There are basic patterns by which physics variables are related to one another. Such patterns include linear relationships, quadratic relationships, inverse relationships and non-relationships. These patterns can be identified by an inspection of data tables and graphs and expressed by verbal statements and equations.
 

There are three representations. You must choose which one is NOT consistent with all the others. The question targets the difference between a linear relationship and a quadratic relationship. The information below about these two types of data patterns will help you to answer this question.

Linear Relationship: Studies of variables in Physics will often show the pattern of a linear relationship. As the name suggests, two variables that are linearly related will be represented by a straight line on an x-y graph. The significance of this is that any given change in the independent variable (x) will always produce the same change in the dependent variable (y). For instance, a 1-unit change in the value of x may produce a 2-unit change in the value of y; if linearly related, then this ratio of the change in y value to the change in x value is always the same ratio. Since this ratio is what we refer to as slope, then one could say that two quantities that are linearly related will result in an x-y graph that has a constant or unchanging slope. This will be demonstrated in a carefully contrived data table by the fact that for the same change in x-value from one row to the another row will always result in the same change in y-value for those two respective rows ... regardless of which two rows are selected. Linearly related quantities have data values that can be described by an equation of the form y = m•x + b where m is the slope of the line on the x-y plot and b is the y-intercept.

A Special Type of Linear Relationship: A very common data pattern that is observed in Physics is the directly proportional pattern. It is a special type or sub-set of the linear relationship. Two variables that are directly proportional to one another will be represented by an x-y plot that shows a straight line (as in a linear relationship) with a y-intercept of 0. As such the typical y = m•x + b equation becomes a y = m•x equation.  In such instances, a doubling of the independent variable (x) results in a doubling of the dependent variable (y). And a tripling of the independent variable causes a tripling of the dependent variable. In more general terms, the value of y will change by the same factor that the value of x is changed by. An x-y data table can be inspected to see if it follows this pattern; if it does, then the quantities x and y are both linearly related and directly proportional to one another.

Quadratic Relationships: There are numerous patterns in Physics that could be described as displaying a quadratic relationship. When an equation  is written for two quantities (call them x and y) that have a quadratic relationship, the highest exponent in the equation will be two (not 1, not 3, but 2). The equation relating x and y has the form

y = Ax² + Bx + C

where A is a non-zero constant and B and C can be any number (including 0). A plot of two quantities that have a quadratic relationship will be curved (not straight) and have the shape of a parabola.  Often times, the constants B and C in the equation above are 0. In such cases, the equation becomes y = Ax² and the value of y is proportional to the square of the value of x. As such, a doubling of x causes a quadrupling of y. And a tripling of x causes a nine-fold increase in the value of y. In general terms, when the variable x is changed by some factor, the value of y will change by the square of that factor. An inspection of the x-y data table will often reveal such a pattern. By strategically comparing two rows of the data table, one will notice that if the value of x in one row is twice the value of x in another row, then the corresponding y value will be four times greater in one row than in the other row.