# Step-by-Step Guide:

# Descriptive Plots - More Options

## Overview

What you will learn here:

to generate different plots for categorical and continuous data

to generate a scatter plot (using Sctr)

to generate bar plots (using Survey Plot)

to generate a dot chart (using JJStatsPlot)

## Pre-requsites

This tutorial introduces you to the following three Jamovi Modules:

scatr: This module allows you to create simple scatterplots. You can also include a regression line

surveymv: This module allows you to create a variety of plots using continuous, nominal and ordinal data. This module supplements the pre-existing plots found within Jamovi.

JJStatsPlot: This module allows you to create a series of plots for all data types as well as correlations between them. Each plot will come with some statistics - you will learn more about how to interpret them in later tutorials

Learn how to install a module here.

## Dataset used

Dataset used for the Example Below

Skoczylis, Joshua, 2021, "Extremism, Life Experiences and the Internet", https://doi.org/10.7910/DVN/ICTI8T, Harvard Dataverse, Version 3.

## Scatterplots

### Variables Used for the Scatter Plot:

Extremism_ScoreScaled - Continuous data (Dependent Variable)

Age - Continuous data (Independent Variable)

Gender - Nominal data (Independent Variable - here used to split results between gender)

## Scatterplots: Step-by-Step Video Guide

Creating a Scatterplot

by Data Advantage

Scatter plots use dots to represent the observations of two continuous variables. You can also split your data using a nominal variable. Each dot represents where the data points of your two variables meet. Scatter plots are used to observe relationships.

Jamovi makes it easy to add in regression lines as well as Marginals (a density or boxplot for each continuous variable in the margins).

## Scatterplot: Step-by-Step Guide

1.

Create a Scatterplot

Navigate to Analyses > Exploration > Scatterplot

Scatter Plots require two continuous variables. You can also add a grouping variable e.g. gender to see if this has an effect on your data.

Usually, you will place the dependent variable on the y-axis (vertical line) and the independent variable on the x-axis.

In the example below we will use the Extremism_scoreScale (dependent variable) and Age (independent variable) with Gender as our Grouping variable.

Jamovi allows you to select what type of regression line you want to fit. You can also include Marginals (these are either a box plot or density plot which is displayed in the margins of the plot.

2.

Select Regression line and Marginals

Now you can select your Regression line.

None

Linear Line

Smooth Line

You can also choose to display a boxplot or density plot on the side of your graph (called Marginals here) - this will give you some idea of how each variable is distributed.

3.

Scatter Plot Results

The Scatterplot shows you the relationship between two variables. Each point in the scatterplot indicates the intersection between the values of the two variables. Scatterplots allow you to fit regression lines which will give you an indication of whether there is a relationship between the variables or not.

You can also group your data using a third categorical variable such as gender. Each option will appear in a different colour. If you include a regression line, you will get one for each of your groups.

In our example, we have used Age and Extremism grouped by Gender. In the scatterplot below we have included a smooth regression line, rather than a linear one - although it looks like a linear line would work well here.

So what does this plot tell us? The below plot tells us the following:

Younger people tend to have higher extremism scores on average

Males tend to have higher extremism scores than females across all age categories.

The age variable is skewed towards older people

Most of the participants have a low extremism score.

Important: Scatterplots are descriptive plots and only provides you with an indication of the possible relationships. Plots like the one below give you a good indication of whether a relationship exists and what form it takes (there might not be any, but it may be linear or curvilinear).

## Survey Plots (surveymv)

### Variables Used for the Box and Density Plots:

Violence_effective_protect_views - Nominal data

Gender - Nominal data (Independent Variable - here used to split results between gender)

Extremism_scoreScaled - Continuous data

## Survey Plots: Step-by-Step video Guide

### Using Survey Plot

Survey plot allows you to create different plots depending on your data type. The survey plots available for continuous data are the same as the ones offered by the Jamovi base packet.

