1 1 1 1.1.............................................. 1 1.2............................................ 1 1.2.1..................................... 1 1.2.2..................................... 2 1.2.3..................................... 4 1.2.4......................... 4 1.2.5...................................... 5 1.2.6.......................................... 5 1.2.7................................... 6 1.3 ICC...................................... 8 1.3.1 Case1........................................... 9 1.3.2 Case2........................................... 10 1.3.3 Case3........................................... 12 1.4 ICC......................................... 15 1.4.1 ICC11 ICC21 ICC31.............. 15 1.4.2 Case1 Case2.................................. 15 1.4.3 ICC...................................... 15 1.4.4............................. 16 1.4.5..................................... 16 1.4.6...................................... 17 1.5............................... 18 1.5.1....................................... 18 1.5.2...................................... 20 1.5.3...................... 20 1.6......................................... 23 1.6.1........................... 23 1.6.2....................................... 25 1.6.3...................................... 26

1 1 1.1 Intraclass correlation coefficients ICC a ICC ICC ICC [1] [2] Fisher [3] ICC Shrout [4] ICC 3 6 Shrout [5][6][7][8] [9] Shrout ICC 3 Case1Case2Case3 6 ICCn1 ICCn k n 123 ICC 6 Bartko [10] ICC One-way ClassificationTwo-way Random ModelTwo-way Mixed Model 3 Case1Case2Case3 Bartko [11] ICC1 ICC2 Two-wayWiner s anchor poin ICC 1.2 1.2.1 ICC [ ] 1. 2. 3. 1. 2. 2. 3. ICC a item characteristic curve) ICC ICC

2 1 1.2.2 classical test theory 1950 item response theoryirt 1950 1.2.2.1 test-retest method test-retest reliability 0 b 1.2.2.2 parallel test method parallel form reliability strongly parallel measurement weakly parallel measurement Cronbach α [1.2.2.4 ] ICC31 1.2.2.3 split-half method 2 2 2 2 odd-even method Spearman-Brown Kuder-Richerdson 20KR-20 Kuder-Richerdson 21KR-21 2 0-1 KR-20 KR-21 1.2.2.4 α Cronbach s coefficient alpha Cronbach α α c A n n b c

1.2. 3 ICC3k α X i T i E i 2 1) 0 E(E i ) 0 (2.1) µ(x) E(X i ) E(T i + E i ) E(T i )+E(E i ) µ(t ) (2.2) 2) Cov(T E) ρ(t E) 0 (2.3) σ 2 (X) Var(T )+Var(E)+2Cov(T,E) Var(T )+Var(E) (2.4) ρ ρ(x) = Var(T ) Var(X) (2.5) Spearman-Brown 2 2 X 1 X 2 T 1 T 2 E 1 E 2 X 1 X 2 X X 1 + X 2 /2 T T 1 + T 2 /2 T 1 T 2 Var(X 1 )=Var(X 2 ) (2.6) Var(T 1 )=Var(T 2 ) (2.7) X 1 (= (X 11 X 12 X 1i ))X 2 (= (X 21 X 22 X 2i )) CovX 1 X 2 = E X 1i µ(x 1 ) X 2i µ(x 2 ) = E T µ(t )+E 1i T µ(t )+E2i = E T µ(t ) 2 + E T µ(t ) E2i + E E1i T µ(t ) + E E1i E 2i = Var(T ) (2.8) ρ(x 1,X 2 ) Cov(X 1,X 2 ) Var(X 1 )Var(X 2 ) = Cov(X 1,X 2 ) 2Var(X 1 ) (2.9)

4 1 2.5 ρ(x) = Var(T ) Var(X) Var(T 1 + T 2 ) Var(X 1 + X 2 ) = Var(T 1)+Var(T 2 )+2Cov(T 1,T 2 ) Var(X 1 )+Var(X 2 )+2Cov(X 1,X 2 ) = 2Var(T 1)+2Cov(T 1,T 2 ) 2Var(X 1 )+2Cov(X 1,X 2 ) = 4Var(T 1 ) 2Var(X 1 )[1 + ρ(x 1,X 2 )] = 2 2Cov(X 1,X 2 ) 2Var(X 1 )[1 + ρ(x 1,X 2 )] = 2ρ(X 1,X 2 ) 1+ρ(X 1,X 2 ) = 2ρ(X 1 ) 1+ρ(X 1 ) (2.10) 2.10 Spearman-Brown 1.2.3 generalizability theory [2] ICC G generalizability study D decision study A A G ICC11 ICC21 ICC 31 G A A D D ICC1k ICC2k ICC3k D G A D 1.2.4 fixed model random effect model mixed model

