ADAPTED PHYSICALACTIVIN QUARTERLY, 1999,16,86-95 O 1999 Human Kinetics Publishers, Inc.

Reliability of Eurofit Physical Fitness items for Adolescent Males With and Without

Mental Retardation

Ciaran Mac Donncha and Anthony W.S. Watson University of Limerick

Terence McSweeney Cope Foundation

Daniel J. O'Donovan University College Galway

The purpose was to examine the reliability of physical fitness items from the Eurofit Test Battery for adolescent males with mild mental retardation (MMR; n = 63, mean IQ = 63.0 k11.5, mean age = 15.5 k 1.2) and those without (n = 22, mean age = 15.6 ? 0.6). Males with MMR scored significantly lower (p I .005) than those without on all items except sum of skinfolds, height, and weight. Intraclass correlations (ICCs) ranged from .94 to .99 for males with MMR and .85 to .99 for those without. Percentage error of the mean for all items ranged from 0.5 to 47.5% for participants with MMR and 0.4 to 32.2% for those without. ICCs indicated that Eurofit physical fitness test items are reliable measures for males with and without MMR. However, the percentage error of the mean is quite large for sit-and-reach and 20-m shuttle test (20- MST) items.

Little information is available on the reliability of physical fitness items for individuals with mild mental retardation (MMR). Many studies report only inter- class correlation coefficients calculated by the Pearson product moment method (e.g., Rarick, Dobbins, & Broadhead, 1976; Pizzaro, 1990), although intraclass correlation techniques are now favored for test-retest data (Thomas & Nelson, 1996). In particular, little research has focused on physical fitness field test items for older adolescents (> 15 years) with MMR.

Ciaran Mac Donncha and Anthony W.S. Watson are with the Physical Education and Sports Science Department at the University of Limerick, Ireland. Terence McSweeney is with the Department of Physical Education at Cope Foundation, Cork, Ireland. Daniel J. O'Donovan is with the Department of Physiology at the University College Galway, Ireland.

Reliability of Fitness Items 87

Rarick et al.'s (1976) study was one of the first to address reliability of fit- ness items for individuals with mild and moderate MR and to provide test-retest coefficients for six subgroups of boys and girls, ages 6-9 and 10-13, with and without MR. Test-retest Pearson product moment coefficients with Spearman- Brown corrections for standing long jump and handgrip strength were r = .94-.98 for individuals with and without MMR. Other items in the Rarick et al. study were not similar to those examined by the present investigators.

Pizarro (1990) concluded that the modified sit-up, sit-and-reach test, 880-yd run, and triceps skinfold measure were reliable for use with male and female ado- lescents (n = 126, ages 12-15) with mild to moderate mental retardation (MR). Pizarro stated that the apparent reliability of the 880-yd run is misleading because a large number of procedural difficulties were recorded during test administration (e.g., judgment of pacing and lack of task comprehension).

One of Pizarro's (1990) most valuable contribution was recording all proce- dural deviations during the testing process. For example, during the sit-and-reach test, individuals with mild (n = 10) and moderate (n = 10) MR committed almost twice as many procedural deviations as participants without MR (n = 11). Bending the knees, jumping of the slide mechanism, and improper positioning were most commonly noted during the sit-and-reach test. Of 117 noted deviations during the 880-yd run, 89% were recorded for individuals with mild and moderate MR. Indi- viduals without MR accounted for the remainder. Numerous other researchers cor- roborate these findings (Dishman, 1978; Koh & Watkinson, 1988; Lavay, Zody, Solko, & Era, 1990; Reid, Seidl, & Montgomery, 1989). Test preparation, famil- iarization, participant-tester confidence, and emphasis on pacing are essential components of reliable physical fitness testing for individuals with MR (Pizarro, 1990). More research is needed that attends to these essential components.

