Additional Plots and Stats with ggquickeda

Samer Mouksassi

2018-09-26

In this vignette we will expand what we learned in the Introduction to ggquickeda vignette. We will again launch the app and select the built-in dataset. Then we will do the following actions:

Change the mapped y variable(s) to Conc and Weight
In the Categorize/cut subtab select Age to be cut into two intervals (default to below and avove median)
Go to Color/Group/Split/Size/Fill Mappings (?) and map Colour By and Fill By to Age and Column Split to Gender

cut a continuous variable to categorical

Go to Median PIs (?) and select Median/PI and then you this get this plot:

MedianPI

This illustrated how to use more than one y variable and how to generate a Median and a Ribbon showing a 95% Prediction interval over the x variable (Time). We can see that Weight does not change over time and that older Females and Males had little difference with respect to concentrations but had higher Weight. Let us look at the Weight distributions in different ways first using a boxplot:

Change the mapped y variable(s) to Weight and x variable to Age
Go to Plot types, Points, Lines(?) and increase Point Size to 4 and make the transparency of the points equal to 0.2
Go to Color/Group/Split/Size/Fill Mappings (?) and map Color By:, Group By: Fill By: and Column Split: to Gender
Go to Boxplots and check the Add a Boxplot? checkbox.
Go To Median PI(?) and uncheck Ignore Mapped Group so the Median PI uses the mapped Gender Group By:

MedianPI

Try to experiment what Label Values? and Label N? do Keep Label N? checked.

In the following part we will generate a descriptive stats table that reflect the plot that we just did. * But first let us fix the fact that Weight is repeated multiple time by subject as it does not change over time. Go to One Row by ID(s) and map it to ID.

MedianPI

Click on the Descriptive Stats Tab

MedianPI

Remove all y variable(s) keeping Age as x variable gives:

MedianPI

Then selecting Weight as x variable gives:

MedianPI

As an exercise play with the options in the Histograms/Density/Bar to reproduce these plots.