Twenty-five:  A tool for understanding global goals and statistics

 

 

Version of 31 January 2007

 

Matt Berkley

 

This document is to help people understand some international goals, such as the United Nations’ Millennium Goals, and some numerical claims about human progress, particularly in economics.

 

The idea is to help resolve some misunderstandings, ambiguities and puzzles.  Clarity about what is being said, and the basis for what is being said, might help overcome some disagreements about past outcomes of policies, and future policy directions.

 

Some distinctions I make below are conceptual.  The presence of a word in the list is not meant to imply that what it refers to is measurable. 

 

For instance, it is important to understand whether someone is claiming to have measured

a) the level of people’s consumption, or

b) the adequacy of their consumption.  

 

Since people don’t agree about the value of different foods, people are unlikely to agree about what adequate consumption is.    Also, what counts as consumption in a broader sense is a bit subjective.  I consume air, whose quality is important to me.   I “consume” the park where I walk.    I consume knowledge, some which might help me live longer.   I consume water, but of what quality?  How much dirty water is of the same value as clean water?  

 

So we might think that any claim to have measured the adequacy of people’s consumption as a whole is false.  Still, it may be useful to distinguish the concept of “level” from the concept of “adequacy” in order to understand what is behind the words of a researcher or politician.   To ask a social scientist whether they are claiming to have measured the level of consumption or the adequacy of consumption may be useful.   To ask a politician whether they are aiming at higher levels of consumption, or higher adequacy, may be useful as well.  

 

As in many areas of life, what may first seem complex can be understood more easily by grasping central principles.   In this case, as in many areas of social science, a key element is imagination.    Another is empathy.   

 

Perhaps many people are frightened by statistics not just because of the numbers, but because of the words.  If you see an abstract noun, try and develop a way of understanding what it means in real life.   A discussion about “consumption” isn’t fundamentally about some complex thing in a mysterious machine called the economy.   It is supposed to refer to real life, so you can ask yourself, or others, what kinds of things it is supposed to refer to in real life.    It is more important to understand what is being discussed than to try to improve on something badly defined.

 

Consumption amount, or level, is a different concept from consumption adequacy, and both are different from consumption expenditure.   That may all sound obvious, but in economics surveys on what people spend are sometimes erroneously described as data on consumption, and then erroneously described as poverty statistics.   What you spend (expenditure) is not what you consumed (consumption), and neither of these are what you lack (poverty).  

 

As I type these words, I am imagining real people.    I do not think it is possible to think about social science meaningfully in any other way.   

 

The aim of this document is to help people decode economics and some other social science.   Behind politicians’ statements about “poverty” going up or down there are real facts (as well as, perhaps, some real exaggeration), and this document may provide pointers to understanding what the facts are.     The facts are about real people, and the person who can imagine  some of them doing different things as the abstract nouns change, may have a good grasp of what is important without knowing anything technical.  

 

 

 

When you are faced with a large-scale social science goal or statistic, the following may be helpful:

 

  1. Take time to think about real-life situations that might occur, to help you spot assumptions, guesswork and opinion.
  2. Think and ask about whether, and if so how, key words might mean something different from their ordinary meaning. 
  3. Think how the goal or measure might be meaningful in real life.
  4. Understand that the burden of proof is on the scientist or politician making a claim or goal.
  5. Think about margins of error in data, considering real-life reasons to trust or not trust it.
  6. Think about margins of error in the reasoning, considering possible real-life situations and possible skewing.
  7. Distinguish survey answers (what people told researchers) from researchers’ inferences as to what really happened.
  8. Distinguish researchers’ samples (people surveyed) from inferences on whole populations.
  9. Distinguish population trends (e.g. rise in the average) from aggregate trends for people (e.g. average rise).
  10. Distinguish spending from income (most global data on “income” are in fact on what people said they spent).
  11. Distinguish spending from items received (economists sometimes confuse “consumption” with “expenditure”).
  12. Distinguish level (of resources) from judgements about adequacy (of resources for need).
  13. Distinguish incidence (frequency) from prevalence (percentage of people in the situation at one time).
  14. Distinguish prevalence from extent (the proportion of rich people doesn’t say how rich they are).
  15. Distinguish commerce from income (e.g. GDP per capita and average income).
  16. Distinguish income from profit (income minus necessary expenses equals profit).
  17. Distinguish prices from judgements about prices relevant to these people.
  18. Distinguish prices from judgements on the cost of living (relevant prices x needs).
  19. Distinguish consumption from personal resources.
  20. Distinguish personal resources from available resources.
  21. Distinguish available resources from judgements on well-being.
  22. Credibility test: Imagine self and/or others in a realistic range of relevant situations. 
  23. Ethics test: Would you apply the assumptions, methods and claims to yourself and people you feel close to, over a realistic range of possible situations?   Note:  The burden of proof is on the scientist or politician to show unusual circumstances are not important to their claim. 
  24. Ask how far the categories of people form distinct and meaningful groups.
  25. Distinguish statistical significance from real-life importance. 

 

...........................................

 


www.mattberkley.com

266 Banbury Road, Oxford OX2 7DL, England
+44 (0)7868 397699
matt@mattberkley.com