Proposal Component Points Topic Title==> Research Question==> Data Source (real or fabricated) If real - identify source ==> Independent Variable==> Independent Variable Type (Nominal, Ordinal, Continuous) ==> Dependent Variable==> Dependent Variable Type (Nominal, Ordinal, Continuous) => Inferential Test - use the decision table below to determine which tests to use. The decision will be based on the type of variables you specified above=> How many samples 3 or more Pearson Correlation Type sample One sample only - one variable Variable Type Continuous One sample only - two variables Both Continuous Both Categorical 1 2 Find a valid reliable substitute measure for height. The Suitability of Arm Span as a Substitute Measurement for Height. Can arm span be used as a reliable substitute for height? Height and arm span measurements of high school students. Arm span measure Continuous Height Continuous Paired sample Unpaired sample Unpaired sample Unpaired Categorical Statistical Test One sample t-test Chi Square Goodness of Fit Pearson correlation Chi Square Categorical Paired t-test Independent t-test Chi Square Continuous ANOVA Continuous 5 5 15 10 10 15 10 15 15 Proposal Component Points Title==> 10 Research Question==> Data Source (real or fabricated) If real - identify source ==> Independent Variable==> Independent Variable Type (Nominal, Ordinal, Continuous) ==> Dependent Variable==> Dependent Variable Type (Nominal, Ordinal, Continuous) ==> Inferential Test - use the decision table below to determine which tests to use. The decision will be based on the type of variables you specified above==> 15 10 10 15 10 15 How many samples Type sample One sample only - one variable Variable Type Continuous One sample only - two variables Both Continuous Both Categorical 1 2 3 or more Paired sample Unpaired sample Unpaired sample Unpaired Categorical 15 Statistical Test One sample t-test Chi Square Goodness of Fit Pearson correlation Chi Square Categorical Paired t-test Independent t-test Chi Square Continuous ANOVA Continuous The Suitability of Arm Span as a Substitute Measurement for Height HLTH 501 David O. Budain 1 Abstract Many anthropometric equations rely on individual height. Accurate height is not obtainable when various skeletal abnormalities exist. Arm span is proposed as a possible substitute for height. Thirteen subjects' arm span and height were measured. The Pearson R for arm span and height was 0.96 (p<0.05). Regression analysis was used to build and equation predicting height from arm span (Height = 0.8655 x Arm Span + 9.3368). Results of this study show that arm span and height are strongly correlated and arm span can be used as a reliable predictor of height. Introduction In many medical, physiological, and human performance measurements the height of human subjects is used as a predictive and/or classification variable. Equations predicting Body Mass Index, pulmonary function, caloric expenditure, and body fat percentage are just a few of the many equations using height as a predictive variable. 1 However, spinal curvature conditions such as kyphosis, scoliosis, lordosis, and kyphoscoliosis make it difficult to determine the correct height of the individual and thereby necessitating the need to identify a substitute anthropometric measurement.2 The need for an anthropometric measurement to serve as a substitute for height has long been recognized. One possible substitute measurement is arm span, that is - the distance from the left middle finger tip to the right middle fingertip of outstretched arms parallel to the ground. This relationship is notably shown in the drawing Vitruvian Man by Leanardo da Vinci (See Figure 1.). Figure 1. Vitruvian Man by Leonard da Vinci. 2 The purpose of this cross-sectional observational study was to determine if there was a significant relations ship between arm span and height to determine if a arm span could serve as a valid and reliable substitute for height. 3 Methods Sample: A convenience sample of 12 high school seniors and 1 senior high school teacher will be used. Equipment: Task Force Hand Tools 25 foot tape measure. Measurements: Each subject height (with shoes off) will be determined with the subject standing flat footed and with erect posture. The arm span will be taken with arms outstretched, parallel to the ground, from the tip of the right middle finger to the left middle finger across the back. All measurements will be recorded to the nearest inch. Statistical Procedures: Mean, median, standard deviation, minimum and maximum will be calculated for the sample. Data will be examined for outliers. Pearson product moment correlation was used to determine the magnitude and significance of the relationship between arm span and height. Hypotheses tested: Null Hypothesis: (rho) =0 There is no significant relationship between arm span and height. 