2021-2022 Canadian Attitudes Toward the Sharing of Tax Information

Final Report

Prepared for the Canada Revenue Agency

Supplier Name: Phoenix SPI
Contract Number: 46637-220662/001/CY
Award Date: 2021-11-25
Contract Value: $133,435.21 (including applicable taxes)
Delivery Date: 2022-03-31

Registration Number: POR #046-21

For more information, please contact: cra-arc.media@cra-arc.gc.ca

Ce rapport est aussi disponible en français.

2021-2022: Canadian Attitudes Toward the Sharing of Tax Information

Final Report

Prepared for the Canada Revenue Agency
Supplier name: Phoenix Strategic Perspectives Inc.
March 2022

This public opinion research report presents the results of a 17-minute telephone survey of 2,200 Canadian residents aged 18 years of age or older from February 3rd to March 14th, 2022.

Cette publication est aussi disponible en français sous le titre : Attitudes à l'égard de l'échange de renseignements fiscaux.

Permission to Reproduce

This publication may be reproduced for non-commercial purposes only. Prior written permission must be obtained from the Canada Revenue Agency. For more information on this report, please contact the Canada Revenue Agency at: cra-arc.media@cra-arc.gc.ca.

Catalogue number:
Rv4-155/1-2022E-PDF

International Standard Book Number (ISBN):
978-0-660-43534-3

Related Publication (Registration Number: POR 046-21):
Catalogue number Rv4-155/1-2022F-PDF (Final Report, French)
ISBN: 978-0-660-43535-0

© Her Majesty the Queen in Right of Canada, as represented by the Minister of National Revenue, 2022

List of Figures

• Figure 1: Devices Used to Access the Internet
• Figure 2: Social Media Platforms Used
• Figure 3: Amount of Time Spent on Social Media
• Figure 4: Awareness of Personalized Ads and Algorithms on Social Media
• Figure 5: Perceptions of Use of Algorithms and Personalized Ads
• Figure 6: Perceptions of Use of Personal information by Organizations
• Figure 7: Views of CRA Sharing Personal Information with Other Canadian Governments
• Figure 8: Views of CRA’s Current Information Sharing Practices with Other Jurisdictions in Canada
• Figure 9: Views of CRA’s Current Information Sharing Practices with Other Jurisdictions in Canada to Locate People who Owe Money
• Figure 10: Views of CRA’s Current Information Sharing Practices with Other Jurisdictions in Canada to Determine Eligibility for Benefits
• Figure 11: Level of Support for Exchanging Personal Information for Different Reasons
• Figure 12: Perceived Impact of Information Sharing on CRA Tax Returns
• Figure 13: Ways Exchange of Information Would Impact Tax Returns of Some People
• Figure 14: Ways Exchange of Information Would Impact Personal Tax Returns
• Figure 15: Views of Information Sharing [Positive Statements]
• Figure 16: Views of Information Sharing [Negative Statements]
• Figure 17: Level of Trust in the CRA
• Figure 18: Views of CRA Sharing Personal Information with Other Canadian Governments

Executive Summary

The Canada Revenue Agency (CRA) commissioned Phoenix Strategic Perspectives (Phoenix SPI) to conduct quantitative research with Canadians to better understand their attitudes toward the sharing of tax information.

1. Research Purpose and Objectives

The sharing of taxpayer information with other federal, provincial, or territorial government organizations reduces duplication across government and ensures that services and programs offered meet the needs of citizens. To maximize its efficiency and value proposition to taxpayers and government clients, the Canada Revenue Agency wants to optimize its use of taxpayer information, including the collection of debts that are not tax debts. The purpose of this research was to collect updated information on Canadians' attitudes and perceptions towards information collection and sharing (previous research was conducted in 1999 and 2009).

The specific objectives were to understand:

• the extent to which Canadians think it is appropriate for the CRA to gather information about citizens from other government sources to ensure compliance with tax law;
• the extent to which Canadians think it is acceptable to disclose or use taxpayer information for purposes other than tax administration; and
• the extent to which the sharing of information affects compliance levels.

2. Methodology

A 17-minute telephone survey was administered to 2,200 Canadian residents aged 18 years or older from February 3rd to March 14th, 2022. Interviews with the target audience were conducted using randomly-selected telephone numbers in each province, including representation from cell-only households. Probability sampling was used; therefore, the results are generalizable to the target population. Based on a sample of this size, the results can be considered accurate to within ±2.1%, 19 times out of 20.

3. Highlights

Internet Use and Digital Profiles

Vast majority use the Internet. Most do this using both a computer and a smartphone. Slightly more than two-thirds use social media, with Facebook being the most widely used by far. Almost three-quarters use social media daily, although the amount of daily use varies considerably.

The vast majority of Canadians (91%) said they use the Internet, whether on a computer, tablet, smartphone or other device. Among those who use it, 84% access the Internet on a computer, 76% use their smartphone, and 53% use a tablet. Slightly more than two-thirds (69%) of surveyed Canadians use social media. Among these respondents, most use Facebook (83%), followed at great distance by 37% who use Instagram, while fewer still use Twitter (19%), and YouTube (13%). Among social media users, almost three-quarters (72%) use it every day. Slightly less than one-third (31%) of daily users report spending 30 minutes to an hour a day on various platforms, while one-third spend more than this each day on social media. Relatively few daily users use it for less than 30 minutes per day (8%).

Views on Algorithms and Big Data

Use of algorithms to personalize ads and tailor services widely noticed. Almost half think use of algorithms, personalized ads is a bad idea. Half believe use of personal information by organizations for service improvement, efficiencies is a bad idea.

More than half of online Canadians (57%) notice the use of algorithms to personalize ads or tailor services to individuals ‘a lot', while 23% notice this a moderate amount. Asked whether the use of algorithms in this way is a good or bad idea, almost half (48%) think it is a bad or very bad idea. Most of the rest are neutral (39%), while only 13% think this is a good idea. All surveyed Canadians (not just those online) were asked whether they believe the use of personal information by organizations to improve how they operate and to provide better customer service is a good or bad idea, responses were similar. In this case, exactly half think the use of personal information in this way is a bad idea, with most of these (31%) viewing it as a very bad idea. Relatively few (16%) think the use of information in this way is a good or very good idea.

Views on Information Sharing

Canadians divided over the CRA sharing personal information with other federal government departments and provincial governments. Almost half think this is a bad idea. Most think the CRA currently shares personal information with other departments and provincial governments. Most believe the CRA shares information to help locate people who owe money to government. Widespread belief that the CRA shares information to determine eligibility for benefits.

The questions that follow focus on the sharing of information by the CRA. Respondents were told this applies to personal information that someone may submit when they file their tax return. About one-quarter (27%) think that the CRA sharing information with other federal and provincial government departments is a good idea. Significantly more view this as a bad idea (47%), with many saying it is a very bad idea (35%); 27% are neutral. That said, just under three-quarters (73%) think the CRA currently shares personal information with other areas of the federal government, while the same proportion believe such sharing occurs with provincial/territorial governments. In both cases, relatively few (16-17%) believe such information sharing does not take place.

Three-quarters believe the CRA currently provides information to other federal departments to locate people who owe money to government. Fewer believe the CRA currently provides information to provincial/territorial governments for this purpose (68%). Eight in 10 believe the CRA provides personal information to other federal departments to determine whether people are eligible for benefits. A somewhat smaller proportion (76%) think the CRA currently shares information for this purpose with provincial/territorial governments.

Support for Information Sharing

Support for the sharing of information by CRA varies significantly depending on type of information and the reason for sharing.

Respondents were told that the CRA is permitted to release personal information only under very limited circumstances. And that the information it collects is often the same as other departments are legally allowed to collect. They were then asked about different types of information-sharing requests the CRA could receive from other federal departments or provincial governments, and to rate their level of support or opposition to each.

The degree to which Canadians support or oppose the CRA sharing information varies significantly based on the type of information being shared and the reason for sharing it. Canadians are most likely to support providing people's names and addresses so that mailing lists can be kept up to date for benefit programs – 63% support this, with 44% offering strong support. The majority also support sharing the number and age of dependents to help with the design of new programs (59%), as well as providing people's income to determine eligibility for benefits (57%). Less than half support providing a person's address to help federal and provincial governments identify people who owe them money (49%), or providing information about people's identification and income so that provinces/territories can administer their programs (45%). Canadians are least likely to support the CRA providing a person's identification and income to Canadian law enforcement agencies (40%). In this case, an almost equal number oppose sharing such information with law enforcement agencies (38%). In other areas, opposition ranged from 18-29%.

Views on Increased Information Sharing

More than half believe increased information sharing would change how some people do their taxes. Few say it would change the way they themselves would do their taxes. Significant minority think individuals might withhold information from tax returns. No consensus on how one's own tax return would be done differently, though the focus is on trying harder to ensure it is accurate.

More than half (57%) believe increased sharing of information between the CRA and other government departments would change the way some people prepare their tax returns. Relatively few (7%) said it would change the way they do their own taxes. Those who believe increased information sharing would change the way some people prepare their tax returns were asked to identify how this might change. The largest proportion (40%) think that some might try to hide or withhold certain information. Identified less often, 25% think that this would encourage people to complete their tax returns correctly and prevent them from withholding information. All other potential changes were identified by small numbers (3% or less). In terms of doing their own taxes, one-third of those who think increased information sharing could have an impact said they don't know how this might impact their tax returns. Fifteen percent would have a professional prepare their return, 12% would try to ensure they completed the return correctly, and 9% say they would be more cautious.

Forty-five percent believe increased information sharing would lead to some people reporting less information on their income tax returns, while 44% think it would lead some people to report inaccurate information. Additionally, 44% think that if CRA exchanges more information with other government agencies, it will be going too far (29% disagreed). Canadians are considerably less likely to agree in the context of increased information sharing by the CRA that it would be okay for some people to withhold information (35%), to not file their tax returns (17%), or to report lower income on their tax returns (15%) if they knew the information was going to be shared with other government departments and agencies. For two of these issues, strong disagreement was expressed by half or more – that it would be okay for some to not file their taxes (50%), or to report lower income (54%).

Most trust CRA to keep their information confidential, but there is less confidence in other government departments.

When asked to rate their level of agreement with statements about information sharing between the CRA and others, Canadians were most likely to say they trust the CRA to keep their information confidential and prevent it from getting into the wrong hands (61%, with 40% strongly agreeing). Approximately half (51%) said they trust other federal departments to keep their information confidential, and 50% think that a greater exchange of information would make the government more efficient and effective. Disagreement ranged from 23% to 31% and was highest for trust that other federal departments would keep their information confidential.

Widespread trust in the CRA in handling of personal information.

Presented with four statements aimed at gauging public trust in the CRA, the vast majority (83%) expect the CRA to have the most secure methods to store tax information. Approximately two-thirds (68%) said they trust the CRA 1) to hold their personal information using the most secure methods, 2) to ensure the transmission of information to other government departments is secure, and 3) to use the most secure methods of transmitting tax information. Disagreement with these items ranged from 8-17%.

Support for the CRA sharing personal information increased significantly by end of the survey.

At the end of the survey, respondents were asked a second time whether they think the CRA providing information about individuals to other federal and provincial government agencies is a good or bad idea. The first time this was asked, 27% said they think this is a good idea. When asked this question at the end of the survey, 41% think this is a good idea – an increase of 14 percentage points. In addition, the proportion who believe this is a very bad idea decreased significantly, from 35% to 21%. Overall, 30% felt it was a bad idea compared to 47% who expressed this view the first time it was asked, a decrease of 17 percentage points.

