Recruiting a Candidate

Introduction

This example is about Jeane who has to evaluate different candidates in scope to engage a new employee for her company. Jeane met 4 candidates today : Dave, Amanda, Jennifer and John. Each of them has of course his own strengths and weaknesses. In order to select the candidate in the most objective way, Jeane has established 5 criteria to base her decision on. Those are :

the experience of the candidate,
his salary expectation,
his technical skills,
his soft skills,
the number of spoken languages.

Figure 1 - Click to enlarge

As the candidates represent the different possible alternatives, they are listed in the "Alternatives" tab of the start interface (Figure 1). The criteria are listed in the Criteria tab (Figure 2).

The Evaluation tab, represented in Figure 3, is a grid that contains an alternative in each row and the criteria in each column. Basically, the values in that table represent the performances of the candidates for each of the criteria.

Figure 2 - Click to enlarge

Figure 3 - Click to enlarge

Analysis

By going into the menu Analysis -> Visualization -> Global Visual Analysis (or by typing CTRL + G), one can see a visual representation of the decision problem. In this chart, the candidates are represented by the points and the criteria are represented by the axes.

If an alternative goes far in the direction of an axis (click on the axis extremity to display the projections), it means that it performs well on the related criterion. For instance (see Figure 4), Jennifer is the most experienced candidate followed by John. On the other hand, Jennifer is not good regarding the Salary expectation criterion as she has the highest salary expectations. Dave is the candidate with the smallest salary expectation, as it can be seen in Figure 5.

If two axes go in the same direction (and thus are close to each other) that means that the related criteria are correlated! On the other hand, if two axes go in two opposite directions, it means that the related criteria are anticorrelated / in conflict. This can be easily observed for the “Salary expectation” and the “Experience” criterion that go in opposite directions. This means that in average and for this specific set of candidate, the most experienced candidates have the biggest salary expectations. This may sound logical but let’s emphasize that the configuration only depends on the data.

Figure 4 - Click to enlarge

Figure 5 - Click to enlarge

Let’s open the Global (numerical) ranking by going in into the menu Analysis -> Ranking and click on the Ranking menu item. We do now have the numerical ranking of the different alternatives. The most preferred candidates are John with a score of 0,23 followed by Amanda with a score of 0,14 (see Figure 6).

Figure 6 - Click to enlarge

Figure 7 - Click to enlarge

Let’s open the stability intervals tools (Analysis -> Sensitivity -> Stability Intervals). They indicate, for each critrion, the interval in which the weight can be changed without affecting the ranking. For instance, you can see in Figure 7 that the experience (which is weighted to 20%) can be changed from 1,8% to 40,7% without changing the candidate who is ranked first. We can check it in real time by opening the walking weights (Analysis -> Sensitivity -> Walking Weights or CTRL + W). You can easily put this new tab in the left part on the main window by drag and dopping it. Select the experience criterion in the tree and move the slider to 41%. For this new weight value, Jennifer arrives first in the ranking as more importance is now given to the experience (Figure 8). On the other hand lower the importance of the experience and you will see that Amanda will come in first position (Figure 9). We can see with this quick analysis that John seems to be a good choice for this recruitment as he is well scored and stays first even for a various range of weights.