ARCHIVED - Drop-off of employment equity groups in recruitment

WarningThis page has been archived.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please contact us to request a format other than those available.



A study by the Public Service Commission of Canada

October 2009

Public Service Commission of Canada
300 Laurier Avenue West
Ottawa, Ontario K1A 0M7
Canada

Information: 613-992-9562
Facsimile: 613-992-9352

© Her Majesty the Queen in Right of Canada, represented by the Public Service
Commission of Canada, 2009

Table of Contents

Executive summary

The purpose of the research project was to determine, under the current Public Service Employment Act (PSEA), whether applicants who voluntarily provide their self-declared employment equity (EE) information, are eliminated (‘dropped-off’) at a greater rate than applicants in the reference group and, if so, at what stages of the recruitment process.

This project consisted of two related studies: a survey of applicants and a study of departmental/ agency appointments to assess drop-off in external advertised recruitment processes for applicants from EE groups. Using the Public Service Resourcing System (PSRS), it compared the representation rates of three EE groups: visible minorities, persons with disabilities and Aboriginal persons with that of a reference group consisting of men and women who are not members of these three EE groups, at different stages of the process from application to appointment or to voluntary withdrawal (‘drop-out’) from the process.

Results of both studies showed that for the visible minority group, there was a drop-off
of three to four percentage points from the first (application) stage to the last (appointment) stage. Persons with disabilities and Aboriginal persons did not show a
drop-off.

From the survey results, visible minorities experienced drop-off mainly at two stages of recruitment: following the review of applications by organizations and following testing. Voluntary drop-out or withdrawal by applicants did not account for the observed
drop-off.

Past studies have used different methodologies. From this study, the Public Service Commission (PSC) found that for visible minorities, the drop-off in percentage points based on applications was about 15%; whereas based on unique applicants, it was about 4%. This difference was due to the finding that visible minorities applied to a higher number of job advertisements compared to the average job-seeker. Furthermore, the study found that the EE self-declaration results obtained from the survey and the applicant data were comparable, supporting the reliability and consistency of the EE data in the PSRS.

The overall conclusion that drop-off is experienced only by the visible minority group warrants attention to ensure that there are no selection biases in the appointment process. The PSC plans to more closely examine the appointment rates of EE groups by department/agency and occupational group in both internal and external processes in order to identify any barriers and to develop strategies for improving EE group representation across all levels of the public service.

Purpose

The purpose of the research project was to determine, under the current Public Service Employment Act (PSEA), whether applicants who voluntarily provide their self-declared employment equity (EE) information, are eliminated (‘dropped-off’) at a greater rate than applicants in the reference group and, if so, at what stages of the recruitment process.

Introduction

Several studies conducted by the Public Service Commission (PSC) since 1999 have indicated that in the federal recruitment process, the appointment rates of some EE groups1 were found to be lower than those for non-EE group members when compared to their respective representation rates in the initial application stage. This phenomenon is referred to as ‘drop-off’. Drop-off was traditionally measured between two points in the recruitment process – application and appointment. These studies were based on information extracted from often incomplete department/agency files, and the unit of analysis was applications rather than applicants.

These limitations precluded a generalization of results and inflated the drop-off results. For example, if a person applied to five appointment processes and accepted one job offer, the drop-off by applications would be four out of five. If one counted by applicants, this person would not be counted in the drop-off results. Furthermore, to the extent that members of an EE group submitted more applications per person than the rest of the applicant population, counting by applications would artificially inflate the observed difference in drop-off between the EE group and the rest of the applicant population. This point is important because it has been reported previously that visible minorities tend to apply to more advertised jobs than the rest of the applicant population.

With recent advances in technology and the requirement to submit applications on-line through the Public Service Resourcing System (PSRS), the quality of applicant information has improved significantly. It is now feasible to use individual applicants (as opposed to applications) as the unit of analysis in studies of drop-off during the job application process.

Challenges remain, however. A study of drop-off by applicants from initial application to final appointment requires the identification of individuals from beginning to end of the process, but there is no common identifier between the PSRS application system and the departmental/agency appointment information that is derived from government pay files.

