RESEARCH SUPPORT
R Visualization with ggplot2 for Research
Figures Made for Journal Review
Figures built in R from your data or fitted model, with observations and uncertainty visible, palettes that survive grayscale and color-vision differences, and files exported at the size and format your journal specifies.
No Commitment Required
Strictly Confidential
In-House PhD-Level Statisticians
Fixed Project Fee
Same-Day Response
What Does Preparing Research Figures in ggplot2 Involve?
R visualization with ggplot2 builds a figure from a dataset by declaring which variables map to position, color and shape, which geometries draw them and which scales, labels and themes finish the result. For research the substantive choices matter more than the code: what the figure must show (distributions, estimates with uncertainty or change over time), whether raw observations stay visible next to the summaries, and whether the encoding survives grayscale printing, color-vision differences and reduction to a journal column. OnlineSPSS, a specialist statistical consultancy for academic, clinical and professional research, provides this support: an in-house PhD-level statistician builds the agreed figures in R from your data or model output and documents the script. What each figure is taken to show stays your decision.
WHEN TO ENGAGE
When a Draft Figure Stops Being Good Enough
A figure is often the last item made for a manuscript and the first one a reviewer reads, so the need usually becomes clear at submission or after a revision letter.
Journal Specifications Arrive The target journal sets a column width, a file format, a minimum resolution and font rules, and the existing plots were made for the screen at another size.
A Summary Hides the Data A bar chart of group means shows nothing of the spread, the sample size or the individual observations, and a reviewer asks to see them.
A Model Needs a Picture A fitted model calls for a figure of predicted values or effects with intervals, since the coefficient table alone does not convey them to readers.
Figures Differ Across a Paper Panels were made at different times with different fonts, colors and scales, and have to be rebuilt from one script into a consistent set.
WHAT THE FIGURE WORK COVERS
Design Decisions Behind a Research Figure
A figure is specified before it is drawn, starting from what it has to show and ending with the file that goes to the journal.
Purpose of the Figure
Each figure is tied to the research question it answers, the comparison it makes and the first thing the reader should take from it, before a plot type is chosen.
Showing the Data
Raw observations are drawn alongside means or model estimates, with jitter, transparency or faceting where points overlap, so that spread and sample size stay visible.
Uncertainty
Confidence or credible intervals are displayed with their meaning stated in the caption, and the plotted estimates come from the fitted model, not from a separate calculation.
Color and Access
Palettes are chosen to stay distinguishable under color-vision deficiency and in grayscale, with point shape or line type as a second channel for each group.
Layout and Export
Panels are combined and text is sized for the final print width, and ggplot2 figures for publication are exported with ggsave in the format and resolution the journal gives.
Figure Notes and Reporting
Axis labels with units, a legend stating what each bar or band shows and the sample behind each panel are set within Results Reporting & Interpretation, so the figure can be read on its own.
COMMON FIGURE ISSUES
Flaws Reviewers Notice in Research Figures
Plotting defaults in R publication graphics suit a screen, not a printed journal column, and reviewers tend to notice the flaws they leave before anything else.
Bars Without the Distribution Means with error bars are shown for small groups, which hides skewness, outliers and the number of observations behind each bar.
Truncated or Dual Axes A bar chart starts above zero, or two y-axes are layered, so the visual difference exaggerates the real one or depends on an arbitrary scaling.
Error Bars Unlabeled The caption does not say whether the bars show a standard deviation, a standard error or a confidence interval, which are different quantities of different width.
Color as the Only Cue Groups are told apart only by red and green or by a rainbow scale, which fails for some readers and in any grayscale printout of the paper.
Defaults at the Wrong Size Text and points sized for the screen become illegible, or crowded, when the figure is reduced to the width of a single journal column.
A Smoother Passed Off as the Model A trend line drawn by the plotting function is presented as the analysis result, though it is a separate fit that can differ from the model reported in the paper.
DELIVERABLES
The Figure Set and Its Supporting Files
Everything needed to regenerate the figures, resize them or change a single panel is returned together, so a later revision needs no rebuilding from memory.
Figure Script One commented R script that builds every figure from the analysis dataset or the model object, so that any panel can be rerun and edited.
Export Files Each figure exported in the formats and dimensions the target journal specifies, usually a vector file and a high-resolution raster file.
Shared Theme A single theme and palette applied across all figures, so that fonts, colors and line weights stay consistent through the manuscript.
Figure Annotation Notes Notes recording what is plotted, what any error bars or bands mean, the sample behind each panel and the model that produced the estimates.
Source Data Table The numbers behind each plot in a table, such as group summaries or predicted values with intervals, for a supplement or a journal that asks for source data.
Accessibility Note A note on how the palette behaves under color-vision deficiency and in grayscale, with an alternative supplied where a version was unclear.
ILLUSTRATIVE EXAMPLE
A Figure That Hid the Pattern in Its Bars
An illustration of how a first-draft figure is rebuilt from the fitted model so that it shows the data and the uncertainty.
