Description

Anonymous
Organising and cleaning data:
Setting working directory as “PenguinProjects” folder (please adjust to suit your working directory):
setwd(“~/RWorkingDirectory/PenguinProjects”)
Loading in libraries, data and functions:
source(“functions/libraries.r”) source(“functions/cleaning.r”) source(“functions/plotting.r”)
Creating a new folder for the raw data and saving the raw data:
#Creating a new data folder dir.create(“data”)
#Creating a new raw data folder dir.create(“data/data_raw”)
Cleaning the data, creating a new folder for the clean data and saving the clean data:
#Using function from “cleaning.r” to clean the data penguins_clean <- cleaning(penguins_raw)
#Creating a new clean data folder dir.create(“data/data_clean”)
Looking at the data:
head(penguins_clean)
Preparing data for analysis:
Removing rows without sex or body mass recorded and defining variables:
#Using function from “cleaning.r” to prepare data penguins_mass <- remove_empty_body_mass(penguins_clean)
#Defining factors and numerical variables penguins_mass$species <- as.factor(penguins_mass$species) penguins_mass$sex <- as.factor(penguins_mass$sex) penguins_mass$body_mass_g <- as.numeric(penguins_mass$body_mass_g)
Looking at the data:
head(penguins_mass, 3)
## # A tibble: 3 x 3
## species sex body_mass_g
## <fct> <fct> <dbl>
## 1 Adelie Penguin (Pygoscelis adeliae) MALE 3750
## 2 Adelie Penguin (Pygoscelis adeliae) FEMALE 3800
## 3 Adelie Penguin (Pygoscelis adeliae) FEMALE 3250

Analysing the data:
Q: To what extent does the species and sex of a penguin influence body mass?
Creating a model with no interaction and testing assumptions of statistical analysis:
#Creating the model lmSpecies_Sex_Mass_No_Interaction <- aov(body_mass_g ~ species + sex, data = penguins_mass)
#Plotting to test assumptions par(mfrow = c(2,2)) plot(lmSpecies_Sex_Mass_No_Interaction)

Overall, the data appears to meet assumptions of ANOVA.
Running ANOVA:
summary(lmSpecies_Sex_Mass_No_Interaction)
## Df Sum Sq Mean Sq F value Pr(>F)
## species 2 145190219 72595110 724.2 <2e-16 ***
## sex 1 37090262 37090262 370.0 <2e-16 ***
## Residuals 329 32979185 100241
## —
## Signif. codes: 0 ’***’ 0.001 ’**’ 0.01 ’*’ 0.05 ’.’ 0.1 ’ ’ 1
Both species and sex appear to have a significant effect on body mass.
Building a model that takes into account interaction between species and sex:
#Creating the model lmSpecies_Sex_Mass_Interaction <- aov(body_mass_g ~ species * sex, data = penguins_mass)
#Plotting to test assumptions par(mfrow = c(2,2)) plot(lmSpecies_Sex_Mass_Interaction)

303
314
Overall, the data appears to fit the assumptions of ANCOVA.
Running ANCOVA:
summary(lmSpecies_Sex_Mass_Interaction)
## Df Sum Sq Mean Sq F value Pr(>F)
## species 2 145190219 72595110 758.358 < 2e-16 ***
## sex 1 37090262 37090262 387.460 < 2e-16 ***
## species:sex 2 1676557 838278 8.757 0.000197 ***
## Residuals 327 31302628 95727
## —
## Signif. codes: 0 ’***’ 0.001 ’**’ 0.01 ’*’ 0.05 ’.’ 0.1 ’ ’ 1
There is a significant effect of interaction between species and sex on body mass.
Comparing the two models:
anova(lmSpecies_Sex_Mass_No_Interaction,lmSpecies_Sex_Mass_Interaction)
## Analysis of Variance Table
##
## Model 1: body_mass_g ~ species + sex
## Model 2: body_mass_g ~ species * sex
## Res.Df RSS Df Sum of Sq F Pr(>F)
## 1 329 32979185
## 2 327 31302628 2 1676557 8.757 0.0001973 ***
## —
## Signif. codes: 0 ’***’ 0.001 ’**’ 0.01 ’*’ 0.05 ’.’ 0.1 ’ ’ 1
Adding an interaction between species and sex significantly improves the model.
Running a Tukey HSD test:
#Using TukeyHSD test to compare between groups (results hidden here to keep document tidy)
TukeyHSD(lmSpecies_Sex_Mass_Interaction)
The Tukey HSD test suggests significant differences between body mass of penguins grouped by species and sex.
The comparisons between groups which are not significantly different are:
1) Chinstrap females and Adelie females
2) Chinstrap males and Adelie males
Producing figures to display the data:
Creating a new folder to store figures in:
dir.create(“figures”)
Producing an interaction plot:
#Using function from “plotting.R” to produce graph plot_mass_figure(penguins_mass)
Interaction Plot of Body Mass, Species and Sex of Penguins in the Palmer Archipelago:

