Skip to contents

jamovi User Guide for ClinicoPath

Source: vignettes/01-jamovi-user-guide.Rmd
01-jamovi-user-guide.Rmd

A Clinician’s Guide to Statistical Analysis in jamovi

This guide provides a comprehensive introduction to performing statistical analysis in jamovi using the ClinicoPath module. jamovi is a free and open-source statistical spreadsheet that provides a user-friendly, point-and-click interface for complex analyses. This makes it ideal for pathologists, clinicians, and researchers who need to analyze data without learning how to code.

Installation

Installing the ClinicoPath Module in jamovi

  1. Open jamovi.
  2. Click the Modules button (a large + sign) in the top-right of the menu bar.
  3. Select jamovi library from the dropdown menu.
  4. In the search box, type ClinicoPath.
  5. Click the Install button next to the ClinicoPath module.
  6. jamovi will install the module and all its dependencies. Once installed, you will see the module’s name appear in the main analysis ribbon.

[Screenshot of the jamovi library showing the ClinicoPath module and the “Install” button.] ***

A Complete Analysis Workflow: A Clinical Example

To illustrate how to use the ClinicoPath module, we will walk through a complete analysis of a clinical research question.

The Clinical Scenario

A researcher is analyzing a dataset of patients who have undergone mammography screening. They want to answer the following question:

Is there a statistically significant difference in age between patients with a positive cancer diagnosis and those with a negative diagnosis?

We will use the breast_cancer_data dataset included with the ClinicoPath module to answer this question.

Step 1: Load the Data

First, we need to load the breast_cancer_data dataset. This dataset is included with the ClinicoPath module, so you can easily open it in jamovi.

  1. In the top-left menu (the “hamburger” menu ☰), click File -> Open.
  2. In the file browser, navigate to the ClinicoPath module’s data folder. The location of this folder varies depending on your operating system. A shortcut is to click on the Examples dataset that comes with jamovi, and then navigate from there.
  3. Select the breast_cancer_data.omv file.

[Screenshot of the jamovi interface with the breast_cancer_data dataset loaded.] ***

Step 2: Perform the Analysis with a Box-Violin Plot

The best way to visualize the difference in a continuous variable (like age) between two categorical groups (like cancer_status) is with a box-violin plot.

  1. From the main analysis ribbon, click on JJStatsPlot -> Categorical vs Continuous -> Box-Violin Plots (Between Groups).

[Screenshot of the jamovi analysis ribbon showing the path to the Box-Violin Plots.] ***

  1. In the analysis window, you will see a panel where you can assign variables.
    • Move the age variable to the Dependent Variable box.
    • Move the cancer_status variable to the Grouping Variable box.

[Screenshot of the analysis window showing the variables being assigned.] ***

  1. jamovi will automatically generate the plot and the statistical analysis.

Step 3: The Output Plot

Here is the plot generated by jamovi:

# Load the data
data("breast_cancer_data", package = "ClinicoPath")

# Create the plot
jjbetweenstats(
  data = breast_cancer_data,
  dep = "age",
  group = "cancer_status",
  type = "parametric",            # Statistical test type
  pairwise.comparisons = TRUE,    # Post-hoc comparisons
  title = "Age Distribution by Cancer Status",
  subtitle = "Two-sample t-test with effect size",
  xlab = "Cancer Status",
  ylab = "Age (years)"
)

Step 4: Interpreting the Plot and Statistics

Now, let's interpret the output.

  • The Plot: The plot shows the distribution of age for both groups (Negative and Positive cancer status).
    • The violin part (the colored, shaded area) shows the density of the data. We can see that the Positive group has a wider distribution of ages.
    • The box part (inside the violin) shows the summary statistics. The thick line is the median age. The box itself represents the interquartile range (the middle 50% of patients).
    • Visually, it appears that the median age of the Positive group is higher than the Negative group.
  • The Statistics: The jjbetweenstats function automatically performs a t-test to compare the means of the two groups. The results are displayed on the plot itself.
    • t-test: The plot shows the results of the t-test: t(198) = -2.9, p = 0.004.
    • p-value: The p-value is 0.004. Since this is less than the standard cutoff of 0.05, we can conclude that there is a statistically significant difference in age between the two groups.
    • Effect Size: The plot also shows Cohen's d, which is a measure of effect size. Here, d = 0.41, which is typically considered a small to medium effect.

Step 5: Reporting the Results

When publishing these findings, you should report the descriptive statistics (mean and standard deviation for each group) and the results of the t-test. Here is an example of how to write this up:

A two-sample t-test was conducted to compare the age of patients with positive and negative cancer diagnoses. There was a statistically significant difference in age between the two groups (t(198) = -2.9, p = 0.004). The mean age of patients with a positive diagnosis (M = 55.4, SD = 12.5) was significantly higher than the mean age of patients with a negative diagnosis (M = 50.2, SD = 11.8). The effect size was small to medium (Cohen's d = 0.41).

Other Common Analyses

The ClinicoPath module provides a wide range of other analyses. The general workflow is the same for all of them:

  1. Select the analysis from the ribbon.
  2. Assign the variables.
  3. Customize the options (e.g., change the type of statistical test, modify the plot appearance).
  4. Interpret the output.

Troubleshooting

“No data to plot” Error

  • Check: Ensure variables are selected correctly.
  • Missing data: Variables with all missing values can't be plotted.
  • Variable type: Ensure variable types match analysis requirements (e.g., don't use a categorical variable as a dependent variable in a continuous analysis).

Statistical Test Failures

  • Sample size: Some tests require a minimum number of data points in each group.
  • Assumptions: Some statistical tests have assumptions (e.g., normality). If the assumptions are not met, the test may fail or the results may be invalid. The ClinicoPath module often provides non-parametric alternatives for these situations.

Getting Help