1.2. 5 d e 2 f 2 A B C 1.2.5 ICC i j x ij i 1 2 nj 1 2 m x 11 x 21 x nm x i x 2 x ij x 2 x i x+x ij x i 2 x ij x i x+x ij x i (2.11) x ij x 2 x i x 2 +x ij x i 2 +2 x i xx ij x i nx mx nx nx mx nx mx x ij x 2 x i x 2 + x ij x i 2 + 2 x i xx ij x i (2.12) i=1 j=1 i=1 i=1 j=1 i=1 j=1 2.11 i i 0 g 2.12 3 0 nx i=1 j=1 mx nx x ij x 2 x i x 2 + i=1 nx i=1 j=1 [ 1.2.2 ] mx x ij x i 2 (2.13) 1.2.6 Case2 Case3 1.1 d e f g x i x x ij i 0

6 1 1.1 1.1 a. A B A B b. c. A B 1 A 2 B BA AB 1.2.7 ICC ICC 1.1 ICC 1.1 [4] df M S n 1 BMS kσt 2 + σ2 W kσt 2 + σ2 I + σ2 E kσt 2 + σ2 E n(k 1) WMS σw 2 σj 2 + σ2 I + σ2 E θj 2 + fσ2 I + σ2 E (k 1) JMS nσj 2 + σ2 I + σ2 E nθj 2 + fσ2 I + σ2 E (n 1)(k 1) EMS σi 2 + σ2 E fσi 2 + σ2 E P f k/(k 1) θ2 J a 2 i /k 1 n k 1.1 2 MS BMSWMSJMSEMS ICC 3

1.2. 7 f θ 1.1 [ ] h Bartko [10] 1.1 h EMS θ J 2 σ2 J σ2 I f

8 1 1.3 ICC ICC Case123 2 6 6 Case 1.2 1.2 a. Case1 1.2 b. Case2 2 a. ICC Case1 ICC Case2b. c. ICC Case3

1.3. ICC 9 1.2 c. Case3 Case2 i Case2 Case3 A B C Case3 3 3 Case3 A a b a 85 b 75 a b 10 Case3 Case2 1.3.1 Case1 1 Intra-rater reliability Case1 1.1 ICC11 ICC1k 1 r k x ij x ij µ + T i + W ij (3.14) µ T i i W ij ij i12 rj12 k σx 2 σ2 T + σw 2 (3.15) σt 2 + σ2 W σ2 T ρ σ2 T σ 2 T + σ2 W (3.16) 3.16 ICC11 1.1 BMS WMS Case1 BMSWMS BMS kσ 2 T + σ 2 W (3.17) WMS σ 2 W (3.18) 3.17 3.18 σt 2 BMS WMS] /k 3.16 i BMS WMS ICC1 /k BMS WMS /k + WMS

10 1 BMS WMS BMS WMS + kwms BMS WMS ICC1 BMS +k 1WMS (3.19) A ICC11 A 1 r m x 11 ( x 111 + x 112 ++ x 11 m )/m 3.16 ρ σ2 T σt 2 + σ2 W /k (3.20) 3.20 3.16 2 σw 2 /k k 3.20 ICC1k 3.17 3.18 3.20 BMS WMS ICC1 k BMS (3.21) 1.4.4 [4] ICC11 100(1 α) j F 0 F U BMS/WMS F 0 F 1 α/2n(k 1)n 1 F L = F 0 /F 1 α/2n 1n(k 1) F L 1 F L +(k 1) ρ ICC(1,1) F U 1 (3.22) F U +(k 1) F 1 α/2nk 1 n 1 α0.05 df 1=nk 1 df 2 n 1 1 0.05/2 0.975 F ICC1k 100(1 α) 1 1 F L ρ ICC(1,k) 1 1 F U (3.23) 1.3.2 Case2 Inter-rater reliability Case2 Case3 [1.2.4 ] Case2 ICC21 ICC2 k k n Case1 x ij µ + T i + J j + I ij + E ij (3.24) j 1 1 [13]

