Misrepresenting the Data
Perhaps the most basic error in the use of empirical data is simply misrepresenting it. This can occur in a number of ways.
One possibility is simply deliberate distortion, claiming that a data set proves something when it doesn’t. If people have an agenda, and set out to prove it, they may reach for the first bit of evidence they can find that even seems to fit their position. Closer examination may show that the evidence isn’t quite as supportive as was first claimed. Alternatively, someone confronted with potentially problematic evidence for their position may misrepresent it to make the problem go away.
A similar error can be committed accidentally. Sometimes when people look at a data-set they see what they want or expect to see, rather than what is actually there. The effect of our presuppositions on our interpretation of evidence should not be under-estimated. It can lead to conclusions being drawn which simply aren’t supported by the evidence.
A further way in which data may be misrepresented is if it is presented selectively. A varied data set can be described focusing in on certain sections of it. The data set as a whole is thus misrepresented; it is effectively replaced by a new set comprising of unrepresentative data.