There have been a few specific requests for carrying out particular applications in R. We will add code to address these requests here throughout the week.

1. Updating R, RStudio, Rpackages
   • If you are getting warning messages while installing packages, your version of R may be out-of-date. To keep your version of R most up-to-date, follow the directions at the webpage linked here.
   • Note for Windows : Keep in mind that if you are using the installr package for Windows, you will be installing a new package to help you. The linked page gives directions for installing this package via your console, but if you do not have internet on your computer, you’ll need to install this package from the source file.
2. Correcting commas from a French dataset

• Many .csv files in Madagascar will have commas in place of periods for numeric entries like weights. If you are not able to import these into R correctly via the read.csv2 command, you can change them manually. Try it with the following code:

## make up a vector where the decimals are entered as commas
comma.weights = c("56,3", "45,6", "67,8", "87,4", "42,1", "75,4")

## look at your vector of weights with commas.
comma.weights

## it is stored as a character because of the commas, which you would like to replace with periods
class(comma.weights)

## you can do this with 'gsub' in base R
comma.weights = gsub(pattern = ",", replacement = ".", x=comma.weights)

## then, turn it into a numeric vector for use in data analysis
comma.weights = as.numeric(comma.weights)

## look at the help file to explore the functions of this command and others like it further
?gsub

3. Correcting NAs in your data

• You can use the following code to subset your data to include only those rows that have no NAs in any column.

## imagine you have a dataset that looks like this
yr = c(1990, 1998, 1997, 2003, NA, 2015)
age =c(5,4,NA,6,12,18)
name=c("Ricky", "Fenosoa", "Lady", NA, "Tahiana", "Tsilavo")
dat.ex = cbind.dataframe(yr, name, age)

## look at your dataframe. you have an NA in column 1/row 5, column 2/row 4, and column 3/row3. the function complete.cases will help you subset your data to include only those rows of data which have no NAs

dat.ex.new = dat.ex[complete.cases(dat.ex),]

##look at your new data frame:

dat.ex.new

## there are no longer any NAs, but be careful! if you were only interested in analyzing your columns for yr and age, you just eliminated a row (row 4) that had all the information you needed. this function will only take rows for which ALL information is complete.

4. Plotting a single shape area

• Warning! This code will only run in the same working directory as Amy’s Intro to Spatial Modeling tutorial. This tells you how to plot a singe shape area within a larger shapefile

## load your required libraries
library(maptools)
library(rgeos)

## read in the entire districts of Madagascar shapefil
mdg_admin2_shp<-readShapePoly('MDG_Shp/MDG_adm2.shp', proj4string = CRS('+proj=longlat'))

## select a subset that only includes the district named "Sava"
subset_shp_file<-mdg_admin2_shp[which(mdg_admin2_shp$NAME_2 == 'Sava'),]

## plot all of Madagascar
plot(mdg_admin2_shp)

## plot the Sava district on top of all of Madagascar and color it blue
plot(subset_shp_file, col = 'blue', add = TRUE)

## plot just the Sava district on its own
plot(subset_shp_file)