# Data Fallacies to Avoid

![Im2]
**Cherry Picking**  
Selecting results that fit your claim and excluding those that don’t.

**Data Dredging**  
Repeatedly testing new hypotheses against the same set of data, failing to acknowledge that most correlations will be the result of chance.
![Im6]
![Im7]

**Survivorship Bias**  
Drawing conclusions from an incomplete set of data, because that data has ‘survived’ some selection criteria.
![Im15]

**Cobra Effect**  
Setting an incentive that accidentally produces the opposite result to the one intended. Also known as a Perverse Incentive.
![Im1]

**Sampling Bias**  
When something happens that’s unusually good or bad, it will revert back towards the average over time.
![Im9]

**Overfitting**  
Creating a model that’s overly tailored to the data you have and not representative of the general trend.
![Im4]

Drawing conclusions from a set of data that isn’t representative of the population you’re trying to understand.
![Im10]

**Regression Towards the Mean**

**False Causality**  
Falsely assuming when two events appear related that one must have caused the other.

**Gerrymandering**  
Manipulating the geographical boundaries used to group data in order to change the result.
![Im12]

**Gambler’s Fallacy**  
Mistakenly believing that because something has happened more frequently than usual, it’s now less likely to happen in future (and vice versa).
![Im11]

**Simpson’s Paradox**  
When a trend appears in different subsets of data but disappears or reverses when the groups are combined.
![Im14]

**Hawthorne Effect**  
The act of monitoring someone can affect their behaviour, leading to spurious findings. Also known as the Observer Effect.
![Im13]
![Im3]

**Publication Bias**  
Interesting research findings are more likely to be published, distorting our impression of reality.
![Im8]

## McNamara Fallacy  
Relying solely on metrics in complex situations and losing sight of the bigger picture.
![Im5]

**Danger of Summary Metrics**  
Only looking at summary metrics and missing big differences in the raw data.