1.3. ICC 11 µ T i i J j j I ij E ij i j i 1 2 nj 1 2 k 3.14 J j I ij J j I ij A a b c B a b c C a b c C Case1 σt 2 σ 2 T ρ σt 2 + σ2 J + σ2 I + σ2 E (3.25) ICC21 1.1 BMS kσ 2 T + σ 2 I + σ 2 E (3.26) JMS nσ 2 J + σ 2 I + σ 2 E (3.27) EMS σ 2 I + σ 2 E (3.28) σ 2 T σ 2 J BMS EMS /k (3.29) JMS EMS /n (3.30) 3.25 BMS EMS /k ICC2 1 BMS EMS /k +JMS EMS /n + EMS BMS EMS BMS EMS + k JMS EMS /n + kems BMS EMS ICC2 1 BMS +k 1EMS + k JMS EMS /n (3.31) Case1 A k n m 3.25 σ 2 T ICC2 k σt 2 +σ2 J + σ2 I + σ2 E /k BMS EMS /k BMS EMS /k +JMS EMS /n + EMS /k BMS EMS ICC2 k BMS +JMS EMS /n (3.32) ICC21 100(1 α) ˆρ ICC2 1

12 1 F J ν F JMS/EMS (k 1)(n 1)kˆρF J + n[1 + (k 1)ˆρ] kˆρ 2 (n 1)k 2 ˆρ 2 FJ 2 +n[1 + (k 1)ˆρ] kˆρ 2 F 1 α/2 (n 1) ν F = F 1 α/2 ν (n 1) n(bms F EMS) F [kjms +(kn k n)ems]+nbms ρ n(f BMS EMS) ICC(2,1) kjms +(kn k n)ems + nf BMS (3.33) ICC2k 100(1 α) ρ L = ρ U = kρ L 1+(k 1)ρ L kρ U 1+(k 1)ρ U (3.34) (3.35) ρ ρ L ρ ICC(2,k) ρ U (3.36) 3.34 3.35 L ρ U 3.33 ICC21 1.3.3 Case3 ICC21 ICC2k [1.2.4 ] ICC31 ICC3k 3.24 ρ σ2 T σ2 I /k 1 σt 2 + σ2 I + σ2 E 3.37 σt 2 σ2 I /k 1 σj 2 k 1.3 Case1Case2 1.1 BMSEMS (3.37) σt 2 (BMS σ E 2 )/k (3.38) σi 2 (k 1)(EMS σe)/k 2 (3.39) 3.37 ICC31 ICC3 1 (BMS σ2 E )/k (k 1)(EMS σ2 E )/k1/k 1 (BMS σ E 2 )/k +(k 1)(EMS σ2 E )/k + σ2 E BMS σ2 E EMS + σ2 E BMS σe 2 +(k 1)(EMS σ2 E )+k σ2 E BMS EMS ICC3 1 BMS +k 1EMS (3.40) k σi 2 0

1.3. ICC 13 Case1Case2 m ICC3k 3.37 σ2 T ρ σt 2 + σ2 E /k (3.41) 1.1 /k 1 σ 2 I BMS kσ 2 T + σ 2 E (3.42) EMS σ 2 E (3.43) a a b b a a 2 2 b 2 b + 1.3 3.41 BMS EMS ICC3 k BMS ICC31 (3.44) F 0 BMS/EMS (3.45) F L F 0 /F 1 α/2 (n 1),(n 1)(k 1) (3.46) F U = F 0 F 1 α/2 (n 1)(k 1),(n 1) (3.47)

14 1 F L 1 F L +(k 1) ρ ICC(3,1) F U 1 F U +(k 1) ICC3k (3.48) 1 1 F L ρ ICC(3,k) 1 1 F U (3.49) k n [ ] I ij V ( kx j=1 I ij ) kv (I ij )+kk 1COV (I ij I i0 j) (3.50) 3.50 0 V (I ij ) V (I i0 j) i i 0 COV (I ij I i0 j)=v (I ij ) COV (I ij I i0 j)= σi 2 /k 1 3.37 Bartko [10] 3.37 σ2 T ρ σt 2 + σ2 I + σ2 E (3.51) 1.1 EMS EMS σ I 2 + σ2 E EMS BMS 3.51 ICC3 1 BMS EMS + σ2 I BMS +k 1EMS + σi 2 (3.52) ICC3 1 BMS EMS = BMS +k 1EMS (3.53) 3.40 EMS σi 2 EMS σ2 I ICC3 1 BMS BMS + kems (3.54) ICC31 3.53 ICC3 1 3.54

