You can use a histogram
of the data overlaid with a normal curve to assess the normality
of your data. A normal distribution is symmetric and bell-shaped, as indicated
by the curve. (Note that the curve is truncated where it extends beyond
the borders of the graph.) It is often difficult to evaluate normality
with small samples.
Example Output |

|

Interpretation |

|
The precipitation data do not follow the normal curve. This is largely
due to the outlier
at the far right of the graph. Without this outlier, the data look much
more normal, as illustrated below.
