Performance Management and Performance Appraisal

Performance Management

The six purpose of performance management systems:

• strategic
• administrative
• communication
• developmental
• organizational maintenance (are some orgs underperforming)
• documentation

Performance management is the process that incorporates goal setting, appraisal, and feedback to make performance-based administrative decisions and help employees improve. It is a combination of three aspects:

• performance goal setting: process of setting SMART goals to direct performance
• performance development: assessment of performance with the goal of providing feedback to facilitate improved performance
• performance appraisal: process of identifying criteria or standards for determining how well employees are performing their jobs

Differences between performance management vs performance appraisal:

• continuing, ongoing process vs one-time event
• importance of alignment among goals, evaluation, and feedback.

An integrated performance management process

1. define performance (set performance goals, establish and communicate performance expectation)
2. evaluate performance (monitor performance, communicate performance strength/weaknesses)
3. review performance (deliver feedback to employees, coach employees)
4. provide performance consequences (provide sufficient and valued consequences)

Setting performance goals, what is a SMART goal?

• specific: what is the goal
• measurable: how will it be measured
• attainable: is this goal attainable
• relevant: is this goal relevant
• time bound: what is your timeline

Performance Appraisal

Evaluating performance. How you appraise a person is related to how you perceive that person. London (2001) outlined the theory as follows: information, motivation, and relationships help us process large amount of information about other people:

1. inputs: target, perceiver, context
2. processes: information processing (how much did we absorb), motivational (do we have other reasons that could affect your evaluation)/affective (how much do you like them), interpersonal/social
3. outputs: consequences for perception of target, consequences for perceiver/decision-maker

So how might the fundamental attribution error contribute to perception of others?

• When something bad happens to someone else, we tend to blame it on the characteristic of the person. However if it happens to ourselves, we tend to blame the environment.

Types of appraisal data:

• object productivity (how much product or service did someone make, how much money did you bring in)
• personnel data (how many days is the employee absent)
• judgement data (how do superiors, peers, view the employee's work in the workplace)

Q: What is one possible objective measure and one possible subjective measure for evaluating a professor's job performance?

Objective production data considerations:

• how much an employee produces, or salesman sells, or faculty publishes, or police write tickets
• production data may be meaningfully compared across employees
• problems with production data: location bias, economy dependencies, funding sources, dependence on equipment (older, faulty), does not take quality into account.

Objective personnel data:

• absenteeism and tardiness (unexcused absences = bad performance)
• safety violations and accidents (frequency and severity are both variables, considers anything that results in injury or property damages)
• problems with using personnel data: may not know the reason for absence or tardiness, safety violations may only apply to a limited number of jobs.

Subjective/judgemental data:

• judgmental evaluations are often used for performance appraisal because:
  • relevant objective measures are difficult to find
  • appraisals can apply to almost all jobs
  • numerous dimensions of interests can be evaluated
  • multiple people can participate in assessment
• Problems with using subjective data
  • rater errors are likely to occur
  • different managers may use different standards
  • rater motivation may influence rating
  • ratees may perceive the data as unfair or biased.

Performance feedback:

1. seeking feedback. There are three motives for seeking feedback:
   1. instrumental motive (self-improvement)
   2. ego-based motive (defend or enhance self-views)
   3. image-based motive (to look better)
2. giving feedback is essential. 1/3 of supervisor feedback results in decreased performance
3. reactions to feedback. It's important for individuals to:
   1. believe they have control over performance
   2. believe system is accurate
   3. those most oblivious to performance deficits least likely to accept feedback.

The 7 characteristics contributing to employee acceptance, the reaction to performance feedback and consequences:

1. solicit and use employee input
2. two-way communication during a feedback interview
3. opportunity for employee to rebut rating(s)
4. rater's degree of familiarity with ratee's work
5. consistent standards applied for feedback and uniform consequences
6. ratings and consequences are based on actual performance
7. recommendations for rewards that are based on ratings.

The supervisor (person given feedback) must be perceived as knowledgeable, objective, and trustworthy.

Rating

The Subjective/judgemental types of data measures are:

• graphic rating scales or "conventional ratings"
• employee comparisons (rankings)
• critical incidents
• behaviorally-anchored checklists

Rating Scales

Graphic rating (conventional rating) scales:

1. targets are rated on 5-20 traits/factors
2. each target performance is judged on 5-7 point scale
3. may be numeric or anchored with descriptors
4. susceptible to rating error

Employee comparison methods:

• rank order (high to low)
• paired comparison (each employee compared with every other employee, best used for small samples)
• forced distribution (5-7 categories where employees must be distributed across all categories, best used for large sample, top-grading/rank and yank)

Behavioral checklist/scales:

• Critical incident method records samples of employee's good performance and bad performance on previously set performance criteria. It requires careful recording by observers of specific incidents, behaviors, and outcomes. Also referred to as "work sampling" and are not usually quantified.
• Behavioral observation scales (BOS) identifies critical incidents (samples of key tasks/behaviors) for a particular job. Its evaluated based on how frequently they exhibit the required behavior, and then totaled. Example job dimension: job knowledge.
• Behaviorally-anchored rating scale (BARS) clear standards upon which an employee is appraised. They are very accurate in appraisal methods and therefore have higher reliability. It is time consuming to develop so its not widely used in business.

