Because both distributions, visualization (V) and no visualization (NV), are skewed, the psychologists conducted a simulation to test for a difference in medians rather than means. For each trial of the simulation, the 20 values of the total number of attempts observed in the experiment were combined into one group and then randomly split into two groups of 10. The difference in the medians (V-NV) of the groups was calculated for each trial. The following dotplot shows the difference in the medians for 100 trials of the simulation. Using the observed difference in medians (V−NV) and the results of the simulation, estimate a p-value for a test for the difference in medians. Show the work needed to calculate this p-value.