The multivariate analyses reinforced this finding—that respondents' support for information sharing increased after being exposed to multiple issues related to the sharing of information by the CRA. This seems to suggest that transparency about information sharing, such as the reason for doing so and the benefit to Canadians, might lead to increased support among Canadians for the sharing of information by the CRA.

Concluding observations

Many Canadians hold fairly negative views about the CRA sharing personal information with other federal departments/agencies and provincial/territorial governments. At the same time, however, Canadians believe – in relatively large numbers – that this is already taking place. There is clear evidence that once people are exposed to issues related to the sharing of personal information by the CRA, including the type of information being shared and the reasons for which it is being shared, they become more supportive of it. This is apparent in both the current survey and the one conducted in 1999.

In short, there is general unease related to the use of personal information and online tracking of views and behaviours. This is likely related to the sense of ‘Big Brother' watching and knowing what individuals are doing. However, such perceptions do not appear to diminish the use of the Internet or provided services.

Overall, there is widespread trust in the CRA when it comes to its holding and transmission of personal information to other government agencies (and noticeably less confidence in other federal departments). Moreover, the multivariate analyses revealed that respondents' expectations that the CRA uses the most secure methods to store tax information drive their support for the CRA sharing information. This has implications for communications strategies: the CRA may want to consider being more transparent about information sharing, including the measures in place to safeguard Canadians' personal information.

It is also clear that Canadians' support for personal information sharing varies considerably depending on the type of information being shared and the purpose for sharing it. People appear to be most reluctant when information is shared with police or for law enforcement reasons, and most accepting of it when the type of information being shared is limited and the reason is for benign administrative purposes and things like designing new programs or determining benefit eligibility. This was supported by the multivariate analyses which found that respondents are more accepting of the CRA sharing personal information with other federal departments/agencies and provincial/territorial governments when it serves specific administrative purposes (e.g., collecting money or verifying ID). In contrast, the more abstract tasks of planning and program design had no significant effect on views of information sharing.

There is also a general belief that increased information sharing could help make government more effective and efficient. Again, this was reinforced by the multivariate analyses which found that respondents are more accepting of the CRA sharing personal information with other areas of the federal when it leads to improvements in program effectiveness and efficiency. There is some evidence that Canadians do not draw a distinction between types/levels of government when it comes to the CRA sharing personal information. And there is a cautionary note that if the CRA shares more personal information, it would be going too far in its mandate. What this means in practice is difficult to determine because most Canadians think a considerable amount of information sharing by the CRA is already taking place.

In terms of Canadians' tax filing behaviours, there is a general belief that increased information sharing by the CRA could have an impact, but few think it would change their own tax filing practices. And those who think it might affect their own behaviour point to things like taking greater care in filing their taxes and trying to make sure their returns are accurate. These perceptions are coupled with strong views held by Canadians that the sharing of personal information by the CRA does not provide a valid reason for reporting less income by Canadians, not filing their taxes at all, or even withholding some information.

It is noteworthy that demographic subgroup differences, while sometimes evident in the data, are not pronounced nor consistent in terms of patterns. This is particularly true with respect to the more important issues explored in the survey – those related to the sharing of personal information by the CRA. The multivariate analyses reinforced this finding. Education, age, and gender did not have a significant impact on respondents' support or opposition to the CRA sharing information with other federal and provincial government agencies. Household income had only a limited impact on acceptance of the CRA sharing.

Looking specifically at the perceptions of urban versus rural residents, a stated objective of this study, there are no significant differences between them on any issues related to information sharing (only with respect to Internet use patterns and perceptions of the use of algorithms). This was supported by the multivariate analyses. Location, like household income, had only a limited impact on acceptance of the CRA sharing: urban residents are less likely than rural residents to favour information sharing with other governments, but only at a marginally significant level.

4. Intended Use of the Results

These results will be used by the CRA to inform decision-making in the design and implementation of the Agency's Digital Strategy initiatives, to identify the level of comfort that Canadians have in inter-governmental information sharing, and to serve as a baseline measure of Canadians' attitudes and perceptions which can be tracked over time.

5. Contract Value

The contract value was $133,435.21 (including applicable taxes).

6. Statement of Political Neutrality

I hereby certify as a Senior Officer of Phoenix Strategic Perspectives that the deliverables fully comply with the Government of Canada political neutrality requirements outlined in the Communications Policy of the Government of Canada and Procedures for Planning and Contracting Public Opinion Research. Specifically, the deliverables do not contain any reference to electoral voting intentions, political party preferences, standings with the electorate, or ratings of the performance of a political party or its leader.

Alethea Woods
President
Phoenix Strategic Perspectives Inc.

Introduction

The Canada Revenue Agency (CRA) commissioned Phoenix Strategic Perspectives (Phoenix SPI) to conduct quantitative research with Canadians to better understand their attitudes toward the sharing of tax information.

Background

Both the privacy and confidentiality of taxpayer information are fundamental to Canada's tax system. Taxpayer information refers to any information either obtained by or on behalf of the CRA for purposes defined by the Income Tax Act and the Excise Tax Act. These acts provide for strict protection regarding the use and exchange of taxpayer information.

The sharing of taxpayer information with other federal, provincial, or territorial government organizations reduces duplication across government and helps ensure that services and programs offered meet the needs of citizens. In order to maximize its efficiency and value proposition to taxpayers and government clients, the Agency would like to optimize its use of taxpayer information, including for CRA's collection of debts that are not tax debts, for instance collection of unpaid Canada student loans or the collection of overpayments of Employment Insurance.

Existing research in this area is outdated and may not reflect the current views of Canadians. Understanding Canadians' views on these issues is key to ensuring that the Agency takes appropriate steps to mitigate any impact on its operations and trust as a result of information sharing/disclosure activities. Public opinion research will help the CRA determine how best to move forward, while maintaining public confidence in the CRA.

Purpose and research objectives

This project directly supports the Minister of National Revenue Mandate Letter. As a longitudinal research project, for which previous iterations were undertaken in 1998/99 and 2008/09, the 2022 research will provide an update on Canadians' attitudes and perceptions towards privacy and security and, any trends in views towards information sharing.

The specific objectives of this research are to better understand:

• The extent to which people think it is appropriate for the CRA to gather information about citizens from other government sources to ensure compliance with tax laws;
• The extent to which people think it is acceptable to disclose or use taxpayer information for purposes other than tax administration, such as providing citizens with more efficient or reliable services, and ensuring citizens meet their responsibilities in compliance with the relevant laws and regulations;
• The nature of regional and urban-rural/smaller centers differences throughout Canada, and age differences; and
• The extent to which the sharing of information might affect compliance levels.

Results will be used to: 1) inform decision-making in design and implementation of the Agency Digital Strategy initiatives; 2) identify the level of comfort that Canadians have in inter-governmental information sharing to make government and CRA more efficient and effective in delivering services; and 3) serve as a baseline measure of Canadians' attitudes and perceptions which can be tracked over time.

Methodology

A telephone survey was administered to 2,200 Canadian residents aged 18 years or older from February 3rd to March 14th, 2022. Interviews with the target audience were conducted using randomly selected telephone numbers in each province, including representation from cell-only households. Probability sampling was used; therefore, the results are generalizable to the target population. Based on a sample of this size, the results can be considered accurate to within ±2.1%, 19 times out of 20. The margins of error are greater for results pertaining to subgroups of the total sample and split samples. For a full description of the methodology, refer to the Appendix: 1. Technical Specifications.

Notes to readers

• All results are expressed as percentages, unless otherwise noted. Throughout the report, percentages may not always add to 100 due to rounding and/or multiple responses being offered by respondents.
• At times, the number of respondents changes in the report because questions were asked of sub-samples of the survey population. Accordingly, readers should be aware of this and exercise caution when interpreting results based on smaller numbers of respondents.
• Where base sizes are reported in graphs, they reflect the actual number of respondents who were asked the question.
• Subgroup differences are evident in the data but are generally not pronounced nor consistent in terms of patterns, particularly with respect to the more important issues explored in the survey – those related to the sharing of personal information by the CRA. As one example, there are no significant differences between urban and rural residents on any issues related to information sharing (only with respect to Internet use patterns and perceptions on the use of algorithms). These differences are presented in the report when they are statistically significant.
  • When reporting subgroup variations, only differences that are significant at the 95% confidence level and that pertain to a subgroup sample size of more than n=30 are discussed in the report, or that are part of a pattern or trend. If one or more categories in a subgroup are not mentioned in a discussion of subgroup differences (for example, if two out of six regions are compared), it can be assumed that significant differences were found only among the categories reported.
• Tracking data is included in the analysis. Specifically, the results from the 1999 survey are referenced. However, in many instances the wording of the questions or preceding preambles differs, which impedes direct comparison. In addition, the CRA did not want the results of the 2009 survey to be included in the tracking of perceptions over time for methodological reasons.
• The survey questionnaire is appended: 2. Survey Questionnaire.

Detailed Findings

1. Internet Use and Digital Profiles

Vast majority use the Internet; most do so using both a computer and a smartphone

Asked whether or not they use the Internet, whether on a computer, tablet, smartphone or other device, the vast majority of Canadians said they do (91%). Among those who use the Internet (n=2,006), 84% access the Internet on a computer, followed by approximately three-quarters (76%) who use their smartphone. Slightly more than half (53%) use a tablet to access the Internet, while few use other devices, such as Smart TVs (3%) or gaming consoles (2%).

Q2. Which of the following devices do you use to access the Internet? [Multiple responses accepted]. Base: n=2,006; respondents who use the internet. [DK/NR: <0.5%].

Not only are the vast majority of Canadians online, but most are also using multiple devices to access the Internet. A clear majority (58%) use both a smartphone and computer. As well, 25% of surveyed Canadians use one device, 26% use two, and 37% use three devices to access the Internet (3% use four devices).

Although equally likely to use the Internet, respondents living in urban settings are more likely to access the Internet via a computer (85% versus 79% of those in rural locations), smartphone (77% versus 69%), or tablet (55% versus 48%).

Individuals aged 65 and older are less likely to use the Internet – 84% do so. In contrast, nearly all Canadians aged 18 to 54 access the Internet (97%), as do 94% of those aged 55-64.

Daily social media users are more likely to use a smartphone to access the Internet (86%), compared to 77% of non-daily users. Additionally, Canadians aged 18 to 34 (94%), 35-54 (92%), and those 55-64 (78%) are also more likely to use a smartphone, compared to about half (52%) of respondents 65 and older.

Canadians in British Columbia, Alberta, and Ontario (86% each) are more likely to use a computer to access the Internet compared to those in Quebec or Atlantic Canada (79% each).

Slightly more than two-thirds use social media; Facebook is most widely used by far

More than two-thirds (69%) of surveyed Canadians use social media. Among these respondents, most have a Facebook account (83%). This is followed at great distance by approximately one-third (37%) who use Instagram, while fewer still use Twitter (19%), and YouTube (13%). Other social media platforms used are reported by fewer than 10%.

Q4. Which social media do you personally use? [Multiple responses accepted]. Base: n=1,378; Social media users. [DK/NR: 2%].

Rural residents are more likely to use Facebook compared to those living in urban settings (88% versus 82% respectively). Urban dwellers are more likely to use Instagram (39%), Twitter (21%), and YouTube (14%).

