# **British Columbia Election Information Ecosystem Project: Baseline Report**
September 2024
## **Contents**
| Executive Summary 1 |
|-------------------------------------------------------------------------------------------------|
| The 2024 British Columbia Election .
1 |
| Attitudes and behaviours of British
Columbians 2 |
| Media consumption
2
Trust and confidence
3
Concern about mis- and disinformation .4 |
| Social Media and Politically Influential
BC Entities .
5 |
| Social Media Usage and Engagement
5 |
| Social Media Platform and Engagement .6 |
| Political Party and Engagement
6 |
| Threats & Vulnerabilities 7 |
| Threat Landscape
8 |
| Vulnerabilities
8 |
| Incident Response Plan .
9 |
| Contributors 10 |
| Methodology
10 |
| Survey
10 |
| Digital Trace
11 |
## **Executive Summary**
**The British Columbia Election Information Ecosystem Project**, established by the Media Ecosystem Observatory, will be monitoring the information ecosystem during the 2024 British Columbia general election.
This baseline report provides a snapshot of the attitudes and online behaviours of British Columbians immediately prior to the election, drawing from a survey fielded in the province in September 2024 and a unique social media data collection and analysis methodology focused on influential individuals and organizations based in British Columbia. We observe that:
- E **Partisan divide across social media platforms**—There is a partisan divide in the use of social media in British Columbia: the Conservative Party of BC is most active on X(Twitter) while the BC NDP is far more active and garners more engagement on Instagram. This is the first time we have observed such an extreme bifurcation of politician discourse.
- E **Public sentiments are positive**—British Columbians are going into the election confident that Elections BC will conduct the election fairly. They are comfortable with vote counting machines and have higher levels of trust in the security of voting machines as compared to other Canadians.
- E **Information threat perception consistent with national level**—Despite being surveyed on the verge of an election, British Columbians are not more tuned to issues of misinformation or foreign interference than other Canadians. This may reflect an overall high-level of attention but may also reflect a complacency that these issues will not be present in the 2024 general election.
- E **Less local news available**—The bankruptcy of Black Press Media, which supported many local news outlets in British Columbia, has reduced the volume of local news on social media in the past year. Alongside the **[Meta](https://meo.ca/work/old-news-new-reality-a-year-of-metas-news-ban-in-canada) [news ban](https://meo.ca/work/old-news-new-reality-a-year-of-metas-news-ban-in-canada)** last year, which differentially impacted local news nation-wide, British Columbians are going into an election with less local news available to them.
During the election period, the team will monitor online spaces related to the election and political information in British Columbia and Canada more generally. We will be identifying information incidents or moments where mis- or disinformation, bot-activity, or other digital media threats may impact the normal flow and/or integrity of information, leading to potential or actual harm. If such an event occurs, we will engage in an Incident Response process which aims to quickly notify, analyze, update, and educate the BC population. While information incidents may occur, adequate information and response can mitigate any negative impacts.
## **The 2024 British Columbia Election**
As the campaign period for the British Columbia (BC) general election on October 19, 2024 begins, concerns about the spread of disinformation online and its impact on electoral security are top of mind in the province and across the country.
The use of the information ecosystem and social media platforms to sow discord and distribute false information has been and remains a threat. The risk of disinformation and misinformation that preys on polarised partisanship and foreign interference is increasingly evident in Canada, as politicians, political elites, and media figures move to the online space to engage with the Canadian public.
While concern for disinformation campaigns has often focused on federal or national elections, provincial elections are not immune. They are significant moments of civic and democratic engagement and inform important federalist relationships within the Canadian political system. Misinformation and disinformation campaigns in Canada have been shown to have had an impact on provincial elections, as in the [2022 Quebec election](https://static1.squarespace.com/static/5ea874746663b45e14a384a4/t/641232839282647078a6b59e/1678914185387/Rapport+final_PDEQ_2023.pdf). The 2024 BC election, which takes place during a year marked by many national contests — and only three weeks before the US presidential election — [is at higher risk of disinformation](https://www.vicnews.com/news/closeness-of-bc-election-to-us-election-likely-to-lead-to-more-false-news-7299349) [around political campaigns.](https://www.vicnews.com/news/closeness-of-bc-election-to-us-election-likely-to-lead-to-more-false-news-7299349) Furthermore, the withdrawal of the BC United Party weeks before the electoral campaign period increases political instability and creates a more active and volatile information ecosystem.
