2002 Graphic Scale

25
Reengineering the Graphic Rating Scale 1 Reengineering the Graphic Scale REENGINEERING THE GRAPHIC RATING SCALE Sven Aelterman Troy State University / Hogeschool Gent TSU Box # 821292, Troy, AL, 36082 1-334-670-4487 [email protected] Dr. Hank Findley Troy State University 200 Bibb Graves, Troy, AL, 36082 1-334-670-3200 [email protected]

description

Appraisal

Transcript of 2002 Graphic Scale

Page 1: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 1

Reengineering the Graphic Scale

REENGINEERING THE GRAPHIC RATING SCALE

Sven Aelterman

Troy State University / Hogeschool Gent

TSU Box # 821292, Troy, AL, 36082

1-334-670-4487

[email protected]

Dr. Hank Findley

Troy State University

200 Bibb Graves, Troy, AL, 36082

1-334-670-3200

[email protected]

Page 2: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 2

ABSTRACT

It appears that few Human Resource Management topics are as controversial as performance

management and/or appraisal. This article is not meant to discuss the merits of performance

appraisal, rather it will attempt to propose two changes to one of the most – if not the single

most – used performance appraisal method: the graphic rating scale. The first proposed changed

is a completely new one and attempts to provide a method for identifying high performers more

clearly and at the same time identifying poor performers more easily. The second part of the

paper advocates the use of automated systems to store and analyze performance appraisals.

Page 3: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 3

REENGINEERING THE GRAPHIC RATING SCALE

INTRODUCTION

Performance Appraisals

Performance appraisals are among the most controversial topics in Human Resource

Management. For every article advocating the use of performance appraisal and/or management,

there seems to be one other article that argues against using performance appraisals. Especially

TQM advocates argue against performance appraisals. Their main concern is that performance

appraisals are contrary to the total quality principle, where striving for quality is an ongoing

effort (Allender, 1995).

There is a need for depicting accurate performance. Many human resource decisions are

made based on performance, such as pay raises, promotions and demotions, and even

termination. Evaluating recruitment results is also often tied with performance. Identifying areas

where development for staff is needed can also be based, at least partially, on performance data.

Research even suggests that formal appraisals help in creating a competitive edge (Longenecker

& Fink, 1999). So at least for now, performance appraisals remain a necessity, for several

reasons, including legal and motivational (Findley, Amsler & Ingram, 1999). Therefore, it is

important to continue to improve the way performance appraisals are performed.

Graphic Rating Scales

Graphic rating scales have come under scrutiny because of several issues related to their

use in performance appraisal. The most cited problems with rating scales are halo and

leniency/strictness (Chiu & Alliger, 1990). The halo effect is a problem that occurs when a rating

Page 4: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 4

on one job dimension affects the rating on another one (Solomonson & Lance, 1997). Leniency

or strictness occurs when a rater tends to rate all ratees respectively low or high.

However, the graphic rating scales also have important advantages. Graphic rating scales

are easy to develop, administer, and interpret. Furthermore, graphic rating scales yield results

that allow comparison across ratee groups (Chiu & Alliger, 1990). Besides, they are generally

recognized as being the most widely used method in performance appraisals.

New research also found that graphic rating scales are as good as or better than two other

methods that are generally deemed better. Tziner, Joanis and Murphy concluded that especially

in terms of ratee attitudes and goal characteristics, the graphic rating scale outperforms

behaviorally anchored rating scales (BARS) and behavior observation scales (BOS) (Tziner,

Joanis & Murphy, 2000).

For all of the reasons above, reengineering should aim to combine the best factors of the

existing methods while trying to eliminate the weak elements of each method. Because of their

widespread use, it is worth spending time trying to improve the graphic rating scales. This text

will try to do exactly that, by suggesting two ways to eliminate problems with the graphic rating

scale. Each method will be explained in detail, after which possible cost issues and fairness

issues will be discussed.

Typical Problems with the Current Situation

Throughout this text, the same cases will be used to illustrate the proposed

improvements. These cases will be introduced here. The first case looks at the performance

appraisal problems in a sales environment. The second case deals with the appraisal of hourly

workers at a major US manufacturer of toys.

