rating scales. As the system becomes fully implemented each user will have a set of relevant and familar approved benchmark descriptions to guide his decisions in evaluating positions. The factor rating scales give the user an overall view of the system and the interrelationship among the several factors. The scales are also used in making interoccupational comparisons. The rating scales are particularly helpful in evaluating new jobs for which job relationship patterns have not yet been established. New benchmark position descriptions can be added to the system with relative ease. This will permit the system to be quickly responsive to the dynamics of a changing Federal work force. This type of system can be used to evaluate a wide variety of jobs with reasonable assurance of consistency of job treatment among agencies. Thus, the system has potentially many positive advantages over the present position classification system. In addition, the total system consisting of factor rating scales backed up by very specific benchmark positions can reduce significantly the opportunities for abuse.

Concept of Administration of the System

It is envisioned that the Civil Service Commission will maintain control over the benchmark positions used on a Government-wide basis. Individual agencies would prepare and use additional benchmark positions subject to approval by the Civil Service Commission. Field installations and subordinate units of an agency would also prepare benchmark descriptions for their unique positions, subject to approval. by their next higher agency echelon. As is true under the current General Schedule system, agencies would have responsibility for assigning individual positions to appropriate skill levels. However, preliminary studies indicate that the technique is one that is relatively easy for line managers and employees to use and understand. Therefore, the system will permit agencies to delegate position evaluation authority to managers well down in an agency's organization.

Basic Procedures Followed in Developing the System

Selecting and Defining Factors

The significant factors and subelements to be used in measuring the relative worth of work were identified through a review of the nature of the work included in occupations in the APTES category. A survey was also made of the factors used in other evaluation systems for similar occupations. The considerations in selection of factors were that they are:

--Measures of significant and distinct characteristics of work.

--Applicable to all positions but in varying degrees.

--Ratable in terms of recognizable and definable differences in level.

This factor measures the nature and extent and level of knowledge
and abilities needed to perform work acceptably.

Factor II. Difficulty of Work

This factor measures the complexity or intricacy of work and mental demands, i.e., judgment, originality, and other mental effort required, as affected by quality and relevance of the available guidelines.

Factor III. Responsibility

This factor measures the assistance and control provided by the supervisor and the impact of work on the accomplishment of the mission of the organization.

This factor measures the skill required in work relationships with others and the importance of such relationships to the success of the work.

Other Requirements

This factor measures any special or unusual requirements or conditions in a job that add to the difficulty of the work but are not adequately considered by the other factors, i.e., unusual physical effort, mental concentration or environmental impact on the work. Factors I through IV should provide adequate means for evaluating most positions; therefore, only atypical positions should be assigned values under this factor.

The five factors selected and their subelements were found to be appropriate for evaluation purposes and produce satisfactory results.

Rating scales were developed for each factor to show the various observable levels within each factor and the interrelationship among closely related or dependent elements. The level definitions in the scales are structured to provide equal differences between successive levels.

Selecting Benchmark Positions

Initially, 152 jobs were selected as tentative benchmark positions to be used in developing APTES. The considerations in their selection were that the:

--position is in an occupation which stands out in the total job population;

--job is common throughout much of the Government or is important to a major phase of Government activity;

--duties can be clearly defined;

--job is broadly related to other occupations;

--Civil Service Commission had published standards;

--job is covered in the PATC or BLS salary surveys.

The class series from which these initial benchmark positions were selected represent 51% of the total job population in the APTES category. Some additions and deletions were made in the original list based on the experience of the writers in preparing definitions for each of the jobs selected. These descriptions followed a uniform format; i.e., a brief summary of the principal duties followed by descriptive data outlining each factor. The tentative factor scales were available during the process of description writing. This permitted the writer to assure that the descriptions provided the job data needed to apply the rating scales. (The package of benchmarks for the model system includes 118 positions.)

