Legal Insights

Is it the end off staff surveys?

Melanie Olynyk

• 02 March 2023 • 11 min read

If you’re in executive management or human resources, you may have heard about last year’s VCAT FOI decision, which resulted in the release of council staff survey results under the Freedom of Information Act 1982 (FOI Act).

The fallout from the decision has implications for how staff surveys are managed. In particular, can we surmise that staff surveys are now dead and buried? Or is there still a way to conduct these internal polls without the results becoming available to all?

The VCAT decision - Wellington v Surf Coast SC (Review and Regulation) [2022] VCAT 942 – was handed down in August 2022. While this decision is clearly relevant to all councils and will need to be considered if any requests are made for the results of staff satisfaction surveys, every request needs to be considered on its merits. This decision doesn’t necessarily mean you can’t run future staff surveys. However, the key implication arising from the decision will be determining how to run any future surveys, given the risk that the data (particularly aggregated data) may become accessible under the FOI Act.

Summary of VCAT decision

The VCAT decision concerned a request under the FOI Act for access to Surf Coast Shire Council’s staff satisfaction surveys from 2016 and 2018. The data was presented in the following formats:

‘verbatim responses’ (including free text comments made by individual participants)
excel spreadsheets setting out overall results on a scale of ‘agree’ to ‘disagree’ and
tabular reports of the overall statistical responses.

The Council had hired a consultant to undertake the survey with its staff. Under its protocols, any reporting group had to include at least ten responses for any selected cohort and more than 30 verbatim responses before a report could be generated. The FOI applicant, who was one of the Council’s own Councillors, contended this was consistent with accepted approaches to achieving anonymity, and it meant results were reported in a statistically safe manner where identities could not be gathered from the data regardless of what information the person may have.

The key aspects of the decision were:

VCAT determined verbatim comments in free text fields could ‘unmask’ or identify individuals and so were exempt from release under s 33 of the FOI Act as well as under s 35(1)(b) on the basis that disclosure of them would be likely to impair the Council’s ability to obtain similar information in the future.
By contrast, VCAT also determined aggregated data, contained in the spreadsheets and reports was not exempt. This data included tabulated statistical breakdowns according to seniority, gender and age. VCAT was not satisfied this data would identify individuals or be contrary to the public interest to disclose as it did not form part of the Council’s deliberative functions and would not impair its ability to obtain similar information in the future.
In addition, VCAT was influenced by the age of the data, being 4-6 years old, and the public interest in its release, contributing to transparency in local government.

Implications for previous surveys

The most significant implication arising from this VCAT decision for previous Council staff surveys is whether this data may need to be provided under the FOI Act whenever a request is made.

This decision is clearly relevant when considering whether information regarding staff satisfaction surveys should be released under the FOI Act. However, it is not as simple as saying the VCAT decision now requires councils to provide copies of all staff satisfaction surveys whenever they are requested. As with any request made under the FOI Act, councils will need to assess exactly what is requested and whether it is exempt from release under the FOI Act.

With this in mind, what do you do if you receive a request for past surveys? First, ensure any request is made formally under the FOI Act and is a valid request. For example, this means the application fee has been paid (unless waived), and the terms of the request are sufficiently clear. This can be an issue where the survey is held in a variety of forms (such as reports, presentations, data in an online portal, etc.). Consider what particular format is being requested.

Once you are clear on what information is being requested, you then need to obtain that information, reviewing each page to determine whether it must be released under the FOI Act. This includes considering whether it is possible to ‘unmask’ individuals from the information provided or whether it would prevent your ability to obtain similar information in the future if it were to be released. For example, if you hold verbatim comments or survey results for a cohort of less than 10 in size, you could consider redacting this information under ss 33 and 35 of the FOI Act.

As part of this assessment process, you will also need to decide whether it is practicable to consult with your staff about the disclosure of the information, particularly where it may identify them or has been given in confidence. If you do consult them, you will need to take into account their views and, if a decision is made contrary to their views, notify them of their right to seek a review of the decision by VCAT.