Extensive research exists regarding the reliability and validity of cardiorespira- tory endurance tests for individuals with MR (Baumgartner & Horvat, 1991 ; Cressler, Lavay, & Giese, 1988; Fernhall & Tymeson, 1988; Rintala, Dunn, McCubbin, & Quinn, 1992), but investigators have not reached a consensus about the most suitable test. Cressler et al. (1988) reported test-retest reliability of four measures of cardiores- piratory fitness for individuals with mild to moderate MR (n = 17, mean age = 35, mean IQ = 54). Results showed intraclass correlations (ICCs) of R = .64 for the modi- fied PWC170 (Wahlund, 1948), .81 for the 12-min runlwalk (Cooper, 1970), .93 for the Balke Ware treadmill test (Balke &Ware, 1959), and .95 for the Canadian step test (Jette, Campbell, Mongeon, & Routhier, 1976). Liu, Plowman, and Looney (1992) reported R = .91 for the 20-m shuttle test (20-MST) of Leger, Mercier, Gadoury, and Lambert (1988) performed by males without MR (n = 12, ages 12- 15). Fernhall et al. (1996) reported a correlation of .95 for the 20-MST when per- formed by individuals with mild and moderate MR (22 boys, 12 girls; mean age = 14.3). Fernhall et al. (1996) did not indicate which type of correlation was used.

Reliable data are required to determine the need for intervention and to evalu- ate the impact of intervention programs aimed at enhancing the physical fitness of individuals with MMR. Several studies have indicated the reliability of cardiores- piratory fitness tests for individuals with mild and moderate MR. Few studies, however, have addressed the reliability of physical fitness items for adolescents with MMR. Thus, the purpose of this investigation was to examine the reliability of items selected from the Eurofit Test Battery (Council of Europe, 1988) for ado- lescent males with and without MMR.

Mac Donncha, Watson, McSweeney, and O'Donovan

Method

Participants

Written informed consent to participate in the study was obtained from parents of all males (ages 14-18) attending two schools, one for individuals with MMR and another for those without. Both schools were located in Cork City, Ireland, and were within a 1-mile radius of each other. A random sample was drawn from a11 males whose parents gave written consent. Participants were 63 males with MMR (mean IQ = 63.0 f11.5, mean age = 15.5 f 1.2) and 22 males (mean age = 15.6 f 0.6 years) with average or better IQ. MMR classification was based on two mea- sures: IQ between 50 and 70, using the Wechsler Intelligence Scale for Children (WISC; Wechsler, 1991), and a measure of social adaptation, based on the =neland Adaptive Behaviour Scales (Sparrow, Balla, & Cicchetti, 1984). Results of both tests were combined for general MMR classification by school personnel.

In Ireland, the term learning difSiculty is preferred over mental retardation, and no distinction is made between these terms. Learning difficulty is further de- fined by specific and general learning difficulty. Specijic learning dificulty refers to a reading, arithmetic, or perceptual problem with no MR. General learning dificulty indicates the existence of MR. When both specific and general learning difficulty are present, the condition is equivalent to MR as used in North America. Although participants in this study were diagnosed as having a general learning difficulty in Ireland, MMR is used in this article to prevent confusion between learning difficulty and disability. The ethical committee of the Cope Foundation sanctioned the study.

Procedure The battery of selected tests was based on those described in the Eurofit handbook (Council of Europe, 1988) and subsequently used in the Northern Ireland Fitness Survey (Riddoch, 1990). This survey examined the physical fitness of 3,211 nondisabled adolescents and is valuable as a normative resource. The Eurofit hand- book defines physical fitness by the following dimensions: cardiorespiratory en- durance, strength, muscular endurance, speed, flexibility, and balance. Anthropometric measures such as height, weight, and skinfold thickness are also included in the Eurofit Test Battery. Tests that represented each dimension of physi- cal fitness were examined, with the exception of balance, which was excluded for practical reasons (i.e., test organization and duration). Balance was not assessed in the Northern Ireland Fitness Survey.

To estimate reliability, tests were administered twice, 1 week apart. All par- ticipants had been tested previously and were familiar with the equipment. Test- retest conditions, such as testing order, day, and time, were standardized for each participant. When individuals with MMR demonstrated a lack of understanding for the testing procedures, practice trials were permitted until test procedures were mastered. For some tests (e.g., sit-ups), in which repeated trials resulted in fatigue, participants were permitted to take the test on another day.

