By Oscar J. Johnson
University of Minnesota, Minneapolis, Minn. The following three experiments were made during the school year 1916-17 in the Department of Psychology at the University of Idaho with the purpose of finding out as accurately as could be, certain measurable effects of smoking on college students.1 It was felt that such a study would be welcomed both by scientific men and the public at large as offering some light on a problem about which there has been much agitation and wide diversity of opinion as well as considerable legislation. There is a good deal of literature on the subject, the larger part of which is pseudo-scientific or too prejudiced to be of much value. What seemed to be needed was a scientific measurement of at least some of the effects in an experiment made either in an intensive way on a few subjects or more extensively on a larger group. As it happened, we were able to use both methods with some degree of success.
Work Done by Other Investigators.
A bibliography of the most important contributions in this field, will be found at the end of this article. Of the investigations especially worthy of notice in this connection, the ones by C. S. Berry, A. D. Bush, Edwin Leavitt Clark, George L. Meylan, and Frederick J. Pack are probably the most valuable.
Dr Bush of the University of Vermont tested fifteen men with a series of ten mental tests and found an average decrease of 10.5 per cent in mental efficiency. The greatest decrease was in the field of mental imagery, 22 per cent, with perception and association coming next. He also found that the greatest loss in efficiency was caused by cigarettes and that they were the only ones in whose smoke nicotine was found. Pyridine seemed to be the principal toxic factor in all tobacco smoke.
C. S. Berry studied the effects of smoking on himself. Each day after lunch for twenty days he added a set of typewritten examples. On alternate days, before adding, he would smoke a cigar; on the other days he would spend this time in conversation or in reading light literature. He found that the effects of smoking 1 The author wishes gratefully to acknowledge his deep debt to Dr H. B. Reed of the University of Idaho whose advice, help, and active participation helped to make the study possible. Thanks are also due to the twenty one men, who with willingness gave of their time and energy, some of them during a period covering several month were more pronounced than those of practice, decreasing the time taken to add by 7.7 per cent. The effects of the smoking were evident, because each day the time taken to add after smoking was less than the time taken the next day without smoking.
Dr George L. Meylan of Columbia College made a careful investigation of the effects of smoking on 223 students. The results show that the average mark of the 115 smokers was 62 per cent and that of the 108 non-smokers was 69 per cent. His conclusions are these: “High School students who acquired the smoking habit before entering college are about eight months older at entrance than the non-smokers,” one reason being “the depressing influences of the use of tobacco on the heart and circulation and hence the retarding effect on growth.” Also, “The scholarship standing of smokers is distinctly lower than that of non-smokers. The intimate connection existing between the smoking habit and the participation in social and athletic activities, makes it impossible to determine how much, if any, direct influence the smoking habit exerts upon scholarship; but the results of this study and similar studies obtained at Clarke College indicate very clearly that the smoking habit is very closely associated with idleness and lack of ambition for scholarly achievement.”
Mr. Clarke of Clarke College found that of 211 students 43.6 per cent smoked, of which only 18.3 per cent won honors as compared with 68.5 per cent of the non-smokers. The smokers exceeded the non-smokers in strength and lung capacity.
Dr Pack of Utah, who has made a wide study of the effects of smoking based on the information received from fourteen colleges and universities, found among other things that? Highest Lowxst Marks. Ma.ru. 101 Non-smokers furnish 11 6 101 Smokers furnish 5 15 And: “Smokers furnish twice as many conditions and failures as do the non-smokers.” And further: “Only half as many smokers as non-smokers are successful in the ‘try-outs’ for football squads.” The Methods in General in this Investigation. The three experiments are radically different as to methods and number of subjects. In the first, twenty men were used, all students of the University of Idaho, except one. Ten of these did not use tobacco in any form; six were habitual smokers; and four smoked occasionally. Three different series of tests were given at four o’clock on Friday afternoons, the tests being identical in kind and as nearly equal in difficulty as possible. Before the first series, no men smoked. Before the two other series each smoker smoked 8.5 cm. of a mild cigar. In this experiment, the relative improvement, or increase in amount of performance, and in accuracy, was used as the measure of the effect of tobacco upon both the habitual and moderate smokers. It happened that the smokers whom we chanced to pick out were superior to the non-smokers, and this may have had a decided effect on their rate of improvement. In the second experiment we used the method of training four subjects in a smaller number of tests until their limit of improvement had been reached and then have them smoke to see the effects. Two faculty members and two students served as subjects and trained themselves in the tests from March until May. In this experiment all work was individual. When a subject had reached the maximum amount of efficiency in all the tests used, he was ready for the first smoking test. After going through the series of tests, he would smoke a cigar. Immediately after smoking, he would again take all the tests and repeat this at intervals of an hour for several hours. In some cases the tests were gone through again the next morning. To find the effects of smoking, we calculated the difference in performance before and after smoking for the several weeks during which the smoking series were repeated.
The third experiment is a survey of the school marks received by all the men of the “University of Idaho during the school year of 1915-16. All the men were classified by personal inquiry into four groups: non-smokers; occasional smokers; moderate smokers; and habitual smokers. The average grades received for these four were calculated and compared.
