Update main.tex

Clarification for violin plot

papers/Gagnon_Kebe_Tahiri/main.tex

@@ -212,7 +212,7 @@ \subsection{Least-Squares Distance}\label{LS}
 
 \section{Results}\label{results}
 
-Figure \ref{fig:fig1} shows the extent and variability of the dataset for two geographical attributes, i.e., latitude (decimal) and longitude (decimal) at the start of sampling, as well as six environmental attributes, such as wind speed (m/s) and depth (m) at the start of sampling, and temperature ($^\circ$C) and the oxygen concentration of the water (mg/L) based on the depth at which the samples were collected.
+Figure \ref{fig:fig1} shows the extent and variability of the dataset for two geographical attributes, i.e., latitude (decimal) and longitude (decimal) at the start of sampling, as well as six environmental attributes, such as wind speed (m/s) and depth (m) at the start of sampling, and temperature ($^\circ$C) and the oxygen concentration of the water (mg/L) based on the depth at which the samples were collected. Violin diagrams display many of the same summary statistics as box plots: the white dots designate the medians, the thickened black bars in the center represent the interquartile range, and the thin black lines on either side evoke the rest of the distribution, except for the points that are rated as "extreme". On either side of the black lines is an approximation of the kernel density to show the shape of the data distribution. The wider areas of the violin diagrams show a higher probability of the variables taking a given value; the thinner areas indicate a lower probability.
 
 \begin{figure}[htbp]
     \centering