All your equity data questions answered
At QuakeLab, we believe strongly in the importance of collecting and analyzing disaggregated data when identifying and addressing inequity. But make no mistake, this is a loaded statement that should not be taken lightly.
The QuakeLab team has dedicated years to learning, researching, iterating and building our approach to collecting data. This is a never ending process based in data ethics, justice and taking guidance from marginalized communities.
So before we jump into answering the question we get most, we want to lay out a few key points:
Collecting equity data, especially demographic data can be incredibly sensitive for individuals. Before you put pen to paper, or send out a survey, you will need to do a lot of planning and trust building. Many of the people of your team may have never self identified in a workplace and it’s critical to approach this with care, intentionality and transparency;
Data collection has been historically weaponized against many marginalized groups, especially Indigenous peoples, disabled folks, queer and trans folks and Black people. So understand that they may not be eager to participate at first - or ever.
The purpose of equity work isn’t diversity, It’s equity. Collecting demographic data shouldn’t be an exercise in counting the marginalized people in your company so you can add more or pat yourself on the back. The idea of collecting demographic data, purely for meeting quotas is incredibly dehumanizing and treats marginalized people as pawns in a diversity game rather than complex human beings who only exist to meet your organizational goals.
Reading this blog post will not make you an equity data expert, we strongly recommend getting external support to do this work if possible because a lot of harm can be done through inequitable and unethical data collection practices. However, if external support isn’t accessible, take your time and ensure you’re doing this right!
It is critical to ensure everything you collect has a purpose. Do not collect data just out of curiosity, this goes against ethical data principles. Be clear about why you are collecting this data and what you will use it for.
Alright, let’s dive in!
Is it illegal to collect demographic data?
No! Let’s hear what the Ontario Human Rights Commission has to say though:
“Many people think that collecting and analyzing data that identifies people on the basis of race, disability, sexual orientation and other Ontario Human Rights Code[1] (the Code) grounds is not allowed. But collecting data on Code grounds for a Code-consistent purpose is permitted, and is in accordance with Canada’s human rights legislative framework, including the Code, the Canadian Human Rights Act[2], the federal Employment Equity Act[3], and section 15(2) of the Charter of Rights and Freedoms[4]. The Ontario Human Rights Commission (the OHRC) has found that data collection can play a useful and often essential role in creating strong human rights and human resources strategies for organizations in the public, private and non-profit sectors.”
That being said, there are provincial and federal laws and guidelines that govern how you can collect and store what is considered personal data, so be mindful of this!
What if no one wants to answer?
It happens. From the Indian Pass System, eugenics, the Indian Act to Police Carding, marginalized people in Canada have had data collection weaponized for further oppression. So it makes sense that many of these same communities are wary of any data collection efforts where they need to self identify. For this reason, trust building and transparency is a critical part of data collection. Even then, some folks may still choose not to participate in collecting data, and that's ok.
That being said, you might run into scenarios where people on your team do not believe in answering demographic questions. In our experience running data collection, there are usually one or two people in every company whose data we have had to clean from the final analysis because they’ve made it clear they will not self identify and/or completely distort or lie. When we see that happen, it tells us that there may be a culture of conservatism or skepticism around this type of DEI work. This is already a red flag.
So recommendations would be setting the stage for this DEI work with employees before starting. Send a clear message that this work is important and that the company believes in it and expects its employees to do the same; setting an example for why employees' experiences are, in fact, impacted by their identity and how they are viewed by others; denouncing the "colourblind" perspective that individuals may ascribe to.
Anonymous vs confidential?
Confidentiality (via informed consent) defines a scenario where individual responses and identities are collected, but not disclosed.
Anonymity defines the process where no personal identifiers (e.g., name, address, telephone number) that link responses to a specific individual are collected.
At QuakeLab, we prefer to collect anonymous data rather than confidential data. We do this to build trust between us and the respondents at the companies we work with. We also do this because we understand that sometimes, the folks causing the most harm in an organisation, may be the ones tasked with collecting data so it’s critical we keep data anonymous at all times. This also informs our decision not to share raw data with our clients.
I don't know all the right terms for all the right identity groups, what should I do?
If you are doing this work, you need to budget quite a bit of time doing some comprehensive research. This includes exploring the discussions within communities around how they self identify, and the most appropriate language to use. We strongly recommend doing this research before or in place of asking colleagues to provide this information especially if they haven’t volunteered and aren’t being compensated.
You’ll want to watch out for common ‘well-meaning’ language that is deeply offensive but has been thought of as ok in the past eg. using ‘differently abled’ or ‘special needs’ versus just using disabled.
What questions do I need to be asking?
Collecting equity data isn’t about figuring out what quotas you need to set, or to satisfy curiosities. We believer equity data should be used to
Identify patterns of inequity.
Identify who is not being well served by your processes, policies and procedures.
