Figure 1 is a cluster and outlier analysis of the workplace 1-home-workplace 2 angle that reveals clusters of high or low values, as well as outliers of high or low values surrounded by clusters of opposite values. Anselin Local Moran’s I statistic for spatial autocorrelation calculates spatial correlation by taking into account the values of features and their location relative to each other. In the case of angles, the higher the value, the more opposite are the directions that partners take to reach their jobs. Clusters of high values are concentrated in the central part of the Island of Montréal, extending to Longueuil and Laval, and in Saint-Jérôme and Saint-Jean-sur-Richelieu, while clusters of low values are frequent around the centre of the Island of Montréal, stretching along Saint Lawrence River. This pattern could be explained by the high number of jobs available in the centre of the region and by high job accessibility by any mode for houses located in the centre. The repetition of the pattern to Saint-Jérôme and Saint- Jean-sur-Richelieu could be due to the fact that these cities still act as local centres and are not dependent on jobs found on the Island of Montréal.