The following physical fitness tests were administered: height (mm), weight (kg), sit-ups (number completed in 30 s), standing long jump (cm), sit-and-reach (cm), handgrip strength (kg), agility as predicted by a 10- by 5-m shuttle run (s), skinfold thickness (rnrn), and cardiorespiratory endurance as predicted by the

Reliability of Fitness Items 89

20-MST (measured by the number of laps completed). Alternate tests, which rep- resented similar physical fitness dimensions as outlined in the Eurofit handbook, were not examined. These included a bicycle ergometer test for assessing cardio- respiratory endurance, a bent-arm hang for muscular endurance, and a plate tap- ping test for fine motor coordination or speed.

Height was measured using a stadiometer (Seca). Measurements were read to the nearest millimeter. Weight was measured using an electronic weighing scale (Seca). Measurements were read to the nearest 0.1 kg. Participants wore only shorts for height and weight measurements.

Abdominal muscle endurance was measured as the number of sit-ups com- pleted in 30 s. Sit-ups were performed with hands placed at the side of the head, knees bent at 90", and the feet securely held by a partner. A full sit-up was defined as touching the knees with the elbows and returning the shoulders to the ground.

Explosive strength was assessed by a standing long jump, using a tape mea- sure attached to foam mats. Participants were asked to stand behind a line drawn perpendicular to the tape measure and jump forward as far as possible. They were advised to swing their arms and flex their knees prior to jumping. They were also advised to keep both legs close together at take-off and landing and not to fall backward. The distance jumped was recorded from the take-off line to the farthest point backward of the participant. The best of two trials was recorded. The dis- tance jumped was recorded to the nearest centimeter.

Flexibility was assessed using the sit-and-reach test, which indicates range of motion in the hamstrings and lower back regions. A standard sit-and-reach box was used. Participants sat down with their shoes removed and placed their feet against the front of the box. The knees remained fully extended throughout the test. The legs were held in position by the investigator, and participants were asked to reach forward slowly and push the ruler forward as far as possible. A smooth stretching performance was encouraged. The best of three trials was recorded. Measurements were read to the nearest 0.5 cm.

Handgrip strength was measured using an electronic handgrip dynamom- eter. Dominant handgrip measures were read to the nearest 0.1 kg. The dynamom- eter was held by the side, with the arm fully extended and palm facing inward. The dynamometer was compressed as hard as possible for 2-3 s. The best of two trials was recorded.

Agility was measured using a 10- by 5-m shuttle run. Each participant was required to sprint 10 times between two lines placed 5 m apart. The track was 1.3 m wide. Measurements were recorded to the nearest 0.01 s.

Cardiorespiratory endurance was estimated using the 20-MST (Leger et al., 1988), which requires participants to run back and forth between two lines set 20 m apart. Running pace is determined by an audio signal. The initial velocity was 8.5 km . hr -' (5.1 mph) and increased by 0.5 km . hr' (0.3 mph)every minute. The test is terminated when the participant cannot reach the end lines in time with two consecutive audio signals. Distance covered was recorded in laps (one lap = 20 m). Approximately 10 participants were grouped together for this test. A staff mem- ber initially joined participants to set the correct running pace and only withdrew when everyone had achieved this pace. A similar procedure was used by Boreham, Paliczka, and Nichols (1990), Riddoch (1990), and Liu et al. (1992). This test is similar to the Progressive Aerobic Cardiovascular Endurance Run (PACER) used in North America (Cooper Institute for Aerobics Research, 1994; Seaman, 1995).

90 Mac Donncha, Watson, McSweeney, and O'Donovan

Skinfold thickness was measured using a Harpenden skinfold caliper. Four sites were used: midbiceps, midtriceps, subscapular, and suprailiac. All were mea- sured on the right side of the body. Measurements were read to the nearest 0.2 mm.

