Dear Tom,

 

I refer to your two requests of 13 November 2024, which we have conflated
into a single request, for information related to professional staff
performance ratings grouped by various demographics, for professional
staff pay bands (B-L).

 

As consultations necessary to make a decision on your request are such
that a response cannot reasonably be given within the original time limit,
the University has extended the time limit for your request by 25 working
days under section 15A(1)(b) of the OIA. The new maximum time limit for
the University’s response is 6 February 2025 (this response date takes
into account the summer holiday period of 25 December to 15 January
inclusive, which are not counted as ‘working days’); we will respond to
your request as soon as reasonably practicable.

 

For more information regarding the summer holiday period please refer to
the ‘Processing Requirements’ section (pages 16-17) of the Ombudsman’s
guide [1]The OIA for Ministers and agencies.

 

You have the right to make a complaint to an Ombudsman if you are
dissatisfied with this extension.

 

Yours sincerely,

Farleigh Quinlivan

 

show quoted sections

Dear Tom,

 

I refer to your two requests of 13 November 2024. We apologise for the
delay in our response.  The University’s response follows:

 

Your first request asked for:

“I’m interested in how the University is fulfilling its commitments to
fairness, respect, wellbeing, and openness, as articulated in Taumata
Teitei. One way to measure the application of these values is by examining
how staff are assessed for their performance, and in particular, if there
are any differences in how staff from different backgrounds are being
assessed and rewarded. The University’s Tupu performance review system
assigns a rating or score to each professional staff member based on their
manager’s perception of their performance, and I’m interested in how these
scores vary across gender, ethnicity, and faculty/service division.

 

For each Tupu assessment year, could you please provide the total number
of staff who received each score for their end of year review for each
professional staff pay band (B-H), and could you please group by gender
and ethnicity, and also by faculty/service division.

 

Please also explain any relevant caveats that should be kept in mind when
analysing or interpreting this information. Please provide this
information in an accessible spreadsheet format (.csv or .xlsx).”

 

Your subsequent request was for the same information as above, but for pay
band (I-L), so we have conflated the two requests into a single request.
We have decided to provide the Tupu performance information requested,
subject to the following:

 

o Your request was for ‘each Tupu assessment year’.  The TUPU
performance rating system was implemented for the year-ending 2022,
therefore, there is no TUPU performance data for the years prior to
2022.
o The analysis of the raw data from 2024 has not yet been completed, and
therefore the requested information (the assessment of that data) is
not held.  This constitutes a partial refusal of your request is
refused under section 18(g).  You are welcome to make a further
request for this information in due course, and it will be responded
to, in accordance with the Act.
o To ensure we do not breach employee privacy and that individual staff
members cannot be identified through their ratings, where the data
points have fewer than 5 staff ratings, this information is withheld
under section 9(2)(a) to protect the privacy of individuals.

 

The data we have decided to provide for bands B-H is attached. The data we
have decided to provide for bands I-L will be provided as soon as
reasonably practicable.

 

In response to your request to explain any relevant caveats that should be
kept in mind when analysing or interpreting this information, our Human
Resources team advises the following:

 

 1. Ethnicity data is not collected for performance reporting, so it
should be interpreted with caution. Employees self-identify their
ethnicity and can select multiple options without specifying a primary
ethnicity. The reported ethnicity reflects the first option selected,
which may not be their primary ethnicity. Additionally, some employees
do not select any ethnicity in the system which means that data is not
fully representative.
 2. The TUPU system has only been running for three years and data for
2024 is not yet available.  It is important not to over-interpret data
for 2022 and 2023, as data for only two years cannot provide true
insights as to trends.  Additionally, 2022 was the first year the TUPU
system was in place and may not be representative of how the system
works when it is bedded in.

 a. Analysis should be undertaken separately for each data set (2022
and 2023) to ensure accuracy.
 b. Year-to-year differences are important to accurately tracking
trends and performance changes. Combining data from both years
hides these variations, making it harder to identify patterns or
improvements specific to each year.
 c. Combining data from two years also ignores potential changes in
annual external factors, team dynamics, and goals of that year,
obscuring key insights.

 3. It is important to exercise caution in implying a value where none is
given (e.g. implying a value of zero where there is no value given). 
This introduces incorrect values into the dataset and can create
biases, particularly in smaller groups.
 4. Caution should be exercised when working with this data, to ensure
that analysis does not have the effect of artificially lowering
metrics and skewing averages and totals.

 

You have the right to make a complaint to an Ombudsman if you are
dissatisfied with this response.

 

Yours sincerely,

 

Landon Watt
Legal Advisor
Office of the Vice-Chancellor

Waipapa Taumata Rau | University of Auckland