This project consisted of two related studies conducted between January 2007 and April 2008 on assessing drop-off in external advertised recruitment processes for applicants from three EE groups: visible minorities, persons with disabilities and Aboriginal persons. The meaning of drop-off was extended to include changes in representation or elimination rates not only between application and appointment, but between key steps in the recruitment process, thus enabling some degree of tracking. Using the PSRS, it compared the representation rates of the three EE groups with that of the reference group at different stages of the ‘General Recruitment’ process from application to appointment or to withdrawal from the process. ‘General Recruitment’ refers to the process where federal government jobs are advertised to the general public using the PSRS and does not include special recruitment programs such as Post-Secondary Recruitment (PSR) or student jobs.

The present research consists of two studies:

1. Survey study: A retrospective survey was completed by those who applied to General Recruitment through PSRS between April 2007 and March 2008. The respondents were asked about their job applications to the federal government since 2006. The survey captured information about each stage of the process from application to appointment or drop-off. The representation rates of the EE groups were compared based on their responses to the survey. The self-declared EE information was captured in this survey.

2. Study of departmental/agency appointments linked to PSRS data: This method utilizes two points in the recruitment process – application and appointment. PSRS data from job applications to advertisements with closing dates between January and December 2007 were compared to departmental/agency data for appointments made between January 2007 and April 2008. Individual applicants were identified based on a combination of name, Personal Record Identifier (if available) and other relevant biographic or demographic information. The representation rates of applicants from EE groups at application (PSRS data) and appointment (information from organizational pay data) were compared. Differing from the traditional PSC method of generating statistics based on EE information, this study used:

  • Current EE data voluntarily provided at the point of application rather than up to a year or more after appointments;
  • Applicants as the unit of measurement rather than the traditional method of using aggregated recruitment applications; and
  • Only positions advertised on the PSRS Web site, for term or permanent positions, excluding casuals, inventories and special recruitment programs such as PSR and student training opportunities.

Part I: Applicant survey study of drop-off

1. Method

A Web-based survey was mounted on the Public Service Resourcing System (PS
RS) site and made available to General Recruitment applicants (it excluded those applying to Post-Secondary Recruitment, student programs or general inventories) between April 1, 2007 and March 31, 2008.

The respondents were asked to provide a retrospective report on their experiences with the General Recruitment process. The survey questions asked about the nature and number of positions applied for, the stages reached in the recruitment process and outcomes to date. When applicants did not reach the appointment stage, they were asked whether this was due to drop-off (the application and assessment process eliminated them) or to voluntary drop-out on their part (for example, they accepted another job).

The survey also requested information about the respondents’ level of education, language background, language skills, age group, employment equity (EE) membership and willingness to relocate.

The responses were collected and analyzed for two different time periods:

  1. Applications submitted between January 2006 and four months prior to completing the survey. The information obtained for this time period was expected to include a large proportion of completed processes. This provided the main source of data for this study.
  2. Applications submitted during the four months prior to completing the survey. While this information was the most current, it would include a high proportion of incomplete cases. Respondents in this group were asked on a volunteer basis to take the survey again in four months, but the number of volunteers proved too small for further analysis. This provided a source of confirmation for the main (first period) data.

The survey results were grouped according to three EE group memberships: visible minorities, persons with disabilities, and Aboriginal persons. After determining that women not belonging to the above three EE groups did not show drop-off or drop-out when compared to the three EE groups, they were included in the reference group only. Thus, the reference group was comprised of men and women who did not belong to the above three groups.

The following analyses were carried out:

  • Descriptive statistics based on biographical information;
  • Representation rates of the four groups (reference group, visible minorities, persons with disabilities and Aboriginal persons) at different stages of the General Recruitment process; and
  • Extent of drop-off experienced by each group across stages of the process.

2. Survey results: Age distribution, university education and language skills of the respondents

The survey was on-line for a 12-month period and 42 110 of about 200 000 applicants or 21% provided responses.