Page 5: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 5

Case 1

One of the salespeople’s interest in the firm’s products has declined to the point where it

starts interfering with his ability to sell the product effectively. His manager has noticed this and

has rated him accordingly on the appropriate job dimension, Product Knowledge. The full

performance appraisal can be found in the appendix (Figure 4: Performance appraisal of Jason

Borman.). Mathematically, he ends up being rated average (using the values from the graph from

Figure 1: Graph showing the current relation between performance dimensions and their

mathematical values.). The fact that he has been rated Unacceptable on Product Knowledge does

not reflect in the overall performance.

Case 2

The supervisor of a work team in a large manufacturing plant is faced with the feared,

traditional year-end performance appraisal. The company uses the graphic rating scale to

perform the performance appraisal for its hourly workers. This particular supervisor supervises a

team of 14 in the main manufacturing plant. As for many supervisors, with the year end comes

what he sees as a major annoyance: performance appraisal.

In this particular plant, the year also means a lot of extra work because of the holiday

season. Two weeks ago, one employee made a mistake that caused his team to have to put in

some serious overtime to catch up on lost production time. That employee put in the most

overtime because she felt guilty. The remainder of the year has been uneventful. The supervisor

failed to note however that she did go through great lengths to increase the team’s overall

productivity, and succeeded.

Her performance form can be found in the appendix (Figure 6: Performance appraisal of

Erika De Wit.).

Page 6: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 6

This case presents two of the problems described above: objectiveness (or rather lack

thereof) and recency errors. As can be seen from the performance appraisal form, John rated

Erika down on Quality of Work (although there have been no other events) and Teamwork. A

fairer review would probably have given Erika an Above Average rating for Teamwork, and

depending on the outcome of an investigation regarding the mistake, Average for Quality of

Work. This would have caused Erika to be rated Above Average on the overall score.

This performance appraisal shows neither objectiveness or diligence. The supervisor

clearly let the recent mistake weigh too heavily on the appraisal (halo and recency effect), while

not taking into account the overall performance the employee exhibited.

FIRST PROPOSAL TO IMPROVE GRAPHIC RATING SCALES

The first proposed improvement will target the identification of poor and high

performers. Identifying poor and high performers is important in maintaining a motivated

workforce.

Problem Background

The traditional relationship between performance dimensions and their mathematical

values is an interval scale, as is shown in the graph below:

Page 7: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 7

Unacc

epta

ble

Excell

ent

Above

Avera

ge

Avera

ge

Below

Avera

ge

Outst

andi

ng

1

6

5

4

3

2

7

Figure 1: Graph showing the current relation between performance dimensions and their mathematical values.

The interval between each performance dimension is equal, in this case one. So someone

with an unacceptable performance (either overall or on one single job dimension) is deemed to

be equally far away from being average than someone with an excellent performance; just on the

other side of the scale. This is not the way it is in reality. However, no one really knows what the

difference between Unacceptable and Below Average on one hand and Below Average and

Average on the other really is.

It is important to note that an unacceptable rating on one or more job dimensions

constitutes a cost to the company. That cost is hard to determine, but can be very high. Think

Page 8: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 8

about an employee getting an Unacceptable rating on Safety in an oil refinery. Therefore, it is

very important to identify poor performers.

Proposed Solution

The solution to this issue would be to use some type of a logarithmic scale, as presented

in the graph below:

Unacc

epta

ble

Excell

ent

Above

Avera

ge

Avera

ge

Below

Avera

ge

Outst

andi

ng

-1

4

3

2

1

0

5

-2

Figure 2: First proposed graph with an alternative relationship between performance dimensions and their mathematical value.

The graph shows clearly that the difference between Unacceptable and Below Average

now is considerably greater than the difference between Below Average and Average. So, if a

person was rated Unacceptable on a certain job dimension, this rating of Unacceptable would

weigh far more heavily on the overall performance than in the existing use of graphic rating

Page 9: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 9

scales. At the same time, for employees being rated average, this proposal has no influence on

their overall rating, which increases the perceived fairness of the method, which is as indicated

above important.

