If an organization is considering whether to collect data on its own or get help from an external consultant, it will need to have enough information to make an informed decision about how to proceed.
This section outlines some of the key considerations that may arise during various steps in the data collection process. There is no requirement that these steps be followed or pursued in the order that they are written. The model presented is offered as a reference tool. How data is gathered and analyzed depends on many factors, including the context, the issue that needs to be monitored, the purpose of the data collection, and the nature and size of the organization.
The main consideration is to make sure that any information collected is done in a way and for a purpose that is consistent with the Code and complies with freedom of information and privacy protection legislation. In the interest of effectiveness and efficiency, it is recommended that efforts be made to collect data that will shed light on issues or opportunities. To protect the credibility and reliability of data, information should be gathered using accepted data collection techniques.
The first step is to identify issues and/or opportunities for collecting data and to decide what next steps to take. To do this, it may be helpful to conduct an internal and external assessment to understand what is happening inside and outside of your organization.
Some organizations, like FCP and Legislated Employment Equity Plan (LEEP) employers, are given specific direction on what issues should be explored and how data must be collected. Other organizations may have more flexibility to decide when and how to collect information to achieve certain goals. Some of the non-exhaustive questions identified below may apply to a diverse range of organizations and audiences, including employees and service users. Depending on the organization, these questions may be considered at Step 1, or at different stages in a data collection process.
Finding the above information can be challenging for smaller organizations, but the internet offers a wealth of resources to choose from. Media reports may offer insights, as well as on-line resources offered by the OHRC, Statistics Canada, the City of Toronto, government agencies, and community organizations that focus on Code and non-Code ground-related topics. Information may also be gathered from various sources using accepted data collection research methodologies discussed in Step 3.
It is to be expected that an internal and external assessment of the organization, in light of the questions listed above, may result in a number of potential issues and/or opportunities for exploring data collection. Before proceeding to Step 2, organizations may wish to consider whether there are any preliminary actions that can be taken to address these issues and/or opportunities, without collecting data (e.g. training, policy development).
Example: The review in Step 1 may have identified the following issues and/or opportunities for collecting data:
The above examples present a potential opportunity or problematic human rights issue, respectively, and could lend themselves to data collection. Decisions need to be made about how best to address the identified opportunities and/or issues and whether it would be appropriate to act, based on the assessments in Step 1 (either instead of or together with further data collection).
If the results of the internal and external assessment seem to show that the organization does not have any pressing problems with discrimination and/or systemic barriers, and is generally in compliance with the Code and OHRC policies, consider whether the organization could still benefit from proactively implementing a data collection initiative (for example, to help monitor the ongoing effectiveness and suitability of policies, programs and intervention strategies).
The focus of Step 2 is choosing a priority issue(s) and/or opportunity(ies) for collecting data, and then setting goals and objectives.
The organization reviews the issues and/or opportunities identified from the internal and external assessment done in Step 1, and picks one or more specific issues and/or opportunities for starting a data collection project from among the list of priorities. Some of the questions an organization can consider when deciding to prioritize an issue and/or opportunity for gathering data include:
Example: An aging taxpayer base provides a government body with a pressing reason to collect data on this group’s projected size, needs and revenue base. This changing demographic also presents an opportunity for the government body to ensure that it is proactively developing policies, programs and services that are accessible and appropriate to meet the needs and concerns of these taxpayers.
While the organization may intend to collect data relating to multiple issues and/or opportunities at the same time, the next steps, including goal-setting, should be individualized for each issue and/or opportunity.
The specific goal(s) defined for each issue and/or opportunity may depend on a hypothesis or guess about what is happening that can be tested using data collection techniques and analysis.
Example: A downtown Toronto hotel receives complaints from guests, who self-identify as being gay, about the unwelcome treatment they received from staff. A hypothesis might be that hotel staff lack sufficient awareness and training about how to deal respectfully with guests who are gay, or are perceived to be from the larger LGBT community. The goal is to get enough evidence to test this hypothesis.
Step 2 can also involve an organization brainstorming a smaller set of questions that may be answered by collecting data. Rather than asking a general question like, “Is there any evidence of discrimination on the basis of sexual orientation or gender identity in this hotel?” one might ask, “What percentage of hotel guests self-identify as being part of the LGBT community?” and “What are the perceptions of the service received by self-identified LGBT patrons?” Ultimately, data that is collected should be rationally connected to the goals set and the overall purpose for collecting the data.
