Illusion of objectivity

While working with data (Data Science) one often runs into mathiness of various forms.

Anything done with data seems objective, devoid of the whimsies of emotion and superstition, but this is not so.

A researcher makes a series of not completely objective decisions while using the data1. Which means that two of them starting with the same data and question can end up with very different conclusions2. There is also the danger of missing obvious things clouded by the hypothesis being investigated3.

And finally there is the corpBS version of it. One can never be sure if they are being intentional about it or not, but illusions of objectivity are very often created by the use of spreadsheets where numbers are filled and weighted averages are calculated, all while the numbers don’t mean anything, are soaked in emotions, or are actually just a Red-Amber-Green4 nominal variable.

  1. Researcher Degrees of Freedom ↩

  2. Many Analysts, One Data Set: Making Transparent How Variations in Analytic Choices Affect Results, Aug 2018 ↩

  3. A hypothesis is a liability, Sep 2020 ↩

  4. No one knows why they call it amber, when almost every sane person calls it orange. ↩

Notes mentioning this note

There are no notes linking to this note.

Here are all the notes in this garden, along with their links, visualized as a graph. If you don't see any nodes try zooming and panning in the grey area.