Table 3. Item difficulty for tasks Single Multiple In
scan and locate tasks, difficulty is conditioned by
the number of pieces of information that the reader
needs to locate, by the amount of inferencing
required, by the amount and prominence of
competing information and by the length and
complexity of the text.
Multiple document
search difficulty is
conditioned by the number of texts,
the complexity of the document
hierarchy (depth and breadth),
familiarity of the structure, the amount
of non-hierarchical linking, the
distance to the goal, the salience and
relevance of the headers and the
dissimilarity of each of the physical
presentation/structure of the sources
(lack of parallelism in different source
texts)
In
literal and explicit meaning and integrate and generate inferences tasks , difficulty is affected by
the type of interpretation required (for example,
making a comparison is easier than finding a
contrast); by the number of pieces of information to
be considered; by the degree and prominence of
competing information in the text; and by the nature
of the text: the less familiar and the more abstract
the content and the longer and more complex the
text, and the lower the coherence of the structure,
the more difficult the task is likely to be.
In multiple documents,
inference difficulty is conditioned on the number
of texts, the salience of the headers,
the similarity of content (e.g.
discrepancy in text
content/arguments, variability in point
of view), the dissimilarity of the
physical presentation/structure of the
sources (lack of parallelism in different
source texts) and the explicitness of
source information.
In