Two major updates at the Tylt (largely occurring behind-the-scenes) are changes to both the duration of debates and our ability to re-launch trending topics — allowing us to gauge public sentiment over a longer period of time. And it’s something we’re excited to share with you today.
Taking a deeper look at news trends is the impetus behind one of our new editorial experiments, where we’re polling users on Donald Trump’s Twitter behavior in our first ever year-long debate. This debate is tightly controlled, from the use of identical editorial language to persistent hashtags. But it has its origins in a series of debates we’ve run on another controversial topic, the Dakota Access pipeline.
We’ve published variations of the Dakota Access pipeline (DAPL) debate five times in the past six months, tied to news pegs related to construction or lack thereof. This initially began instinctively; editors were just trying to craft debates around trending news topics and we were seeing considerable chatter on the issue.
But after the first two debates, we began to realize that it could be incredibly valuable to look at how people feel about a particular subject over time… something we’ve talked about internally, but never executed until recently due to a few technical hurdles. Those are now cleared, but the DAPL debates still provide the clearest insight on what that could mean for future Tylt debates.
What have we learned from all of this? A good amount. The data indicates people are willing to talk about the same topic more than once, which was one of our earliest fears when we began to relaunch debates. But what we’ve seen is velocity and news saturation can move the needle in regards to voting behavior over time — take a look at this chart comparing debate results and Google Trends.
The only real outlier here was our October debate, which makes sense when you look at the political climate in context: it was the first time the Dakota Access pipeline became a legitimate campaign issue for both Hillary Clinton and Donald Trump.
While defenders of the Sioux tribe have been mobilized on this particular issue for months (as reflected in our data), October was really the first and last time the issue became a topic of interest for Trump’s highly enthused and mobilized base. That seems to have tailed off significantly following the election, as indicated by the resurgence of support for the Sioux and a sharp decrease in interest via Google Trends data.
This only begs the question, how will sentiment change on a topic where cultural norms shift over a period of time? And that’s why we’re so excited about the ability to track debates over months or even years.
The debates also pose unique challenges for the Tylt and our community, especially as we move forward with longer and repeat debates over a period of time.
For starters, each debate gives us a lot of information on influencers, but mobilizing all those influencers in subsequent debates still remains a challenge we’re eager to tackle. We’re also tinkering with how to let users know when community consensus has changed from debate-to-debate.
The DAPL debates are messy experiments in comparison to what we have planned in the future. But they give us a good foundation to see what works, what doesn’t and what impact the news cycle has on reoccurring debates. We’re lucky to have a top-notch staff that’s working on new ways to visualize your opinions over a period of time. And we are stoked to share more with you in the coming weeks, after we collect some more data on our first ever year-long debate.
Because, as always, we are obsessed with making sure your opinion counts.
