Statistics – 1

Statistics – you either love them or loath them!

Here is a definition from Wikipedia ‘Statistics is a branch of mathematics dealing with the collection, analysis, interpretation, presentation, and organization of data.[1][2] In applying statistics to, e.g., a scientific, industrial, or social problem, it is conventional to begin with a statistical population or a statistical model process to be studied. Populations can be diverse topics such as “all people living in a country” or “every atom composing a crystal”. Statistics deals with all aspects of data including the planning of data collection in terms of the design of surveys and experiments.’

I have a problem with statistics – I am not sure that they are that accurate.

The issue is that that they can often be used to distort the truth about a subject.  For example, the television companies rely on a very small sample of what people watch to identify the ratings of any given programme.  But, in reality it cannot give a true picture of what people actually watch because not everyone registers their viewing habits.  Then there is fact that we often change channels during a programme for a variety of reasons which distorts figures.

Another area that is not very accurate is when it comes to counting people in certain countries.  Here there are  many issues such as to whether the figures are correct in terms of who belongs to what family; corruption – counting people several times for political reasons or to show that certain people groups are larger than they really are; the non-professionalism of the counters; are all people counted eg the homeless and refugees;and also many countries have not undertaken any kind of census for quite a few years. When it comes to counting refugees – does the UN and others exaggerate or under-report the numbers.? As far as I know there never has been any double-checking of any UN statistics since the organisation was created back after World War 2.  When you consider the UN is probably the biggest provider of statistics, one would have thought that checks should be made, especially as a lot of money is involved.

Many statistical models have been created, but I wonder how accurate they really are?  There never has been any really testing of them despite that they are only theories.

One area that particular concerns me is the counting of those who follow a particular religion.  For example, in Islam, whole families are counted as Muslims as well as subsequent generations, because it is a cultural ‘thing’, despite that there are many ‘categories’ of Muslims from ‘secular’ ones to ‘extremists’ – none of these are accurately counted, often because of fear of shame or dishonour.  When it comes to Christianity, it is almost as bad, especially when it comes to Evangelicals.  If a minister of church is evangelical, then the whole congregation is counted as such when the truth maybe very few actually have a living faith.  But, even those who consider themselves as evangelical may not practice their ‘faith’.  So, when statements are quoted that such a country has a very large population of Evangelicals, like Guatemala, I take them with a ‘pinch of salt’ when you see how much violence, corruption and children born outside of marriage exists.  Even in the UK, the figure of 7% evangelical is in reality is far too high when you start to dig into the working out of the faith of those part of ‘evangelical’ churches.  It also does not help that there are at least 17 ‘evangelical tribes’ with varying beliefs about key doctrines and practice.  Quoting ‘high’ figures can easily become ‘triumphalist’.  Instead, we should be looking at the real figures and take appropriate action in the area of reproducible discipling.

So, we can see that statistics can give a distorted picture of reality and therefore does not properly inform what our views should be about a given issue.  We need a better understanding of the contexts that statistics are used and get behind the ‘headlines’.  We also need to understand, in the case of religion, the belief systems of the various sub-groups, to properly understand them and therefore to be able to report objectively about them.

Advertisements

2 thoughts on “Statistics – 1

  1. I agree that statistics aren’t the be all and end all, and we need to dig beneath the numbers to understand what’s really going on in the world. But, speaking as someone who uses and generates statistics every day, I would argue that statistics do have their place!

    Taking a sample of the population is a very cost-effective way of representing the views and characteristics of the whole population. If every single person with a TV were to report their viewing, this would be a very costly and intrusive exercise, and would yield exactly the same results as the current sampling method, assuming the sample is representative and is of a sufficient size to generate small confidence intervals, which I’m sure is the case.

    You are right to suggest that we should be sceptical about official figures relating to religion. In the UK census and most national surveys, religion is ascertained from a single question asking the respondent to select from a list of recognised religions. It follows that people who identify themselves as Christian will range from those who have a vague belief in God and may have been baptised as a child through to those who are avid followers of Jesus. Such high-level statistics do not make these distinctions, but can still be useful to show trends over time – for example, the increase in the percentage of the UK population identifying themselves as Muslim or identifying as having no religion. There are some other surveys (for example the World Values Survey) which ask many more questions about particular behaviours and attitudes, and statistics from these sources would shed further light on the more interesting questions you identify. However, you can only really understand the true reality of someone’s faith by talking to them in-depth (i.e. qualitative research) and it is impossible to do this in a standardised way across a large population.

    • Thanks for your comments. I am still not convinced that sampling is anywhere near the reality, especially as most of them are of a small group. For example, in the USA, a major survey company only consults 2,000 people out of a population of 300 million.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s