Elections BC has [raised public attention](https://elections.bc.ca/2024-provincial-election/election-integrity/threats-to-election-integrity/) to the increasing risks of disinformation and misinformation and has taken steps to [protect electoral integrity against cyber threats and foreign](https://elections.bc.ca/docs/Report-202223ARSP-October2023.pdf) [interference through the](https://elections.bc.ca/docs/Report-202223ARSP-October2023.pdf) *[Election Amendment Act](https://elections.bc.ca/docs/Report-202223ARSP-October2023.pdf)*, 2023[.](https://elections.bc.ca/docs/Report-202223ARSP-October2023.pdf) Key disinformation and misinformation threats include the transmission of false online statements that could impact candidate evaluations and voter registration, ultimately leading to voter suppression. The threat of AI- or other manipulated media is also of increasing concern. Fact-checking by platforms is inefficient and there are insufficient regulation and monitoring measures. A lack of digital literacy and cybersecurity hygiene amongst the electorate and candidates increases vulnerability as the risk of exposure and impact of potential disinformation and misinformation campaigns is greater. These key issues expose the need for heightened awareness of and concern over the distribution of false information by bad actors during the election. Monitoring and securing the online information system is imperative to the electoral and democratic security of the BC election.
This baseline or pre-election report is part of the British Columbia Election Information Ecosystem Project. In this report, and throughout the election, we evaluate and will be evaluating attitudes and behaviours of British Columbians as well as news media, journalists, politicians, and other influencers in online spaces. This study of the British Columbia information ecosystem is the first-ever of its kind.
## **ATTITUDES AND BEHAVIOURS OF BRITISH COLUMBIANS**
This baseline report provides a snapshot of the attitudes and behaviours of British Columbians immediately prior to the 2024 British Columbia general election.
It explores the sources British Columbians use to get information about the election. It also quantifies their general levels of trust, their trust in Elections British Columbia, and their trust in electronic tabulators. We also examined the extent to which British Columbians are concerned about mis- and disinformation.
The forthcoming post-election report will capture any changes in these metrics that occurred during the campaign. When appropriate, we compare the attitudes of British Columbians to a comparable sample of Canadians from other provinces.
All data is drawn from a representative survey conducted in early September 2024, with results weighted by age and gender (see the "Methodology" below for additional details).
### **[Media consumption](#page-3-1)**
British Columbians have a strong preference for receiving their news from digital platforms. As illustrated in Figure 1, news websites are the most popular source of information, with 67.4% of respondents following the news closely on them. Television continues to be a popular choice, particularly amongst older British Columbians (59.3%). Social media is more closely followed (49.4%) than traditional media, such as radio (39.3%) and print (37.7%). Podcasts rank the lowest, with only 19.3% of respondents closely following news through this medium.
The top three social media platforms used by British Columbians are Facebook, Youtube, and Instagram. Figure 2 shows the percentage of British Columbians that access a given platform at least weekly (general use). All respondents who indicated they used the platform were asked about political use, and we also report the percentage of British Columbians that use a given platform for politics at least weekly (political use).
Facebook is the most widely used social media platform, with more than two-thirds of respondents using it for general purposes (68.8%) and more than a quarter to engage with political
content and current affairs (27.3%). The decision in 2023 by [Meta, Facebook's parent company, to block news](https://meo.ca/work/old-news-new-reality-a-year-of-metas-news-ban-in-canada) on the platform has not resulted in major displacement of Facebook and Instagram usage. Youtube is the second most commonly used platform, with 58.7% for general purposes and, like Facebook, 25.5% for political purposes. Instagram has 51.6% general users and noticeably less use for political content (19.3%). X/Twitter usage is split with 25.8% of users engaging for general purposes and 17.0% for political content. This is the smallest difference between political and general usage, reflecting Twitter's dual role as a platform for both general and political conversations. Despite its rapidly growing popularity in recent years, TikTok continues to have the lowest general (22.1%) and political usage (12.7%), although usage is high among young British Columbians: among those under 25, 47.0% use TikTok and 20.1% for political content.
