As technical instruments, the current military performance appraisal systems are no worse than the rating scales used in varied sectors of industry and education. But our concern is not with relative merit; too much is at stake for us to be content with simple comparisons.
That our present system apparently has functioned adequately is more a testimony to the quality of our officer corps than to the reporting program. There can be little argument that the current evaluation systems readily distinguish the marginal and the unsatisfactory performers. But what do they tell us about the large percentage of extremely capable officers? We need not know only about those requiring additional training or elimination. We need a better method of selecting the “giants”—those who are head and shoulders above their contemporaries.
More than a few knowledgeable officers have stated that we can achieve such a method with forms—but not with those now in use. The need, they say, is for a more rigid set of marking standards, or a return to “objective honesty,” or a set of well-defined procedural guidelines, or more support from reporting officers. These exhortations are pot new nor have they gone unheeded in the past. The forms have been changed before. During the period 1865-1956, the U. S. Navy used 48 different fitness report formats. All the changes and revisions incorporated in the various formats were for the purpose of making officer evaluations more objective and more realistic. With each change, the forms demanded increasing effort of the reporting seniors.
From their inception, military appraisal systems have relied for their success upon the reporting official. The basic assumption has been that with due effort, the man entrusted with a military mission was singularly qualified to appraise the performance and behavior of his subordinates. When military organizations consisted largely of unskilled men performing repetitive and identical tasks, evaluation gravitated toward the commanding officer and his staff. Considering the needs which originally existed for some type of appraisal, the existent state of the “psychological art,” and military tradition, it is little wonder that responsibility for performance appraisal came to rest exclusively with the commanding officer. He had the right and the obligation to evaluate, because he more than anyone had the ability.
If, because of the increasingly technical complexity of the military environment, reporting officers are becoming less able to evaluate performance accurately; and if “formology” is, at best, a mutable science, where, then, does this find us? It finds us exactly where we are now, and where we have been subsequent to World War II—in a deplorable circle.
Our mechanized and specialized military environment has created a much more severe human relations problem than that which automation once promised to eliminate. The advanced technology which is reducing much of the routine human input is the same technology which is demanding better humans.
The Armed Forces have been aware of this dilemma, and have attacked it with varied approaches. Yet, with rare exceptions, we have not seen the evaluation/appraisal system used in the attack.
The fact is that now, when we need to use every developmental tool available, we have overlooked one with perhaps the greatest potential—one that has been in our midst since the early 19th century—the fitness report. Throughout their long history, however, the report systems have only provided opportunities for reporting officers to exercise leadership; none of them have demanded leadership from these seniors.
Evaluation systems can and should be motivational and informational devices as well as administrative tools. They should demand that reporting officers continually measure results, and they should demand, as well, that seniors continuously develop the professional abilities of those in their charge.
Those who decry any attempt to overhaul the military evaluation systems are not granting the same recognition to the advances in the social sciences as have been given to advances in the physical sciences. The early military appraisal systems were the forerunners of evaluation systems later founded in public and private fields. It is disconcerting to note that a number of progressive industrial concerns, and some of our federal agencies, viz., FAA and VA, have acknowledged social science gains and have incorporated them in their appraisal and development programs. Why those agencies and not the military?
Even to begin to answer the question, certain assumptions must be made. These assumptions are neither particularly controversial nor original, but, upon their acceptance or rejection will rest largely the acceptance or rejection of the recommendations which will follow.
Assumption No. 7. Career officers take pride in their profession and in the performance of their duty. The promise of financial reward is an easily understood motivation, but altruistic pride in one’s work requires further elaboration. Peter F. Drucker has expressed himself on this matter:
. . . there are few relations which so completely embrace a man’s entire person as his relation to his work. Work was not, Genesis informs us, in man’s original nature, but it was included soon thereafter. “In the sweat of thy brow shalt thou eat thy bread” was both the Lord’s punishment for Adam’s fall and His gift and blessings to make bearable and meaningful man’s life in his fallen state. Only the relationship to his Creator and that to his family antedate man’s relationship to his work; only they are more fundamental.
