Recruitment and Selection in Canada 6th Edition Solution Manual

CHAPTER TWO: FOUNDATIONS OF RECRUITMENT AND SELECTION I: RELIABILITY AND VALIDITY Emotional Intelligence 1. What do you think? Do the data that you collected from the Queendom.com exercise help you to answer this question? Is there a relationship between job performance and EQ? Can you support your answer with any empirical data? How can the construct of EQ be improved? Is it too broad? Is EQ simply another aspect of personality? ANS: It is only possible for us to discern the relationship between IQ, EQ, and personality from our Queendom.com data. If these scores demonstrate low correlations with one another, as expected, then we can say with some assurance that the three measures are measuring different constructs (i.e., they have divergent validity). Further, if the test scores remain consistent over the two waves of data collection then we can also assert that our data is fairly reliable. However, because it is not clear that Queendom.com uses the Bar-on EQi, and because Queendom.com does not include a measure of performance, we cannot answer whether the Bar-On EQi scores are statistically related to performance scores. We also cannot really say if EQ is just another form of personality from the Bar-On EQi data, though the Queendom.com data might imply that the two are statistically unrelated. It is possible that personality and EQ share common variance though this would not necessarily indicate that the variables are empirically identical constructs. Current measures of EQ may be too broad to fully capture the content domain of the construct. Scales can be improved via scale validation strategies at both the construct and measurement levels (see Figure 2.2). 2. If you planned to use EQ as part of your selection system, discuss the steps that you would take to ensure that you were able to make reliable and accurate inferences about job performance in your work situation. That is, what would you have to do to show that your measure was reliable and valid? ANS: There are a number of steps to take to ensure the reliability and validity of our chosen measure of EQ. As an HR manager, you should start by checking the existing reliability and validity data of your measure in the extant literature (these indices are sometimes easily obtained by conducting a literature review on the topic/construct at hand). In the absence of any such data, or if the data are unclear, you should ensure that factors impacting the potential reliability, including chance, lack of standardization, and temporary characteristics, are minimized at test time. In terms of ensuring validity, you may take strategies outlined in Figure 2.2 in the textbook and, for small samples, the steps outlined in Recruitment and Selection Notebook 2.1. To avoid range restriction, you should look at your validity coefficients to ensure that scores from subgroups are similar to those from the overall sample. To avoid measurement error, you should ensure the test is reliable over at least two points of data collection. To avoid sampling error, you should ensure the tests are taken by the appropriate samples (e.g., applicants and job incumbents) in as large a population as possible (estimates from small samples will likely be quite variable so corrections may need to be applied). You should also take steps to ensure that your test is perceived to be fair and unbiased by your employees, as described in Recruitment and Selection Notebook 2.2, because adverse reactions to tests can negatively impact, or attenuate, validity scores.