In Step 3, organizations will make decisions about who will be surveyed, how data will be collected, the sources of data that will be used, and the duration of the data collection project, among other questions. These decisions may be made in consultation with an expert. The methods and approaches will flow from the goals set in Step 2, and will vary significantly depending on a number of factors, including the organization’s context, size, resources, and the purpose and complexity of the issue(s) or opportunity(ies) selected.
Some of the questions to consider at this stage include:
The “group of interest” (e.g. youth service users of a local community centre who cannot read and speak English as a second language) will be the focus of the study, and the data collection methods used will refer to this group, or the persons within it, depending on the goals of the project.
Example: A South Asian male youth service user, who cannot read and speaks limited English, may face discrimination on any of the grounds of age, race, colour, ancestry, ethnic origin, place of origin, gender, disability or perceived disability (e.g. could be seen as having a learning disability). However, he may also be exposed to discrimination on intersecting grounds based on being identified as a “young, illiterate Indian male from a foreign country,” based on the various assumptions or stereotypes that are uniquely associated with this socially significant interaction of multiple identity factors.
To better understand the potential impact of multiple identity factors, or intersectionality, when collecting and analyzing data about a group of interest, it may be helpful to consult with communities, and review applicable research and other relevant documents that highlight how the dynamic of discrimination and disadvantage can play out in a practical way for persons identified by Code and non-Code grounds. The OHRC’s recent edition of Human Rights at Work is a useful reference for this purpose. The OHRC has also developed policies and guidelines that provide a more detailed outline of how the Code applies to the various grounds (see Appendix G for a list of OHRC guides, policies and guidelines).
The “comparator group” should be persons who share one or more characteristics with the persons in the group of interest, but differ in the key characteristic(s) being studied (e.g. youth service users who cannot read but can speak English fluently). The experiences of youth service users who cannot read and who speak English as a second language can then be compared to youth service users who cannot read but can speak English fluently.
Some data collection initiatives require gathering data from multiple sizes, groups or communities located in different locations and geographical areas. When determining where to collect information from, key factors to consider include who the data will be collected about and who the data will be compared to.
Example: A local community centre is interested in making its current youth literacy program more responsive to the needs of an increased number of youth in the surrounding area who cannot read and who speak English as a second language. The community centre plans to gather information about the community it serves and the geographical region it is located in. Data is gathered from the community centre’s pre-existing records relating to its service users, including people who attend the youth literacy program or have expressed an interest in it. Publicly available information about the characteristics of the surrounding neighbourhood is also explored, among other data sources.
Choosing categories provides a way to organize the information that is collected. This can be done either before collecting data, as discussed in this step, or after data is collected (see Step 5).
In some cases, although it is not required, it is preferable to use pre-determined categories such as those developed by Statistics Canada. There are certain benefits to this approach.
Example: Organizations can be confident that the 12 racial groups used by Statistics Canada will represent how the majority of Canadians racially classify themselves. In addition, use of these categories is most likely to produce reliable and valid results and enable researchers to directly compare the results of their studies to Census data collected by Statistics Canada.
The limitations are that if these categories are used, some respondents may not identify with them or may object to them. Another limitation is that Statistics Canada does not produce Census data on all grounds (for example, on sexual orientation).
For a fee, Statistics Canada will customize its data. For example, it can break it down to “disaggregated” data for a local labour market or for a particular occupational category.
Another limitation is that the Statistics Canada categories may be too broad depending on the goals selected in Step 2.
Example: Using a broad category such as “racialized” can mask important differences between racialized groups, since racialized groups are not subject to exactly the same experiences, racial stereotypes and types of discrimination. However, when it is necessary to describe people collectively, the term “racialized person” or “racialized group” is preferred over terms like “racial minority,” “visible minority,” “person of colour” or “non-White” as it expresses race as a social construct rather than as a description based on perceived biological traits. Also, these other terms treat “White” as the norm that racialized persons are to be compared to, and have a tendency to group all racialized persons in one category, as if they are all the same.
Consider other categories to describe the groups selected (for example, relating to job or service categories). Organizations may ultimately choose the categories that best reflect where the organization is at in terms of achieving its human rights, equity and diversity goals.
In the context of human rights, social-science researchers are commonly asked to lead or help with data collection projects. Two types of data are used in social science research: qualitative and quantitative. A good research effort involves the use of both types. Both approaches, while distinct, can overlap and rely on the other to produce meaningful data, analysis and results.