Assumption No. 2. Appraisal is required for administrative, informative, and motivational purposes. Promotion, elimination, and job assignment are but three of the more obvious administrative purposes served. The informative purposes and, indeed, the motivational purposes of appraisal cannot be served, however, unless the man being marked desires and appreciates constructive criticism. Substantial controversy exists as to whether a subordinate in fact wants to know of his shortcomings. Many feel that, however tactfully criticism is proffered, it creates undue anxiety and inhibits any further meaningful communication between the rating officer and the subordinate. It is the position of this paper that if the criticism is timely, skillfully offered, and in relation to jointly-determined standards, friction is not necessarily the final result.
Assumption No. 3. All job assignments have certain requirements in common. One of these common denominators is planning, or working out the necessary details for fulfilling the mission of an organization plus the methods of achievement and of measuring performance. Another is directing: the continuous task of making decisions as the leader of an organization, and of incorporating them into orders, instructions, directives, etc. A third is controlling, or influencing the behavior of individuals toward the achievement of organizational objectives. A fourth is motivating through the setting of high standards; demanding peak performance of one’s self and one’s subordinates. And a fifth is innovating, through the application of advanced skill and knowledge in all areas of responsibility; developing new processes and techniques; and improving upon the old.
Assumption No. 4. A way must be found to measure results rather than behavior. Throughout the Armed Forces, stress is placed upon performance, but rating officers have difficulty in appraising it. In the end, the raters have little beyond the behavior of their officers to observe and judge. This creates a host of perceptual and judgmental problems, among which are: the rater sees nothing but behavior; yet, he is asked to interpret it, place a value on it, and record the evaluation on a scale; he must continually strive for objectivity; and he must, of course, understand his own involvement.
Because “behavior-reporting” is difficult, it has a tendency to become an end in itself. Instead of restricting their concern to “intelligent behavior,” for instance, diligent raters find themselves evaluating an officer’s “intelligence”—and this, even the raters would agree, they are not qualified to do. Informed as most seniors are concerning human behavior, they are not psychologists. Perhaps this explains in part why seniors experience discomfort in completing evaluation reports, because in the hands of conscientious individuals the system induces them to be something which they are not.
Assumption No. 5. A valid and reliable source of information concerning job results is the man being rated. Let the ratee assist in an appraisal of his own performance? Why not? Who knows more about the requirements of the job than the officer currently in that assignment? If we trust an officer to develop his officer and enlisted subordinates, can he be trusted less in his own development?
Numerous social studies have shown that the output, growth, and morale of civilian workers rise when they are able to exert some measure of control over their work environment. There is reason to believe that the same studies among the officer corps would produce similar results. Another reason for the ratee’s participation in his own appraisal is that, although most assignments in the Armed Forces are explicitly defined, all of them are perceived and implemented in a variety of ways by their officer incumbents. Officers continuously search for ways of accomplishing jobs that are better than any previous method. In attempting to control and improve the work environment, then, it is essential that the officer and his senior understand, and agree upon, the nature of these efforts and the yardsticks by which results will be measured. When such accord is achieved, the ratee will feel more assured and committed, because he has contributed measurably to the planned betterment of the organization, and because he now knows exactly what is expected of him by his senior, and also the manner in which his fulfillment of these expectations will be measured.
The appraisal systems in use by the military can be improved considerably through adaptations based upon the foregoing assumptions. No experience in this matter exists in the military, however, and that which can be pinpointed in industry, though promising and exciting, is nevertheless, sparse. In this area as in any, the military must “walk” before it “runs.” It must build up its own body of specific knowledge through gradual implementation starting first with regular, active-duty career officers in the more senior ranks. Once the assumption can be tested and refined, it can be expected that more precise techniques and skills will emerge to apply to the more junior ranks.
As with the current appraisal program, we envisage that a system developed from our assumptions would be as applicable in war, perhaps more so, as under conditions of peace.
The problems which have been encountered in administering the present-day rating scale system clearly indicate to many that there is a more profound change required than merely changing the administrative procedures or forms.
Proponents of rating scales contend that the answer to the dilemma cited is three-fold: improving the style of the items contained in the forms; improving the scale format; and training the raters. It just has not worked. If it has not worked in the simple environments of the past, then why should we believe that rating scales with all their improvements could withstand the increased demands of the current environment?