4 Alternative Hypothesis: (rho) 0 There is a significant relationship between arm span and height. Hypotheses tested at the 0.05 level of significance. If a significant relationship between arm span and height is determined then regression analysis was used to derive an equation to predict height from arm span. Data analysis and graph creation were accomplished using SPSS 20.0 5 Results Arm span and height measurement are shown in Table 1. Table 1. Raw data and correlation parameters Student Name Arm Span Height (Inches) (Inches) A 61.5 63.5 B 70.5 70.0 C 66.5 66.0 D 68.0 65.5 Dr. Barton 67.0 68.0 E 60.5 61.5 F 71.5 72.5 G 77.0 76.5 H 62.0 64.5 I 64.0 65.0 J 64.0 64.5 K 72.5 71.0 L 71.0 72.0 6 Descriptive statistics for arm span and height are shown in Table 2. Table 2. Descriptive Statistics ArmSpan N Valid Height 13 13 0 0 Mean 67.38 67.73 Median 67.00 66.00 64 65 4.946 4.347 24.465 18.901 Skewness .330 .582 Std. Error of Skewness .616 .616 Kurtosis -.625 -.439 Std. Error of Kurtosis 1.191 1.191 Range 17 15 Minimum 61 62 Maximum 77 77 25 63.00 64.50 50 67.00 66.00 75 71.25 71.50 Missing Mode Std. Deviation Variance Percentiles Figure 2. Box plot of arm spam measurements. 7 Figure 3. Box plot of height measurements. Correlation between armspan and height are shown in Table 3. Table 3. Correlations ArmSpan Spearman's rho ArmSpan Correlation Coefficient Sig. (2-tailed) N Height Correlation Coefficient Sig. (2-tailed) N Height 1.000 .963** . .000 13 13 .963** 1.000 .000 . 13 13 **. Correlation is significant at the 0.01 level (2-tailed). 8 Scatterplot of arm span and height is shown in Figure 3. Figure 3. Scatterplot of Arm Span and Height 9 Results of regression analysis is shown in Table 4. Table 4. Regression Analysis r r Std. Error ANOVA table Source Regression Residual Total SS 191.8190 15.5977 207.4167 0.925 0.962 1.249 df 1 10 11 12 1 Height MS 191.8190 1.5598 F 122.98 p-value 6.11E-07 confidence interval 95% upper 21.1675 1.0394 Regression output coefficient std. variables s error Intercept 9.3368 5.3097 n k Dep. Var. Arm Span 0.8655 0.0780 t (df=10) 1.758 11.090 p-value .1092 6.11E07 95% lower -2.4940 0.6916 Discussion In an effort to determine whether or not a significant correlation between arm span and height, measurements were obtained from 13 \"normal\" subjects. The results shown in Table 2 and Figures 2 and 3 indicate there were no outliers and that the data were almost normally distributed. Therefore all data were included in the statistical analyses. Results of the correlation analysis in Table 3 indicate a significant (p<0.05) strong positive (r=0.925) correlation between arm span and height. This strength and direction of the correlation is further demonstrated by the scatterplot shown in Figure 3. 10 The significant correlation between arm span and height allowed for subsequent regression analysis, the results of which are shown in Table 4. The resulting regression equation is as follows: Height = 0.8655 x Arm Span + 9.3368 The results of this study suggest that arm span measurement can be used as a substitute for height in normal subjects. Limitations of this study include the small sample size, narrow range of arm spans and height, and the fact that all subjects were healthy and had no observable spinal curvature. Caution must be exercised in generalizing these results to populations other than described above. 1. Use of anthropometric measures to assess weight loss;George A. Bray,4 M.D., Frank L. Greenway,5 M.D., Mark E. Molitch,6 M.D., William T. Dahms,7 M.D., Richard L. Atkinson,8 M.D., and Kare 2. The use of arm span as a predictor of height: A study of South Indian Women SP Mohanty, S Suresh Babu and N Sreekumaran Nair Kasturba Medical College and Hospital, Manipal, Karnataka, India 11 The Suitability of Arm Span as a Substitute Measurement for Height HLTH 501 David O. Budain 1 Abstract Many anthropometric equations rely on individual height. Accurate height is not obtainable when various skeletal abnormalities exist. Arm span is proposed as a possible substitute for height. Thirteen subjects' arm span and height were measured. The Pearson R for arm span and height was 0.96 (p<0.05). Regression analysis was used to build and equation predicting height from arm span (Height = 0.8655 x Arm Span + 9.3368). Results of this study show that arm span and height are strongly correlated and arm span can be used as a reliable predictor of height. Introduction In many medical, physiological, and human performance measurements the height of human subjects is used as a predictive and/or classification variable. Equations predicting Body Mass Index, pulmonary function, caloric expenditure, and body fat percentage are just a few of the many equations using height as a predictive variable. 1 However, spinal curvature conditions such as kyphosis, scoliosis, lordosis, and kyphoscoliosis make it difficult to determine the correct height of the individual and thereby necessitating the need to identify a substitute anthropometric measurement.2 The need for an anthropometric measurement to serve as a substitute for height has long been recognized. One possible substitute measurement is arm span, that is - the distance from the left middle finger tip to the right middle fingertip of outstretched arms parallel to the ground. This relationship is notably shown in the drawing Vitruvian Man by Leanardo da Vinci (See Figure 1.). Figure 1. Vitruvian Man by Leonard da Vinci. 