Event History Calendar

Key Components

The primary aim of the event history calendar approach is to maximize the accuracy of autobiographical recall. Just as event history calendars represent the past both thematically and temporally, the structure of autobiographical knowledge is believed to be organized in a similar fashion. Theoretically, the thematic and temporal associations of events within the structure of autobiographical knowledge afford retrieval cues that can be implemented in event history calendar interviewing and aid respondents to reconstruct their pasts more completely and accurately. One type of retrieval cue involves the sequencing of periods of stability and the transitions between them with regard to what happened earlier and later in time within the same timeline. For example, one may remember that one's employment period in one company immediately preceded another period of employment with a different company. In between these periods resides the transition point from one period to another, and both the length of periods and the timing of transition points are recorded within event history calendar timelines. In addition to sequential retrieval, the use of parallel retrieval cues [Page 247]involves the remembering of events across timelines and domains that happened contemporaneously, or nearly so. For example, one may remember a period of unemployment that was contemporaneous with a change in residence from one location to another. Parallel retrieval is particularly effective if the timing of one of the events is especially memorable, as this memory will locate the timing of the other event as well. It is the use of the respondents' own remembered events as cues to recall less easily retrieved information that is hypothesized to lead to benefits in data quality, and it is this requirement that necessitates conversational flexibility in interviewing.

Other key components of event history calendar design include representing thematic aspects of the past into domains, such as residence and labor, and the capturing of temporal changes by the inclusion of one to several timelines within each domain. For example, a labor history domain may include separate timelines to collect temporal changes in the amount of work, periods of unemployment, and periods in which one had been out of the labor force. In addition to the length of the reference period, instrument designers need to determine the smallest units of time in which life events are to be located, whether years, months, or in some instances, thirds of a month. For longer reference periods, larger time units should be used, so as to provide participants with a level of temporal detail that is best matched to how finely tuned the timing of events can be reconstructed in memory and to lend to both interviewers and respondents a manageable number of units. The first domains that should be queried are those whose events are most easily remembered, to motivate responding and also to lay out a framework in which more easily remembered events can be used as cues in the remembering of events that are queried later in the interview. Requesting respondents to provide “landmark events,” such as the timing of holidays and birthdays, can be an effective first domain when used in this fashion, but landmarks appear most beneficial for shorter reference periods. With longer reference periods, it may be best to ask respondents to trace their residential histories, which helps respondents to map temporal locations with physical ones. As part of their use of flexible conversational interviewing, event history calendars can be implemented by allowing interviewers and respondents to return to domains once covered, although usually the interview flows by proceeding from one domain to the next.

...