Shaw Task Mapping Question Megathread

[xehu]

Use this thread to ask questions about task mapping related to Shaw (1963).

To start, here is an introduction of Shaw's dimensions, as written by Larson (2010):

Very few systematic attempts have been made to understand the multidimensional structure of group tasks. By far the most comprehensive effort of this sort is a study conducted by Marvin Shaw (1963). Shaw analyzed the multidimensional structure of 104 group tasks. These tasks were gathered mostly from the extant experimental literature, although a few were developed by Shaw himself specifically for this study. Each task was fairly brief (i.e., could be accomplished in an hour or less), suitable for use in experimental research with college students, and could be described on a single sheet of paper. The full set of 104 tasks covered a diverse array of activities, including mathematical puzzles, arithmetic problems, sorting and transfer problems, word and sentence construction tasks, idea generation tasks, tasks that require the coordinated operation of some piece of apparatus, ranking tasks, and target search tasks. An example of a transfer problem analyzed by Shaw is the Milkman’s Quandary, shown in Figure 2.1.5.

Shaw (1963) analyzed the structural relationship among these 104 tasks by having 49 judges, mostly graduate students in psychology, rate the tasks on each of 10 attributes using the method of “equal appearing intervals” developed by Thurstone and Chave (1929).

...Based largely on the results of the factor analysis, Shaw (1963) concluded that the differences among the 104 tasks he studied could best be explained in terms of six underlying dimensions, which he labeled task difficulty, solution multiplicity, cooperation requirements, intellectual-manipulative requirements, population familiarity, and intrinsic interest. As can be seen in Figure 2.2, the first two dimensions were each defined by three attributes. Scores on the three attributes defining the first dimension (task difficulty) were very highly intercorrelated, as were scores on the three attributes defining the second dimension (solution multiplicity). Thus, on empirical grounds, it can be argued that these scores tap just two underlying features, not six. The remaining four dimensions were defined by just one attribute each. According to Shaw, however, the last two of these (population familiarity and intrinsic interest) are relatively weak, and he recommended that they be interpreted with caution. This was partly because the attributes defining these dimensions also had moderately strong loadings on one or more of the other dimensions. Further, a question can be raised about whether these two dimensions really tell us about the tasks per se, or whether they tell us, instead, about the experience and interests of those most likely to perform such tasks (college students in research settings). Still, all six dimensions identified by Shaw appear to be psychologically meaningful, in the sense of having a plausible impact on group process and performance. Population familiarity, for example, is likely to affect the extent to which different group members perceive a task in the same way, and so share a common conceptualization of how it should be performed.

Here is a screenshot from Larson (2010) summarizing these dimensions (Figure 2.2):

For the original work, please see: Shaw, Marvin E. (1963). Scaling group tasks: A method for dimensional analysis.
Shaw 1963 paper - small.pdf