Experiment I. Tests Used. 1. The Genus-Species Test. 2. The Opposites Test 3. Test of “Logical” Memory or Reproduction of Ideas. 4. The Addition Test. 5. Accuracy, or Precision of Movement: Aiming. 6. Strength of Grip. These tests were done in the usual way except that of “logical” memory. In this one, the substance of two passages of uniform and easy prose of equal length was written down as soon as read, and the average of the two passages used at each series of tests was the subject’s score.
Results.
Accuracy and efficiency were the two things especially considered in working up the results. Efficiency was calculated by two methods: first by the absolute amount of work done; secondly by the formula given in Whipple’s “Manual of Mental and Physical Tests,” p. 313: E = eA, where “e” is the number of symbols (3 ? y examined and “A” is obtained from A = , “c” is the number c -f- o of symbols crossed, or attempted; “w” is the number of symbols wrongly crossed; and “o” is the number of symbols erroneously omitted. Owing to their nature, this method was impossible with tests three, five, and six. The second formula also gives another measure of accuracy, the results of which are given in table III. Table I.?Experiment I. Improvement of Smokers and Non-smokers. Smokers. Tests. Genus-Species Opposites “Logical” Memory. Addition Aiming Strength of Grip Average Scores. First Series. 5.0 8.5 6.8 61.3 5.9 38.6 Second Series 7.1 9.2 7.1 68.6 5.4 41.7 Third Series. 8.3 11.6 7.9 76.1 5.2 41.8 Average percentage of improvement Average percentage of improvement, leaving out the test of strength of grip Percentile Improvement, Third Series over First 66.0 36.4 16.1 24.1 11.8 8.2 27.1 30.8 Percentile Improvement, Average of Second and Third Series over First 54.0 22.3 10.2 17.9 10.1 8.0 20.4 22.9 Non-Smokers. Genus-Species Opposites “Logical” Memory. Addition Aiming Strength of Grip 5.8 7.0 6.3 53.2 5.8 40.8 6.7 9.7 6.9 64.0 4.8 42.1 8.1 10.8 7.2 69.6 4.6 41.4 Average percentage of improvement Average percentage of improvement, leaving out the test of strength of grip 39.6 54.3 14.2 30.8 20.6 1.4 26.8 31.9 27.5 45.7 11.1 25.5 18.9 2.2 21.8 25.7 136 THE PSYCHOLOGICAL CLINIC. * ?- ? * ttC-?^<l ^r,^711 tr> _ y /oo <J -tj nj / 0} f $U b o r-c/i 7-7 &/e s u * Lb <g~ic&/ Memory’ Test , t Te.$-t jTT ? 1st fes k TnTY jTt ” a.7idT J?/r jT2> <2 s, ?ty’Ut of drip T*s~L ? Jj<y?-S>riok<,Y? ir ~ ? ??? ‘i’mo TferS /0O ^ 0 ? jr e ^ | S OppoSLr^s ? > <? t> Aiming’ /*,st _ ? S?*y o Jre >-s , f Jbrofj o7i c cvv ?Jti^tiS c mfrv U’TrJ <rit) ? H>| ^ ^ 0 _ __ J.st7*& Jt-ndT ~T 1st Test 7 JdT -1st TestJvST 3d T ^ o fltddt i ^ Test Figure I.
The improvement of the non-smokers, which was considered normal, was compared with that of all the smokers and then with the habitual and occasional smokers separately. The tables giving the results of the last two have been omitted because of lack of space. It was found, however, that the average improvement in all tests from the first to the third series for the non-smokers was 7.4 per cent less than for the habitual smokers, but 8.9 per cent more than for the occasional smokers. If we leave out the strength of grip test which was found valueless in an experiment of this kind, the above values become 6.2 per cent and 10.3 per cent respectively. From table I and figure I it will be seen that the smokers improved more than the non-smokers in the genus-species test, the “logical” memory test, and the strength of grip test when efficiency was calculated by the amount of performance of the groups. But it will be seen that improvement in “logical” memory is only 1.9 per cent in their favor, which is so little as not to mean anything. As has already been said, the strength of grip test was generally so unsatisfactory that the only thing it does show is its uselessness. The nonsmokers improved more in the opposites, addition, and aiming tests. When the average improvement from the first series to the third for all tests is taken, the result for the smokers is 27.1 per cent and for the non-smokers 26.8 per cent. If the strength of grip test be left out, the improvement for the smokers is 30.8 per cent and for the non-smokers 31.9 per cent.
By referring to table II, it will be seen that the smokers made a total of 26.5 errors in the genus-species, opposites, and addition tests (the only ones in which errors could be counted) in the first series, before smoking, as against 35.0 for the non-smokers. Incidentally, this fact again illustrates the superiority of the smokers. In the third series, these same smokers made 33.0 errors to 25.0 by the non-smokers. In other words, the smokers increased by 24.5 per cent their number of errors, while the non-smokers decreased theirs by 28.5 per cent. The per cent of all errors made to the total amount of work done during the whole experiment is 4.9 per cent for the smokers and 6.2 per cent for the non-smokers.