Does this graph come close to depicting the real difference between Unacceptable and

Below Average on one hand and Below Average and Average on the other? As stated before,

that difference has not been determined. The recommendation here is to use a common policy

throughout the company, again to increase the perceived fairness.

Refining the Solution

To continue on the issue of perceived fairness: high performers may find the use of the

graph above unfair because their high performance is now less recognized than the performance

of above average employees. The solution to this issue would be to change the direction of the

curve at a certain point, to allow a greater distance between Outstanding performance and

Excellent performance. This is shown in the graph below:

Page 10: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 10

Unacc

epta

ble

Excell

ent

Above

Avera

ge

Avera

ge

Below

Avera

ge

Outst

andi

ng

-1

4

3

2

1

0

5

-2

Figure 3: Graph displaying the relationship between performance dimensions and their mathematical value.

The point of inflection in the curve can occur sooner, so that the difference between

Above Average and Excellent would be greater. Furthermore, the curvature of the graph is also

variable. The only requirement is that within a company or department, the same graph is used to

project performance ratings. If not, employees may perceive it to be easier for some job

categories to get an Outstanding overall rating. This will affect the perception of fairness and

therefore citizenship behavior (Chan Kim & Maugorgne, 1997).

The tables below show an example performance rating using graphic rating scales before

and after the use of the first proposed improvement. The table uses traditional performance

dimensions, such as Quality of Work and Quantity of Work. This example is not meant to

provide a sample performance appraisal form, rather is it used to show how this proposal works.

Page 11: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 11

Below the performance (i.e. Unacceptable, etc) is the numerical value that is assigned to it,

according to the graphs above.

Job Dimension

Unacceptable (1)

Below Average

(2)

Average (3)

Above Average

(4)

Excellent (5)

Outstanding (6)

Quality of Work 4

Quantity of Work 3

Job Knowledge 1

Attendance 4 Reliability 3 Safety 4 Overall Performance 3.167

Table 1: Performance appraisal form using the traditionally assigned values.

Job Dimension

Unacceptable (-1.5)

Below Average

(1.9)

Average (3)

Above Average

(3.75)

Excellent (4)

Outstanding (5)

Quality of Work 3.75

Quantity of Work 3

Job Knowledge -1.5

Attendance 3.75 Reliability 3 Safety 3.75 Overall Performance 2.625

Table 2: Performance appraisal form using the weighted performance dimensions.

As the example above shows, by weighting the performance dimensions, different results

can be obtained.

This refinement can also have a drawback. Assigning a higher value to Outstanding

might undo the effect of assigning the lower value to Unacceptable in those cases where an

employee has been rated Outstanding and Unacceptable. The table below illustrates this.

Page 12: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 12

Job Dimension

Unacceptable (-1.5)

Below Average

(1.9)

Average (3)

Above Average

(3.75)

Excellent (4)

Outstanding (5)

Quality of Work 5

Quantity of Work 3

Job Knowledge -1.5

Attendance 5 Reliability 3.75 Safety 4 Overall Performance 3.208

Table 3: If the employee has both Unacceptable and Outstanding job dimensions.

As can be seen from the table above, the fact that this particular employee scores low on

job knowledge, but average to high on the other job dimensions, causes the advantage of the

solution to be lost. When returning to the values of the original graph (Figure 2: First proposed

graph with an alternative relationship between performance dimensions and their mathematical

value.), we get this result:

Job Dimension

Unacceptable (-1.5)

Below Average

(1.9)

Average (2.9)

Above Average

(3.5)

Excellent (3.9)

Outstanding (4)

Quality of Work 4

Quantity of Work 2.9

Job Knowledge -1.5

Attendance 4 Reliability 3.5 Safety 3.9 Overall Performance 2.800

Table 4: Using the values from the original improvement proposition.