### **Trust and confidence**
Trust in the news media is lower in BC than nationally. We evaluated respondents' trust in various groups to 'act in the best interests of Canadians'. British Columbians had notably different trust levels than Canadians from other regions (see Figure 3). Less than half BC respondents (42.4%) expressed trust in the news media, compared to 51.6% across the country. BC respondents also show lower levels of trust towards journalists: with 47.8% expressing some or a lot of trust, compared to 55.4% at the national level.1 When it comes to trust in elected officials, 32.8% of respondents from BC report that they trust these officials, which is slightly lower than the national average of 33.4%.
**1 This comparably low level of trust in journalists in the BC-only sample may be the cause of a specific event or context when the survey was fielded, with previous BC samples in August, July, and June showing comparable levels of trust at 61%, 51%, and 56%.**
**Figure 2: Social media use by British**
**Columbians**
**Figure 3: Trust levels amongst British Columbians**
Given that electronic vote tabulators will be introduced for the first time in the October 19 election, we measured public perceptions of this new technology (Figure 4). The vast majority of respondents in BC (83.1%) believe electronic vote-counting machines are as or more secure than hand counting and 85.2% are comfortable with the use of tabulators.
We also find that an overwhelming majority of the public (89.2%) has confidence in Elections BC's ability to conduct the upcoming election fairly, with more than half indicating they are very confident (57.2%). Only a smaller portion of respondents expressed doubts: 7.3% were not very confident and 3.4% not at all confident. A strong majority of the population maintains high levels of confidence in the fairness of Elections BC.
### **Concern about mis- and disinformation**
Concern about mis- and disinformation among British Columbians about online misinformation closely reflects national levels (Figure 5). In BC, 47.7% of respondents reported being very or extremely concerned about online misinformation, while 47.2% of respondents across Canada expressed the same level of concern.
A significant portion of respondents in both BC (55.8%) and across Canada (58.3%) are very or extremely concerned about AI-generated content misleading the public. Apprehension about AI-generated content is widespread and fairly consistent, suggesting that the issue resonates strongly with both provincial and national populations.
There is likewise a lot of concern about foreign interference in the election. More than half of British Columbians (51.5%) indicated they were very or extremely concerned about foreign governments influencing Canadian media and politics. This level of concern is comparable to the national level (53.8%). Concern about foreign influence specifically on the BC elections is less but again comparable to the national election. Despite being on the verge of an election, respondents were not more tuned to these issues than other Canadians.
Using the digital trace data (described in the next section), we evaluated the extent to which conversations about misinformation were present in the British Columbia information ecosystem. We find that 0.33% of all posts from January to September 2024 referred to misinformation, disinformation, or foreign interference. This is higher than we generally observe at the national level (0.32% BC versus [0.28% national in June 2024](https://www.cdmrn.ca/publications/2024-08) and 0.64% BC versus [0.22% national in August 2024](https://static1.squarespace.com/static/65427f5b140649321cd829e9/t/66df4e3bc60a4141e729e1e6/1725910588303/SitRep24-08_VF.pdf)). Based on our analysis, this high level is driven by several prominent influencers who use the terms 'misinformation' and 'disinformation' when referring to traditional media outlets and politicians with whom they disagree.
**Figure 4: Confidence in Elections British Columbia and vote counting machines**
**Figure 5: Concern about mis- and disinformation going into the British Columbia Election**
### **SOCIAL MEDIA AND POLITICALLY INFLUENTIAL BC ENTITIES**
Based on an exploratory analysis of social media content and engagement (here measured as the aggregation of the number of likes or hearts across posts) among individuals and organisations relevant to the upcoming BC election, we find a high level of segmentation in online political discourse. There are distinct patterns of social media engagement among BC's major provincial political parties and among the categories of entities we studied: politicians (incumbents and candidates), news outlets, influencers, and civil society organisations.
Through an extensive process detailed in the Methodology section, we developed a seed list of 755 politicians (incumbents and candidates), news outlets, journalists, influencers, and civil society organisations, all of whom are arguably relevant to political discourse in British Columbia. Having identified these individuals and organisations (henceforth entities), we confirmed their corresponding official social media accounts on Facebook, X/Twitter, Instagram, Youtube, and TikTok (1,057 accounts). We then collected all of their social media content from each of those four platforms between January 1 and September 1, 2024. This left us with 652 entities who had posted on one of the four platforms we were able to collect for this report. To facilitate cross-platform comparability, we treat the number of posts published by a given entity as indicative of its rate of usage, and the number of likes garnered by each entity as a measure of engagement. Further explanation on the inclusion criteria and collection methods for the seed list can be found in the Methodology section below.
