Conducting workplace surveys

Queensland government agencies1 often conduct workplace surveys of some or all of their staff. They generally relate to topics such as staff satisfaction, workplace concerns, perceptions of their work, feedback on supervisors and/or other agency issues.2

When agencies deal with personal information, they must comply with the privacy principles3 in the Information Privacy Act 2009 (Qld) (IP Act). Information is personal information if it can reasonably identify an individual.

This guideline outlines issues agencies will need to consider when conducting workplace surveys. Agencies should also refer to Surveys and the privacy principles for more information on designing privacy compliant surveys.

Outsourcing the survey

If agencies outsource the workplace survey to a contractor, the agency must take reasonable steps to bind the contractor to comply with the privacy principles.4 A failure to do so will make the agency liable for any privacy breach by the contractor.

For more information refer to Contracts and other agreements.

Anonymous surveys

Workplace surveys are often conducted anonymously. However, even where surveys are designed to be anonymous there can be factors that can allow survey participants to be identified. These include the size of the agency or the community, questions that relate to unique characteristics of the participants, and the use of free text fields.

Size of the agency

Where an agency is large, eg a department with several thousand staff, the pool of participants will generally be large enough to obfuscate individual identities when dealing with generic responses. However, if the agency is small, eg a statutory body with 50 staff, individual staff members may be identifiable from even generic answers to questions, even on an anonymous survey.

Unique participants

For staff with unique characteristics that are the subject of survey questions, the recording of the characteristic could effectively de-anonymise the survey for those staff. The smaller the agency, the more likely this is to occur, but it can happen even in large departments.  For example, if the survey records the position held by the participant and this position is relatively unique, eg the Director-General, that participant will be readily identifiable.

Free text fields

Free text fields have the potential to enable the identification of both participants and other individuals in an otherwise anonymous survey, for example if a staff member writes the following: I am the Manager of the agency’s ICT Procurement Unit. I have concerns that the Principal Procurement Officer, James Brown, has taken shortcut with his assessment of due diligence for new ICT contracts.

In some instances, the writing style of an individual when entering responses in a free text field can provide sufficient context to identify the writer.

When preparing surveys, agencies should carefully consider whether a free text field is necessary.

Controlling free text fields

Adding a statement to free text fields that specifically asks participants not to include any personal information, of themselves or other people, may assist in keeping the survey anonymous. Agencies may want to consider advising participants that if personal information is provided in the free text field:

  • that it must be limited to their own information,
  • that they do so voluntarily
  • what it will be used for; and
  • if it will be given to other parties.

Advising staff who are completing the survey

Collection notices

When collecting personal information directly from someone, agencies must provide the individual with specific information5 (referred to as a collection notice). These include why the information is being collected and anyone it will be disclosed to. If the survey is being collected under a law, details of that law must also be included in the notice.6

Where a survey is not anonymous, agencies must include a collection notice. Where it is intended to be anonymous but there is a reasonable chance that it could identify the staff member completing it despite the agency's intent, agencies should consider including a collection notice.

For more information refer to All agencies – What to tell people when collecting personal information.

Confidentiality

Agencies often want to tell participants that the survey, its results, and/or any reports or documents produced from those results will be confidential. Confidentiality can be a complex area of the law that requires specific factors to be satisfied before it will apply. As such, care should be taken before giving assurances of confidentiality and agencies may wish to seek legal advice first.

Refer to the Basic guide to confidentiality for more information.

Potential release under RTI

Documents in the possession or control of an agency can be applied for under the Right to Information Act 2009 (Qld) (RTI Act). While personal information is generally subject to a high level of protection, de-identified information or anonymous information may not be subject to the same protection. There is generally a community interest in the results of a workplace survey and staff survey results or related documents may be released if someone applies for them under the RTI Act.7

Agencies may want to ensure, in consultation with their RTI decision maker, that staff understand what is, and is not, likely to be released in the event of an RTI application, so they feel comfortable about participating in the survey.

Refer to Applications for Workplace Surveys for more information.

Surveys for a purpose

Every workplace survey should be conducted for a purpose relevant to the agency's needs and functions. The questions asked should only collect information that relates to why the survey is being conducted.

Where questions collect sensitive information, particularly information related to EEO data, gender, sexuality and domestic violence, agencies should consider making the questions optional—particularly where the pool of participants is small and the answer might identify the person completing the survey.