In this example, the assigned ratings remained unchanged, but the mathematical values

assigned to each performance dimension have been taken from Figure 2. Using these values, the

Page 13: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 13

overall rating falls back into the Below Average category. We can conclude that organizations

must make a choice between using a curve with a point of inflection and using a regular

exponential-like curve. When using a graph with a point of inflection, it may be advisable to

further increase the difference between Unacceptable and Below Average.

Cost

The added cost of adopting this method is minimal. Initially, time must be spent finding

fair weights to apply to the performance dimensions. However, once these numbers have been

found and introduced on the performance appraisal forms or, better yet, introduced in the

performance appraisal software, the use of this proposal becomes completely transparent. It may

be necessary to spend time explaining the new method to the raters and ratees, in order to make

sure that the new scales are perceived as being fair.

Fairness

This method is probably not entirely without critique. For example, it is necessary to

inform employees of the use of this method. If not, employees may consider it unfair that

unacceptable performance is treated differently than performance rated below average. It is to be

expected that especially poor performers would perceive this method as less fair, while high

performers are expected to think positively of it.

It is also important to mention that this method can be combined with a previously found

method: weighting the job dimensions. If a company finds that safety is three times as important

as quantity of work, the performance rating for safety is multiplied by three before the totals are

made. This proposal complements and possibly enhances the weighting of job dimensions. That

Page 14: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 14

is because the use of a negative number for an unacceptable performance that is also weighted

will increase the influence of that unacceptable rating (through the negative number) even more.

Example Solution Using the First Proposal

To provide another example of the use of this proposal, Case 1 (see above) will be

reviewed using the improved graphic rating scale. The employee was rated average using the

traditional method of assigning linearly increasing numbers to the performance dimensions. In

the appendix, the performance appraisal form using weighted values for the performance

dimensions can be found (Figure 5: Performance appraisal of Jason Borman using the proposed

improvement.). Note that in this instance, different values have been assigned to the performance

dimensions that are shown in the previous examples.

SECOND PROPOSAL TO IMPROVE GRAPHIC RATING SCALES

The second improvement will attempt to effectively combat recency errors and leniency

or strictness. Recency errors occur when the raters only include recent events and performance

while doing the performance appraisal. Research has shown that with paper-based systems, only

performance of the last 6 to 12 weeks is taken into account (Dutton, 2001). When performing

annual or bi-annual performance reviews, raters should strive to include the performance of the

affected period, i.e. either 12 or 6 months, and more importantly, to weight each performance

incident equally, regardless of when it occurred. Strictness or leniency occurs when raters have a

tendency to rate either low or high and occurs more with graphic rating scales than with a

method that uses ranking.

Page 15: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 15

Problem Background

Often, performance appraisal is done at the end of the year. Many managers consider the

performance appraisal to be a necessary evil and want to get it over with as soon as possible. On

top of that, the end of the year may be a busy period for the firm, causing the issue of

performance appraisal and review to be taken lightly. Research shows that because of these

factors, objectivity suffers and raters tend to use the most recent performance during the

performance appraisal. Usually, paper-based performance appraisals don’t take raters back more

than 12 weeks (Dutton, 2001). This last factor is known as recency error. Other negative effects

affecting performance appraisal are halo and rating inflation, as discussed above.

"The reason managers dread doing performance appraisals and employees dread getting

them is that they see them as an event and not a process," says Dick Grote, author of "The

Complete Guide to Performance Appraisals" (Grote, 1996).

Proposed Solution

To avoid recency errors, companies must consider automating their performance

management system. While many businesses have automated many of their administrative

processes, the automation of performance management in general and performance appraisal in

particular has only recently attracted attention. Through the use of new software, it becomes

easier to implement an automated system. Some of the applications that are available now are

also accessible to small and medium-sized businesses.

While a discussion of the different software packages that are available for performance

management is beyond the scope of this paper, it is worth mentioning the advantages that come

with such software. Afterwards, some of the changes that would need to be made to the

performance management practices in a company will be highlighted.

