Chapter 4.5: Relations with survey respondents
Although there is no doubt that collecting data is in the public interest, it is still both a burden on respondents and an intrusion into their private lives. With this in mind, national statistical agencies have a moral, ethical and professional obligation to act respectfully and sympathetically toward respondents while protecting the confidentiality of the information collected.
Data collection must be carried out in an effective, exemplary and timely manner. It should target optimal respondent participation (quantity and quality) to get the most comprehensive survey results. Data collection activities must employ user-friendly survey tools, as well as concepts and terminology that respondents understand well. These tools must be tested in advance on a representative sample of respondents (individuals, households or businesses) to ensure that these criteria are respected.
As part of its respondent relations management program, Statistics Canada has established the Directive on Informing Survey Respondents. This document clearly defines the roles and responsibilities of divisions that collect information directly from respondents with respect to the management of data collection and relations with respondents.
Participation is voluntary for all social surveys, except for the Census of Population, the Labour Force Survey and the Canadian National Health Survey, given their major role for Canadians and the scope of their results. Consequently, it is important that respondents be encouraged to participate. Analyses are currently under way to determine whether participation in other social surveys should also be mandatory.
In the case of surveys of businesses and agricultural operations, including the Census of Agriculture, participation is generally mandatory (only certain special business surveys are voluntary). This is because most business and agriculture surveys are necessary to establish the national accounts, or because they are directly or indirectly related to legislated programs.
In its respondent relations management program, Statistics Canada emphasizes the following strategic pillars:
- Promoting the agency's positive image and credibility – These elements are trademark of Statistics Canada, which help it to conduct successful surveys.
- Protecting the confidentiality of respondent information – This requirement is extremely important to Statistics Canada. The Statistics Act guarantees Statistics Canada's commitment to protect the confidentiality of information obtained from Canadians. This policy is anchored in the agency's organizational culture (Chapter 4.6 of this report deals exclusively with privacy and confidentiality).
- Working continuously to reduce the response burden as much as possible – This is done by rigorously and systematically managing the response burden, employing user-friendly collection tools and an increasingly integrated survey method, and ensuring that personnel responsible for data collection are properly trained, in terms of both technical aspects and awareness of the organizational culture.
- Encouraging respondents to participate in surveys – This is done by showing respondents how important and useful survey results are to individuals and communities.
- Paying special attention to respondents' agreement to participate in surveys. This helps to maintain acceptable response rates, particularly with voluntary surveys.
Strategies, mechanisms and tools
Strategies, mechanisms and tools used to interact with respondents can be divided into three categories:
- strategies and tools common to households and businesses
- strategies and tools specific to households
- strategies and tools specific to businesses
1. Strategies and tools common to households and businesses
Statistics Canada has developed a Strategic Communications Plan for all respondents, with the aim of creating and using communication products that help to promote the importance of its surveys and encourage respondent participation.
Not all Statistics Canada surveys face the same issues with respect to collecting respondent data. To target communication needs effectively, Statistics Canada has developed a strategic approach: the pyramid of communication requirements with survey respondents (Figure 4.5.1). This approach makes it possible to divide surveys into two groups: response burden and risk of non-response. The pyramid below depicts the three levels of communication:
Figure 4.5.1: Statistics Canada's pyramid of communication requirements with survey respondents
Level I represents the basic level of communication needs. It includes communication activities and products that are systematically offered with all surveys:
- Standard letters or emails – This refers to correspondence such as letters or emails of invitation, reminder or refusal. Design and content templates are revised regularly to make them as user-friendly and motivating as possible. Their content highlights the utility of the data, the importance of confidentiality and information about collection.
The Editorial Board of the Communications Division writes, revises and approves all letters and emails that are addressed to survey respondents. In this way, the board applies the elements of the Directive on Informing Survey Respondents, uses the most recent templates and procedures, and maintains uniform language quality.
The content of invitation letters is regularly subject to qualitative testing and testing of written-communication persuasion techniques to determine factors that could influence survey participants to respond more often.
- Information for Survey Participants (ISP) web module – This module is accessible through the Statistics Canada website, and provides information about its data collection and surveys, particularly regarding collection periods, collection methods, voluntary or mandatory participation, confidentiality, sharing and linkage agreements, and frequently asked questions.
- Interviewer training and reference material – This training enables interviewers to use data collection methods that are consistent, efficient and of high quality. The training is designed by and offered by Collection and Regional Services Branch staff (for more information, see Chapter 3.4: Data collection planning and management). Training includes a national module that applies to all surveys, as well as a module on sharing best practices. The Branch also offers training specific to certain surveys, which includes detailed information about collection methods and concepts specific to these surveys.