1.4. ICC 15 1.4 ICC 1.4.1 ICC11 ICC21 ICC31 ICC11 ICC21 ICC31 ICC11 ICC21 ICC31 [8] 0 ICC11 ICC21 ICC Case Case 1.4.2 Case1 Case2 Case1 [1.3.1 ] [4] [8] ICC(11) ICC21 ICC11 BMS BMS kσt 2 + σw 2 3.17 ICC21 BMS BMS kσt 2 + σi 2 + σe 2 3.26 σw 2 1.1 σw 2 σ2 J + σi 2 + σe 2 (4.55) 3.30 3.28 ICC21 BMS ICC11 σt 2 σ 2 T BMS ([JMS EMS]/n + EMS) = BMS EMS (JMS EMS/n) (4.56) (4.55) (4.56) ICC11 σt 2 ICC(1 1) σ2 J σt 2 + σ2 J + σ2 I + σ2 E σ2 J (4.57) ICC21 σj 2 (= (JMS EMS)/n) ICC11 l ICC1 1 1.4.3 ICC ICC 1.2 Landis [12] Kappa ICC l

16 1 [6] [9] ICC 0.7 ICC ICC 1.2 [12] ICC 0.0 0.20 slight 0.21 0.40 fair 0.41 0.60 moderate 0.61 0.80 substantial 0.81 1.00 almost perfect 1.4.4 D ICCn kn =1, 2, 3 ICC11 0.7 0.9 m k ρ 11 ρ 2 ρ 2 1 ρ 1 (4.58) ρ 1 ρ 2 0.9 k 1 0.7 3.8571 (4.59) 0.7 1 0.9 4 4 Eliasziw [13] 1.4.5 ICC Eliasziw [13] n H 0 ρ0 0 1.2 H 0 ρ 0.6 sabstantial m Spearman-Brown [1.2.2 ] n

1.4. ICC 17 1.4.6 SEM ICC Case SEM SEM q SEM σtotal 2 (1 ρ) (4.60) σtotal 2 ICC11 SEM 1 SEM1 2 σ2 W WMS (4.61) ICC12 ICC13 SEM 2 SEM 3 SEM2 2 = σj 2 + σi 2 + σe 2 JMS EMS /n + EMS (4.62) SEM3 2 = σi 2 + σe 2 EMS (4.63) SEM Stratford [14] SEM " # WMS WMS, (4.64) χ 2 χ 2 α/2,df1 α/2,df χ 2 α,df df α/2 χ2 2 SEM SEM A SEM B H 0 2 df A df B F F df Adf B SEMA 2 /SEM B 2

18 1 1.5 ICC ICC ICC 1.5.1 Case2 T i N(0, σt 2 ),J jn(0, σj 2) I ijn(0, σi 2),E ijn(0, σe 2 ) 3.14 3.24 A k n k ICC ICC21 J j T i X (x 11 x 12 x ij )(i =1, 2,,nj =1, 2,,k) x ij µ + T i + J j +(TJ) ij + e ij (5.65) (TJ) ij µ 0XN(µ X σx 2 ) X x n1 x n2 x nk k k 5 10 20 i i 10 30 r 0.0 0.5 0.9 r k k R R AA t A n k DN0 1 A t D X ICC21 1,000 1.4 n k ricc 0 Fisher [3]Shrout [4] o ICC ICC 0 ICC 1 o Fisher + 1 n0 log 2 n 0 1 + 1 2n 0 1 n0 n +2

1.5. 19 k=5 n20 r0 n20 r0.5 n20 r0.9 n20 r0 n20 r0.5 n20 r0.9 n10 r0 n10 r0.5 n10 r0.9 n20 r0 n20 r0.5 n20 r0.9 n30 r0 n30 r0.5 n30 r0.9 k=10 n10 r0 n10 r0.5 n10 r0.9 n30 r0 n30 r0.5 n30 r0.9 k=20 n10 r0 n10 r0.5 n10 r0.9 n30 r0 n30 r0.5 n30 r0.9 1.4 ICC21

20 1 a. r0 95 b. r0.5 95 c. r0.9 95 1.5 ICC21 95 1.5.2 1.3 1.5.1 1.5.1 X A t X ICC21 95 k 5n 20 r 00.50.9 r 100 1.5 Eliasziw [13] 1 1.5.3 ICC ICC 1.6 ICC11 ICC21 ICC31 1.6-a. AD ad ICC 1 b. A d +2 ICC A d +2 1.6-c.

