When the prosecuting attorney asks the classic question, Where were you on the night of January 3, 1965: the witness always seems to know. This may be the result of prior coaching; but the witness saw a murder on that night, it is quite likely that it would be remembered for the rest of the witness’ life. Thus, the importance of the event has much to do with its recollection. Few consumers will recall the brand of canned peas purchased three weeks ago, but most can remember the brand of car bought years ago. Unusual events are remembered well. Most people have trouble remembering all of the television shows they watched the night before, but everyone who saw the first game of the World Series probably remembered it for at least several days. Events that are part of an established pattern or that interrupt an established pattern are remembered better than unrelated events. Consumers who always buy Colgate toothpaste will remember what brand they bought last time than the person who has no established favorite. This is true even if the Colgate consumer bought Crest the last time.
Any woman remembers her wedding date, but most of the bridesmaids would be unable to recall it a few years later. The difference is in importance of the event to the individual. Researchers cannot change the ability of person to remember things, but can try to ask the person most apt to remember.
Traditionally the homemaker has been sought as the respondent on questions about household purchases because she made most of such purchases and established habits that helped her to remember purchases and factors surrounding them. With the advent of two career families and the changes taking place in the roles of husband and wife, the latter is no longer a clear choice as the family respondents.
Researchers can control the length of time between an event and the date of their questioning only by limiting their questions to events that have happened recently. This means that questions on purchases are usually limited to the most recent purchase; questions on television programs watched are limited to the previous day, the immediately preceding hours or even to the current moment; and questions on the readership of magazines are asked on the latest issue.
Ability to remember is influenced by the stimulus that calls for the remembering. Thus, it is easier for respondents to recognize a past event without any clues. The latter is referred to as unaided recall. Suppose a respondent is asked. What TV programs did you watch last night? Experiments have shown that such a question may locate as few as 10 percent of those who actually watched a program. Many more can recall the same programs if they are given a list of all programs from the previous evening and asked which ones they watched. This is called aided recall.
The considerable advantage of aided recall that comes with stimuli to the respondent’s memory is, at least partially offset by the bias resulting from the suggestions offered by each such stimulus. Every multiple choice question is subject to suggestion bias of this type. The extent of this bias and methods of reducing it are discussed in the on multiple choice questions separately.
An extreme of the aided recall approach is the recognition method of stimulating recall. Instead of describing possible answer, researchers show the respondent the actual items in which they are interested. To determine the readership of advertisement researchers often go through magazines with respondents showing the each advertisement and asking if they remember having seen it before. This recognition method furnishes an excellent stimulus to memory but it is also subject to suggestion bias as is all aided recall.