The Rate of Improvement in Efficiency : Part 1

Increasing Efficiency In Business: A Contribution to the Psychology of Business
by Walter Dill Scott, Ph.D.

(Page 10 of 16)

No novice develops suddenly into an expert. Nevertheless the progress made by beginners is often astounding. The executive with experience is not deceived by the showing made by new men. He has learned to accept rapid initial progress, but he does not assume that this initial rate of increase will be sustained.

The rate at which skill is acquired has been the subject of many careful studies. The results have been charted and reduced to curves, variously spoken of as "efficiency curves," "practice curves," "learning curves," according to the nature of the task or test. Some of these dealt with the routine work of office and factory. In others typical muscular and mental activities were observed in a simpler form than could be found in actual practice.

Five of my advanced students joined me in strenuous practice in adding columns of figures for a few minutes daily for a month. Our task was to add 765 one-place figures daily in the shortest possible time. No emphasis was placed on accuracy, but each one tried to make the highest daily record for speed. The results of our practice are graphically shown in Curve A of Fig. 1. As shown in that curve for the first day our average speed was only forty-two combinations per minute, but for the thirtieth day our average was seventy-four combinations per minute, We did not quite double our speed by the practice, and we made but little improvement in accuracy. The most rapid gain was, as anticipated, during the first few days. We made but little progress from the sixteenth to the twenty-third day, and also from the twenty-fourth to the thirtieth day.

Of the six persons practicing addition, five of us also practiced the making of a maximum grip with a thumb and forefinger. Just before beginning the adding each day this maximum grip (or pinch) was exerted once a second for sixty seconds, first with the right hand and then with the left. Likewise at the completion of the addition sixty grips were taken by the right hand and sixty by the left. The total pressure exerted by each individual in the 240 trials (four minutes) was then recorded and expressed in kilograms. The result of the experiment is shown in curve B of Fig. 1. The average total pressure for each of the five persons was for the first day 620 kilograms; for the twenty-fourth day 1400 kilograms. Our increase was very rapid for the first few days, and no general slump was encountered till the last week of practice. In one particular our results in the test on physical strength were not anticipated - we did not suppose that by practicing four minutes daily for thirty days we could double our physical strength in any such a series of maximum grips with the thumb and forefinger.

It is a simple matter to measure day by day the accomplishment of one learning to use the typewriter. All beginners who take the work seriously and work industriously pass through similar stages in this learning process. Figure 2 represents the record for the first eighty- six days of a learner who was devoting, in all, sixty minutes daily to actual writing. The numbers to the left of the figure in the vertical column indicate the number of strokes (including punctuations and shifts) made in ten minutes. The numbers on the base line indicate the days of practice. Thus on the ninth day the learner wrote 700 strokes in the ten minutes; on the fifty-fourth day 1300 strokes; on the eighty-sixth day over 1400 strokes.

In this curve it will be observed that the increase in efficiency was very great during the first few weeks, but that during the succeeding weeks little improvement was made. - BOOK, W. R, "The Psychology of Skill," p. 20.

The progress of a telegraph operator is determined by the number of words which he can send or receive with accuracy per minute. In learning telegraphy, progress is rapid for a few weeks and then follow many weeks of less rapid improvement. Figure 4 presents the history of a student of telegraphy who was devoting all his time to sending and receiving messages. His speed was measured once a week from his first week to the time when he could be classed as a fully accomplished operator. By the twentieth week this operator could receive less than 70 letters a minute, although he could send over 120 letters a minute. At the end of the fortieth week he had reached a speed of sending which he would probably never greatly excel even though his speed was far below that attained by many operators. The receiving rate might possibly rise either slowly or rapidly until it equaled or exceeded the sending rate. - BRYAN & HARTER, "Studies in the Physiology and Psychology of the Telegraphic Language," Psychological Review, Vol. IV, p. 49.

There are certain forms of learning and practice which do not readily admit of quantitative determinations. Nevertheless very successful attempts have been made even in the most difficult realms of learning. A beginner with the Russian language spent 30 minutes daily in industrious study and then was tested for 15 minutes as to the number of Russian words he could translate. Figure 5 shows diagrammatically the results of the experiment. Thus on the thirteenth day 22 words were translated; on the fiftieth day 45 words. Improvement was rather rapid until the nineteenth day, and then followed a slump till the forty-sixth day. Improvement was very irregular. - SWIFT, E. J., "Mind in the Making," p. 198.

These five figures are typical of nearly all practice, or learning, curves. They depict the rate at which the beginner increases his efficiency. In every case we discover very great fluctuations. On one day or at one moment there is a sudden phenomenal improvement. The next day or even the next moment the increase may be lost and a return made to a lower stage of efficiency.

There are certain forms of skill which cannot be acquired rapidly in the beginning. In such instances a period of time is necessary in which to "warm up" or in which to acquire the knack of the operation or the necessary degree of familiarity and self-confidence before improvement becomes possible. This is true particularly in the "breaking in" of new operators on large machines like steam hammers, cranes, and the like, where the mass and power of the machine awes the new man, even though he has had experience with smaller units of some kind. It applies also to new inspectors of mechanical parts and completed products in factories - especially where the factor of judgment enters into the operation. Such instances are exceptions, however, and differ from those cited only in having a period of slow advance preliminary to the rapid progress.

Apparently, improvement should be continuous until the learner has entered into the class of experts or has reached his possible maximum. As a matter of fact the curve which expresses his advance towards efficiency never rises steadily from a low degree to a high one. Periods of improvement are universally followed by stages of stagnation or retrogression. These periods of little or no improvement following periods of rapid improvement are called "plateaus" and are found in the experience of all who are acquiring skill in any line.

These plateaus are not all due to the same cause.

They differ somewhat with individuals and even more with the nature of the task in which skill is being acquired. With all, however, the following four factors are the most important influence: -

1. The enthusiasm dependent upon novelty becomes exhausted.

2. All easy improvements have been made.

3. A period of "incubation" is needed in which the new habits under formation may have time to develop.

4. Voluntary attention cannot be sustained for a long period of time.

About the Author

Walter Dill Scott (1869-1955) was one of the first applied psychologists. He applied psychology to various business practices such as personnel selection and advertising. In 1900 he was appointed instructor of psychology and education and director of the psychological laboratory at Northwestern University. In 1905, Dr. Scott was made professor of psychology and head of the department of psychology.