Representation of survey respondents by group

Table 1 provides the representation of survey respondents by group. Two hundred and thirty-one (231) respondents belonged to more than one EE group. When compared to the PSRS EE data, Table 1 shows that the survey sample from Period 1 is a reasonable representation of the applicant population for that period.

Table 1: Employment equity group membership
  Sample/
Applicants
Reference
group**
Visible
minorities
Persons with
disabilities
Aboriginal
peoples
Period 1: Survey respondents
– January 2006 to four months
prior to survey*
30 165 21 567
71.5%
6 710
22.2%
1 079
3.6%
1 040
3.4%
PSRS applicants to
General Recruitment in 2007
225 258 162 895
72.3%
48 402
21.5%
7 157
3.2%
8 581
3.8%

* Some respondents self-declared to more than one EE group.
** The reference group is composed of men and women who are not members of the three EE groups.

Choice of occupation and departments/agencies

The top three occupational groups to which EE group members applied were: Clerical and Regulatory, Program Administration and Administrative Services. These are the occupational groups most commonly advertised on the PSRS site.

Respondents applied most frequently to the following four departments: Human Resources and Skills Development Canada, Health Canada, National Defence and Transport Canada. Veterans Affairs Canada was the fourth choice of persons with disabilities, while Citizenship and Immigration Canada was the fourth choice of Aboriginal persons. These departments accounted for approximately a third of all applications by the survey respondents.

Home province

The top home provinces listed by respondents were Ontario, Quebec and British Columbia. Aboriginal persons listed Ontario, Quebec and Manitoba as the top three.

Age distribution

Table 2 provides the age distribution of the four groups under study. With the exception of persons with disabilities, more than 50% of applicants in each group are between the ages of 21 and 35. It is this group that the federal government and other organizations are trying to recruit in order to address the aging workforce.

Table 2: Age distribution by group
Age range in years Reference group Visible minorities Persons with
disabilities
Aboriginal
peoples
Less than 21 0.8% 0.8% 0.2% 1.1%
21-35 57.4% 53.6% 36.9% 50.7%
36-45 25.5% 31.3% 31.9% 30.7%
46-55 14.3% 12.4% 25.3% 15.5%
More than 55 1.9% 1.9% 5.6% 2.0%

University degree

As shown on Table 3, the percent of visible minority applicants with a university degree was higher than that of the other groups studied. A similar finding applies to the total PSRS data for calendar year 2007.

Table 3: Percentage of group members with university degrees
Sample Reference group Visible minorities Persons with
disabilities
Aboriginal
peoples
Survey Respondents 50.3% 72.6% 48.5% 39.7%
PSRS* 48.8% 69.9% 51.5% 38.4%

* PSRS data for calendar year 2007.

Mobility of respondents

As willingness to relocate plays an important role in finding a job, respondents were asked if they were willing to relocate. Visible minority group members indicated the strongest interest in relocating to obtain a job, as shown in Table 4.

Table 4: Willingness of groups surveyed to relocate
Question: Are you willing to relocate?
Response Reference group Visible minorities Persons with
disabilities
Aboriginal
peoples
Yes 32.8% 53.1% 36.2% 34.4%
No 21.0% 9.9% 18.5% 18.7%

* Remainder of respondents answered, "it depends."

Linguistic characteristics of respondents

First official language chosen by respondents when applying for a job

The survey allowed applicants to choose among the following as the official language they generally chose when applying for a job: First official language is French, English or both. The PSRS system does not ask applicants to report on both, so exact comparisons with PSRS data were not possible. (See Table 5.) Visible minority responses resembled the reference group, while fewer persons with disabilities and Aboriginal persons reported French as their first official language.

Table 5: First official language reported by survey respondents
Language Reference group Visible minorities Persons with
disabilities
Aboriginal
peoples
French 24.8% 22.8% 13.1% 11.3%
English 68.1% 70.6% 82.3% 83.6%
Both 7.8% 6.6% 4.5% 5.3%

Language of upbringing and fluency

Respondents were asked whether their language of upbringing was English, French, both, or other. It is noteworthy that 40.9% of visible minorities reported a language other than English or French as their language of upbringing. (See Table 6.)

