Today is World Mental Health Day.
Recently, I stumbled upon an engaging visualization representing the increase in the 'deaths of despair' due to the pandemic. Here is a link for the same.
Do use the arrow icons to get more information. For example, in one of the panels, you can play around the different levels of 'mental health care', 'employment status', and 'social connect' to see the effect of these variables on the number of deaths of despair.
In recent years, the overbearing use of social media has been cited as one of the leading causes of mental health issues. This is because so many of us, at one point or other, have felt jealous of someone else, looking at their lives on social media. But, as students of statistics, we must remember that the data on social media are the classic example of data 'missing not at random'.
What is meant by that?
We all have heard the problem of missing data in statistics and how it can affect our conclusions. There have been many developments in the field of probability and statistics about handling missing data. However, most of these focus on the data 'missing completely at random'.
On the other hand, what we see missing on social media is not at all random. We see hundreds of stories of weddings, new cars, new jobs, new babies, family picnics, festivities and whatnot. But we rarely see things like breakups, divorces, job losses, prolonged illnesses, infertility struggles, piled up debts on these platforms, even though these could be happening to the same people.
So, as statistics students, let us remember that the grass always looks greener on the other side because the data are missing not at random.
and here is another example of 'data missing not at random', ironically, received thanks to the social media ЁЯШО