Scatterplot

The scatterplot (or @scattergram) helps us analyzing @numerical variables, and we can quickly identify several features of our variables by making a scatterplot of our data. Like a bivariate table, the scatterplot has two dimensions: X (independent) and Y (dependent). The scatterplot can find out if the X-variable influences the Y-variable, if there is a connection between them and if the there is a pattern.

The scattergram tells us: - if there is a relationship between the varibales. Two variables are associated if the distribution of Y change for the various conditions of X. - how strong the relationship is, by drawing a regression line through the centre of the spread. The strength is the spread. - what the direction of the relationship is. The direction can be detected by observing the angle of the regression line; positive or negative.

When we have the scatterplot, we are interested in checking for linearity. We can see if we can find some relationship between the variables and use that difference to predict something about a new value in our data. Each case has a value in each @variable, and is marked as a "dot" between X and Y. The "dot" is the combination of the two variables. If there is linearity in the scatterplot, the dots sholuld form something like a straight line. If not, the data are not suitable for analysis, or there is a lack of association between them. To find if there is linearity, we draw the regression line. The regression line is an average value through our data, and indicates the direction (whether the association between the variables is weak or strong) within the data. If the association is strong, the model for the regression line can be used to find one variable when you know the other.

See how to make scatterplot in excel and @scatterplots in R.