Table 6: Language of upbringing reported by the groups under study
Language Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
French 27.4% 16.2% 13.9% 12.8%
English 54.7% 39.2% 69.2% 74.0%
Both 5.9% 3.7% 6.5% 6.6%
Other Language 12.0% 40.9% 10.4% 6.6%

Fluency in the two official languages

Tables 7 and 8 provide the degree of official languages fluency reported by the survey respondents. Since these are self-reports without reference to language standards or language tests, they should be interpreted with caution.

Most respondents self-reported high fluency in English. Fewer than 50% reported high fluency in French. The least fluency in French is reported by persons with disabilities and Aboriginal persons. One may conclude, with caution, that these two groups would find it more challenging than the other groups to obtain bilingual imperative positions. Visible minorities reported the least fluency in writing in English, but this result is not readily interpreted without additional data.

Table 7: Proportion of respondents reporting themselves as very fluent in French
Skill Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
Oral 39.8% 36.1% 23.0% 22.1%
Reading 42.8% 39.9% 26.6% 23.3%
Writing 33.6% 18.6% 18.6% 17.5%

Table 8: Proportion of respondents self-reporting as very fluent in English
Skill Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
Oral 81.9% 75.3% 89.1% 92.4%
Reading 88.2% 82.7% 91.5% 93.4%
Writing 80.9% 75.0% 88.4% 91.0%

Language requirements as potential barrier to job applications

The survey respondents were asked if language requirements were a barrier to them in the recruitment process. The responses of all groups were comparable, with the exception that 37.9% of persons with disabilities reported language requirements as a barrier that they had experienced often or always in their job applications.

Table 9: Overall language requirements as possible barriers
Question: Have language requirements posed a barrier in applying for a bilingual position? Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
Often or always 23.5% 26.5% 37.9% 26.1%

3. Survey results: Number of advertisements (applications) per applicant

Table 10 provides the number and percentage of survey respondents in the groups under study and, for each group, the average number of advertisements to which the group members had applied during the time period covered by the survey. It is noteworthy that visible minorities and persons with disabilities reported applying to 74% and 64% more advertisements per person respectively than the reference group. This trend has been observed in previous studies for visible minorities and persons with disabilities.

Table 10: Distribution of respondents by employment equity group and average number of advertisements (applications) per applicant in each group
Reported by survey Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
Number and % of applicants 21 567
71.5%
6 710
22.2%
1 079
3.6%
1 040
3.4%
Average number of advertisements per applicant 8.1 14.1 13.3 7.9

4. Survey results: Voluntary drop-out from the recruitment process

The basic steps in the General Recruitment process are:

  1. application;
  2. retention by the department or agency for further processing;
  3. testing;
  4. interview;
  5. job offer; and
  6. appointment.

During each of these steps, applicants can be eliminated (dropped-off) or they may decide to withdraw on their own (drop-out) from the process.

Before proceeding to an analysis of drop-off, analyses of voluntary drop-outs were conducted. It was found that, although a certain number of applicants drop out at each stage (when invited for testing or interview and when offered a job), the drop-out rates are comparable for the reference group and the three EE groups under study. The following were the most frequently cited reasons for dropping out of the process:

  • At the testing stage, the main reasons for not taking the test were: scheduling (could not get off work, too busy, etc.), location (testing site too far, etc.), or no longer interested in the job;
  • At the interview stage, the main reasons for refusing the interview were: no longer interested in the job, found another job, and/or location of the interview or the job; and
  • Following a job offer, the main reasons for turning down the job were: money, length of the term offered, no longer interested in the job, or found another job.

5. Survey results: Drop-off by employment equity groups in comparison to reference group at stages of recruitment from initial application to job offer

The key purpose of this project was to determine whether members of EE groups, when they apply to General Recruitment, are eliminated from the process in higher proportions than members of a reference group consisting of all other applicants, and if so, at what stages or steps in the process. The results are presented in the following table. Drop-off at each step was determined by subtracting the percentage representation at the preceding step from that at a given step. (See Table 11.)

