Ch. 5 - Data Visualization

230-268

plots should do the following

• show the data

• induce the reader to think about the data being presented

• avoid distorting the data

• present many numbers with minimum risk

• make large data sets

• encourage the reader to compare different pieces of data

• reveal underlying message of the data

chartjunk

• distracting visual elements in a chart or graph

• fail to communicate quantitative data clearly

boxplot

• median at the center of the plot

• box at the top & bottom of the statistic

• boxes = limits within whch the middle 50% of observations fall (IQR)

three types of boxplots

• 1d boxplot

• simple boxplot

• clustered boxplot

clustered boxplot

• splits data by a second categorical variable

• boxplots for this second variable are produced in different colours

simple boxplot

• multiple boxplots for the chosen outcome

• splitting the data by a categorical variable

1d boxplot

• single boxplot of all scores for the chosen outcome

simple bar

• means of scores across different groups or categories of cases

clustered bar

• different coloured bars to represent levels of a second grouping variable

stacked bar

• different coloured bars are stacked on top of each other

simple 3d bar

• second grouping variable is displayed by an additional axis

clustered 3d bar

• add a third categorical variable on an extra axis

simple error bar

• mean is represented by a dot

• lien represents the precision of the estimate of the mean

clustered error bar

• mean displayed as a dot with an error bar around it

line charts

• bar charts but with lines

simple line

• display the means of scores across different groups of cases

multiple line

• equivalent to clustered bar chart

• plot means of an outcome variable for different categories/groujps of a predictor variable

• produce different coloured lines of each category/group

scatterplot

• plot each entity’s score on one variable against their score on another

• visualizes relationship between the variables

simple dot plot/density plot

• like a histogram

• individual scores displayed as dots

• shape of the distribution of scores