1 Data management

Load data

vignette("ordinal-knee1")
data(knee, package = "catdata")
dtaK<-knee
head(knee)
##   N Th Age Sex R1 R2 R3 R4
## 1 1  1  28   1  4  4  4  4
## 2 2  1  32   1  4  4  4  4
## 3 3  1  41   1  3  3  3  3
## 4 4  2  21   1  4  3  3  2
## 5 5  2  34   1  4  3  3  2
## 6 6  1  24   1  3  3  3  2

2 Analysis

資料應該分割為Th=1, Th=2(治療前後)，以Th=1(治療前)進行poisson分析。
但試分割資料失敗，所以暫將全部資料一起分析 結果顯示，治療前後資料一起分析時，性別年齡無法顯著預測R4(疼痛程度)

library(base)

knee$Th <- as.factor(knee$Th) 
knee$Sex <- as.factor(knee$Sex)
head(knee)
##   N Th Age Sex R1 R2 R3 R4
## 1 1  1  28   1  4  4  4  4
## 2 2  1  32   1  4  4  4  4
## 3 3  1  41   1  3  3  3  3
## 4 4  2  21   1  4  3  3  2
## 5 5  2  34   1  4  3  3  2
## 6 6  1  24   1  3  3  3  2
m3<-glm(R4~Sex+Age, family = poisson, data=knee)
summary(m3)
## 
## Call:
## glm(formula = R4 ~ Sex + Age, family = poisson, data = knee)
## 
## Deviance Residuals: 
##     Min       1Q   Median       3Q      Max  
## -1.1126  -0.9955  -0.2662   0.6255   1.4507  
## 
## Coefficients:
##              Estimate Std. Error z value Pr(>|z|)    
## (Intercept)  0.975656   0.208788   4.673 2.97e-06 ***
## Sex1         0.003780   0.124446   0.030    0.976    
## Age         -0.002594   0.005857  -0.443    0.658    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for poisson family taken to be 1)
## 
##     Null deviance: 79.313  on 126  degrees of freedom
## Residual deviance: 79.110  on 124  degrees of freedom
## AIC: 426.57
## 
## Number of Fisher Scoring iterations: 4