Sunday, 19 October 2014
Basis Of Singaporean Spirit - People Power: Please Spare Us the Statistics
[This is part of the book: Basis Of Singaporean Spirit - People Power, click here to go to the main page]
A peculiarity of government agencies is that they like to publish statistics. Whenever, there is any controversy, there is always a statistics 'readily available' for them to support any claims!
Well, we all know numbers don't lie. But we also know that statistics does! Why do I state that statistics is unreliable? First, let's take a look at the definition below.
Statistics is the study of the collection, organization, analysis, interpretation and presentation of data. It deals with all aspects of data including the planning of data collection in terms of the design of surveys and experiments.
Noticed the underline words? Yes, any faults at ANY of the underline words, either intentional or unintentional, would make the statistics meaningless. So how reliable do you think statistics really is?
I am always amazed with the efficiency of how government agencies are able to publish statistics in such a short time.
Whenever there is any negativity, they always seem to have 'ready statistics' to rebut. How could this ever be possible? Well, it is possible, if they had used old data and surveys that were not meant for this purpose.
Alternatively, they could also have sent out surveys and questionnaires to people sitting beside them. So how can these statistics be reliable?
A real survey may take weeks or even months to design, not to mention executing them. Have you come across survey questions that asked the same question but in a different way? Well, you are right, that was intentionally designed to identify any wilful or careless answering.
Unless, you are able to inspect all the data points and procedures used in the statistics, using statistics to convince people is simply meaningless and stupid.
But is statistics useless? No, it depends on its usage. Whenever statistics is used to support a claim or as proof, I always take it with a pinch of salt.
Again, unless we can validate all the steps taken to come up with the statistics, unless we can see how the survey and experiment are designed, unless we can see how the sample pool is decided, there is really no reason why we should treat it as any form of fact.
As a business process consultant, I love statistics because that is how I measure the performance of organisation and their processes and set KPIs[i] to improve them! Without those statistics as a benchmark, there is no reference for improvement initiative.
So remember, understand the basis of statistics and don't treat it as a fact because it was never meant to be.
[i] KPI stands for Key Performance Indicator. It is a performance measurement. In simple terms, if you meet it, you are good; if you don't, you are bad.