Classics in the History of Psychology

An internet resource developed by
Christopher D. Green
York University, Toronto, Ontario
ISSN 1492-3713

(Return to index)

Functional Periodicity

Leta Stetter Hollingworth (1914)
Posted July 2000

[p. 86] X



It was felt to be highly desirable that as many women as possible should be tested in order that some intelligent idea might be gained as to the range of individual differences which might exist in the matter of mental and motor ability during menstruation. Accordingly, through the great courtesy and kindness of Miss Caroline Stackpole and Professor Maurice Bigelow of the Department of Biology in Teachers College. Columbia University, an invitation was extended to the women students in that department to take part in an experiment as outlined below. Seventeen women volunteered their services. All of them were professional women, their professions including teaching, nursing, sanitary engineering, and domestic supervision. They ranged in age from 20 to 40 years. They were not selected by the experimenter on any basis except that described above, i. e., their interest in responding to an invitation to assist in executing a piece of scientific work. It is, however, true that they were a selected group of women, for their presence as professional workers and students in a great university indicates at once that the pathological, the feeble, the ignorant and the lazy had at least been eliminated from their group. Some of them knew the purpose of the experiment, some of them suspected it, and some were ignorant of it.

The aim of the extensive experiment was the same as that of the intensive experiment.

The method of the extensive experiment was very much less exact than in the case of the intensive work. The tests were given in the same order at each trial, but they were not always. given at the same hour of the day, nor by the same person.[p. 87] Instead of being given daily, the tests were made only on every third day. The experiment extended over thirty days, thus yielding ten records, and including at least one menstrual period, for every subject. For the sake of reliability, two trials were made at each sitting, and the record for any given test on any given subject for any given day is the average of two trials. At each sitting a list of questions (to be discussed later), was answered by each subject.

The scope of the intensive experiment was greater as respects number of subjects included; it was less as respects number of traits tested and period of time covered. Only three of the tests used in the intensive work were used here, i.e., (1) Steadiness, (2) Tapping, (3) Opposites.


The data from the extensive experiment will not be presented in great detail, as were the data from the intensive experiment. All the data are, however, on record at Teachers College, and may be consulted for purposes of criticism or comparison by properly qualified persons.

The method of presenting the data is as follows: Each record was reduced to a multiple of the Mean Variation of the group by first obtaining the variation of each individual at each trial from the average of the group at that trial, and then dividing these individual variations through by the Mean Variation of the group. Thus, to illustrate, the first trials for the seventeen individuals in the tapping test ran as follows:

Subject A        47.0 sec.
      "     B        68.5   "
      "     C        57.9   "
      "     D        58.7   "
      "     E         53.4   "
When these records for all seventeen subjects were averaged, the result was 60.9 seconds (the average performance of the group as a whole). When the variation of each individual from this average was found these variations ran as follows:
Subject A        +13.9 sec.
      "      B        -  1.6   "
      "      C        +  3.0  "
      "      D        +  2.2  "
      "      E         + 7.5   "
                        etc.[p. 88]

The + sign signifies better than the average of the group, and the - sign signifies poorer than the average of the group, When these variations are averaged, the Mean Variation for the group of seventeen is found to be ± 5.3.

Now dividing each individual variation through by this Mean Variation of the group, we have the record of each individual for the first trial in the tapping test reduced to a multiple of the Mean Variation of the group. The record now runs as follows, and these are the figures which appear in Tables XVIII, XIX, and XX:

Subject A     +2.60 sec.
     "      B     -1.43    "
     "      C     +0.57   "
     "      D     -0.41    "
     "      E     +1.41   "
The same procedure was carried through for the second trial, the third trial, etc., in all three of the tests, until every individual record had been reduced to a multiple of the Mean Variation of the group. This method tends to minimize the influence of individual differences in absolute ability, of practice, and of other factors that might tend to obscure the specific problem here under consideration. The method has been used by Norsworthy [1] and by Sleight,[2] and has been recommended by Thorndike.[3]

The next step in the statistical procedure was to place all the critical days in two columns, with the ordinary days running out from the critical period in both directions as might chance. It would obviously happen by chance that certain of the subjects would be menstruating just as the experiment was begun; others, in the midst of the experiment; others, at the very end of the experiment. for example, the critical period for Subject A occurred at the very end of the experiment; for Subject B it occurred in the midst of the experiment; for Subject C it occurred on the fourth trial. Thus their records are arranged as follows:

[p. 89] There are two columns of numbers in the critical period because for about one-half of the subjects two of the five days of the menstrual period chanced to be included.

