Normal Distributions, sometimes called Bell curves, are a standard curve defined by two values, the Mean and Standard Deviation (SD).
Refer:
http://en.wikipedia.org/wiki/Normal_distribution
Your data has a mean of 3.47 and a SD of 0.561447
This means that data has an average of 3.47 and that 99.7% of the data will be between Mean - 3 x SD and Mean + 3 x SD = 1.78 - 5.15
The Blue line is the Normal curve for the Mean and SD used based on you data
The Blue line is not based on a range but is calculated using Named Formula specifically x and y
Your data has been plotted as Circles
Each circle is plotted from Column N & P
N is known as a bucket list, specifically it is a collection of buckets which is used to tally (Column O) the values that surround each centroid
eg: the Value 3.6 (N16) has 5 values in +/- 0.1 from 3.6
so between 3.5 and 3.7 there are 5 values in your original list from Column K
Column P is purely the frequency of column O ie: Bucket Count / Total Count
You can see that the circles on the chart don't forma smooth curve as the blue line does, in fact it is heavily weighted between 3.4 and 4.0 with a small tail around 2.4-2.6
This may be either because :
1. There is only 20 data points or
2. The data may not be Normally Distributed
The Red Crosses simply show where each data point falls on the Normal Curve, supporting that the sample size may not be normal as discussed above
I am not a statistician or mathematician and so you should probably do some more research on this if you are going to start modelling something based on a Normal Curve of this data.
