95. 9391939’Ws999: 999E CED "F 1979-95 9'992‘9'9'9'A9'ED STEP TESE' “99': ‘99 ""939": Cf 99939999.” 999999555 959.9993 99913939993 999329;; 999; 999' 9:: 919.9999: . '09 999. 99.. 99I99CHE999‘9999 ST?“ 9:? 9.999%? 9‘9 $799999 .E‘9‘ 9999999929. BRQZENDE HE 1993.? ' LIBRARY. Wm .Udvuity A COMPARATIVE STUDY OF THE HARVARD STEP TEST WITH TESTS OF VARIED TIMES AND HEIGHTS By Stanley Farrel Brizendine AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Health, Physical Education, and Recreation 1967 janék4/ )Zgéfifidmuxh~'i Approved ABSTRACT A COMPARATIVE STUDY OF THE HARVARD STEP TEST WITH TESTS OF VARIED TIMES AND HEIGHTS by Stanley Farrel Brizendine Statement of the Problem The study was undertaken to determine whether or not a test could be derived having a shorter stepping time and a lower stepping height than the Harvard Step Test. The problem was, of course, to get significant results which would be as good or better than the results achieved by the Harvard Step Test. Methodology Fifteen subjects were selected from the campus at Andrews University. Each subject was either a major or minor in physical education on the initial test. They were all in average or above average condition. Each subject was given fifteen step tests, including the Harvard Step Test. Graphs and statistical calculations showed a need for study between the Harvard Step Test and the B-1 test, an 8—inch high step with a one-minute stepping time. A second group of twenty-nine subjects was more randomly picked from the same population. The Second group had a larger range than that of the first group, Stanley Farrel Brizendine ranging from very poor to very good. The second group was subjected to the 8-1 test and within a 2A—hour period the Harvard Step Test was administered. A rank order correlation was computed between the two tests. This data was then statistically treated using the Analysis of Variance. Following the application of the F test, the Scheffe test compared the means of the between and the within groups on both the Harvard Step Test and the B-1 Test. Conclusions Within the confines and limitations of this study the following principle conclusions may be drawn. 1. The B-l Step Test compares with the Harvard Step Test by rank order correlation with a high posi- tive correlation of .89. The 8-1 Step Test produces the same results as the Harvard Step Test, that is, it classifies individuals into groups of good, average, and poor in relationship to cardiovascular fitness. The results of the study reveal that the 8-1 Step Test and the Harvard Step Test both distin- guish at the .05 level of confidence between the good and poor groups, and between the average and poor groups. Neither the 3-1 Step Test or the Harvard Step Test statistically differentiated between the good and average groups. A COMPARATIVE STUDY OF THE HARVARD STEP TEST WITH TESTS OF VARIED TIMES AND HEIGHTS By Stanley Farrel Brizendine A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Health, Physical Education, and Recreation 1967 DEDICATION To my parents, Harvey and Eloise Brizendine, for without them none of this would be possible. 11 ACKNOWLEDGMENTS The author wishes to express his deep appreciation to the many pe0ple who have contributed to the success of this study. To Dr. William Heusner, his advisor, the author is deeply indebted for valuable suggestions and statistical assistance. To the students of Andrews University, who willingly participated as subjects in the study, for without them the task would have been impossible. To Mr. Robert Fadeley, whose assistance with statisti— cal calculations contributed immensely. To my wife, Barbara, for her patience and kind under- standing; and lastly, to three boys, Stanley, Steven, and Sidney my three growing inspirations. iii TABLE OF CONTENTS Page DEDICATION I I I I I I I I I I I I I I 11 ACKNOWLEDGMENTS . . . . . . . . . . . . iii LIST OF TABIJES 0 I I I I O I I I I I I V LIST OF FIGURES . . . . . . . . . . . . vi Chapter I. INTRODUCTION . . . . . . . . . . 1 Purpose of the Study . . . . . . . 2 Need for the Study . . . . . . . 3 Limitations of the Study . . . . . . 3 Definitions of Terms . . . . . . . . A II 0 REVIEW OF LITERATURE I I I I I I I I 6 Brief History of Cardiovascular Studies . 6 PUlSe-Ratio T8813 o o o. o o o o o o 7 Harvard Step Test . . . . . . . . . 9 III. METHOD OF PROCEDURE . . . . . . l2 Selection of Subjects . . . . . . 12 Measurement Procedure . . . . . . . l2 IV. RESULTS AND ANALYSIS . . . . . . 19 Presentation of Data . . . . . . . 19 Analysis of Data . . . . . . . . . 23 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . 30 Summary . . . . . . . . . . . 30 Conclusions . . . . . . . . . . . 31 Recommendations . . . . . . . . . 32 BIBLIOGRAPHY 0 0 I I I I I I I I I I O 33 iv Table LIST oF TABLES Page Mean Recovery Rates, Experimental Group 1 . . 