Example: A restaurant chain wants to improve service and access to customers with disabilities. Management decides to collect qualitative information using focus groups consisting of a range of stakeholders, including customers and representatives of organizations from the disability community.
Example: A simple 1- 5 rating variable for the survey statement, “My union handles human rights grievances in a sensitive and efficient manner” gives respondents the option of circling: 1 (Strongly Disagree), 2 (Disagree), 3 (Neutral) 4 (Agree) and 5 (Strongly Agree).
A respondent circles “2 = Disagree.” To understand the value of “2” here, a researcher must consider some of the judgments and assumptions that are behind this choice. Did the respondent understand the term "human rights grievance"? Has the respondent had experience filing a grievance with the union? Does the respondent like unions generally?
a focus on numbers and rankings alone can overly simplify or lead to an inaccurate understanding of complex situations and realities, unless a broader context is provided
Example: An employment data survey of the Custodial Services Division of a large organization reveals that 80% of the cleaning staff are women and that 6 of 7 Custodial Services supervisors are men. A comparison between these figures and gap data from Human Resources and Skills Development Canada (HRSDC) shows that, while there is an overrepresentation of women in the ranks of cleaners, there is no gap for women in the ranks of supervisors.
The reason for the seeming discrepancy is that HRSDC gap data is based on availability. Nationally, so few women are Custodial Services supervisors that there is a statistically insignificant availability, giving rise to the conclusion that there is no numerical gap with respect to women supervisors. This conclusion, however, does not make sense since the organization knows that the 200:40 women to men cleaning staff ratio is supervised by a 6:1 male to female supervisory staff ratio. The organization decides to ignore the HRSDC data and apply common sense by setting up career advancement mentoring and other policies and programs to increase the number of female supervisors in its workforce.
Qualitative and quantitative data are generally gathered from more than one source. Where possible, two or more of the following sources should be used together to strengthen reliability and consistency in results.
Pre-existing or official data is information that has already been documented (e.g. newspaper clippings, case law, Statistics Canada census data, photographs) or is created by an organization during its routine business operations (e.g. employee personnel files, student registration forms, annual reports, occurrence reports). This data may contain information that directly relates to specific Code grounds like race, but more commonly will relate only indirectly (for example, in the form of names, place of origin or ethnicity). This type of information could be used as proxies or stand-ins for race, but would be less reliable than actually having self-reported racial data.
Example: Outcomes of workplace recruitment, hiring, promotions and terminations can be recorded, as can events such as interventions by security guards and customer complaints. When recording these events, relevant Code ground and non-Code classifications could also be included. This data could then be examined for trends over time to show whether discrimination or systemic barriers exist, may exist or do not exist.
Survey research is a broad area and generally includes any measurement procedures that involve asking respondents questions. A "survey" can range from a short paper-and-pencil questionnaire to an in-depth one-on-one interview (interviews will be discussed further below).
In designing a survey, it is important to consider the specific characteristics of the respondents, to make sure that the questions are relevant, clear, accessible and easy to understand. Some practical considerations to keep in mind are whether the respondents can read, have language or cultural barriers, have disabilities, and can be easily reached.
Example: The TDSB’s 2006 Student Census, Grades 7-12 System Overview included a component on how senior and secondary school students generally perceived their schooling and out-of-school experiences in 10 areas, including school safety and home support and involvement.
Example: A transgender employee may self-identify as female but a third party may identify her as male.
Interviews and focus groups (also referred to as “group interviews”) allow for information to be provided orally, either individually or in a group setting. The data can be recorded in a wide variety of ways including written notes, audio recording and video recording.
In focus groups, the interviewer facilitates the session. A select group of people are brought together, asked questions, encouraged to listen to each other's comments, and have their answers recorded. The same set of questions may be used for a number of different groups, each of which is constituted slightly differently, and for a range of purposes.
Focus groups may be facilitated by professionals, but this is not always needed. The decision to hire a professional facilitator may depend on the goals of the focus group research, the nature of the questions asked, the skills and experience of staff taking part, and the need for confidentiality or anonymity.
Example: To get the unique perspective of each group, an organization may wish to hold separate focus groups for representatives of each of the organization’s internal and external stakeholder groups, such as senior management, front-line employees, service users, union representatives and community groups. Or, it may be of greater value to organize a group that includes people representing all key internal and external stakeholders, to allow for contrasting ideas to be expressed and discussed.
