Count violations of contradiction rules defined in the cross-item-level metadata. These rules are written using a REDCap inspired format. The higher the number or percentage of rule violations, the lower the data quality.
Note: if you see a 1 in the legend of the plot, this indicates that there is a rule violation.

• Logical contradictions
• Empirical contradictions

Check applied to categorical variables. The categories observed in the study data should be present in the allowed categories listed in the metadata. The higher the number or percentage of non-matching categories, the lower the data quality.
Note: OBSERVED_CATEGORIES = categories present in the study data; DEFINED_CATEGORIES = categories defined in the metadata; NON_MATCHING = categories present in the study data but not defined in the metadata; NON_MATCHING_N = total number of observational units with mismatches; NON_MATCHING_N_PER_CATEGORY = number of observational units with mismatches per undefined but found category.

• Inadmissible categorical values

Check applied to numerical or time-date variables. If a range of values is provided in the metadata, the presence of values outside the interval is checked. These can be: inadmissible values (hard limits), improbable but plausible values (soft limits), or values outside measurement ranges (detection limits). The higher the number or percentage of values outside the limits, the lower the data quality.
Attention: values outside hard limits are removed from the following quality checks.

• Inadmissible numerical values
• Inadmissible time-date values
• Uncertain numerical values
• Uncertain time-date values