Further, the rating scale concept, as we have known it has not provided for introduction of information from the most knowledgeable source—the rated man himself.
Let us pause for a moment and review the present rating process. Although there are minor procedural variations between branches of the military service in preparing and submitting performance appraisals, there is a general similarity in the rating process, as well as in the underlying assumptions (see Figure 1). This rating process involves observation and evaluation of an officer by his supervisor or commander who periodically prepares and submits a report. Such a report is then reviewed and/or endorsed at each succeedingly higher command level and is finally deposited in the permanent file of the officer reported on at the administrative headquarters of his particular branch of the service. Except for Army officers, the man whose performance has been appraised is not shown the report unless the appraisal indicates his performance is below minimum standards. In such cases, the officer is notified, shown the report, and given an opportunity to express his comments or rebuttal before the report is forwarded. Since a very small percentage of performance appraisals result in substandard ratings, most reports are routed up through channels without ever having been seen by the officer who has been evaluated.
The appraisal form used for the report typically includes the officer’s job description, a series of personal traits to be rated, and space for a narrative description of his performance. Also, an over-all summary evaluation indicates his comparative standing on a distribution curve of officers of equal rank.
The basic purpose of this appraisal system, of course, is to provide reliable information upon which to base personnel decisions. For example, the appraisals allegedly represent the best available information regarding the officer’s comparative value to the military service. More specifically, the appraisals represent the most important single document for determining the officer’s qualifications for promotion, assignments, and special training opportunities.
This is only a small fraction of the functions which are desired in a performance appraisal system. Performance appraisals should provide valuable aids in such functions as personnel development, job satisfaction, and motivation toward achievement of organizational objectives. Yet, as we shall see, the present system not only fails to assist in these functions, but it also produces somewhat less than reliable information for assessing the quality of the officer personnel inventory and for making decisions affecting individual officer’s careers.
There are several assumptions underlying the current rating system which have a marked affect upon the results achieved from its use. The assumptions which are implied in the present procedures are:
The reporting officer is the only reliable and competent source of information for appraising the performance and behavior of his subordinates. This assumption places the responsibility for the success or failure of the system squarely on the shoulders of the rater. In three of the four services—the Army uses a rater-endorser system wherein both fill out all portions of the report—the rating is prepared by the reporting senior alone. Although it is reviewed and endorsed by higher authority, the reviewing official acts as little more than a rubber stamp because of his limited contact with the rater.
This presupposes that all reporting officers are capable of making unbiased, objective appraisals of the subordinate’s performance. Experience has shown that this assumption is not valid. To illustrate this point, a few significant variables are listed below:
• The rating officers differ in competence and experience in the skill or technology in which the subordinate is performing.
• Rating officers differ in management methods, i.e., authoritarian as compared to permissive; a subordinate’s response will differ under each method.
• Rating officers interpret the content and words of the rating procedures and forms in varying ways.
• Rating officers differ in the thoroughness and perceptiveness of their observations of subordinates.
We do not feel it necessary to expand this list further to illustrate that bias alone seriously reduces the validity of the present system.
The main fallacy in this assumption is that the rater fails to make full use of a better source of information, i.e., the subordinate himself. In the rating process the subordinate is assigned a passive role of being appraised and must rely upon the judgment of his senior for an unbiased appraisal of his work. There is, as noted previously, an ever-in- creasing body of knowledge which clearly shows that when workers are given the opportunity to participate in establishing their own work rules, their performance is significantly improved. By the same criterion, the quality of performance appraisals will be improved if the subordinate is given the opportunity to participate in setting his own work objectives and in evaluating his performance against those specific objectives. This procedure would considerably decrease reliance on the single source of information and would tend to reduce the amount of bias.
An officer's performance can be evaluated in terms of observed personal behavior characteristics. This assumption requires the rating officer to place a quantified value on such intangible factors as moral courage, loyalty, planning ability, leadership, and many others which have no quantifiable measurements. These are admittedly subjective values dealing with the personal characteristics of the rater. It is simply not possible to isolate any one of them, measure it, weigh it and place it into context as a quantified value of such a complex entity as a human being. Yet, this is what is required of the rating officer.
Given the variables listed in the discussion of the first assumption, the process of translating the subordinate’s performance into personal characteristics further reduces the validity and reliability of the rating.
