Statistics for everyone: ch.39 Continuous variable as interval variable

easier R than SPSS with Rcmdr : Contents

ch.39 Continuous variable as interval variable

A numeric variable can be nominal variable by dividing it into several intervals.

Name the new variable (red square), divide it by how many (green square), and set the new variable to be numeric and the same number (blue square).

The new variable has been coded as 1, 2, and 3.

Let’s summarize this data.

You can see that the new variable is distributed by a number of (67, 66, 67) as a nearly equal number.

Now let’s name it and do it with the K-means clustering method.

Give it a proper name.

I got the result and it was coded in A, B, C.

In summary, of course, the numbers are different. K-means clustering uses a method of tying similar things together according to the nature of those variables. When you run a marathon, divide it appropriately, just as the leading group and the group that runs a little further behind it. Developed contry and developing contry are divided according to the nature of those variables.

‘K-means clustering’ is also available on other menus, so let’s try it.

‘K-means clustering’ can also be clustered by selecting 1 or more variables , but here we will select only one variable as defined earlier.

Let’s set it up as above.

A new ‘KMeans’ variable has been created. If you compare it to the variable next to it, you will see that A = 2, B = 1, C = 3.