#### **Social Media Usage and Engagement**
Although news outlets published the highest volume of posts (Figure 6) — 49.9% of all posts — influencers garnered the majority of engagement (63.9% of all engagement) among BC entities in our dataset. Engagement with politicians' posts has been decreasing slightly in recent months, even though the number of posts published by politicians has increased since the beginning of the year. This may signal that BC candidates are struggling to keep their social media audience engaged, even with the approach of the election period.
We also see a steep decline in social media posts from news outlets at the end of March. This may be due to the sale and restructuring of Black Press Media, a BC-based publisher that owns dozens of local BC news media properties. Many of Black Press Media's subsidiary news outlets stopped regularly posting content on X/Twitter shortly before the [announcement](https://www.blackpressmedia.com/black-press-media-concludes-transition-of-ownership/) of the publisher being sold on March 25, 2024. The economic vulnerability of local news production is hardly unique to British Columbia, but belongs to a widely recognized trend across Canada. Indeed, in [our previous analysis of Meta's ongoing ban on](https://meo.ca/work/cdmrn/when-journalism-is-turned-off) [news](https://meo.ca/work/cdmrn/when-journalism-is-turned-off) content on Facebook, we found that among Canadian news organisations, local news outlets were the hardest hit, suffering the largest proportional losses in engagement and being the least diversified across social media platforms. This decline in independent local news outlets has important implications for the quality of local news content, and the vulnerability of local news to disinformation and interference.
### **Social Media Platform and Engagement**
Overall, X/Twitter accounts form the highest proportion of online engagement (54.8%). This, however, reflects a comparatively low rate of engagement given that X/Twitter posts account for 90.8% of total posts. By contrast, 29.2% of total engagement is on Instagram, despite it only accounting for 8.1% of total posts. Similarly, 14.6% of engagement is with TikTok, despite it contributing only 0.6% of total posts; and Youtube accounts for 1.4% of total engagement and 1.1% of total posts.
There is a substantial difference in what constitutes political or news content on each social media platform. Figure 7 shows the weekly engagement of each entity category broken down by social media platform. Civil society entities derive most of their engagement from Instagram (89.1% of their total engagement). There is also significantly more engagement with politicians on Instagram (76.4% of their total) than other platforms like X/Twitter (22.1% of their total), with federal BC-based politicians, such as Jagmeet Singh, driving the largest proportion of this engagement. Influencers amass most of their engagement from X/Twitter (75.8%) And most X/Twitter engagement is with influencers, particularly right-wing accounts such as Kat Kanada and Billboard Chris. With their engagement restricted on Meta platforms, news outlets derive the majority of their engagement from TikTok (64.9%). This engagement is often dependent on non-political news content, such as videos featuring animals, going viral and gaining a disproportionate amount of engagement.
### **Political Party and Engagement**
Online engagement with the BC NDP and with the BC Conservative Party is comparable: the BC NDP constituting 40.8% of total engagement and the BC Conservative Party, 45.2%. The BC Green Party generally sees low levels of engagement on social media, constituting only 14.0% of the total engagement, and most of this can be attributed to occasional viral posts by a single MLA candidate, Nicola Spurling, running in Coquitlam-Maillardville.
The BC NDP and the BC Conservative Party show distinct patterns depending on the platform. The BC NDP dominates engagement on Instagram, while the Conservative Party dominates engagement on X/Twitter. This suggests that left-leaning political discourse tends to occur in a different social media space than right-leaning political discourse.
This aligns with [our findings from the Alberta general elections](https://meo.ca/work/election-misinfo/alberta2023) [in May 2023](https://meo.ca/work/election-misinfo/alberta2023). Comments on Instagram posts published by NDP and UCP candidates during the electoral writ period indicated that there was minimal overlap in readerships, even on a single platform.
Our findings in BC suggest an even starker divide. Conservative and left-leaning audiences likely do not even frequent the same platforms: conservatives congregate on X/Twitter, while NDP supporters gravitate toward Instagram.
