The analysis provided by the analyst is recast in a way that looks like it supports a desired conclusionĬonfirmation bias not only sneaks its way into the process of applying data analysis, but it can also influence the entire process:.

The analysis is re-done until it eventually yields a palatable conclusion.

Someone wants to use analytics to support their agenda or aim.

Towards Data Science lists three ways how confirmation bias sneaks to our process of data analysis: While it’s a human trait, confirmation bias clouds our judgment and distorts our perception of reality. Fs.blog defines confirmation bias as our tendency to cherry-pick information that confirms our existing beliefs or ideas. Everyone has committed it at least once in a while. You have just fished for data.Confirmation bias is one of the most common biases humans have. You would like to show progress, so you decide to include them in your report to management. Your company has no business in Ohio or Louisiana, and it there is no reasonable conclusion why these correlations exist, even though the p-values are less than 5%. Of the 100, two show statistical significance: interest rates on boat loans in south Louisiana and household debt in north Ohio. You download a database from a government website and decide to run a series of automated regressions. Imagine that instead of choosing only four macroeconomic indicators, you chose 100.

This could have gone another way…īUT wait! This could have gone another way. An honest analyst would not make this claim, but show that there is no relationship. You had conflicting evidence of the statistical significance of two correlations, but since one side of the evidence supports a personal claim you would like to make, you decide to accept it. Since you’re in a bind, you decide to claim that GDP and Employment rate have a strong relationship to revenue, and that you’re going to explore this further.

Data dredging def