Page 16: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 16

Advantages of Automating Performance Appraisal Management

Lately, 360-degree performance appraisals have gained acceptance. While adding

reviewers to the appraisal process may be beneficial to the outcome of the appraisal, it does

create a significant overhead for the manager involved. By automating the actual recording of the

performance appraisal, a lot of time can be won. Instead of having each reviewer fill out a paper

form, then submitting that form to the responsible manager who has to compile the information.

Automating this system could mean that the reviewers log on to an internal web site and fill out

the form electronically1. The responsible manager can then check on the progress of the review

and simply request the final, compiled data from the computer system.

Of equal usefulness is the ability to keep a history of performance appraisals without the

need for an archiving system and storage space to store the paper documents. Tied to keeping an

extended history without added cost or effort is the ability to search that history quickly. Keeping

a history of performance appraisals is useful because that data can be analyzed to predict future

performance of new hires, an important aspect of the selection process.

A third advantage of using automated systems for performance appraisals is because

information is a very important asset that must be safeguarded. Electronic data can be backed up

easier that paper records. And while electronic data is usually viewed as less secure than paper-

based records, implementing access controls to electronic records is actually more feasible than

doing the same thing with paper records.

While keeping an electronic history allows you to analyze the data easier than do paper

files, they also allow different views on the data to be created with ease. The performance

appraisal history can be viewed by department, by age, by education level, etc. With the

1 While at the same time receiving assistance from the software on how to fill out the form appropriately!

Page 17: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 17

technologies that are available today, non-technical managers can create those reports

themselves, without being dependent on a IS department to create the necessary queries.

Since compiling the performance appraisal data is so easy and quick, it becomes possible

to schedule performance reviews more than once a year. This has been advocated by many

performance appraisal specialists as a way to improve the entire performance management

process.

Finally, using software to perform performance appraisals ensures that company-wide the

same policies and procedures are used. Having a company-wide policy that is enforced helps

ensuring the validity and the fairness of the whole process.

Disadvantages As Well

While automating the performance appraisal process clearly does have a lot of tangible

advantages, care should be given to possible pitfalls, the first of which is data security and

privacy. As has been mentioned above, electronic data is often regarded as being less secure. But

by taking the necessary precautions, it is possible to create an electronic system that is both

secure and efficient. This is shown in practice by Red Hat Corp., a distributor of Linux operating

system software. Red Hat has installed an internet-enabled performance management software

package, and has so far successfully secured the data (Dutton, 2001). More than that, while it

may be hard to limit access to paper-based performance appraisals, especially inside the HR

department, the software allows for strict rules determining the access permissions of users.

Next to security issues, there is the issue of cost. Acquiring or implementing software that

is capable of the functions listed above is costly. Then there is the added cost of training, support,

and maintenance. However, in the total cost of performance appraisals, the cost of the

technology is minimal. Research shows that the annual cost of performance appraisal per

Page 18: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 18

employee can be as high as $3,200. The majority of the costs are in preparing the appraisal,

conducting reviews, designing the appraisal system, etc. (Dutton, 2001). An automated system

will actually help reduce the time spent on the most costly parts of performance management,

thereby directly giving a return on investment.

Companies must also be alert to over-automating the process. In the words of Robert

Bacal (1999): “Performance appraisal is an interpersonal communication process.” While

software may assist reviewers in gathering and analyzing the data, the performance review

sessions do remain a human affair.

Effective Use of Software

To use software effectively for performance appraisals, it should offer an employee log

functionality. The employee log could be redesigned to include an immediate evaluation of

critical incidents as soon as they have been fully investigated. This solution comes forth from the

belief that when evaluating performance based on critical incidents, the evaluation tends to be

different after time has passed. In general, the evaluation tends to reduce extreme performance

(good or bad) to average performance (Mitchell & James, 2001).