**Figure 7: Engagements rates in 2024 with British Columbiabased politically influential entities, separated by type of entity and type of platform**
**Figure 8: Engagements rates with political candidates, separated by political party and highlighting Instagram and X**
## **Threats & Vulnerabilities**
**Disinformation and misinformation** campaigns and incidents can negatively shape the process or outcome of an election. The project team has identified a range of information threats and vulnerabilities that could have considerable impact on the 2024 election in British Columbia.
Threats range from the spread of toxic and inaccurate messaging to more sophisticated attacks on the security of electoral institutions and systems. Vulnerabilities range from contentious political figures to insufficient public communications. If exploited, these vulnerabilities can be leveraged by political figures, operatives, and influencers who seek to gain political prominence or influence through stoking disinformation and division. Foreign influence on key debates relating to affordability, public safety, and political stability are also of concern. False or misleading stories that go unchecked may be used to generate political tension and can emerge from legacy or alternative and far-right media outlets. An information incident2
**2 An** *information incident* **is a disruption in the information ecosystem, including both sudden and prolonged interruptions, that significantly impacts the normal flow and/or integrity of information, leading to potential or actual harm to the public, government, Canadian democracy, and/or the broader information ecosystem.**
can impact voter perceptions, behaviours, and, ultimately, election outcomes.
The following section provides a more detailed outline of existing threats to the BC election and subsequent vulnerabilities.3 Note that this is not an exhaustive list of existing threats and vulnerabilities and this discussion aims to conceptualise a risk landscape that is complex and constantly evolving. Critical threats and vulnerabilities are expected to change over the course of the election period. During the election period, the research team will monitor the information ecosystem4 in BC using this threat landscape and these vulnerabilities as a baseline for detecting information incidents.
- **3 Note, this is not an exhaustive list of existing threats and vulnerabilities. This discussion aims to conceptualise a risk landscape that is complex and constantly evolving. Critical threats and vulnerabilities are expected to change over the course of the election period.**
- **4 An** *information ecosystem* **is the sum of a complex but analyzable set of relationships found in and across digital media. An online information ecosystem is thus composed of interconnected but distinct communities across social and traditional media: distinct because of the different platforms that constitute it but interconnected by a common set of individuals and organizations (entities) who share information and relationships across multiple platforms.**
#### **Threat Landscape**
A threat is a potential information incident that could impact the behaviour or reputation of a political party or leader; voter knowledge, behaviour, awareness, or attention; or voter turnout and election results. We identify prominent potential threats that could impact the electoral information ecosystem.
**Mis/Disinformation** Misrepresentation of a candidate or an organisation; election tools or technologies; the voting process, location, or eligibility; with the intent to impact election outcomes or suppress voters.
> False claims regarding election fraud or electoral policies that could have an impact on the role of electoral officials and/ or mislead voters.
> False claims about the attitude and behaviour of British Columbians with the intent to falsely inform voters.
The use of false statements to alter public debate or of advertising campaigns to drive attention towards particular political events or opinions.
#### **Leaked or Adverse Use of Information**
Leaked personal or confidential information that could jeopardise the safety or security of a voter, candidate, or party leader, and affect the result of an election.
Unauthorised access to or distribution of confidential material or information that could jeopardise election systems and infrastructure or the security of an individual.
Ransomware attacks that use unauthorised access to personal or confidential information against a voter, candidate, or party leader to impact voter choice or political behaviour.
Direct threats to or harassment and intimidation of individual voters, voter demographics, or election workers and officials to change political behaviours and outcomes.
#### **Attacks on Infrastructure**
Direct threats to online technologies and platforms or election infrastructure and systems with the intent to affect the electoral process or outcome.
Ransomware attacks to electoral infrastructural systems to leverage and influence political behaviour.
#### **Bot & Troll Networks**
Use of online bot and troll networks to alter online discourse, disrupt information instructure, or change political behaviour.
### **Vulnerabilities**
To facilitate threat detection and the identification of incidents, the incident response team will remain vigilant of vulnerabilities that could be exploited. We define vulnerabilities as the characteristics, dynamics, and complexities of the existing electoral information ecosystem that exposes it to potential harm. Some of vulnerabilities we are monitoring include:
**Political leadership** Political or leading figures that are considered to be more contentious or divisive in public opinion may be more likely to be implicated in mis- or disinformation campaigns or targets of information incidents.
