R script for within subjects mediation

space2/spacing_analysis.R

@@ -0,0 +1,24 @@
+remove(list=ls())
+setwd("/Users/matt/Desktop")
+input.filename <- "SPACE2_eeg_recall.csv"
+mydata <- read.csv(input.filename, sep= ",", header=TRUE)
+levels(mydata$space) <- c("mass", "spac2", "spac12", "spac32")
+
+# linear model
+Y <- cbind(mydata$theta, mydata$recall)
+fit <- aov(Y ~ space + Error(sub), data=mydata)
+summary(fit)
+
+# I don't understand why this doesn't show the massed condition
+fit <- lm(Y ~ space, data=mydata)
+summary(fit)
+
+# MANOVA
+runfactor <- ordered(as.factor(mydata$space), levels = c("mass", "spac2", "spac12", "spac32"))
+#fit <- manova(cbind(mydata$theta, mydata$recall) ~ runfactor + Error(mydata$sub))
+fit <- manova(cbind(mydata$theta, mydata$recall) ~ runfactor)
+summary(fit, test="Pillai")
+summary.aov(fit)
+
+# put data in wide format
+mydatawide <- reshape(mydata, timevar = "space", idvar = c("sub"), direction = "wide")