Suppose that a researcher does a study to see how level of anxiety (A1 = low, A2 = medium, A3 = high) is used to predict exam performance (Y).
Set up orthogonal coded dummy variables (O1, O2) to represent linear and quadratic trend, and run a regression analysis to predict exam scores from O1 and O2. What conclusions can you draw about the nature of the relationship between anxiety and exam performance?