Statistical Analysis

Statistical analysis was conducted using SPSS for Windows 6.0. Descriptive sta- tistics were computed, and test-retest reliability was established using ICCs. The 95% confidenceintervals (the standard deviation of the mean difference between test and retest multiplied by 2.0) was calculated for each measurement procedure. The percentage error of the mean was also calculated by dividing the 95% confi- dence interval by the maximum mean score. Planned group comparisons were conducted using independent student t tests. A Bonferroni adjustment (alpha = .05 divided by number of comparisons = 9) was applied to the alpha level because multiple t tests result in a greater probability of a Type I error. Alpha was calcu- lated a s p I .005. Skewness of test data was calculated by dividing the raw value by the standard error for skewness. The resultant values were interpreted as Z scores (i.e., values + 1.96 exceedp I .05, and values f 2.58 exceedp I .01). Data were considered normally distributed if Z I 2.0 (Vincent, 1995).

Results

ICCs, corresponding 95% confidence intervals (95% CIS), and percentage error of the maximum mean for individuals with and without MMR are presented in Table 1. All test data were normally distributed, except for sum of skinfolds in both groups.

ICCs for physical fitness items were R = .94-.99 for males with MMR and .85-.99 for those without. ICCs provide unitless estimates of measurement reli- ability. To provide a measure with the original measurement units, 95% CI was calculated. In an intervention study, if an individual does not change more than the 95% CI between pre- and posttest scores, no change can be claimed, as this may be due only to the measurement error inherent in the test and testing procedures. Also, correlation coefficients are influenced by the dispersion of the scores. In the current study, the standard deviations of the individuals with MMR are generally greater than those without. Largely dispersed scores usually result in higher corre- lation coefficients (Baumgartner & Horvat, 1991; Bland & Altman, 1990; Sale, 1991; Watson, 1995) The use of 95% CIS thus offers a more accurate and appli- cable estimate of reliability.

Independent student t test group comparisons (see Table 2) indicated that male adolescents with MMR scored significantly poorer ( I .005) on all variables except for height and weight. No significant difference existed between groups for the sum of skinfolds.

Discussion

ICCs for physical fitness items were R = .94-.99. for males with MMR and 35- .99 for those without. These correlations suggest that the reliability of items in this study were high for both groups. ICCs for sit-and-reach (R = .97), standing long

Reliability of Fitness Items 91

Table 1 Intraclass Correlations (ICCs), 95% Confidence Intervals (95% CIS), and Percentage Error of the Mean for Males With (n = 63) and Without Mental Retardation (n = 22)

Variable

With mental retardation Without mental retardation % Error % Error

ICC 95% CI of mean ICC 95% CI of mean

Height (mm) .99 8.4 Weight (kg) .99 1.6 Sit-and-reach (cm) .97 4.9 Standing long jump (cm) .98 19.6 Sit-ups (reps) .95 4.7 Handgrip strength (kg) .95 6.9 Shuttle run (s) .97 2.0 20-MST (laps) .94 21.6 Sum of skinfolds (mm)" .99 2.4

"Sample sizes for sum of skinfolds were 20 and 21 for males with and without MMR, respectively.

Table 2 Comparison of Male Adolescents With (n = 63) and Without (n = 22) Mental Retardation on Fitness Measures

Variable

With Without mental retardation mental retardation M SD M SD t

Height (mm) Weight (kg) Sit-and-reach (cm) Standing long jump (cm) Sit-ups (reps) Handgrip strength (kg) Shuttle run (s) 20-MST (laps) Sum of skinfolds (mrn)"

"Sample sizes for sum of skinfolds were 20 and 21 for males with and without MMR, respectively. * p 5.05. **p 5.005.

92 Mac Donncha, Watson, McSweeney, and O'Donovan

jump (R = .98), sit-ups (R = .95), and handgrip strength (R = .95) were higher for males with MMR. ICCs for males without MMR were R = .91, .93, 3.5, and .91, respectively, for the measures in question. This difference may be explained by the greater dispersion of scores as reflected by the standard deviations of males with MMR. ICCs were higher than the Pearson correlations reported by Pizarro (1990) for individuals (ages = 12-15) with MR (n = 44), where r = .83 for modified sit- ups and .90 for sit-and-reach. Pizarro's (1990) test items were similar to those in the present study. However, Pizarro (1990) used interclass correlations (Pearson product moment) to estimate reliability. Interclass correlations tend to be smaller than ICCs when based on the test-retest scores of the same population (Erbaugh, 1990).