If these two points are your guiding light, you can begin to bring together the data you should collect. Additionally, we believe in the importance of capturing as many identifiers as possible so your analysis is intersectional. This includes race, gender, ethnicity, visible and invisible disabilities and more.
Should I only collect demographic data?
Remember that you are not running an investigation, which means you also want to collect non-demographic data that helps identify patterns without forcing respondents into trauma mining. You want to explore how often people take time off, how quickly or slowly they have been professionally mobile (eg. promotions), if they have accessed accommodations for their disability etc. This will allow you to run a cross analysis to identify if certain groups are navigating specific barriers at a higher level eg. Black women moving up into senior positions at much slower speed or disabled folks taking less parental leave for fear of punishment.
In fact, we strongly discourage you to only collect demographic data because your equity actions will follow this stream of thinking that focuses on diversification and not systems analysis and change.
How do you use the info you've collected?
Data analysis and creating actions based on that analysis are your next steps. Raw data on its own can only do so much, whereas data analysis is the process of inspecting, cleaning, transforming, and modelling your data with the goal of discovering useful information, informing conclusions, and supporting decision-making.
As is the ongoing theme, it’s important to take as much time as possible to dive into research and best practices for data analysis. Here’s where we truly recommend working with experts who already have the processes, tools and knowledge to properly clean data, identify what is statistically significant, etc.
This analysis should then help you identify your organizational equity challenges so that you can begin focusing on focused solutions to these challenges.
How do you share the insights you learn?
Sometimes, we work with organizations that have already dipped their toes into data collection. Often, we find that after they collect data, they do not effectively communicate their findings to the rest of the team or how the insights will inform a strategy or plan going forward. At QuakeLab we follow three key principles in data collection:
Clearly communicate what you’re collecting and why
Clearly communicate how you’re going to be using and storing the data
Clearly communicate what you found and how they can access that information
Throughout data collection, communication and transparency is critical. However, when communicating your insights, it’s important that you set up mechanisms to ensure you are maintaining anonymity. We do this through sharing analyzed data rather than raw data - however this may have an added level of complexity when the analysis is coming from within the company rather than an external company. Additionally, we group our insights to ensure there is no way for individual responses to be pulled out of the insights.
Once again, it is critical that you build out an internal (and potentially external) communications plan to lay out the insights of your collection process and what next steps will be taken. Consider sharing:
The methodology you used;
The equity challenges you found, who they affected and how;
What next steps you’ll be taking to either develop solutions or to implement the solutions that have already been developed
How you’ll be measuring the progress and effectiveness of your solutions
How does collecting demographic data work in service of building equity?
Edmonton Social Planning Council did an excellent job outlining the importance of collecting race-based data in their 2021 report:
“History has shown that race-based data can be used to uphold racist systems and discriminatory practices; but data can also help to dismantle them. Currently, race-based data is collected in only a few key systems, and data collection strategies are woefully inadequate for current needs (in areas such as health, justice, and education). The limited data that is available does not provide adequate evidence to support targeted policy change and intervention. Race-based data is crucial to develop effective anti-racism frameworks, and to understand the diverse, intersectional, needs of racialized communities in Canada.”
We would extend this explanation further than race, but to identity in general.
To expand this a little further, collecting equity data (which includes questions to help identify patterns of inequity), are critical in identifying the equity challenges that need to be addressed, who they affect, and how they affect them.
How do you analyze demographic data?
We’re going to be really honest, at QuakeLab, we have invested the time and resources into building the infrastructure and gaining access to the tools we need for robust analysis. We have a data analysis team (TheLab) who use tools like paid Survey Monkey, Google Spreadsheets, and coding for analysis.
We understand these tools and expertise may not be accessible to you, so we recommend the below:
Develop questions you want answers to ahead of time so you can clearly search for the answers in your data;
Earlier, we recommended including questions about how your team navigates systems like professional mobility, access to accommodations, etc. We encourage you to measure the answers for these questions against demographics eg. Are Indigenous women accessing professional development resources less than other folks?
Use demographic data not to figure out how you need to be diversifying but who is already at the organization and is being underserved eg. you find out 30% or respondents are disabled (visibly and invisibly), and then explore the accommodations made available, the avenue for requesting accommodations, etc.
Use the demographic data to learn about who is here, and who isn’t. However, we want to stress once again that using demographics to go through a ‘plug and play’ is dehumanizing and won’t solve inequity. So work on focusing on the systems that have created barriers for specific groups in recruitment.
Be transparent about the process you took at each stage.
Do I need to hire externally to collect this info?
Ideally, yes. However, if resources are truly not available, we recommend taking a lot of time doing your research into how to properly do this work.
Obviously this hasn’t been an exhaustive list of questions and answers, so if you’re looking for some support we are thrilled to work with you, give us a shout and let’s talk!