For the non-zero case, what makes the change qualitative is that its direction but not its magnitude is specified.
[1] Typical exercises of qualitative economics include comparative-static changes studied in microeconomics or macroeconomics and comparative equilibrium-growth states in a macroeconomic growth model.
A simple example illustrating qualitative change is from macroeconomics.
The hypothesized relationships can be equivalently represented as signed functional relationships and signed partial derivatives (suitable for more than one independent variable): Qualitative hypotheses occur in earliest history of formal economics but only as to formal economic models from the late 1930s with Hicks's model of general equilibrium in a competitive economy.
[3] There Samuelson identifies qualitative restrictions and the hypotheses of maximization and stability of equilibrium as the three fundamental sources of meaningful theorems — hypotheses about empirical data that could conceivably be refuted by empirical data.