Basic workflow
library(ggplot2)
library(GoldenVizR)
p <- ggplot(mtcars, aes(wt, mpg, colour = factor(cyl))) +
geom_point() +
labs(
title = "Fuel economy by weight",
subtitle = "Motor Trend cars",
x = "Weight",
y = "Miles per gallon",
colour = "Cylinders",
caption = "Source: mtcars"
)
report <- analyze_plot(p)
report$summary
## checked_rules implemented_checks pass warning fail info status
## 1 25 25 24 1 0 0 warning
What the report contains
A GoldenVizR report is a list with four stable fields:
plot_summary |
Basic metadata extracted from the ggplot2 object,
including title, axis labels, geoms, layers, data size, facets, and
coordinates. |
rule_results |
One row per GoldenViz rule, with the rule id, family, name, status,
message, and suggestion column. |
status_counts |
Counts of PASS, WARNING,
FAIL, and INFO. |
summary |
One-row overview with checked rules, implemented checks, status
counts, and report status. |
Inspect rules that need attention:
subset(report$rule_results, status != "PASS")
## rule_id family rule status
## 15 R15 Readability Direct labeling when useful WARNING
## message suggestion
## 15 Direct labels may improve readability. <NA>
Manual analysis
Manual analysis means passing a specific ggplot2 object
to analyze_plot() and inspecting the returned report
yourself. This is useful in notebooks, teaching material, package tests,
dashboards, or review workflows where you want explicit control over
which chart is checked.
manual_plot <- ggplot(mtcars, aes(factor(cyl), mpg)) +
geom_boxplot() +
labs(
title = "Fuel economy by cylinder count",
x = "Cylinders",
y = "Miles per gallon",
caption = "Source: mtcars"
)
manual_report <- analyze_plot(manual_plot)
manual_report$summary
## checked_rules implemented_checks pass warning fail info status
## 1 25 23 22 1 0 2 warning
Inspect the full rule table:
manual_report$rule_results
## rule_id family rule status
## 1 R1 Completeness Clear title PASS
## 2 R2 Completeness Axis labels PASS
## 3 R3 Completeness Missing values and gaps PASS
## 4 R4 Completeness Legend clarity INFO
## 5 R5 Completeness Source and context PASS
## 6 R6 Completeness Annotation for key message PASS
## 7 R7 Readability Readable text PASS
## 8 R8 Readability Accessible color choices PASS
## 9 R9 Readability Avoid clutter PASS
## 10 R10 Readability Grid and guide balance PASS
## 11 R11 Readability Consistent formatting PASS
## 12 R12 Readability Appropriate chart type PASS
## 13 R13 Readability Readable scale labels PASS
## 14 R14 Readability Overplotting management PASS
## 15 R15 Readability Direct labeling when useful INFO
## 16 R16 Readability Visual hierarchy PASS
## 17 R17 Integrity Truthful scale PASS
## 18 R18 Integrity No distorted baseline PASS
## 19 R19 Integrity Avoid misleading encodings PASS
## 20 R20 Integrity Handle outliers honestly PASS
## 21 R21 Integrity Proportional areas PASS
## 22 R22 Integrity Uncertainty when needed WARNING
## 23 R23 Integrity Avoid cherry picking PASS
## 24 R24 Integrity Respect data granularity PASS
## 25 R25 Integrity Neutral framing PASS
## message suggestion
## 1 Title text is present. <NA>
## 2 Both axis labels are present. <NA>
## 3 No missing values or gaps were reported. <NA>
## 4 No legend information was supplied. <NA>
## 5 Source or contextual caption is present. <NA>
## 6 No annotation need was reported. <NA>
## 7 Text size appears readable. <NA>
## 8 Color choices were reported as accessible. <NA>
## 9 No high clutter signal was reported. <NA>
## 10 Grid or guide information was supplied. <NA>
## 11 Formatting was reported as consistent. <NA>
## 12 Chart type information was supplied. <NA>
## 13 Scale label information appears readable. <NA>
## 14 No overplotting signal was reported. <NA>
## 15 Direct-labeling need was not supplied. <NA>
## 16 Visual hierarchy information was supplied. <NA>
## 17 Scale was reported as truthful. <NA>
## 18 No zero-baseline requirement was reported. <NA>
## 19 No misleading encoding was reported. <NA>
## 20 No outliers were reported. <NA>
## 21 No area encoding was reported. <NA>
## 22 Uncertainty appears needed but was not reported. <NA>
## 23 No cherry-picking risk was reported. <NA>
## 24 Data granularity was reported as respected. <NA>
## 25 Framing was reported as neutral. <NA>
Most workflows start by filtering to rows that are not
PASS:
subset(manual_report$rule_results, status != "PASS")
## rule_id family rule status
## 4 R4 Completeness Legend clarity INFO
## 15 R15 Readability Direct labeling when useful INFO
## 22 R22 Integrity Uncertainty when needed WARNING
## message suggestion
## 4 No legend information was supplied. <NA>
## 15 Direct-labeling need was not supplied. <NA>
## 22 Uncertainty appears needed but was not reported. <NA>
You can also inspect a specific rule family:
subset(manual_report$rule_results, family == "Integrity")
## rule_id family rule status
## 17 R17 Integrity Truthful scale PASS
## 18 R18 Integrity No distorted baseline PASS
## 19 R19 Integrity Avoid misleading encodings PASS
## 20 R20 Integrity Handle outliers honestly PASS
## 21 R21 Integrity Proportional areas PASS
## 22 R22 Integrity Uncertainty when needed WARNING
## 23 R23 Integrity Avoid cherry picking PASS
## 24 R24 Integrity Respect data granularity PASS
## 25 R25 Integrity Neutral framing PASS
## message suggestion
## 17 Scale was reported as truthful. <NA>
## 18 No zero-baseline requirement was reported. <NA>
## 19 No misleading encoding was reported. <NA>
## 20 No outliers were reported. <NA>
## 21 No area encoding was reported. <NA>
## 22 Uncertainty appears needed but was not reported. <NA>
## 23 No cherry-picking risk was reported. <NA>
## 24 Data granularity was reported as respected. <NA>
## 25 Framing was reported as neutral. <NA>
Because the report is structured, it can be used
programmatically:
any(manual_report$rule_results$status == "FAIL")
## [1] FALSE
manual_report$status_counts
## PASS WARNING FAIL INFO
## 22 1 0 2
Render an HTML report
Use render_report() when you want a formatted HTML
version of the same structured review.
html_report <- render_report(report)
html_report
In an R Markdown document, the rendered report can be printed from a
code chunk. To save the same report as a standalone HTML file, use
save_report():
save_report(report, "goldenviz-report.html")
If your installed version includes view_report(), use it
for quick local inspection:
The HTML report is expected to include the same core sections as the
structured object:
## [1] "plot_summary" "rule_results" "status_counts" "summary"
The report can also be embedded directly inside documentation,
notebooks, and HTML reports:
The main areas of the report are annotated below:
