Table 4. Area indices and optimal cut-off values obtained from ROC curve analysis for height, weight, BMI, 15 specified SAT-layers, 4 Compartments, Total SAT and TBF% of 21 female athletes and 21 female non-athletes.
Personalparameters | Area index1 | P | Optimal cutoff2 | Sensitivity | Specificity | Correctly classifiedcases | |
H0: small | H0: large | [mm] | [%] | [%] | |||
Height (m) | 0.595 | n.s.3 | |||||
Weight (kg) | 0.728 | <0.05 | 66.0 | 95.2 | 9.5 | 52.4% (22 of 42) | |
BMI (kg/m2) | 0.717 | <0.05 | 18.8 | 4.8 | 100.0 | 52.4% (22 of 42) | |
TBF% | 0.757 | – | <0.01 | 30.5 | 100.0 | 47.6 | 73.8% (31 of 42) |
Total SAT9 | 0.866 | – | <0.001 | 83.6 | 61.9 | 100.0 | 81.0% (34 of 42) |
SAT-Top 4 | |||||||
Neck | 0.901 | – | <0.001 | 4.8 | 90.5 | 71.4 | 81.0% (34 of 42) |
Triceps | 0.908 | – | <0.001 | 10.4 | 95.2 | 76.2 | 85.7% (36 of 42) |
Biceps | 0.853 | – | <0.001 | 3.8 | 71.4 | 85.7 | 78.6% (33 of 42) |
Upper back | 0.888 | – | <0.001 | 3.3 | 81.0 | 95.2 | 88.1% (37 of 42) |
Front chest | 0.881 | – | <0.001 | 4.1 | 76.2 | 85.7 | 81.0% (34 of 42) |
Lateral chest | 0.866 | – | <0.001 | 3.3 | 71.4 | 95.2 | 83.3% (35 of 42) |
Upper abdomen | 0.746 | – | <0.01 | 4.7 | 57.1 | 85.7 | 71.4% (30 of 42) |
Lower abdomen | 0.663 | – | n.s. | ||||
Lower back | 0.689 | – | <0.05 | 11.6 | 85.7 | 47.6 | 66.7% (28 of 42) |
Hip | 0.616 | – | n.s. | ||||
Front thigh | 0.859 | – | <0.001 | 9.5 | 90.5 | 66.7 | 78.6% (33 of 42) |
Lateral thigh | 0.824 | – | <0.001 | 9.1 | 71.4 | 95.2 | 83.3% (35 of 42) |
Rear thigh | 0.641 | – | n.s. | ||||
Inner thigh | 0.842 | – | <0.001 | 9.6 | 71.4 | 85.7 | 78.6% (33 of 42) |
Calf | 0.825 | – | <0.001 | 5.6 | 85.7 | 66.7 | 76.2% (32 of 42) |
Compartments | |||||||
Arms5 | 0.923 | – | <0.001 | 15.9 | 100.0 | 76.2 | 88.1% (37 of 42) |
Trunk6 | 0.909 | – | <0.001 | 13.9 | 76.2 | 95.2 | 85.7% (36 of 42) |
Abdomen7 | 0.707 | – | <0.05 | 34.9 | 71.4 | 66.7 | 69.0% (29 of 42) |
Legs8 | 0.854 | – | <0.001 | 44.5 | 90.5 | 66.7 | 78.6% (33 of 42) |
There are two possible hypotheses (H0): that either small/large values provide stronger evidence for positivity.
Optimal cut-off value estimated by Youden-Index (Youden. 1950).
Not significant (p>0.05).
SAT thickness of 15 body sites in mm.
Body sites biceps+triceps.
Body sites front thigh+lateral thigh+rear thigh+inner thigh+calf.
Body sites upper abdomen+lower abdomen+lower back+hip.
Body sites neck+upper back+lateral chest+front chest.
Body sites 1–15.