As shown in Table 11, a drop-off of 3.1% from application to acceptance of a job was observed for visible minorities. Some of this drop-off took place when the organization determined (Step 2) which applicants to retain for further processing. Further drop-off took place between testing (Step 3) and interview (Step 4).

The results presented in Table 12, which apply to advertisements posted within the four months prior to the survey and are therefore less complete than those in Table 11, provide confirmation of the extent of drop-off observed for visible minorities and the fact that it is observed primarily at Step 2 and between Steps 3 and 4 of the process.

Drop-off was not observed for the other groups. The drop-off observed for visible minorities in the recruitment process may be reduced by ensuring that there are no biases introduced in the process. Departments and agencies should carefully review the initial applications with attention to the job requirements to ensure that none of the applicants has been eliminated at this stage on irrelevant grounds. Tests and assessment tools used in the process should demonstrably assess the qualifications posted in a valid and reliable manner.

Table 11: Drop-off at each stage of General Recruitment process from application to job offer
Results for advertisements posted in 2006 and 2007, and up to 4 months prior to completing the Survey
Steps in recruitment process Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
1. Active applicants
(30 165)
71.5% 22.2% 3.6% 3.4%
2. Screened in by departments
(11 433)
72.3% 21.1% 3.8% 3.5%
3. Testing
(9 854)
71.9% 21.9% 3.8% 3.4%
4. Interviews
(5 148)
73.7% 19.0% 4.0% 4.0%
5. Job offered
(2 544)
74.2% 18.4% 4.2% 3.9%
6. Accepting job offer
(2 020)
73.5% 19.1% 4.2% 4.1%
Drop-off by step
After initial screening
(Step 2 minus 1)
+0.8% -1.1% +0.2% +0.1%
After further screening
(Step 3 minus 2)
-0.4% +0.8% 0.0% -0.1%
After testing
(Step 4 minus 3)
+1.8% -2.9% +0.2% +0.6%
After interview
(Step 5 minus 4)
+0.5% -0.6% +0.2% -0.1%
From job offer to appointment
(Step 6 minus 5)
-0.7% +0.7% 0.0% +0.2%
Overall drop-off from application to appointment
(Step 6 minus 1)
+2.0% -3.1% +0.6% +0.7%

Table 12: Exploring Drop-off at each stage of the General Recruitment process from application to job offer based on recent job-seeking activity (advertisements posted within the 4 months immediately preceding the survey)
Responses for applications during the 4 months immediately prior to completing Survey
Steps in recruitment process Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
1. Active applicants
(23 778)
67.5% 25.6% 4.2% 3.7%
2. Screened in by departments
(6 792)
68.5% 24.4% 4.3% 3.9%
3. Testing
(6 305)
68.4% 24.6% 4.3% 3.6%
4. Interviews
(2 707)
69.2% 22.5% 4.8% 4.5%
5. Job offered
(616)
As many of the advertisements applied to in the 'past four months' may not have reached the appointment phase, these data were not reported.
Drop-off by step
After initial screening
(Step 2 minus 1)
+1.0% -1.2% -0.1% +0.2%
After further screening
(Step 3 minus 2)
-0.1% +0.2% 0.0% -0.3%
After testing
(Step 4 minus 3)
+0.8% -2.1% +0.5% +0.9%

As the survey asks respondents to recall their job-seeking activity over a long period of time, these data reported here are intended as a check using more recent recall information. While appointments may be incomplete within a four month period and are likely evenly distributed, the shorter term information supports the longer term trends observed in this survey study.

6. Conclusions from applicant survey study

Using a Web-based survey, job applicants were invited to provide information on their General Recruitment experience. Their responses were then analyzed. Drop-off was computed and distinguished from voluntary withdrawal (drop-out).

