Schiemann and Ulrich's (Reference Schiemann and Ulrich2017) discussion of opportunities for industrial–organizational (I-O) psychology speaks directly to the growing movement of companies eliminating their performance ratings. One of the main issues that the focal article highlights is the need for strong analytics and measurement. A core domain of industrial psychology historically has been employee performance measurement. However, by some accounts, as of September 2015, at least 51 large companies were in the process of eliminating their performance ratings (Rock & Jones, Reference Rock and Jones2015). Whether eliminating performance ratings is a good idea is a subject of debate in the I-O psychology community. In addition to a focal article in the summer of 2016 edition of Industrial and Organizational Psychology (Adler, Campion, et al., Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016), several sessions addressed this topic at the 2016 Society for Industrial and Organizational Psychology convention (e.g., Adler, Colquitt, et al., Reference Adler, Colquitt, Kamin, Huth, Walzer and Merges2016; Hettal, Garza, Levy, & Cleveland, Reference Hettal, Garza, Levy and Cleveland2016; Hunt et al., Reference Hunt, Mansfield, Burlacu, Karavis, Kim and Levy2016; Roch & Gorman, Reference Roch and Gorman2016). The I-O psychology community appears to be divided on the value of performance ratings. In my opinion, the recent drive to abandon rigorous performance measurement appears to be a step backward rather than a step forward.
One of the seven areas of opportunities for I-O psychologists discussed by Schiemann and Ulrich is analytics/information. Even though employee performance measurement is just one domain within this area, it appears extremely salient given its importance to our profession's relevance and impact in terms of workplace performance. Schiemann and Ulrich (p. 14) point out that “For many HR professionals, measures have been [a] foreign language” and that measures and their development and use represent one of our core areas of expertise. Schiemann and Ulrich (p. 15) include arguments stressing the importance of differentiating among employees, focusing on “Who are the high performers, what are the critical roles, and what are the most strategic requirements of the business?” They give the example of Mark James, named chief human resources (HR) officer of the year by Human Resource Executive magazine in 2013, who attributed his organization's success to constant measurement. Schiemann and Ulrich conclude that according to their thought leaders, a more strategic approach to measurement and analytics is needed; there is no mention that HR metrics, such as performance ratings, should be eliminated.
In fact, Schiemann and Ulrich's discussion of analytics can be seen as a call for what we have been doing since the founding of I-O psychology: conducting a work analysis (“knowing the business—how widgets are produced”; p. 16) and assessing employees’ performance to differentiate between low, average, and high performers. It should be noted that the emphasis is on distinguishing high performers and not on identifying potential high performers. Identifying potential high performers is a selection issue, distinct from identifying who is indeed a high performer, an issue of employee assessment.
Performance ratings are indispensable when differentiating high performers from average or low performers in most cases, regardless of whether the ratings are on a traditional numerical measure from 1 to 7 or on a measure consisting of written performance descriptors, such as “needs improvement,” “solid performer,” and “high performer.” I-O psychologists have a long history of designing measures to assess employee performance. Even though there is debate regarding how well we measure performance (Murphy, Reference Murphy2008; Woehr, Reference Woehr2008), I would argue that managers, coworkers, customers, and so forth do a much better job of evaluating employee performance when using decision aids, such as rating measures, than when relying on gut intuition (Kleinmuntz, Reference Kleinmuntz1990). In the absence of rating measures to assess performance, if a company wishes to differentiate its high performers, the company is often left with gut impressions. Relying on bottom-line measures instead of ratings is often not advisable; often performance ratings are the only feasible way to measure and compare employee performance (e.g., Hunt, Reference Hunt2016; Murphy, Reference Murphy2008; Murphy & Cleveland, Reference Murphy and Cleveland1995; Tziner & Roch, Reference Tziner and Roch2016).
Too often bottom-line measures are used to assess employee performance because they are easy and not because they convey a complete picture of performance. The example that Aharon Tziner and I give in our commentary to Adler, Campion, et al.’s (Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016) focal article discussing getting rid of performance ratings is that the sales of detection devices of explosive materials will grow in regions suffering terrorist attacks, thus improving sales numbers of the sales personnel in these regions regardless of sales behaviors (Tziner & Roch, Reference Tziner and Roch2016). As Schiemann and Ulrich mention, organizations should measure what is important and not what is easy to measure. Bottom-line measures, at best, only provide one piece of information regarding employee performance.
