Software > Household Registration System > What's New in HRS 2.2?

What's new in HRS 2.2?

HRS 2.2 was released in October of 2000. It includes a number of changes and features that are the result of comments from HRS 2 users. Experienced developers should be able to alter previous versions of the HRS to incorporate the new features. At the moment, there is no automated machinery in place to update previous versions of the HRS.

Specific changes incorporated into the HRS 2.2 include

  1. The demographic rates and life table programs now include options to do calculations on subsets of the population. Users specify a subset of individuals or locations in selection files. These files consist of two columns: the ID and a field called inclflag. If the inclflag is set to 1, then the individual or location is included in the analysis. This general mechanism allows for almost any subset of the population to be specified.
        
  2. Consistency logic has been changed to allow inconsistent records to coexist with consistent data. Previous attempts to bring legacy data into the HRS were made difficult because of the tight regime of consistency logic. The row level checks in the table were the primary places where the inconsistencies were detected. The row level checks of the HRS2.2 now include code to skip over any record that has the status_dat (data quality status of that record) field set to "F". The output calculations can skip over this inconsistent data.
        
  3. The find option in the data entry screens has been made more efficient. Previous versions would filter out all records that did not meet the specified criteria. Filtering can take a relatively long time (up to 20 seconds for large datasets). HRS 2.2 includes the filter option as well as an option to locate the first record that satisfies a user specified criteria. Finding the first record takes a negligible amount of time.
        
  4. HRS 2.2 has two forms for managing updates to the population. The default is the old update form. A new update form allows a data entry clerk to type in the ID of an individual who has had some demographic event occur to him or her. Some data entry clerks prefer to type the ID rather than scroll down a list of IDs. The new form is named "updatebyid" and is found in the list of forms in the project.
      
  5. The new distribution comes with no data in the tables. It also has a file with sample data for those users who would like to have examples.

Bruce MacLeod
macleod@usm.maine.edu



Print this page

@
E-mail this page

This page updated
8 April 2005