The paper presents a strategy for designing relational databases with the program FileMaker to study the histories of individuals and organizations. The approach facilitates efficiency in entering data and flexibility for constructing statistical analyses from the raw data. The key feature of the strategy is to define the basic unit of observation in the database in terms of an agent, an event, and a date. Given that programs such as FileMaker can easily sort data by agent and date, once you structure the data correctly you can construct well-ordered event histories for agents, even if the researcher may enter the data in an unordered fashion. By using events that happened to an agent at a particular time as the basic unit of observation, you maintain maximum flexibility to do statistical analysis that aggregate basic data in different ways. The paper illustrates the power of the approach by outlining how to analyze changes in geographic distances between two events marking the life history of chemists.
[1]
J. Murmann.
Knowledge and competitive advantage: The coevolution of firms
,
2003
.
[2]
Charles Tilly,et al.
Popular Contention in Great Britain
,
1996
.
[3]
John Scott.
Social Network Analysis
,
1988
.
[4]
Stanley Wasserman,et al.
Social Network Analysis: Methods and Applications
,
1994
.
[5]
G. Alter,et al.
Casting spells: database concepts for event history analysis.
,
1999
.
[6]
S. Lenway,et al.
Knowledge and competitive advantage, the coevolution of firms, technology and national institutions
,
2003
.
[7]
Charles Tilly,et al.
Popular Contention in Great Britain, 1758-1834.
,
1995
.