This is not to say that work analysis and employee performance measurement systems do not need to evolve. One clear message in Schiemann and Ulrich's discussion of HR thought leaders’ opinions is that business strategy needs to drive measurement; work analysis and employee performance measures should be designed to reflect business strategy. However, this does not mean that performance ratings should necessarily continue to be used in performance management systems focused on providing developmental feedback to employees. As demonstrated by Cleveland, Murphy, and Williams (Reference Cleveland, Murphy and Williams1989), performance ratings tend to be used for three general purposes: (a) employee development; (b) administrative purposes, such as pay raises, promotion, placement, and termination; and (c) research purposes, such as validating selection measures and tracking talent in an organization; performance appraisal is more than just assessing employee performance. Perhaps the time has come to end the practice of using performance ratings for developmental purposes for the reasons discussed by Adler, Campion, et al. (Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016) and Pulakos and O'Leary (Reference Pulakos and O'Leary2011), but performance ratings are still needed for both administrative and organizational research purposes. How can an organization “focus on the connection of talent, customer-facing processes, and technology to customer and market goals” (Schiemann & Ulrich, p. 16) without good measures of talent—this is not possible by simply examining the number of widgets produced or monthly sales. The type of sophisticated analytics suggested by Schiemann and Ulrich are predicated on understanding employee performance (not potential performance), the strengths and weaknesses exhibited by employees, and connecting this information to other business information. Schiemann and Ulrich (p. 15) discuss this as an opportunity to leverage one of our primary distinctions from other HR professionals, our knowledge of measurement and the scientific method, “perhaps . . . the single biggest area for I-O impact—and visibility.” Now is not the time to abandon measures of employee performance, that is, performance ratings, just because they are difficult and often viewed negatively by employees and their managers. Measurement is something that we do know how to do well. Why not leverage this strength?
Schiemann and Ulrich's (Reference Schiemann and Ulrich2017) discussion of opportunities for industrial–organizational (I-O) psychology speaks directly to the growing movement of companies eliminating their performance ratings. One of the main issues that the focal article highlights is the need for strong analytics and measurement. A core domain of industrial psychology historically has been employee performance measurement. However, by some accounts, as of September 2015, at least 51 large companies were in the process of eliminating their performance ratings (Rock & Jones, Reference Rock and Jones2015). Whether eliminating performance ratings is a good idea is a subject of debate in the I-O psychology community. In addition to a focal article in the summer of 2016 edition of Industrial and Organizational Psychology (Adler, Campion, et al., Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016), several sessions addressed this topic at the 2016 Society for Industrial and Organizational Psychology convention (e.g., Adler, Colquitt, et al., Reference Adler, Colquitt, Kamin, Huth, Walzer and Merges2016; Hettal, Garza, Levy, & Cleveland, Reference Hettal, Garza, Levy and Cleveland2016; Hunt et al., Reference Hunt, Mansfield, Burlacu, Karavis, Kim and Levy2016; Roch & Gorman, Reference Roch and Gorman2016). The I-O psychology community appears to be divided on the value of performance ratings. In my opinion, the recent drive to abandon rigorous performance measurement appears to be a step backward rather than a step forward.
One of the seven areas of opportunities for I-O psychologists discussed by Schiemann and Ulrich is analytics/information. Even though employee performance measurement is just one domain within this area, it appears extremely salient given its importance to our profession's relevance and impact in terms of workplace performance. Schiemann and Ulrich (p. 14) point out that “For many HR professionals, measures have been [a] foreign language” and that measures and their development and use represent one of our core areas of expertise. Schiemann and Ulrich (p. 15) include arguments stressing the importance of differentiating among employees, focusing on “Who are the high performers, what are the critical roles, and what are the most strategic requirements of the business?” They give the example of Mark James, named chief human resources (HR) officer of the year by Human Resource Executive magazine in 2013, who attributed his organization's success to constant measurement. Schiemann and Ulrich conclude that according to their thought leaders, a more strategic approach to measurement and analytics is needed; there is no mention that HR metrics, such as performance ratings, should be eliminated.