When entering new critical incidents in the employee log, the rater should be given

maximum support from the information system. For example, companies that currently use the

graphic rating scale and coupled it with descriptions of what performance level relates to what

behavior2 may choose to have the rater select from a list of behaviors instead of a list of

numbers. In that case, the rater is relieved from the duty of having to rate someone Average,

Below Average, or Excellent. Instead, he selects the behavior that was exhibited by the employee

2 Instead of Graphic Rating Scales (GRS), these scales are Behaviorally Anchored Rating Scales (BARS).

Page 19: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 19

from a list. The software assigns the actual rating in the background. Then, when the time for the

performance appraisal comes, it suffices to retrieve the list of critical incidents with their

associated rating.

Performance appraisals today are based on more than critical incidents. For example, the

goals that are set for employees and work units should also be entered in the system. When a

goal is met or not met, this should be appropriately entered into the system. The system can then

rate the performance of the individual or team. Rating the achievement of goals is a separate

appraisal method, Management by Objectives (MBO). However, it is possible to combine

graphic rating scales with MBO. The advantage of using software when combining different

appraisal methods is that the different methods can be transparent to the user.

This continuous use of software is consistent with the belief that performance appraisals

should be an ongoing assessment instead of a once-in-a-year event (Fandray, 2001). While the

performance review with the employee may still take place only once a year, it facilitates and

encourages a continuous evaluation of employee performance. This proposal also achieves three

of the six improvements suggested by Weizmann, namely ‘Link the performance-management

calendar to the organization’s business calendar’, ‘Conduct a mid-year review’, and ‘Don’t get

bogged down in paperwork.’ (Weizmann, 2001)

Fairness

This proposal has to potential to increase employees’ perception of fairness with

performance appraisals. Since an unbiased computer system actually does the rating, issues with

performance appraisals such as discrimination based on sex or race can be avoided. Since

software-based performance-appraisals tend to focus on results and actions rather than

personality traits, employees are more likely to view them as fair (Dutton, 2001).

Page 20: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 20

To use this solution effectively, it is necessary for supervisors to include every critical

incident promptly. This is the only way to ensure that all behavior is logged and will be taken

into account during the performance appraisal interview. Companies can enforce the use of this

type of employee log by training managers to recognize and rate behavior and by regularly

checking if the employee logs are filled out conscientiously.

Example Solution Using the Second Proposal

By revisiting the case 1, we can show that by diligently keeping records of performance

on each employee, this employee would not have been rated Below Average. Her efforts would

have been recognized properly. Although she did cause overtime, something that must

undoubtedly be included in the performance appraisal, it should not reflect on other job

dimensions and it should not be the only event included in the appraisal.

As described above, using performance appraisal software has the potential of reducing

recency and objectiveness errors when using graphic rating scales. As always, introducing a new

method should be followed by training of both supervisors and employees.

RECOMMENDATION

While other solutions may exist to improve the efficiency of the graphic rating scale, the

solutions that have been presented in the paper may very well tackle the problems more

effectively than others, since they attempt to reduce the influence of human behavior more than

solutions.

The reader should also be aware of the fact that no research has been conducted as to the

feasibility of the proposed improvements. The ideas presented in this paper are purely

theoretical. However, the example cases present frequent occurrences of problems with

performance appraisals. In those cases that are presented in this paper using the proposed

Page 21: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 21

improvements definitely helps. These errors may be reduced greatly by improving the training

for the supervisors and managers performing the performance appraisal. However, the need for

training supervisors has been known for a long time (Buzzotta, 1988; Eyres, 1989). One would

expect that companies taking performance appraisal serious have already implemented training

and awareness programs for their supervisors and managers. And yet, performance appraisals

continue to be a source of controversy.

One final remark: as with any proposed solution to a problem, when first implementing it,

care should be given to combine the guidelines described in the paper with common sense. This

is especially important in this case because as stated before, no research as to the actual

usefulness of the proposals.

Page 22: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 22

APPENDIX

Performance Appraisal Forms for Case 1

Figure 4: Performance appraisal of Jason Borman.

General Manufacturers

Performance Appraisal Form for: Jason Borman Job Title: Sales Representative

Date: 12/10/2001

Prepared by: Belinda Gomez Job Title: Sales Manager

Category Unacceptable Below Average

Average Above Average

Excellent Outstanding

Quality of Work 2 Quantity of Work 4 Teamwork 4 Attendance 4 Product Knowledge 1 Overall 3 Comments:

Employee has seen and read this performance appraisal form on 12/11/2001:

Jason Borman (signature of employee)

Page 23: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 23

Figure 5: Performance appraisal of Jason Borman using the proposed improvement.

