How Did We Arrive at Today's Best Practice

Today’s performance management best practices are the result of ongoing efforts to address two key challenges that have plagued performance measurement since its inception:

• 

What type of performance should be measured – abilities, skills, behaviors, results? 

• 

How can we measure performance most reliably, accurately, and fairly?

To understand where we are today with performance management and why certain approaches have become best practices, you need to understand how they evolved over time, based on trial and error.

The start was defining rating standards  .  .  .

 • 

The beginning of formal performance management can be attributed to principles of “scientific  management,” developed by industrial engineers in the early 1900s. These principles emphasized the importance of defi ning standards against which to measure performance, an important best practice today.

Next, abilities were evaluated  .  .  . 

• 

During World War I (1914–1918), the performance of Army officers was evaluated, with a focus on assessing officer ability. These efforts marked the fi rst large-scale use of judgmental assessment and began to solidify the use of performance management systems in government and industry.

Scales were developed to measure job-relevant traits  .  .  . 

• 

In 1922, the Graphic Rating Scale was introduced.3 This scale was designed to elicit ratings of traits relevant to a job and was anchored with verbal anchors, numerical anchors, or both. Use of a rating scale was a signifi cant step forward. However, one problem with graphic rating scales was that the rating points were not well defined. For example, a scale might have assessed whether an employee “Exceeded,” “Met,” or “Failed to Meet” expectations, without articulating exactly what those expectations were. This left managers to develop their own interpretations of the rating scale points. Since some managers inevitably expect more than others. The result was that employees were held to different standards. Thus, graphic rating scales were limited because they did not provide suffi ciently defi ned standards that managers could use to systematically and fairly evaluate employees. The problem of suffi ciently defi ning performance standards to guide evaluations has continued to plague performance management.

Performance measurement evolved to assess work outcomes and behaviors  .  .  . 

• 

In the 1950s and 1960s, there was development and expansion of the types of performance rated, beyond job relevant traits. This happened because it was recognized that traits (e.g., conscientiousness), thought to underlie job performance, were not the most direct and meaningful thing to evaluate. As a result, attention shifted to more direct and observable measures of performance. These included both objective outcome measures (e.g., dollar volume of sales) and subjective (e.g., behavioral) measures. For example, in 1954, a behavioral measure, called the critical incident technique, was introduced by Flanagan.7 This technique focused on eliciting specifi c examples of effective and ineffective job behaviors and moved performance measurement away from traits towards observable behavior.

Objectives-based performance measures took center-stage  .  .  . 

• 

Also in 1954, Management by Objectives (MBO) was introduced by Peter Drucker in his book, The Practice of Management. MBO involves defi ning concrete and specifi c objectives so that both managers and employees understand what the employee is expected to achieve. MBO systems increased in their popularity and use, particularly in the 1970s. However, experience with MBO revealed several diffi culties, including the time it took to set measurable objectives for each and every employee, unforeseen events that required objectives to be continually modifi ed throughout the rating period, and defi ning objectives that were entirely within the employee’s control. These diffi culties led many organizations to abandon the MBO approach to performance management.

Behavior based performance measures took center-stage  .  .  . 

• 

The trend increasingly moved more towards measuring job behavior, including the development of pre defined behavioral standards against which employees could be evaluated. Assessing job behavior circumvented the problems associated with measuring objectives – that they often needed to be revised during the rating period, accomplishing objectives can be infl uenced by things outside an employee’s control, and considerable time is required to define individual, customized objectives for each employee. A major influencing force was the development of Behaviorally Anchored Rating Scales (BARS) in the early 1960s, which focused on quantifying behavioral job performance.8 These scales focused on assessing performance dimensions that represent the major job requirements (Figure 2.1). Within each dimension, specifi c behaviors anchor different rating levels, as shown in the example on p. 10. These behaviors provide managers with concrete exampel of the type of performance that is associated with different effectiveness levels.

Behavior based performance measures improved  .  .  . 

• 

The BARS rating format was an important step in developing well defined rating scales. However, one limitation of these scales was that it could be difficult to match an employee’s performance to the very specific behavioral examples used to anchor the rating scale. Even though an employee might be performing at a level represented by an example behavior, the employee probably would not have exhibited the exact behavior that appears on the scale. This required managers to infer which of the few scaled behaviors best matched an employee’s performance. Several variants of the behaviorally-based method followed in the late 1970s and early 1980s to address this issue.10,11 Rather than using only a few very specific behaviors to anchor the rating scale points, more general behavioral descriptions were used to anchor different effectiveness levels. The use of more general behavioral descriptions made it easier for managers to match their observations of employee performance to a rating scale point. Today, well defi ned behavioral standards remain a hallmark of effective performance management systems.

Civil rights put a focus on fairness  .  .  . 

• 

The civil rights movement of the late 1950s and early 1960s drew attention to the fact that minorities had systematically been denied equal opportunity in areas such as housing, education, and employment. The Civil Rights Act of 1964 and subsequent legislation was passed to rectify these inequities and prohibited discrimination in employment practices. Performance appraisals, which often serve as the basis for pay, promotions, and terminations, were required to be job relevant. This requirement was reiterated in 1979 with the publication of the EEOC Uniform Guidelines on Employee Selection Procedures. The implication of this legislation was that certain procedures to ensure job relevance needed to be followed in developing performance management systems. For more guidance on these procedures, see the Society for Industrial and Organizational Psychology’s, Principles for the validation and use of personnel selection procedures.

Multi source ratings gained popularity  .  .  . 

• 

Developing out of work in the 1940s, the 1960s and 1970s saw an increased focus on gathering performance information from rating sources other than managers, such as peers and customers. The idea behind collecting performance information from different sources is that, depending on one’s relationship to the employee, he or she will see different aspects of performance. For example, a customer is probably in the best position to judge someone’s customer service skills. Likewise, a direct report will see aspects of performance dealing with providing feedback and mentoring that an individual’s manager may never directly observe. In the early 1990s, formal multi source or 360 degree feedback programs further evolved out of organizational trends such as employee involvement, self-managed work teams, and an increased focus on customer satisfaction. These programs quickly gained widespread popularity in the workplace.

Competencies took center-stage  .  .  . 

• 

Through the 1990s and into the 2000s, organizations have increasingly adopted competency based human capital systems, including the use of competency models as the basis for performance management.17 Although extremely popular, there has been debate about what “competencies” are and how to most effectively measure them. Sometimes, competencies refl ect knowledge areas and skills, sometimes they refl ect performance factors, sometimes they refl ect values, and sometimes they refl ect personality traits. Here, competencies are defi ned as the knowledge, skills, abilities, and other personal characteristics that are most instrumental for achieving important job outcomes that contribute to organizational success (Figure 2.2). A best practice in the use of competencies for performance management purposes has been defi ning them in terms of behavioral performance standards that describe different levels of effectiveness in each competency area.

Results became the bottom line  .  .  . Critical thinking

• 

The most recent trend in performance management is incorporation of a “results focus.” The emergence of this trend is largely due to organizations becoming increasingly focused on achieving results, not just driving effective behaviors. The idea is that each employee needs to be accountable for producing results that contribute to the organization achieving its goals. It has thus become best practice today to assess both the results employees achieve as well as how they went about achieving these  or in other words, their job behavior.