20 Rank Order Correlations . . . . . . . . 2A Mean Recovery Rates, Harvard and B-1 Tests . 2A Comparison of Means, Pilot Study with Initial Test . . . . . . . . . . . . . . 25 Analysis of Variance . . . . . . . . . 28 Scheffe Multiple Comparison of Means . . . 29 LIST OF FIGURES Figure Page 1. Harvard Step and Experimental Steps . . . 15 2. Four Experimental Test Means . . . . . 22 26 3. Mean Scores, Harvard and B-1 Tests . . . vi CHAPTER I INTRODUCTION In 1943, Lucien Brouha and associates in the Harvard Fatigue Laboratories developed the Harvard Step Test. The purpose of this test was to measure the ability of the body to adapt itself to hard work and to recover from the same. Another useful measure the test gives is a means of classifying young men into various levels of fitness. The test classifies young men into three groups: least fit, fit, and most fit. Another way of expressing the three categories could be, poor, average, and good. This, one must understand, is a general classification of cardio- vascular—circulatory efficiency. It is not a specific rating which will allow one to pin-point exacting problems in the cardiovascular system. If no medical examination is to be given the student, the Harvard Step Test will indicate a basic level of fitness so the teacher can there— by determine the level of conditioning to be administered. The Harvard Step Test is, in some respects, an easy test to administer; with the exceptions of time involved, and the height of the step for shorter individuals. Purpose of the Study For a reasonably fit individual it will take six and one-half minutes to complete the test, that is, if he completed the total test of five minutes of stepping plus the one and one-half minutes of pulse recovery time. This is quite time consuming if a large group of people are to be tested. The second problem is that of height. A short per- son has a considerable amount of discomfort with the eighteen—inch height of the stepping bench. Even though the literature indicates the effect of leg length on the Harvard Step Test has no significant effect upon heart rate, it is found, through administration of the test, that shorter subjects have extreme difficulty in keeping cadence. This fact remains true not only for the unfit subjects but for the fit subjects as well. In reviewing the literature little information was found regarding the study of stepping at lower levels than the modified eighteen-inch height. The author set out to study this problem by comparing pulse rates in recovery time on the Harvard Step Test with recovery rates on four- teen experimental step levels lower than eighteen inches, and combinations of various levels. In connection with the various heights, the time of stepping at each level will be studied with a cadence that is constant. Need for the Study The problem as stated previously should be studied because of a great waste of time and effort if there is no need to step for the full five minute period, or at the unbearable height of eighteen inches for the shorter individual. If, for example, a student could step for one minute at an eight-inch level and get the same results as the Harvard Step Test gives, then why waste the extra time which could be used for class instruction? Another need for the study is to determine whether the height of eighteen inches is necessary. If eight inches, for example, proves just as successful then administration, as far as equipment is concerned, is less costly. As previously stated very little information is available concerning stepping at the lower levels. For this reason the author is interested in studying this problem further. Limitations of the Study For the initial study, fifteen college age males from Andrews University, Berrien Springs, Michigan, were chosen. Their ages ranged from eighteen to twenty-three years of age. All subjects used were physical education majors or minors, varying from average to excellent overall condition. No training period preceded the time of testing, and each subject was instructed to not alter his daily activity pattern. The subjects led as normal a college life as could be expected during the testing period. For the second group of subjects in the final study, thirty male students of the same age group were taken from the college population. The second group was more of a random sample of the normal population with their physical condition ranging from poor to excellent. Definition of Terms In the following study, certain terms will be used throughout. These terms will be clarified now so harmony with terminology will be carried out. The Harvard Step Test is the test used in this study as the criterion for comparison. The Harvard Step Test was devised by Lucien Brouha and associates in the Harvard Fatigue Laboratories during World War