How Political Campaigns Use Personal Data

This introductory workshop is for researchers, activists, technologists, and anyone curious to know more about the use of personal data in political campaigns. The workshop addresses foundational information about personal data and explores its multifaceted relevance to political influence. Participants, regardless of their expertise, will engage in discussions and activities to grasp the intricacies of different data categories, real-world examples with relevance to current politics, and share experiences from their personal or professional realms, facilitating the identification of unique investigative paths.

Learning Activities

1.Opening (15 minutes)

Session aim: Introduce yourself and the presentation, have participants introduce themselves

Instructions:

• Welcome participants and introduce yourself and your team.

• Introduce the project (if the workshop is part of a larger initiative) and provide participants with some background information about how this workshop came to be and why you think the topic is important.

• Introduce the workshop objectives and agenda.

• Discuss ground rules for the workshop. Make it a collaborative process and encourage participants to suggest rules too.

• Encourage participants to note down themes, questions, people, political parties, tools, technologies and other aspects discussed during the workshop which might be relevant for them if they decide to continue investigations on the topic.

• Allow participants to introduce themselves by asking them to answer the following questions. You and your team can join in as well.

  • What is your name and what do you do?
  • What is one piece of information about yourself (other than what you’ve already shared) that we can find online?

2.What is personal data (45 minutes) 

Aim: Understand and define the meaning of personal data.

Tools / Materials: Sticky notes and pens or access to a collaborative digital whiteboard.

Preparation: 

• For the "Individual Reflection" activity in this session:

  • for an in-person workshop, ask the participants to write down their responses on a sticky note.
  • for online workshops, it is recommended to use a free online tool to collect participant responses.

• For the activity “Different Types of Personal Data”:

  • for an in-person workshop, prepare in advance a collection of statements centered around personal data to be shared with the participants (one statement on each sticky note, see examples below in the activity). Prepare copies of the same statements to share with each group, depending on your audience size. It is recommended to have 3-5 people in each group. All groups must receive the same stack of information.
  • for online workshops, create a document or shared whiteboard with the statements, with a separate area or document for each group.

• Resource to read when preparing the session presentation: Data: From Personal to Political

Instructions:

• Ask participants how they define “personal data” as opposed to “data”. Allow them a minute to share their answers.

Prepare and give a presentation focusing on the following points.

• “Personal data is any data that can be used to identify or re-identify an individual, and any large-scale data sources that are produced by the behaviours and actions of individuals”. (Source: Influence Industry Guidebook).

• “Personal data” can be used and applied in a number of situations:

  • Personal data is a broad term that can take diverse forms, including demographic, opinion and behavior data.
  • Personal data can be connected to an individual’s name, contact, or ID.
  • Personal data can also refer to anonymized statistics such as website traffic or event attendance numbers.

• This workshop will look at how political parties use personal data, digitally collected or stored, to help them create, gain or maintain political power through political influence.

• Ask participants, and allow them a minute to share, what they know about possible data collection:

  • Where is information collected from?
  • What information is possible to collect?
  • Who is doing the collection?

• What information is possible to collect? There are three main ways that personal data is collected:

  1. Data you create such as posts that you like on social media, when you fill in a form online to sign up to an event, or when you send an email.
  2. Data that others create about you such as photos that you are tagged in or being part of another person’s ‘contact’ or ‘friend’ list.
  3. Data that is inferred about you created through analysis of the trends and connections between data that you create or data created about you, and data of other people such as deciding who you are most likely to vote for based on where you live. Can also be referred to as shadow or trace data.

• Political parties can use data-driven technologies to optimise the content and format of their messages. For example, they can track which of their emails generates the most click-throughs to their website.

• Political candidates can target individuals directly through their email address and personalise the content based on any personal data they have gathered on them, such as demographic details or their previous interactions with them on social media or email.

• The practice of collecting these types of data existed before the inception of the internet including through public surveys or focus groups.

