Improving behaviour classification consistency: a technique from biological taxonomy



AARE2002: Paper CHO02101

1. Determine the observable features for each category.

2. Choose one observable that will provide a suitable starting point for the key. Ideally, the
first decision in the key should be the least error-prone, and should divide the categories
into two roughly equal groups.

3. Continue repeating the previous step for the remaining observables within a group of
behaviours produced by a previous decision, until only one behaviour is left.

4. Check the observable features of the remaining behaviour category. If it is possible to
reach this end-point of the key without the remaining observables (if any) being present,
then the original list of categories must be incomplete. Create a new category of necessary.

5. Repeat steps 3 and 4 until all categories have been separated.

If two categories cannot be distinguished from each other by observable features, they must
be combined into a single category. If a particular category contains two distinct sets of ob-
servables, the category can be split in two separate categories. This can result in a more logical
classification scheme. However, the combined category may be more useful for the intended

Non-play

Absence of play

Exploratory

Explores play environment (examines or searches for play objects or watches play part-
ner ’s play)

Functional Play

Obvious play

Uses play object

Play object used in purely physical manner

Constructive Play

Obvious play

Uses play object

Play object used in purely physical manner

Spatial arrangement of play object(s)

Pretend Play

Obvious play

Symbolic or representational elements to play

Game with Rules

Obvious play

Play involves rules

Plays with partner

Figure 2: List of observable features of behaviours



More intriguing information

1. Migrating Football Players, Transfer Fees and Migration Controls
2. The name is absent
3. The name is absent
4. Imputing Dairy Producers' Quota Discount Rate Using the Individual Export Milk Program in Quebec
5. Effects of red light and loud noise on the rate at which monkeys sample the sensory environment
6. The name is absent
7. The magnitude and Cyclical Behavior of Financial Market Frictions
8. International Financial Integration*
9. The name is absent
10. FOREIGN AGRICULTURAL SERVICE PROGRAMS AND FOREIGN RELATIONS