II. It is a step test which uses an eighteen-inch bench upon which the sub- jects step to a cadence of thirty steps per minute for a total time of five minutes. Pulse rate is recorded for the final one to one and one—half minutes of recovery. No previous pulse rate is needed. This test will be referred to, throughout the remainder of the study, as the step test. The Pulse-Ratio Test is a test devised by W. W. Tuttle prior to Brouhas work. With the Pulse-Ratio Test there is a ratio between the resting pulse rate before exercise and the pulse rate for two minutes after exercise. The standard pulse rate recovery measurement used in recording the Harvard Step Test score will be used as standard measurement procedure on all experimental tests for the study. This score is computed by counting the total post exercise heart beats in a thirty-second period, starting at one minute and ending at one and one-half minutes. The subjects used in the initial study were physical education majors and minors at Andrews University. In_the final study, students from various disciplines were used from the same population. CHAPTER II REVIEW OF LITERATURE Brief History of Cardiovascular Studies Researchers have spent years in developing a single test that will measure physical fitness. Mosso experi- mented with the effect of exercising a muscle on an ergo- meter as early as 18A” (8). In doing so, he hypothesized that muscular efficiency was primarily dependent upon circulatory factors. Mossos' work developed interest for further study by researchers. Following Mossos' work experiments were conducted to show that cardiovascular factors are related to good physical condition. In 1921, Hambly and Hunt experimented with the pulse rate following exercises of running and walking up and down stairs (6). This method was given up in 1922 in favor of a more standardized form of exercise. To stand- ardize the exercise a thirteen-inch bench was used. The rate of exercise was determined by stepping to cadences varying from six to thirty-six steps per minute. Hambly, Pembrey, and Warner in 1925 used the standard- ized thirteen-inch step with two cadences, eighteen and twenty-four steps per minute (7). The use of these tests were to compare various tests for assessing physical fit- ness. Their main objective was to obtain a pulse ratio of less than 2.5 and another ratio slightly higher than 2.5. The only conclusive statement recorded is the follow- ing: "For many reasons the pulse affords the best test of efficiency." Campbell did a study in 1925 using the techniques of Hambly and Hunt. He used a thirteen-inch step, and a cadence of twenty-eight steps per minute. For his sub- jects he used eighty medical students. Their condition ranged from good to poor. Campbell concluded that, "His study showed a greater difference between the unfit and the average than between the average men and the athletes." (3) Pulse-Ratio Test One of the first to refine the original work by Hambly,.Hunt, and McClintock was W. W. Tuttle (9). In the laboratory at the State University of Iowa, Tuttle and his associates made their primary interest the application of the pulse—ratio that directly related to specific test problems in physical education. Of course application of this test is not only limited to physical education, but could be administered by any group of individuals where there is need of comparing physical efficiency. The pulse— ratio, introduced by Tuttle, is a ratio of the resting pulse to the rate after a certain time in recovery. This is obtained by simply dividing the total pulse for two minutes, after a known amount of exercise, by the normal resting pulse for one minute. Tuttle found many pitfalls and chances for variation which could destroy the usefulness of the test. The first thing Tuttle had to establish was a standard rate of work, or standard exercise. Hambly and his staff tried many types of exercises such as walking, stair climbing, and running. Hambly's investigation, along with that of Tuttle's, led to the adoption of stool climbing as a standard exercise (9). The stool used was thirteen-inches high and so constructed that it could be mounted and dis- mounted easily. Tuttle's work displayed the use of varying the rate of stepping and also the use of an increase in the number of steps at a uniform rate. The standard pattern of stepping would be to use a four count step pattern. At cadence l, the left foot is placed on the top of the stool; at the count of 2, the right foot is placed on the top of the stool. The subject is now standing wholly upon the stool. At the count of 3, the left foot is placed on the floor, and at the count of A, the right foot is placed on the floor. This completes one full cycle of stepping; the subject is now ready to start the second repetition (9). The correct cadence