The truth about voter turnout: What can we actually learn from over 100 voter turnout experiments?
The truth about voter turnout: What can we actually learn from over 100 voter turnout experiments?
Want to know what really works in campaigning? Campaign Lab has launched an exciting new series bringing cutting-edge academic research directly to campaigners, and our first session with LSE's Associate Professor Florian Foos did not disappoint. Drawing on a massive analysis of 122 individual level Get Out The Vote (GOTV) experiments across multiple countries, Foos challenged some of our most basic assumptions about campaign effectiveness.
The research tackled crucial questions that every campaigner grapples with: Which voters should we target? What methods actually work? And can we really apply US campaign tactics in European contexts? The answers might surprise you.
First up - voter targeting. Conventional wisdom often pushes campaigns to focus on either high-propensity voters (they're easier to reach!) or low-propensity voters (bigger potential impact!). But the data tells a different story. The “sweet spot”? Those middle-propensity voters, with a 40-60% likelihood of voting. These voters are both reachable and responsive - the campaign equivalent of striking gold.
When it comes to outreach methods, the study delivered some fascinating insights for GOTV. Email campaigns? Not effective. Door-to-door canvassing? It's complicated. While US campaigns see an average 1.5 percentage point boost in turnout from door-knocking, European results are more mixed. This highlights a crucial point: we can't simply import American campaign tactics wholesale.
Notably, phone banking showed better results in Europe than the US, while significant variations appeared between countries like Denmark and the UK. This reinforces that we can't treat Europe as a single campaign context - each country needs its own evidence base.
Age demographics threw up some interesting patterns too. While younger voters proved harder to reach, they were actually quite responsive when contacted. The most consistent results came from the 55-75 age bracket, suggesting this might be where campaigns can get the most bang for their buck.
But perhaps the most intriguing finding concerns text messaging. Despite limited data (only six experiments), texting showed impressive results and potential cost-effectiveness compared to traditional phone banking. However, there's a catch - getting those phone numbers in the first place, especially with GDPR regulations, remains a significant hurdle.
The research highlighted crucial gaps in UK campaign data quality compared to the US. Without party registration data and comprehensive voter information, UK campaigns face additional targeting challenges. However, this also presents an opportunity - improving data collection and targeting capabilities could significantly enhance campaign effectiveness.
Social media impact emerged as a key discussion point. While direct ad campaigns show modest results, the research suggests we need better ways to measure broader digital impact - from algorithmic changes to organic content reach. Traditional metrics may not capture the full influence of social media on voter behavior.
Key takeaways for campaigners:
1. GOTV efforts are most effective with "middle propensity" voters - those with a 40-60% likelihood of voting. While high-propensity voters are easier to contact, they're less responsive to outreach. Low-propensity voters are more responsive when reached but harder to contact. This suggests campaigns should focus resources on middle-propensity voters for maximum impact. It also means progressives should accurately record turnout data and ensure marked registers are requested and entered to keep this targeting up to date!
2. Door-to-door canvassing shows mixed results between US and non-US contexts. While highly effective in the US (1.5 percentage point increase in turnout), door-to-door efforts showed limited impact in European countries. However, more UK-specific research is needed since current data is limited. We need more doorknocking experiments - if you’re interested in running one - get in touch!
3. Text messaging appears highly effective for GOTV, potentially more cost-effective than phone banking. However, the challenge lies in obtaining phone numbers, especially under GDPR regulations. This suggests campaigns should prioritize collecting mobile numbers, particularly from middle-propensity voters, while building compliant data collection systems. If you’ve got ideas for phone number collections - get in touch!
This research underscores the importance of evidence-based campaigning and the need for more UK-specific studies. Campaign Lab's new series promises to keep bringing these vital insights directly to campaigners, helping bridge the gap between academic research and campaign practice.
This session marks the beginning of our commitment to bridging the gap between academic research and campaign practice. Join us for our next session on February 21st, focusing on relational campaigning with Evie Monnington-Taylor, Director of Research at Vote Rev (sign up here).
Interested in collaborating on UK-based experiments? We're actively seeking partners to build a stronger evidence base for what works in UK campaigning. Get in touch by emailing douglas@campaignlab.uk
-------------
All of our seminars will be available to watch in full on our youtube channel - watch Florian's analysis below: