INDEXING AS PROBLEM-SOLVING - A COGNITIVE APPROACH TO CONSISTENCY

Citation
C. David et al., INDEXING AS PROBLEM-SOLVING - A COGNITIVE APPROACH TO CONSISTENCY, Proceedings of the ASIS annual meeting, 32, 1995, pp. 49-55
Citations number
26
Categorie Soggetti
Information Science & Library Science","Information Science & Library Science
ISSN journal
00447870
Volume
32
Year of publication
1995
Pages
49 - 55
Database
ISI
SICI code
0044-7870(1995)32:<49:IAP-AC>2.0.ZU;2-J
Abstract
It is well known that indexers differ significantly in their judgment as to which terms reflect most adequately the content of a given docum ent. For more than three decades, studies on interindexer consistency identified several factors associated with low consistency, but failed to provide a comprehensive model of this phenomenon. Our research pro gram is an attempt to apply theories and methods from cognitive psycho logy to the study of indexing behavior. From a theoretical standpoint, indexing is considered as a problem solving situation. The research e ffort then consists in identifying the subject's problem space, that i s, the set of declarative and procedural knowledge used to solve the p roblem, and uncovering the strategic decisions that may lead to differ ences in chosen keywords. In order to access to these cognitive proces ses, three kinds of verbal reports are used: 1) thinking aloud protoco l collected while the task is in progress, 2) retrospective comments a nd answers provided by the indexer when confronted with a video record ing of the task, 3) indexer's comments on lists of keywords produced b y several other indexers for the same document. We will present result s of an experiment in which eight indexers (four beginners and four ex perts) were asked to index the same four documents with two different thesauri. It will be shown that the three kinds of verbal reports prov ide complementary data on strategic behavior, and that it is of prime importance to consider the indexing task as an ill-defined problem, wh ere the solution is partly defined by the indexer him/herself.