is counted by a metronone which can be calibrated for use at various cadences per minute. The pulse-ratio test has been applied to some sports performances, horizontal bar (10), and swimming (1), for example. "It is generally agreed that of all the measurements used for testing physical efficiency the pulse rate during an exhausting exercise seems to be the most reliable" (A). In this case the use of the pulse-ratio index would be antiquated because it has been determined that the resting heart rate has too many variables. The reliability of the resting heart rate is so low it pulls the pulse-ratio index too far off scale. According to some researchers, oxygen intake during exercise is probably the most valid measure of cardio- vascular condition and efficiency. This, of course, takes costly equipment to administer. Many physical education departments cannot afford such costly equipment, so must rely on some type of test which is easy to administer and still is within the limit of their budget. Harvard Step Test Of all the tests available, the most promising is the Harvard Step Test (2). It is one of the easiest tests to administer, and in fact, the least expensive. Tests re- quiring the use of a treadmill or oxygen analyzing techni- ques would be completely out of the question because of the cost to small schools. This leaves us, primarily, with 10 the Harvard Step Test as the most economical means of interpreting general cardiovascular fitness ratings. The Harvard Step Test was developed by Lucien Brouha in an effort to assess physical fitness. In an address delivered at Springfield College, December 3, 1992, Brouha revealed the following results of a study of Harvard stu- dents (2): l. The test picks out the best, the average, and the worst men in a group of unknown men. 2. Comparison of-the scores in June with the scores in September indicate the effect of training. 3. Some students deteriorated during that period as indicated by a lower-second-score than that of the first. These men were excused from regular exercise on account of illness or minor injuries. A. It-is seen that the progress made by the low group is more marked than that made by the good group. This indicates that the training program was, adequate for the unfit, but not hard enough for many of the fit-men. Scores on the Harvard Step Test ranged from 15 to l20 with 50 or below being poor, 51 to 80 average, and 81 and above being good. The average score of 2200 students, irrespective of training, was 75. Varsity athletes scored above.90 when in full condition. Highest average physical fitness is achieved when men-are placed in groups of equal fitness and training. In this study. Lucien Brouha stated, 11 It is worth mentioning that only the heart rate dur— ing the recovery period following exercise need to be determined. The initial heart rate before exer- cise is not important for the purpose of the test. Studies reveal that the heart rate before exercise is not usually related to an individuals fitness (2). CHAPTER III METHOD OF PROCEDURE Selection of Subjects Subjects for the study were physical education majors and minors recruited on a volunteer basis. None of the subjects were forced or paid to participate in the study, in fact all who participated did so very willingly. Fifteen college males from Andrews University were selected. The age level of the subjects ranged from eighteen to twenty-three years of age. All subjects used in the initial test were, from visual evidence and from previous knowledge of activity levels, in average to excellent physical con- dition. Measurement Procedure Since the Harvard Step Test was used for comparison throughout the study, the techniques of measurement were standardized. To get reliable pulse rates a heart recording machine was used with attachment to the body by the use of surface electrodes on the skin. crass electrodes were placed at three areas of the body. Two electrodes were placed on the left chest. One, one and one—half inches above the left nipple and one—inch toward the sternum. l2 13 The other electrode was placed one and one—half inches directly below the nipple. The third electrode, the ground, was placed on the small of the back, one-inch to the left of the Spine. The electrodes were placed on the subject prior to exercise so heart rate could be checked and re- corded immediately after exercise. A Sony 350 tape re— corder was plugged into the heart recorder so pulse rate could be recorded immediately after exercise for reliable heart measurement, rather than relying on the palpation method of recording. Pulse rates were recorded every fif— teen seconds in recovery for a total expired time of three and one-half minutes in length. At_fifteen-second inter- vals, throughout the total tape, one—eighth-inch splicing tape markers were placed