Magna est Veritas
I've always been interested in where our family name came from. It's not common, and causes lots of confusion with others. Give it over the phone and I can be assured that I will be recorded as Barlow, or Farlow, occasionally Varley. Even Google changes my name to Barlow when put into a search.
A while ago I contacted someone with the same surname on Facebook (sadly he is no longer with us). He kindly sent through the Varlow coat of arms, that you can see here. I don't know what all the symbols stand for, but there is a latin inscription included.
Magna est Veritas
Now I didn't do Latin at school so the only bits of Latin that I really know are "Cogito ergo sum" and "In canis corpore transmuto". The first of these was, of course, coined by Descartes and its translation is commonly known - "I think therefore I am". The second is lesser known, probably because it comes from the Disney film, the Shaggy D.A. and means "Turn into a dog". The first has great relevance as one of the stages of Decartes' logical endeavour to prove the existence of God (although flawed, this was a great example of the use of the scientific method, of which Descartes is seen to be one of the pioneers). The second quote has no relevance but for some reason sticks in my mind. I don't know why, because the film really wasn't that good (I'd like to say the book was better, but it wasn't). So I'd like to replace it with something else, something of relevance. Magna est Veritas seems to fit the bill - "Great is Truth".
You may ask why I have chosen to use my family coat of arms as a logo for the training I now deliver? Well, to me, training is all about truth. Training imparts knowledge, and knowledge is power. There is, therefore, a great responsibility for the trainer to ensure that what is being taught is true, and is taught in such a way that it empowers others to take the truth forwards. In my working life, (nearly 30 years working in the NHS and academia both locally and nationally) I have often come across situations where the truth had to be 'sweetened' or 'watered down'. Sometimes I have felt uncomfortable not being able to be honest with others, and sometimes I have been criticised for doing so. Often I was told that people couldn't handle the truth (there's a film quote in there somewhere), and to an extent in business all of that is true. The same, however, cannot be said for teaching and training. It is the trainer's responsibility to impart truth, to ensure that those who go on to use their new found skills do so appropriately. It is the trainer's aim that those they have trained add to the volume of truth rather than detract from it.
Most of my training concentrates on analysis and research methods. Much of this is fundamentally centred on data. We have more data at our fingertips than ever before. Analytics, information, analysis, statistics, data science etc. are all terms that we bandy about. We talk about the power of data, of big data, of the advance of Artificial Intelligence. We consider the differences between qualitative and quantitative data, of differing levels of data quality. We debate the use of big data with that of 'small data' or 'thick data'. We dabble in SAS and SPSS. The more fashionable among us use Python and R. Ultimately it doesn't matter. Underpinning all of this should be the requirement for truth. If we are to base our decisions on data, we have to be sure that the data is 'true', and that how it has been analysed is 'true'. This means that we have to understand when data is good enough (and to be honest about when it isn't), when quality of data may be more important than speed of data, what context the data can be used in, and what analytical methods can (and can't) be applied. We need to understand the challenges and pitfalls of repeated data testing, the likelihood of chance results, the role of natural variation in processes, the nature of cause and effect etc. The quote "There are three kinds of lies: lies, damned lies, and statistics." (often erroneously attributed to Mark Twain who in turn probably wrongly attributed it to Disraeli) emphasises that we can use data incorrectly, either accidentally or purposefully. Either way, statistics used in this fashion can be misleading and dangerous.
So truth is vitally important in data analysis. It is an essential consideration before we pick our statistical tests, or build our data models. It is fundamental to our interpretation of results. But more than that, it is essential for our data collection, for the appropriate setting up of research studies, for asking the right questions and formulating the correct hypotheses. Truth is the lifeblood that has to run through everything that we do when handling data, otherwise our wealth of data will be at best misleading, and at worst dangerous.
Of course my view isn't shared by everybody. Often we get so caught up in providing what others want that we put aside some (or all) of these fundamentals. As human beings we love things that please the eye. Great visualisations of data, interactive graphics, heat maps etc. There is an industry out there creating more and more impressive visualisation software. All of this is great, and empowering, providing the tools that we use contain the right data i.e. 'the truth'. When we use visualisations themselves as an analytical tool, when the visualisation becomes more important than the data, when we keep tweaking the visualisation so it says what we want it to (or what the customer wants it to) then we have sadly missed the point. In these cases we have let our love for the bright and shiny overshadow our quest, and responsibility, for the truth.
My aim in providing training to others is to help individuals understand the right ways of asking questions, formulating hypotheses, collecting data, analysing and drawing conclusions. If we are truthful about the limitations of our data, and apply appropriate statistical techniques and visualisations then we have created something new, and hopefully gained new understanding.
The training that I offer delves into all of this in more detail. For now I would encourage you to keep the motto "magna est veritas" in your mind. It doesn't matter whether you are an analyst or statistician, a data scientist or researcher. It doesn't matter if you are only a user of the data or analysis. Keep the motto in your mind when you are looking at performance reports, when you are looking at new data, when you are performing analyses, when you are creating new indicators etc. It is so tempting to use the data we have with insufficient understanding, to use it incorrectly to support our points of view, to assume that 'doing data is easy'. But every one of our decisions is based on some kind of data. The wrong data, the wrong indicator (or KPI) or the wrong analysis leads to the wrong conclusions and the wrong decisions.
I look at my family crest and the knight's armour. I like the idea that my distant relatives were knights, fighting for justice. I'm not much of a fighter, but as an analyst, as a manager, as a leader, as a teacher it is up to me to defend the truth. It is my responsibility to protect it and ensure that those who are trained by me also understand the importance of data and how they can also become guardians of its truth.
So in all of the pressures and restraints of work, the challenges and short deadlines, the requests to push out data remember:
Magna est Veritas - Great is Truth