When we have bivariate data, we want to determine what sort of relationship the two variables have. As the independent variable (x) changes notice how the dependent variable (y) tends to change. Just by observation, we may notice the following:
A simple relationship: if the distribution of points appears to follow a trend either linear or non-linear depending on if the points appear to follow the shape of a line or not.
Consider being given an x-value that doesn't correspond to any data point we have. Does the data set give us an idea of what y-value that point should have to fit in with the rest of the data? If yes there might be a relationship. If not, there might be no relationship between the variables.
Outliers: in a scatterplot, any data points that are very different from the other data points will be quite obvious especially if the rest of the points appear to have a relationship.
Even when two variables have a relationship, it may not be a causal relationship. We cannot say for sure that a change in the value of x causes y to change or that the value of y causes a corresponding value of x even when a relationship is apparent. It may be that both x and y have a relationship with some other hidden variable, which creates an indirect relationship between x and y.
The scatter plot shows the relationship between sea temperature and the amount of healthy coral.
Which variable is the dependent variable?
Which variable is the independent variable?
The following scatterplot shows the height and weight of six students:
How tall is the student who weighs 42kg?
How tall is the tallest student?
The price of ten houses are graphed against the house's land area on the following graph:
Describe the relationship between land area and house price in the data.
To determine the relationship two variables have: as the independent variable (x) changes describe how the dependent variable (y) tends to change.