Abstract
The Intermediate Data Structure (IDS) provides a common structure for storing and sharing historical demographic data. The structure also facilitates the construction of different open-access software to extract information from these tables and construct new variables. The article Using the Intermediate Data Structure (IDS) to Construct Files for Analysis (Quaranta 2015) presented a series of concepts and programs that allow the user to construct a rectangular episodes file for longitudinal statistical analysis using data stored in the IDS. The current article discusses, in detail, each of these programs, describing their technicalities, structure and syntax, and also explaining how they can be used.
Highlights
The Intermediate Data Structure (IDS) was developed as a strategy aimed at simplifying the collecting and sharing of longitudinal demographic data (Alter & Mandemakers 2014; Alter, Mandemakers & Gutmann 2009)
Historical demographic data obtained from population registers or family reconstitutions is generally analyzed statistically by using event history analysis, for example, Cox models (Cox 1972; Therneau & Grambsch 2000)
The programs presented allow the creation of the Extended Intermediate Data Structure (EIDS) tables and append imported data to them, to construct a variable to indicate household size, to select variables for analysis from those included in the IDS and the EIDS tables, to produce a data extraction from them and to transform this data extraction into a rectangular episodes file that is ready for statistical analysis
Summary
The Intermediate Data Structure (IDS) was developed as a strategy aimed at simplifying the collecting and sharing of longitudinal demographic data (Alter & Mandemakers 2014; Alter, Mandemakers & Gutmann 2009). The structure provides a common platform to store data from different databases, regardless of their original form of storage Further development to such structure is being made by members of the European Historical Population Samples Network (EHPS-Net), who are creating open access programs to use the IDS to construct files for analysis. Historical demographic data obtained from population registers or family reconstitutions is generally analyzed statistically by using event history analysis, for example, Cox models (Cox 1972; Therneau & Grambsch 2000) To conduct these types of analyses, the data must be formatted as a rectangular episodes table. The article proposed a format for data extractions and for the output produced by open-access extraction programs It consists of a Chronicle file containing one row per each declaration of any variable or event to be included in the analysis and a Variable setup file describing the information contained in such extraction. Rectangularizes and reformats the information contained in the data extraction, giving an episodes table during output, which is a rectangular file that is ready for statistical analysis
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
