The other day I heard on NPR that “Jobless claims jumped to 520,00.” That’s all they said with regards to the number. There was no background given before or after. No information on what the jumped from. What the historical % changes are during recessions…Nothing. What a worthless statement when given by itself! I see numbers reported like this all the time. Not just by news organizations, but also by supposedly scientific studies.
If you are reporting numbers like this, please, for the love of all that is good and mathy, stop.
There are two elements that are required to give meaning to a number. First and foremost, there must be a relative vector. What I mean is that we must be given not only the direction of the number, but the % change in that direction. In order for there to be “relativeness,” there must be a before and after state – ideally with one raw number and % change. For example(I’m making all the following example numbers up), “The Jobless claims jumped to 520,000 - a 5% increase since the beginning of 2008.” In order for it to have “vectorness”, it needs to be spelled out in multiple dimensions. For example, “50% of the new jobless claims are related to the auto industry.”
The second element that must be present is environment. The environment is the background information that frames the number and puts its relative vector in perspective. This is the area that we, as willing information receivers, are failed the most. We need to know what has historically happened during similar periods (recessions). For example, “During recessions, jobless claims tend to jump 8% before falling again.” Also, “The auto industry tends to where a large % of the new claims come from.” Not only that, but we should be told what the new jobless claims situation is like across the world. Are we losing jobs at a faster or slower rate than other large countries? For example, “Germany has had a 15% jobless claims jump over the last 9 months.” The enviroment can be expanded upon from many angles and it is up to the data reporter to find the telling environmental numbers and give it to us.
The following is an example statement of a non-naked number just barely fulfilling requirments to give the number meaning:
Jobless claims jumped to 520,000 – a 5% increase since the beginning of 2008. 50% of the new jobless claims have come from the auto industry and 30% have come from the financial industry. During recessions, jobless claims tend to jump 8% before falling again and the auto industry tends to be where the biggest loss of jobs comes from. Germany has experienced a 15% jump in job losses. Japan 10%. Switzerland 2%. And Canada has actually experienced a 4% decline in job losses.
It really must be hard for people to look up and give us that information so the number can have meaning. Or are we expected to get everything after the original number just by tone of voice? I don’t know, but its frustrating to continue to hear people state a naked number and then act as if they’ve actually told me something. “Scientific” studies breach the naked number public contract in a different way, but I’ll leave that for another time.