1.5. 21 k 4 n 4 A B C D ICC(1,1)= 1.0000 a 1 1 1 1 1 ICC(2,1)= 1.0000 b 2 2 2 2 2 ICC(3,1)= 1.0000 c 3 3 3 3 3 SEM1= d 4 4 4 4 4 SEM2= 2.5 2.5 2.5 2.5 2.5 SEM3= A B C D ICC(1,1)= 0.9684 a 1 1 1 1 1 ICC(2,1)= 0.9684 b 2 2 2 2 2 ICC(3,1)= 0.9684 c 3 3 3 3 3 SEM1= d 5 4 4 4 4.25 SEM2= 2.75 2.5 2.5 2.5 2.5625 SEM3= A B C D ICC(1,1)= 0.9684 A B C D ICC(1,1)= 0.9684 a 11 11 11 11 11 ICC(2,1)= 0.9684 a 110 110 110 110 110 ICC(2,1)= 0.9684 b 12 12 12 12 12 ICC(3,1)= 0.9684 b 120 120 120 120 120 ICC(3,1)= 0.9684 c 13 13 13 13 13 SEM1= c 130 130 130 130 130 SEM1= d 15 14 14 14 14.25 SEM2= d 150 140 140 140 142.5 SEM2= 12.75 12.5 12.5 12.5 12.563 SEM3= 127.5 125 125 125 125.63 SEM3= A B C D ICC(1,1)= 0.4286 A B C D ICC(1,1)= 0.4286 a 11 10 9 8 9.5 ICC(2,1)= 0.5000 a 110 100 90 80 95 ICC(2,1)= 0.5000 b 12 11 10 9 10.5 ICC(3,1)= 1.0000 b 120 110 100 90 105 ICC(3,1)= 1.0000 c 13 12 11 10 11.5 SEM1= c 130 120 110 100 115 SEM1= d 14 13 12 11 12.5 SEM2= d 140 130 120 110 125 SEM2= 12.5 11.5 10.5 9.5 11 SEM3= 125 115 105 95 110 SEM3= A B C D ICC(1,1)= - A B C D ICC(1,1)= -0.2698 a 110 90 70 50 80 ICC(2,1)= 0.2000 a 110 90 70 50 80 ICC(2,1)= 0.0361 b 120 100 80 60 90 ICC(3,1)= 1.0000 b 110 90 70 50 80 ICC(3,1)= 1.0000 c 130 110 90 70 100 SEM1= c 110 90 70 50 80 SEM1= d 140 120 100 80 110 SEM2= d 120 100 80 60 90 SEM2= 125 105 85 65 95 SEM3= 112.5 92.5 72.5 52.5 82.5 SEM3= A B C D ICC(1,1)= -0.3169 A B C D ICC(1,1)= -0.1111 a 110 90 70 50 80 ICC(2,1)= 0.0093 a 110 111 112 113 111.5 ICC(2,1)= 0.1304 b 110 90 70 50 80 ICC(3,1)= 1.0000 b 110 111 112 113 111.5 ICC(3,1)= 1.0000 c 110 90 70 50 80 SEM1= c 110 111 112 113 111.5 SEM1= d 115 95 75 55 85 SEM2= d 111 112 113 114 112.5 SEM2= 111.25 91.25 71.25 51.25 81.25 SEM3= 110.25 111.25 112.25 113.25 111.75 SEM3= 1.6 ICC b. ICC c. 10 1.6-d. ICC a. b. A d +2 A d +20 ICC ICC SEM 1.6 b. c. d. SEM 1.6-e. ICC ICC31 1 e. 10 f. ICC

22 1 g. ICC11 ICC21 ICC31 ICC11ICC21 ICC31 1.6-h. d +10 ICC 11 ICC21 AD ICC31 1.6-i. d +5 d 1.6-h. d ICC11 ICC21 h. i. h. i. j. ICC11 ICC21 ICC SEM ICC ICC ICC 0 ICC

1.6. 23 1.6 110 10 AD 4 A B C D A B C D 1 126 122 131 125 1 6 5 4 7 2 137 143 141 141 2 6 8 6 8 3 113 119 115 105 3 15 14 12 15 4 153 143 135 144 4 4 4 1 0 5 146 157 150 149 5 11 10 11 11 6 161 157 160 160 6 15 14 15 18 7 110 109 105 113 7 9 12 9 12 8 145 151 152 156 8 5 2 4 5 9 126 141 132 122 9 14 12 14 16 10 114 126 130 125 10 9 8 7 8 133.1 136.8 135.1 134.0 9.4 8.9 8.3 10.0 17.95 16.79 16.75 18.63 4.20 4.23 4.67 5.50 1.6.1 1.3 df M S 10319.5 9 1146.6 BMS 40.4 76.1 3 25.4 JMS 0.9 765.9 27 28.4 EMS 740.6 9 82.3 BMS 47.5 15.7 3 5.2 JMS 3.02 46.8 27 1.7 EMS p 0.01 p 0.05 BMSJMSEMS ICC21