Video by J Skoczylis

## Survey Plot - Bar Charts: Step-by-Step Guide

1.

Create a Bar Chart

Navigate to Analyses > Exploration > Survey Plots

Now all you have to do is select your nominal/ordinal variables and select what kind of output you want.

In terms of Plot types, you can select either stacked or Groups. You can also select whether to show your data as counts or in percentages and whether to have the labels in the plot or on the x-axis. For bar plots, you should usually use percentages. Finally, you can decide whether to include missing values or not (most of the time you probably should exclude them, but there may be cases where you want to include them.

### Stacked Bar Plot:

The below plot is a stacked bar plot using percentages and excluding missing values. The labels are displayed on the x-axis rather than in the bar.

### Grouped Bar Plot:

Below is the output for a grouped bar plot using percentages and excluding missing values. Again the labels are displayed ion the x-axis.

2.

Results: Bar Charts

### Stacked Bar Plot:

The below plot is a stacked bar plot using percentages and excluding missing values. The labels are displayed on the x-axis rather than in the bar.

### Grouped Bar Plot:

Below is the output for a grouped bar plot using percentages and excluding missing values. Again the labels are displayed ion the x-axis.

## Survey Plot - Box/Denisty Plot: Step-by-Step Guide

1.

Create a Box/Density Plot

Navigate to Analyses > Exploration > Survey Plots

Select your variables. You can now select the type of plots you want - a Box plot or density plot (or both together). ]

If you select data it will show you the data points on your plot.

### Your Continuous Plot:

Here we have combined a box plot, denisty plot showing the data points.

2.

Results: Box/Density Plot

### Your Continuous Plot:

Here we have combined a box plot, density plot. The plot also shows actual data points

## JJStatsPlots

### Variables Used for JJStatsPlots:

Racism_score - Continuous data (Dependent Variable)

Highest_Qualification - nominal data (Independent Variable)

Gender - Nominal data (Independent Variable - here used to split results between gender)

## JJStatsPlots: Step-by-Step video Guide

### Using JJStatsPlots

This module gives you lots of different plot options. Not all of them will be covered, but the concept is easy, drag and drop the variables you want and select the settings for your graph.

video by J Skoczylis

## JJStatsPlots: Step-by-Step Guide

1.

Using JJStatsPlots to create graphs

Navigate to Analyses > Exploration > JJStatsPlot

This module allows you to create plots for continuous and categorical variables. It also makes it easy to create plots that show relationships between variables.

JJStatsPlot really gives you two options:

Let Jamovi decide which plot is best for your selected variables (selected Graphs and Plots). Note, this only works when you have a dependent and independent variable.

Select one of the main plots from the list (see image below). Selecting your plot is easy as the dropdown menu tells you what type of data you need.

Once you have selected what plot you want, all you have to do is drag and drop your variables into the right boxes and select your settings.

The plots come with lots of additional statistics. At this point, ignore them, but as we go along you will learn how to read the extra information with ease.

1.

Results: Dot Plot

You should already be familiar with most of the plots available, so we will only show you one example: a Dot Plot.

Navigate to Analyses > Exploration > JJStatsPlots

As outlined above select your variables. The JJStatsPlot dot plot is actually a so-called Cleveland dot plot

This dot graph plots points (in this case the mean) belonging to one of the several categories of a categorical variable. In many ways, this plot is similar to a bar chart, where the dots replace the bar.

As you will see from the plot below, there is a lot of information in them. As we have split the data by Gender, we get three graphs. One for males, one for females and one for everything together.

The Black dots represent the mean for each category. The dotted line gives you the mean for all the categories together and the scale at the bottom gives you the racism score.

So, males with no qualifications have a mean racism score of ~-0.61. This is compared to a mean of just over ~0.19 for those with a Postgraduate Degree. These results can be compared to the mean which is -0.13 for males.

Additional Information: As you can tell these plots also provide you with additional information. Depending on what analysis you select you will get the relevant output. More on interpreting this type of output in later tutorials.