Skip to main content

Quick Explore

The Quick Explore sub-menu provides several tools for rapid, automated exploration of a dataset. These are particularly useful when working with a new dataset and wanting to get an overview before diving into specific analyses.

Access via Advanced > Quick Explore.

Missing Values​

Opens a summary window showing the extent of missing data across all variables in the dataset. The output includes:

  • A text summary table listing each variable and the number/percentage of missing values
  • A combination plot showing which combinations of variables have missing values together

This uses the iNZightMR package's calcmissing() and plotcombn() functions to compute and visualise missing data patterns. Understanding which variables are missing and whether missingness is related across variables is an important first step in any analysis.

All 1-Variable Plots​

Automatically steps through every variable in the dataset, plotting each one individually. For each variable, iNZight produces the appropriate default plot:

Use the Next and Previous buttons to cycle through variables. This is a quick way to spot unusual distributions, outliers, or unexpected variable types.

All 1-Variable Summaries​

Produces a text summary for every variable in the dataset in a single window. For each variable, the summary includes:

  • Numeric variables: Mean, standard deviation, min, quartiles, max, and number of missing values
  • Categorical variables: Frequency table of levels

This provides a comprehensive numeric overview of the entire dataset at a glance.

Explore 2-Variable Plots​

Select a reference variable, and iNZight will step through plotting every other variable in the dataset against it. This is useful for quickly identifying which variables have interesting relationships with a variable of interest.

The plot type is chosen automatically based on the variable types:

Pairs​

Creates a scatter plot matrix (pairs plot) for a user-selected set of variables. Each variable is plotted against every other variable in a grid layout, making it easy to identify correlations and patterns across multiple variables simultaneously.

  • Hold CTRL (or CMD on Mac) to select multiple variables
  • Optionally select a colour variable to colour all plots by a grouping variable

This uses R's gpairs package to produce the matrix.