Wed | Dec 7, 2016

How to lie with statistics

Published:Friday | November 7, 2014 | 12:00 AM

Most of us who went to UWI in the 1970s were exposed to a nice little book, How To Lie With Statistics, by journalist Darrell Huff, as a recommended text for the 'Use of English' university course. I was reminded of it this week when I saw the reportage based on the STATIN bulletin dated November 3, 2014, titled 'The Labour Force in July 2014'.

The Observer screamed: 'STATIN says unemployment down 11%' between July 2013 and July 2014. The Gleaner was more modest: 'Unemployment down, says STATIN'. That's not the biggest news coming out of the data that STATIN presented but did not properly analyse.

I have prepared a data table summarising STATIN's press release. There are only two types of people in the labour force: employed people and unemployed people. But not everyone who is of labour-force age (14+) is counted. To be in the labour force, you must either be employed, or, if not working, you must be willing and able to work, and must have been actually seeking work over the last month. People in prison, students, housewives, the aged and the disabled are not included in the labour force (therefore, they cannot be called 'unemployed').

Who is employed?

To be considered 'employed', you must have worked in some form of economic activity for at least one hour during the week previous to being visited by the STATIN interviewers. STATIN does not have a category called 'part-time' or 'underemployed'; if you work only one hour for the week you are employed. If you sell oranges on the street for one hour each week, and earn a few dollars, you are considered just as employed as Lascelles Chin.

If you help your friend sell peanuts on the street for one hour during the week previous to being visited by the STATIN interviewers - free, gratis and for nothing - you are considered employed!

Here is an important point: If I am not working in the month previous to being visited by the STATIN interviewers, but I wrote a job application, or went to a business place to enquire if they have any vacancies, or perused the want ads in a newspaper, I am considered to be in the labour force and, therefore, unemployed. But if I am not working in the month previous to being visited by the STATIN interviewers and I did none of those things (maybe I am frustrated at seeking over a long period and finding nothing), I am considered not to be in the labour force and, therefore, cannot be considered unemployed.

So in analysing labour-force data, it is not enough just to look at those who are employed and unemployed; you must also look at those who have dropped out of the labour force, and why.

The first observation is that the population of labour-force age (both male and female) increased between July 2013 and July 2014, which is what you expect, since Jamaica's population increases every year. The second observation is that the labour force (both male and female) declined. Where is our productive potential going? The third observation is that those not in the labour force (both male and female) increased between July 2013 and July 2014.

And so, yes: Those employed (both male and female) increased between July 2013 and July 2014 (by 1.51%); and those unemployed decreased (-11.01%). But that decrease is not because they all found work, but because almost half of them (10,000-plus) dropped out of the labour force altogether! And they are not all suddenly disabled, or put in prison.

Men are much more seriously affected than women.

These data cannot be used to support the claim that the IMF programme to fix the economy is working well. In fact, it exposes a deepening crisis!

In seeking to paint a rosy picture, STATIN did not present all the data (I had to calculate some of it), and presented a superficial analysis. (Is an election coming)? And the mass media simply regurgitated their press release. Journalist Darrell Huff must be turning in his grave: people are still lying (or at least, not telling the whole truth) with statistics!

Peter Espeut is a sociologist, survey research specialist, and data analyst.