3. Data preparation and limitations

View the most recent version.

3.1 The Registered Apprenticeship Information System

The Registered Apprenticeship Information System (RAIS) survey obtains information on the number of apprentices registered in each province and territory across Canada. The information is requested from provincial and territorial ministries responsible for apprenticeship training and extracted from administrative files.

The purpose of the survey is to gather information on individuals who receive training and those who obtain certification within a trade where apprenticeship training is being offered. The survey compiles data on the number of registered apprentices taking in-class and on-the-job training in trades that have been designated as either Red Seal trades or non-Red Seal trades. It also compiles data on the number of certificates granted to apprentices and trade qualifiers in Red Seal and non-Red Seal trades.

The survey gathers data annually on the number of individuals which are registered and continuing their training, newly registered, discontinued, successfully completed and received certificates in qualified apprenticeship programs over a calendar year. Certification information is obtained on registered apprentices who have successfully completed, as well as trade qualifiers, receiving provincial and Red Seal qualifications in a designated trade. Since 1991, the survey has been requesting individual micro level information on each apprentice concerning, not only the above mentioned information, but also their gender, age, sponsorship, date of registration, time of completion, in-class and on-the-job credits, reason for leaving the program and prior trade certification.

3.2 Methodology for creating a longitudinal apprenticeship-training database

The first stage of the analysis consisted of preparation of the longitudinal database. Data from the annual Registered Apprenticeship Information System (RAIS) survey were linked for the period 1993 to 2003 using the unique provincial identifiers attached to the individual Registered Apprenticeship Information System (RAIS) records. ² Only those variables that were considered to be reliable across provinces were kept in the database; these were the same variables that were included in the 1992 database. These variables provided information on the province/territory, age, sex and trade of apprentices and their learning status over the year as well as information on their certification. Once the database was created, some adjustments to the data were made to improve the coherence of the longitudinal information. This was necessary because the Registered Apprenticeship Information System (RAIS) annual survey was not designed originally for longitudinal analysis and, therefore, the data had not been verified for consistency on an individual basis from year to year.

Most of the specifications that were used to improve the coherence of the 1993 data were the same as those that had been applied to the 1992 data. As a result, comparisons between the two cohorts were not affected. In general, problems in the data that were identified were similar to those in the 1992 data:

• wrong trade changes (usually these were frequent trade changes over the years for the same apprentice that were likely due to changes in a jurisdiction's trade code structure);
• not real starters (these were apprentices who were recorded as having completed in the starting year (1993) a program for which the duration was nominally two years or more); ³
• apprentices recorded as having completed in the same trade twice;
• apprentices recorded as having discontinued in the same trade in two consecutive years;
• apprentices who were recorded as having skipped years with no indication of discontinuation or new registration or reinstatement;
• continuers with no records in the following years;
• completers in a trade who were recorded as having continued or discontinued in the same trade in the following years; and
• individuals who were recorded as being continuers in a new trade with no new registration.

Differences may also exist across provinces, since each province has its own set of apprenticeship programs and its own apprenticeship information system. ⁴ First, a year of apprenticeship may be defined differently across provinces. Second, provinces do not necessarily proceed the same way in the 'discovering' of cases of discontinuation and, as a consequence, there may be differences across provinces in the estimation of the timing of discontinuation. Third, provinces may or may not track the number of hours of on-the-job training and therefore, may not have an accurate indicator of the activity status of the apprentices. Finally, we note that individual identifiers are not transferable from one province to another; as a result, it is not possible to track an apprentice's status across provinces.