Canadians earning household incomes exceeding $100,000 are more likely to be social media users. More specifically, 74% of those with household incomes of $100,000 to just below $150,000, and 77% of those with household incomes of $150,000 and above.

The following subgroups are more likely to use Facebook:

• Canadians in Quebec (91%) and Atlantic Canada (88%);
• Those who are 55-64 (86%) and 65 and older (92%);
• Women (86%); and
• Trade school and college graduates and those with a university diploma below a bachelor's degree (86%).

Almost three-quarters use social media daily; amount of daily use varies considerably

Among social media users (n=1,378), almost three-quarters (72%) use social media every day. Slightly less than one-third (31%) of daily users reported spending 30 minutes to an hour a day on various platforms, while exactly one-third spend more than this each day on social media. Relatively few daily social media users use it for less than 30 minutes per day (8%).

Turning to non-daily social media users, 11% spend about an hour per week on their accounts (11%), while fewer spend between 2 to more than 10 hours a week.

Q6. How much time do you spend on social media in a typical day? Base: n=1,370; social media users; [DK/NR= <0.5% - 3%].

Social media users aged 35-54 and 55-64 are more likely to report spending between 30 minutes to an hour on social media per day (38% and 33% respectively). Moreover, men are also more likely to report spending this amount of time on social media (35% versus 29% of women).

2. Views on Algorithms and Big Data

Use of algorithms to personalize ads and tailor services is widely noticed

Ahead of being asked their level of awareness of the use of personalized ads and algorithms on social media, respondents were presented with the following:

"As part of their internet strategy, organizations are making increasing use of algorithms to tailor personalized ads or services to specific users. The algorithms take into account the things you look at, interact with, and ‘like' on the Internet, and then suggest other things you might like."

With this description in mind, more than half of online Canadians notice this a lot (57%), while more than one in five notice it a moderate amount (23%). At the other end of the spectrum, 11% said they don't notice this very much, and 8% do not notice this at all. Notably, virtually no respondents volunteered that they have never heard of the use of algorithms in this way (<0.5%).

Q7. When using the Internet for social media or other activities, would you say you notice this a lot, a moderate amount, not very much, or not at all? Base: n=1,981; Social media users. [DK/NR: 1%].

The following subgroups are more likely to report noticing the use of personalized ads a lot or a moderate amount:

• Canadians in Alberta (85%) compared to those in Ontario (78%) and the Prairies (75%);
• Those under the age of 65: 90% of 18-34 year olds, 88% of 35-54 year olds, and 82% of 55-64 year olds versus 67% of those aged 65 and older;
• Women (82% compared to 78% of men);
• Trade school and college graduates or those with a university diploma below a bachelor's degree, as well as those with a bachelor's degree or higher (82% and 81%, respectively, versus 75% of those with less education); and
• Daily social media users (90% versus 75% of non-daily users).

Almost half think use of algorithms to personalize ads, tailor service is a bad idea

Asked whether the use of personalized ads and algorithms is a good or a bad idea, almost half of surveyed Canadians (47%) believe it is a bad or very bad idea. Most of the rest are neutral (39%). In contrast, significantly fewer think this is a good idea (13%).

Q8. Do you think the use of algorithms in this way is a good or bad idea? Base: n=1,929; Social media users. [DK/NR: 4%].

Canadians living in urban centres are more likely to think that the use of algorithms to personalize ads and tailor services is a good idea – 14% compared to 10% of those in rural settings. So too, were those who view CRA information sharing with provincial and federal departments as a good thing (19% versus 10% of those who think it is a bad idea).

The following subgroups are more likely to view the use of algorithms as a bad idea:

• Canadians in British Columbia (55%), Alberta (55%), the Prairies (51%), and Ontario (47%);
• English speakers (52%) as compared to French speakers (40%);
• Those who do not use social media (55% compared to 45% of users);
  • Non-daily users (50%) as compared to daily users (43%); and
• Those who have completed trade school/college/or who have a university diploma (50%).

Moreover, respondents under the age of 35 are less likely to think the use of algorithms is a bad idea (34%).

Half believe organizations' use of personal information for service improvement and efficiencies is a bad idea

The following questions were asked of all respondents, not just online Canadians. Prior to the next question, respondents were read the following preamble:

"Many organizations are making greater use of personal information they have about us, like age, income, and our use of websites. They use this information for different purposes, including to improve how they operate and to provide better customer service."

Asked whether they believe the use of personal information by organizations in this way is a good or a bad idea, responses were similar to those offered regarding the use of algorithms and personalized ads (discussed above). In this case, exactly half think the use of personal information in this way is a bad idea, with most of these (31%) viewing it as a very bad idea. One-third were neutral (34%). Relatively few (16%) think the use of information by organizations in this way is a good or very good idea.

Q9. Many organizations are making greater use of personal information: In general, do you think this is a good thing or a bad thing? Base: n=2,157; all respondents. [DK/NR: 2%].

Those living in urban centres are more likely to think the increased use of personal information by organizations is a good or very good idea (17% versus 12% of those in rural settings). Those under 35 are also more likely to view this as a good thing: 22% compared to 15% of Canadians aged 35 and older.

Internet users are significantly more likely to view the use of information in this way as a bad idea (52%), compared to just 29% of non-Internet users. Moreover, Canadians in Atlantic Canada (54%) are more likely to judge this use as a bad idea compared to those who reside in Ontario (47%).

3. Views on Information Sharing

The questions that follow focus on the sharing of personal information by the CRA with other federal departments and agencies or provincial/territorial governments. Before answering, respondents were told that this applied to personal information that someone may submit to Canada Revenue Agency when they file their tax return.

Canadians divided over the CRA sharing personal information with other departments and provincial governments; almost half think this is a bad idea

About one-quarter of surveyed Canadians (27%) think the disclosure of information about individuals by the CRA to other federal and provincial/territorial government departments is a good idea (13% say it is a very good idea). In contrast, significantly more view this sharing of information as a bad idea (47%), with most of whom saying it is a very bad idea (35%). About one-quarter (27%) are neutral on the subject, saying it is neither a good nor a bad idea.

Q10. Do you think the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good idea or a bad idea? Base: n=2,149; all respondents. [DK/NR: 2%].

Canadians living in Quebec are significantly more likely than those in every other region to say the sharing of personal information by the CRA is a good idea (38%). Among them, 21% say this is a very good idea.

Conversely, Canadians under the age of 35 are less likely to think this form of information sharing is a bad idea. More specifically, while just 37% of those aged 18 to 34 believe this is a bad idea, approximately half of those 35-54 (46%), 55-64 (52%), and 65 or older (49%) believe this is a bad idea to some extent (scores of 1 or 2).

Most think the CRA currently shares personal information with other departments and provincial governments

Just under three-quarters of Canadians (73%) think the CRA currently shares information about individuals with other areas of the federal government. Moreover, the same proportion believes this exchange of information also occurs between the CRA and provincial or territorial governments. In both cases, relatively few (16-17%) believe this type of information sharing does not take place.

Clearly, Canadians are not drawing a distinction between types/levels of government when it comes to the CRA sharing personal information about individuals. They assume other federal departments and agencies and provincial/territorial governments are being treated the same by the CRA when it comes to sharing information about individual Canadians.

Q11. Do you believe the CRA and other areas of the Canadian federal government currently exchange information like this? Base: n=2,200; all respondents. [NR: <0.5%].
Q12. Do you believe the CRA and provincial/territorial governments currently exchange information like this? Base: n=2,200; all respondents.

Respondents in Quebec (67%) are less likely to think the CRA shares personal information with other departments or provincial and territorial governments compared to those in other regions of Canada.

Individuals aged 65 and older are significantly less likely to believe the CRA shares personal information with other federal departments compared to younger Canadians – specifically, 66% of those 65 and older versus around three-quarters of those 18-34 (77%), 35-54 (77%), and 55-64 (79%). So too are Canadians with less education: 67% of those who have completed high school or less compared to 76% of college/trade graduates and those with a university diploma and 74% of university graduates.

Most believe the CRA shares information to help locate people who owe money to government

Three-quarters of Canadians believe the CRA currently provides information to other areas of the federal government to locate people who owe money to the government. Conversely, 13% believe the CRA does not currently do this, while 11% don't know if this is the case.

Fewer Canadians believe that the CRA currently provides information to provincial or territorial governments for this purpose (68%). For this specific purpose – helping to locate people who owe money to the government – it is noteworthy that there is greater differentiation with respect to the sharing of information by the CRA with other levels of government than there was for the more general issue of sharing personal information about individuals (without a purpose stated).

Note: A split sample was used to determine if Canadians believe the CRA currently provides information for this purpose to 1) other areas of the federal government, and 2) provincial/ territorial governments.

Q13. Do you believe the CRA is currently providing information to other areas of the federal government to locate people who owe money to the government? Base: n=880; SPLIT SAMPLE.
Q13. Do you believe the CRA is currently providing information to provincial/territorial governments to locate people who owe money to the government? Base: n=877; SPLIT SAMPLE.

Social media users are more likely to believe the CRA shares this type of information with other areas of the federal government in order to help locate individuals who owe money (78% versus 68% of non-social media users).

Moreover, those with a high school diploma or less (78%), and those who have completed trade school/college/or hold a university diploma (81%) are more likely to believe this information is shared with federal departments compared to 68% of those with a university degree who said the same. These same respondents are also more likely to believe this sharing of information also occurs between the CRA and provincial and territorial governments.

Widespread belief that the CRA shares information to determine eligibility for benefits

Eight in 10 respondents believe the CRA provides personal information to other areas of the federal government to determine whether people are eligible for benefits. Ten percent believe information is not currently shared for this purpose, while an additional 10% did not provide a response.

A somewhat smaller proportion of Canadians (76%) think the CRA currently shares information for this purpose with provincial or territorial governments, with 13% thinking this is not the case.

Note: A split sample was again used to determine if Canadians believe the CRA currently provides information for this purpose to 1) other areas of the federal government versus 2) provincial/territorial governments.

Q14A. Do you believe the CRA is currently providing information to other areas of the federal government to determine whether people are eligible for benefits? Base: n=864; SPLIT SAMPLE.
Q14B. Do you believe the CRA is currently providing information to provincial/territorial governments to determine whether people are eligible for benefits? Base: n=893 SPLIT SAMPLE.

Canadians in the Prairies (89%) are more likely to believe that the CRA shares information with other areas of the federal government to determine eligibility for benefits compared to those in Ontario (79%) or Alberta (74%). Along with respondents in the Prairies, those aged 35-54 (86%) are more likely to say the same compared with 78% of those aged 55-64, and 77% of those 65 and older.

Comparing results among Canadians who were asked about sharing with provincial and territorial governments, those aged 18-34 are more likely to believe that the CRA shares information with other governments, particularly compared to those who are 35-54 (83% versus 71% respectively).

4. Support for Information Sharing

Respondents were told that under current laws, the CRA is permitted to release information about individuals only under very limited circumstances. And that, "the CRA often collects the same information as other government departments. These government departments have the legislative authority to collect this information on their own. Having CRA share this information could make governments operate more efficiently."

Support for sharing of information by CRA varies significantly depending on type of information and reason

Canadians were asked about different types of information-sharing requests the CRA could receive from other areas of the federal government or provincial governments, and to rate their level of support or opposition to the sharing of information for each, using a 5-point scale.