To improve the usefulness of the performance appraisal, as well as to reduce it to a more objective rating, we propose that instead of trying to quantify personal characteristics, the rating should be based on what was done. How much more meaningful is it to know that a supply officer did reduce his inventory value by $2,000,000 while increasing his supply effectiveness rate from 80 per cent to 85 per cent than it is to know what his reporting senior feels about his moral courage or loyalty or leadership?
To the extent that merit is a factor, the appraisal system provides a basis for organization rewards and punishments.
The performance appraisal is an important factor in decisions to promote, eliminate, and assign personnel. The fact that it is used for these purposes makes it a powerful personnel control device. The subordinate’s career progression or lack of it depends largely on the ratings he receives. He is therefore motivated to become “boss-oriented,” because the senior subjectively determines the future of his career. In this situation, the subordinate’s integrity may suffer compromise when his opinions or decisions differ from those of his senior. Therefore, motivation and personnel development are largely a function of how the subordinate is managed and do not necessarily arise from organizational requirements.
If the rating system were modified to permit the individual subordinate to establish his own targets toward accomplishing the organizational mission, establish the criteria against which his performance is measured, and appraise his actual performance against those targets, many of the problems would be eliminated. First, his motivation would be “mission oriented” instead of being “boss oriented.” Second, he would be relatively free to innovate and bring the full potential of his ability into the challenge of achieving the organizational objectives without the threat of being unfairly rated. Finally, he would be able to achieve a greater degree of self-fulfillment.
The Services have been confronted with three serious problems in administering the rating system. These problems are the tendency toward rater leniency, the “blending” tendency, and rater bias.
Rater Leniency, of course, is that almost universal inclination to give “good” ratings— with a corresponding reluctance to give “low” ratings. The tendency toward rater leniency is understandable, and in the broad human sense even praiseworthy, but it inevitably reduces the usefulness of any effectiveness reporting system. The Chief of Naval Personnel in a memorandum to all officers signing fitness reports remarked that only five months after a new rating form was prescribed there was already a noticeable “high marking” tendency. This tendency reflects the skepticism with which supervisors approach the onerous task of passing formal judgments on other human beings.
In addition to the obvious inequities this tendency causes in promotion decisions, a more serious problem may develop from failure to eliminate the unfit as early as possible. Every year that a substandard officer is retained, his service correspondingly becomes more deeply committed. The longer such an officer is retained, the greater the injustice that is done to him, the service, and his dependents. When this happens, the system which generated the lenient ratings is as much at fault as the man himself.
The Blending Tendency thwarts the intent of the rating system, which is to rate separately each of the characteristics of the rating form without regard to the ratings given on the others. Analysis of a large number of reports, however, discloses that many rating officials, despite their sincere desire to rate the various characteristics independently, show a tendency to rate according to a general impression of the officer concerned.
Because of this tendency, the performance records do not reflect the specific tasks that the officer does well or the ones he does poorly. Under these conditions, it is difficult to make the most effective use of manpower. Moreover, training programs may not be most prudently directed.
Although Rater Bias caused by differences in the background or philosophy of individual rating seniors is recognized, there has been no satisfactory method found by which to adjust ratings to a comparable scale. Studies to determine how to correct for rater bias have met with three currently unresolved obstacles: lack of rater histories; lack of individual rater consistency; and rating officers resistance to having their evaluations changed. There is doubt that these problems can be resolved in the present system.
The new reporting system being discussed in this paper draws its inspiration essentially from two sources: first, The Human Side of Enterprise by Douglas McGregor, and second, Professor Alva F. Kindall, of the Harvard Business School, and his performance appraisal concepts. We have attempted to “pick up the ball and run with it” insofar as application, innovation, and adaptive creativity are concerned.
An illustration of our approach can be best achieved by examining a hypothetical report format (see Figure 2). But, first, a word of explanation of the term, “worksheet,” as used hereinafter, is in order. When the rating officer fills in a Fitness Report under the present system of appraisal, he takes into account recommended markings made by an officer junior to himself, but senior to the man being rated. These recommendations to the marker generally are made on a worksheet. [Only the U. S. Army does not use the worksheet concept.]
