Unaided Recall

One of the first decisions a designer of a questionnaire must make is whether the information sought can best [Page 916]and most appropriately be gathered from respondents by asking questions with or without clues to the possible or most appropriate answers. A question-asking strategy that relies on as few cues as possible, or even none whatsoever, in order to encourage the respondent to spontaneously mention items of interest is known as unaided recall. Such question tools are commonly used in marketing research and other situations in which it is desirable to know the respondent's reaction to questions unprompted by previous questions or topics mentioned in previous questions. In social science surveys, it is often desirable to elicit answers unstructured by other specifie political or social questions. These types of subtle cues and reminders have a way of prompting, structuring, and defining the range of suitable ideas and events to be recalled.

Unaided recall questions typically ask respondents to mention specifie instances of more general phenomena that are typically construed rather broadly. For example, an unaided recall question might ask respondents to describe all television news programs they watched in a given day or week. Such a question design places the cognitive burden on the respondent to remember, name, and categorize his or her answer. The strength of such a question is that it allows the respondent to respond naturally, that is, unprompted by various cues as to what kind of specific information is sought. However, this question format can be subject to a variety of errors, including accuracy and omission. On occasion, for certain question topics, accuracy can be assessed by comparing the respondents' unaided responses to various types of records. Omission of useful information is likely when questions are unaided because particular information might easily be forgotten or considered irrelevant by the respondent. Cues, as in aided recall, can improve memory for suitable events and ideas by providing examples, but they can also introduce error by suggesting inappropriate ideas to the respondent. Most importantly, however, cues can affect and even structure the kinds of ideas and events that the respondent mentions.

People's memory for particular things may fail for many reasons, including that they may have never really known the information, have honestly forgotten it, or they are unwilling to spend the required time and mental effort to recall completely and make sure their answers are as accurate as can be. People might remember the information incompletely, recall it incorrectly, or express it poorly so that it is misunderstood by the researcher.

However, sometimes researchers want to know exactly whether a respondent can instantaneously remember (recall) something without being prompted with any clues, such as when branding studies are done and respondents are asked on the spur of the moment to come up with a name of a company (e.g., Please name a company that makes digital video recorders). In these cases, it is informative to the survey sponsors (e.g., the client company) to know what portion of the sample is able to spontaneously generate their name and their competitors' names without any prompting. In such a case, unaided recall would be the preferred research strategy. This does not preclude use of additional aided recall questions later in the survey.

...