Solt Maze a
- Author:
Hehman H. Young, Ph.D.,
University of Pennsylvania.
The slot maze designed by the author employs the fundamental principles of the pencil and stylus mazes in such a manner as to set an interesting practical problem by such mechanical construction of the test that only one solution is actually possible. This is accomplished by cutting the pattern of the maze out of a panel and mounting it in such manner that the slot with all its blind alleys furnishes a free passage-way for the movement of a slider so constructed that it cannot be removed from the board except at the ends of the course. This makes the test practically “fool proof” and eliminates all necessity for negative instructions. An entire series of these tests is being devised and standardized. This article, as a preliminary statement, is confined to a report on the use of the first of the series, Slot Maze A, in testing 1304 children ranging in age from four to nine.
The accompanying diagram shows the pattern of the maze. The narrow unshaded strip between the parallel lines is the cut out part of the panel and consistutes a continuous slot from the point A to the point B. The blind alleys are equally continuous with the direct path. The slots are ^ of an inch wide and an inch apart. The entire board is nearly a foot square. The starting point is indicated on the diagram by the letter A and the end or goal by the letter B. A toy boy clown about 1| inches high is set in the slot at B as the objective for the goal. He stands with his back toward the starting point and faces the slot down which the slider will approach him. The slider carries on its top a toy shoe just large enough to be easily grasped by young children. The circles in the slots at A and above B enable the slider to be inserted or removed at these points. When in use the circular opening at the starting point A is closed so the child cannot remove the slider by returning to the starting point.
Method.
The test is placed in front of the child so that the right side of the goal boy is toward the child. This position of the test makes the slot leading from the starting point A, point directly toward the child.
The instructions to the child are: “See this boy? (Pointing to the boy). This (grasping the shoe) is his shoe. The boy wants his shoe. I want you to take it to him. See how quickly you can take the shoe to the boy.” (With these last instructions the shoe being firmly grasped by the examiner is slowly moved several inches along the slot directly toward the child.)
The above are the full set of exact instructions given the child. If they are not given verbatim, the examiner is apt to give the child a cue by unwittingly using the word around in some such statement as, “Push the shoe around to the boy.”
The time is counted in seconds from the instant the above instructions are completed until the child gets the shoe in the slot in front of and against the boy. If the child does not complete the test by the end of five minutes, his time record is marked DNC, and the test is stopped. Sometimes a child stops, thinking the test completed, when the shoe is in a slot opposite the goal boy, but not in the right one. When a child does this or gives up before the end of the five-minute time limit, he is urged to continue by repeating part or all, as the case may require, of the original instructions, as, “The boy wants his shoe. Take it to him.” Only in rare instances is it necessary to give the additional urge, “Go on, put the shoe in the boy’s hands.” If the child says he cannot do it, he is encouragingly told, “Oh, yes you can. The other boys (or girls) did it and you can too.” No other instructions, verbal or otherwise, are ever given. Encouragement, but no assistance dare be given, because the purpose of the test is to learn whether the child with the objective clearly defined can discover the route by which to deliver the shoe. At the end of each trial the examiner lifts out the slider through the opening in front of the goal boy and inserts it at starting point A, ready for the next trial.
Children who complete the test within the time limit are given a second trial immediately with instructions, “See if you can take it (pointing to the shoe) to him quicker this time.” The time is kept and recorded as on the first trial. Five minutes is the time limit here also.
Type and Distribution of Children Tested.
The children tested for this report came from four schools in different districts of the City of Philadelphia.* In one district, children came from homes considered to be above average so far as social status and type of occupation of parents is concerned. A second district is rated below average in these same respects, while the other two districts are rated as about average. In the first two schools every child up to and including 7 years and 11 months of age was tested. In one of the others every child up to and including 8 years and 11 months of age was tested, while in the fourth school only children under 6 years of age were tested.
Results.
Table I gives the ten percentile distributions for boys and girls separately. Under age 4 are reported all children according to sex who at the time of examination were 4 years old but not 5. All age groups are constructed on this same plan. In Table I immediately under each age is given the number of children tested at that age. The best score of the poorest ten per cent of scores made on the first trial in each age is set down in the table opposite the 10th percentile. In the same way the remainder of the table shows the best score for every tenth percentile. Thus the 100th percentile line shows the best score made at each age.