When the records for the seventeen subjects had been all arranged (as may be seen in Tables XVIII, XIX, and XX) the columns were added algebraically, and the average efficiency of the group as a whole at every trial was found. Tables XVIII, XIX, and XX present the complete data as thus treated, and show the individual records, as well as the performance of the group as a whole.

Table XVIII gives the record of the steadiness test. If there were a characteristic inefficiency at critical periods for the group as a whole, we should expect that the average performance of the group at the critical period would be registered as poorer than on any or most of the other days. Inspection shows that on 7 of the 17 ordinary days the records were the same as or poorer than the record of the group at the critical period.


The tapping test was exactly the same as that described in the intensive experiment under Part I of the tapping test. (See p. 15.) The data were treated and are presented exactly as in the case of the steadiness test just described. Table XIX includes all the records made in the tapping test, each record here being presented as a multiple of the Mean Variation of the group. Inspection shows that on 11 of the 17 ordinary days the records were the same as or poorer than the record of the group at the critical period. This result indicates no influence either favorable or detrimental on speed of voluntary movement at the menstrual period.


The opposites test was exactly the same as that described under the intensive experiment. The data were treated and are presented in Table XX, in exactly the same way as in the steadiness and tapping tests just described. Inspection shows that on 6 of the 17 ordinary days the records were the same as or poorer than the record of the group at the critical period. In 4 out of the 6 cases the performance of the group on ordinary days was much poorer than at the critical period.[p. 90]


At each sitting every subject recorded answers to the following questions:

1. Did you retire at the usual hour last night?

2. Were your evening occupations of the usual sort?

3. Did you sleep as well as usual?

4. Did you rise at the usual hour?

5. Did you lunch as usual? Was your diet of the usual kind?

6. Have you had any digestive disturbance?

7. Have you taken exercise to-day? Usual amount? Unusual?

8. Have you experienced any emotional disturbance in the last 24 hours? Irritation? Depression? Excitement?
Anger? etc.

9. Have you suffered in any way during the last 24 hours?

10. Note the beginning of menstruation.

11. Note the end of menstruation.

Many of these questions have no direct bearing on the psychological problem of determining mental and motor efficiency at a given time. But the long list of questions served to divert attention from the last two questions in the case of those who did not know the exact purpose of the experiment, and all of the questions were important as bearing on measurements of blood pressure, temperature, and pulse which were taken throughout the month by Miss Stackpole.

The purpose of this study, as has been stated, was simply to determine whether there is any measurable change in efficiency in mental and motor processes during menstruation. It was no part of the original aim to go into the matter of introspection, or to discuss the affective processes. However, a brief summary of the replies obtained from these questions will not be entirely out of place here.

Of the 17 subjects, 4 reported physical disturbances on critical days, as follows:

Subject K -- "Abdominal discomfort, and slight headache." (Headache also reported by this subject on days when she was not menstruating.)

Subject I -- "Very slight abdominal pain."[p. 91]

Subject D -- "Headache." (Also reported on many other days, when not menstruating.)

Subject A -- "Headache," on the fourth day of the period. (Never reported when not menstruating.)

Of the remaining 13 subjects, 12 simply entered the word "no" under the question, "Have you suffered in any way?" on critical days, and one subject had "a cold in the head" on the critical day.

Of the 17 women only two reported emotional disturbance of any kind on or near the critical day. Thus

Subject I -- "Worry." (This subject's sister was very ill.)

Subject D -- "Depressed and uncomfortable on account of the cold."

It is to be noted that both of these subjects reported physical disturbance also on critical days. The remaining 15 subjects simply entered the words "no" or "none" under the question, "Have you experienced any emotional disturbance within the last 24 hours?" on their critical days.

None of the 17 subjects reported that her occupations had been different, except that some of them "went to the theater," or to some social affair, as was frequently reported at other times.

As for exercise, none of the 17 women reported any change at critical periods. Those who did not exercise on other days did not on critical days; those who took exercise on other days did so on critical days. Thus for example, Subject B reported "moderate walking" during menstruation, the same report being made on other days. Subject I reported, "Just came from the Gymnasium." Subject L reported, "Walked to 132nd Street and back, as usual."

Briefly. the replies to the miscellaneous questions which were submitted to the subjects fail to yield any positive result. Had the replies been written without any key to the occurrence of critical days, the investigator would have been utterly unable to determine from the replies when these critical days occurred, with the possible exception of two cases.


[1] Naomi Norsworthy, The Psychology of Mentally Deficient Children, 1906.

[2] W. G. Sleight, Memory and Formal Training, Brit. Jour. Psych., 1911.

[3] E. L. Thorndike, Educational Psychology, 1910.