Presidential Travel to Change the Narrative: Recorded Future and Foreign Media Analysis

Posted: 24th November 2010
Presidential Travel to Change the Narrative: Recorded Future and Foreign Media Analysis

President Obama recently completed a 10 day trip through Asia, which began only a few days after the Democratic Party suffered significant losses in the Congressional midterm elections.  Historically, Presidents have used foreign travel such as this as a way to turn the page on unfavorable domestic news, as they result in the media writing stories detailing the full aura of the Office of the President.  In addition, trips such as these cause out-sized coverage by the foreign press leading to a boost internationally.  With Recorded Future, we can analyze both the domestic and the foreign media to calculate exactly how successful President Obama’s trip has been.

To set up the model, we use Recorded Future to find the dates of the trip.  A query for “Barack Obama — Person Travel — last 30 days” lays out the results on a timeline:

After the November 2nd elections, we see President Obama traveled through Asia from the 5th through the 15th.  We can even generate a KML file showing the places he visited and view it in Google Earth.

We then extract all Obama mentions using the Recorded Future API from 8 weeks before the trip through one week after.  For each instance, Recorded Future calculates both a positive and a negative sentiment score, which is based on the vocabulary used to describe the instance.  We organize them by published date across four time periods: from September 10th through November 3rd (instances occurring far before the trip), November 4th and 5th (during the lead up to the trip), November 6th through the 15th (the trip itself), and the 16th until the 22nd (the week post-trip).  We then sort them by source country and average the positive and negative score for each time period based.  The results are below.

Domestic vs. International Press Coverage (Positive and Negative)

The above chart shows both average positive and negative sentiment scores per mention of President Obama in the US press and in the foreign press.  We find that while there was a big spike in how positively the US media reported on President Obama while he was on his trip, it had no lasting impact, and both the positive and the negative score for the week following the trip are nearly exactly where they were for the weeks leading up to it.  Average domestic positive sentiment rose on November 4-5 from the weeks prior (as the story changed from the Congressional losses to the overseas trip) and then rose nearly 50% while the President was overseas.  The White House succeeded in changing the story and recasting the President in a positive light, at least for a little while, before returning home to the average daily sentiment.

Internationally, the trip caused foreign media to be both more positive and more negative.  Average positive sentiment is now 17% higher than before, but average negative sentiment is 27% higher though still negligible.  As will be discussed, this is because of the mixed results achieved by President Obama on the trip.

Positive Sentiment on Obama Instances in Countries of Interest

We can get a more detailed analysis by breaking down the foreign media by country.  This analysis of positive sentiment shows that in these countries of interest, the media on average covered the President with more positive sentiment after his trip than before.  Of the countries where this is not true, Japan and Indonesia had unnaturally inflated positive sentiment scores earlier this month due to the President soon visiting their country, and Russia has had a fall-off because of problems President Obama has been having in getting the Senate to ratify the US-Russian START II treaty.

Looking at the average negative sentiment per story by country, the trend is one of slight rise over time, although the results are much more scattered than with positive sentiment.  These results reflect the general attitude that President Obama did not achieve everything he had hoped to do while overseas.  This especially involves a perceived failure to exert American influence at the G20 meeting to both control China’s currency manipulation and to procure a Free Trade Agreement with Korea.  The simultaneous disappointments caused the average negative sentiment across all Obama instances, foreign and domestic, to rise about 10%.

Anything concerning India also concerns Pakistan, and a visit to one country but not the other could generate grievances.  Pakistan experienced a spike in negative sentiment in the days leading up to the trip, but by the end of the trip it had fallen below pre-trip levels and its positive sentiment had risen to the highest of the countries surveyed.  This is because prior to the trip President Obama promised to travel to Pakistan some time in 2011.  We can see this laid out on a Recorded Future timeline of President Obama’s future travel plans.

Looking at the results for sentiment, some anomalies stand out.  Russia’s average negative sentiment spikes right before the trip, China’s spikes right after, and Japan’s positive sentiment spikes right before the President’s arrival.  The Japanese boost is understandably about President Obama’s soon-to-happen visit.  To understand the other results, we can dive further in to the data by looking at source type within each country.

Negative Sentiment
_Media Source_ _Media Type_ _Through Nov 3_ _Nov 4-5_ _Nov 6-15_ _Nov 16-22_
United States Blog 0.028 0.047 0.033 0.020
United States Mainstream 0.030 0.028 0.026 0.025
United States News Agency 0.026 0.028 0.035 0.029
China Blog 0.015 0.000 0.013 0.019
China Mainstream 0.010 0.000 0.011 0.113
China Niche 0.000 0.000 0.001 0.000
Russia Blog 0.045 0.000 0.096 0.000
Russia Mainstream 0.014 0.000 0.000 0.034
Russia News Agency 0.044 0.106 0.047 0.015

The pre-trip spike in negative sentiment in Russia happened entirely in the in news agency sources.  Diving deeper, we see a series of negative instances still discussing the election results and what they mean for US-Russian relations in regards to the START treaty.  The negative sentiment calculated by the system towards President Obama in these stories was not found over the same time period in similar stories in the Russian mainstream media or in Russian blogs.

The spike in Chinese negative sentiment following the President’s trip happened mainly within the Chinese mainstream media, which was by a factor of six more negative when discussing President Obama than than the Chinese blogosphere at the same time.  These stories in the mainstream media centered on the currency manipulations and potential round of quantitative easing by the US Federal Reserve.  Chinese blogs did not nearly have as significant a spike in negative sentiment when discussing the same event.

By examining media based on foreign vs domestic, based on source country, and based on source media type, we can get a unique insight into how the world is covering an entity.  With Recorded Future’s positive and negative sentiment calculations, we can say that the White House was successful in changing the story of their midterm defeat, but the success was temporary.  The world still writes about the President much more positively than negatively, and the President received better coverage in our biggest rivals’ blogs than in their mainstream media sources.  These insights are just the beginning of what we can find; stay tuned for more examples using Recorded Future for foreign media analysis.

Interested to learn more about OSINT analysis using Recorded Future? Check out our web site or contact us.