Microsoft Office Tutorials and References

In Depth Information

**Stuck in the Middle with AVERAGE, MEDIAN, and MODE**

Customer

Total Amount Spent Last Year

A

$300

B

$90

C

$2,600

D

$850

E

$28,400

F

$300

The mean (using the AVERAGE function) is $5,423.33. The median is $575,

and the mode is $300. These three amounts aren’t even close! Which one best

represents the typical amount that a customer spent last year?

The issue with this set of data is that one value — $28,400 — is so much

larger than the other values that it skews the mean. You may be led to

believe that each customer spent about $5,423. But looking at the real

values, only one single customer spent a lot of money, relatively speaking.

Customers A, B, C, D, and F spent nowhere near $5,423.33, so how can that

“average” apply to them?

Figure 9-2 shows this situation in which one value is way out of league with

the rest — sometimes called an
outlier
— which makes the average not too

useful. Figure 9-2 also shows how much the mean changes if the one

spendthrift customer is left out, but if you leave out any other customer, there is

very little change in the mean.

Figure 9-2:

Deciding

what to

do with an

unusual

value.

In Scenario 2, Customer E is left out. The mean and the median are much

closer together — $968 and $850, respectively. Either amount reasonably

represents the mid value of what customers spent last year.