The Difference between Projections and Forecasts
Open, published methods and data used to create well defined scenarios, versus pretending to foretell the future
by William Warren Munroe, March 12, 2014

If you could see into the future what would you like to know about? Some people may look for ways to address personal interest, looking to gain wealth without working. Finding treasure - golden nuggets? On the other hand, perhaps one of the most important topics of global interest is human population.

Indeed, there are few topics as important as human population projections. Population change impacts, and is impacted by, social and economic activity as well as earth’s environments. Population projections are useful when setting priorities, such as considering opening and closing public and private facilities, including public schools, transportation, electrical generation, water works, Official Community Plans, etc. Indeed, looking into the future referring to current trends is fundamental to decision-making:

"How can society intelligently participate in the setting of national priorities if the population does not understand where we are, and where we are heading on current trends."1 (Ivan Fellegi, 1999)

Wanting to look into the future is not new. Around the world for thousands of years people tried many ways to see into the future, including examined patterns in animal entrails, flocks of flying birds, clouds, scattered bones, tea leaves, looking at lines on palms of hands, looking at the stars, reading special cards, using hallucinogens, and hearing voices from above (or below), as well as looking at statistical data gathered using rigorous well documented methods, in a way, sifting numbers through one’s fingers to find the important bits.

"Yet a large part of society is not used to analyzing statistical data. As a consequence we are really failing to serve them well if data are all of what we produce - most people will not sift through the millions of printed pages we produce in order to find the occasional golden nuggets." (Ivan Fellegi, 1999)

Canadians have arguably the best data set in the world, namely Statistics Canada’s Census of Population conducted every five years, with which to examine the most important topic of our time, human population projections. The Population Projection Project uses a method that the United Nations calls "census survival ratios" (aka cohort change ratios) because it is understandable and

"very useful ... transparent and easy to explain to a wide range of audiences."2 (Swanson et al, 2009)

When census numbers from Statistics Canada are plugged in

"the resulting projections not only have internal validity but, also, ‘‘face validity’’"3 (Swanson et al, 2001)

Projections simply take what we have seen in the past and carry these into the future. If people living in an area continue to have similar fertility, life expectancy, migration patterns, what will the future look like? Forecasting on the other hand simply pretends to foretell the future.

"Forecasting pretends to foretell the future, while the projection is an analytic tool which -- within the constraints of a tightly specified model -- enables the analyst to consider the implications of alternative scenarios." (Ivan Fellegi, 1999)

We can create scenarios of high or low fertility, increased or decreased economic activity, etc.

"The issue of scenarios is important: in order to underline the analytic character of projections, it is our policy to publish always a set of possible projections, each corresponding to a well defined an analytic scenario." (Ivan Fellegi, 1999)
Information Age or Disinformation Age?

We finally have a wonderful source of information and a wonderful way to look at possible future scenarios based on current trends found in the well documented census; however, some people want to stop the census, shutting the briefly opened window of opportunity, closing the curtains, shutting out the light to see the ways into the future.

Sadly, if the census is discontinued as is being considered in the UK and USA, Canadians will be forced to use numbers that are made up by a small group of people in positions of authority who will not make public the methods they use, pretending to foretell the future. These people become increasingly abusive when questioned just as BC Statistics did when asked why they did not inform taxpayers about the way population numbers were created, including using the change in the number of telephone landline hookups as an indicator of population change, despite the rapid increase in cell phone use.

Thus we may continue to move from an active involved, informed and consulted citizenry to being misinformed and easily misled. Already, forecast numbers that cannot be explained, verified nor reproduced, are being used to justify closing public schools in BC. Instead of moving from democracy to despotism, insist methods accompany "forecasts" and don't use "forecast" numbers that can not be verified and reproduced.


Use the Canadian Population Projection Project:

- to be able to verify and reproduce population projections.

- if you prefer population projections that make sense rather than population projections that don't make sense

- if you want to be able to see how the population projections are created, not only to make sure that the correct methods accompany the "findings" but to also to understand the strengths and weaknesses of the methods and data used so that we better understand what the "findings" mean.


Related reading

See the chart provided in 2010 and the charts developed from additional information showing the "Population Change and the Fluctuating Enrolment"

See the "Enrolment forecast changed without explanation "


Endnotes

1 "Analytic Activities At Statistics Canada", Ivan Fellegi, former Chief Statistician, Statistics Canada, Conference Of European Statisticians, June 1999

2"Forecasting the Population of Census Tracts by Age and Sex." David A. Swanson, Alan Schlottmann, Bob Schmidt, 2009, p. 60.

3"State and local population projections: Methodology and analysis." Smith, S., Tayman, J., & Swanson, D., 2001, pp. 282-285.



Back to Top

Website content, code, and design by W. W. Munroe. Copyright ©