Safrit (1990) reported correlation coefficients for 11- to 14-year-old boys without MR ranging from .94 to .97 and .62 to .93 for the sit-and-reach and sit-up tests, respectively. The type of correlation used was not stated. In the present study, individuals with MMR had ICCs equal to the upper level of the correlation ranges described by Safrit (1990), but participants were older than those in Safrit's study.

Rarick et al. (1976) described interclass correlation coefficients greater than r = .90 for boys with MMR (ages 10-13) for the following tests: right and left grip strength, standing long jump, and forward spinal flexion (toe touch). Balogun, Adenlola, and Akinloye (1991) reported a reliability coefficient (Pearson) greater than r = .90 for handgrip strength of college male students. The present study supports these researchers' findings.

Reliability coefficients for the shuttle run and the 20-MST in the present study were similar for males with and without MR. Correlation coefficients for the 20-MST in the present study (R = .94 and .95 for individuals with and without MMR, respectively) are greater than those reported by Liu et al. (1992) for partici- pants without MMR and similar to those of Fernhall et al. (1996) for participants with MMR. Based on correlation evidence, the 20-MST shows promise as a reli- able test of aerobic endurance for individuals with MMR. Very few procedural difficulties were noted during this test. Participants adapted very quickly to test requirements. This test seems suitable for individuals with MMR as the intensity level is controlled by the test, unlike distance runs. Also, in the earlier stages, the participant can adapt to test demands, as the level of intensity is low. A potential difficulty with the 20-MST is that participants are generally measured in groups, and an individual's score may be influenced by others dropping out or persever- ing. In the current study, 20-MST reliability for individuals with MMR was not as high as that reported by Cressler, Lavay, and Giese (1988) for the Balke Ware treadmill test and the Canadian step test. In the present study, participants with and without MMR scored 59.1 + 24.6 and 85 + 20.0 laps, respectively. Boreham et al. (1990) reported a score of 81.7 + 15.9 laps for 15-year-old males without MMR. Riddoch (1990) reported a similar score (86 + 20 laps) for same-age peers. Liu et al. (1992) found that 12- to 15-year-old males without MMR scored 71.8 + 27.0 laps on the 20-MST.

Research into the validity of 20-MST as a measure of cardiorespiratory en- durance for individuals with MMR is required. Montgomery, Reid, and Koziris (1992) reported a validity correlation of r = .78 between a modified shuttle run and the V0,max score of 18 individuals with mild to moderate MR (ages 20-35). However, the shuttle run underestimated V0,max values by 28%. Femhall et al. (1996) reported r = .88 between 20-MST and V0,max scores of individuals with mild and moderate MR (22 boys, 12 girls, mean age = 14.3).

Reliability of Fitness Items 93

The reliability of nonperformance tests, such as height, weight, and skinfolds, was high for both groups in the present study. To achieve a reproducible score, test procedures must be standardized. A difference in the 95% CI and the percentage error for the mean for the sum of skinfolds for both groups are presented in Table 1. The greater 95% confidence interval for male adolescents without MMR was accounted for by variability in one individual's test-retest score.

The magnitude of the 95% CI and percentage error of the mean of some tests in the present study do not support the corresponding ICCs. A high correlation should suggest a low 95% CI and low corresponding percentage error of the mean. The magnitude of the 95% CI and percentage error for the sit-and-reach and 20- MST do not agree with the strength of the corresponding ICCs in both groups (see Table 1). For example, the ICC for the sit-and-reach test for male adolescents with MMR is R = .97, the corresponding 95% CI is 4.9 cm, and the percentage error of the mean is 47.6%. This lack of agreement may be explained by the effect that the dispersion of scores has on correlations (Baumgartner & Horvat, 1991; Bland & Altman, 1990; Sale, 1990; Watson, 1995). Thus, investigators should carefully consider which analysis method is used to establish reliability.

Procedural deviations similar to those recorded by Pizarro (1990) were noted in the current study, but these were overcome by insisting that participants become adept at each test and by administering each item so as to meet performance crite- ria (e.g., knees were held down by an assistant in the sit-and-reach test, jerking forward was discouraged, sit-ups were not counted unless elbows touched the knees, and multiple trials were permitted so that participants could learn a test item). Procedural difficulties do exist when testing individuals with MMR, but reliable results can still be obtained, as demonstrated by several studies.