Overall, the results of this study revealed that there was drop-off for visible minority group members from the short-listing to the job-offering phases and that this was not the case for the remaining groups being compared. The magnitude of drop-off was small, but persisted in a downward direction from the applicant to job offer stages. On the other hand, the representation rates through the same stages of assessment increased for the remaining groups, even though their job-seeking activities were not as high.

From further examination of the survey responses, there was no evidence to support the differential voluntary withdrawal (drop-out) for visible minority group members. In fact, once offered a job, they tended to accept the offer more than others. While drop-out rates could not account for drop-off, the reasons that the respondents dropped out were analyzed for additional qualitative information. The main reasons for not taking a test were the same across groups – scheduling problems and location of the test. Interviews were turned down for the same reasons across groups – no longer interested in the job or found another job, and location of the job or scheduling issues. It was expected that because visible minority group members and persons with disabilities applied to significantly more positions, scheduling issues would be bigger reasons than for the other groups; however, this was not supported by the data. While location issues surfaced, which included where the test, job, or interview was located, when respondents were asked about their willingness to relocate for a job opportunity, visible minority group members provided the highest interest in relocating among all groups.

As level of education is commonly one of the merit criteria in job requirements, it was expected that, as visible minority group members have been observed in this and other studies to have higher levels of education, their job opportunities would be greater. In addition, to the extent that higher education improves a person’s testing skills, knowledge and work experience, a reasonable inference might be that visible minority group members would have an advantage as job-seekers. These expectations were not supported by the survey data.

PART II: Statistical study of drop-off by departmental/agency appointments

1. Method

Organizational appointment data compiled by the Public Service Commission are derived from pay records. Since the objective of applying to advertised jobs through the Public Service Resourcing System (PSRS) is to be appointed to a job in a federal department or agency, this study asked the following question: When the proportions of employment equity (EE) group members appointed by departments (i.e. job offer made and accepted) are compared to proportions of EE group applicants, is there evidence of drop-off for members of the EE groups in comparison to the reference group?

The current study analyzes job opportunities advertised on the PSRS Website, whose closing dates were from January 1 to December 9, 2007 (not quite a calendar year). For this period, there were approximately 16 783 term appointments of three months and over, and indeterminate appointments made from the General Recruitment between January 2007 and March 17, 2008, with the understanding that the advertisements closed at the end of 2007 would have appointments extending into 2008. Of these, 8 354 appointees were identified by linking departmental/agency appointment data (derived from pay records) to PSRS applicants. The PSRS and departmental/agency data were linked through combinations of the following indicators: employee Personal Record Identifiers (when available), partial dates of birth, last name, first name, and, sometimes, algorithms based on staffing process identifiers.

There was no EE data in the departmental/agency appointment files. Therefore, the
self-declaration of EE status provided when applying to General Recruitment through PSRS was used in this study. To verify the reliability of this information, the EE
self-reports provided by persons in the survey were compared with the self-declaration reports on their initial application through PSRS. The two sources of information (survey and PSRS) yielded the following percentages of agreement in EE self-declarations: 93.6% for visible minorities, 84.5% for persons with disabilities, and 72.5% for Aboriginal persons. While the degree of correspondence for two of the three groups was acceptable, the same degree was not observed for Aboriginal peoples.2

2. Results of statistical study

Table 13 shows the representation rates of EE and reference group members at two steps of the recruitment process: applicants to General Recruitment and acceptance of the job. Drop-off was observed only for the visible minority group: its representation rate dropped from 21.5% to 17.8% of those accepting a job offer, that is, -3.7%. This
drop-off is consistent with that observed in the survey study (Part I of this report).