Rating Errors and Biases

Rater errors and biases include the following:

• leniency errors: positive vs negative leniency (severity)
• central-tendency errors: avoidance of extreme ratings
• contrast error: comparisons against others vs against set standards
  • Likely to happen when a rater rate numerous people within a short time period. Evaluating a weak employee will make other employees be rated higher because they are being compared to that weak employee.
• halo and horn error: rater's general feelings about the employee.
  • Halo errors are when the rater generally has a favorable attitude. Horn is unfavorable attitude about an employee that permeates evaluation. Typically, the rater has strong feelings about one aspect of the employee's performance, and that feeling is generalized to the other performance factors.
  • Two halos: 1) failure to differentiate an employee's performance across different dimensions 2) giving uniformly consistent ratings to an employee who earns in
  • valid and invalid halo (considered the most serious and pervasive error or bias).

Unlike errors and unconscious bias, rater motivation is referred to as "conscious rating distortion", which is deliberate. They purposely give ratees scores that are higher or lower than they actually deserve.

• Rater inflation: intend to be accurate (but not rewarded by oganization), intend to avoid conflict, benevolence for employy, impression management so that the rater appears in a better light
• Rater deflation: shock employee and alert them to their problems, intend to teach the employee a lesson or encourage them to look for other work, provide compelling documentation to justify future action (e.g., termination).

Rater Training

Types of rater training:

• Rater error training: enhance awareness of possible errors, taught to make fewer errors through example, does not necessarily result in increase accuracy
• Frame of reference training: particularly promising, calibrates rater, shows them what to look for, grading of performance is uniform at least within a company, generally results in increased accuracy.

Authenticity of appraisals: Murphy and Cleveland (1985) found that accurate ratings are more likely to occur when

• good/poor performance is clearly defined,
• distinguishing workers in terms of performance level is accepted
• there is high trust in the system
• low ratings don't automatically result in loss of reward
• rewards are linked to accuracy in appraisals.

Authors don't know of any organizations in which all conditions are met.

Self-assessments.

• individuals have poor understanding of their own strengths and weaknesses.
• positive leniency is a problem; will often rate performance higher than others
• fundamental attribution error effects
• Anderson, Warner and Spencer (1984) study found that when they asked applicants to rate their own abilities in real and fake (but believable) clerical tasks, the applicants rated themselves high on real tasks, their ability was not verified. But they also rated themselves on the fake tasks that they did not perform.

Peer Assessments

Peer-assessments have three types:

• peer nomination: each person nominates a specific number of group members as being highest on a particular dimension of performance
• peer ratings: each group member rates the other on a set of performance dimensions using one of the several kinds of rating scales
• peer rankings: each member ranks all others from best to worst on one or more performance dimensions.

They found that their response was biased by friendships, peers do not like to assess each other, peer assessments are more accurate than self-assessment, and there is a consensus that it may be better for developmental (vs administrative) purposes.

360 Feedback

360-degree feedback (multisource feedback) is the practice of using multiple raters from different perspectives in the assessment of individuals.

The 360-degree feedback relationships go from

• x-axis: peers self peers
• y-axis subordinates self supervisors

The difference in rater perspective is regarded as valuable.

• differences provide opportunities for professional development and personal learning to understand why individuals are perceived as they are by others
• organizational should clarify whether 360-degree feedback will be used for development purposes or administrative purposes and stick to it.

Sources of measurement data- strengths and weaknesses:

• student:
  • strengths: provide insight on individual perspective, allows for self-reflection, can capture aspects that others cannot observe
  • weakness: subject to bias, positive leniency, limited perspective
• peer:
  • strengths: diverse perspective, reflects performance on team dynamics, insight to interpersonal skills and collaboration
  • weakness: subject to personal bias, potential conflict of interest, not a full comprehensive understanding of the individual's work, influenced by personal relationships.
• supervisor
  • strengths: insight to overall performance, more weight, has a broader perspective
  • weakness: subject to personal bias, not aware of their day-to-day, power dynamic lead to lack of honest and constructive feedback
• self-evaluation
  • strength: encourage self-awareness, tool for personal development
  • weakness: prone to overestimation/underestimation, lack of objectivity, struggle with blind spots.