Quality of Data Collected from a Vulnerable Population: Using the Calendar Method and Third-Party Help

In the field of social sciences, collecting extensive retrospective data such as life histories has been fraught with recall errors compromising the quality of the data. Collection of life histories gets even more challenging when it has to be done among vulnerable groups, such as poor households who are victims of involuntary displacement and resettle­ment due to disasters or infrastructure projects. Recognizing this, a tailor-made life history calendar was designed to gather data from households in Indonesia who were involuntarily resettled in a newly-built community. The substantive focus of this study is to obtain ‘social capital’ histories on the level of neighbourhood, households, and heads of households.  In addition to the usual aided recall features the calendar procedure included the option to allow, in certain parts of the interview, the help of “third parties” in the recollection of the requested information. These “third parties” are either household members or friends of the respondent who also reside in the same community. Collaborative recall generally increases accuracy of information that is shared socially between community members, since different people may remember different aspects of the memories. A natural field experiment was designed in which data quality was compared between the conditions ‘with help of third parties’ and ‘without help of third parties’. Data quality was assessed regarding ‘numbers’ (of household related transitions), ‘names’ (of community leaders) and ‘dates’ (years of public services). The study examines whether integration of “third-party help” in the calendar method can enhance the recollection process among respondents and may lead to better data quality. In addition, it was also discussed how “third-party help” may interact with calendar procedures and what consequences arise for related data collection procedures.