• At a very high-level, this data is collected then used as the foundation to make decisions, where we get terms like, ‘data-driven decision-making‘.

• Close the section by summarising that every internet connected device collects data – more data than we can imagine is collected and stored.

• This data can be about the person using the device, can be about another person implicated by the user, or can be about the person’s habits and behaviors.

• All of this data can be and is used by political groups in their attempts to gain political power.

[10 minutes] Instructions + Group activity

• Divide participants into small teams of 3-5 people each.

• Provide each group with a stack of sticky notes with prepared statements with different types of personal data and information (10-15 statements to each group).

• If online, create a document or shared whiteboard with the statements, with an separate area or document per group.

• Make sure to include information that captures different types of personal data explained in the next section (personality, behaviour, contact, opinion). The names, behaviours, and traits you share must be relatable and relevant to the culture and context of the participants.

• The following are sample statements you can use and adapt for this activity:

  • Jane spends 25 minutes every day commuting.
  • 11% households in Ireland bring their recycling to a recycling centre.
  • David prefers listening to classical music during his free time.
  • Katie lives at 54 Panama Way, Apartment 3D, Blue Haven.
  • Gina, a teacher in a community college, does not like to tip service staff.
  • Fifth graders in larger school districts have a keen interest in extracurricular activities such as karate and self-defence.

• Ask the participants to read each sticky note and instruct them to find three or four ways to categorize the information based on the type of data mentioned in the note. Give an example such as four of the above statements use ‘Identified name’ and two use ‘Anonymised statistics’.

[10 minutes] Activity debrief

• Back in the main group/room, ask teams to share the categories identified.

• Discuss similarities and differences between the groups’ categories.

• Ask what makes certain categories easy to identify and label across all groups.

  • For instance, if all groups had a category called ‘demographics’ or ‘opinions’, you can discuss whether these are easier to discern than other categories. 

• Ask the participants to identify categories that are unique and why other groups may have labeled them differently.

  • For instance, if one group labeled David’s taste in music as ‘opinion data’ and another group labeled it as ‘behavioural data’, encourage the groups to share and discuss why they picked these different labels.

• Discuss which statements were harder to categorize and why.

• Close the discussion by highlighting that regardless of their category, all types of personal data can be used for analysis and are valuable to political campaigns.

Instructions for presentation

• Introduce the participants to the following types of personal data which are collected, analysed, and leveraged by groups and can be applied to political influence campaigns:

  • Public opinion data - Sentiments expressed in social media posts, comments, and discussions can provide insights into public opinions on various topics.
  • Demographic data – Data collected when users register on websites or from self-reported forms – including age, gender, location, and marital status.
  • Contact data – Email addresses collected through newsletter sign-ups and website registrations.
  • Behavioural data – Information gathered through an individual’s website browsing history – including how many times the person visited the website, what pages they viewed, and how much time they spent on each page.
  • Personality data – Controversial categorisation inferring parts of an individual's personality through collected data such as psychometric profiles used by Cambridge Analytica.
  • Financial data – Information collected through online banking or credit card transactions – including data on purchases made online and offline.

Prepare: Find a relevant example for the types of data from your own context. You can see some examples we’ve prepared here

• Data collection processes are often built into the structure of various digital platforms and devices, but remain hidden from the user and are often collected without consent.

  • For example, the email you share when you sign up for a newsletter could be data that is shared and sold without your knowledge.
  • Similarly, the number of clicks on a website or likes on a social media platform can also be data that is collected and stored for years without your explicit consent.

• Personal data can be obviously political such as our voting history or if we attended a rally.

• Other non-political data such as our purchases or location can become political. This data may be used to decide our political information such as where we will vote or who we might vote for. For example:

  • Contact data can be used to reach out and create direct communication between individuals and political groups.
  • Financial data can be collected so that a political organisation can receive maximum donations.
  • Behavioural data can be collected to understand what topics people want to hear about and when.

