How to Export a Graph from RStudio for a Poster

A practical guide to making poster-ready figures in R with ggplot2 and exporting them at fixed dimensions using ggsave().

Overview

When creating a poster in PowerPoint, figures should be designed and exported from R at their final size rather than resized later.

This requires a deliberate workflow that controls both the appearance of the plot and how it is saved.

• Create figures using ggplot2
• Use a large base font size suitable for posters
• Export figures with ggsave() using fixed dimensions (in inches)
• Insert into PowerPoint without major resizing

This produces figures that are readable, properly scaled, and consistent across the poster.

The Problem

A common issue is getting a figure from R into PowerPoint that is actually readable on a poster.

Many students use approaches that seem convenient but produce poor results:

• copying a figure from the Plots tab in RStudio
• using the Export button in the Plots tab adjusting image size and font size
• taking a screenshot of the graph

These approaches do not preserve control over:

• figure size
• text size
• resolution
• aspect ratio (length to width ratio)

As a result, when the figure is placed into PowerPoint and resized:

• text becomes too small to read
• images appear blurry or pixelated
• plots become stretched or distorted
• facet panels become cramped
• figures look inconsistent across the poster

General Approach

The solution is to design and export figures in R at the exact size they will appear in PowerPoint.

This means:

• deciding how much space the figure will occupy (e.g., half-column or full-column)
• setting text size appropriate for poster viewing
• exporting the image with fixed width and height in inches
• avoiding major resizing after insertion into PowerPoint

This approach ensures:

• readable text
• clean image quality
• correct proportions
• consistent layout across figures

Recommended workflow

1. Create and store the graph

• Use ggplot2 to build your figure
• Assign the plot to an object using a descriptive name (e.g., penguin_scatter)
• Build the full plot before saving (do not rely on the last displayed plot)
• Use clear, consistent naming for figure objects
• Preview the plot in RStudio to check appearance before exporting

2. Choose a theme and set base_size = 20

• Start with your existing plot object (e.g., penguin_scatter)
• Add a theme layer with an appropriate base size, and assign the result to a new object (e.g., fig1)
  • Example: fig1 <- penguin_scatter + theme_minimal(base_size = 20)
• This ensures:
  • the plot is readable in the RStudio Plots tab
  • the saved image will also have appropriately sized text for a poster
• Recommended theme options:
  • theme_minimal(base_size = 20)
  • theme_bw(base_size = 20)
  • theme_classic(base_size = 20)
  • theme_light(base_size = 20)
• The base_size setting controls most text, including:
  • axis titles and labels
  • legend text
  • facet labels
• Adjust other visual elements directly in geoms as needed:
  • point size → geom_point(size = ...)
  • line width → geom_line(linewidth = ...)
  • text labels → geom_text(size = ...), geom_label(size = ...)

3. Export with ggsave()

• Use ggsave() to save your figure as a PNG file
  
• Include the following arguments:
  • filename = (name of the output file, e.g., "penguin_scatter.png")
  • plot = (the plot object, e.g., fig1)
  • width = (in inches)
  • height = (in inches)
  • units = "in"
• Choose width based on how the figure will be used in PowerPoint:
  • full-column figure: width = 13.5
  • half-column figure: width = 6 to 7
• Set height intentionally based on the content:
  • adjust to maintain good proportions
  • ensure text, points, and facets are not crowded
  • there is no single correct value; expect to iterate
• Use PNG format for clean, high-quality images suitable for posters

4. Open and inspect the saved file

• Go to the Files tab in RStudio
• Locate the exported image file (e.g., "penguin_scatter.png")
• Click the file to open it
• Check the figure carefully:
  • text size is readable
  • points and lines are appropriately sized
  • nothing looks crowded or cut off
  • proportions look correct (not stretched)
• If the figure does not look right:
  • adjust width and/or height in ggsave()
  • revise theme or geom sizes as needed
  • re-save and check again

5. Insert the image into PowerPoint

• In RStudio, go to the Files tab
• Check the box next to your image file (e.g., "penguin_scatter.png")
• Click More → Export…
• Save the file (typically to your Downloads folder)
• In PowerPoint:
  • Go to Insert → Pictures → This Device
  • Navigate to your Downloads folder
  • Select the image and insert it
• After inserting:
  • avoid large resizing (especially enlarging)
  • position the image in its intended space on the poster

Examples

The example below uses the penguins dataset from the palmerpenguins package.

Example 1: Full-column figure

This example saves a figure that takes up the entire width of a poster column, 13.5 inches.

Setting base_size = 20 yields a good font size for this image.

Example 2: Half-column figure

This version uses the same graph but creates a smaller image suitable for half of a poster column.

Because the image is smaller, there isn’t as much room for text on the figure. You can alleviate this problem by moving the legend to the bottom of the graph.

Then save the image using ggsave().

• Make the base font size smaller when saving the image (e.g. 16)
• Note: The theme() layer is repeated here because adding theme_bw() resets any theme elements that had been customized in the previous step.

Facetted graphs

• Facetted plots require additional layout control to remain readable on a poster
• Use facet_wrap() with the ncol argument to control the number of columns
• Adjust ncol based on the width of the figure:
  • half-column figure:
    • use ncol = 1 or ncol = 2
  • full-column figure:
    • use ncol = 3 or more
• Choose a layout that:
  • keeps each panel large enough to read
  • avoids overcrowding
  • makes efficient use of available space
• The best choice depends on:
  • the type of plot
  • the number of facets
  • how much detail each panel contains
• Expect to adjust both ncol and figure height to get a clean result

Example 3: Facetted figure

This example uses facets and adjusts the facet layout with ncol.

Note that the color legend can be removed because it is now redundant with the facet labels.

To export that image for a full column width:

Example 4: Facetted figure with vertical layout

For a narrower version of the same facetted graph, you might change the facet layout to ncol = 1. This creates a vertical stack of panels that can be saved as a half-column figure.

Like the half-column example above, the resulting figure will have a title that is too large and a color legend that takes up too much space on the side of the graph. Rather than making the base font size smaller on this graph (its still a large, featured graph, just vertically oriented), we can tweak some other elements:

• Keep the title from being too long, by giving the plot a new title using labs():

  • Make the title shorter, e.g. title = "Penguin Measurements"

  • Move info to an optional subtitle, e.g. subtitle = "Body mass vs. flipper length"

  • Keep the original title, but put in a manual line break \n, e.g. title = "Penguin body mass vs.\nflipper length"

• Move the legend to the bottom of the graph, and style is so it looks nice there, using theme():

  • Move it to the bottom with legend.position = "bottom"

  • Put the legend title above the legend with legend.title.position = "top"

  • Center the legend title with legend.title = element_text(hjust = .5)

Here is the figure with a single column of facets, a wrapped title, and a repositioned legend:

To export this version for a half-column width: