Speed and Accuracy of the Feebleminded on Performance Tests
The Psychological Clinic Copyright, 1931, by Lightner Witmer, Editor Vol. XIX, No. 9 February, 1931 :Author: W. E. McClure, Ph.D.
Director of The Juvenile Adjustment Agency, Toledo, Ohio. If mental age were the only factor which influences one’s scores m speed and accuracy on performance tests, the inference could be drawn that the average performance of feebleminded adults with a given mental age 011 a group of performance tests should be the same 111 speed and accuracy as the performance of normal boys who have the same mental age. In this study the performances of a group ?f feebleminded subjects, highly selected on the basis of mental age, were compared with the performances of a control group composed of an equal number of normal boys with the same mental age. A group of standard performance tests (see Table II) was given to 41 feebleminded male subjects who ranged in chronological ages from 16 to 50 years, and in mental ages from 8 years 6 months to 9 years 6 months. Their mean chronological age was 20.7 years and their median was 19.0 years. (See Table I). The same tests were given to a group of 41 normal 9-year-old boys. They were selected on the basis of their mental and chronological ages. They ranged chronologically from 8 years 10 months to 9 years 2 months, and mentally from 8 years 6 months to 9 years 6 months.
The Stanford-Binet Test was used to determine their general intelligence. The feebleminded subjects had an average I.Q. of 56, while the average for the normal boys was 100.
The feebleminded subjects were selected from among the inmates of two institutions for the feebleminded, The Ohio State Institution and The Oklahoma State Institution. The normal boys Were selected from the public schools of two cities in Oklahoma, Norman and Oklahoma City.
The tests employed in this study are well known and need no explanation. Eleven of them are contained in the Pintner-Paterson
Table I The chronological ages of 41 feebleminded subjects Age Number 16 7 17 6 18 7 19 4 20 2 21 1 22 5 23 3 24 2 26 1 28 1 42 1 50 1
Total 41 Scale of Performance Tests1 and were administered and scored according to the directions given by Pintner and Paterson. The only other test was the Whipple-Healy Tapping Test. It was administered and scored according to Hewes.2 All of the tests excepting the Seguin Form Board, the Knox Cube Test and the Feature Profile Test were scored on the basis of both speed and accuracy of performance. The Seguin Form Board and the Feature Profile Test were scored on the basis of speed, and the Knox Cube Test was scored on the basis of accuracy.
The Computation of Comparable Scores The results of the various tests could not be compared with each other until the raw scores had been converted into comparable scores. The formula for computing Kelley’s Z-scores3 was used, which is as follows:
X-M Z = Sigma In this formula “Z” is the standard score, “X” the raw score, “M” the arithmetic mean of the total number of scores on each 1 Pintner, E. and Paterson, D. G., A Scale of Performance Tests. D.
Appleton and Company.
‘Hewes, A., Standardization of the Whipple-Healy Tapping Test. JAppl. Psychol., 1922, 6, 113-119. * Kelley, T. L., Statistical Method, p. 115, Macmillan. test made by both the normal and feebleminded subjects, and ‘Sigma” is the standard deviation from the mean of the total dumber of scores on each test.
Kelley makes the following statement regarding the Z-scores: “Such measures as these may be called ‘standard scores’ as they are measures of deviation expressed in terms of standard deviation.” Table II presents the mean raw score, the P.E. of the mean raw score and the mean Z-score made by each of the two groups of subjects on the tests measuring speed and accuracy of performance. The raw scores made on the tests marked “Time” represent the number of seconds required to perform the tests.
Table II This table shows the mean raw score, the P.E. of the mean raw score and the mean Z-score made on each test by the normals and the feebleminded. Mare and Foal-Time Mare and Foal-Errors Seguin Form Bd.-Time 5-Figure Form Bd.-Time 5-Figure Form Bd.-Errors 2-Figure Form Bd.-Time 2-Figure Form Bd.-Moves Casuist Form Bd.-Time Casuist Form Bd.-Errors Manikin-Time Manikin-Score Feature Profile-Time Ship Test-Time Ship Test-Score Picture Comp.-Time Picture Comp-Score Knox Cube-Score Substitution-Time Substitution-Errors “Tapping-Speed *Tapping-Errors Normals Mean Raw Score 40.12 2.34 24.48 103.67 7.93 103.68 21.71 105.62 7.93 41.01 4.51 220.26 106.34 13.86 286.94 386.34 6.59 140.22 .74 46.54 1.15 P.E. Mean Raw Score ?2.29 ? .17 ? .40 ?6.53 ? .89 ?10.04 ?1.53 ?6.91 ? .77 ?1.98 ? .08 ?10.86 ?6.18 ? .53 ?11.29 ?16.29 ? .17 ?2.87 ? .10 ? .90 ? .14 Mean ZScore - .073 - .119 - .573 -.039 + .206 -.297 -.270 + .123 + .266 -.129 + .234 -.153 - .255 ? .105 -.162 + .083 + .266 + .176 + .196 - .664 - .040 Feebleminded Mean Raw Score 37.19 2.02 17.28 96.34 11.44 53.90 14.76 126.12 13.48 35.36 4.05 186.60 68.80 12.74 247.46 368.78 5.66 160.44 2.45 60.44 1.07 P.E. Mean Raw Score ?2.26 ? .23 ? .52 ?6.67 ?1.10 ?6.38 ?1.01 ?10.06 ?1.38 ?2.73 ? .13 ?10.59 ?4.21 ? .63 ?13.84 ?15.62 ? .26 ?7.44 ? .59 ?1.78 ? .12 Mean ZScore + .072 + .119 + .760 + .045 -.204 + .297 + .269 - .123 -.238 + .129 - .189 + .177 + .255 - .097 + .162 - .023 -.266 -.380 -.209 + .310 + .019 * Whipple-Healy Tapping Test. Included in this group of tests is the Whipple-Healy Tapping Test which is marked “Speed.” The raw score made on this test represents the number of “taps” made in 30 seconds. The tests in Table II measuring accuracy (most of them are listed twice because they measured both speed and accuracy) are indicated by “Errors,” “Moves,” or “Scores.”
