John Bosley, Monica Dashen, and Jean Fox (1999)
"When Should We Ask Follow-up Questions About Items in Lists?," Proceedings of the Section on Survey Research Methods, American Statistical
Surveys often require respondents to recall a number of items belonging to a class or category, and also to recall significant facts or attributes about each recalled item. For example, the survey may ask for recall of purchases within a category such as "furniture" during some reference period, and to also recall for each item retrieved from memory some descriptive details (item attributes), such as price, date of purchase, model number, and the like. This purpose of the study was to examine the effects of two different ways of presenting a complex free recall task on the amount and accuracy of information retrieved from memory. The recall task focused on retrieving multiple instances of item within named categories, along with descriptive item details. The study findings are intended to apply to recall of purchased items, and so an experimental procedure simulating actual purchases was carried out on all study participants. Their simulated purchase and pertinent details about each one were recorded to serve as a basis for assessing accuracy of recall. After an interval during which they worked on a distracter task, participants were presented with a recall task. In one condition, they were asked to recall several items within a category before being asked to supply detail about any of the items. In the second condition, they supplied details about each item as soon as it was recalled. Outcome measures included the number of items per category recalled, accuracy of recall of both items and item detail (attributes), ratings of perceived task difficulty, and frequency of verbal and non-verbal behaviors indicative of fatigue, boredom and other signs of motivational or affective reactions to the task, or of cognitive burden. Study findings will be presented and discussed from the perspective of what cognitive and motivational factors account for any differences in the effectiveness of the two methods for eliciting information of this nature.
Last Modified Date: July 19, 2008