• Close the session by summarising that:

  • data collection has been around much longer than the internet,
  • the data collection that comes from digital technologies now has a much larger scale than before,
  • different types of data can be used to understand different things about people and their personal lives.

3. Data-driven technologies and political influence (65 minutes)

Aim: Understand the use and history of personal data in politics.

Materials: The sticky notes with examples of personal data used in the activity “Different Types of Personal Data”

Preparation / materials: 

• In this session, you will be utilizing both materials and discussion points shared in sessions 1 and 2. Make sure to refer to your notes. Some parts of this session, as highlighted in relevant areas, require further reading:

  • Defining Political Influence
  • The History of Industry Mediation in Campaigns

[3 minutes] Introduction

Instructions

Prepare a short presentation addressing:

• What is influence? - Here we talk about political groups using personal data for political influence.

• Influence is the capacity to affect the opinions, behaviours and actions of others.

• Influence is different to:

  • informing – which aims to provide information without necessarily changing behaviour, and
  • regulation – which aims to change behaviour through enforcement rather than choice.

• Influence is providing information with the aim to have an effect on the behaviour or actions of an individual or a group.

[8 minutes] Spectrogram activity

Instructions

• Set up a series of slides with statements from below:

  • A political party uses accurate information about their policies to swing voters to support/vote for their candidate.
  • An environmental campaign encourages people who are interested in climate action to fly less by targeting people who they have data on to show they are frequent flyers.
  • A misinformation campaign aims to disrupt the trust in a local media outlet by posting fake news with their branding.

• Set up a physical spectrum from one side of the room as ‘this is good for democracy’ to the other side of the room ‘this is bad for democracy’

• If online, run this exercise using a poll tool choosing between the two sides of the spectrum.

• Show each slide one at a time and ask participants to stand along the spectrum or vote in the poll.

[4 minutes] Activity debrief

• Reflect as a group, during the exercise, on the different choices made by participants.

• If you have time, ask the participants to share more examples of influence, and use these examples within the spectrum or online poll.

• Note that data collection can also be applied to causes that many would categorise as a “good cause” and not just negative ideas of political influence.

[10 minutes] Groups activity: Influence choices by media

Instructions

• Separate the participants into four groups.

• Assign each group a form of communication media: TikTok, Print, Radio and Facebook.

Prepare: you can choose other communication media options that are more relevant to your audience.

• Each group must identify features of successful political campaigns carried out through the assigned media.

  • For example, if a group was assigned ‘Radio’, then the group must identify features of a successful radio campaign conducted by a political party. How long should it be? Who should voice it: male or female? What kind of tone must they use: relaxed, persuasive, humorous or passionate?
  • Similarly, if the group was assigned ‘Print’: What kind of print media would be ideal? What images should they use? On which page/section should it be printed?

• You can also encourage the participants to share an example of a successful political campaign conducted in the medium if possible.

[10 minutes] Activity debriefing

• Each group must then present (1 - 2 minutes each) the medium assigned to them and share their findings with the other groups.

• Discuss the different group’s work together.

• Encourage participants to think about how this campaign would be different compared to those conducted on other mediums.

  • For example, how is a radio campaign different from one on social media?
  • Ask participants who they think the political party would need to work with to create those adverts.

• Conclude by mentioning that political groups, candidates, and civil society have always adapted their influence strategies to new technologies. Emphasize how political campaigning communications, including the actors involved in developing the public influence strategies of politicians and political parties, have changed with changes in technology.

Instructions

Prepare a presentation including:

• Watch: Political Persuasion Trailer (English)

• Most campaigns use several strategies, tactics and tools to create influence.

• The scope and impact of political influence are often enhanced with the support of professionals.

• Politicians and other campaign strategists engage a variety of services and consultants, frequently sourced from specialized companies with significant expertise in human and technological aspects of data collection, processing, and targeted commercial outreach.

• Political groups now work with commercial marketers, advertisers and other private sector companies to apply their expertise to political campaigns and to shaping a candidate's / group's image.