Whatever format is chosen, it is important that the focus group is structured and managed in a way that cultivates a “safe space” for people to share their experiences. In some cases, this may not be possible without setting up separate focus groups or hiring a professional facilitator who is not connected to the organization.
Typically, interviews involve a set of standard questions being asked of all respondents, on a one-on-one basis, so that accurate trends and gaps can be drawn from the data. Interviews are commonly conducted face-to-face, but for more rapid results, can also be done over the telephone, or, as technology advances, through video-conferencing and other means.
Trained staff or external experts can gather data by identifying and recording the characteristics and behaviour of research subjects through observation, either within or outside of an organization. Observed data can include information gathered using all of the senses available to the researcher, including sight, hearing, smell, taste and touch.
Example: A human rights organization that offers a mediation service hires a mediation expert to observe mediators and service users and provide feedback about any issues of concern related to human rights. To minimize potential stress and anxiety experienced by the people being observed, staff and service users are informed in advance of the purpose and goals of the exercise. Service users’ consent is sought. Staff is advised that the observed data gathered will only be used for research purposes and not shared with their managers. The expert maintains access to the data, and the results are reported on an aggregated and summarized basis to prevent individuals from being identified.
Hiring experts, while potentially expensive, can add validity and credibility to research analysis because they are often perceived as having no vested interest in the research results.
Information gathered using observation techniques differs from interviewing, because the observer does not actively ask the respondent questions. Observed data can include everything from field research, where someone lives in another context or culture for a period of time (participant observation), to photographs that show the interaction between service providers and service users (direct observation). The data can be recorded in many of the same ways as interviews (taking notes, audio, video) and through pictures, photos or drawings.
Each source of data used to collect information has its strengths and weaknesses. Some of the more common potential strengths and weaknesses identified above have been highlighted. Analyzing data from multiple perspectives and relying on data from different sources can strengthen the conclusions drawn from research. A combination of statistical analysis, observational data, legal analysis, documentary analysis, in-depth interviews and external and/or internal consultation can help maximize understanding of a given situation. Organizations should choose the sources of data that best suit their program goals, context, resources and organizational culture.
Data can be collected and analyzed on a short-term or project basis in response to situations or needs that arise from time to time. A short-term data collection project would include a start and a finish date, with set deliverables to be carried out over a certain period of time.
The best practice is to collect data on an ongoing, permanent basis, and to analyze this data as often as is needed to identify, address and monitor barriers to Code-protected persons or other persons based on non-Code grounds.
Data collected in a time-limited study may be less complete than data collected through ongoing monitoring. This is because short-term studies do not allow for the assessment of trends, patterns or changes over time. However, where costs, time and resources are a factor, short-term studies may be the preferred choice to fulfil a need and project goals.
Other factors may also influence the reliability of the data. For example, people may modify behaviour while under scrutiny during the data collection period.
When planning on how best to collect data in Step 4, it is important to be aware of the practical considerations and best practices for addressing logistical challenges organizations often face at this stage of the process. Implementing a data collection plan requires attention to matters such as:
Step 5 involves analyzing and interpreting the data collected. Whether quantitative and/or qualitative methods of gathering data are used, the analysis can be complex, or less so, depending on the methods used and the amount of data collected.
Explaining the technical steps involved in analyzing and interpreting data is beyond the scope of this guide. An organization will have to determine whether it has the internal capacity and expertise to analyze and interpret data itself, or whether it will need the help of an external consultant.
A smaller organization that has basic data collection needs may be able to rely on internal expertise and existing resources to interpret the meaning of gathered data.
Example: An organization with 50 employees wants to find out if it has enough women working in management positions, and if there are barriers to equal opportunity and advancement. The organization counts the number of female employees it has (25), and determines how many of these employees are working in supervisory and management positions (two). A few motivated employees identify some issues of concern, like gender discrimination, that may have broader implications for the organization as a whole.
After deciding to do an internal and external assessment (Step 1), and gather qualitative data using focus groups and interviews with current and past employees, senior leadership decides that barriers exist for women in the organization’s recruitment, hiring, promotion and human resources policies, processes and practices. Efforts are made to work with female employees, human resources and other staff to address these barriers. The organization makes a commitment to foster a more equitable, inclusive work environment for all employees.
Once an organization has analyzed and interpreted the results of the data collected, it may decide to act on the data, collect more of the same type of data or modify its approach.