on the tape. Through the use of this technique the heart rate was counted for each fifteen— second interval. As soon as a test was completed the pulse rate was recorded on a master sheet for further analysis. To help the subject keep cadence, an electric metro— nome was used. The metronome was set at 120 beats per minute, which gave a perfect rhythmical count of 30 steps per minute. The metronone had an electric bulb on the front which flashed at each beat. This helped the subjects with poor rhythm keep better time as they stepped. The elapsed time of stepping was kept on a twelve-inch dial timer that also had a thirty—second sweep hand. The sub— ject, as well as those administering the test, could see the time expired on each test. IA The standard work criterion was the Harvard Step Test. A stool eighteen-inches high was used so all sub- jects stepped on a standard bench. A second apparatus for stepping was constructed (see Figure 1). For consistency and correct application each step level was assigned a letter. The four-inch level was A, the eight—inch level B, the twelve—inch level C, and the sixteen—inch level D. Throughout the study this lettering procedure was used as well as combinations of these letters. For example, ABCD-l, indicates four levels; four—inch, eight-inch, twelve-inch, and sixteen-inch steps, while the number 1 incidates the total time in minutes of stepping at each level. A less complex example would be, B-l. This symbol indicates the eight-inch level of height with one minute of stepping time. A schedule sheet was drawn up with the tests for each subject assigned in random order. The subject never knew which step test he would take next, but was given instructions as to the testing procedure prior to each individual test. Before the subject was tested he sat in a chair for three to five minutes and his resting pulse rate was taken and recorded. This was done primarily to pretest the equipment, although it was felt that the data might prove of some value later. A total of fifteen tests were given to each of the fifteen subjects to determine if any level or combination of levels would yield a step test having a high relationship l5 :3 . wfim m 99 H on: doom Hassosanoqu can ampm onm>p m cm . =m =NH zma Qm<>m¢m =mH 16 with the Harvard Step Test. After completion of the 225 tests, evidence of a good relationship was obtained between the Harvard Step Test and each of the fourteen experimental tests. The author proceeded to gather further data by using thirty additional subjects. In picking the second group of subjects a cross section of poor, average, and good cardiovascular fitness levels was obtained. This was done by subjective knowledge of the subjects' activity patterns and by visual evaluation. Subjects in the good cardio- vascular efficiency group consisted of physical education majors and non majors who were on a consistent training pro- gram of daily activity. They also participated in every intra-mural sport offered at Andrews University. The average subjects consisted of students that participated sparingly in activity, possibily obtaining their exercise from walk- ing and participating in light activities. The poor sub- jects were chosen because of their complete lack of activity. In interviewing each of the poor subjects they admitted that they had not participated in any type of physical activity for the past two years. In all cases this group was over— weight. Each of the thirty subjects in the second testing group was given only the Harvard Step Test and the 8—1 Test. The 8—1 Test was determined as the best experimental test to use because of: (a) its ease of administration (an l7 eight-inch high step-—approximately one standard stair tread); (b) its conservative time limit (one minute); and (c) the high relationship which was observed between the recovery curves obtained from that test and those obtained from the Harvard Step Test and each of the thirteen other experimental tests. The same method of procedure as was used on the first testing group was duplicated on group two, with the exception that only two minutes of pulse recovery time was recorded. The highest rank order correlation was obtained be- tween the Harvard Step Test score and the B-1 Test score when the recovery pulse was counted over the 0 to 45 seconds range. For this reason it was felt there was no need of recording the extra one and one-half minutes of recovery. The B-l Test was administered first and within twenty-four hours the Harvard Step Test was given. The author attempted to give both tests within a two-three hour interval of each other on corresponding days. This was attempted, due to the fact that heart rate will vary during the day for different individuals. Resting pulse prior to the administration of the Harvard Step Test was observed and the