24 1 BMS EMS ICC2 1 BMS +k 1EMS + k JMS EMS /n 1146.6 28.4 = 1146.6+(4 1) 28.4+4 25.4 28.4 /10 1118.2 1230.6 ICC2 1 0.909 95 [1.3.2 ] ˆρ F J ν ICC2 1 0.909 JMS/EMS 0.89 (k 1)(n 1)kˆρF J + n[1 + (k 1)ˆρ] kˆρ 2 (n 1)k 2 ˆρ 2 FJ 2 +n[1 + (k 1)ˆρ] kˆρ 2 F (4 1) (10 1) 4 0.909 0.89 + 10 [1 + (4 1) 0.909] 4 0.9092 = (10 1) 4 2 0.909 2 0.89 2 +10[1 + (4 1) 0.909] 4 0.909 2 29.95 F 1 α/2 (n 1) ν = F (0.975929.95) 3.561 F = F 1 α/2 ν (n 1) F (0.97529.959) 2.576 n(bms F EMS) F [kjms +(kn k n)ems]+nbms ρ n(f BMS EMS) ICC(2,1) kjms +(kn k n)ems + nf BMS 10 (1146.6 3.561 28.4) 3.561 [4 25.4+(410 4 10) 28.4]+101146.6 ρ ICC(2,1) 10 (2.576 1146.6 28.4) 4 25.4+(410 4 10) 28.4+102.576 1146.6 0.7232 ρ ICC(2,1) 0.963 q q SEM = JMS EMS /n + EMS25.4 28.4 /10 + 28.45.30 95 0.7232 ρ0 ρ 0.7 5 ICC2 1 0.906 0.776 ρ ICC(2,1) 0.973 SEM 1.43 H 0 ρ 0.7 5 SEM

1.6. 25 1.6.2 A B C D A B C D a. c. d. 1.7 1.7 2 box plot a. b.c. d. a. b. SEM SEM

26 1 1.6.3 D ICC21 0.909 ICC21 0.906 0.900.99 0.01 1.4.4 1.8 12 10 8 6 4 2 0 0.9 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99 0.98 5 1.8 ICC ρ 0.98 5 ρ0.99 10

1.6. 27 036-8564 66-1 FAX 0172 39 5981 E mailpteiki@cc.hirosaki-u.ac.jp URLhttp://www.hs.hirosaki-u.ac.jp/pt/eiki/

29 [1] 1995. [2] 1992. [3] Fisher RA 1975p167-197. [4] Shrout PE,Fleiss JLIntraclass Correlations:Uses in Assessing Rater ReliabilityPsychol Bull 86 420-4281979. [5] James WY,Connie LB,et.al.:Reliability of goniometric measurements and visual estimates of ankle joint active range of motion obtained in a clinical setting.arch Phys Med Rehabil 74:1113-1118,1993. [6] Reliability. 41:945-952,1993. [7]. 12:113-120,1997. [8] Gabrielle R,Maria S:Reliability of assessment tools in rehabilitation:an illustration of appropriate statistical analyses.clinical Rehabilitation 12:187-199,1998. [9] Portney LG,Watkins MP:Foundations of clinical research-applications to practice-,appleton Lange,USA,1993,p505-516. [10] Bartko JJThe intraclass correlation coefficient as a measure of reliability.psychological Reports 19 3-111966. [11] Bartko JJOn various intraclass correlation reliability coefficients.psychological Bulletin 83762-765 1976. [12] Landis JR,Koch GG:The measurement of observer agreement for categorical data.biometrics. 33,159-174,1977. [13] Eliasziw M,Young SL,et al.:statistical methodology for the concurrent assessment of interrater and intrarater reliability:using goniometric measurements as an example.physical Therapy 74:777-788,1994. [14] Stratford PW,Goldsmith CH:Use of the error as a reliability index of interest:an applied eample using elbow flexor strength data.physical Therapy 77:745-750,1997.