Level II of the pyramid applies to surveys that have a heightened risk of response burden (e.g. length of interview, number of questionnaires, number of respondents in household) or a heightened risk of non-response related to particular situations (e.g. importance of the survey in the national statistical system, survey topic, age group). Consequently, these surveys require additional effort in terms of communicating with respondents. Level II communication activities that may be offered along with Level I communication activities include the following:
- Newsletters, brochures and infographics sent with letters of invitation – These communication products have audience appeal and provide relevant information about the surveys, including objectives, examples of how the data is used, information about confidentiality and data sharing (if applicable), and a link to the Information for Survey Participants module on the Statistics Canada website (primarily for household surveys).
- Social media campaigns and messages on Twitter and Facebook – Did you know? Bulletins offer the highlights of recently published survey results.
- A series of media articles – In general, these articles are aimed at groups of small or medium-sized businesses and focus on data that are of interest to these groups.
- Blog posts – These blogs, managed by federal departments or business associations, publish information to generate respondent interest and promote the benefits of taking part in Statistics Canada surveys.
Level III pertains to surveys deemed to be at high-risk because of their high visibility or sensitive nature. In addition to Level I and II communication activities, Level III may also include the following:
- A communication strategy that provides a deeper analysis of the communication needs of the survey. It will contain, among other things, an examination of the public environment, measurable communication objectives, a communication approach adapted to the audience, and targeted communication activities to increase the response rate. The Respondent Communications Team conducts the activities and follow-up in collaboration with the partners involved.
- Videos published on the Statistics Canada website and on the Statistics Canada YouTube channel – These videos aim to demystify certain surveys or to address topics of interest to respondents. Some examples include
- We Are Statistics Canada
- Statistics Canada Surveys—Your Participation is Important
- Statistics Canada Business Surveys
- An Overview of Canada's Consumer Price Index (CPI)
- What is Gross Domestic Product (GDP)?
- Talking Business—Manufacturing
- Canadian Health Measures Survey
- Talking Business—Getting to know your market and industry
- Fundamental Principles of Official Statistics
- Articles in regional and local media – Statistics Canada can identify cities in regions where data collection is a challenge and publish articles in regional and local media describing the surveys underway in the area. This helps to share relevant information about the survey with a larger audience. Media coverage heightens the visibility of surveys, which leads to higher household survey participation.
- Collaboration with external partners – Statistics Canada sends letters and relevant information (e.g., brochures, infographics, tweets) about a survey to municipal governments, targeted associations and, in some cities, police departments. The agency asks them to use their communication networks to promote the survey to their members and communities. The information provided also enables these groups and organizations to answer questions from the public and to confirm the validity of Statistics Canada surveys.
In terms of collection tools, the Integrated Collection and Operation System (ICOS) enables more intensive use of dynamic data-collection management, an approach through which response rates by geographic area or respondent profile are monitored in real time. This tool makes it possible to target collection efforts and resources continuously, obtaining response rates that optimize the quality of estimates (for more information, see Chapter 3.4: Data collection planning and management).
2. Strategies and tools specific to households
Since household surveys are primarily voluntary, the collection method must be more accommodating to respondents. Interviewers must be patient and persuasive, demonstrating interpersonal skills, both in person and over the phone. Numerous follow-ups must be carried out with diplomacy. The most sensitive or difficult cases should be handled by specialized or senior employees.
To encourage respondents to participate in surveys, it is important to thank them for giving up their time. When possible, Statistics Canada provides participants with the survey results or other information that they can use for personal planning or comparing their situation with the general population.
3. Strategies and tools specific to businesses
Statistics Canada has developed a number of tools to facilitate data collection from businesses and to reduce their response burden.
The collection strategy is supported by the agency-wide governance of the survey portfolio. On the one hand, this refers to managing the sample selection process in order to prevent small businesses from participating in more than two surveys. On the other hand, it refers to implementing a "survey collision" examination process to avoid collecting the same information, often simultaneously, with different questionnaires. The Business Register (use of which is mandatory) and the integrated approach for business surveys are elements that contribute to the success of this strategy (for more information, see Chapter 3.5: Collection management and planning).
To reduce the response burden, Statistics Canada offers business survey participants the services of an ombudsman. The role of the ombudsman is to examine complaints from participants who feel they have too high a response burden or who have misgivings about their interactions with Statistics Canada staff. The services of the ombudsman are impartial and free of charge.