Table III shows that the average improvement in efficiency as measured by the formula given above is practically the same as that obtained by comparing the actual amount of work done without the use of the formula. The tests in which the smokers improved most are the same here as before.
“Logical” memory is the only test in which the smokers improved more in accuracy than the non-smokers; the respective per cents of improvement are 24.6 and 16.6. In genus-species the Table II.?Experiment I. This table shows the number of errors made in three tests in the three Series. Smokers. Tests. Genus-Species. Opposites Addition Total. First Series. No Smoking. 0.0 11.5 15.0 26.5 Second Series. Smokers Smoked. 0.5 9.0 38.0 47.5 Third Series. Smokers Smoked. 1.5 9.5 22.0 33. Total No. of Errors. 2.0 30.0 75.0 105.0 Percentage of Increase of Errors in Third Test over First. ?17.4 46.0 24.5 Non-Smokers Genus-Species. Opposites Addition Totals. 0.0 18.0 17.0 35.0 1.0 15.0 11.0 27.0 1.0 10.0 14.0 25.0 2.0 43.0 42.0 97.0 -44.5 -17.7 -28.5 Smokers. Tests. Genus-Species. Opposites Addition Total No. of Units of Work Correct. 204.0 293.0 2060.0 Total No. of Errors Made. 2.0 30.0 75.0 Average percenta ;e of errors Percentage of Errors. 0.9 10.2 3.6 4.9 Non-Smokers. Genus-Species. Opposites Addition 206.0 275.0 1868.0 2.0 43.0 42.0 0.9 15.6 2.2 Average percentage of errors 6.2
smokers improved 4.5 per cent to 13.1 per cent of the non-smokers; in opposites 6.3 per cent to 35.5 per cent; and in addition 1.0 per cent to 3.2 per cent. The average improvement from the first series to the third series is 9.1 per cent for the smokers and 17.1 per cent for the non-smokers. (See table III and figure II.) It was felt that one reason why some of the smokers improved as rapidly as they did, besides the probable effect of their superior
Table III.?Experiment I. Comparison of amount of improvement of Smokers and Non-smokers in accuracy and efficiency from the first to the third series of tests, when the two following formulas were used in the calculation. Accuracy = -??. Efficiency = eA. c ?o Smokers. Testa. Genus-Species Oppositiea “Logical” Memory. Addition First Series. No Smoking. 0.87 0.79 0.61 0.95 E. 5.0 8.3 6.1 60.9 Third Series. Smokers Smoked. 0.91 0.84 0.76 0.96 E. 8.2 11.2 7.6 75.8 Average percentage of improvement. Percentage of Improvement, Third Series over First. 4.5 6.3 24.6 1.0 .1 E. 64.0 34.9 24.6 24.4 36. Non-Smokers. Genus-Species Opposites “Logical” Memory. Addition 0.84 0.62 0.60 0.94 5.8 6.3 6.0 52.9 0.95 0.84 0.70 0.97 8.0 7.0 69.3 Average percentage of improvement. 13.1 35.5 16.6 3.2 17.1 37.9 57.3 16.6 31.0 35.7 intelligence, was that they exerted themselves more than did the non-smokers. The experimenter is sure that he saw evidences of this greater effort, which in some cases at least undoubtedly was unconscious. There is another factor which may have influenced the results and which is closely connected with the one just mentioned. Unlike other experiments on the effects of drugs, it was impossible to disguise the taking of it into the body. The value of such disguise was recognized by Dodge in his experiment on the effects of ethyl alcohol, and by Hollingworth in his experiment on the effects of caffeine. If this factor would materially change the subject’s reactions, we are here facing a problem in technique which is very difficult to solve.
In general we may conclude from the results of experiment I that smoking reduces the accuracy and to some extent the efficiency of mental and motor activity. Furthermore, that there may be a great deal of difference in effects on different individuals and even upon different functions of the same person. This is born out by experiment II in which the method of experimentation lended itself particularly to the discovery of effects on different individuals or processes and the results of which are more definite than those of experiment I.
{To be concluded.) /&Qft XOfo Subordi Tndie s resit kV?>> ^ 3?r , Cppcst-Le* S&%[ i <52^ 75/ 7ej^” Test -Zstjest 3</ 7c~i? ‘Loaned J fatrporf ‘jest iC?j ?;j&/grs -7tu?-jgtfKii? ?*7< yo% /<??* 6??f 6~0?o Ad Ji tLO~n Test j.&t> Jest 3d Test -2 $t Jest -^7<Xlt Figure II. Figure II.
Disclaimer
The historical material in this project falls into one of three categories for clearances and permissions:
Material currently under copyright, made available with a Creative Commons license chosen by the publisher.
Material that is in the public domain
Material identified by the Welcome Trust as an Orphan Work, made available with a Creative Commons Attribution-NonCommercial 4.0 International License.
While we are in the process of adding metadata to the articles, please check the article at its original source for specific copyrights.