Sub-series consists of data files from the System for Human Resources Monitoring (SHuRM) and the successor Employment Equity Data Base (EEDB). These records are a partial reconstitution of the original data files by Treasury Board effected in 1996 and they do not represent exact copies of the original data files as they would have appeared in the live systems. The data from these two small applications permit the analysis of employment equity performance of government departments and the federal government as a whole in relation to the three designated groups who can only be accurately determined through self-identification: aboriginal peoples, individuals with disabilities and members of visible minorities. (Women, the fourth designated group under employment equity policy, are not included because employment equity data related to women is routinely collected through normal staffing procedures and because larger mainframe based information systems such as the Incumbent System can provide a comprehensive basis from which to analyse and monitor female participation rates.) The SHuRM and Employment Equity Data Base are true personnel data files which profile every (known) member of these three employment equity target groups willing to self-identify as members of the Public Service outside the executive category as defined by Schedule I, Part I of the Public Service Staff Relations Act. The two data bases represent the only source of reliable service-wide data in automated format in relation to these three groups. Automated data at the Public Service Commission, for example, only includes individuals who enroll in Special Measures Initiatives Programs, not the complete universe of employees to whom the programs are directed.
To begin building up the data bank, the Treasury Board mounted a government-wide survey in April 1985. A questionnaire distributed to every public employee invited any one in the target groups to identify themselves. About 10,000 or 14 per cent of the federal public servants chose to self-identify. Beginning in 1987, Departments collected the basic self-identification information from new (post-1985) employees on standardized forms. The Public Service Commission has supplemented the departmental input with data from Application for Employment forms and their data bases detailing participants in Special Measures programs in order to build a more complete data bank capable of accurately measuring participation rates of the three designated groups. Not until 1988 did all this effort actually produce a reasonably accurate file. The SHuRM master data file is comparatively simple, being composed of 19 core data elements and expanded modestly to 22 when the revised system was renamed Employment Equity in 1994. Apart from personal identification number and designated group identity code, the bulk of the fields are extracted from the Incumbent System. The common fields to both versions of the data base include name, initials, age, sex, three fields to cover the three designated group identities possible, salary, years of continuous service, department, occupational category, occupational group, subgroup (where applicable), level within occupational group, employment type, coded geographical location, a province code, and a text version of the city in which the work location is situated.
The Personal Identification Number appears in the National Archives version of all data files, it having been substituted for social insurance number retroactively on older SHuRM files by the creator before formal transfer to the National Archives. Four original principal fields in the SHuRM master file have been omitted from the reconstituted version: two internal control date fields, a death flag indicator for entries no longer found in Incumbent and therefore earmarked for deletion from the system, and a source field which explained whether the data came from the original SHuRM survey, the PSC, other Treasury Board sources or departmentally collected data. The extended portion of the SHuRM master data file incorporating 29 additional fields (composed of subcodes further breaking down designated group status for original survey entries) were not maintained by Treasury Board when the SHuRM system was superseded by the EEDB and that portion of the data files was lost. These minor changes and omissions mean that each archival data file does not look exactly like the original snapshots of the data base at calendar year ends, 1988 to 1993. The archival version of the data file contains the portion of the master file with enduring value. The three modest data elements added to the employment equity data base are coded bargaining unit designator, Yearsspen (pensionable service)and Cenaggl (another geo code). The common set of 19 variables in the two versions of the data file permit a modest level of quantitative analysis and verification of designated group participation rates and distribution. Cross referenced with Incumbent data, the data bases permit a certain level of longitudinal analysis and career profiles. The universe is defined as the 15,000 to 20,000 public servants whom the Treasury Board is able to verify as belonging to a designated group.
The master files for the older data files had to be reconstituted in 1996 by taking the annual snapshots of personal identification numbers from ShuRM/ EEDB and using them to extract the matching record from the corresponding year-end Population Reporting File from the Incumbent system. In addition, in creating the archival version of the file for transfer, the Treasury Board made an effort to eliminate all records that did not qualify for inclusion in the target group under the parameters of the policy (term employees of less than six months, students, casual employees, and EX or management category, in effect, the divisional director level up). While this effort does not appear to have been completely consistent at least for older files, this transaction means once again that the archival data file for a given year does not perfectly reflect the data file as it would have existed when the original snap shot was taken at a given year-end. The justification for this editing procedure is that the original data files were compiled out of a data collection process that, for administrative simplicity, was inclusive of all employees whether they qualified as part of the employment equity target population or not. The edited files in effect reflect the data incorporated into statistical output reports for the purpose of documenting group participation rates, consistent with Treasury Board policies, but the procedure has the consequence of limiting potential longitudinal studies, particularly in regard to the portion of the population moving on to the Executive Category. The archival record is composed of data in the standard .dbf format, together with the essential documentation package (layouts, field definitions, ShuRM User Guide, the Employment Equity User Manual and the Employment Equity Employee Identification Data Collection Head Office Procedures Manual) needed to read and decode the data and evaluate methodology used to build the data base. The archival data files comprise annual slices of the data taken at the end of the calendar year for the years 1988 to 1993 inclusive and taken at the end of the fiscal year (31 March) for 1994 and subsequent years.
The older the data file, generally, the lower the degree of accuracy attained. As well, all of ShuRM and Employment Equity data files are subject to the Access to Information and Privacy Acts and to the additional requirement that Treasury Board be consulted regarding requests for data files less that five years old. Data must be anonymized before research can be undertaken. Consult the portfolio archivist for access to the Finder Data Base under "EEDB" or "ShuRM" in the title field.(RG55, G0002770-G0002777)