Mean Body Mass Male Female
Error bars represent 95% confidence intervals
Saving the figure as a PNG file:
#Using function from “plotting.R” to save figure as PNG
Saving the figure as an SVG file:
sessionInfo()
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## Running under: Windows 10 x64 (build 22000)
##
## Matrix products: default
##
## locale:
## [1] LC_COLLATE=English_United Kingdom.1252
## [2] LC_CTYPE=English_United Kingdom.1252
## [3] LC_MONETARY=English_United Kingdom.1252
## [4] LC_NUMERIC=C
## [5] LC_TIME=English_United Kingdom.1252
##
## attached base packages:
## [1] stats graphics grDevices utils
##
## other attached packages: datasets methods base
## [1] svglite_2.1.0 ragg_1.2.2 dplyr_1.0.7
## [4] janitor_2.1.0 ggplot2_3.3.3
##
## loaded via a namespace (and not attached): palmerpenguins_0.1.1
## [1] Rcpp_1.0.8.3 lubridate_1.8.0 lattice_0.20-41
## [4] deldir_1.0-6 png_0.1-7 assertthat_0.2.1
## [7] digest_0.6.27 utf8_1.2.1 R6_2.5.1
## [10] backports_1.4.1 evaluate_0.18 highr_0.9
## [13] pillar_1.6.0 rlang_0.4.11 data.table_1.14.2
## [16] rstudioapi_0.14 rpart_4.1-15 Matrix_1.4-1
## [19] checkmate_2.0.0 rmarkdown_2.18 labeling_0.4.2
## [22] textshaping_0.3.6 splines_4.0.5 stringr_1.4.0
## [25] foreign_0.8-81 htmlwidgets_1.5.4 munsell_0.5.0
## [28] compiler_4.0.5 xfun_0.22 pkgconfig_2.0.3
## [31] systemfonts_1.0.4 base64enc_0.1-3 htmltools_0.5.2
## [34] nnet_7.3-15 tidyselect_1.1.1 htmlTable_2.4.1
## [37] gridExtra_2.3 tibble_3.1.1 Hmisc_4.7-0
## [40] fansi_0.4.2 crayon_1.5.2 withr_2.5.0
## [43] grid_4.0.5 gtable_0.3.1 lifecycle_1.0.0
## [46] DBI_1.1.3 magrittr_2.0.1 scales_1.1.1
## [49] cli_3.0.1 stringi_1.7.6 farver_2.1.0
## [52] latticeExtra_0.6-30 snakecase_0.11.0 ellipsis_0.3.2
## [55] generics_0.1.3 vctrs_0.3.8 Formula_1.2-4
## [58] RColorBrewer_1.1-3 tools_4.0.5 interp_1.0-33
## [61] glue_1.4.2 purrr_0.3.4 jpeg_0.1-9
## [64] fastmap_1.1.0 survival_3.2-10 yaml_2.3.5
## [67] colorspace_2.0-0 cluster_2.1.1 knitr_1.33