#### **Public Communications**
Public communications by news organisations, political parties and organisations may lack effective planning or protocol to respond to false information, resulting in unclear, insufficient, or slow communications responses.
Resource and staff/journalist reductions due to **[recent cutbacks in major news](https://www.cbc.ca/news/federal-funding-journalists-expire-1.7116950) [organisations](https://www.cbc.ca/news/federal-funding-journalists-expire-1.7116950)** and the **[bankruptcy of](https://www.blackpressmedia.com/black-press-announces-corporate-restructuring-and-sale-transaction/) [Black Press media](https://www.blackpressmedia.com/black-press-announces-corporate-restructuring-and-sale-transaction/)** can lead to limited or delayed public communications related to information incidents.
**Digital Platforms** Lack of effective platform regulation and transparency, such as policies and programs for fact-checking, bot detection, and monitoring protocols, can lead to the underdetection and delayed reporting of information incidents.
#### **Digital Infrastructure**
Technological and digital shortfalls and inefficiencies, such as outdated electoral technologies and the inability to detect and respond to the use of artificial intelligence (AI) for adverse intent and of deep-fakes.
**Digital Literacy** Lack of digital literacy among political influencers, candidates, and British Columbians regarding disinformation and false information. This could lead to increasing polarisation and false information related to existing and important provincial issues such as housing, addiction and healthcare, immigration and refugees, and the Carbon Tax.
## **Incident Response Plan**
To ensure ongoing monitoring of incidents and response to incidents as they happen, the team follows a five step operational process:
#### **1. Detect & Assess (0 hrs)**
During **normal election operations** (periods where no incidents occur) the team scales up resources to be able to conduct ongoing media monitoring daily to be able to detect and assess potential information incidents should they arise. Events are identified as incidents on a case-bycase basis using a pre-determined set of criteria.
*Incident declaration: Factors used to assess an incident include speed of impact, extent of online engagement, nature of population affected (e.g. marginalized and/or vulnerable groups), scope of systems affected (e.g. political, economic, social, ideological), complexity, intervention required (resources needed), and/or researchability (ability to understand the incident through research).*
#### **2. Activate & Plan (within 24 hours)**
In the event an incident is identified, we activate the **incident response team**.
*Incident response team: a specialised group of experts that convene to identify, coordinate and conduct a rapid, distributed investigation for the duration of the incident response period.*
This group collectively identifies key research questions, allocates research leads, constructs a plan for sharing findings through incident updates and identifies a 1-2 deliverable timeline. Areas of rapid research are selected based on their ability to facilitate understanding the nature of the event, potential impact and/or support the identification response strategies. In addition, questions are selected to add to our ongoing election survey and additional data collection infrastructure is assessed and modified if additional data is needed.
#### **3. Notify (within 24 hours)**
Upon activation, a **public notification** is drafted that summarizes the incident.
*Incident Notification: consists of a description of the event, event timeline, key questions under investigation and subject matter experts for contact.*
Prior to public release this notification is shared with relevant stakeholders for comment. The notification is then shared via social media and via our public facing website.
#### **5. Debrief (post election)**
Following the election, the final report on the election will include **incident debriefs** for all incidents that occurred. This document will be reviewed through internal peer review, and shared online.
*Incident debrief: a summative report that summarizes any incident(s) detected, synthesizes key finding and/or observations, and identifies key considerations to inform prevention, mitigation and preparedness for future events. This may include lessons learned, emerging threats and/or vulnerabilities identified.*
#### **4. Analyze & Inform (within 24-72 hours)**
Research leads (from the incident response team) **initiate research** by collecting data, analyzing, and preparing a draft incident update addressing their research question. Incident updates are assessed through rapid peer-review, revised and published online.
*Incident Updates: insights on the information incident as they emerge, including initial incident summary (or reference to our timeline or notification), key questions being addressed, brief summary of research approach (if applicable), research findings and analysis, conclusions and recommendations.*
## **Contributors**
Contributors to this report (listed alphabetically) are: Alexei Abrahams, Thomas Bergeron, Aengus Bridgman, Esli Chan, Mika Desblancs-Patel, Danielle Bohonos, Ashley Vu, Jennie Philips, Zeynep Pehlivan, Saewon Park, Chris Ross, Ben Steel, and Junyan Zhu.
