Final stats of the NewsTrust Baltimore pilot
Now that the NewsTrust Baltimore pilot has ended, we are publishing our final reports about our local news experiment. For the past six months, we have provided a free online service to help local residents find good journalism about their city -- and become more discerning news consumers in the process, thanks to the support of the Open Society Foundations.
Be sure to check our other reports on our Baltimore pilot, which include personal observations from our staff, an analysis of the Baltimore news ecosystem, our editorial report, educational and community report and survey results.
For more information about the wrap-up of this project and about NewsTrust's new directions, check out this blog post.
Here are the final statistics from our six-month NewsTrust Baltimore pilot, from Jan. 31 to July 31, 2011 (see also our early findings from our first three months). These results suggest that our local news experiment was well-received, with encouraging levels of participation from the Baltimore community. Here are some of the highlights.
Web traffic: How many people checked out the site?
A total of 21,048 people visited NewsTrust Baltimore during our six-month pilot, from Jan. 31 to July 31, 2011, according to Google Analytics. This exceeded our goal of 20,000 unique visitors for that period. Over the six months, we averaged 3,913 monthly unique visitors to our site, which is more than 10 percent of a typical month's traffic on our national NewsTrust.net site.
This timeline from Google Analytics shows daily visitors to the NewsTrust Baltimore site during the project. Overall, visits to our site were steady, typically higher at the start of the week and lower on the weekends. We observed high traffic peaks after launch (4,415 unique visitors in February), as well as during our educational programs (4,218 visitors in April), and traffic trailed off in our last month, when we cut back on community engagement activities (3,382 visitors in July). This level of traffic seems to be on par with other independent nonprofit news sites in Baltimore.
These traffic statistics for our local website do not include the additional traffic from our widgets, which reached about 115,000 monthly unique visitors during the period, on average. This additional traffic was largely generated by widgets on The Baltimore Sun's local news pages, as well as on independent sites like the Baltimore Brew. Because these widgets were typically placed toward the bottom of the pages on our partner sites and not promoted editorially, they did not generate much click-through, though we believe they increased awareness of our project.
Throughout the six-month pilot, people viewed 140,146 pages on our site, or an average of 3.58 pages per visit, for an average of 4 minutes and 18 seconds. We view these numbers as a positive indicator of participation, suggesting that visitors were taking the time to read and engage with the stories we listed on our site.
Here are the main sources of traffic to NewsTrust Baltimore, according to Google Analytics:
Member stats: Who used NewsTrust?
Of the 21,408 people who came to NewsTrust Baltimore during the pilot, 535 signed up and became NewsTrust members.
These members can be broken down into these groups:
Adults 18-34 64%
Adults 35-49 21%
Adults 50+ 16%
High school only 39%
College graduate 26%
No experience 25%
1-4 years 40%
5 or more years 35%
Note that these demographics are not necessarily representative of our entire membership or visitors to our site but present a reasonable estimate of the most active participants. The high ratio of members between 18 and 34 is largely due to our partnerships with universities like Towson University, and many journalism students indicated that they had 1-4 years experience.
Of the people who signed up, 62 percent reviewed a story on NewsTrust, while 38 percent did not, based on information from our internal SQL database.
An interesting way to visualize who did and did not review is to break down reviewers and non-reviewers separately. The colorful chart on the left is the breakdown, by group, of the 65 percent of members who reviewed stories. The gray chart on the right breaks down by group the 36 percent who did not review:
Content stats: What kinds of stories did we find and review?
Over the six months, members read (or at least clicked on) 7,550 news and opinion stories. That includes stories that came from our RSS feeds, as well as the 5,382 stories that were manually posted by members. Of these 5,382 stories, 88 percent were posted by NewsTrust staff, as we had expected based on prior experience with local news hunts.
Of all those stories, 1,619 received a rating from 3,503 reviews. Here's how the reviews broke down by regular members, members who received Trusted Member status, and staff:
We reviewed a range of stories across a wide variety of topics. Here are the top 5 most-reviewed stories from the first three months of the pilot:
- "Supreme Court Rules For Military Funeral Protesters"
- from the Associated Press (on WJZ site) -- 51 reviews
- "Made (once again) in America"
- from The Baltimore Sun -- 46 reviews
- "On the Trail of Addiction"
- from Urbanite -- 43 reviews
- "More renewable energy policies aim to save money and environment"
- from Maryland Reporter -- 37 reviews
- "The Sun Also Rises"
- from Urbanite -- 21 reviews
(Note: excluded from this list are several stories we asked all students to review for news literacy tests.)
Our team wrote 52 blog posts on the NewsTrust Baltimore Blog, providing informative summaries of the nine news hunts we held in the first three months of the pilot, as well as other important events, accomplishments and milestones. These activities and campaigns produced some positive results; 117 participants advanced to a higher member level, and we identified 58 Trusted Members, who earned the trust of our community through their thoughtful reviews. We also certified 62 student reviewers and gave three superlative awards for their contributions to the site.
Those are our quantitative stats for the NewsTrust Baltimore pilot. In our next post, we'll share a final overview of what we learned from this pilot. Stay tuned for more later this month.
-- by Jon Mitchell and Fabrice Florin