Define what was counted
Separate confirmed deaths, euthanasia where recorded under the chosen definition, unaccounted-for animals and other removals. Keep the reporting definition explicit and consistent. Different summaries may use different categories, so their totals should not be combined without reconciliation.
Use individual or stable event identifiers where the records permit. Repeated messages about one death should not become multiple deaths, while a later additional loss should remain a new event with its own time and source.
Establish the population and interval
Record actual group membership during the stated period, including births, arrivals, transfers and other changes. A population that changes over time may need a more careful analysis than one opening count. Ask the relevant analyst or professional which method answers the intended question.
| Reporting question | Information required |
|---|---|
| How many died? | Confirmed count and category definition |
| In which group? | Actual population and membership changes |
| Over what period? | Consistent start and end basis |
| Compared with what? | Comparable unit, selection and recording |
| With what uncertainty? | Missing animals, dates or source gaps |
Work through equal counts in unequal groups
In a fictional arithmetic example, two groups each record four confirmed deaths. One group contains forty animals during the defined interval and the other contains four hundred under otherwise comparable counting assumptions. Equal counts do not mean equal proportions.
The comparison still requires compatible periods and definitions. If one figure covers a month and the other a year, the simple proportions answer different questions. Preserve the raw counts and context rather than selecting whichever comparison looks most dramatic.
Keep missing records out of the zero category
No death recorded may mean no confirmed event, an uncompleted check or unavailable reporting. State which is known. A period when records were not collected should not create a reassuring run of zeros in a graph.
If a count is revised after reconciliation, retain the previous value, reason and date. A later reader should be able to distinguish a genuine additional loss from correction of an earlier duplicate or misclassification.
Separate the number from the cause
A mortality series describes events under a definition. It does not diagnose why animals died or identify the effect of a treatment. Keep veterinary findings, laboratory results and causal interpretations with their own scope and evidence.
When unexpected deaths are ongoing, obtain prompt professional and relevant official advice rather than waiting for statistical completeness. Use sudden-mortality reporting and prevalence-claim review for related distinctions. A clear count with an honest limitation is more useful than a precise-looking mortality percentage unsupported by the population record.
Keep revisions out of the event-date trend
If an old duplicate is removed today, the correction belongs to the original period and its revision history. It should not appear as a negative death on today’s event graph. Preserve both event date and correction date so the series can be reproduced and a reader can understand why a historical total changed.
Reconcile a changing goat inventory before choosing a denominator
Consider a fictional monthly goat ledger: 50 animals at opening, ten births, five purchases, eight sales and two recorded deaths. With no other entries or departures, the closing inventory reconciles to 55. That check can reveal an omitted movement; it does not make 55 the automatic denominator for the two deaths.
The animals entered and left at different times. Preserve those actual dates and identify whether the intended question concerns the opening cohort, a birth cohort or all animals contributing observation time during the month. If the relevant population history remains unavailable, report the two recorded deaths with the defined period and the denominator gap. Neither the closing inventory nor a guessed mid-month count repairs the missing history. Keep kids and adults distinct where the reporting question calls for separate cohorts, without inventing age-specific risk from a combined total.
Compare pond counts through their actual population histories
For farmed fish, retain stocking, transfer and harvest records as the actual inventory events they describe. Preserve the actual event type and date instead of treating every removal as a recorded death. If two location totals cover different membership histories or observation periods, display those limits with the counts before attempting a comparison.
Cite this reference
LivestockDefense. “Report farm-animal mortality counts without inventing a rate.” Updated 2026-10-30. https://livestockdefense.com/guides/farm-animal-mortality-statistics-without-denominators/
Add the date you accessed the page. This identifies the current reference; keep a permitted saved copy if your work depends on its exact wording. A citation alone does not establish support for a claim.
How to cite a page and its revision ↗