Appendix 1: APS Workforce data
APS Employment Database
The Australian Public Service Employment Database (APSED) contains employment, diversity and education details for all people employed in the APS under the authority of the Public Service Act 1999 (Cth).
Information on staffing, including trends in the size, structure and composition of the APS, contributes to research and evaluation work on the changing nature of the APS. This, in turn, assists agencies to formulate their people management policies and practices.
APSED is the definitive source of APS employment data, supporting strong evidence-based APS workforce policy, people management and advice.
APSED Scope and collection methodology
APSED stores the employment data of all current and former APS employees. The database was established in 1999 but contains data on APS employees from 1966. The most recent snapshot, conducted on 30 June 2021, contains records relating to 153,945 employees.
APSED is maintained by the APSC and the data is supplied to APSED from the HR systems of APS agencies.
Two types of data files are used to update and maintain APSED: movement files and snapshot files. In general, both file types contain the same data items, but they differ in purpose.
- Movement files are provided to the APSC from each agency every month. They are used to document changes in employment history (for example, engagements, promotions and maternity leave) for all people employed under the Public Service Act 1999 on a monthly basis.
- Changes in employment characteristics every month are recorded using movement codes. Movement files contain a record for every movement relevant to updating and maintaining employee records in APSED that has been processed in an agency’s HR system during the month. Therefore, if an employee undertakes multiple movements within a reference period, the corresponding movement files will contain multiple records for that employee. Conversely, if an employee has no movements during the reference period, they will not appear in the movement file.
- Snapshot files are provided to the APSC from each agency on a six-monthly basis. They are used to verify that all the information stored in APSED is correct and current at 31 December and 30 June each year. Snapshot files contain a single record for every APS employee employed by a particular agency on 31 December and 30 June.
Agency HR systems supply APSED with unit records containing this personal information:
- Personal particulars: Australian Government Staff number, name, and date of birth.
- Diversity data: gender, Indigenous identification, country of birth, year of arrival, first and main languages spoken, parents’ first languages, disability status.
- Employment data: classification, email address, date of engagement, employment status, standard hours, workplace postcode, any movement details and dates, operative status, previous employment, job family, job function, job role, current and previous agencies.
- Educational qualifications and main fields of study.
Under Section 50 of the Australian Public Service Commissioner’s Directions 2016, an Agency Head must ensure measures are in place to collect information from each employee in the agency and give collected information to the Australian Public Service Commissioner. While individuals do not explicitly consent to the collection of their movement and employment data, they can choose to supply or withhold all diversity data except gender data. An Agency Head must allow APS employees to provide a response of ‘choose not to give this information’.
Management and administration
Agency HR systems collect relevant data items and supply these to the APSC through secure or encrypted means. Agencies are responsible for the collection, security, quality, storage, access, use, and disclosure of their HR data as well as compliance with the Australian Privacy Principles. While agency HR systems capture detailed information on each APS employee’s pay, leave history and entitlements, these are out of scope for APSED. Only data fields supplied to the APSC are in scope.
Upon receipt, each data file is corrected in an iterative process. Once validated and transferred to the APSC, error checks on the new files are performed by the APSC against the extant data in APSED. The APSC and the agency work together to resolve these differences. Once resolved, cleaned data is incorporated into APSED.
APSED data is stored on a secure information technology system that is password protected and accessible only by a team in the APSC who have been granted access by team supervisors and trained in protecting and using these collections. Standard operating procedures dictate when personal information can be added or changed. All changes to the database are logged in an audit file.
Privacy and confidentiality
Data protections within APSED include secure transfer of information between agencies and the APSC, storage of data on APSC servers requiring individual logins to access, restriction of access to a small number of authorised users, and ensuring public release of data is undertaken in aggregate format only.
APS Employee Census
The APS Employee Census is an annual employee perception survey of the APS workforce. All eligible personnel employed under the Public Service Act 1999 are invited to participate. The Employee Census has been conducted since 2012 and collects APS employee opinions and perspectives on important issues, including employee engagement, wellbeing, leadership, and general impressions of the APS.
APS Employee Census collection methodology
In light of the COVID-19 pandemic and its impacts on the workforce, the 2020 APS Employee Census was postponed from its usual May to June delivery timeframe and deployed from 12 October to 13 November 2020. This change in timeframes meant that 2020 APS Employee Census results were not available for reporting in time for publication within the 2019–20 State of the Service Report, but are included within this report.
In 2021, the APS Employee Census returned to its usual timeframe and was administered to eligible APS employees during the period 10 May to 11 June 2021.
Although participation is encouraged, the APS Employee Census is voluntary. If a respondent chooses to participate, only a limited number of demographic-type questions must be answered. The remaining questions do not require a response.
APS Employee Census design
Questions from previous years were used as the basis for the 2020 and 2021 APS Employee Census questionnaires with the aim of maintaining a reliable longitudinal dataset. Questionnaire content was evaluated, reviewed and updated; new questions were included to address topical issues or to improve the quality of the data collected. In particular, an underlying framework was applied to the 2021 APS Employee Census questionnaire.
