Comparative Evaluation of Performance Using Multiple Criteria

Critique of “COMPARATIVE EVALUATION OF PERFORMANCE USING
MULTIPLE CRITERIA” by HERBERT TERRY

The author talks about the main point that the comparative evaluation of performance may be necessary for purposes of selection, assignment, or improvement. The three major point are being mentioned like this: Selection aims at choose the one that is suitable most, assignment means assign the task to the people who can do it best, and improvement, the most difficult one, means helps the individual or product to improve. This three major point is so important and being discussed in the whole article.

In order to explain the main point, the author mentions about the example of average grades, which we become so familiar with when we are in school. And he mentions that one important further step is to assemble the available data in tabular matrix form. He also mentions the way that transform verbal rating of the sort to a numerical. The author also provide several methods for rating overall performance.

As far as I am concerned, the comparative evaluation of performance is of great value to discuss and it can be applied to many real problems and analysis. Although the opinion of Dr. Herbert is really splendid, intelligent, and outstanding, some part of the article is still lack some evidence and can be discussed about for more details, and some points may still have some space for improve. Strength and weakness of this article are illustrated as following.

Strength,

Dr. Herbert carefully discuss about several methods for rating overall performance. Method I, which is the most common one, just use the average rating of the product we talk about. As for Method II, it uses the exponential weighting, which is more accurate but it is difficult to visualize its significance in any given situation. The last method use the regression analysis, as a well-known statistical technique which can be effectively utilized for purposes of assignment. However, this method need a great number of data so it is not easy to use in most cases. The author describe the three methods with explanation and example, which makes the methods really clear to the readers.