The degree to which Canadians support or oppose the CRA sharing information varies significantly based on the type of information being shared and the reason for sharing it. Canadians are most likely to support providing people's names and addresses so that mailing lists can be kept up to date for benefit programs – 63% support this, with 44% offering strong support. Note that this is very much a purely administrative use of the information. The majority of Canadians also support the sharing of the number and age of dependents to help with the design of new programs (59%), as well as providing people's income information to determine eligibility for benefits (57%).

Q15. Please tell me how strongly you would support or oppose exchanging information for each reason, using a 5-point scale, where ‘1' means you strongly oppose it, ‘5' means strongly support, and ‘3' means neither support or oppose. Base: n=2,200; all respondents. [DK/NR: 1 - 2%].

Somewhat less than half support providing a person's address to help other levels of government identify people who owe them money (49%) or providing information about people's identification and income level so that provinces and territories can administer their programs (45%).

Canadians are least likely to support providing a person's identification and income to Canadian law enforcement agencies (40%). In this case, an almost equal number oppose sharing such information with law enforcement agencies (38%). For other types of information and purposes, opposition ranged from 18-29%.

Canadians in Quebec are more likely to express support for all items as compared to most other regions of Canada.

Support for providing people's names and addresses in order to keep mailing lists up to date for benefit programs declined as age increased – 72% of respondents aged 18-34 expressed support, 63% of those 35-54, 63% of those 55-64, and 60% of those 65+ said the same.

Social media users (62%), along with those with a university degree (63%), are more likely to express support for providing the number and ages of dependents to help in the design of programs.

5. Views on Increased Information Sharing

More than half believe increased information sharing would change how some do their taxes; few say it would change the way they do their own taxes

More than half of surveyed Canadians (57%) believe that an increase in the exchange of information between the CRA and other government departments/agencies would change the way some people prepare their tax returns. Almost one third (31%) thought this would not be the case (11% were uncertain).

When asked if this increase in information sharing would change the way they prepare their own tax returns, relatively few Canadians (7%) said it would. Conversely, 89% said there would be no change in how they prepare their tax returns.

[LEFT] Q16. Suppose the CRA and other government departments and agencies increased the extent to which they exchange information. Do you think that some people would change the way they prepare their tax returns for CRA? Base: n=2,200 all respondents. [NR: <0.5%].
[RIGHT] Q18. And would the increased exchange of information between CRA and other government departments and agencies change the way you prepare your tax returns for CRA? Base: 2,200 all respondents. [NR: <0.5%].

Respondents in Alberta are more likely to believe that the increased exchange of information would lead some people to change the way they prepare their tax returns, with 64% thinking it would. Additionally, the likelihood of thinking that the increased exchange of information would influence how some people prepare their tax returns declined as age increased – 66% of Canadians 18-34, 61% of those 35-54, and 58% of those 55-64, compared with exactly half of those 65 and older – and it was higher among those who are daily social media users (61% versus 54% of non-daily social media users).

Asked if the degree of information exchanged between the CRA and other departments would change how they prepare their own tax returns, Canadians with a university degree are more likely to say they would not change how they prepare their taxes (94% compared to 86% of Canadians with less education).

Significant minority think individuals might withhold information from tax returns

Respondents who believe the increased exchange of information between the CRA and other departments and agencies would influence the way some people prepare their tax returns (n=1,233) were asked to identify ways in which people might change the way they do their taxes.

The largest proportion (40%) think that some people might try to hide or withhold certain information. Identified much less often, one-quarter believe that increased information sharing would encourage people to complete their tax returns correctly and prevent them from withholding information. All other responses were identified by small numbers only (3% or less) and can be seen in the graph below.

Q17. In what ways might people change the way they prepare their tax returns for the Canada Revenue Agency? [Multiple responses accepted]. Base: n=1,233; respondents who believe increased exchange of information would influence tax returns. [NR: 1%].

Notably, a significant proportion of respondents did not hazard a guess in terms of how increased information sharing might influence individuals to change how they complete their taxes (26%).

Compared to those living in all other regions of Canada, Quebec residents (30%) are less likely to believe increased information sharing would encourage individuals to hide or withhold information from their tax returns. Canadians with a university degree are more likely to believe people would withhold information from their returns (43%) compared to those with a high school diploma or less (35%).

No consensus on how one's own tax return would be done differently, though focus is on ensuring accuracy

Canadians who said the increased exchange of information between the CRA and other government agencies would influence how they complete their own tax returns (n=164) were asked to identify ways in which they might change the way they do their taxes.

There is no consensus on how tax preparation might be impacted. The single largest proportion noted they don't know how this might influence their tax returns (33%). Following this, 15% would have a professional prepare their return, 12% would try to ensure they completed the return correctly, and 9% say they would be more cautious. All of these respondents indicated, in one way or another, that they would try harder to make sure their tax return was done accurately. Conversely, 12% state they would withhold certain information from their returns. Eleven percent think this would simplify the process and reduce paperwork (11%).

Q19. In what way would you change the way you prepare your tax returns for CRA? [Multiple responses accepted]. Base: n=164; respondents who say they would change how they prepare their tax returns. [NR: 3%].

There are no subgroup differences to report due to low base numbers.

Most trust CRA to keep their information confidential, less confidence in other government departments

Surveyed Canadians were asked to rate their level of agreement with a set of statements about information sharing between the CRA and others, using a 5-point scale. These statements were framed in a positive manner.

Canadians were most likely to agree that they trust the CRA to keep their information confidential and prevent it from getting into the wrong hands. Here, 61% expressed agreement, with 40% strongly agreeing. Approximately half (51%) said they trust other federal government departments to keep their information confidential, with one-third expressing strong agreement. Notably, one in five strongly disagree that other federal departments can be trusted to keep information from getting into the wrong hands. Exactly half think that a greater exchange of information would make the government more efficient and effective.

Disagreement with these statements ranged from 23% to 31% and was highest for trusting that other federal government departments would keep their information confidential.

Q20A/D/G. I am going to read you a list of statements. For each one, please tell me whether you agree or disagree with it using a 5-point scale, where ‘1' means you strongly disagree, ‘5' means strongly agree, and ‘3' means neither agree or disagree. Base: n=2,200; all respondents. [DK/NR: 1 - 2%].

Quebec residents are more likely than those in Alberta to agree that they trust other federal government departments to keep their information confidential (55% versus 45% respectively). Moreover, Quebec residents are also more likely to strongly agree with this statement (36%), compared to those in Alberta (26%) and British Columbia (29%).

Canadians who are 65 or older are more likely to trust the CRA (68% compared to 57% of 18–54-year-olds and 58% of 55-64 year olds) and other federal government departments (58% compared to 48% of younger Canadians) to keep their information confidential.

Quebec residents are more likely to believe that a greater exchange of information would make governments more effective and efficient (56%). Along with Quebec residents, university graduates are also more likely to agree with this statement (55%), compared to college graduates (44%), and those with a high school diploma or less (49%).

Views on information sharing vary considerably depending on type of information, purpose

Surveyed Canadians continued to rate their level of agreement with a set of statements about information sharing between the CRA and others, using the same 5-point scale. These statements were framed in a negative manner.

Forty-five percent of Canadians believe an increase in the exchange of information would lead to some people reporting less information on their income tax returns, while almost the same number (44%) think it would lead some people to report inaccurate information. Approximately one-quarter expressed disagreement with these statements.

Additionally, 44% think that if CRA exchanges more information with other government agencies, it will be going too far, with 29% expressing disagreement.

Canadians are considerably less likely to agree that it would be okay for some people to withhold information (35%), to not file their tax returns (17%), or to report lower income on their tax returns (15%) if they knew the information was going to be shared with other government departments and agencies. Moreover, for two of these issues, strong disagreement was expressed by half or more of surveyed Canadians – that it would be okay for some people to not file their tax returns (50%), or to report lower income on their taxes (54%).

Q20B/C/E/F/H/I. I am going to read you a list of statements. For each one, please tell me whether you agree or disagree with it using a 5-point scale, where ‘1' means you strongly disagree, ‘5' means strongly agree, and ‘3' means neither agree or disagree. Base: n= 2,200; all respondents. [DK/NR: 2 - 4%].

Younger respondents, specifically those aged 18-34 (50%) and 35-54 (47%), are more likely to agree that an increase in the exchange of information would lead to some people reporting less information on their tax returns. This is compared to 41% of Canadians 65 and older who said the same. Those between the ages of 18 and 34 are also more likely to agree it would be okay for some people to withhold personal information from the CRA (41%), in particular compared to 33% of those aged 35-54 who said the same.

Forty-four percent of Canadians with a high school diploma or less, and 49% of college graduates agree that if the CRA begins to exchange more information with other government agencies it would be going too far, compared to 37% of university graduates.

Widespread trust in the CRA in handling of personal information

Presented with four statements aimed at gauging public trust in the CRA, the vast majority of Canadians (83%) expect the CRA to have the most secure methods to store tax information. Approximately two-thirds (68%) said they trust the CRA to hold their personal information using the most secure methods, while the same proportion trust the CRA to ensure the transmission of information to other government departments is secure (68%) and trust the CRA to use the most secure methods of transmitting tax information (68%). The last two items were asked of a split sample of respondents. Disagreement with these statements ranged from 8-17%.

Q21A/B. I am going to read you a list of statements about the CRA specifically. For each one, please tell me whether you agree or disagree with it using a 5-point scale, where ‘1' means you strongly disagree, ‘5' means strongly agree, and ‘3' means neither agree or disagree. Base: n=2,200; all respondents. [DK/NR: 1%].
Q21C. Base: n=1,101; SPLIT SAMPLE. [DK/NR: 2%].
Q21D. Base: n=1,099; SPLIT SAMPLE. [DK/NR: 1%].

Trust in the CRA to use the most secure methods to transmit tax information increased with age. Seventy four percent of Canadians 65 and older, and 71% of those aged 55-64 agree compared to 54% of those who are 18-34, and 65% of Canadians who are 35-54. Canadians under 35 (73%) were less likely to expect the CRA to have the most secure methods to store tax information (compared to 85% of Canadians aged 35 and older).

Trust in the CRA to hold personal information using the most secure methods also increased with age. Most significantly, respondents who are 65 and over (75%) were more likely to agree with this statement compared to all other age groups.

Support for the CRA sharing personal information increased significantly by end of the survey

Respondents were asked earlier in the survey whether they think the CRA providing information about individuals to other federal and provincial government agencies is a good or bad idea. At that time, 27% of surveyed Canadians said they think this is a good idea. They were then asked this same question at the end of the survey. After being exposed to multiple issues related to the sharing of information by the CRA, 41% of surveyed Canadians now think this is a good idea – an increase of 14 percentage points.

In addition, the proportion of respondents who believe this is a very bad idea decreased significantly, from 35% to 21% by the end of the survey. Overall, 30% felt it was a bad idea compared to 47% who expressed this view the first time it was asked, a decrease of 17 percentage points.

Q10. Do you think the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good idea or a bad idea? Base: n=2,149; all respondents. [DK/NR: 2%].

Q22. Do you think the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good or bad idea? Base: 2,179; all respondents. [DK/NR: 1%].

Quebec residents are more likely to agree the exchange of information between the CRA and other government departments is a good idea (51%) as compared to those in all other regions of Canada. Moreover, those with a university degree were more likely to express support (46%) as were daily social media users (46%).

6. Profile of Survey Respondents

The following tables present the characteristics of respondents (using weighted data).

7. Driver Analysis

A multivariate analysis was conducted to determine which attitudes and perceptions drive Canadians' views on the CRA providing information about individuals to other federal and provincial/territorial government departments. The results of this analysis are presented in this section of the report.

