1. Get data.
load("/Users/hnegami/Desktop/DATA VIS/sampDat.dat")
2. Descriptives.
summary(subS)
## pos orient judgori rt
## left :14662 Min. : 0.01 Min. :-110 Min. :0.102
## right:14714 1st Qu.: 45.11 1st Qu.: 47 1st Qu.:0.929
## Median : 90.38 Median : 90 Median :1.215
## Mean : 90.21 Mean : 89 Mean :1.343
## 3rd Qu.:135.45 3rd Qu.: 132 3rd Qu.:1.615
## Max. :180.00 Max. : 315 Max. :4.942
##
## angdiff validcue blocknum trialnum
## Min. :-89.89 FALSE : 0 Min. :1.00 Min. : 1.0
## 1st Qu.: -8.69 TRUE : 0 1st Qu.:1.00 1st Qu.: 38.0
## Median : -0.06 invalid: 5908 Median :2.00 Median : 76.0
## Mean : 0.16 valid :23468 Mean :2.42 Mean : 75.6
## 3rd Qu.: 8.89 3rd Qu.:3.00 3rd Qu.:113.0
## Max. : 89.89 Max. :5.00 Max. :150.0
##
## userid tilt orientToVert angdiff_TV
## 701 : 750 l:14735 Min. : 0.0 Min. :-89.89
## 704 : 750 r:14641 1st Qu.:22.7 1st Qu.:-11.68
## 705 : 750 Median :45.2 Median : -2.50
## 715L : 750 Mean :45.0 Mean : -3.62
## 717L : 750 3rd Qu.:67.5 3rd Qu.: 5.96
## 718L : 750 Max. :90.0 Max. : 89.89
## (Other):24876
## angdiff_TV_MC
## Min. :-104.65
## 1st Qu.: -8.16
## Median : 0.00
## Mean : -0.44
## 3rd Qu.: 8.03
## Max. : 130.31
##
3. Subselect the participants who were in experiment 1.
subexp1 <- subset(subS, (grepl("exp", userid)))
4. Average their performance by trial.
aveperf <- aggregate(angdiff ~ trialnum, data = subexp1, mean)
5. Plot the average performance (y) versus trialnum (x).
library("ggplot2")
p <- ggplot(data = aveperf, aes(x = trialnum, y = angdiff))
p + geom_point() + theme_bw()
6. Loess line with confidence intervals.
p + geom_smooth(fill = "light blue") + theme_bw()
## geom_smooth: method="auto" and size of largest group is <1000, so using loess. Use 'method = x' to change the smoothing method.
