Single Variable Quality Check

Single Variable Quality Check

Usage

checkdata(
  data,
  var = NULL,
  showOutliers = TRUE,
  showDistribution = FALSE,
  showDuplicates = FALSE,
  showPatterns = FALSE,
  rareCategoryThreshold = 5,
  clinicalValidation = TRUE,
  unitSystem = "auto",
  outlierTransform = "none",
  mcarTest = FALSE,
  cvMinMean = 0.01,
  showSummary = FALSE,
  showAbout = FALSE,
  showCaveats = FALSE
)

Arguments

data

.

var

.

showOutliers

Detect potential outliers using three methods - z-score (|z| > 3), the IQR fence (1.5 x IQR), and the modified Z-score (|M| > 3.5, MAD-based) - and report points flagged by at least two of them. Below n = 10 single-method flags are shown and labelled informative-only.

showDistribution

Display descriptive statistics and distribution characteristics.

showDuplicates

Identify and count duplicate values in the dataset.

showPatterns

Analyze patterns in missing data and value distributions.

rareCategoryThreshold

Percentage threshold for flagging rare categories (important for chi-squared assumptions and modeling).

clinicalValidation

Perform context-specific validation for clinical variables (age, lab values, etc.). Heuristic ranges may need adjustment.

unitSystem

Unit system for clinical plausibility checks. Auto-detect attempts to infer from data ranges.

outlierTransform

Apply transformation before outlier detection to handle skewed distributions (especially right-skewed lab values).

mcarTest

Explain whether the missingness mechanism can be tested formally for this variable. Little's MCAR test is multivariate and cannot be computed from a single variable, so this analysis reports heuristics about where the missing values sit rather than a test of the mechanism; enable this to state that limitation explicitly in the output.

cvMinMean

Suppress coefficient of variation when absolute mean is below this threshold (avoids instability).

showSummary

Display a plain-language summary of data quality suitable for copying to reports.

showAbout

Display information about data quality assessment methodology.

showCaveats

Display important limitations and assumptions of the quality assessment.

Value

A results object containing:

results$notices					a html
results$todo					a html
results$qualityText					a preformatted
results$missingVals					a table
results$noOutliers					a html
results$outliers					Shows outliers detected by at least 2 of 3 methods: Z-score (|z|>3), IQR (1.5×IQR rule), Modified Z-score (MAD-based |z|>3.5). Points flagged by only 1 method are NOT shown.
results$outlierMethodSummary					Summary of each outlier detection method. These are heuristic approaches; consider skewness and sample size when interpreting.
results$distribution					a table
results$duplicates					a table
results$patterns					a table
results$naturalSummary					a html
results$aboutAnalysis					a html
results$caveatsAssumptions					a html

Tables can be converted to data frames with asDF or as.data.frame. For example:

results$missingVals$asDF

as.data.frame(results$missingVals)