Speed of Performance: Normal Versus Feebleminded Subjects In Figure 1, as well as in all the following figures, the tests are represented on the horizontal axis and the mean Z-score for
Figure 1 Showing the difference between the normal and feebleminded subjects in their performance on tests measuring speed. .70 .60 Normal 1 .40 / .50 S .30 o .20 r w .10 r-K ‘ ^ qq MeazryQ*- frotgl of 41 Nor, ana 41 P.M. aJ -.10 C -.20 -2 -.30 W -.40 -.50 -.60 ~?70 1 ‘ i l I I i I i q ?????????(D’tJ SSSSSSSSSS? “H H ? *H ?H -H t-I *H *H *H ? fHE-tEHEHEHEHEHEH EhEh ft ? ??l III l I I CO ^’d’dH-Poci+J’d’d i 03^^aScQ<How?H^b0 ? a5 aJ O ? ?H ?H ? cd c3 C EH55^EHCH-pt-’ O O vM CQCQ ? CD OQ CQ ft p, O SSC-hOiDh-h SS aJ ?POO -H (D O OQ O O _ 3 ? ? (L, O >T P U & 2 r-?HP ? cd S P ? ? d aJ ?PrafjS ctf u f-i *H ? w -j ?3 3 3 W ,Q P fcO Pm P bO to I 3 ro ?H o -H ? ? ?CQ ttf (l, *rH (z< CO M o P ft ? Oft 5 g 3 pt< S each test is represented on the vertical axis. The performance of the normals is represented by the unbroken line while the performance of the feebleminded subjects is represented by the broken line. The straight line extending horizontally from ‘’ .00 ” represents the mean performance by the total of 82 normal and feebleminded subjects. Figure 1 shows that the feebleminded subjects are superior to Figure 2
Showing the difference between the normal and feebleminded subjects in their performance on tests measuring accuracy.
the normal boys on 9 of the 11 tests measuring speed of performance. The mean Z-score of these 11 tests (the mean of the means) for the feebleminded is .17, while for the normals it is ?.20. Thus, the mean difference between the normals and the feebleminded in speed of performance is .37 of one standard deviation.
Accuracy of Performance: Normal Versus Feebleminded Subjects Figure 2 shows that the normals are superior to the feebleminded on 7 of the 10 tests measuring accuracy of performance. The mean Z-score of these 10 tests (the mean of the means) for the normals is .09, while for the feebleminded it is ?.07 Thus, the mean difference between the normals and the feebleminded in accuracy of performance is .16 of one standard deviation. The Effect of Age on Speed and Accuracy of Performance To determine roughly the effect of age on speed and accuracy of performance, the mean scores of the 20 younger feebleminded
Table III This table shows the mean Z-score of the^groupof 20 youngerfeebleminded subjects and of the group of 21 older feebleminded subjects on each test. Younger Group Mean Z-score Older Group Mean Z-score Mare and Foal-Time Mare and Foal-Errors Seguin Form. Bd.-Time 5-Figure Form Bd.-Time 5-Figure Form Bd.-Errors 2-Figure Form Bd-Time 2-Figure Form Bd.-Moves Casuist Form Bd-Time Casuist Form Bd.-Errors Manikin-Time Manikin-Score Feature Profile-Time Ship Test-Time Ship Test-Score Picture Completion-Time Picture Completion-Score Knox Cube-Score Substitution-Time Substitution-Errors Whipple-Healy Tapping-Speed Whipple-Healy Tapping-Errors + .266 + .401 + .833 + .413 + .144 + .457 + .418 + .184 -.013 + .306 - .095 + .355 + .189 -.140 + .162 + .254 -.127 -.021 -.292 + .489 -.002 - .230 - .500 + .348 - .182 - .371 + .144 + .144 -.430 - .610 - .042 -.277 - .001 + .316 - .054 + .155 - .416 -.371 -.318 -.139 + .139 -.037
subjects were compared with the mean scores of 21 older ones. The 20 younger ones ranged in age from 16 to 18 years, the mean being 17.0 years. The 21 older ones ranged in age from 19 to 50 years, the mean being 24.3 years. Table III shows the mean Z-scores of these two groups of feebleminded subjects.