This report was made possible by the considerable work of a large number of researchers in Canada. In no particular order, Shane Littrell, Ashley Vu, Nader Hedfi, Olivia Melanson, Taylor Owen, Peter Loewen, Isabelle Corriveau, and Lee Slinger. We also thank Brian Morgan for his design work.
## **Methodology**
### **Survey**
Survey findings for this report draws upon responses from 1,005 BC-based adults conducted from August 28th to September 5th using a commercial survey panel provider. The margin of error for a comparable probability-based random sample of the same size is +/-3.09%, 19 times out of 20. All results are weighted by age and gender.
Comparisons to the nationally representative population is drawn from the Canadian Digital Media Research Network monthly tracking survey which surveyed 1,463 Canadian adults conducted from July 24th to 29th using a commercial survey panel provider. We excluded respondents from British Columbia in this dataset, resulting in 1,270 remaining respondents. The margin of error for a comparable probability-based random sample of the same size is +/-2.6%, 19 times out of 20. All results are weighted by age, gender, and region.
### **Digital Trace**
Digital trace data was collected for influential BC entities belonging to the following categories: influencers, politicians (both incumbents and candidates), news outlets, and civil society organisations. A team of research assistants worked together to identify influencers whose content revolved around key topics in BC, including immigration, housing, economy, environment, healthcare, drug policy, and LGBTQ+ issues. To find influencers, they searched on Google and Reddit threads for popular BC political influencers. They explored relevant hashtags and keywords on Twitter, Instagram, TikTok, and Youtube, such as "#BCpoli" and "#VanPoli". On X/Twitter, Instagram, and TikTok, they additionally used features such as suggested accounts, retweets, and reposts to find influencers with similar content. They recorded the entities' social media handles and ensured that they fit the requirements to be a seed. Such criteria included being Canadian, having at least 1,500 followers, not being a current political candidate (these are captured separately), being active on social media in the past year, and having at least a portion of their content focused on BC politics.
The list of politicians is composed of federal MPs elected in BC ridings, current BC MLAs, and candidates who were either registered with BC elections or announced as candidates from official party websites as of Aug 30, 2024. For this report, this list includes BC United candidates who were still listed as a candidate as of the aforementioned date. The initial list of news outlets came from [Media Cloud](https://www.mediacloud.org/)'s "national" and "state-local" datasets of Canadian news outlets. Research assistants reviewed
this list for accuracy and also supplemented it with province information for each news outlet. From the reviewed news outlet list, only the BC news outlets were included. The largest and most active BC civil society organisations were collected from [Canada Revenue Agency](https://open.canada.ca/data/en/dataset/08ae3944-1a9d-483a-a7ae-116bc58199fd)'s list of registered charities dataset as well as [Office of the Registrar of Lobbyists British Columbia](https://www.lobbyistsregistrar.bc.ca/the-registry/open-data/)'s lobbying activity reports. Lastly, during the election period, we will continue to expand the number of entities tracked and we may also collect data for specific topics or incidents.
Having identified 755 politicians, news outlets, influencers, and civil society organisations, all plausibly relevant to the BC elections, we then identified their official social media accounts across X/Twitter, Instagram, Youtube, TikTok, and Telegram. In total, these 755 entities operate 1,057 social media accounts: 541 on X/Twitter, 358 on Instagram, 112 on Youtube, 46 on Tik-Tok. We found just one account on Telegram, and accordingly excluded this platform from our analysis. We identified many accounts on Facebook, but unfortunately due to Meta's decision to shut down CrowdTangle API access in August 2024, data is not currently available for those accounts.
Having identified these 1,057 accounts, we collected all of their social media content on X/Twitter, Instagram, Youtube, and TikTok, for the eight-month period of January 1 to September 1, 2024. Collectively these 1,057 accounts posted 606,641 times across these four platforms, garnering various engagement from viewers. While each platform supports different affordances for engagement, all platforms permit viewers to like content, so we employ like count as a platform-agnostic concept of engagement. By like count we mean the number of likes garnered by a particular piece of content, such as a tweet on X/Twitter or a video on Youtube. We interpret content receiving more likes as more engaging than content with fewer likes. By extension, we treat accounts as more engaging if their content received more likes than the content of other accounts.