An FOI applicant also has a right to appeal where some or all of the information is not provided to them, first to the Office of the Victorian Information Commissioner and then on to VCAT.

Implications for future surveys

Looking ahead, the key implication of the VCAT decision will be determining how future staff surveys are managed. This will now involve a series of considerations, including the following check-list:

Key questions and considerations

What kind of data is sought?

For surveys that only collect free text responses, this may mean the information is more likely to identify individuals (either the survey respondents themselves or those mentioned by them in their responses).

In turn, this may also make it more likely to be exempt under the FOI Act. Conversely, aggregated statistical data which does not identify individuals are more likely to be required to be released.
How is the survey to be undertaken?

Ask if survey responses will only be given verbally or if there will be a variety of means to provide answers (for example, by email only to the CEO, via an internal mailbox, or though an online survey, etc)
How will the information be produced? Will responses only be verbal from a consultant, through reports produced by a consultant, or through access to data on a consultant’s portal?

There is a risk all of this kind of data may be accessible under the FOI Act, where it cannot identify individuals.

The size of each cohort of respondents also needs to be considered. It is generally recognised that a response size of 10 or more makes the information statistically safe. Anything less may make it more likely for individuals to be identified.
If information is made available through a portal, what kind of data will be included?

Where a portal is used to collect responses, particular consideration needs to be given to whether the council has instant access to data in a portal and can extract it in various ways, as that data is also likely to be considered in its possession.
Who has access to the data?

The more strictly controlled the access to the data is, the better the argument is that its disclosure would be unreasonable. Think about this in terms of access by the executive leadership team, other staff (including those with management responsibility, such as mangers and team leaders) and councillors.
How will the information received be used by your council?

In addition to thinking about who will have access to the data, think about how the information received is to be used by your Council. Will it be used to prepare further reports, or to hold briefings (of staff or councillors)? Does your council need detailed written reports if it is only going to use them to hold verbal briefings?
What are the terms of engagement for consultants?

The FOI Act can extend to documents held by consultants in certain circumstances. This is because a council can be considered to have ‘constructive’ possession of information if it has a right of control over it, even if it is not in its actual possession. This is influenced by the terms of engagement for any consultant. For example, if a council engaged a consultant to undertake a survey but the agreement gave the council a right to the data collected, then the council is likely to be found to have constructive possession of all of the data collected by the consultant, even if it is not in the council’s actual possession.

Instead, the council could provide the consultant ownership of the data, only requesting information in particular forms or asking to view information at the consultant’s offices (or via screen sharing when meeting online) but not retain a copy in the council’s own records.

You should also ensure any agreement with a consultant does not prohibit your Council from providing the information to others without the consultant’s consent, given councils have obligations under the FOI Act.
How long is information kept on file?

Consideration needs to be given to how long the information must be retained. This includes understanding what information is able to be destroyed under normal administrative practice, such as rough notes and draft documents.
What information is conveyed to staff prior to the survey (where confidentiality is important but also cannot be overstated)?

Carefully consider what information is provided to staff about how the survey is conducted, including how the survey responses will be used and disclosed. There is a difference between anonymity and confidentiality. While your council may want to give an assurance of both, this may not actually reflect the reality.

From a privacy perspective, staff need to be told about the purpose for which the information is being collected and how it will be used and disclosed, including within the organisation. Is there some information that will be treated more confidentially than others – such as verbatim comments (the access to which is strictly limited) versus the qualitative scale-based questions?

What this means for councils

All of this means that your Council may wish to revisit the way in which it undertakes staff surveys by reviewing its approaches to obtaining feedback from your people Four simple steps to begin the process are:

Revisiting communications to staff about how surveys are conducted and how the survey responses will be used and disclosed;
increasing the amount of information received verbally -potentially by both survey respondents and from any consultant engaged to assist in the process -and reducing the amount of information received in writing to only high level summaries or themes;
not having access to a portal or limiting it to ‘view only’ functionality;
destroying information as soon as it is no longer required to be retained.

For further questions about how your Council should conduct future surveys or any other FOI or privacy issues, please contact Partner Melanie Olynyk from our Privacy Team.