This framework applied a model from the field of organisational psychology and served to better define a core question set and how results could be interpreted and reported. While a standardised questionnaire was employed, at each administration of the APS Employee Census, agencies could ask their employees a limited number of additional questions.
APS Employment census development
The 2020 and 2021 APS Employee Census questionnaires both included 137 individual questions. These questions were grouped into sections addressing key aspects of working for an APS agency.
Each year the content of the APS Employee Census is reviewed to ensure each question has value and meets a specific purpose. The APSC researches and consults broadly to develop and select questions for inclusion. In 2020 and 2021, the APSC:
- considered strategic-level priorities coming from the Secretaries Board and other senior-level committees to ensure the Employee Census would capture appropriate information to inform these priorities
- consulted with subject matter experts from within the APSC and other APS agencies to seek their input to question design and information requirements for supporting APS-level policies and programs
- researched contemporary understanding of issues and options for questionnaire content
- provided participating agencies with an opportunity to give feedback and input to questionnaire design.
The resulting questionnaires covered numerous themes and measures. Central to these are three indices addressing employee engagement, innovation and wellbeing.
APS Employee Census delivery
In 2020 and 2021 the APS Employee Census was administered using these collection methods:
- online, through a unique link provided to each employee by email from Engine, the contracted Employee Census administrators
- telephone surveys with employees who did not have available supportive information technology to provide reasonable adjustment for their disability
- paper-based surveys for employees who did not have access to an individual email account or did not have suitable access to the Internet.
Sampling and coverage
The population for the APS Employee Census is provided by participating agencies. It is up to individual agencies to set their eligibility criteria. For example, some agencies removed employees on long-term leave, while others contact employees on long-term leave to check if they would like to participate.
2020 APS Employee Census
The 2020 APS Employee Census covered eligible ongoing and non-ongoing employees from 100 APS agencies. One APS agency elected not to participate.
Invitations to participate in the 2020 APS Employee Census were sent to employees from 12 October 2020. The number of invitations was adjusted as new employees were added, separations processed, and incorrect email addresses corrected. The deadline for survey completion was 13 November 2020.
The final sample size for the 2020 APS Employee Census was 137,897. Overall, 108,085 APS employees responded, giving a response rate of 78%.
2021 APS Employee Census
The 2021 APS Employee Census covered eligible ongoing and non-ongoing employees from 102 APS agencies. Four APS agencies elected not to participate.
Invitations to participate in the 2021 APS Employee Census were sent to employees from 10 May 2021. The number of invitations was adjusted as new employees were added, separations processed, and incorrect email addresses corrected. The deadline for survey completion was 11 June 2021.
The final sample size for the 2021 APS Employee Census was 142,493. Overall, 109,537 APS employees responded, giving a response rate of 77%.
Management and administration
The APS Employee Census is managed and coordinated by the APSC’s Strategic Policy and Research Group. The APSC contracts an external service provider to support survey administration and reporting activities. Engine was this service provider in both 2020 and 2021.
The public release of agency Employee Census data is aggregated and de-identified to protect individuals’ privacy. The release complies with the Privacy Act 1988 (Cth), and the principles and codes under the Act.
Individual agencies may elect to publically release their agency-level Employee Census results on their respective websites after this report is tabled. These results can also be centrally accessed on the APSC website.
The APSC employs specific rules around how results are reported. Such rules protect the privacy of respondents and individual agency results.
De-identification of data is undertaken in accordance with guidance provided by the Office of the Australian Information Commissioner.
Employee engagement idex
The APS Employee Census uses a model of employee engagement developed by Engine.
This model addresses three attributes associated with employee engagement and measures the emotional connection and commitment employees have to working for their organisation. In this model, an engaged employee will:
- Say—the employee is a positive advocate of the organisation
- Stay—the employee is committed to the organisation and wants to stay as an employee
- Strive—the employee is willing to put in discretionary effort to excel in their job and help their organisation succeed.
First introduced in 2017, the Say, Stay, Strive employee engagement model is flexible and the APSC has tailored the questions for the APS context.
The 2021 APS Employee Census addressed innovation through a set of dedicated questions that contribute to an index score. This Innovation Index score assesses both whether employees feel willing and able to be innovative, and whether their agency has a culture that enables them to be so.
The Wellbeing Index in the APS Employee Census measures both the practical and cultural elements that allow for a sustainable and healthy working environment. The APS has long been focused on the wellbeing of its employees. As employers, APS agencies have obligations under work health and safety legislation. Together with these obligations is an acknowledgement that high performance of employees and organisations cannot be sustained without adequate levels of employee wellbeing.
The results for the wellbeing index are presented in Appendix 4.