author gave the test only after the resting pulse was within a few beats of the resting pulse achieved prior to the 8-1 Test. Data was collected on twenty-nine of the subjects in the second testing group. One subject was eliminated after the testing due to an injured left knee. This information 18 was not obtained until discomfort was apparent while the subject was stepping on the Harvard Step Test. The sub— ject terminated after three and one-half minutes of stepping due to the discomfort. It was quite apparent that he could not give an all out effort, which invalidated his step test SCOPE . CHAPTER IV RESULTS AND ANALYSIS Presentation of Data The investigation dealth with the administration of an initial testing procedure which involved 15 subjects. Each individual subject was given 15 completely different step tests. After each test was completed the subjects pulse recovery rate was recorded for a full three and one- half minutes. Table 1 gives the mean heart rates in -recovery at fifteen-second intervals. This was necessary to study the recovery patterns for each specific step test. This first initial test was a pilot study to see if any step or combination of steps would produce a pulse rate recovery pattern that would significantly correlate with the Harvard Step Test. In looking at the pulse recovery curves of all 1A experimental tests, in relationship to the Harvard, it was found that little, if any significance could be determined at the one to one and a half minute recovery level. In Table 1 it can be seen by the means of all recovery rates that only in the Harvard Step Test is there a rapidly declining pulse rate. In all of the 1A other tests the pulse rate has leveled off, to a certain degree, by the end of one minute in recovery. 19 2O mHH mmH mOH mOH HOH NHH mHH AHH HNH mNH mNH mmH HA mum NA mA HA om om ow mm mm mm mOH mHH HMH 0A H 9oom< Hm Hm Nm Nm mm mm Am mm Na mm OHH mNH HA Nnoma NA HA NA NA NA NA HA HA HA mm NOH mHH mm Hnom< MA mA NA NA NA mA HA HA HA Hm Hm 00H we mum HA NA HA HA HA HA HA HA HA Hm mm mOH mm Hum HA 0A HA HA NA MA NA mA mA Hm om 00H NA msm mm mm mm mm Am Am As we as HA om moH mm Nnm mm mm mm mm mm Am mm mm mm mA mm HOH 0A Ham NA HA 0A NA so HA mA mo Hm om mm 00H Aw m9ma HA NA mA BA NA Am NA NA Am mA mm AOH mm N9m< Ho HA HA mm mm mm om 0A As 0A mA OOH Ho HumH HA NA NA 0A 0A 0A 0A mm mm HA mA mm mm m9< HA 0A 0A HA HA 0A 0A 0A 0A 0A MA mm HA NIH mm Am Am Am Hm mm mm Hm mm Am HA mm Am HIH . . . . . . . . . . . . mmHsm oo.m mH.N om.N mH.N oo.N mH.H om.H mH.H oo.H mH. om. mH. weHumom pass 2 .mHuc mHm>noch pcooom mH on .H asoaw HapcmEHaogxo .mmpma mao>oooh :NoZII.H oHnme 21 It must be realized that the Harvard Step Test is unique, in that it has its own specific means of measuring cardiovascular fitness. The Harvard, because of the intensity of exercise, gathers significant pulse rates later in recovery. Why then, if a different technique of measurement in testing is used, is not a different means of collecting pulse rates used? In the first place, the problem being studied is that of developing an easier, faster means of determining cardiovascular efficiency. This being the case, the testing and the heart rate recovery time being shortened, will add to a solution of the problem. In looking at the heart rates in recovery for all experimental tests, it was surprising to find, that in-most tests, from exact termination of exercise to A5 seconds in recovery produced almost parallel recovery patterns. The author took a cross sampling of the recovery curves to demonstrate the recovery patterns for four tests. The results of tests ABCD—l, B—3, B-1, and A—1 may be observed. (See Figure 2). The parallel of the recovery pattern from termination of exercise to A5 seconds in recovery is very apparent. With this evidence the author proceeded to do further statistical work in comparing the B—1 test with the Harvard Step Test. The B—l test was administered on an 8—inch step level for 1 minute. This is very convenient, for where can a person not find a common stair tread which is 8—inches in height? This height of 8~inches is within the realm of physical ability for many more individuals 22 .mcmme pmmp HmpcoEHaoaxm Adomll.m maswwm mEHB moo VOA Tow rem IOOH noHH fioma IQMH rOHH 9omH HIVH ESTfld 23 than the Harvard Step Test. Another ideal factor is that the 5—minute stepping time on the Harvard Step Test is cut to 1 minute with the B-1 test. Analysis of Data Rank order correlations were computed, comparing the estimated rank of fitness against a series of 8 recovery time intervals for the B-1 test. As can be observed, the highest correlation was found in the recovery interval of 0 to A5 seconds. A correlation of .68 was achieved, which is not as high as anticipated, but does show a need for further study of the problem (See Table 2). At this point