Blog

Statistics used inadvertently

Many people think that statistics is a complex and difficult subject. That only professionals who have received training will know how to use its skills and knowledge. In some common areas where statistics is used, such as in weather forecasting, the census, election, etc., they all need to use a lot of manpower and material resources to collect the data, summarise it and perform complex calculations to obtain the required results.

Common terms in statistics include count, mean, median, mode, maximum, minimum, range, variance, standard deviation, sample size, coefficient, correlation, normal distribution, parameter, population, etc. Okay, that’s enough already, please let me go. Just seeing these technical terms makes me dizzy, not to mention understanding what they actually mean and how to apply them. Let alone the more complex mathematical formulas and different types of graphs.

Image 1: An esoteric perception of statistics
 
But is statistics so esoteric and difficult to understand? This reminds me of some examples of my day-to-day life in the UK, where I unwittingly apply statistics.
 
I take my son to school every morning by bus. It always arrives on time, rarely delayed, and never absent. But things get complicated when I set off to work. Delays became frequent, and occasionally the bus doesn’t even turn up. After I’ve waited a lengthy amount of time, sometimes two buses come at once.

Image 2: Indefinite bus waiting
 
At this point, I have to open the bus company's mobile app. I check the bus schedule again and see where the bus is in real time (this is only used as a reference by the way, as often the bus’s precise location is not shown). These disappointing experiences have taught me to check the real-time location of the bus before leaving the office, which improves my chances of catching the bus.
 
In autumn last year, I was going to buy a television. Every November in the UK, major stores will organise the best deals in their Black Friday sales. There are attractive discounts on many items including home appliances. Everyone loves a discount, so why not wait a bit to get it? Chances of no more Black Friday sales are very small. If I am willing to start enjoying my new television later, it is worth taking the chance to exchange a little time for a discount that has a considerable chance?
 
It's only now summer in the UK, and I'm already thinking about what home appliances to buy for this year’s Black Friday sales. A vacuum cleaner? An instant pot? An air conditioning unit? What new toys are you looking to buy this year?

Image 3: The annual Black Friday Sales
 
Competition is fierce in UK supermarkets, and certain supermarkets have recently introduced innovative and thoughtful personalised offers for its members. They will offer you a discount (sometimes quite a bit) on an item based on what you have recently bought.
 
For customers, if the quality of the item is good enough, the customer will be happy to buy it again. Even if the goods at home have not been used up; it feels good to buy an extra one to lower the average unit cost price.
 
For supermarkets, compared to the traditional weekly specials, they can now promote almost every single item. And as a consequence, they can more accurately forecast sales and inventory. More importantly, they can increase customer loyalty and bring back customers. After all, it is a lot easier to sell a customer a product they have already tried, than to convince a brand-new customer altogether.

Image 4: Supermarket promotions
 

In these very ordinary life examples, from catching the bus to saving on living expenses, I unconsciously apply some statistical concepts including data collection, summarising data and predicting outcomes.
 
As I continue to accumulate experiences, both good and bad, on bus rides and shopping at the supermarket, more and more data are collected, and the aggregated results will help me make more accurate predictions. This way, with a greater sample size, I have a better chance of getting the results I expect. Of course, these are not strictly formal statistical behaviours. These behaviours are just what we do very roughly in the brain. It is just with a little bit of statistics applied.

Image 5: Data collection, data aggregation, result forecasting
 
Having said so much, I would like to point out that statistics is not as difficult and unreachable as we imagine it to be. It is very close, and so close to us in fact, we may be applying statistics every day without realising it.
 
Currently, at Stats4SD, we are working on some new applications of statistics for agriculture. We hope that through applied statistics, we’ll be able to provide some advice that farmers would benefit from knowing about. For example, the best dates for sowing and harvesting, short-term and long-term weather forecasting, as well as what soil composition, temperature and rainfall are best for particular crops. These all help to maximise crop yields and minimise crop losses, and to protect and preserve farmers' efforts and provide enough food as possible.
 
For you, where have you unconsciously used statistics? Where would you apply statistics, if at all? What can be done to give you the desired results you expect? I’d be interested to hear ?.

Author: Dan Tang