Our system of appraisal also requires a worksheet. But it in no way resembles those now in use. The form we have devised to replace the fitness report—we call it, for obvious reasons, an Officer Progress Report— could not be filled in without a worksheet. Yet, as we shall see, our worksheet requires no third person participation. There are two other innovative aspects to our proposed appraisal system. One is a deletion, and the other is an addition. The deletion is that section of currently used reports which requires reporting seniors to rate officer characteristics or personality traits. The addition is the incorporation of a “goal-setting” technique and the concept of appraising a man on his specific achievements.
The “goal-setting” device is a reasonable and easily understood concept. Goal-setting is the procedure whereby an individual establishes performance targets for each of his responsibilities for the rating period.
It is not within the scope of this study to criticize or change military job descriptions. Thus, when a superior and a subordinate agree on the subordinate’s goals for the coming period, they will be doing so within the confines of a specific job description which has been designed for the over-all needs of the service. In the process, the superior, rather than acting as a rubber stamp, analyzes the goals and suggests such changes, additions, or deletions as he feels are required. A give-and-take of this nature could occur several times before a mutually agreed upon set of goals is established.
What are the criteria for these goals? First, a clear-cut relationship must exist between the individual’s goals and the organizational mission. Too, the goals must be measurable. (The person being rated must determine the method to be used in measuring his goals.) Goals should be time-phased, in that periodic target dates are set for joint review and evaluation. A challenge must be present in each goal in order to further organizational purposes as well as to require the best effort of the goal setter. If the goal is too easy, an injustice is done both to the development of the man and the organization. Goals should represent improvements over past performance; they should be mutually exclusive, and should be realistic and attainable.
Hence, it can be seen that when mutually agreed upon goals meet these criteria, the rater becomes a reviewer rather than a policeman, amateur psychologist or a conscience-stricken “good guy.” In addition, the rated man can easily rate himself, and the appraisal interview then becomes the discussion of mutually recognized results.
The rated man should develop his own measuring system in the appraisal of his achievements for two reasons: first, by committing himself to a specific goal and then by defining the terms on which it will be measured, he is totally involved; and this is consistent with the assumption that the subordinate likes to work and likes to know where he is going and how he is doing. Second, it makes the target dates more meaningful in that progress or lack of it will be more easily recognized.
Specific target dates are used to measure progress toward goal achievement for several reasons. First, they afford a convenient way of checking progress. Second, they allow assessment of the realism and attainability of the goals. Third, they have an immediate motivating function. Fourth, they encourage and demand forward planning. Finally, use of targets offers training for both the rater and the rated in gaining proficiency in achievement techniques.
The worksheet, then, serves as a basis for a portion of the Officer Progress Report. (The Officer Progress Report is not illustrated on these pages because its final form doubtless would continue to vary as greatly between services as the present fitness reports do.) It is important only that the Officer Progress Report consist of five separate parts, Part I is the Personal Data section. This is the standard information currently appearing on all fitness reports, e.g., Name, Rank, MOS, Periods of Non-availability, etc.
Part II is the Achievement Appraisal section. This would be filled in, as amplified on page 84, exclusively from information transcribed, almost verbatim, from the Worksheet.
Part III of the Report is the General Comments section. In this portion of the report, the marker states his attitude toward the man as a member of his command in comments ranging from “Particularly desire to have” to “Prefer not to have.” In this section, too, the reporting senior must give a concise statement of his general impressions of the man being marked, and of his potential worth to the service. Here he will also note whether the goals had been changed and, if so, why; and he will state whether previously unforeseen circumstances had either helped or hindered the attainment of the goals. At first glance, it would seem that this section, calling as it does for an “opinion” by the marker, contradicts this paper’s thesis. One point should be understood clearly: the Achievement Appraisal section substitutes more reliable data for the “opinion” of the superior. But the marker clearly must express an opinion in Part III. The major innovation, then, is that the opinion is isolated and clearly identified for what it is.
In Part IV, the Recommendations section, the marker makes any recommendations he desires. But, specifically, he suggests what the man’s next duty assignment should be.
Part V, the Signatures section, is devoted to the signatures of the marker, the man being marked, and of the reviewing officer.