The results which are presented graphically in Figures 3 and 4 show the younger group superior to the older group in both speed Figure 3 Showing the difference between the younger and older feebleminded subjects in speed of performance. Younger F.M. group Older F.M. group ? and accuracy. On the speed tests the younger group was superior on all but one out of the 11, while on the accuracy tests they were superior on 8 of the 10. The mean Z-score of the means of the 11 tests measuring speed Figure 4 Showing the difference between the younger and older feebleminded subjects in accuracy of performance. Younger P.M. group Older P.M. group Wean of total^f 41 Norf and 41 F. M A > / * / / ‘ ? / t . j s..’ J ‘ “**?’ V’’ / / V I I I I I I I ?? I ‘ caooQtamoQjo a> k J-i ^ p-i ?-4 S-< ?-< fn 000000000 utnuuutntntn u w i w w w I l i w 1 $3 ? I 1+3+3+3 | rH 0?d tjO’d tQ aQ tQ +3 a5?HfcCka)a>(Dco O +3 aj <rl aj Eh H Eh (D fo a> o ft o eh ?H CQ ft (Q <D ft ?3 ft 03 -H ,Q -H ? f3SSEHSi*3,dO a o u Jh-how-h Q O >> O g 43 <D PC rH fo 05 X 2 U <D aj so 43 CJ <D 0 4* ?} ?H s 3 W O td 43 pj | iH C9 O & <D 0 rO “H ?H rH Q> pJ A* 1*4 ft aJ CQ ft o ? > S3 fa is .33 for the younger group and ?.09 for the older group. The difference between the two groups is .42 of one standard deviation. The mean Z-score of the means of the 10 tests measuring accuracyis .06 for the younger group and ?.27 for the older group, making a difference of .33 of one standard deviation.
Improvement in Speed of Performance on the Sequin Form Board Are normal individuals more easily motivated than the feebleminded? The results of the three trials on the Seguin Form Board might lead to such a conclusion.
The directions require that the Seguin Form Board be given three times to each subject. On the first trial the tester says, “Let us see how quickly you can put the blocks in place.” No further motivation is permitted. On each of the second and third trials, however, the subject is urged and encouraged in every way to make the best possible record.
Table IY shows the time in seconds it took the normals and feebleminded to perform the test on each of the three trials. On the first trial the feebleminded performed it in 6.40 seconds less time than the normals. Both groups improved on each of the two following trials. The average time for the feebleminded group on the third trial was 19.13 seconds, which was an improvement of 8 seconds over the first trial. The average time for the normals on the third trial was 25.72 seconds, which was an improvement of almost 8 seconds.
Table IV The time required by the normal and feebleminded groups to perform the Seguin Form Board on each of the three trials.
Normals Feebleminded Mean time Mean time in seconds in seconds First trial 33.57 27.17 Second trial 28.16 20.07 Third trial 25.72 19.13 Table V shows the per cent of improvement made by each group on the second trial over the first, on the third trial over the second, and on the third trial over the first. On the second the improvement of the normals over the first trial was 16 per cent and the improvement of the feebleminded was 26 per cent. The improvement on the third trial over the second was 9 per cent for the normals and 5 per cent for the feebleminded. The improvement on
Table V The percent of improvement on the Seguin Form Board by the normal and feebleminded groups shown on the second and third trial over the first. 2nd trial 3rd trial 3rd trial over 1st over 2nd over 1st Normals 16% 9% 23% Feebleminded 26% 5% 30% the third trial over the first was 23 per cent for the normals and 30 per cent for the feebleminded.
The data in Table V indicate that the normals were motivated to work more nearly up to their limit of speed on the first trial than the feebleminded by the words, “Let us see how quickly you can put the blocks into place.” This is shown by an improvement of only 16 per cent on the second trial over the first by the normals while the feebleminded improved 26 per cent. Furthermore, the normals improved only 23 per cent on the third trial over the first while the feebleminded improved 30 per cent. It is not certain that either group was working up to the limit of speed on the third trial although it seems reasonably safe to make this assumption. Then, if the results of the third trial do represent the limit of speed of performance by both groups the conclusion may be drawn that the stimulus which motivated the subjects on the first trial elicited greater speed from the normals than from the feebleminded because the former improved on the second and third trials over the first by a smaller per cent than did the latter.
Summary
1. The feebleminded subjects whose mean age was 20.7 years chronologically and 9 years mentally showed a superiority of .37 of one standard deviation in speed of performance on a group of standard form board tests over the normal 9-year old boys. 2. The normal boys showed a superiority of .16 one standard deviation over the feebleminded group in accuracy of performance on the same tests.
3. The group of 20 younger feebleminded subjects with a mean chronological age of 17 years showed a superiority of .42 of one standard deviation in speed of performance and .33 of one standard deviation in accuracy over the group of 21 older feebleminded subjects with a mean chronological age of 24.3 years. 4. The stimulus, “Let us see how quickly you can put the blocks in place,” for the Seguin Form Board elicited on the first trial a greater effort from the normals than from the feebleminded.
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.