• There are many companies working in each country, sometimes with one political party and sometimes across all of them. Some of the companies also work across more countries.

Prepare: You may be able to find examples from your country or, if you have time, ask the participants if they know any. You can find examples by country or region in this Influence Industry database.

[10 minutes] Activity: Political ad discussion

Instructions

• Invite participants to identify and share an example of a political advertisement. They can paste a link in the chat (if online) or send links to the trainer to project them on a screen (if offline).

• Ask participants to identify the different professionals who could have been involved in producing this advertisement. They could be: script writers, voice artists/actors, animators, graphic designers, directors, translators, editors, etc.

• The following are some prompts to facilitate a discussion with participants based on some of the identified examples above:

  • Who do you think was part of creating this advertisement?
  • How long do you think it took for them to create this?
  • What are some key decisions they made (type of media used, length of advertisement, language used etc) and why were these decisions made?
  • Do you think the people who created this advertisement only make political adverts? Or would they also be involved in advertising commercial products and services?

• Conclude the activity by emphasizing how data-driven technologies have created a new set of professionals employed by political groups to manage their communications.

[10 minutes] Presentation and discussion

Instructions

• Follow up on the previous discussion to inform the participants about the different kinds of professionals who are involved in the process of converting personal data into political data. For example:

  • Professionals involved in the conversion of personal data into political data include data analysts, data scientists, and data engineers who gather, process and analyse data.
  • Compliance experts ensure compliance with data protection laws, while political analysts interpret insights for political context.
  • Information security specialists and ethical hackers safeguard data, and software developers create necessary tools.
  • Communication experts translate complex data, and policy analysts assess implications and recommend regulations.
  • Campaign strategists use data for election strategies, social media analysts monitor online conversations, market researchers collect political preferences, and data privacy officers oversee compliance with privacy laws.

• Highlight that these experts don’t just play the role of controllers, but also have qualitative input into the content, frequency and style of messaging that happens through these channels.

• Share some of the key functions these professionals play in the political process. The following are some examples, but encourage participants to share their own:

  • They control and manage the use of the technologies
  • Mediate between citizens' voices and political groups
  • Provide strategy and direction for how to engage with technology

• These services are often referred to as the “influence industry” defined by Tactical Tech as the collection of private firms who sell their services (including collecting, analysing and trading of data) to candidates or parties who want to influence political opinions particularly during, but not limited to, election campaigns.

4.Targeted political influence (40 minutes)

Aim: Explore the strategic use of data-driven technologies to present information to target groups to influence their opinion.

Preparation / materials: 

• Prepare your slides ahead and include key definitions, examples, and sources in your presentation to share with participants when necessary.

• Throughout this session, you will need to share technology examples. You can find several examples for the corresponding technologies in the guide Data as Political Influence.

• For the activity ‘Location data as insight’ you will need several pieces of A3 paper or an online whiteboard.

Instructions:

Prepare a presentation including the following points:

• Different communication channels are employed to target users with advertisements and content that are intended to directly appeal to them as the content is based on their preferences.

• Data-driven technologies have changed channels, tactics, and tools of influence.

• While there is debate as to whether the impact of the ads is successful, political groups invest money and time into data-driven contact with voters.

• We often think of Facebook adverts, but there are other channels used for such influence.

• Ask participants to talk about some of the places where they’ve seen adverts already covered in the workshop (using examples from previous activities).

Prepare: Add more technologies that the participant’s don’t cover themselves, and include real-life examples from their context. Some tools you can cover or add include:

• Robocalls, or a robocalling service, automatically dials a list of phone numbers in order to deliver a pre-recorded message or, in more technically advanced scenarios, even conduct a live call. (Source "Robocalls and Mobile Texting: Automated campaign outreach").

• SMS: Campaigns can also use messaging platforms such as peer-to-peer SMS and bulk texting to initiate conversations between campaign volunteers and voters, as well as to administer surveys and polls, and to encourage people to vote. (Source "Robocalls and Mobile Texting: Automated campaign outreach").

