You will be creating a plot showing the relationship between the metalicity (feha) and the galactic lattitude (b). In order to do this you will have to read in the feha.csv file (refer back to the tutorial if you are unsure how to do this.) and plot the b (column 1) on the x-axis and feha (column 3) on the y axis.

 Once you have done this it should look something like this. If it does not please go back and try again. Once you are satisfied with your result take a moment and look at the plot. See if you can distinguish any trends or structure in the plot.

If it still looks like a scattering of random points do not be discouraged, data will often look random unless you know what you are looking for. In order for us to see the structure of this data we will use a technique which will bring out an aspect that is very difficult to see with the naked eye.

What we will use is a histogram. This is a plot which shows the frequency of certain values, and will show us if there are some values that are more common than others. Lets make a histogram of the feha values. Begin by creating a histogram with the default number of breaks (do not include a break argument). It is clear that there is some clear structure, in fact it is a classical gaussian distribution, a statistical distribution of values present throught all of science.

There is however further structure that can be found in the data. Create a series of histograms with an increasing number of breaks (you will not need to go past 50 or so). As you increase the number of breaks the "resolution" of the plot increases and you are able to make out new structures.

What you are seeing is called a bimodal distribution, or simply a distribution of values centered around two specific values.

Back Next