ࡱ> Root EntryRoot EntrypGT(@ FileHeader\DocInfoJBodyText K(K( HwpSummaryInformation.BPrvImage SPrvTextDocOptions pGT(pGT(Scripts pGT(pGT(JScriptVersion c DefaultJScript`_LinkDocd  !"#$%&'()*+,-./0123456789:;<=>?@ACDEFGHIKLMNOPQRSTUVWXYZ[]^_abefghijklnopqrstuvwxyz{|}~ ] ȩ : Performance of Intra-class Correlation Coefficient for bilateral cases in Medical Research Ŭ :  P [DYP] | : 9 4|(|) $ 5:00-6:00 nj :  6218 Reliability is an important concept in medical practice about the study of  consistency or  repeatability of measurements. Intra-class Correlation Coefficients (ICC) are commonly used in medicine as reliability coefficients. In medical research, there are rare the study of ICC that the fundamental unit is not the patient but the side. Most of such cases are considered to be independent despite their correlation. Therefore, intra-class correlations coefficients for cases such as the values of sides are highly correlated are considered assuming the dependence between bilateral cases. In this talk, I introduce ICC and compare performance of ICCs through simulation in situations that: (1) be considered one side among both sides, (2) a randomly selected side under the independent cases, and (3) incorporating sides under the dependent casGIF89aɻxxxkkk]]]PPPCCC555((( d`LH00`Hذ̘`H0|pdXL@`0H `H0`dHL`H0xp`XH@0d LȰ``HH00dLؘȀ`H0xp`XH@d0L `H0d`LH`H0|pdXL@0`$H`dHL00`HذȘdL0xp`XH@0` HdH0ȳ``HHdL0xp`XH@`0H Ը̰ĠऀؘpȈXxHp@h@d8`8X0xT0pH dxddxd`d`p`|``x`p``x`!, H*\ȰÇ#JHŋ3jȱǏ CIɓ(S\ɲ˗0cʜI͛8sɳϟ@ JѣH*]ʴӧPJJիXjʵׯ`ÊKٳhӪ]˶۷<3e(w?t;._zc//E̸ǐ#KL2#Ќ`sK_Q}tҪa=5֦s뎽ܴ}๳qY+=vq3_=9sʱ=en&~ӫ{ruϿ7 `蠃>x`&x!b](`"hB-@.25 #7x㎍UX :h$9Hd:*#(VieN !$FFHaf)&hlp)tֹ]Z8bBRb*4hLB&J"HY 颀2:i^!ʩY(x>)^*~ +EbI[P.'IMhR:괠Jicj-MNڬ^lJ 8Qj!\[,gRXv+ ɒ+.[vHrn3 ޶J-LB fW,wiL *ɻҳ-Ljޞ§K~Ҵ/l%ܳ//|Ϋ9ݴ"/-9.)|)V6>9貹z˄<eۊY&8 ,y+EN(Ly=E\&| r%kw5σң7g!~8dw-! .ƿ5JyF!F`a|f\3eP_X<9σP.-r䍨ڣ􈸥:Kŏb J~Q6t84UO2~ @JЂMBІ:D'JъZͨF7т;`a``a`0x  ` l x]ajou2014D 9 22| Ɣ| $ 1:29:43ajou8, 5, 8, 1453 WIN32LEWindows_7@P'@K(@VkAfmHMOM ZjTPQ+`'az?@azA J<ޣD7lM6"(ydf~ޛeRACM]tV5{*'IjD/1V X `U—3y9F[Jjaڋ+1>0äciwJ-w>^*=sXbyGi«۝.܎g9ZG|:~__1KFgRmPRXƒ!G=stG_/CT*]Sc1K8E R0BjkĶZlBEjf{jԲW7 ߲~&"~jԠը(5?7Section0mSMFjӗWhE\ MQn&eP9e;!6Z]8@Ny9H-}M"׾I;}p{wG dߙH"|< GF3eg"%f*s3!iTf˜uB2S b;s9c9 kAǿ3y&M Rڟ{\DEրśbj&Ad=kD (^4O5XCAl֤vgTz#y3u~? l ˉ`$a]tlr3kvN(EoaT3g6Śe͟N6Zd{ڟXLaU,"S]H|k{˄Lp7($lpVeMb% (3`:[ƌ+]~;xG0|pp^ z5O#̏?|~FZYgVEיCގ`[Bb(lU*2gUXUBjj:pHWP Document Fileu@DO 6C,H,#GB!k0~ 6*7=M*Z:ҙ#?*?ӿT@Y~X+mj3'm[v|:j)%C$Mϼ5#cd߈ In this talk, I introduce ICC and compare performance of ICCs through simulation in situations that: (1) be considered one side among both sides, (2) a randomly selected side under the independent cases, and (3) incorporating sides under the dependent casVOhe3;MSR0rH06m5)Kbl2t( {C "zBOi0I/S9IC}f;޼}#4v#1Ni6M(_ ijTyZM~ގ k ]d<{pܚ7CUՏl7 |x ?@9$9nAX'G>Sƨ^/_y5G֍.;fcWo{j}F 3v^w P E8/u |}(ӆiPd pq^UrWQRz[ٷ2Y ʊla.[9[9~C٢T!PVq' FM&ۖmu~(]pܣhyiMBkpվx.]{x~9YWmSf\N aHЏg_‡K۫k aycW./otx[6B'FԽڡݡ|ǀ/_6v~5vͻ?CF&`?LފS~D?{CX'&mN|nޮ.4{^O2ެٜcNXrA3?\dj`IYTSܭ\Ƥ:Ȍ>5xME2\w>ُl[1{xoٝ^b{cI^C+F41Kx GpO)u Yde0{m}l,S݋gD"|"{eԬn>W'⪢%ձաʪpѪĵ>(ϲذ£BŲ('ĖɌX;qKܶd8݄+˱(-lٯt%N.>c+'g;3o[T{u4lG2|w[rw̪/+tUߥ!VzW5KN?P3i@7~M c