ILLUSTRATIVE EXAMPLE · ALPINE PLANT PHENOLOGY RESEARCH · UNITED STATES
A postdoctoral researcher in plant ecology had recorded first-flowering dates for several species at sites along an elevation gradient over many years and needed the main figure for a manuscript.
The first draft was a bar chart of the mean flowering date by elevation band, which suggested a tidy ordering. The complication was that the species differed in sample size and in year-to-year variation, and the bars hid both. The statistician rebuilt the figure from the fitted model, with the observed dates as semi-transparent points, the model-predicted trend with its 95% confidence band, one panel per species and a palette that stayed distinguishable in grayscale, with point shape as a second cue. The type size was set for a single journal column, and the figure notes recorded what the band represented. The trend was worded as an association across sites, since elevation had not been assigned.
Outcome: The researcher received the figure script, the export files in the journal's format, the shared theme, the figure annotation notes and a table of the predicted values behind the plot.
This example does not describe a real client or project. We never share details of actual client work, in line with the confidentiality we promise every researcher. This illustrative example is included to show the type, scope, and standard of work involved in this kind of analysis.
SOFTWARE
Plotting Tools Beside ggplot2
Figures are built where the data and models already live, and a second tool is used when a collaborator or a journal workflow calls for it.
HOW WE ASSESS YOUR PROJECT
We Review Your Actual Research
Your assessment is based on a review of your actual research materials and analytical requirements, not a generic project category.
You Submit Your Project Details
Share whatever is available at the current stage, such as the research aim, protocol or proposal, dataset, variable list, manuscript or statistical requirements.
We Review Your Research and Define the Scope
Your statistically trained Research Coordinator reviews the materials to understand the study design, research questions, data structure, analytical requirements and expected deliverables. If clarification is needed, additional documents or information can be requested before the assessment is finalized.
Your Analysis Is Completed and Delivered
Once the scope and fixed project fee are agreed, the in-house statistician completes the work and the Research Coordinator remains the communication point through delivery, with the analysis, deliverables and timeline as agreed.
RESEARCH RESPONSIBILITY
What a Figure Claims Is Still Your Call
OnlineSPSS builds and documents the figures, while the message a figure carries and the conclusions drawn from it remain with the researcher.
Agreed Statistical Support OnlineSPSS provides the agreed methodological, analytical and statistical reporting support, defined in a scope that is confirmed before any work begins.
No Intellectual Contribution OnlineSPSS does not claim any intellectual contribution to your manuscript, report or article. Authorship remains the researcher's.
Your Research Decisions The research question, the choice of what to emphasize, the interpretation of each figure and the conclusions remain yours, with overall responsibility.
Confidential Handling Research data and project details are handled confidentially, project information is not published, and an NDA is available where required.
FREQUENTLY ASKED QUESTIONS
Questions About Publication Figures in R
A figure that looks right on screen can still fail review, and export formats, error bars and model-based plots are where it most often does.
What does ggplot2 figure preparation for a manuscript include?
R visualization with ggplot2 at OnlineSPSS covers designing each figure around its research question, building it from your data or fitted model, showing observations and uncertainty, choosing accessible colors, and exporting files to your journal’s specifications. Each figure in the agreed set is listed with its panels and the export formats the journal requires.
Which file format should a ggplot2 figure be exported in?
The journal’s author guidelines decide, so they are read before exporting. Vector formats such as PDF or SVG keep line art sharp at any size, while some journals ask for EPS, or for TIFF at a stated resolution. The ggsave function can write several formats, and dimensions are set in the units the journal gives, so the figure is not rescaled afterward.
Should error bars show standard deviation, standard error or a confidence interval?
They answer different questions. A standard deviation describes the spread of the observations, a standard error describes the precision of a mean, and a confidence interval gives a range of values compatible with the estimated quantity. The choice follows the message of the figure and is stated in the caption. With few observations, showing the points is more informative than any of them.
Can a ggplot2 figure display the results of my model?
Yes, when the plotted values come from the fitted model: predicted values or marginal estimates with intervals, computed from the model object and then drawn. A smoother added by the plotting function is a separate fit, not the model in the paper. The figure then matches the results table and carries the same limits of interpretation.
How do you protect my research data and confidentiality?
Your data and project details are handled confidentially, and project information is not published. Researcher names and institutions are not used publicly without permission. An NDA is available where required and can be arranged before sensitive materials are shared. Read more about confidentiality and data security.
What happens when I request an assessment, and how long does it take?
You share your research question, design and relevant materials, such as the dataset or model object, the target journal’s figure guidelines and any draft figures, and a statistically trained Research Coordinator reviews the request and responds the same day. The analysis, deliverables, timeline and one fixed project fee are then confirmed before any work begins, so the timeline follows from the agreed scope.
Still have a question? Find more answers in our full FAQ.
NEXT STEP
Figures Sized for the
Journal That Will Print Them
The journal's figure guidelines, the model behind each plot and any drafts a coauthor has seen decide how much is redrawn and how much is only resized.
No Commitment Required
Strictly Confidential
Fixed Project Fee
In-House PhD-Level Statisticians
Same-Day Response
Serving Researchers Since 2014