Table 13: Proportional representation of employment equity and reference group members at the initial application and acceptance of job stages of General Recruitment, 2007-2008
Representation of EE and reference group members at the first and last stages of recruitment based on linked organizational and PSRS data
  Reference group Visible minorities Persons with
disabilities
Aboriginal peoples
PSRS applicants to General Recruitment in 2007 (225 258) 162 895
72.3%
48 402
21.5%
7 157
3.2%
8 581
3.8%
Persons appointed to jobs between January 1, 2007 and March 17, 2008 through General Recruitment 6 267
75%
1 483
17.8%
271
3.2%
386
4.6%
Percentage point change in representation:
drop-off by applicants
+2.7% -3.7% 0% +0.8%

3. Conclusions from statistical study

This study, based on the information contained in organizational files, yields essentially the same conclusion as the drop-off study, based on a survey of candidate responses. The only group that experiences drop-off across stages of the General Recruitment process is the visible minority group. In this study, the drop-off rate was 3.7 percentage points, that is, from 21.5% at initial PSRS General Recruitment application to 17.8% upon acceptance of the job. The extent of this drop-off is contrasted with a comparable increase in the proportional representation of the reference group.

For reporting purposes, a drop-off of 3.7 percentage points from the study on organizational appointments is used as it is a more direct and objective measure than the statistics based on retrospective recall of job-seeking experiences.

Part III: Overall conclusions and implications for future research

Two different methods for assessing drop-off yielded similar results regarding drop-off for visible minorities in the General Recruitment process. The drop-off rates in both studies were consistent. The survey depended upon self-reported estimates of job-seeking activity. The study of departmental/ agency appointments depended upon links between initial application through the Public Service Resourcing System (PSRS) and acceptance of the job as indicated in the organization’s files.

The current research provided unique information that was previously unavailable. A large sample was used that reached across many government departments and agencies, and ranged from the application to the appointment stage in the recruitment process. The survey results helped to clarify and eliminate some hypothesized reasons for drop-off. Drop-off is not due to differential drop-out rates. All three employment equity (EE) groups and the reference group had similar drop-out rates and for similar reasons. Language obstacles, while a concern for all groups, were comparatively low for visible minority group members.

Self-declaration as members of EE groups in this study was current as it was derived from the same source that was being studied and was matched to the periods under study. The EE information reported in the survey and derived from the PSRS application was found to be consistent.

Drop-off was observed for visible minorities mainly at two stages in the recruitment process: when organizations determined who should be retained for further processing, based on their review of applications and when determining which applicants were to be interviewed after the testing stage. To ensure that no biases are introduced at these stages, all departments and agencies should carefully review the initial applications with attention to the job requirements to ensure no-one has been eliminated on irrelevant grounds.

Comparison with previous study

The 2006 Public Service Commission (PSC) study entitled Drop-off Rates for Employment Equity Groups – Automated Screening Reports & Appointments 2000-2005 reported that visible minorities averaged 25.7% of applications but only 10.5% of appointments, i.e. a drop-off of 15.2 percentage points. This study used applications as the unit of measurement as opposed to applicants.

While the current study consistently used applicants as the unit of measurement for drop-off, it was observed that if applications were used instead, the drop-off was about 15 percentage points also, and much higher than the 3.7 percentage points observed based on applicants. This difference was due to the finding that visible minorities applied to a higher number of job advertisements compared to the average job-seeker and therefore, using applications instead of applicants would inflate the drop-off.

Implications for future research

The PSC plans to more closely examine the appointment rates of EE groups by department/agency and occupational group in both internal and external processes. This will help to identify any barriers and to develop strategies for improving EE group representation across all levels of the public service.

Study Team

This study was conducted by the Equity and Diversity Directorate, Policy Branch and the Research and Development Division, Personnel Psychology Centre, Staffing and Assessment Services Branch of the Public Service Commission.

Equity and Diversity Directorate:
Paula Green, Director General
Jonas Ma, Senior Advisor

Personnel Psychology Centre:
Henry Edwards, Director, Research and Development Division
Penny Faulkner, Principal Investigator

Foot Notes

  1. Under the Employment Equity Act, the four designated employment equity groups are Aboriginal peoples, persons with disabilities, visible minorities and women. [Return]
  2. The degree of correspondence was determined as follows: if a person self-declared in the survey but did not when applying to PSRS, the correspondence was zero. Similarly, if a person self-declared in both situations, the correspondence was one. These results were summed up and converted to percentages for each EE group. [Return]