Job Evaluations

Approaches to Classifying Employees in Government

Whole job ranking is the simplest form of job content evaluation. It involves a whole job, job to job comparison of each job resulting in an array of jobs from the highest to the lowest ranked job. The result of the ranking is only an indication of order; it does not tell you anything about the relative degree of distance between jobs.

The whole job ranking method is by far the simplest of all job evaluation methods and works well in small organizations which have a singular purpose. Whole job ranking is inexpensive, quickly learned and implemented. This method will not work in most governments because it leaves no documentation, the criteria on which jobs are ranked is not stable, is too superficial and will not control grade or classification growth. Furthermore, it could not be supported in a legal proceeding.

Classification method is a job evaluation method that has traditionally been associated with government pay systems. Classification is a method which compares jobs on a whole job basis. Pre-defined class specifications (or job descriptions) are established for a series of job classes, and a job (held by an employee) is placed in whichever classification best describes it.

The classification method is simple to administer, inexpensive to implement and maintain, can be implemented quickly, little training is required, and thousands of governmental units throughout the country rely on this form of job evaluation. Its disadvantages are that jobs are forced to fit into pre-defined classifications since most jobs do not fit neatly. The result is the creation of new classes when they may not be necessary, job descriptions can be manipulated to fit into a pre-defined classification thus causing grade creep when non is justified, and it is not practical for organizations that have a diverse set of job families or occupational groups because there is no good way to compare the value of one classification level against another for determination of pay purposes, unless another job evaluation method is used.

Slotting is a method for placing jobs into a category, classification or any other ranking scheme that has been developed by another job evaluation method. Slotting involves the comparison of a job to some already existing standard or benchmark job hierarchy.

The slotting technique is simple to administer, inexpensive to use, quickly implemented and little training is required. Its disadvantages are that it cannot be used as a stand-alone system of job evaluation, there are no specific criteria on which to determine the relative value of one job against another and there usually is insufficient documentation on which to justify a decision. Furthermore, since it relies on an existing hierarchy to be useful, it does not meet the basic needs of the most governments to develop a new system.

Most organizations are now using quantitative job evaluation methods. There are basically three types:

• job component
• factor comparison
• point factor

The job component method is a statistically developed job evaluation method which relies on scored questionnaires, multiple regression analysis and factor analysis. Employees complete a standardized questionnaire which either identifies the behavioral components of the job or the task based components of the job. The incumbent indicates for each task or behavior, the importance of the category and the frequency of performance. The questionnaire is input into a computerized mathematical model and a job score is determined.

These methods have been in existence for the past 20 years. The behaviorally based method was originally developed in commercial form at Purdue University and is now known as the Position Analysis Questionnaire (PAQ). The task based approach was developed by the Air Force for training purposes and has been used commercially by many different companies.

Both of these methods are attractive techniques. They appear to eliminate the subjectivity of the job evaluation process, they are numerically scored and computer driven. They objectively assess each job in the same fashion, they are comprehensive, can be done on a PC and streamlines administration because job analysis (desk audits) are not absolutely required. Because of the reduced administrative time requirement, they achieved great popularity approximately 6-8 years ago.

These methods have not been in great use today because of the following disadvantages. They are highly complex and require a well trained statistician to maintain. They are extremely difficult to communicate to employees because they are perceived to be a black box approach. In addition, even though they appear to be objective, the hierarchy created is only as good as the information that is entered. In our experience, you cannot eliminate a thorough job analysis if you want to provide a control on the consistency of the data input. Second, governments have experienced that they need some subjectivity in the process to fairly evaluate jobs that are difficult to describe and document. Finally, the behavioral approach does not result in job descriptions in the normal sense of the word, has never to our knowledge been used successfully in a government setting and does not, in and of itself, control classification creep. In the task based approach, individual questionnaires are needed for each job family, the method has not successfully been used in the government sector in large organizations nor does it assist in developing a job hierarchy which meets standard statistical tests of acceptability.

The factor comparison method is a quantitative job evaluation method in which a series of rankings are conducted on separate compensable factors to assess which jobs contain more of a factor than the other jobs being evaluated. Factor rankings for each job are assigned numerical values and then combined to form a total job score. The job hierarchy is determined by the total scores for the group of jobs being evaluated.

Many public sector organizations use a form of factor comparison, although it is unlikely they call it factor comparison. More than likely governments review classifications and compare and contrast jobs in the same job family by looking at the supervision required, the education or experience required and the level of complexity of the job responsibilities. As a result of this exercise they determine the grade the job being reviewed should be placed.

One of the commercially available methods that is a factor comparison method is the Hay Guide Chart. While this method has been converted to look like a point factor method, it has its roots as a factor comparison approach.

That advantages of this method is that it is relatively reliable, the compensable factors can be tailored to the unique characteristics of the government jobs and values and it does a good job of integrating market data into the system. The disadvantages are that there are no standardized definitions to justify the ranking decisions, it cannot be used for new or changed jobs because the whole ranking process would need to be redone, and it is viewed by employees as too complex and it will not control the expansion of job classes and levels.