General Manufacturers

Performance Appraisal Form for: Jason Borman Job Title: Sales Representative

Date: 12/10/2001

Prepared by: Belinda Gomez Job Title: Sales Manager

Category Unacceptable (-1.5)

Below Average

(1.5)

Average (3)

Above Average

(4.5)

Excellent (6)

Outstanding (8)

Quality of Work 1.5 Quantity of Work 4.5 Teamwork 4.5 Attendance 4.5 Product Knowledge -1.5 Overall 2.7 Comments:

Employee has seen and read this performance appraisal form on 12/11/2001:

Jason Borman (signature of employee)

Page 24: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 24

Performance Appraisal Forms for Case 2

Figure 6: Performance appraisal of Erika De Wit.

REFERENCES

Allender, H. D. (1995). Reengineering performance appraisals the TQM way. Industrial Management , Nov/Dec 1995, 10.

Bacal, R. (1999). Seven Stupid Things Human Resource Departments Do To Screw Up Performance Appraisals. Retrieved on January 13, 2002 from http://www.work911.com/performance/particles/stuphr.htm

Buzzotta, V. R. (1988). Improve your performance appraisal. Management Review, Aug. 1988, Vol 77, 40-44.

T’oys Corp.

Performance Appraisal Form for: Erika De Wit Job Title: Production Line

Date: 12/15/2001

Prepared by: John Smith Job Title: Manufacturing Supervisor

Category Unacceptable Below Average

Average Above Average

Excellent Outstanding

Quality of Work 1 Quantity of Work 3 Teamwork 2 Attendance 4 Safety 3 Overall 2.6

Comments: Caused team to put in overtime

Employee has seen and read this performance appraisal form on 12/17/2001:

Erika De Wit (signature of employee)

Page 25: 2002 Graphic Scale

Reengineering the Graphic Rating Scale 25

Chan Kim, W. & Maugorgne, R. (1997). Fair process: Managing in the knowledge economy. Harvard Business Review, July/August 1997, 65-76.

Chiu, C. & Alliger, G.M. (1990). A proposed method to combine ranking and graphic rating in performance appraisal: The quantitative ranking scale. Educational and Psychological Measurement , Fall 1990, Vol. 50, Issue 3, 493-505.

Dutton, G. (2001). Making reviews more efficient and fair. Workforce, April 2001, Vol. 80, Issue 4, 76-82.

Eyres, P. S. (1989). Legally defensible performance appraisal systems. Personnel Journal, Juli 1989, 58-62.

Fandray, D. (2001). The new thinking in performance appraisals. Workforce, May 2001, Vol. 80, Issue 5, 36-40.

Findley, H. M., Amsler, G. M. & Ingram, E. (1999). Reengineering the performance appraisal. National Productivity Review, Winter 2000, 39-42.

Longenecker, C. O. & Fink, L. S. (1999). Creating effective performance appraisals. Industrial Management , Sep/Oct 1999, 18-23.

Mitchell, T. R. & James, L. R. (2001). Building better theory: Time and the specification of when things happen, The Academy of Management Review, October 2001, Vol. 26, Issue 4, 530-547.

Solomonson, A. & Lance, C. (1997). Examination of the relationship between true halo and halo effort in performance ratings. Journal of Applied Psychology, Vol. 82, Issue 5, 665-674.

Tziner, A., Joanis, C. & Murphy, K. R. (2000). A comparison of three methods of performance appraisal with regard to goal properties, goal perception and ratee satisfaction. Group and Organization Management , Vol. 25 No. 2, June 2000, 175-190.

Weizmann, J. (2001). Quote found in Fandray, D. (2001).

The companies, people, and events presented in this paper are fictitious. Any resemblance to actual companies, people, or events is entirely coincidental.