Calculating and interpreting index scores
The questions comprising the Employee Engagement, Innovation and Wellbeing indices were asked on a five-point agreement scale. To calculate the index score, each respondent’s answers to the set of questions were recoded to fall on a scale between 0 and 100%. The recoded responses were then averaged across the five or more index questions to provide the index score for that respondent. An individual only receives an index score if they have responded to all questions comprising that index.
|Example question 1
|Example question 2
|Example question 3
|Example question 4
|Example question 5
|Sum of question weights for this employee
|Index score for this example respondent (325/5)
Index scores for groups of respondents are calculated by averaging the respondent scores that comprise that group.
An index score on its own can provide information about the group to which it relates. Index scores, however, have the most use when compared with scores over time or between work units, organisations and demographic groups.
Treatment of responses of 'don't know' and 'not applicable'
Specific questions included within the 2020 and 2021 APS Employee Censuses enabled respondents to provide responses of ‘don’t know’, ‘not applicable’ or similar. Responses of this nature were typically excluded from the calculation of results for inclusion within this report. This was so that results reflected respondents who expressed an informed opinion to the relevant question.
Depending upon the intent, other products generated from the 2020 and 2021 APS Employee Censuses may not apply these same rules. The method in analysis and reporting will be made clear within these products.
APS Agency Survey
The APS Agency Survey is conducted annually and collects information on a range of workforce initiatives, strategies and compliance matters, including the number and type of APS Code of Conduct breaches. APS agencies with at least 20 APS employees complete the survey. The information collected through the APS Agency Survey is used to inform workforce strategies and for other research and evaluation purposes.
Since 2002, the APS Agency Survey has been administered to APS agencies with employees employed under the Public Service Act 1999. The survey assists the Australian Public Service Commissioner to fulfil a range of duties as specified in the Act. These include, but are not limited to:
- informing the annual State of the Service Report
- strengthening the professionalism of the APS and facilitating continuous improvement in its workforce management
- monitoring, reviewing and reporting on APS strategies, initiatives, and policies.
APS Agency Survey collection methodology
In 2021, the APS Agency Survey was administered to 95 agencies during 21 June to 30 July 2021. The response rate for 2021 was 100%, which is typical for this survey.
Each year the APS Agency Survey is sent to the contact officers nominated for each agency. These contact officers are responsible for coordinating the input from relevant areas and uploading responses to an APS Agency Survey portal managed by ORIMA Research.
The survey requires each Agency Head to verify the agency’s submission for completeness and accuracy of responses.
APS Agency Survey management and administration
The APSC’s Strategic Policy and Research Group manages and coordinates the APS Agency Survey. The APSC contracts an external service provider to support survey administration. ORIMA Research was this service provider in 2021.
All APS Agency Survey data are stored in a secure password-protected environment. Where results are included in reporting, agency results are de-identified or aggregated.
Collection of diversity data
Concise demographic information enables the APS to produce accurate workforce descriptions to support the management of its workforce. The APSC collects personal data including diversity information concerning all employees employed under the Public Service Act 1999. On engagement to the APS, employees provide personal information to their employing agencies. This information is provided to the APSC to enable the execution of the APS Commissioner’s functions.
With the exception of gender, the provision of diversity data is voluntary for APS employees. As a result, diversity rates represent the proportion of employees who identify as belonging to that diversity group, and actual diversity rates may be underestimated. The APS Employee Census, in which responses are de-identified, records higher rates for some diversity groups.
This difference is historically largest for the proportion of employees with ongoing disability. The Employee Census likely attracts higher disclosure rates as responses are not readily identifiable to individual employees, unlike the more formal human resources system. Many people are likely to regard their diversity information as sensitive information and it has long been established that people avoid reporting details they regard as sensitive.123 Employees may decide to withhold information when they believe it does not affect their ability to carry out their role. Some may have concerns about possible impacts to their future employment or how it will affect them socially within the workplace. The APS takes pride in its ability to foster inclusive workplaces but also respects employees’ rights to privacy regarding personal information.
Collecting gender information
Individuals may identify and be recognised within the community as a gender other than the sex they were assigned at birth or during infancy, or as a gender which is not exclusively male or female. Therefore, where gender information is collected and recorded in a personal record, individuals should be given the option to select M (Male), F (Female) or X (Indeterminate / Intersex / Unspecified).
Since 2014, respondents to the APS Employee Census could identify their gender as X (Indeterminate/Intersex/Unspecified). This question was updated in 2021 to align with the updated standard released by the Australian Bureau of Statistics124 and better provide all employees the opportunity to more accurately describe their situation. For the first time, APS Employee Census respondents could describe their gender as man or male, woman or female, non-binary or that they used a different term.
123Tourangeau R, Yan T. (2007). Sensitive questions in surveys. Psychological bulletin. 133(5):859.
124Australian Bureau of Statistics. (2021). Standard for Sex, Gender, Variations of Sex Characteristics and Sexual Orientation Variables 14 January.