SCI 105
"The Hawthorne Effect"
Text and tables drawn from:
Roethlisberger, F. J. & Dickson, W. J. (1941). Management and the Worker: An Account of a Research Program Conducted by the Western Electric Company, Hawthorne Works, Chicago. Cambridge, MA. Harvard University Press.
Imagine that you are a member of the management at a large factory. You would like to increase the output of your production crew, but you are unsure about the best way to do this (assuming that leg irons and flogging are out of the question). You have some ideas for changes that you might make to the work environment, but you don’t know which will be most effective (or deleterious!). What might you do? And, more importantly, how would you do it?
The management at the Hawthorne Works of the Western Electric Company (which produced a variety of telephone equipment for Ma Bell) was faced with just such an issue. In 1927, they launched a series of investigations to try to answer these questions. The following are excerpts of the Roethlisberger and Dickson account of their research:
It was decided to isolate a small group of workers in a separate room somewhat removed from the regular working force, where their behavior could be studied carefully and systematically. Although there was no idea of a complete control of all the factors affecting work performance, it was thought that by selecting a small group of employees the number of variables which inevitably creep into a large group situation could be somewhat diminished. Such influences as the amount of work ahead of the operators, changes in type of work, the introduction of inexperienced operators, the shifting of personnel because of fluctuation in work schedules could be largely eliminated. In a small group it would be possible to keep certain variables roughly constant; experimental conditions could be imposed with less chance of having them disrupted by departmental routines. It would also be easier to observe and record the changes which took place both with- out and within the individual. And lastly, in a small group there was the possibility of establishing a feeling of mutual confidence between investigations and operators, so that the reactions of the operators would not be distorted by general mistrust.
The Operators Chosen for the Test Room
Certain problems arose in selecting the girls for the test. First, in order to avoid the influence of the element of 'learning" upon the results, only operators who were thoroughly experienced in relay assembly work were selected. Secondly, it was desirable that the girls selected should be willing and cooperative, in order that their reactions to the changing conditions of the test would be normal and genuine. The illumination experiments had demonstrated that an employee's response to an experimental change could not be ascribed solely to a simple physiological reaction. Some girls were on the defensive or suspicious and held back their output, while others, overly anxious to cooperate, increased their output by 'spurting" when illumination was increased. In planning this test, the investigators wished to secure a kind of relationship with the participants which would insure their working at a natural pace and "as they felt."
The method adopted for selecting such a group was to invite two experienced operators who were known to be friendly with each other to participate in the test and ask them to choose the remaining members of the group. The group thus selected consisted of six girls: five to do the actual assembly operation, which has already been explained, and the sixth to act as layout operator. The latter's duties were of a minor supervisory character and consisted of assigning work and procuring parts for each assembler. This arrangement of having a layout operator serve the assemblers was identical with that in the regular relay assembly department, with the exception that quite frequently in the regular department one layout operator served six or seven girls instead of five in the test room.
Organization of the Test into Periods
At the beginning of the inquiry, the investigators had certain specific questions to which they hoped to find answers. All the factors which influence the reactions of the worker could not be studied at the same time; certain factors had to be studied before others. The original inquiry started with six questions, all of which were related, more or less to the problem of fatigue.
1. Do employees actually get tired out?
2. Are rest periods desirable?
3. Is a shorter working day desirable?
4. What are the attitudes of employees toward their work and toward the company?
5. What is the effect of changing the type of working environment?
6. Why does production fall off in the afternoon?
These questions were originally chosen because they were typical of the questions being asked at the time. Rest pauses and fatigue were controversial topics in industrial circles. Furthermore, the illumination experiments had cast doubt on the method by means of which most conclusions on such topics had been reached.
The test was organized into periods, each period representing the number of weeks during which a specific condition of work was in force. The exact nature of the experimental conditions for all periods was not determined at the beginning of the experiment. It was thought best to plan only one step at a time and to let the results obtained in one period determine the conditions of the next period. The original questions with which the investigators started dictated the first set of experimental conditions imposed.
For convenience, the schedule of test conditions for the first thirteen periods is presented in Table III. The periods were numbered consecutively. Periods I - III constituted an introductory phase, the purpose of which was preparation for experimentation. During Period I the operators were still in the regular department. Period II, which covered the first few weeks the operators were in the test room, was planned to permit the girls to become familiar with their new surroundings. In period III a change in wage payment was introduced, a necessary step before the experiment proper could begin. Periods IV - VII were concerned entirely with rest periods and constituted the second phase of the test. In Periods VIII - XIII the investigators experimented with a shorter working day and week.
TABLE III - Schedule of Test Periods - Relay Assembly Test Room Period Number Special Feature Dates Included Duration
(Weeks)Times of Rest Pauses
(see note 1 below)A.M.
P.M.
I In regular department (see note 2 below) 5-25-27 to 5-10-27
Approx. 2
None
II Introduction to test room 5-10-27 to 6-11-27
5
None
III Special group rate (see note 3 below) 6-13-27 to 8-6-27
8
None
IV Two 5-min. rests 8-8-27 to 9-10-27
5
10:00
2:00
V Two 10-min. rests 9-12-27 to 10-8-27
4
10:00
2:00
VI Six 5-min. rests 10-10-27 to 11-5-27
4
8:45, 10:00, 11:20
2:00, 3:15, 4:30
VII 15-min. AM "snack" and 10-min. PM rest (see note 4 below) 11-7-27 to 1-21-28
11
9:30
2:30
VIII Same as VII but 4:30 stop 1-23-28 to 3-10-28
7
9:30
2:30
IX Same as VII but 4:00 stop 3-12-28 to 4-7-28
4
9:30
2:30
X Same as VII 4-9-28 to 6-30-28
12
9:30
2:30
XI Same as VII but Sat. AM off 7-2-28 to 9-1-28
9
9:30
2:30
XII Same as III (no "snack" or rests) 9-3-28 to 11-24-28
12
None XIII Same as VII but operators furnish own snack, company furnishes beverage 11-26-28 to 6-29-29
31
9:30
2:30
- "In the test room the operators normally worked a 48 hour week consisting of five 8 and 3/4 hour days, from Monday to Friday inclusive, and one half day on Saturday. The hours of work were from 7:30 to 12:00 and from 12:45 to 5:00. On Saturdays they worked from 7:30 to 12:00. ...On weekdays, they received time and a half (pay) for the 45 minutes they worked in excess of eight hours." (Roethlisberger & Dickson, 1941, p.33). Periods IV through XIII deviated from this schedule as indicated. Employees were served a company-provided lunch during the 12:00-12:45 time slot.
- During this period the five test subjects were observed and their output recorded while they still working in the main relay assembly department. This was a factory-like work area, with close to one hundred relay assemblers working in one large room. During subsequent periods (II - XIII) the 5 subjects were isolated in their own "private" room.
- The Hawthorne employees were generally paid under a complicated system comprised of an experienced-based grade rate, a straight piecework incentive rate and/or group piecework incentive rate. When the test subjects were moved into the test room (reducing their group size from 100 to 5), this required that the group piecework rate be adjusted. Thus, the "special" pay rate was NOT a pay increase. In fact the change in the pay scale was designed to make certain that the operators would not make more than they would have in the normal workroom.
- This was a company provided snack, consisting of sandwiches (pb & j, egg salad or tongue!) or soups, and some fruit, generally.