Conclusion

Findings of this study indicate that physical fitness items as outlined by the Eurofit handbook (Council of Europe, 1988) are appropriate for individuals with MMR. Moreover, ICC as a reliability estimate indicates that the physical fitness items in this study are reliable for male adolescents with and without MMR.

Using a 95% CI may be a more appropriate method for establishing test item reliability because it is not affected by the dispersion of the scores, and it is pre- sented in the same measurement units as used in the test. Future studies should indicate whether interclass or intraclass correlations were used to estimate test- retest reliability and should also present a 95% confidence interval for each vari- able examined.

References

Balke, B., &Ware, R. (1959). An experimental study of physical fitness of air force person- nel. U S A m e d Forces Medical Journal, 10,675-688.

Balogun, J.A., Adenlola, S.A., & Akinloye, A.A. (1991). Grip strength normative data for the Harpenden dynamometer. Journal of Orthopaedic and Sports Physical Therapy, 14, 155-160.

Baumgartner, T., & Horvat, M. (1991). Reliability of field based cardiorespiratory fitness running tests for individuals with mental retardation. Adapted PhysicalActivity Quar- terly, 8, 107-114.

94 Mac Donncha, Watson, McSweeney, and O'Donovan

Bland, J.M., & Altman, D.G. (1990). A note on the use of the intraclass correlation coeffi- cient in the evaluation of agreement between two methods of measurement. Comput- ers in Biology and Medicine, 20,337-340.

Boreham, C.A.G., Paliczka, V.G., & Nichols, A.K. (1990). A comparison of the PWC170 and the 20-MST tests of aerobic fitness in adolescent school children. Journal of Sports Medicine and Physical Fitness, 30, 19-23.

Cooper Institute for Aerobics Research. (1994). The Prudential FITNESSGRAM test ad- ministration manual. Dallas, TX: Author.

Cooper, K.H. (1968). A means of assessing maximal oxygen intake. Journal of the Ameri- can Medical Association, 203,201-204.

Council of Europe, Committee for Development of Sport. (1988). Eurofit: European tests of physicalfitness. Rome: Edigraf Editoriale Grafica.

Cressler, M., Lavay, B., & Giese, M. (1988). The reliability of four measures of cardiores- piratory fitness with mentally retarded adults. Adapted Physical Activity Quarterly, 5,285-292.

Dishman, R.R. (1978). Aerobic power, estimation of physical activity and attraction to physi- cal activity. Research Quarterly, 49,285-292.

Erbaugh, S.J. (1990). Reliability of physical fitness tests administered to young children. Perceptual and Motor Skills, 71, 1123-1128.

Fernhall, B., Pitetti, K., Vukavich, M., Stubbs, N., Hensen, T., Winnick, J., & Short, F. (1996). Validation of cardiorespiratory fitness field tests in children with mental re- tardation. Medicine and Science in Sports and Exercise, 28(Suppl. 5), S50.

Fernhall, B., & Tymeson, G.T. (1988). Validation of a cardiorespiratory fitness field test for adults with mental retardation. Adapted Physical Activity Quarterly, 549-59.

Halle, J.W., Silverman, N.A., & Regan, L. (1983). The effects of a data-based exercise program on physical fitness of retarded children. Education and Training of the Men- fally Retarded, 18,22 1-225.

Jette, M., Campbell, J., Mongeon, J., & Routhier, R. (1976). The Canadian Home Fitness Tests as a prediction of aerobic capacity. Canadian Medical Association Journal, 114,680-682.

Koh, S.M., & Watkinson, E.J. (1988). Endurance run pacing of moderately mentally handi- capped children. Journal ofthe Canadian Association for Health, Physical Educa- tion and Recreation, 54, 12- 15.

Lavay, B., Zody, J., Solko, C.M., & Era, K. (1990). Effects of a 7-month runtwalk program on the physiological fitness parameters of adults with mental retardation. In G. Doll- Tepper, C. Dahms, B. Dolt, & H. Von Selzam (Eds.), Adaptedphysical activity: An interdisciplinary approach. Proceedings of the 7th international symposium (pp. 289- 298). Berlin: Springer-Verlag.

