untitled

Size: px

Start display at page:

Download "untitled"

ゆずさふしはら
5 years ago
Views:

1 R M1 M2 1 8 Takahashi, Y., Roberts, B.W., Yamagata, S., & Kijima, N. (in press). Personality traits show differential relations with anxiety and depression in a non-clinical sample. Psychologia: An International Journal of Psychological Sciences. 2 (BIS BAS) 2 ( ), 2 ( ) ( ) 2

2 BIS BAS BAS,,, 2 (Gray, 1970) (Behavioral Inhibition System: BIS), (Behavioral Activation/Approach System: BAS), BIS BAS 3 BIS, BAS (e.g. Kimbrel et al., 2007), BAS (Johnson et al., 2003), (Coplan et al., 2006)? (BIS) (BAS) 4

3 BIS BAS 2, BAS ( ), Time 1 BIS/BAS (Carver & White, 1994;, 2007) (Zung, 1965) (Spielberger et al., 1970) (2 ) Time 2 (Zung, 1965) (Spielberger et al., 1970) 6

4 (2007) 7 depanx <- read.csv( depanx.csv ) # head(depanx) # bis 1~7 (4, 7 ) bas 1~13 (4, 13 ) a01_t1 ~ a20_t1 1 (4, 20 ) d01_t1 ~ d20_t1 1 (4, 20 ) a01_t2 ~ a20_t2 2 (4, 20 ) d01_t2 ~ d20_t2 2 (4, 20 ) 8

5 BIS 2 (4, 5) BIS BAS attach(depanx) # depanx$bis1r <- 5 - bis1 # depanx$bis6r <- 5 - bis6 # detach(depanx) # depanx attach(depanx) # depanx$bis_t1 <- bis1r + bis2 + bis3 + bis4 + bis5 + bis6r + bis7 #BIS ( ) depanx$bas_t1 <- bas1 + bas2 + bas3 + bas4 + bas5 + bas6 + bas7 + bas8 + bas9 + bas10 + bas11 + bas12 + bas13 #BAS 9 rowsums (), colsum () depanx$anx_t1 <- rowsums(depanx [c(24:43)]) # 1 depanx$dep_t1 <- rowsums(depanx [c(44:63)]) # 1 depanx$anx_t2 <- rowsums(depanx [c(64:83)]) # 2 depanx$dep_t2 <- rowsums(depanx [c(84:103)]) # 2 detach(depanx) attach(depanx) depanx1 <- cbind(sex, age, bis_t1, bas_t1, anx_t1, dep_t1, anx_t2, dep_t2) # 6 depanx1 <- data.frame(depanx1) # attach(depanx1) # library(psych) #psych describe(depanx1) # 10

6 ( 1 ) alpha(depanx [c(24:43)]) #1 alpha(depanx [c(44:63)]) # 1, BIS =0.80, BAS =0.81,1 =0.90, 1 =0.81,2=0.91 2= BIS BAS( ) round(cor(depanx1[,3:8], use="pairwise.complete.obs"), 2) 2, 揃 complete.obs 12

7 pairs (depanx1 [,3:8]), BIS 1, BAS cor.test () X Y Z,, BAS X <- cor(depanx1 [,c("bas_t1", "dep_t1", "anx_t1")], use="pairwise.complete.obs") #3 library(psych) #psych partial.r(x, c(1, 2), 3) # X

8 Y <- cor(depanx1 [,c("bas_t1", "dep_t2", "anx_t2")], use="pairwise.complete.obs") library(psych) partial.r(y, c(1, 2), 3) 12, 5% 15 corpcor cor to pcor X <- cor(depanx1 [,c("bas_t1", "dep_t1", "anx_t1")], use="pairwise.complete.obs") #3 install.packages("corpcor") #corpcor library(corpcor) #corpcor X.pcor <- cor2pcor(x) rownames(x.pcor) <- rownames(x) # colnames(x.pcor) <- colnames(x) # X.pcor 16

9 1 BIS BAS 17 y ) = a ( ) + b ( ) x + e ( ) y^ ( ) = a ( ) + b ( ) x ( - ) 2 reg.anx_t1 <- lm(anx_t1 ~ bis_t1, data = depanx1) summary(reg.anx_t1) 18

10 ( ) (R 2 ) F p, = BIS BIS , BIS ( ) X Y =X Y (Y /X ) X Y ( 0, 1), SDx <- sd (bis_t1, na.rm = TRUE) SDy <- sd (anx_t1, na.rm = TRUE) R <- cor (anx_t1, bis_t1, use="pairwise.complete.obs ) B <- R * ( SDy / SDx ) B

summary(reg.bis_t1), BIS =14.86 0.16 0BIS 14.