• WhatsApp, Telegram or other chat apps are used to share information about upcoming events or to encourage people to share information with friends.

Instructions:

Prepare a presentation including the following points:

• Data-driven influence is often much more targeted and personalised than influence through traditional means such as newspapers or billboards.

• The difference between “general” ads and targeted ads:

  • Targeted ads, at their core, are a personalised form of advertising that should reach audiences based on their preferences and behaviors.
  • General, traditional ads, cast a much wider net and are shown to a broad, non-select audience.
  • Targeted (including micro-targeted) ads use collected, processed and analysed data to divide broad audiences into smaller groups based on data collected from their daily routines, assumed personality traits and online behavior.

• Where might people see targeted ads? Have the participants seen any supposed targeted ads?

  • For example: "As a person who practices yoga every morning on YouTube and likes several yoga pages on social media, I have noticed several ads for brands that I’ve never interacted with about yoga clothes and products."

• Ask the participants - “Can you recall a moment when a targeted add felt (too) personal, making you question the level of information about you that is out there?”

  • NOTE: as this is a personal question, not all participants may feel comfortable sharing the answer. Encourage responses only if participants volunteer. You can share some general examples if there are no volunteers, like getting an advertisement for a product that you bought once.

• Ask the participants to consider how this targeted content also exists in political campaigns.

• Ask participants - "How do tailored political messages influence our opinions, shape our narratives and potentially impact democratic processes differently to general ads?" - elicit some answers from the audience.

[2 minutes] Brief presentation

Instructions:

Share an introduction / slides with the following:

• Ads can be targeted based on location through a process called “geo-targeting”.

• This method of targeting ads involves tailoring messages based on a user’s location.

  • For example: when you connect to a coffee shop wi-fi or are in the vicinity of a coffee shop, you might be served ads with coupons for that coffee shop.
  • Or a political example, certain neighborhoods might see certain political ads while other neighborhoods see others.
  • During a political conference such as COP20 (The United Nations Climate Change Conference) people in the area will be targeted with relevant ads, for instance on environmental policies.

[10 minutes] Individual activity: Location data as insight

Instructions

• Place participants in small groups of 3 – 4.

• Provide them with a piece of paper each or an online whiteboard.

• Set a timer for 5 minutes:

  • Ask each group to create a list of locations they have all visited this week. For example: home, coffee shop, work, gym, different countries, airport.
  • Next, ask the participants to draw a fake new map putting all these locations together on one page.
  • If online they can create a chart diagram of someone’s journey throughout a few days.

• After the first 5 minutes, set another 5 minute timer for the second half of the activity:

  • Ask participants, what might people be able to guess or assume about you through these locations?

  • Some points that you can suggest include:

    • Demographics – neighborhood could suggest income level
    • Interests and hobbies – places you visit in a week
    • Place of work – where do you go during workdays?

[5 minutes] Activity debrief

• Based on the assumptions from the last task, what types of ads would participants - if they were a politician - choose to serve to the devices in the area?

• Some optional suggestions include:

  • Television ads or social media ads around a gym, where people might either watch TV or check social media.
  • Billboards on streets that lead to popular office complexes.
  • Targeted adverts on any mobile app games that are based on their hobbies of interest.

[3 minutes] Brief presentation

Instructions:

Brief presentation emphasizing the following:

• Campaigns have long practiced basic geotargeting by treating swing districts and stronghold districts differently.

• Some of the most common geotargeting methods and technologies include:

  • Geofencing: creating a virtual perimeter around a point of interest to promote a message only to individuals inside that area. Geofences can be cast around individual buildings or around areas with a radius of several miles.
  • IP targeting: collecting location-based information from IP (Internet Protocol) addresses and targeting messages based on IP location information such as on your laptop and browser.
  • Mobile geotargeting: targeting political messages to specific mobile phones in certain locations via digital ads.
  • Property geotargeting: targeting political messages on social media or printed leaflets to specific post codes.