Example

The gender question on the Working for Queensland survey 2018 offered three choices—male, female, and X (intersex, transgender, or gender diverse)—and was optional. It also explained why data from the question was being collected and why it was valuable. This allowed participants to make an informed decision about whether to provide the information.

As noted above, use of free text fields can risk collecting personal information that may not be required. Generally, using free text fields should only be done where there is a demonstrated need. Automatically including a free text field in every survey can result in collecting information for which the agency has no need, or in de-anonymising a survey which was intended to be anonymous.

Before adding a free text field, agencies should consider if they can elicit the desired information through specific questions.

Using and disclosing staff survey information

If the results of the workplace survey, or documents/reports produced from the results, do not identify specific staff or cannot be used to identify specific staff, then they do not contain personal information. If it is not personal information, agencies can use or disclose it without reference to the privacy principles.

However, as discussed above, the size of the agency can have a significant impact on whether the survey participants are potentially identifiable from the survey results. The smaller the agency, and the more specific and sensitive the questions, the more likely the individuals can be identified.

Agencies will need to take care when, for example, sharing the results of a workplace survey with all staff, and may need to aggregate the results of some questions or skip certain questions altogether to avoid identifying participants.

Using personal information from workplace surveys

Where a workplace survey has collected personal information either explicitly (eg it has been entered in a free text field) or because an individual can reasonably be identified from their responses, any use of it must comply with the privacy principles.

Personal information can be used for the primary purpose of collection. The primary purpose of collection is the reason why the survey was conducted. Agencies should carefully consider why they are conducting the survey and ensure that communications about the survey and/or any statements provided with free text fields reflect that purpose. This will allow the agency to use any personal information inadvertently collected (where the survey is intended to be anonymous) or provided as part of the survey for that purpose.

Example

A survey participant uses the free text field to make comments about another staff member. The agency wants to pass those comments to the other staff member for that staff member's information. Because doing so will identify the survey participant, it will be a use of personal information, so it must comply with the privacy principles.

The purpose of the survey was to evaluate staff satisfaction with their jobs. The agency included no information about what it would do with responses that contained personal information about other people. That makes giving the comments to the other staff member a secondary use of personal information that has to comply with IPP 10(1)—which it would not.

If there had been a statement on the survey/near the free text field advising, for example, that the agency would disclose responses to any third party discussed in the responses to the survey to the third party, then giving the comments to the other staff member would have been a use for the primary purpose of collection. However, agencies would need to consider whether a statement of this kind this could have a negative impact on participation in the survey.

Disclosure of personal information from workplace surveys

Personal information can only be disclosed to the individual it is about unless one of the exceptions in the privacy principles apply. Agencies will need to carefully consider at the time they are preparing the surveys if there is any entity to whom the survey results will be disclosed. This information should be included in communications with staff about the survey. Any other disclosures of personal information resulting from the survey results will need to be carefully considered in light of the requirements of the privacy principles.

Refer to the Use and disclosure guidelines for information.

De-identified information

Agencies may want to consider whether they can achieve desired outcomes using de-identified information. For example, where dealing with identifiable answers to surveys or personal information provided in free text fields that raise issues of concern, identifying specific staff members may not be necessary. Agencies should consider whether they can communicate and/or address the concerns in a more general way that captures the core or theme of the concern, without singling out the staff member or members who raised them.

Refer to Privacy and De-identification for more information.

  • 1 Including Ministers, State departments, local governments, Hospital and Health Services (HHSs), universities and statutory bodies.
  • 2 One example is the Working for Queensland survey conducted annually by Queensland State government agencies.
  • 3 Health agencies (Queensland Health and Hospital and Health Services) must comply with the National Privacy Principles; all other agencies must comply with the Information Privacy Principles.
  • 4 Under chapter 2, part 4 of the IP Act.
  • 5 Set out in NPP 1 for health agencies and IPP 2 for all other agencies.
  • 6 Health agencies also have to ensure participants know how to contact the agency, know they can access their personal information, and understand the consequences, if any, for not completing the survey.
  • 7 Star News Group Pty Ltd  and Southern  Downs  Regional Council [2019] QICmr 39 (12 September 2019)

Current as at: April 30, 2020