You divide measurements into groups labelled “0–10” and “10–20.” A value of exactly 10 appears, and the labels no longer tell you where it belongs. The arithmetic may be simple, but the grouping rule is incomplete.
State which edge is included
For an invented exercise, define the first group as values at least 0 and less than 10, and the second as values at least 10 and less than 20. Under that rule, 10 belongs only to the second group.
Other conventions can be valid, but they must be stated consistently. Check the first and last edges as well. A value of 20 falls outside these two groups as defined; it needs another group or an explicit out-of-range treatment.
Test the awkward values before the easy ones
Try 0, 9.9, 10, 19.9 and 20 on paper. Then run the same examples through the tool you intend to use. If the output differs, inspect its interval options rather than relabelling the chart to hide the discrepancy.
Our article on combining filter conditions is a useful companion. A bin normally needs both a lower and an upper condition, and a missing condition can admit values you did not intend to count.
Keep the rule beside the chart
Write a short legend or note that describes the boundaries in language your reader can follow. Preserve the precise rule in the analysis record, especially if labels are shortened for display.
Finally, reconcile the bin counts with the number of included observations. Check separately for missing or out-of-range values. A tidy set of bars is easier to trust when every eligible observation has a clear place and no boundary value has been counted twice.