evidence was inconclusive due to a lack of further data. At the suggestion of Dr. W. W. Heusner, the authors advisor, another group of subjects was tested. This time a more random sample of the total population at Andrews University was chosen. It was felt that a need for subjects in the poor range of cardiovascular fitness was needed. The author proceeded to enlist the help of students as subjects in the final testing. In all, 30 men were tested. One subject was not used in the study because of a knee injury which invalidated his Harvard Step Test score. The B-l test was administered first and within a 24 hour period the Harvard Step Test was given. The mean recovery rates were recorded for the B-1 test and the Harvard Step Test (See Table 3). 24 Table 2.-—Rank order coefficient of correlations estimated fitness with B—l test n = 15. m Recovery rate intervals Pearson r (aprox.) (seconds) 1. 0-15 .56 .57 2. 15-30 .57 .59 3. 30-A5 .58 .60 A. 0—30 .60 .62 5. 15—95 .61 .63 ' 6. 0-95 .68 .70 E 7. 1 00-1 30 .A1 .A3 9 8. Sum of 3-30 sec. intervals .09 .51 “ ‘T9th'; “Wk . Table 3.—-Mean recovery rates on the final Harvard and B-1 Tests n = 29. m Test Rest :15 :30 :AB 1:00 1:15 1:30 1:95 2:00 H—5 74.0 169 162 150 104 136 127 12A 121 B—l 73.1 116 99 87 83 78 7A 75 7A It was unnecessary to record total heart beats to a full three and one-half minutes in recovery, since the Harvard Step Test needs only one and one-half minutes in recovery and the B—1 test needs only A5 seconds for comparison. The author decided to keep a record of the recovery heart rates for a period of 2 minutes on the final group of subjects. A comparison was made of the means of the first group on the Harvard Step Test and the means of the second group. 25 The same was done with the means of the B-1 test for testing groups 1 and 2 (See Table A). Table A.--Comparison of means, pilot study with initial test. Test Rest :15 :30 :45 1:00 1:15 1:30 1:A5 2:00 =15 71 166 1A8 138 129 125 121 117 116 n=29 7A 169 162 150 199 136 127 12A 121 =15 70 10A 86 75 69 69 68 67 68 n=29 73 116 99 87 83 78 7A 75 7a It is interesting to note how close the two means of the Harvard Step Test and the B-1 test parallel one another (See Figure 3). The difference between the means in both groups can be explained because of the sampling in each instance. The first group was much more fit, with no subjects falling in the poor category on the Harvard Step Test. In contrast the second group had 5 men who fell in the poor category on the Harvard Step Test. This difference can be attributed to the lower means on the first group tested. After testing the second group of subjects a rank order coefficient of correlation was computed. Subjects were ranked in order from first to last, using the Harvard M4 «Vania... -. -31” :. - - 26 .mpmmp Hum sew ooua ma" — _ II“ opm>pmm .mmnoom cmmzll.m mEHB om" ma“ _ _ onsmfim ooH I om ow OOH omH oza ova oma ELVH ESTfld 27 Step Test as the evaluating device. Each subject was then ranked correspondingly with the B—1 test. A high positive correlation of .89 was achieved between these two tests. This first indicates that the B-1 test is very good, and in fact should produce essentially the same results as can be achieved with the Harvard Step Test. The capacity of these two tests to differentiate between good, average, and poor groupings of cardiovascular fitness required further analysis. To test the results, a one way Analysis of Variance was applied. The Harvard Step Test produced an F of 66.67 which is highly significant. With the B-1 test an F of 79.75 was produced. This again is highly significant, for an F of 5.53 is all that is required for significance at the .01 level (See Table 5). Before the Analysis of Variance was completed a Bartlett test was administered. A Bartlett Chi-square of 2.16 was produced on the Harvard Step Test and a Bartlett Chi-square of u.uu on the B-1 test. A Chi—square of 5.991 for these tests would have been required for significance at the .05 level, consequently the assumption of homo- geneity of variance which is required for an analysis of variance is tenable. The next information needed was a meaningful comparison of pairs of means. At this point, it was known that the B-1 test is as good a test as the Harvard Step Test and corre- lates well with it. At this point it is not known how well 28 Table 5.—-Ana1ysis of variance: Harvard Step Test and B-1 Step Test. W Source S.S. D.F.' M.S. F Harvard Step Test: Between 10,626 2 5,313.00 Within 2,061 26 79.69 Total 12,687 28 66.67 B-l Step Test: Between 4,761 2 2,380.50 Within 776 26 29.85 Total 5,537 28 79.75 the two tests distinguish between good and average, good and poor, and average and poor groups of cardiovascular fitness. For a meaningful interpretation of the data a comparison of two means at a time was applied. The technique applied was that of H. Scheffe (5). The results of this test showed that in both the Harvard Step Test and the B-1 Test differentation between the good and poor groups, and between the average and poor groups at the .05 level was obtainable. One factor that was not substantiated was adequate resolution between the good and average groups using either test. (See Table 6) 29 Table 6.