Chapter 3 Regression and ANOVA

3.1 Packages needed

Include this chunk at the top of you analysis to ensure that you have all the packages. It also includes the wrapper to add buttons to a data table if you want to use this. Remember that data tables can only be included in HTML documents.

library(ggplot2)
library(dplyr)
library(mgcv)
library(DT)
theme_set(theme_bw())
dt<-function(x) DT::datatable(x, 
    filter = "top",                         
    extensions = c('Buttons'), options = list(
    dom = 'Blfrtip',
    buttons = c('copy', 'csv', 'excel'), colReorder = TRUE
  ))

3.2 Univariate

3.2.1 Data

d<-read.csv("https://tinyurl.com/aqm-data/mussels.csv")
dt(d)

3.3 Data summaries for individual variables

Change the name of the variable to match a numerical variable in your own data set. The command removes NAs just in case you have them

summary(d$Lshell,na.rm=TRUE)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    61.9    97.0   106.9   106.8   118.7   132.6

3.4 Individual statistics for a single variable

Mean, median, standard deviation and variance.

mean(d$Lshell, na.rm=TRUE)

## [1] 106.835

median(d$Lshell, na.rm=TRUE)

## [1] 106.9

sd(d$Lshell, na.rm=TRUE)

## [1] 14.84384

var(d$Lshell, na.rm=TRUE)

## [1] 220.3397

3.5 Simple boxplot of one variable

Useful for your own quick visualisation.

boxplot(d$Lshell)

3.6 Simple histogram of one variable

Useful for your own quick visualisation.

hist(d$Lshell)

3.7 Neater histogram of one variable

This uses ggplot. Change the bin width if you want to use this.

g0<-ggplot(d,aes(x=d$Lshell))
g0+geom_histogram(color="grey",binwidth = 5)