Chat with our AI personalities
The mean will increase very substantially, the median will remain unchanged.The mean will increase very substantially, the median will remain unchanged.The mean will increase very substantially, the median will remain unchanged.The mean will increase very substantially, the median will remain unchanged.
If the wide range is evenly spread between the very small and the very large (the distribution is symmetric) then there is not much to choose between the median and the mean. If not, the median will have some advantages as a measure of central tendency.
The Median is a form of average. Mathematically, the average is the sum of all values, divided by the number of values there are. This is known as the arithmetic mean. An Average is meant to give a representative value of all of your values. If you have a lot of numbers that are very similar and then one that is massively out of proportion with all the others, then the mean doesn't give a representative value. There are also two other forms of average, other than the arithmetic mean. They are the Mode and Median. The Mode is the most common value in a list and the Median is the middle value when all are arranged in numerical order. Sometimes it is better to use them.
All three are types of averages. The mean is what the typical person means when he says "average": add up all the values and divide by the number of values. The median is the middle value: if you arrange the numbers from low to high and then take the one exactly in the middle of the sequence (or the mean of the two middle values, if there are an even number), that's the median. The mode is the value that occurs most often.In what statisticians call a normal distribution, the mean, median, and mode will all be the same or at least very close. However, many sets of data are not "normal distributions", and in those the three values can be very different. For example, consider the set 5,5,5,20,20,55,100. The mode is 5, the median is 20, and the mean is 30.
The mean is not always better. It is better in the sense that it uses information from each observation but that also means that it is more likely to be distorted by outliers. Under the law of large numbers the mean has some very useful and well understood properties and that make it a good test statistic.