The variables

The main dependent variable is Q20 in the survey questionnaire:

Now that we have talked about the issue for some time, let me ask you a question I asked at the beginning of this interview. Do you think the idea of the CRA providing information about individuals to other federal and provincial [territorial if G=11-13] government departments is a good or bad idea? Please use a 5-point scale, where ‘1' means a very bad idea, ‘5' a very good idea, and ‘3' means neither a good nor bad idea.

The proximate independent variables include the following:

• Q9. Many organizations are making greater use of personal information: In general, do you think this is a good thing or a bad thing?
• Q10. Do you think the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good idea or a bad idea?
• Q15. Support/Oppose exchanging information for each reason:
  • Q15A. Providing people's name/address so that mailing lists can be kept up to date for benefit programs and databases such as vital statistics registries could be kept up to date.
  • Q15B. Providing people's income information so that federal and provincial/territorial government departments can determine whether people are eligible for benefits.
  • Q15C. Providing the number and ages of dependents people have to help in the design of new programs.
  • Q15D. Providing information about people's identification and income levels so that provinces/territories can administer their programs.
  • Q15E. Providing a person's address information to help federal or provincial/territorial departments identify people who owe money to them.
  • Q15F. Providing a person´s identification and income information to Canadian law enforcement agencies, such as the RCMP.
• Q20. Agree/Disagree with the following statements:
  • Q20A. A greater exchange of information would make governments more effective and efficient.
  • Q20B. An increase in the exchange of information would lead to some people reporting less information on their income tax returns.
  • Q20C. It would be okay for some people to not file their tax returns to the CRA if they knew the information was going to shared with other government departments.
  • Q20D. I trust the CRA to keep my information confidential and prevent it from getting into the wrong hands.
  • Q20E. It would be okay for some people to withhold personal information from CRA if they knew the information was going to be given to other government agencies without their consent.
  • Q20F. An increase in the exchange of information would lead to some people reporting inaccurate information.
  • Q20G. I trust other federal government departments to keep my information confidential and prevent it from getting into the wrong hands.
  • Q20H. It would be okay for some people to report lower income on their tax returns to CRA if they knew the information was going to be given to other government agencies.
  • Q20I. If the CRA begins to exchange more information with other government agencies, it will be going too far.
• Q21 Agree/Disagree with the following statements:
  • Q21A. I trust the CRA to hold on to my personal information using the most secure methods.
  • Q21B. I expect the CRA to have the most secure methods to store tax information.
  • Q21C. When it submits tax information to other federal and provincial/territorial government departments, I trust the CRA to ensure that the transmission of the information is secure. [split sample]
  • Q21D. I trust the CRA to use the most secure methods of transmitting tax information. [split sample]

The standard socio-economic and demographic variables were treated as control variables to see if they change the relationship between the proximate independent variables and the main dependent variables. The background variables used included education (highest level of formal education completed), household income, age, gender, location (urban versus rural), and employment status (employed full time versus those who are not employed full time) .

The methods

An ordinary least squares (OLS) linear regression was conducted to determine the impact of the proximate independent variables on the main dependent variable. The first regression included all the independent variables except for the split sample variables (Q21C and Q21D); the analysis was then repeated three times. Twice it was repeated with minor modifications to incorporate the split sample variables. The regression was then repeated a third time after the control variables were added to the model. All analyses were conducted with weighted data (i.e., the weight variable was activated).

The results

The initial regression performed very well, with the independent variables explaining 62.5% of the variation in the dependent variable. Recall that this analysis did not include the split sample variables:

• Q21 Agree/Disagree with the following statements:
  • Q21C. When it submits tax information to other federal and provincial/territorial government departments, I trust the CRA to ensure that the transmission of the information is secure. [split sample]
  • Q21D. I trust the CRA to use the most secure methods of transmitting tax information. [split sample]

Results are provided in the following R Square table:

In addition to the independent variables performing well overall, this is also highly significant in an inferential sense with a significance level approaching .000.

The regression coefficients for each independent variable are presented in the table below. Variables shaded orange are conventionally significant (significant .05 or less) and variables shaded blue are approaching conventional significance (significance greater than .05 but less than .1). Rows not shaded contain variables that are not conventionally significant.

The following observations about the regression coefficients are noteworthy:

• Q20A has the largest significant beta coefficient as well as a conventionally significant and positive impact on the dependent variable, indicating that those who think governments could be more efficient and effective if they exchanged information also tend to think that the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good idea.
• Q10 had the second largest beta coefficient as well as a conventionally significant and positive impact on the dependent variable. This means that respondents who thought it was a good idea for the CRA to provide information to other federal and provincial agencies at the beginning of the interview also thought this later in the interview after having responded to a series of questions that could have impacted their opinion on this topic.
• The next largest (but negative) significant beta coefficient is associated with Q20I. This indicates that those who agree with the view that the CRA sharing information with other agencies would be going too far are less likely to agree that the CRA sharing of information with other federal and provincial/territorial government departments would be a good idea.
• The next largest (positive) and significant beta coefficients are those associated with Q15E and Q15D. This indicates that those who support the CRA 1) providing a person's address to other agencies so they may pursue money owing or 2) providing information about people's identification and income levels to the provinces/territories for program administration are more likely to think that the CRA providing information to other government departments would be a good idea.

The variables with smaller impacts that are significant (or that are marginally significant) offer the following insights:

• Q20B and Q20C are noteworthy because of the (negative) significant beta coefficients associated with each variable. Those who agree with the views that an increase in the exchange of information would lead to some people reporting less information on their income tax returns or that it would be okay for some people to not file a return if they knew the information was going to be shared with other government departments are less likely to agree that the CRA sharing of information with other federal and provincial/territorial government departments would be a good idea.
• Q20F has a smaller impact on the dependent variable but is noteworthy. Those who agree with the view that an increase in the exchange of information would lead to some people reporting inaccurate information are more likely to agree that the CRA sharing of information with other federal and provincial/territorial government departments would be a good idea. While variables Q20F and Q20B are seemingly similar, there is an important distinction: Q20B is grounded in information reporting on income tax returns. The reporting of inaccurate information does not deter people from holding the view that information sharing is a good idea. In contrast, fairness around income taxes does—that is, people potentially reporting less information on their return.
• Q20G and Q21B have smaller impacts on the dependent variable but are also noteworthy. Those who trust other federal government departments to keep their information confidential or who expect the CRA to have secure methods to store tax information are more likely to agree that the CRA sharing of information with other federal and provincial/territorial government departments would be a good idea.

Following the initial regression, two additional regressions were conducted using each of the split samples:

• Q21 Agree/Disagree with the following statements:
  • Q21C. When it submits tax information to other federal and provincial/territorial government departments, I trust the CRA to ensure that the transmission of the information is secure. [split sample]
  • Q21D. I trust the CRA to use the most secure methods of transmitting tax information. [split sample]

Results are provided in the following R Square table:

The R Square does not change dramatically when either split sample variable is added. In both instances, the R Square is large and highly significant.

The coefficients for the individual independent variables also do not change in a substantive way. The table below presents the variables and indicates in which regression (if any) each was significant. Cells shaded in orange mean the effect of a variable was significant at less than .05. Cells shaded in blue indicate .05 to .1 significance for the variable's coefficient. Cell not shaded indicate that the variable was not significant.

Neither split sample variable is significant in the regression, and both regressions yield fewer coefficients that are significant or nearly significant. This is the result of the smaller sample sizes: n=903 for the regression with Q21C and n=908 for the regression with Q21D. Overall, the general direction of the findings from the regression do not change with the addition of variables Q21C and Q21D.

When the socio-economic and demographic variables are added as control variables to the initial regression (which excludes the split variable: Q21C and Q21D), results are similar:

The table below presents the variables and indicates in which regression (if any) it was significant. Cells shaded in orange mean the effect of a variable was significant at less than .05 in the regression. Cells shaded in blue indicate .05 to .1 significance for the variable's coefficient. Cell not shaded indicate that the variable was not significant.

For the most part, the initial independent variables are not greatly affected by the introduction of the socio-economic and demographic variables. When the control variables are added, the only noteworthy change in the original independent variables is that Q15F (Providing a person's identification and income information to Canadian law enforcement agencies, such as the RCMP) is no longer significant.

Looking at the control variables, household income has a marginally significant impact on one's acceptance of the CRA sharing information with other governments. Urban residents are less likely than rural residents to favour information sharing with other governments, but only at a marginally significant level. Education, age, gender, and full-time employment status, however, do not seem to have much impact on the dependent variable.

Conclusions and implications

Based on the multivariate analyses, the following conclusions and implications can be drawn:

• The relationship between Q10 and Q22 is very strong. These are essentially the same variables in terms of meaning (Do you think the idea of the CRA providing information about individuals to other federal and provincial/territorial government departments is a good idea or a bad idea?). The implication is that respondents can change their views about the CRA sharing information with other levels of government when exposed to various questions that prompt a kind of deliberative process. This seems to suggest that transparency about information sharing, such as the reason for doing so and the benefit to Canadians, might encourage Canadians to support the CRA sharing information.
• The consistently strong effects of Q15D (providing information about people's identification and income levels so that provinces/territories can administer their programs), Q15E (providing a person's address information to help federal or provincial/territorial departments identify people who owe money to them), and Q20A (a greater exchange of information would make governments more effective and efficient) suggest that Canadians are more accepting of the CRA sharing information when it serves specific administrative purposes (e.g., collecting money or verifying ID) or could lead to improvements in program effectiveness and efficiency. In contrast, the more abstract tasks of planning and program design did not resonate to the same extent with respondents. For example, Q15C (providing the number and ages of dependents to help design new programs) and Q15A (providing people's name/address so that mailing lists can be kept up to date for benefit programs and databases) have no significant effect on views of information sharing.
• Several variables relate to trust and expectations of security. Trusting the CRA to keep information confidential (Q20D) and to use the most secure methods (Q21A) have no impact on the dependent variable , suggesting a lack of knowledge or understanding of the CRA's practices in this area. Expecting the CRA to use the most secure methods to store tax information (Q20B), however, does drive respondent support for the CRA sharing information, suggesting that Canadians already have high expectations of the CRA when it comes to security measures. In addition, Q20G draws attention to the other side of the equation—the need for Canadians to trust other federal government departments that would be receiving information from the CRA. This has possible implications for communications strategies: the CRA may want to consider sharing information with Canadians about the measures in place to safeguard their personal information, both at the CRA's end and at the receiver's end.
• There is a theme relating to how respondents view the behaviour of others when it comes to tax filing. Data from several questions suggest that Canadians might be less likely to support the CRA sharing information because they think this would encourage ‘others' to underreport or not report tax-related information. This also has implications for potential communications strategies: the CRA may want to consider the value of information campaigns designed to remind Canadians of their tax responsibilities and the consequences of non-compliance.
• The socio-economic and demographic variables only had a modest impact on the dependent variable suggesting that communications plans need to target all Canadians at this time rather than focus on specific segments of the population.

