Learn more about the on-going projects StoryBench is tackling.

2020 Election Tracker

  • Reopen schools narrative spreads across shadowy local news sites
    Americans are divided about reopening schools this fall and polls show that this disagreement falls along party lines with Republicans more in favor of children returning to schools. Which is why it wasn’t surprising when we found a pro-school reopening narrative popping up this summer on a large network of local and business news websites
  • How news media are setting the 2020 election agenda: Chasing daily controversies, often burying policy
    It’s a paradox of examining political coverage. Are news media just reporting what the political candidates are talking about? Or does political journalism really set the agenda by selecting stories around specific news items, scandals and issues du jour? Our topic analysis of ~10,000 news articles on the 2020 Democratic candidates, published between March and
  • Facebook continues to run political ads with misinformation
    Facebook claims it has enacted measures to stem the spread of misinformation on its platform since the 2016 election. But the platform seems to be applying a different set of standards to politicians. Earlier this month, Elizabeth Warren’s campaign released an ad on Facebook falsely claiming that Mark Zuckerberg had endorsed Donald Trump for re-election
  • Even before Ukraine news, Joe Biden was the 2020 candidate with the most media coverage
    Former Vice President Joe Biden has dominated the mainstream media’s coverage in the last month, even before the whistleblower’s complaint was reported in The Washington Post, prior to the Wall Street Journal suggesting that President Donald Trump urged the Ukranian President to work with personal Attorney Rudy Giuliani, and before the White House released the

Reinventing TV

Climate Journalism Lab

  • How the Allen Coral Atlas is mapping and monitoring coral reefs worldwide
    Although coral reefs occupy less than one percent of the ocean floor, their importance extends well beyond their size. The National Oceanic and Atmospheric Administration estimates that 500 million people survive on coral reefs for income, and their economic value in the U.S. is estimated at $3.4 billion each year. More importantly, healthy coral reefs
  • How The New York Times visualized racist historical redlining and urban heat
    In late August, The New York Times published “How Decades of Racist Housing Policy Left Neighborhoods Sweltering,” a visual reckoning on historical redlining, persistent racial inequality and who feels increased urban heat across U.S. cities today. The project, led by Times climate graphics editor Nadja Popovich and Times climate reporter Brad Plumer, opens with a
  • How Circle of Blue investigated freshwater issues in Texas
    While 70% of our planet’s surface is covered in water, only 3% of it is fresh water – and most of that is packed away in glaciers and polar ice caps, out of human reach. The limited freshwater resources that we do have access to — groundwater, lakes, rivers and ponds — are shrinking under
  • Five ways organizations are visualizing carbon emissions
    Climate change is a global issue and scientists and journalists alike are increasingly making climate-related data more visible and accessible. Understanding the major causes and contributors to emissions of CO2, the most prevalent of greenhouse gases, through data visualizations can help corporations, governments and the public address areas in need of reduced environmental impact. Here

Data Journalsim in R

  • How to calculate a rolling average in R
    Rolling or moving averages are a way to reduce noise and smooth time series data. During the Covid-19 pandemic, rolling averages have been used by researchers and journalists around the world to understand and visualize cases and deaths. This post will cover how to compute and visualize rolling averages for the new confirmed cases and
  • Update: How to geocode a CSV of addresses in R
    This post is an update from the previous post, “How to geocode a CSV of addresses in R”. We will be using the ggmap package again, and be sure to investigate the usage and billing policy for Google’s Geocoding API. The API now has a pay-as-you-go model, and every account gets a certain number of requests free per month, For
  • How to use hierarchical cluster analysis on time series data
    Which cities have experienced similar patterns in violent crime rates over time? That kind of analysis, based on time series data, can be done using hierarchical cluster analysis, a statistical technique that, roughly speaking, builds clusters based on the distance between each pair of observations. Basically, in agglomerative hierarchical clustering, you start out with every
  • Diagnosing the accuracy of your linear regression in R
    In this post we’ll cover the assumptions of a linear regression model. There are a ton of books, blog posts, and lectures covering these topics in greater depth (and we’ll link to those in the notes at the bottom), but we wanted to distill some of this information into a single post you can bookmark