The Enterprise Portfolio Management Program is responsible for overseeing relations with respondents from Canadian large businesses, maintaining business structures in the Business Register and managing business survey responses. These activities are carried out by business portfolio managers with the help of specialized research analysts in the relevant sectors.
Electronic data collection is the mode favoured by Statistics Canada for the majority of business surveys because it makes it easier for respondents to participate in surveys. This is an important component of ICOS.
As part of its engagement with businesses and business associations, Statistics Canada takes additional steps to better understand the concerns of businesses and to highlight that they are also data users. This is particularly true for large businesses, where data users and survey respondents are rarely the same people. Users work in planning and marketing, while respondents tend to work in accounting or administration. One should not assume that users discuss the importance of good statistical data with respondents. Instead, one must take the necessary measures to encourage this type of communication within businesses.
Since 2007, Statistics Canada has actively contributed to the Government Paperwork Burden Reduction Initiative for Canadian small businesses. This initiative develops strategies for eliminating excessive regulation, overlap, duplication and redundancy, as well as simplifying regulatory compliance. Statistics Canada has a concrete objective to reduce the time businesses devote to participating in surveys, whether by reducing the number of surveys or questions in the surveys, limiting the period during which businesses are included in the sample, using more user-friendly data collection methods, or by replacing surveys, in whole or in part, by using more administrative data.
Key success factors
Statistics Canada strategically invests in research initiatives, be they in qualitative testing to improve the content of invitation letters, or in a more complex analysis of suitable criteria for persuasion, based on a respondent profile. These initiatives support the principles of behavioural economicsEndnote 1, by which people in general, or respondents in our particular context, do not base their decisions solely on rational arguments or principles, but, instead, also react to emotional or behavioural elements. The goal of these initiatives is to pinpoint the criteria that influence participants to answer and, in particular, the specific point in the survey process when they make the decision to participate. The challenge with these research initiatives is to determine to what extent these strategies, also known as nudging,Endnote 2 enable the agency to influence a person to take part in a survey, and thereby maintain or improve response rates.
Statistics Canada uses many new media as communication tools to complement more conventional tools. The purpose of using these new communication tools is to foster more interactive exchange with the public; these tools use clear and accessible language both to explain the significant role statistics play in shaping people's lives, and to provide relevant, strategic and timely information to respondents.
Statistics Canada is aware that it is vital for interviewers to receive effective training and materials. This material might include infographics or visual aids that enable interviewers to better explain Statistics Canada's mandate. The training and materials also show how important it is to participate in surveys and explain how the data collected may be used.
Challenges and looking ahead
Maintaining acceptable response rates for all surveys, particularly social surveys, is a major challenge for all statistical agencies in light of the following constraints:
- greater difficulty reaching respondents as a result of the increasing use of smartphones and avoidance strategies (call display, number blocking, etc.)
- increasing respondent concern with privacy
- decreased willingness to participate in surveys because they lack time or interest, or because they are increasingly solicited by private telemarketing firms
With the ultimate goal of maintaining or improving response rates, and thereby preserving the quality of its survey results, Statistics Canada has committed to continue investing in
- better reminder and refusal strategies;
- more online questionnaires for data collection, when appropriate for the survey;
- optimal respondent cooperation by using persuasive, relevant and interactive communication strategies;
- tests, techniques and research initiatives that better identify levers of persuasion and more specific criteria for persuading respondents to participate in surveys; and
- more systematic and strategic use of administrative data to support the survey process and reduce the response burden.
- Endnote 1
Return to endnote 1 referrer
- Endnote 2
Thaler and Sunstein, 2009.
Return to endnote 2 referrer
Government of Canada (2005). Statistics Act, R.S.C. 1985, c. S-19 amended by 1988, c. 65, s. 146 1990, c. 45, s. 54 1992, c. 1, ss. 130, 131. 2005, c. 31, 2005, c. 38. Retrieved from http://laws-lois.justice.gc.ca/eng/acts/S-19/FullText.html.
Kahneman, Daniel (2013). Thinking Fast and Slow, Presentation on Slideshare. Consulted on the 11 of March 2016 and retrieved from http://www.slideshare.net/steddyss/writing-to-persuade-3604130.
Sunstein, Cass R.; Thaler, Richard E (2009). Improving Decisions About Health, Wealth, And Happiness, BWD Creative (Australia). Consulted on the 11 of March 2016 and retrieved from http://bwdcreative.com.au/thoughts/nudge-people-act-using-behavioural-economics/.
Statistics Canada (2011). Directive on Informing Survey Respondents, Ottawa. Internal document. Accessible on demand.
- Date modified: