for large difficulties in any trend estimating (however, with a little foresight you can anticipate some of them). The trend may be simply displaced (as in the example), or the trend disrupted completely.

 

    Hype or Bandwagon Effects

When something is "new" and others mimic it, it produces rapid growth. After a while, the growth poops out and you either get a bust or a correction to more reasonable growth rates. When some event starts big, everyone may think it will be the next major trend, so even more people get involved, and the trend overshoots its normal course of events and falls back (often rapidly). It then corrects itself to the appropriate rate to sustain growth. This can be called the hype curve (where the hype gets ahead of reality) or a representation of the bandwaggon effect (everyone joins in whither they know anything about the subject or not).

 

   Trend Extrapolation

Trends are often extrapolated into the future, with a "range" indicated to allow for historical variation. This works fine for situations where there are a lot of historical data and the ranges in variation are fairly consistent (e.g., world grain production) but is not useful when there is high variation (e.g., stock market values). In the diagram, the middle line is the average value over a long time (such as world grain yield). The lower and upper lines bracket the variation of the average, so they bracket the average by accounting for the long term variation. This is often summarized as the high, low, and medium changes on whatever subject is being studied.

Return to Instructor's Viewpoints