&textDisplay0=In this section we're going to show how we go about analyzing this data, and in particular if we calculate the appropriate percentages we can really start to understand the relationship between the two variables we've got here, music genre and cause of death.

&textDisplay1=Let's look at the table carefully; I'll describe some parts of it. First of all you'll notice that it's got two rows, that is one for rhythm and blues and one for the rock musicians and five columns relating to the five types of death. Notice it's fairly arbitrary which way round we've drawn it - we could have drawn it so that it had two columns and five rows - that would have been just as clear. You'll also notice because it's a five by two table, we've got ten what we call cells. And even just by looking at the cells as they are now some of the patterns start to become apparent. For instance there are it seems not a single case of a rhythm and blues musician who's died from a drugs or alcohol related death - this is what we call an empty cell. Whereas the same can't be said of rock musicians where you can see from the top of the list there are many cases of rock musicians who have died from alcohol and drugs related death.

&textDisplay2=You'll see from the table now that we've put numbers into each cell instead of the list of the names of the bands. These numbers are known as the counts in each cell sometimes called the frequency or the number of cases. For instance the empty cell tells us that there were zero cases of rhythm and blues deaths from alcohol or drugs, but underneath it we can see that there were 42 rock musicians whose deaths were attributed to alcohol or drugs.

&textDisplay3=What we're going to do now is transform these numbers into percentages and there's a lot more to this stage than it might first appear but it's very important to get this right if you're going to correctly interpret the data. There are three ways in which one might convert these numbers into percentages. One could have them as percentages of the total number of bands - in this case we've got 208 ‘cases’ altogether. Or we could have percentages that add up to a hundred within each of the columns for each of the different causes of death. Or finally we could have row percentages adding up to one hundred percent separately for the rhythm and blues and for the rock musicians. Let's think about how we're going to make that decision. The critical thing is to decide between these two variables, if there's a cause and effect relationship that one's assuming and if so which is the cause and which is the effect. There's nothing in the data to tell you this and there's nothing the computer can do to determine it, this is your decision. From your knowledge of the world, you need to make that decision. Sometimes it's straightforward, sometimes it's quite difficult or contentious, but here we can argue that it's the sort of music that one's involved in that might affect one's lifestyle and therefore the cause of death. So we're going to assume here that the music genre is the cause and the type of death is the effect or the outcome. Having made that decision now we will run the percentages so that they add up to a hundred within each of the categories of the 'causally prior variables'. So here the percentages are going to add to one hundred, separately for the rhythm and blues musicians and for the rock musicians - in other words the percentages are going to add to one hundred along each of the rows.

&textDisplay4=Let's start putting percentages into the table instead of just the simple counts or frequencies. Starting with the first cell we've got 12 cases there out of the total row of 37 cases. The total number of rhythm and blues musicians so to calculate the percentage we'll divide 12 by 37 which is just over a third of cases or about 34%. The next cell, zero is clearly going to be 0%. Four of the R%26B musicians committed suicide, that's 4 out of 37 or just over 10 percent, 6 of them were murdered, which makes 16 percent and the final 15 who died of medical causes make up 40 percent. So here we can see among those categories of death the single largest one is medical. Going on to the rock musicians we're going to do the same here but because there are a lot more in this category we're going to divide by 171 here. So starting again in their first cell, 46 of them, which is just over a quarter or 27 percent died of accidental deaths; 42, almost a quarter of their deaths were related to alcohol or drugs. Twenty, or 11.7 percent, committed suicide, 7 of them were murdered, which is 4 percent of the Rock musicians, and finally the other 56 we've lumped together as “medical causes of death”.