Documenting animal behavior with an ethogram
Build a short behavior catalog, define each entry so another observer can recognize it, choose a sampling rule, pilot the sheet, and record uncertainty instead of guessing at an animal's motive.
Scope: A non-invasive field method for describing visible animal behavior across taxa. It introduces operational behavior labels and common sampling rules, but it does not replace a study design tailored to a species, question, site, or animal-care review. · Last updated
Define actions, not interpretations
Begin with a small list such as feeding, traveling, resting, grooming, scanning, and out of view. Give each label a boundary that another observer could apply: “head down, manipulating or ingesting food” is testable, while “hungry” is an inferred state. Categories should cover the question without overlapping at the same level, and a notes field can preserve unusual actions before the catalog is revised. [1][2]
Separate states from events
States have measurable duration—resting or swimming, for example—whereas events are brief points such as a peck, leap, or alarm call. That distinction determines what you record. Instantaneous scans can estimate the proportion of sampled moments spent in states, while continuous recording can capture durations and event rates. A count of scans is not automatically a count of behavior bouts. [2][3]
Choose who and when before watching
Focal sampling follows one identified individual for a stated interval; scan sampling records members of a group at set moments; event sampling follows specified conspicuous acts. Ad libitum notes remain valuable for discovery but overrepresent what is dramatic or easy to see. Fix observation length, scan interval, time of day, and rules for switching subjects so sessions can be compared honestly. [3][4]
Pilot, calibrate, and keep uncertainty
Test the sheet before collecting the main series. Two observers can independently code the same short sequence, compare disagreements, and sharpen definitions rather than forcing agreement afterward. Record start and end time, subject identity if known, visibility, distance, weather, and disturbance. Mark “occluded,” “unknown,” or “not sampled” when appropriate; invented precision is harder to repair than missing data. [1][4]
Related guides
Identify it and save the field note.
Where this guide comes from
Source-checked editorial guide. Last updated . This guide teaches identification and field skills; it is not a substitute for expert verification when it matters.