2 The purpose of this cross-sectional observational study was to determine if there was a significant relations ship between arm span and height to determine if a arm span could serve as a valid and reliable substitute for height. 3 Methods Sample: A convenience sample of 12 high school seniors and 1 senior high school teacher will be used. Equipment: Task Force Hand Tools 25 foot tape measure. Measurements: Each subject height (with shoes off) will be determined with the subject standing flat footed and with erect posture. The arm span will be taken with arms outstretched, parallel to the ground, from the tip of the right middle finger to the left middle finger across the back. All measurements will be recorded to the nearest inch. Statistical Procedures: Mean, median, standard deviation, minimum and maximum will be calculated for the sample. Data will be examined for outliers. Pearson product moment correlation was used to determine the magnitude and significance of the relationship between arm span and height. Hypotheses tested: Null Hypothesis: (rho) =0 There is no significant relationship between arm span and height. 4 Alternative Hypothesis: (rho) 0 There is a significant relationship between arm span and height. Hypotheses tested at the 0.05 level of significance. If a significant relationship between arm span and height is determined then regression analysis was used to derive an equation to predict height from arm span. Data analysis and graph creation were accomplished using SPSS 20.0 5 Results Arm span and height measurement are shown in Table 1. Table 1. Raw data and correlation parameters Student Name Arm Span Height (Inches) (Inches) A 61.5 63.5 B 70.5 70.0 C 66.5 66.0 D 68.0 65.5 Dr. Barton 67.0 68.0 E 60.5 61.5 F 71.5 72.5 G 77.0 76.5 H 62.0 64.5 I 64.0 65.0 J 64.0 64.5 K 72.5 71.0 L 71.0 72.0 6 Descriptive statistics for arm span and height are shown in Table 2. Table 2. Descriptive Statistics ArmSpan N Valid Height 13 13 0 0 Mean 67.38 67.73 Median 67.00 66.00 64 65 4.946 4.347 24.465 18.901 Skewness .330 .582 Std. Error of Skewness .616 .616 Kurtosis -.625 -.439 Std. Error of Kurtosis 1.191 1.191 Range 17 15 Minimum 61 62 Maximum 77 77 25 63.00 64.50 50 67.00 66.00 75 71.25 71.50 Missing Mode Std. Deviation Variance Percentiles Figure 2. Box plot of arm spam measurements. 7 Figure 3. Box plot of height measurements. Correlation between armspan and height are shown in Table 3. Table 3. Correlations ArmSpan Spearman's rho ArmSpan Correlation Coefficient Sig. (2-tailed) N Height Correlation Coefficient Sig. (2-tailed) N Height 1.000 .963** . .000 13 13 .963** 1.000 .000 . 13 13 **. Correlation is significant at the 0.01 level (2-tailed). 8 Scatterplot of arm span and height is shown in Figure 3. Figure 3. Scatterplot of Arm Span and Height 9 Results of regression analysis is shown in Table 4. Table 4. Regression Analysis r r Std. Error ANOVA table Source Regression Residual Total SS 191.8190 15.5977 207.4167 0.925 0.962 1.249 df 1 10 11 12 1 Height MS 191.8190 1.5598 F 122.98 p-value 6.11E-07 confidence interval 95% upper 21.1675 1.0394 Regression output coefficient std. variables s error Intercept 9.3368 5.3097 n k Dep. Var. Arm Span 0.8655 0.0780 t (df=10) 1.758 11.090 p-value .1092 6.11E07 95% lower -2.4940 0.6916 Discussion In an effort to determine whether or not a significant correlation between arm span and height, measurements were obtained from 13 \"normal\" subjects. The results shown in Table 2 and Figures 2 and 3 indicate there were no outliers and that the data were almost normally distributed. Therefore all data were included in the statistical analyses. Results of the correlation analysis in Table 3 indicate a significant (p<0.05) strong positive (r=0.925) correlation between arm span and height. This strength and direction of the correlation is further demonstrated by the scatterplot shown in Figure 3. 10 The significant correlation between arm span and height allowed for subsequent regression analysis, the results of which are shown in Table 4. The resulting regression equation is as follows: Height = 0.8655 x Arm Span + 9.3368 The results of this study suggest that arm span measurement can be used as a substitute for height in normal subjects. Limitations of this study include the small sample size, narrow range of arm spans and height, and the fact that all subjects were healthy and had no observable spinal curvature. Caution must be exercised in generalizing these results to populations other than described above. 1. Use of anthropometric measures to assess weight loss;George A. Bray,4 M.D., Frank L. Greenway,5 M.D., Mark E. Molitch,6 M.D., William T. Dahms,7 M.D., Richard L. Atkinson,8 M.D., and Kare 2. The use of arm span as a predictor of height: A study of South Indian Women SP Mohanty, S Suresh Babu and N Sreekumaran Nair Kasturba Medical College and Hospital, Manipal, Karnataka, India 11