Ranking and Application of Tentative Rating Scales to Benchmark Positions

In order to establish a base line for development of the ranking system, and also to better identify anomalies in the existing occupational relationships, a six-man panel was selected to rank a smaller sample of 48 jobs from the initial list of benchmark positions according to level of difficulty and responsibility of work. To this sample was added two jobs at overlapping levels from the COMOT benchmarks. The APTES benchmarks selected were for occupational series with populations representing about 40% of the total.

The paired-comparison method was used in ranking the terms of each factor and then on a whole job basis. compares each job with each other job by pairing the

sample of benchmark jobs, first in The rater following this technique jobs with each other. This technique

makes it possible to compare two jobs without being influenced by the remaining jobs. The pertinent literature and textbooks suggest that under this technique the:

--rankings are individually simpler;

--assumptive bias is reduced significantly;

-- results are likely to be reproducible.

The panel members were then given a set of tentative rating scales and asked to apply them to the benchmark positions. The results of these ratings, along with the rankings described above, provided basic data essential to the development of the system including the firming up of the preliminary identification of levels within each factor element.

Developing Point Values and Factor Weights

Upon completion of the factor scale definitions, a system of appropriate factor weights and point values was established.

It is evident that each factor need not have equal weight or significance in measuring the relative worth of work.

Factor weights were obtained by systematically testing possible combinations of factor weights until the combination was identified which produced an acceptable rank order of the tentative benchmark positions in comparison. The weights used in the COMOT system produced a satisfactory relationship. The weights are:

Each level in each factor was assigned a point value which is a multiple of ten. For example, the lowest level in Factor I has a value of 20 (2 x 10) and the highest level has a value of 160 (16 x 10). The actual number of levels in Factor V has yet to be determined. (See discussion in the guide chart for Factor V.) A minimum score of 10 was established to discourage its use in evaluating minor job differences. An arbitrary 20 point maximum was established. The factor scale values are:

All of the tentative benchmark positions were rated according to the factor rating scales as thus weighted. The positions were then placed into groups or "skill levels" by judgmentally considering two characteristics:

--the similarity in point value and hence in level of difficulty and
responsibility; and

--the pay relatives for each position.

The objective of the grouping was to achieve pay differences proportioned to work differences for positions and occupations in the APTES category.

A table was then constructed to convert the total point score for any position to an appropriate skill level. The table provides a consistently wider range of points for each higher level of skill, thereby reflecting typical patterns of pay differences and also the broader range of work in successively higher levels of jobs.

Provision for Trainee Positions

The evaluation of truly trainee positions on the basis of duties and responsibilities is always difficult and becomes especially controversial when trainee positions are placed in the same skill levels used for other types of productive workers. APTES proposes to avoid this problem by not placing trainee positions into skill levels. Rather, trainee positions will merely be paid a salary which will be a percentage of the entry salary for the first target position for which the employee is being trained. The exact pay for the trainee can be made proportional to his relative skill and the amount of training needed.

For example, the lowest productive level for an engineer would be Level V. The training level salary for a recent graduate engineer would be based on a percentage of the salary for Level V. Upon satisfactory completion of training, the engineer would move to Level V.

OCCUPATIONS COVERED BY THE EVALUATION SYSTEM

for

ADMINISTRATIVE, PROFESSIONAL, AND TECHNOLOGICAL POSITIONS

APTES includes all occupations in which the principal nonsupervisory positions at the full performance level are:

--of an administrative, professional, or technological (or paraprofessional)
nature; and

--like exempt jobs as defined in the Fair Labor Standards Act of 1938;
and

--like jobs typically paid on a national scale basis by other employers;
and

--most typically require either a baccalaureate, graduate level, or
professional level of education, or an extensive body of specialized
knowledge acquired through on-the-job training and experience.

APTES is applicable only to nonsupervisory positions. All supervisory and managerial positions, including working leader type positions, are evaluated and paid under the Supervisor and Manager Evaluation System (SAMES).

APTES specifically includes the following occupations:

(This list will be developed and issued by the Civil Service Commission

prior to installation of the system, and will be revised thereafter as
necessary to assure its accuracy at all times.)