11 ( ) 1 BIS reg.bis_t1 <- lm(bis_t1 ~ anx_t1, data = depanx1) summary(reg.bis_t1), BIS = BIS 14.86, 1 BIS ( ),, z.depanx1 <- scale(depanx1) # z.depanx1 <- data.frame(z.depanx1) # attach(z.depanx1) # reg.z.anx_t1 <- lm(anx_t1 ~ bis_t1, data = z.depanx1) summary(reg.z.anx_t1), 2 22

frame (bis_t1 = seq (7, 28, 1)) # range reg.anx_t1_95ci <- predict (reg.

12 plot(anx_t1 ~ bis_t1, data = depanx1, xlab="bis", ylab=" ") # abline (reg.anx_t1) # X Y 23, 100, 95 confint (reg.z.anx_t1, level = 0.95) 95% range <- data.frame (bis_t1 = seq (7, 28, 1)) # range reg.anx_t1_95ci <- predict (reg.anx_t1, range, interval = "confidence") # reg.anx_t1_95ci matplot (range, reg.anx_t1_95ci, lty=c(1, 2, 2), type="l", add=true) # 24

13 (21.93) ( ) () 25,, 1 26

14 mreg.anx_t2 <- lm(anx_t2 ~ bis_t1 + bas_t1, data = z.depanx1) summary(mreg.anx_t2), 2,BIS BAS BIS 1, mreg.dep_t2 <- lm(dep_t2 ~ bis_t1 + bas_t1, data = z.depanx1) summary(mreg.dep_t2), 2,BIS BAS BIS 1, 2.37, BAS 1,

15 ,, install.packages ( car ) #car library (car) #car mreg.anx_t2 <- lm (anx_t2 ~ bis_t1 + bas_t1, data = z.depanx1) # vif (mreg.anx_t2) #VIF mreg.dep_t2 <- lm (dep_t2 ~ bis_t1 + bas_t1, data = z.depanx1) vif (mreg.dep_t2) 29 VIF,,, 0.2VIF 5, 80, VIF 1, (BIS BAS ) 30

16 , ( ), 1 2 (2 ) 3 BIS BAS 31, 2 z.depanx1.nona <- na.omit (z.depanx1) # attach (z.depanx1.nona) # reg1 <- lm (dep_t2 ~ sex + age, data = z.depanx1.nona) #1 summary (reg1) reg2 <- lm (dep_t2 ~ sex + age + anx_t2, data = z.depanx1.nona) #2 summary (reg2) reg3 <- lm(dep_t2 ~ sex + age + anx_t2 + bis_t1 + bas_t1, data = z.depanx1.nona) # 3 summary (reg3) anova (reg1, reg2, reg3) # 32

17 Step1 Step2 Step3 33 2, BIS BAS 2 (2) 34

18 , 2 z.depanx1.nona <- na.omit (z.depanx1) # attach (z.depanx1.nona) # reg1 <- lm (anx_t2 ~ sex + age, data = z.depanx1.nona) #1 summary (reg1) reg2 <- lm (anx_t2 ~ sex + age + dep_t2, data = z.depanx1.nona) #2 summary (reg2) reg3 <- lm(anx_t2 ~ sex + age + dep_t2 + bis_t1 + bas_t1, data = z.depanx1.nona) # 3 summary (reg3) anova (reg1, reg2, reg3) # 35, () 36

19 BAS, BAS BIS BAS 37 8 (cor()) (pairs()) (partial.r(),cor2pcor()) / (lm()) (lm()) (abline()) (confint(),predict()) (matplot()) VIF (vif() ) 38

Use R

Use R Use R! 2008/05/23( ) Index Introduction (GLM) ( ) R. Introduction R,, PLS,,, etc. 2. Correlation coefficient (Pearson s product moment correlation) r = Sxy Sxx Syy :, Sxy, Sxx= X, Syy Y 1.96 95% R cor(x,