Satisficing

Forms of Satisficing

Respondents who devote less-than-optimal effort to the task of answering questions can engage in weak or strong satisficing. Weak satisficing occurs when a respondent performs all four cognitive steps but performs one or more of these less carefully or attentively than is needed to optimize. A respondent implementing weak satisficing may be less thoughtful in inferring the intended meaning of a question, less thorough in searching memory for all relevant information, less balanced in integrating the retrieved information into a summary judgment, and more haphazard in selecting the appropriate response option from the list offered. Strong satisficing occurs when a respondent skips the retrieval and judgment steps altogether and seeks merely to identify a plausible answer based on cues provided by the question, without reference to any internal psychological cues directly relevant to the attitude, belief, or event of interest to the researcher. If no cues pointing to such an answer are immediately evident in a question, a satisficing respondent may choose a response at random. Strong satisficing allows a respondent to provide a reasonable and seemingly defensible answer while applying very little effort. Rather than making a sharp distinction between weak and strong satisficing, Krosnick proposes that an individual's response to any given question can fall somewhere along a continuum ranging from optimizing at one end to strong satisficing at the other.

Conditions under Which Satisficing is Likely

Krosnick has hypothesized that the likelihood a survey respondent will satisfice is a function of the respondent's ability to perform the cognitive tasks of optimizing, the respondent's motivation to perform the tasks, and the difficulty of the tasks. Satisficing should be more common when the respondent has less ability to optimize, when the respondent is less motivated to optimize, and when the tasks are more difficult.

...