Appendix

1. Technical Specifications

The following specifications applied to this survey:

• A 17-minute random digit dialling (RDD) telephone survey was administered to 2,200 Canadian residents, 18 years of age or older, between February 3rd to March 14th, 2022.
• The questionnaire was pre-tested in advance of the fieldwork to ensure that it measured what it intended to measure and that respondents understood the questions.
• An overlapping dual-frame (landline and cell phone) sample was used to minimize coverage error. The same random selection process was used for both the landline and cell phone sample, thereby ensuring the integrity of the probability sample.
• The sample frame was geographically disproportionate to improve the accuracy of regional results. The distribution of completed surveys was as follows:

  -	Completed Interviews
  Newfoundland and Labrador	100
  Nova Scotia	100
  Prince Edward Island	102
  New Brunswick	101
  Quebec	400
  Ontario/Nunavut	604
  Manitoba/Saskatchewan	200
  Alberta/Northwest Territories	251
  British Columbia/Yukon Territories	342
  Canada	2,200

• Based on a sample of this size, the overall results can be considered accurate within ±2.2%, 19 times out of 20 (adjusted to reflect the geographically disproportionate sampling).
• The following table presents information about the final call dispositions for this survey and calculation of the response rate:

  -	Total	Landline	Cell
  Total Numbers Attempted	107,445	15,872	91,573
  Out-of-scope - Invalid	55,985	1,895	54,090
  Unresolved (U)	31,741	8,327	23,414
  No answer/Answering machine	31,741	8,327	23,414
  In-scope - Non-responding (IS)	17,038	4,272	12,766
  Language barrier	226	36	190
  Incapable of completing (ill/deceased)	160	99	61
  Callback (Respondent not available)	1,327	343	984
  Refusal	14,771	3,582	11,189
  Termination	554	212	342
  In-scope - Responding units (R)	2,681	1,378	1,303
  Completed Interview	2,200	1,206	994
  Quota Filled	60	16	44
  Not Qualified – Age, Region, Non-filer	421	156	265

• The overall response rate was 5% (10% for the landline sample and 3% for the cell phone sample). The response rate formula is calculated as follows: [R=R/(U+IS+R)]. This means that the response rate is calculated as the number of responding units [R] divided by the number of unresolved [U] numbers plus in-scope [IS] non-responding households and individuals plus responding units [R].
• The survey data have been weighted by region, age and gender using population figures from Statistics Canada's census data. The weights correct for the disproportionate regional sampling. Any respondents who refused to provide their age were given a neutral weight so as not to skew the weighting proportions. The table below shows the unweighted and weighted proportions for the variables used to create the weights:

  -	Unweighted	Weighted
  Base	n=2,220	n=2,220
  Region
  Atlantic Canada	403	189
  Quebec	400	498
  Ontario*	600	803
  Prairies	200	110
  Alberta**	251	197
  British Columbia***	336	310
  Age
  Under 35	280	349
  35 to 54	539	529
  55 to 64	427	405
  65 or older	944	824
  Gender
  Male	1,060	957
  Female	1,128	1,149
  Another gender	11	11

*Includes Nunavut.
**Includes Northwest Territories.
***Includes the Yukon Territory.

• A non-response analysis was conducted to assess the potential for non-response bias. Survey non-response can bias results when there are systematic differences between survey respondents and non-respondents. To undertake the analysis, characteristics of survey respondents—region, gender and age—were compared with those of the target population. Below, these comparisons are discussed and any differences between the survey sample and the population are evaluated in terms of the potential for non-response bias.
• The following table compares the survey sample on a regional basis to the population parameters by gender. All survey data are unweighted.

  Regions	Male		Female
  -	% of Population	% of Sample	% of Population	% of Sample
  Atlantic	47.0%	47.6%	52.5%	52.1%
  Quebec	43.3%	48.5%	56.7%	51.5%
  Ontario*	46.1%	47.2%	53.2%	51.8%
  Prairies	49.0%	46.2%	50.1%	53.8%
  Alberta**	42.8%	52.6%	56.1%	46.6%
  B.C.***	45.0%	48.5%	54.3%	50.9%
  Canada	45.2%	48.2%	54.3%	51.3%

*Includes Nunavut.
**Includes Northwest Territories.
***Includes the Yukon Territory.

Overall, the survey sample fairly closely approximated the target population on a national level. On the regional level, the sample generally approximated the population in five of six regions. The sample from Alberta showed the highest discrepancy from the population in terms of gender with a margin of 10%. Weights were applied to adjust for the discrepancy between the survey sample and the population. Weighting serves to reduce bias should it be present, but not to eliminate it completely.

The following table compares the survey sample on a regional basis to the population parameters by age. All survey data are unweighted.

*Includes Nunavut.
**Includes Northwest Territories.
***Includes the Yukon Territory.

Overall, the survey sample also fairly closely approximated the target population on a national level in terms of age. On a national level, Canadians under 35 were under-represented slightly in the survey sample (by 3.8%) and those aged 65 and older were slightly over-represented in the survey sample (by 4.0%). On the regional level, the survey sample overrepresents those aged 65 and older who live in the Prairies (by 14.4%) and British Columbia (by 10.1%). Conversely, 18–34-year-olds are underrepresented in Alberta (by 7.4%) and in Ontario (by 5.5%). Across the other regions, there were smaller discrepancies. In order for the results to be representative of the population of Canada, weights were applied to correct for this discrepancy between the sample proportions and the population.

2. Survey Questionnaire - English

Introduction

Hello/Bonjour, my name is [Interviewer's name] and I am calling from Phoenix SPI on behalf of the Government of Canada. We are conducting a survey on current issues of interest to Canadians. Would you prefer that I continue in English or French? Préférez-vous continuer en français ou en anglais? [IF NEEDED: Je vous remercie. Quelqu'un vous rappellera bientôt pour mener le sondage en français.]

The survey takes about 15 minutes and is voluntary. Your responses will be kept confidential and anonymous, and the information provided will be administered according to the requirements of the Privacy Act, the Access to Information Act, and any other pertinent legislation.

This survey is registered with the Canadian Research Insights Council's survey validation system. Should you have any questions about the survey, I can give you a contact person within the Government of Canada. May I continue?

[LANDLINE SAMPLE]

A. May I please speak with the person in your household who is 18 years of age or older and who has had the most recent birthday? Would that be you?

[CELL SAMPLE]

B. Are you 18 years of age or older?

C. Are you in a place where you can safely talk on the phone and answer my questions?

INTERVIEWER NOTE: If you think respondents' setting may endanger them (e.g., driving a vehicle), hinder their participation (e.g., background distractions), or hamper their ability to respond openly (e.g., lack of privacy), suggest rescheduling the interview.

D. Does your household subscribe to a home phone service, also known as a landline?

[EVERYONE]

E. Have you ever filed a tax return?

F. Do you, or does anyone in your family or household, work in any of the following areas? [READ LIST]

G. In which province or territory do you live? [DO NOT READ]

H. In what year were you born?

I. [ASK IF H=99] Would you be willing to tell me in which of the following age categories you belong?

[READ LIST; STOP WHEN RESPONDENT SELECTS AN ANSWER]

J. What is your gender?

[READ LIST; STOP WHEN RESPONDENT SELECTS AN ANSWER]

Section 1: Online/Digital Profile

To start,

1. Do you use the Internet, whether on a computer, tablet or smart phone?

IF YES, ASK

2. [IF Q1=01] Which of the following devices do you use to access the Internet? Do you use a…? [READ LIST. ACCEPT ALL THAT APPLY]

3. [IF Q1=01] Do you use any social media? [INTERVIEWER NOTE: IF RESPONDENT ASKS WHAT IS MEANT BY SOCIAL MEDIA, SAY THIS INCLUDES THINGS LIKE FACEBOOK, TWITTER, INSTAGRAM AND WHAT'S APP, AMONG OTHERS]

IF YES, ASK:

4. [IF Q3=01] Which social media do you personally use? [DO NOT READ LIST. ACCEPT ALL THAT APPLY. RECORD FIRST MENTION]

5. [IF Q3=01] Do you use social media every day?

6. [IF Q5=01] How much time do you spend on social media in a typical day?

7.B. [IF Q5=02] How much time do you spend on social media in a typical week?

Section 2: Views on Algorithms and Big Data

ASK ALL INTERNET USERS:

7. As part of their internet strategy, organizations are making increasing use of algorithms [AL-GR-I-THMS] to tailor personalized ads or services to specific users. The algorithms take into account the things you look at, interact with, and ‘like' on the internet, and then suggest other things you might like. When using the Internet for social media or other activities, would you say you notice this a lot, a moderate amount, not very much, or not at all? [READ SLOWLY AND CLEARLY]

8. Do you think the use of algorithms in this way is a good or bad idea? Please use a 5-point scale, where ‘1' means you think it is a very bad idea, ‘5' means a very good idea, and ‘3' is neither a good or bad idea. [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.]

ASK EVERYONE:

9.Many organizations are making greater use of personal information they have about us, like age, income, and our use of websites. They use this information for different purposes, including to improve how they operate and to provide better customer service. In general, do you think this is a good thing or a bad thing? Please use a 5-point scale, where ‘1' means you think it is a very bad idea, ‘5' means a very good idea, and ‘3' is neither a good or bad idea. [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.]

Section 3: Views on Information Sharing

This survey is about federal and/or provincial [territorial if G=11-13] governments exchanging information about individual people. By information, I mean personal information that someone may submit to Canada Revenue Agency, or C-R-A, when they file their tax return. T1999

10. Do you think the idea of the CRA providing information about individuals to other federal and provincial [territorial if G=11-13] government departments is a good idea or a bad idea? Please use a 5-point scale, where ‘1' means a very bad idea, ‘5' a very good idea, and ‘3' means neither a good or bad idea. [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.] T1999 Modified

11. Do you believe the CRA and other areas of the Canadian federal government currently exchange information like this? T1999

12. And do you believe the CRA and provincial [territorial if G=11-13] governments currently exchange information like this? T1999-MODIFIED

[ROTATE PRESENTATION OF Q1 Q13 AND4]

13. [ASK IF Q11=01 OR Q12=01] Do you believe the CRA is currently providing information to [SPLIT SAMPLE; 50% Q13a "other areas of the federal government"; 50% Q13b "provincial [territorial] governments"] to locate people who owe money to the government? T1999-MODIFIED

14. [ASK IF Q11=01 OR Q12=01] Do you believe the CRA is currently providing information to [SPLIT SAMPLE; 50% Q14a "other areas of the federal government"; 50% Q14b "provincial [territorial] governments"] to determine whether people are eligible for benefits? T1999-MODIFIED

Section 4: Support for Information Sharing

[READ SLOWLY AND CLEARLY. IF RESPONDENT ASKS A QUESTION, RE-READ RELEVANT PORTION OF QUESTION, OR RE-READ ENTIRE QUESTION. READ VERBATIM.]

I would like to take a minute to explain some of the issues in having the CRA share its information. Under current laws, CRA is permitted to release information about individuals only under very limited circumstances.

CRA often collects the same information as other government departments These government departments have the legislative authority to collect this information on their own. Having CRA share this information could make governments operate more efficiently.

We are interested in your views about what types of information sharing you would support, and in what cases, if any, you would be comfortable with the CRA sharing information without seeking your consent. T1999

15. I am now going to read some information-sharing requests the CRA could receive from other areas of the federal government or provincial [territorial if G=11-13] governments. Please tell me how strongly you would support or oppose exchanging information for each reason, using a 5-point scale, where ‘1' means you strongly oppose it, ‘5' means strongly support, and ‘3' means neither support or oppose. [READ AND ROTATE ITEMS; ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.]

a. Providing people's name and address so that mailing lists can be kept up-to-date for benefit programs and databases such as vital statistics registries could be kept up to date.
b. Providing people's income information so that federal and provincial [territorial if G=11-13] government departments can determine whether people are eligible for benefits
c. Providing the number and ages of dependents people have to help in the design of new programs
d. Providing information about people's identification and income levels so that provinces [territorial if G=11-13] can administer their programs
e. Providing a person's address information to help federal or provincial [territorial if G=11-13] departments identify people who owe money to them
f. Providing a person's identification and income information to Canadian law enforcement agencies, such as the RCMP.

