Welcome to the Analytix Thinking blog! The blog that is intended to help people think rightly about data and deciding what is true.
First, the intended audience is those who are analytically/quantitatively minded, but the exposition is also meant to be consumable by those who are curious about such matters, but without formal mathematical, statistical or algorithmic training. I hope that’s you! I hope to attract scientists from disciplines of science who generate data and try to make sense of it in the presence of variability and uncertainty. I hope to attract the non-scientific public audience who are trying to make sense of all the “breaking news” and headline stories about the latest research findings. To accommodate a broad audience, I will steer clear of complex math/stats and equations in favor of concepts and how to think about evidence – data, analysis and interpretation. This, in fact, can be more difficult than the math/stats itself since it gets to the heart of how we think about questions/problems with uncertain answers. I like to use stories and analogies to make the concepts more clear.
Second, the premise of this blog is two-fold (in no particular order):
- Provide commentary and guidance in response to the explosion of information – a lot of hype and some substance – on big data and analytics, including biomarkers (biologic and digital), electronic medical records, observational research, etc.
- Advocate for some fundamental changes in statistical thinking, including inference and estimation of drug treatment effects; changes in analytical thinking, including validation of analyses and interpretation of results.
This will lead to some commentary on the relationship between various analytical disciplines, by which I mean statisticians, data scientists, epidemiologists, economists, computer scientists, operations researchers, mathematicians, health outcome scientists and, in short, anyone who collects, analyzes and interprets data … data from the natural world (i.e. physical or social science) or data from the human-created world (e.g. business or politics).
As implied above, my view of “analytics” is quite inclusive – broad and deep.
Third, I realized that I will be using examples of research studies published in scientific literature or news reports that span a very wide range of topics and disciplines. I cannot hope to be knowledgeable in all of these areas. I do hope that my experience and knowledge about statistical concepts – deriving evidence from data, ascertaining what is likely or unlikely to be true, understanding variability, bias, multiplicity and a host of other inferential matters – can be successfully applied across a broad swath of scientific and social problems.
I intend to be fair and balanced in my blogs by addressing the research findings or content while refraining from attacking the scientists or the reporters themselves. I realize that this is a fine line to walk. I also realize that putting my ideas – criticisms, questions, alternative explanations, novel ideas – in the public view via this blog will make me vulnerable to criticisms as well. Such is life. “If there is no struggle, there is no progress.” (Frederick Douglas) [I have also heard that George Bernard Shaw said, “There is no progress without conflict.” Or some such statement. Any help on identifying this is appreciated.]
Lastly, as a follow-on to the Third item above, I am open to your comments and queries and even disagreements. I look forward to you helping me refine and improve my analytics thinking. I am anxious to hear about your stories or interesting articles, whether they be from a scientific journal or the lay press. Contributions are welcome.
With that very brief review of the purpose and scope of this blog, I hope you are ready to read the first entry. Welcome as you join me and (hopefully) many others on our collective journey through analytics thinking!