Leger, L.A., Mercier, D., Gadoury, C., & Larnbert, J. (1988). The multistage 20 metre shuttle run test for aerobic fitness. Journal of Sports Sciences, 6,93-101

Liu, N.Y.S., Plowman, S.A., &Looney, M.A. (1992). The reliability and validity of the 20- meter shuttle test in American students 12 to 15 years old. Research Quarterly for Exercise and Sport, 63,360-365.

McConaughy, E.K., & Salzberg, C.L. (1988). Physical fitness of mentally retarded indi- viduals. In N.W. Bray (Ed.), International review of research in mental retardation (pp. 227-258). New York: Academic Press.

Montgomery, D.L., Reid, G., & Koziris L.P. (1992). Reliability and validity of three fitness tests for adults with mental handicaps. Canadian Journal of Sports Sciences, 17, 309-315.

Reliability of Fitness Items 95

Pitetti, K.H., & Campbell, K.D. (1991). Mentally retarded individuals: Apopulation at risk? Medicine and Science in Sports and Exercise, 23,586-593.

Pitetti, K.H., Rimmer, J.H., & Fernhall, B. (1993). Physical fitness and adults with mental retardation. Sports Medicine, 16,23-56.

Pittetti, K.H., &Tan, D.M. (1991). Effects of a minimally supervised exercise program for mentally retarded adults. Medicine and Science in Sports and Exercise, 23,594-601.

Pizarro, D.C. (1990). Reliability of the health related fitness test for mainstreamed educable and trainable mentaIly handicapped adolescents. Adapted Physical Activity Quar- terly, 7,240-249.

Rarick, G.L., Dobbins, D.A., & Broadhead, G.D. (1976). The motor domain and its corre- lates in educationally handicapped children. Englewood Cliffs, NJ: Prentice Hall.

Reid, G., Seidl, C., &Montgomery, D.L. (1989). Fitness tests for retarded adults. Journal of Physical Education, Recreation and Dance, 60,76-78.

Riddoch, C.R. (1990). Northern Irelandfitness survey-1989. Belfast: Queen's University of Belfast, Division of Physical and Health Education.

Rintala, P., Dunn, J.M., McCubbin, J.A. & Quinn, C. (1992). Validity of a cardiorespiratory fitness test for men with mental retardation. Medicine and Science in Sports r . ~ d Exercise, 24,941-945.

Safrit, M.J. (1990). The validity and reliability of fitness tests for children: Areview. Pedi- atric Exercise Science, 2,9-28.

Sale, D.G. (1991). Testing strength and power. In D. MacDougall, H.A. Wenger, & H.J. Green (Eds.), Physiological testing of the h ighpegomnce athlete (2nd ed., pp. 21- 107). Champaign, IL: Human Kinetics.

Seaman, J.A. (Ed.). (1995). Physical best and individuals with disabilities: A handbook for inclusion in fitness programs. Reston, VA: AAHPERD, American Association for Active Lifestyles and Fitness.

Sparrow, S.S., Balla, D.A., & Cicchetti, D.V. (1984). Vineland Adaptive Behaviour Scales. Circle Pines, MN: American Guidance Service.

Thomas, J.R., & Nelson, J.K. (1996). Research methods in physical activity (3rd ed.). Champaign, IL: Human Kinetics.

Vincent, W.J. (1995). Statistics in kinesiology. Champaign, IL: Human Kinetics. Wahlund, H. (1948). Determination of the physical working capacity. Acta Medica

Scandinavica, Suppl., S215. Watson, A.W.S. (1995). Physical fitness and athletic pegomance (2nd ed.). London:

Longman. Wechsler, D. (1991). Wechsler Intelligence Scale for Children-Ill. San Antonio, TX: Psy-

chological Corporation.

Author's Note

This study was supported in part by the National Rehabilitation Board and the Cope Foundation, Cork, Ireland. Requests for reprints should be sent to C. Mac Donncha, Physical Education and Sports Science Department, University of Limerick, Ireland.