Section 5: Views on Increased Information Sharing

16. Suppose the CRA and other government departments and agencies increased the extent to which they exchange information. Do you think that some people would change the way they prepare their tax returns for CRA? T1999

17. [ASK IF Q16=01] In what ways might people change the way they prepare their tax returns for the Canada Revenue Agency? T1999 Modified [DO NOT READ LIST. ACCEPT ALL THAT APPLY]

18. And would the increased exchange of information between CRA and other government departments and agencies change the way you prepare your tax returns for CRA? T1999

19. [ASK IF Q18=01] In what way would you change the way you prepare your tax returns for CRA? T1999 Modified [DO NOT READ LIST. ACCEPT ALL THAT APPLY]

20. I am going to read you a list of statements. For each one, please tell me whether you agree or disagree with it using a 5-point scale, where ‘1' means you strongly disagree, ‘5' means strongly agree, and ‘3' means neither agree or disagree. T1999 Modified [READ/ROTATE ITEMS; [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.]

a) A greater exchange of information would make governments more effective and efficient
b) An increase in the exchange of information would lead to some people reporting less information on their income tax returns
c) It would be okay for some people to not file their tax returns to the CRA if they knew the information was going to shared with other government departments
d) I trust the CRA to keep my information confidential and prevent it from getting into the wrong hands
e) It would be okay for some people to withhold personal information from CRA if they knew the information was going to be given to other government agencies without their consent
f) An increase in the exchange of information would lead to some people reporting inaccurate information
g) I trust other federal government departments to keep my information confidential and prevent it from getting into the wrong hands
h) It would be okay for some people to report lower income on their tax returns to CRA if they knew the information was going to be given to other government agencies
i) If the CRA begins to exchange more information with other government agencies, it will be going too far

21. I am going to read you a list of statements about the CRA specifically. For each one, please tell me whether you agree or disagree with it using a 5-point scale, where ‘1' means you strongly disagree, ‘5' means strongly agree, and ‘3' means neither agree or disagree. [READ/ROTATE ITEMS; [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.]]

a) I trust the CRA to hold on to my personal information using the most secure methods
b) I expect the CRA to have the most secure methods to store tax information

[SPLIT SAMPLE Q21D AND Q21D]
c) When it submits tax information to other federal and provincial [territorial if G=11-13] government departments, I trust the CRA to ensure that the transmission of the information is secure
d) I trust the CRA to use the most secure methods of transmitting tax information

22. Now that we have talked about the issue for some time, let me ask you a question I asked at the beginning of this interview. Do you think the idea of the CRA providing information about individuals to other federal and provincial [territorial if G=11-13] government departments is a good or bad idea? Please use a 5-point scale, where ‘1' means a very bad idea, ‘5' a very good idea, and ‘3' means neither a good or bad idea. [ACCEPT "DON'T KNOW" OR "REFUSED" IF VOLUNTEERED; RECORD SEPARATELY.] T1999 Modified

Demographics

We have a few final questions for statistical classification purposes. Be assured that your responses will be held in strict confidence.

23. What is the highest level of formal education that you have completed?

[READ LIST; STOP WHEN RESPONDENT SELECTS AN ANSWER]

24. Which of the following best describes your total household income last year, before taxes, from all sources for all household members?

[READ LIST; STOP WHEN RESPONDENT SELECTS AN ANSWER]

25. Which of the following categories best describes your current employment status? Are you… [READ LIST - ACCEPT ONE ANSWER ONLY]

26. What language do you speak most often at home? [ACCEPT ALL THAT APPLY]

[DO NOT READ]

27. What are the first three digits of your postal code?

28. RECORD: LANGUAGE OF SURVEY COMPLETION

CONCLUSION

That concludes the survey. Thank you very much for your thoughtful feedback. This survey was conducted on behalf of the Canada Revenue Agency.

Survey Questionnaire- French

Introduction

Bonjour/Hello, je m'appelle [nom de l'intervieweur] et je téléphone de la part du Phoenix SPI au nom du gouvernement du Canada. Nous menons un sondage sur les sujets d'intérêt actuels pour les Canadiens. Préférez-vous continuer en français ou en anglais ? (AU BESOIN : Thank you. Someone will call you back soon to conduct the survey in English

Ce sondage dure environ 15 minutes et est entièrement volontaire. Vos réponses demeureront confidentielles et anonymes, et les renseignements fournis seront administrés selon la Loi sur la protection des renseignements personnels, la Loi sur l'accès à l'information et toute autre loi qui s'y applique.

Ce sondage est enregistré dans le système de validation des sondages du Conseil de recherche et d'intelligence marketing canadien. Si vous avez des questions au sujet du sondage, je peux vous donner le nom d'une personne-ressource au sein du gouvernement du Canada. Puis-je continuer?

[ÉCHANTILLON AVEC LIGNE DE TÉLÉPHONE FIXE]

K. Puis-je parler à la personne de votre ménage qui est âgée de 18 ans ou plus et dont l'anniversaire est le plus récent? Est-ce vous?

[ÉCHANTILLON AVEC CELLULAIRE]

L. Êtes-vous âgé de 18 ans et plus?

M. Êtes-vous dans un endroit où vous pouvez parler au téléphone en toute sécurité et répondre à mes questions?

REMARQUE À L'INTENTION DE L'INTERVIEWEUR : Si vous croyez que le contexte des répondants peut les mettre en danger (p. ex., conduite d'un véhicule), nuire à leur participation (p. ex., distractions en arrière-plan) ou nuire à leur capacité à répondre ouvertement (p. ex., manque d'intimité), suggérez de reporter l'entrevue.

N. Votre ménage est-il abonné à un service téléphonique résidentiel, aussi appelé téléphone fixe?

*INDIQUEZ LE CODE POUR TÉLÉPHONE CELLULAIRE SEULEMENT [SURVEILLEZ LES QUOTAS]

[TOUT LE MONDE]

O. Avez-vous déjà produit une déclaration de revenus?

P. Est-ce que vous ou un membre de votre ménage travaillez dans l'un des secteurs suivants? [LISEZ LA LISTE]

Q. Dans quelle province ou quel territoire vivez-vous? [NE LISEZ PAS CETTE OPTION]

R. Quelle est votre année de naissance?

S. [POSEZ LA QUESTION SI G=99] pourriez-vous me dire à quelle catégorie d'âge vous appartenez?

[LISEZ LA LISTE; ARRÊTEZ LORSQUE LE RÉPONDANT CHOISIT UNE RÉPONSE]

T. Quel est votre genre?

[LISEZ LA LISTE; ARRÊTEZ LORSQUE LE RÉPONDANT CHOISIT UNE RÉPONSE]

Section 1 : Profil en ligne/numérique

Pour commencer,

29. Utilisez-vous Internet, sur un ordinateur, une tablette ou un téléphone intelligent?

DANS L'AFFIRMATIVE, POSEZ LA QUESTION SUIVANTE :

30. [SI Q1=01] Lequel des appareils suivants utilisez-vous pour accéder à Internet? Utilisez-vous un/une...? [LISEZ LA LISTE. ACCEPTEZ TOUTES LES RÉPONSES QUI S'APPLIQUENT.]

31. [SI Q1=01] Utilisez-vous des médias sociaux? [REMARQUE À L'INTENTION DE L'INTERVIEWEUR : SI LE RÉPONDANT DEMANDE CE QUE L'ON SIGNIFIE PAR « MÉDIAS SOCIAUX », DITES QUE CELA COMPREND DES CHOSES COMME FACEBOOK, TWITTER, INSTAGRAM ET WHAT'S APP, ENTRE AUTRES]

DANS L'AFFIRMATIVE, POSEZ LA QUESTION SUIVANTE :

32. [SI Q3=01] Quels médias sociaux utilisez-vous personnellement? [NE LISEZ PAS LA LISTE. ACCEPTEZ TOUTES LES RÉPONSES QUI S'APPLIQUENT. ENREGISTREZ LE PREMIER ÉLÉMENT MENTIONNÉ]

33. [SI Q3=01] Utilisez-vous les médias sociaux tous les jours?

34. [SI Q5=01] Combien de temps consacrez-vous aux médias sociaux au cours d'une journée habituelle?

7.B. [SI Q5=02] Combien de temps consacrez-vous aux médias sociaux au cours d'une semaine habituelle?

Section 2 : Points de vue sur les algorithmes et les mégadonnées

DEMANDEZ À TOUS LES UTILISATEURS D'INTERNET :

35. Dans le cadre de leur stratégie Internet, les organismes utilisent de plus en plus des algorithmes pour adapter les publicités ou les services personnalisés à des utilisateurs précis. Les algorithmes tiennent compte des choses que vous regardez et « aimez » et avec lesquelles vous interagissez sur Internet, puis vous suggèrent d'autres choses que vous pourriez aimer. Lorsque vous utilisez Internet pour les médias sociaux ou d'autres activités, diriez-vous que vous remarquez cela beaucoup, de façon modérée, pas beaucoup ou pas du tout? [LIRE LENTEMENT ET CLAIREMENT]

36. Pensez-vous que l'utilisation d'algorithmes de cette façon est une bonne ou une mauvaise idée? Veuillez utiliser une échelle de cinq points, où « 1 » signifie que vous pensez qu'il s'agit d'une très mauvaise idée, « 5 » signifie une très bonne idée, et « 3 » n'est ni une bonne ni une mauvaise idée. [ACCEPTEZ « NE SAIT PAS » OU « REFUSE DE RÉPONDRE » S'IL S'EST PORTÉ VOLONTAIRE; ENREGISTREZ SÉPARÉMENT]

DEMANDEZ À TOUT LE MONDE :

37. De nombreux organismes utilisent davantage les renseignements personnels qu'ils ont à notre sujet, comme l'âge, le revenu et notre utilisation des sites Web. Ils utilisent ces renseignements à différentes fins, notamment pour améliorer les sites Web, la façon dont ils fonctionnent et pour fournir un meilleur service à la clientèle. En général, pensez-vous qu'il s'agit d'une bonne ou d'une mauvaise chose? Veuillez utiliser une échelle de cinq points, où « 1 » signifie que vous pensez qu'il s'agit d'une très mauvaise idée, « 5 » signifie une très bonne idée, et « 3 » n'est ni une bonne ni une mauvaise idée. [ACCEPTEZ « NE SAIT PAS » OU « REFUSE DE RÉPONDRE » S'IL S'EST PORTÉ VOLONTAIRE; ENREGISTREZ SÉPARÉMENT. ]

Section 3 : Points de vue sur l'échange de renseignements

Ce sondage porte sur l'échange de renseignements sur les gens entre les gouvernements fédéral et provinciaux [territoriaux si G=11-13]. Par renseignements, je veux dire les renseignements personnels qu'une personne peut soumettre à l'Agence du revenu du Canada (ou ARC), lorsqu'elle produit sa déclaration de revenus. T1999