In fact, Schiemann and Ulrich's discussion of analytics can be seen as a call for what we have been doing since the founding of I-O psychology: conducting a work analysis (“knowing the business—how widgets are produced”; p. 16) and assessing employees’ performance to differentiate between low, average, and high performers. It should be noted that the emphasis is on distinguishing high performers and not on identifying potential high performers. Identifying potential high performers is a selection issue, distinct from identifying who is indeed a high performer, an issue of employee assessment.
Performance ratings are indispensable when differentiating high performers from average or low performers in most cases, regardless of whether the ratings are on a traditional numerical measure from 1 to 7 or on a measure consisting of written performance descriptors, such as “needs improvement,” “solid performer,” and “high performer.” I-O psychologists have a long history of designing measures to assess employee performance. Even though there is debate regarding how well we measure performance (Murphy, Reference Murphy2008; Woehr, Reference Woehr2008), I would argue that managers, coworkers, customers, and so forth do a much better job of evaluating employee performance when using decision aids, such as rating measures, than when relying on gut intuition (Kleinmuntz, Reference Kleinmuntz1990). In the absence of rating measures to assess performance, if a company wishes to differentiate its high performers, the company is often left with gut impressions. Relying on bottom-line measures instead of ratings is often not advisable; often performance ratings are the only feasible way to measure and compare employee performance (e.g., Hunt, Reference Hunt2016; Murphy, Reference Murphy2008; Murphy & Cleveland, Reference Murphy and Cleveland1995; Tziner & Roch, Reference Tziner and Roch2016).
Too often bottom-line measures are used to assess employee performance because they are easy and not because they convey a complete picture of performance. The example that Aharon Tziner and I give in our commentary to Adler, Campion, et al.’s (Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016) focal article discussing getting rid of performance ratings is that the sales of detection devices of explosive materials will grow in regions suffering terrorist attacks, thus improving sales numbers of the sales personnel in these regions regardless of sales behaviors (Tziner & Roch, Reference Tziner and Roch2016). As Schiemann and Ulrich mention, organizations should measure what is important and not what is easy to measure. Bottom-line measures, at best, only provide one piece of information regarding employee performance.
This is not to say that work analysis and employee performance measurement systems do not need to evolve. One clear message in Schiemann and Ulrich's discussion of HR thought leaders’ opinions is that business strategy needs to drive measurement; work analysis and employee performance measures should be designed to reflect business strategy. However, this does not mean that performance ratings should necessarily continue to be used in performance management systems focused on providing developmental feedback to employees. As demonstrated by Cleveland, Murphy, and Williams (Reference Cleveland, Murphy and Williams1989), performance ratings tend to be used for three general purposes: (a) employee development; (b) administrative purposes, such as pay raises, promotion, placement, and termination; and (c) research purposes, such as validating selection measures and tracking talent in an organization; performance appraisal is more than just assessing employee performance. Perhaps the time has come to end the practice of using performance ratings for developmental purposes for the reasons discussed by Adler, Campion, et al. (Reference Adler, Campion, Colquitt, Grubb, Murphy, Ollander-Krane and Pulakos2016) and Pulakos and O'Leary (Reference Pulakos and O'Leary2011), but performance ratings are still needed for both administrative and organizational research purposes. How can an organization “focus on the connection of talent, customer-facing processes, and technology to customer and market goals” (Schiemann & Ulrich, p. 16) without good measures of talent—this is not possible by simply examining the number of widgets produced or monthly sales. The type of sophisticated analytics suggested by Schiemann and Ulrich are predicated on understanding employee performance (not potential performance), the strengths and weaknesses exhibited by employees, and connecting this information to other business information. Schiemann and Ulrich (p. 15) discuss this as an opportunity to leverage one of our primary distinctions from other HR professionals, our knowledge of measurement and the scientific method, “perhaps . . . the single biggest area for I-O impact—and visibility.” Now is not the time to abandon measures of employee performance, that is, performance ratings, just because they are difficult and often viewed negatively by employees and their managers. Measurement is something that we do know how to do well. Why not leverage this strength?