The mean is one of those statistics that is more sensitive to outliers, and hence to mistakes in data, than it is to uncontaminated data.
Here's a pseudorandom sample of normally distributed values with mean 10 and variance 1 in sorted order:
8.3, 8.3, 9.0, 9.1, 9.2, 9.3, 9.5, 9.5, 9.7, 9.8, 9.9, 9.9, 10.2, 10.3, 10.4, 10.7, 10.9, 11.1, 11.3, 11.8
(Actually I've rounded them to the nearest 10th for easier reading.)
Their mean is 9.91.
Now let me add one outlier to the sample, say 4.2. Now the mean is 9.64. The original mean was only 0.09 away from the true mean, this one is four times as far away.
Of course it must also be said that one must be extremely careful about discarding data. Sometimes what appears to be an outlier has the most interesting information.
Chat with our AI personalities
The mean is "pushed" in the direction of the outlier. The standard deviation increases.
The median is least affected by an extreme outlier. Mean and standard deviation ARE affected by extreme outliers.
An outlier does affect the mean of the data. How it's affected depends on how many data points there are, how far from the data the outlier is, whether it is greater than the mean (increases mean) or less than the mean (decreases the mean).
An outlier will pull the mean and median towards itself. The extent to which the mean is affected will depend on the number of observations as well as the magnitude of the outlier. The median will change by a half-step.
The range.