38. Pensez-vous que l'idée que l'ARC fournisse des renseignements sur les personnes à d'autres ministères fédéraux et provinciaux [territorial si G=11-13] soit une bonne idée ou une mauvaise idée? Veuillez utiliser une échelle de cinq points, où « 1 » signifie qu'il s'agit d'une très mauvaise idée, « 5 » signifie une très bonne idée, et « 3 » n'est ni une bonne ni une mauvaise idée [ACCEPTEZ « NE SAIT PAS » OU « REFUSE DE RÉPONDRE » S'IL S'EST PORTÉ VOLONTAIRE; ENREGISTREZ SÉPARÉMENT. ] T1999 modifié

39. Croyez-vous que l'ARC et d'autres secteurs du gouvernement fédéral du Canada échangent actuellement des renseignements de cette nature? T1999

40. Et croyez-vous que l'ARC et d'autres gouvernements provinciaux [territoriaux si G=11-13] échangent actuellement des renseignements de cette nature? T1999-MODIFIÉ

41. [POSEZ CETTE QUESTION SI Q11=01 OU Q12=01] Croyez-vous que l'ARC fournit actuellement des renseignements [DIVISEZ l'ÉCHANTILLON; 50 % Q13a « à d'autres secteurs du gouvernement fédéral ou aux gouvernements provinciaux [territoriaux] » 50% Q13b « aux gouvernements provinciaux [territoriaux] »] pour repérer les personnes qui doivent de l'argent au gouvernement? T1999-MODIFIÉ

42. [POSEZ CETTE QUESTION SI Q11=01 OU Q12=01] Croyez-vous que l'ARC fournit actuellement des renseignements à [DIVISEZ l'ÉCHANTILLON; 50 % Q14a « d'autres secteurs du gouvernement fédéral »; 50% Q14b « gouvernements provinciaux [territoriaux] »] afin de déterminer si les gens sont admissibles à des prestations? T1999-MODIFIÉ

Section 4 : Soutien à l'échange de renseignements

[LISEZ LENTEMENT ET CLAIREMENT. SI LE RÉPONDANT POSE UNE QUESTION, RELISEZ LA PARTIE PERTINENTE DE LA QUESTION OU RELISEZ LA QUESTION ENTIÈRE. LISEZ TEXTUELLEMENT.]

J'aimerais prendre une minute pour expliquer certains des enjeux liés au fait que l'ARC communique ses renseignements. Selon des lois actuelles, l'ARC est autorisée à divulguer des renseignements sur les personnes seulement dans des circonstances très limitées.

L'ARC recueille souvent les mêmes renseignements que d'autres ministères du gouvernement. Ces ministères ont le pouvoir législatif de recueillir ces renseignements de leur propre initiative. En communiquant ces renseignements, l'ARC permet aux gouvernements de fonctionner plus efficacement.

Nous souhaitons connaître vos points de vue sur les types d'échange de renseignements que vous appuieriez, et s'il y a lieu, dans quels cas vous seriez à l'aise que l'ARC communique vos renseignements sans demander votre consentement. T1999

43. Je vais maintenant lire certaines demandes d'échange de renseignements que l'ARC pourrait recevoir d'autres secteurs du gouvernement fédéral ou de gouvernements provinciaux[territoriaux si G=11-13]. Veuillez me dire dans quelle mesure vous appuieriez l'échange de renseignements pour chaque raison, ou vous y opposeriez, en utilisant une échelle de cinq points, où « 1 » signifie que vous vous y opposez fortement, « 5 » signifie que vous l'appuyer fortement et « 3 » signifie ni appuyer ni opposer. (LISEZ ET FAITES UNE ROTATION DES ÉLÉMENTS) ); ACCEPTEZ « NE SAIS PAS » OU « REFUSE DE RÉPONDRE » S'IL S'EST PORTÉ VOLONTAIRE; ENREGISTREZ SÉPARÉMENT. ]

a. Fournir le nom et l'adresse d'une personne afin que les listes d'envoi puissent être tenues à jour pour les programmes de prestations et que les bases de données, comme les registres de l'état civil, puissent être tenues à jour.
b. Fournir des renseignements sur le revenu d'une personne afin que les ministères fédéraux et provinciaux [territoriaux si G=11-13].puissent déterminer si les gens sont admissibles à des prestations
c. Fournir le nombre et l'âge des personnes à charge pour aider à la conception de nouveaux programmes.
d. Fournir des renseignements sur l'identification et le niveau de revenu des personnes qui permettra aux provinces d'administrer leurs programmes.
e. Fournir les renseignements sur l'adresse d'une personne pour aider les ministères fédéraux ou provinciaux [territoriaux si G=11-13 ]à identifier les personnes qui leur doivent de l'argent.
f. Fournir des renseignements sur l'identification et le revenu d'une personne aux organismes d'application de la loi canadiens, comme la GRC.

Section 5 : Point de vue sur l'augmentation de l'échange de renseignements

44. Supposons que l'ARC et les autres ministères et organismes gouvernementaux donnent leur pleine mesure en matière d'échanges de renseignements. Pensez-vous que certaines personnes changeraient la façon dont elles préparent leurs déclarations de revenus pour l'Agence? T1999

45. [POSEZ CETTE QUESTION SI Q16=01] De quelle façon les gens pourraient-ils changer la façon dont ils préparent leurs déclarations de revenus pour l'Agence du revenu du Canada? T1999 modifié [NE LISEZ PAS LA LISTE. [NE LISEZ PAS LA LISTE. ACCEPTEZ TOUTES LES RÉPONSES QUI S'APPLIQUENT.]

46. L'échange accru de renseignements entre l'Agence et les autres ministères et organismes gouvernementaux changerait-il la façon dont vous préparez vos déclarations de revenus pour l'Agence? T1999

47. [POSEZ CETTE QUESTION SI Q18=01] De quelle façon changeriez-vous la façon dont vous préparez vos déclarations de revenus pour l'Agence? T1999 modifié [NE LISEZ PAS LA LISTE. ACCEPTEZ TOUTES LES RÉPONSES QUI S'APPLIQUENT.]

48. Je vais vous lire une liste d'énoncés. Pour chacun d'eux, veuillez m'indiquer si vous êtes d'accord ou en désaccord en utilisant une échelle de cinq points, où « 1 » signifie que vous êtes tout à fait en désaccord, « 5 » signifie que vous êtes tout à fait d'accord, et « 3 » signifie que vous n'êtes ni d'accord ni en désaccord T1999 modifié [LISEZ/FAITES UNE ROTATION DES ÉLÉMENTS] [ACCEPTEZ « NE SAIS PAS » OU « REFUSE DE RÉPONDRE » S'IL S'EST PORTÉ VOLONTAIRE; ENREGISTREZ SÉPARÉMENT.

j) Un plus grand échange de renseignements rendrait les gouvernements plus efficaces.
k) Une augmentation de l'échange de renseignements inciterait certaines personnes à fournir moins de renseignements sur leurs déclarations de revenus.
l) Il serait acceptable pour certaines personnes de ne pas produire leurs déclarations de revenus à l'ARC s'ils savaient que les renseignements seraient acheminés à d'autres ministères du gouvernement
m) Je fais confiance à l'ARC pour garder mes renseignements confidentiels et éviter qu'ils ne tombent entre de mauvaises mains.
n) Il serait acceptable pour certaines personnes de ne pas divulguer des renseignements personnels à l'ARC si elles savaient que les renseignements seraient fournis à d'autres organismes gouvernementaux sans leur consentement.
o) Une augmentation de l'échange de renseignements inciterait certaines personnes à fournir des renseignements inexacts.
p) Je fais confiance aux autres ministères du gouvernement fédéral pour garder mes renseignements confidentiels et éviter qu'ils ne tombent entre de mauvaises mains.
q) Il serait acceptable pour certaines personnes de déclarer à l'ARC un revenu moins élevé dans leurs déclarations de revenus si elles savaient que les renseignements seraient fournis à d'autres organismes gouvernementaux.
r) Si l'ARC commence à échanger plus de renseignements avec d'autres organismes gouvernementaux, cela ira trop loin.

49. Je vais vous lire une liste d'énoncés au sujet de l'ARC en particulier. Pour chacun d'eux, veuillez me dire si vous êtes d'accord ou en désaccord avec celui-ci sur une échelle de 5 points, où « 1 » signifie que vous êtes tout à fait d'accord, « 5 » signifie tout à fait d'accord, et « 3 » signifie ni d'accord ni en désaccord, plutôt en désaccord ou tout à fait en désaccord avec ce qui suit. ou pas d'accord. [LIRE/FAIRE LA ROTATION DES POINTS]; [LISEZ/FAITES UNE ROTATION DES ÉLÉMENTS] ACCEPTEZ « NE SAIT PAS » OU « REFUSE DE RÉPONDRE S'IL SE PORTE VOLONTAIRE; ENREGISTREZ SÉPARÉMENT.

e) Je fais confiance à l'ARC pour conserver mes renseignements personnels en utilisant les méthodes les plus sécuritaires.
f) Je m'attends à ce que l'ARC ait les méthodes de stockage des renseignements fiscaux les plus sécuritaires possible
[DIVISEZ L'ÉCHANTILLON Q21 ET Q21D]
g) Lorsqu'elle soumet des renseignements fiscaux à d'autres ministères fédéraux et provinciaux, je fais confiance à l'ARC pour veiller à ce que la transmission des renseignements soit sécurisée.
h) J'ai confiance que l'ARC utilisera les méthodes les plus sécuritaires pour transmettre des renseignements fiscaux.

50. Maintenant que nous avons discuté de la question pendant un certain temps, permettez-moi de vous poser une question que j'ai posée au début de cette entrevue. Pensez-vous que l'idée que l'ARC fournisse des renseignements sur les personnes à d'autres ministères fédéraux et provinciaux [territoriaux si G=11-13 ] soit une bonne idée ou une mauvaise idée? Veuillez utiliser une échelle de cinq points, où « 1 » signifie qu'il s'agit d'une très mauvaise idée, « 5 » signifie une très bonne idée, et « 3 » n'est ni une bonne ni une mauvaise idée. T1999 modifié [ACCEPTEZ « NE SAIT PAS » OU « REFUSE DE RÉPONDRE S'IL SE PORTE VOLONTAIRE; ENREGISTREZ SÉPARÉMENT].

Démographie

Nous avons quelques dernières questions à vous soumettre aux fins de classification statistique. Soyez assuré que vos réponses demeureront strictement confidentielles.

51. Quel est le niveau de scolarité le plus élevé que vous ayez atteint?

[LISEZ LA LISTE; ARRÊTEZ LORSQUE LE RÉPONDANT CHOISIT UNE RÉPONSE]

52. Lequel des énoncés suivants décrit le mieux le revenu total de votre ménage pour l'année dernière, avant impôts, de toutes les sources pour tous les membres du ménage?

[LISEZ LA LISTE; ARRÊTEZ LORSQUE LE RÉPONDANT CHOISIT UNE RÉPONSE]

53. Choisissez l'énoncé qui décrit le mieux votre situation d'emploi actuelle. Êtes-vous... [LISEZ LA LISTE – ACCEPTEZ UNE SEULE RÉPONSE]

54. Quelle langue parlez-vous le plus souvent à la maison? [ACCEPTEZ TOUTES LES RÉPONSES QUI S'APPLIQUENT]

[NE LISEZ PAS CETTE OPTION]

55. Quels sont les trois premiers caractères de votre code postal?

56. ENREGISTREZ : LANGUE DE RÉALISATION DU SONDAGE

CONCLUSION

Voilà qui met fin à ce sondage. Merci beaucoup pour vos commentaires perspicaces. Ce sondage a été mené au nom de l'Agence du revenu du Canada.