The point factor method uses defined factors and degree levels to establish job value. Job descriptions are compared to the definitions of degree levels in order to determine the most appropriate level. The corresponding points for that level are then added to the other levels on other factors to derive a total score.

Of all the methods of job content evaluation, this type of method is most widely used. It is relatively reliable, relatively objective and easy to evaluate new or change jobs. The compensable factors can be tailored to the City/County jobs and values. Of the disadvantages, the point factor method is very time consuming to develop or tailor to the organization. It is very time consuming to use to evaluate jobs; it can take approximately one hour to evaluate each job when a committee is used. The meanings of the definitions do not cover all possible complexities and nuances of jobs in the public sector and thus, the job points can fluctuate widely from person to person. In fact, it is not uncommon that incumbents change a minor task or wording of a responsibility and the points increase, thus leading to classification and grade creep. We believe that many point factor systems have supported the creation of larger departments and steeper organizational pyramids because of the desire to receive more points and thus a higher grade.

While it is viewed a relatively reliable, they have been shown to have an inter-rater reliability in the .70-.80. This means that the total points that a committee arrives at can vary as much as 15%. Variations of this magnitude can mean the difference between 2-3 salary grades. Finally, in our experience, while the point factor method can do a reasonably good job of reducing the number of classifications, it has a less than adequate record in terms of maintaining or controlling grade creep.

A commonly used point factor method in the government sector is the Oliver System. This method has been adopted in a variety of states and some local governments. However, because of the subjectivity of this method, which is also inherent in point factor methods, it has been our experience that this method is rarely implemented fully or used for any length of time. In fact, most organizations that originally adopted the method, no longer use it for job evaluation purposes because it has failed to meet their on going needs.

Recommendation

Most of our clients use a job evaluation method that combines the best features of the non-quantitative methods with the best features of the quantitative methods. The method is commercially known as the Decision Band™ Method (DBM) or the Paterson Plan.

The Decision Band™ Method (DBM) of job evaluation is a highly effective method of objectively evaluating the worth of a job to an organization. A sound theoretical framework serves as the basis of the method which provides a consistent and valid approach to comparing and contrasting jobs.

Initially developed over thirty years ago by Professor Emeritus Thomas T. Paterson, and further refined by Fox Lawson's compensation consultants, through hands on experience in government settings, the Decision Band™ Method has been tested in organizations throughout the world as a means of identifying rational distinctions in pay for all jobs. DBM has been used successfully in both public and private sector organizations. Organizations converting their current job evaluation plans to the DBM framework can be assured that their pay hierarchy will be reliable, valid and justifiable.

The basic premise of DBM is that the value of a job to an organization is based on its level of responsibility. Responsibility is reflected, and therefore measured, by the decision-making requirements of the job. Because all jobs require incumbents to make decisions in order to perform their jobs, decision-making is a logical and equitable basis on which to compare jobs within an organization.

DBM job evaluation involves three basic steps. First, jobs are classified into one of six different "Decision Bands" based on the characteristics of the decisions that a job requires. This classification process reflects the level of responsibility of the job within the organization. The second step involves classifying the jobs within each "Decision Band" into one of two "Grades" based on the differential supervisory difficulty and effort required of the job. Each of the six bands is subdivided into two grades, (coordinating and non coordinating) or subclasses, depending on the requirements to monitor or supervise subordinate jobs. Exhibit 1 is an overview of the DBM Decision Structure.

At this point in the process, jobs will have been classified into a hierarchy of twelve different classes of jobs, based on the job's level of responsibility and supervisory difficulty and effort.

The third and final step involves further subdivision of jobs within each grade into subgrades by reference to the complexity, difficulty, and skills required of the job in relation to other jobs that have been classified into the same Band and Grade. Frequently, this Subgrading is done within each occupational group in order to facilitate the making of relative judgments in comparing the complexity, difficulty, and skills required of jobs across different occupations. The Subgrading process shares many characteristics with point-factor methods of job evaluation.

Finally, the jobs are priced within the relevant labor market.

DBM is easy to use, highly reliable, administratively fast and efficient, can be tailored to the government's needs, and has been proven effective in reducing the number of classifications, reducing grade creep highly difficult to manipulate the value of the job by changing a few words or tasks in a job description. It has been used throughout the country to replace ineffective point factor and classification methods and bring simplicity and rationality back into the process.

Approximately one year ago it was critically examined by the State of North Carolina as a possible tool to be used throughout the State to classify over 119,000 job titles. The State subsequently adopted the method because it allowed them to move toward broad banding and simplification of their classification and compensation structure. Other government organizations have experienced a reduction in the number of classifications using DBM. Finally, because of the way in which it processes jobs for evaluation purposes, we have found that a minor change in the wording of the job or the tasks typically does not result in an increased grade.

We recommend the Decision Band™ Method to any organization that is interested in simplifying its classification and compensation system either in terms of broader bands of classification or because it can be used to rationalize the logic behind pay.