• Location data is powerful tool in campaigns especially when combined with assumptions and interpretations.

  • For example: A post code might suggest an income level and a political persuasion. Coupled with which supermarket you frequently shop at, this might suggest your feelings on certain political topics.

Prepare: Find two or three examples to highlight the use of geotargeting by political groups relevant to you own context. You can see some examples in this article: "Geotargeting: The Political Value of Your Location".

• Ask participants: "Can you think what other forms of targeting might exist?"

• Use this moment to point at what can be covered in a next possible workshop, such as the related module "Understanding Political Data as an Asset and Intelligence"

5.Reflection and closing (15 minutes)

Aim: Summarise the key points and ask participants to critically reflect on what they’ve learned.

Presentation instructions:

• Summarise the workshop points at a high-level. Some key points can include:

  • How useful data transforms based on it’s analysis and application.
  • Digital technologies and their use in the current media system encourages the collection and increasing use of personal data to inform political communications.
  • Data points are used in decision making for who sees which ads and where.
  • This is not a new phenomenon, technology has always impacted political communications, and data has always been collected before technology made it easy.
  • The scope, scale and speed of digital technology has increased the amount of data available.
  • The hidden nature makes these new methods different as the average user and voter is unaware of the collection and impact of the data.

• Ask participants what they think about the technologies that you have discussed so far: Do they see any problems or ethical dilemmas with these tools?

• Some considerations can include:

  • Increased segmentation and profiling – these technologies can intensify the segmentation and profiling of the electorate,  the incentive of which might lead to privacy concerns
  • Misinformation and fake news – unregulated or unsubstantiated messages or scams can spread widely.
  • Nefarious use – the data in the wrong hands can send threatening, untrue and offensive messages to groups of people.

Instructions

Ask the participants the following questions to understand their experience with the workshop. They can share it verbally or write it down on a sticky note/digital whiteboard. If time is short, not all participants need to answer all questions.

• What was the main takeaway from this workshop?

• What was one completely new thing you learned from this workshop?

• Which of these topics are you curious to learn more about and investigate further?

• Were you able to identify any topics or themes that seemed relevant to your country or region?

• What does the local legislation in your country say about personal data and the protection of privacy? What do you think about these legislations?

• What is the conversation surrounding the political use of personal data like in your country or context? Do you think the average voter is aware?

Thank the participants for their cooperation, enthusiasm, and interest.

Contact Us

Please reach out to us at Exposing the Invisible if you:

• have any questions about this workshop plan and facilitation guidelines,
• use this workshop plan and want to share feedback and suggestions that can help to improve them,
• adapt the workshop plan to a specific context and want to share the results with us,
• want to suggest new activities, tips or examples that can be added to this workshop,
• want to share your expertise and collaborate with us on developing and testing new workshops.

Contact: eti@tacticaltech.org (GPG Key / fingerprint: BD30 C622 D030 FCF1 38EC C26D DD04 627E 1411 0C02).

Credits and Licensing

CC BY-SA 4.0

This content is published by Tactical Tech's Exposing the Invisible project, and licensed under a Creative Commons Attribution-ShareAlike 4.0 International license

• Workshop authors: Amber Macintyre, Cassie Cladis, Dharini Priscilla The Influence Industry Project, Tactical Tech.
• Editorial and content: Christy Lange, Amber Macintyre, Cassie Cladis, Laura Ranca, Jasmine Erkan
• Graphic design: Exposing the Invisible, Tactical Tech
• Website development: Laurent Dellere, Saquib Sohail
• Project coordination and supervision: Christy Lange, Amber Macintyre, Laura Ranca, Wael Eskandar, Jasmine Erkan, Cassie Cladis, Marek Tuszynski, Safa Ghnaim.

This curriculum was published as part of the Collaborative and Investigative Journalism Initiative (CIJI) co-funded by the European Union. 

Disclaimer:

Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Education and Culture Executive Agency (EACEA). Neither the European Union nor EACEA can be held responsible for them.