6.7 Putting It All Together

This chapter focused on visualizing data in jamovi. Graphs are not separate from statistics. They are part of how we understand our data.

Chapter Recap

You should now be able to:

  • explain why visualizing data matters
  • choose graphs based on variable type
  • use bar plots for categorical variables
  • use histograms, density plots, box plots, violin plots, data points, and mean markers for continuous variables
  • visualize continuous variables across groups
  • use graphs to notice skew, outliers, unexpected categories, and possible data-cleaning problems

Common Mistakes

Watch for these common mistakes:

  • making a graph that does not match the variable type
  • using a bar plot for a continuous variable
  • ignoring individual data points when group sizes are small
  • treating visual differences as statistical significance
  • failing to notice duplicate categories caused by spelling or spacing
  • choosing a graph because it looks nice rather than because it answers a question

Applied Practice

Imagine you have these variables:

Variable Type
Gender_Recoded Categorical
Class_Level Categorical
Boredom_Mean Continuous
Mind_Wandering_Mean Continuous

Which graph would you use for each goal?

  1. Show the distribution of Boredom_Mean.
  2. Show the frequency of each Class_Level.
  3. Compare Mind_Wandering_Mean across Gender_Recoded groups.
Answers
  1. A histogram, density plot, box plot, violin plot, data points, or mean marker could help show the distribution of Boredom_Mean.
  2. A bar plot would show the frequency of each class level.
  3. A grouped visualization such as box plot + violin plot + jittered data + mean would help compare mind wandering across gender groups.

Looking Ahead

Up to this point, we have focused on understanding the data we have: what variables we measured, how they are set up, whether they need to be cleaned or transformed, how to describe them, and how to visualize them.

That is always the first step. Next, we will begin asking inferential questions: what can these data tell us beyond this sample? That is where hypothesis testing comes in.