Table I.?First Trial Percentiles in Seconds for 642 Boys and 662 Girls. Boys. Age. Girls. No. of Cases. 39 126 168 191 118 36 128 206 200 92 Percents 100 90 80 70 60 50 40 30 20 10 0 35 89 120 180 188 DNC DNC DNC DNC DNC DNC 20 40 63 81 120 170 240 DNC DNC DNC DNC 18 35 45 60 77 101 120 190 300 DNC DNC 20 29 37 44 54 75 98 117 169 300 DNC 14 22 28 36 43 55 63 86 135 260 DNC 60 133 160 265 DNC DNC DNC DNC DNC DNC DNC 27 82 117 177 198 DNC DNC DNC DNC DNC DNC 15 58 85 120 140 225 DNC DNC DNC DNC DNC 17 36 53 78 110 146 215 295 DNC DNC DNC 21 35 45 59 77 105 131 210 275 DNC DNC
In every instance, except in the 100th percentile which is the best trial score and hence cannot be expected to be uniformly consistent, the time values for the older children are shorter than for any younger age group within the same sex. Six per cent of the eight-year-old and nine per cent of the seven-year-old boys did not complete the test within the five-minute time limit. In all other ages the number of DNC’s was greater than ten per cent.
SLOT MAZE A. 77
Figure 1 is a graphic representation of the percentile distribution of Table I. The numbers across the top indicate the ten percentile limits, and those at the side indicate the time in seconds. The number on each curve indicates the age it represents. As DNC signifies an arbitrary time limit and is not a time value, it means that Figure 1. Percentiles op First Trial Time. Full Line Boys.
Broken Line Girls. 78 THE PSYCHOLOGICAL CLINIC. the actual time for completing the test, if sufficient time had been allowed, would have been over three hundred seconds for those scored DNC. Because the direction of the time curves beyond three hundred seconds is not known they stop just over the 300second line. This figure shows more clearly than Table I how within each sex the different age groups maintain their identity, never crossing or even approximating one another except at the 100th percentile. This is a convenient chart for comparing the time record of an individual child. His percentile rating can be determined by locating the level of his score on the left of the chart and following it in on a horizontal line to his sex and age group. This point of intersection can be read into a percentile by noting its position on one or relative position between two of the vertical ten percentile lines. Table II.?Second Trial Quintiles and Medians in Seconds for the 523 Boys and the 412 Girls who Completed First Trial.
Age. Boys. Girls. No. of Cases. 19 85 136 172 111 13 62 120 143 74 Percents 100 80 60 50 40 20 0 21 37 79 90 156 182 DNC 17 26 39 48 70 131 DNC 13 22 30 36 50 91 DNC 13 20 24 26 30 51 DNC 12 16 19 21 24 40 275 30 51 62 162 205 300 DNC 21 36 58 88 125 220 DNC 12 27 36 52 72 158 DNC 14 24 30 34 43 81 DNC 14 22 28 40 47 101 DNC
Table II is constructed for the second trial in the same manner as Table I for the first trial, except that only the median and every twentieth instead of every tenth percentile are given. That there are fewer cases here is due to the fact that those who failed to complete the test on the first trial within the five-minute time limit did not get a second trial. For the boys there is a consistent decrease in time with increasing age, but for the girls this is not true. For the 20th, 40th and 50th percentiles the time for eight-year-old girls is longer than for the seven-year-old girls. The eight-year-old boys constitute the only group where every child successfully completed the second trial within the five-minute time limit. Of the eightyear-old girls only one failed. Figure 2 gives the medians for both sexes on both trials. Itshows the rapid decrease in time which accompanies increasing age, especially on the first trial. The young children do the second trial so well that on it it would be impossible to have the same rate of improvement with increasing age.
Table III is employed to show that on an average within certain limits there is a direct ratio between rate on first trial and rate on second trial. The boys and girls are treated separately. The first column shows first trial time intervals, the second column the number of boys who completed their first trial within the time interval of the first column, the third column gives the median time for these same boys on their second trial, and the fourth column gives the Figure 2. Medians op First and Second Trials. Full Line Boys.
Broken Line Girls.