--Scheffé Multiple Comparisons of Means. Harvard Step Test and B-1 Step Test. m Comparison F Harvard Step Test: 1. Good, Average 1.20 2. Good, Poor 6.62* 3. Average, Poor 5.99* B-l Step Test: 1. Good, Average 2.3A 2. Good, Poor 7.01* 3. Average, Poor 8.73* *significant at .05 level CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary Lucien Brouha and associates developed the Harvard Step Test for the purpose of testing the cardiovascular efficiency of men during World War II. He succeeded in doing so by the development of a step test which classified men into good, average, and poor groups of cardiovascular efficiency. The prime purpose of this study was to investigate a variety of step levels and the time in stepping at various levels. The author desired to develop, if possible, a step test which would be much easier to administer to certain individuals,namely shorter and heavier people. Another purpose was to investigate the need for 5 full minutes of stepping, as is required in the Harvard Step Test. The group of subjects for the study was tested at Andrews University. The subjects were students ranging in age from 18 to 23 years. A group of 15 subjects were used in the initial study. This group took 15 seperate tests, including the Harvard Step Test. Data compiled from this group showed a need for further study. 30 31 A second group of 29 students was randomly picked for the second and final testing from the same population. The second group covered the total spectrum of good, average, and poor cardiovascular fitness. Rank order correlations were run, comparing the Harvard Step Test ranking with the B-1 Step Test using a 0 to 45 second recovery interval. Analysis of Variance was applied to both the Harvard Step Test and the B-1 Test. Both tests showed significance and were, therefore, treated with the Scheffe test, which compared the means between each group for both tests. Conclusions The following conclusions were drawn from an analysis of the data: 1. The first group of 15 subjects were not randomly picked. Too many of the subjects were in the good and high average classification on the Harvard Step Test, causing the rank order correlation to be low. 2. The second group was a more randomly selected group, producing a very high correlation between the B-1 Test and the Harvard Step Test. 3. The B-l Test will yield the same information as the Harvard Step Test. A. The B—l Test has the same problem of significantly categorizing the good from the cardiovascular fit person. 32 5. The B-l Test distinguishes between good and poor, average and poor with a higher level of significance than the Harvard Step Test. Re commendation: 1. The B41 Test needs further study before it replaces the Harvard Step Test. 2. Computation of a table similar to the table for the Harvard Step Test should be produced so all ranges of cardiovascular fitness may be evaluated. 3. Tests of reliability and validity should be administered to the B—1 Test. 10. BIBLIOGRAPHY Armbruster, David, and W. W. Tuttle. "The use of the pulse-ratio test for ranking efficiency in swimming," Unpublished Report, 1931. Brouha, Lucien, M.D. "The step test: a simple method of measuring physical fitness for muscular work in young men," Research Quarterly, 1A:31—36, 1982. Campbell, J. M. H. "Vital capacity, pulse rate before and after exercise, and physical fitness in health," Guyjs HOSpital Report, 75:263—281, 1925. Consolazio, Frank, and others. "Physical fitness and performance in man," Ppysiological Measurements of Metabolic Functions in Man. New York} McCraw-Hill, 19 33 pp. BW’é—BST Ferguson, G. A. Statistical Analysis in Psychology and Education. “New—York: MEGraw—Hill, 1966, pp. 295-297. Hambly, W. D., G. H. Hunt, L. E. L. Parker, M. S. Pembrey, and E. C. Warner. "Tests for physical efficiency," Guy‘s Hospital Report, 72:367-385, 1922. Hambly, W. D., M. S. Pembrey, and E. C. Warner. "The physical fitness of men assessed by various methods," Ggy's Hospital Report, 75:388—39“, 1925. Matthews, D. K. Measurement in Physical Education. Philadelphia: W. B. Saunders, Cb., 196A. Tuttle, W. W. "The use of the pulse—ratio test for rating physical efficiency," Research Quarterly, 2:5—17, 1931. Tuttle, W. W., and R. C. Wilkens. "The application of the pulse~ratio test to efficiency in performing on gymnasium apparatus, the horizontal bar," Arbeits Physiol., 3:“49-A55, 1930. 33 3 m R A R m L Y n S R E w N U E T Amnv T"). ”qu77