As was indicated earlier, the rating of personality traits has been deleted. Significantly, the Services place a great deal of emphasis on command responsibility, judgment, and decision-making under stress. Officers of suc- ceedingly higher grades are assumed to possess more experienced judgment. Thus, it is felt that we are not asking a senior to be omniscient by requiring of him an opinion in the General Comments section of the Progress Report concerning the value of this man to the organization.
It is suggested that there are three beneficiaries to the Reports’ General Comments' section: the promotion boards, the rater, and the rated. In the case of the promotion boards, the military promotion system is faced with problems unheard of in industry. It is possible, for example, that the Army may have 15,000 lieutenant colonels from whom to select 1,500 for promotion to colonel. This is a huge task—to evaluate each candidate and make a meaningful decision that each member of the board can live with. The need is foreseen by the writers for an additional tool such as the General Comments section with which to delineate between officers. As Dr. Howard C. McArthur of Harvard University has said, it is not so hard to delineate between fair and poor, but much harder when the choice is between good and better.
A very basic element of every job, and military jobs are prime examples, is compatible interaction with fellow men. It would be extremely valuable and significant if a man who had achieved 100 per cent or more of his goals over a long period of time was doing so at the expense of his subordinates and at the constant irritation of his fellow officers. If no one liked working with this man, albeit his unit met goals, but had extremely poor morale, this would be the type of thing that could be responsible for losing men in combat. A promotion board has the right to such information. Moreover, the commanding officer has the obligation to inform the board.
This appraisal forces the rater to think more critically about the man, and most important, aids him in suggesting goals for the man being rated for the coming period. Correspondingly, it should help the rated man for the same reasons.
The foregoing discussion highlights a rather obvious fact: namely, that a massive study is needed in the area of implementation of a new system which is based on the assumptions that have been discussed. What is presented here is an introduction and a framework for experimentation and further study. Not only could the system itself be tested, but answers could be obtained to such questions as: what procedures will be taken to provide for an officer whose goal period is interrupted for any one of many possible reasons?
In summary, then, there are advantages and disadvantages in the use of the Officer Progress Report. Some of the advantages are: it demands new and more productive assumptions about the motivation of officers; it no longer requires the rating officer to play God; it accomplishes the desired effect of tailoring the job to the man, and, conversely, avoids stereotyping a job or a man; it gives promotion boards a new yardstick—what has the man achieved? Further, the reviewing officer can make judgments of his own based on the goals selected and the priority given to them by both rated and rater. The Officer Progress Report also forces critical thinking about responsibilities and weaknesses both on the part of the rater and the rated; it will improve superior-subordinate relationships, because very little, if any, hostility can arise from measuring performance against mutually agreed upon goals; and, in addition, the various roles are much more clearly defined.
On the other hand, certain problems will result from the use of this new system, which might be categorized in the following manner:
Logistics: The matter of educating the Services generally and individual officers, specifically, poses a big job. Regardless of where, how, and on what scale it is started, the problems will be similar; only the scale will change as the system begins to spread out, or down, or both.
Administration: Promotion boards will be hardest hit at first, because they will have to adopt a new rating system in “midstream,” so to speak, and consider the careers of many officers at the same time. Morale is the immediate concern, and the answer again turns to the word “implementation.”
Inertia: Resistance to change is to be expected. Not the least will come from those who would like to be considered omniscient. But, more damaging will be the voice of the traditionalist and the conservative who regard present methods as time-tested and sacred: “You will ruin the service,” they will say.
Collusion. The rated and' rater may enter into a conspiracy. What is to prevent the rater who wants to be liked and the rated who wants a good report from setting, jointly, an easy set of goals? There are two preventatives. First, self-control on the part of each, especially the rated man, is viewed as a powerful stimulus and motivator. It is assumed that a man who is given the opportunity to chart his own future in terms of accomplishments will want to attempt to accomplish as much as he possibly can. Secondly, the rater is also being rated and is, himself, setting goals. Thus, the echelons of goal-setting are progressive and self-serving as long as the man at the top knows what he wants and is strong enough to inspire high standards from his subordinates. Lastly, a reviewing officer will doubtless be most interested in the goals, for, therein, both rater and rated will be giving their joint opinion of what they consider to be the objectives of the command.