Geoscience Reference
In-Depth Information
estrus. It is logically impossible to be certain that clip marks had no effect on
behavior, but as in the case of Hoogland's (1995) Nyanzol dye-marked ground
squirrels, we could detect no evidence that it did. However, absence of evi-
dence is not evidence of absence, so caution is necessary. For example, bird
rings or ear tags have been demonstrated to affect judgments of asymmetry
and mate choice in some species (Burley 1988). This topic is further discussed
by Bekoff and Jamieson (1997) (see also Moehrenschlager et al. in press). In a
similar way, names of animals should be carefully chosen to avoid biasing
behavioral observation of social role. The name given to an animal, be it that
of a colleague, friend, or personality, can inadvertently influence all but the
most objective observer. Even ascribing “male” and “female” names can cause
recording biases. Numbers are better (although “A1” may still find deferential
recording to “C5”), but can be rather hard to remember. In our own badger
study we primarily used names derived from the appearance of the identifying
fur clip mark, giving asexual and fairly connotation-free names such as Line,
Dot, Dash, and Corners.
SAMPLING AND RECORDING RULES
Sampling rules dictate which subjects to watch and when. Martin and Bateson
(1993) distinguish ad libitum, focal, scan, and behavior sampling. Recording
rules specify how the behavior is to be recorded. The two options are continu-
ous recording and time sampling. Time sampling may be undertaken on an
instantaneous or one-zero basis.
The qualities of each sampling rule are easily envisaged through the eyes of
contrasting onlookers at a football match. The typical spectator engages in ad
libitum sampling and samples everything. He or she is likely to record the
goals and any spectacular fouls, with a bias toward the fouls committed by cer-
tain star players and those committed on the ball. In contrast, a talent scout
might be a focal sampler, observing one player at a time. He or she may gain a
poor impression of the game as a whole, but a clear impression of the qualities
of particular players. Of course, the scout must be alert to biases such as the
stage of the match at which a given player becomes the focal subject (it would
be inappropriate to compare one player observed in the first 5 min with
another observed in the final 5 min). A team manager might opt for scan sam-
pling, censusing the behavior of each player in turn, at rapid intervals, to get
an overall impression of how the team functions as a unit. Finally, a gambler,
interested only in the outcome of the game, may undertake behavior sampling,
recording only goals. By focal or scan sampling one could miss these events.