Table III.?Second Trial Median Time Values as Related to Time on First Trial for the 523 Boys and the 412 Girls who Completed First Trial. First Trial, Seconds. Boys. No. of Cases. M. P. E. Girls. No. of Cases. M. P. E. 10- 19. 20- 29. 30- 39. 40- 49. 50- 59. 60- 69. 70- 79. 80- 99. 100-119. 120-139. 140-159. 160-179. 180-199. 200-249. 250-300. 53 52 72 38 35 33 45 43 26 22 25 15 26 30 13 18 22 25 30 30 33 33 50 57 33 43 79 65 81 0.5 2.5 3.5 9 8.5 22 12 18 37 26 27 47 83 46 36 3 16 32 26 26 27 18 34 41 41 24 24 28 32 40 20 21 23 25 26 35 50 44 64 89 46 69 125 55 70 ? 4.5 5.5 3 5.5 19 19 34 38 53 70 65 84 54 48
probable error of this median. The P. E. here is one-half the difference between the upper and lower quartiles. Thus the figures in the first line of the table mean that eight boys finished their first trial somewhere in the time interval between ten and nineteen seconds, and that the median time for these same boys on their second trial was thirteen seconds with a probable error of 0.5 seconds. The last three columns of the table give the same information about the girls as columns 2, 3 and 4 do for the boys. The second trial probable errors are very much smaller for those children who completed the first trial in less than one minute, than for those children consuming more than a minute on their first trial. Second trial times vary directly with first trial times up to one minute, but beyond this the second trial medians are much longer and do not show any consistent relationship to first trial times. No child who completed the first trial in less than forty-five seconds failed the second trial. Only three children of those completing first trial under one minute failed second trial. Sex Differences. Probably the consistently large sex differences were the most unexpected result obtained in this investigation. The first trial median times for the girls of every age are approximately twice as SLOT MAZE A. 81 large as those for the boys of the same age. There is considerable variation on the second trial medians, the least sex difference coming at seven years of age where the median for the girls is only 31 per cent above that for the boys. Letting 100 per cent represent the median time for boys at each age, the percentage values of the median times for girls at the various ages are as follows: Age 4 5 6 7 8 First trial.. ? 177+(?) 223 195 192 Second trial 180 183 144 131 190 Considering the number of children failing to complete the test within the five-minute time limit, the percentage of failures is as follows: 19 per cent of the boys failed on the first trial, 38 per cent of the girls failed on the first trial, and of those completing the first trial, 3.4 per cent of the boys failed on the second trial, and 6.3 per cent of the girls failed on the second trial. These differences are graphically represented on Figures 1 and 2, and by the blocks of DNC in Table I. Figure 1 also shows how nearly identical throughout are the time values for the four-year-old boys and the five-year-old girls, and also those for the six-year-old boys and the eight-year-old girls.
Discussion.
As it is the purpose of this article only to introduce the slot maze and furnish a preliminary standardization of slot maze A, no attempt is made to analyze the various mental and physical factors involved in its performance or to give an interpretation of the sex differences revealed. After testing several hundred children it was decided that slot maze A would hardly be of value as a test for children nine years of age and older. Accordingly only children under nine were taken for this investigation. The distribution tables indicate that boys at least one year older and that girls two and possibly three years older should have been included.
There are great differences in the time taken by children of the same age in performing this test and there is much overlapping in the successive ages. These variations and overlappings, however, are probably no greater than the variations and overlappings in mental make-up of the children. These differences and overlappings in a test make it distinctly useful to bring out individual differences, but seemingly do not enable the validity of the differences it reveals to be satisfactorily established by statistical treatment.*
As a test free from the classical fitting together of blocks in one form or another, the slot maze fills a gap in the array of clinical tests. Slot maze A is useful in testing children four to ten years of age. The time required for the performance of the test by children of these ages is sufficiently long to allow ample time for the examiner to make significant qualitative observations of the child’s behavior in addition to getting his reaction time. The average decrease in time of the second trial over the first is not the result of increased efficiency in the physical manipulation of the test. It is the average time required by the children to learn how to solve the problem of delivering the shoe.
Fernberger, “Statistical vs. Non-Statistical Interpretation of Test Results,” Psychological Clinic, 1922, XIV, 68.
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.