CODEWITHSUNDEEP
rggplot2
Captain Murphy
Captain Murphy

Reputation: 865

Linetype and Guide options for ggplot2 when using geom_smooth with continuous variables

I am using ggplot2 and the geom_smooth functions to make predicted lines from a lm model. I am using continuous y and x functions in my ggplot code.

I am trying to create a different line type for each line in the plot (which is the variable "location" ranging from -1,0, or 1), and I would like each location listed in the guide (where -1 = Location A, 0 = Location B, and 1 = Location C). A replicable excerpt of my code is below:

#test scores that range from 1 to 7
score <- c(6, 3, 5, 6, 7, 2, 4, 6, 3, 5, 4, 3, 3, 1, 3, 3, 3, 5, 2, 3, 2, 2, 7, 3, 7,     5, 4, 1, 3, 2, 7, 6, 6, 3, 6)
#location of the school
location <- c(1,  0,  0,  0,  1,  1,  1, -1,  1, -1,  0, -1,  0,  1,  0,  0,  0, -1,     0, -1,  0, -1,  0,  0,  0,  0,  0, -1,  0, -1,  0,  0,  0,  1,  0)

IQ <- c(0.7171425604,    0.7056850461,    1.3929736220,    0.0633936624,   -0.6872828336,   -1.3665840767,    1.4368569944, 0.7297599487,   -0.5735047485,   -0.6752912747,   -0.6213572428,   -0.6110533924,   -0.7090921238,    0.0501744806, 1.3916802944,   -0.0055243194,    1.3619753292,    1.4406369365,    0.6529601586,    0.0538097896,    1.3821853866, 1.3870600993,    0.0040551996,    0.6600558495,    1.3550162100,    1.3081187951,   -1.7541949601,    1.3768167017, -0.6232446826,   -1.2793074919,    0.0560708725,   -0.5993356051,   -0.5857733192,   -0.6005459705,   -0.6659873442)

df <- data.frame(score, location, IQ)

p1 <- ggplot(data=df,aes(y=score,x=IQ,shape=factor(location))) + 
geom_smooth(method = "lm", se=F) + scale_y_continuous("Test Score",limits=c(1,7)) +
scale_x_continuous("IQ level", limits=c(-2.45, 1.45)) + 
theme_bw() +
theme(axis.text.x=element_text(size=rel(1.2), color = 'black'),
    axis.title.x=element_text(size=rel(1.3)),
    axis.title.y=element_text(size=rel(1.3)),
    axis.text.y=element_text(size=rel(1.2), color = 'black'),
    panel.grid.minor=element_blank(),
    panel.grid.major.x=element_blank(),
    panel.grid.major.y=element_blank(),
    axis.ticks.y = element_blank(), 
    panel.border = element_blank(),
    plot.title = element_text(size = rel(1.4))) 

print(p1)

Whenever I try to add linetype or the manual linetype, I keep getting errors about how linetype cannot be mapped with continuous variable (only discrete). Is there any way around this or should I be using a different function? Maybe create labels for the values in location (School 1, School 2, and School 3?) beforehand? Or should I be using some alternative to linetype? I cannot find a solution online or in Wickham's book. Thank you!

Upvotes: 2

Views: 8201

Answers (1)

Didzis Elferts
Didzis Elferts

Reputation: 98459

You are getting this error message because location is continuous variable (numeric variables are treated as continuous in ggplot) and for linetype= only discrete variables can be used. To use it for linetype= convert location to factor. To change labels in legend use scale_linetype() and provide argument labels=.

ggplot(data=df,aes(y=score,x=IQ,linetype=factor(location))) + 
  geom_smooth(method = "lm", se=F) + scale_y_continuous("Test Score",limits=c(1,7)) +
  scale_x_continuous("IQ level", limits=c(-2.45, 1.45)) +
  scale_linetype(labels=c("Location A","Location B","Location C"))

Upvotes: 2

Related Questions