Andy Wills

```
data
```

```
[1] 1 2 2 3 3 3 4 4 50
```

- The median is the middle number when the data is put in order

```
median(data)
```

```
[1] 3
```

```
data
```

```
[1] 1 2 2 3 3 3 4 4 50
```

- The mean is the sum of all the numbers (72), divided by the sample size (9), giving:

```
mean(data)
```

```
[1] 8
```

```
data
```

```
[1] 1 2 2 3 3 3 4 4 50
```

In this case, the mean (8) is bigger than nearly all the numbers. This means it's not very

*representative*of its sample.The mean will always be unrepresentative when there are a few numbers that are very high (or very low), compared to the rest.

In these cases, we sau the distibution is

*skewed*.

- Our
*income*data is skewed, so the median gives a better indication of average salary than the mean in this case.