Quantitative and qualitative information can provide a solid basis for creating an effective action plan designed to achieve strategic organizational human resources, human rights, equity and diversity goals identified through the data collection process. If an organization feels it has enough information to develop an action plan, it should consider including the following elements:
In some cases, an organization may decide that it needs to collect more information because there are gaps in the data collected, or areas where the data is unclear or inconclusive. This may prompt them to conduct a more detailed internal and external assessment (go back to Step 1) or try another approach.
In the end, there is no one or “right way” to conduct a data collection initiative. The experiences of Mount Sinai Hospital, KPMG Canada, the Keewatin-Patricia District School Board, TD Bank Financial Group, the University of Guelph and the DiverseCity Counts project and featured in the Appendices reflect this statement, yet also show some similarities in terms of the best practices and lessons learned.
Step 1: Identify issues and/or opportunities for collecting data
Step 2: Select issue(s) and/or opportunity(ies) and set goals
Step 3: Plan an approach and methods
How should data be collected?
What sources of data should be used to collect information?
How long will the data be collected (the scope of data collection)?
Step 4: Collect data
Step 5: Analyze and interpret data
Step 6: Act on result
 The Employment Equity Act (the Act) applies to federally regulated employers, like banks, transportation and communication companies with 100 or more employees, as well as to Crown corporations and the federal public service. Employers covered by the Act are known as Legislated Employment Equity Plan (LEEP) employers.
 Statistics Canada online: www.statcan.gc.ca.
 The City of Toronto offers many publications and reports on its website relating to an array of topics by sector or topic, including the labour force. See City of Toronto, Publications and reports, online: www.toronto.ca/business_publications/publications.htm.
 The term “comparator group” is used to determine whether human rights “discrimination” in fact exists in a scenario. Comparison is made between a group claiming discrimination and another group that shares the relevant characteristics, to determine if disadvantage, denial, devaluation, oppression or marginalization has been experienced. A comparator group must share relevant characteristics with the group of interest in the area being questioned for comparison to be meaningful. Who the appropriate comparator group is will depend on the context and is often contested between litigants. Often the comparator group is a more privileged group in society, often the dominant group.
 S. Wortley, The Collection of Race-Based Statistics Within the Criminal Justice and Educational Systems: A Report for the Ontario Human Rights Commission (Centre of Criminology, University of Toronto) [unpublished], online: www.ohrc.on.ca.
 Data collection based on certain grounds, such as ethnic origin, sex and disability, has been done for many years under federal employment equity legislation, the national census that takes place every five years or in accordance with international requirements. In comparison, data collection on other grounds, such as sexual orientation, has not been done much in the past. Notably, the national Census does not include a question about sexual orientation, although sexual orientation has been included on other non-mandatory surveys and has been the subject of testing. Statistics Canada, Ministry of Industry “2006 Census Content Consultation Report, Catalogue No. 92-130-XE (2003, Revised in February 2004).
 C. Agocs, “Surfacing Racism in the Workplace: Qualitative and Quantitative Evidence of Systemic Discrimination” (2004) 3:3 Canadian Diversity at 26, online: www.ohrc.on.ca. For more information about Statistics Canada’s “custom services” see Statistics Canada, supra note 27.
 S. Wortley, The Collection of Race-Based Statistics Within the Criminal Justice and Educational Systems: A Report for the Ontario Human Rights Commission (Centre of Criminology, University of Toronto) [unpublished], online: www.ohrc.on.ca at 6.
 See Ontario Human Rights Commission’s Policy and Guidelines on Racism and Racial Discrimination (2005), online: www.ohrc.on.ca at 9-10.
 Social Science is defined as the scientific study of human society and social relationships. The Concise Oxford Dictionary Ninth Edition, s.v. “social science.”
 J.-C. Icart, M. Labelle, R. Antonius, Indicators for Evaluating Municipal Policies Aimed at Fighting Racism and Discrimination, Report presented to UNESCO, Fight against Discrimination and Racism Section, Division of Human Rights and Fight against Discrimination Sector for Social and Human Sciences (MontréaI, Québec: International Observatory of Racism and Discrimination: Centre for Research on Immigration, Ethnicity and Citizenship (CRIEC), Université du Québec à Montréal, 2005) at 47, online: CRIEC www.criec.uqam.ca/pdf/CRIEC%20Cahier%2028%20(en).pdf.