Plane crashes and web site traffic

Summary
AirSafe.com, which has been in operation since July 1996, provides the aviation safety community and the general public with useful information about aviation safety and security. The site highlights a particular class of events, which it defines as significant events, a category that typically includes events involving airliner deaths or other aerospace events that attract significant amounts news media attention. These significant events all all listed in one or more places on AirSafe.com, but at a minium are listed on summary pages that list the significant events from a given year.

Between May 2006 and November 2015, the period covered by this study, there were a total of 269 events with significant traffic spikes on AirSafe.com, specifically days where the estimated number of visits to the site exceeded the average number of visit in a recent 28-day period, by at least two standard deviations. While many of these spikes occurred within seven days of widely reported aviation-related events, these spikes can occur months or even years after an event, and some can't be associated with a particular event.

Using a combination of Google Analytics data about web site traffic, along with insights gleaned from Google search results, traffic behavior implied the following trends over this over nine-year study period:

• 56.1% of the spike events were associated with AirSafe.com-desigated significant aviation events that occurred during the study period.
• 62.5% of the spike events were associted with AirSafe.com-desigated significant aviation events that occurred either before or during the study period.
• 68.8% of the spike events were associted some identifiable aerospace-related event.
• Celebrities were associated with 21 of the 269 spike events (7.8%), yet only two of those spikes were due to significant events that occurred during the time period coverd by this study.
• The 269 spike days were not distributed evenly during the week, and the null hypothesis for a distribution of spike events independent of the day of the week was rejected.

Why study traffic spikes?
One of the reasons that AirSafe.com exists is to call attention to avation-related events that may have an impact on airline safety or on the public's perception of airline safety, so understanding what kind of events lead to significant traffic spikes will help to guide future content development. Particularly close attention is given to what the site calls significant events, which include a wide range of events, including many that have nothing to do with airline flights. Given that AirSafe.com is a web site, and one of the goals of this, or any other, web site is to generate traffic, the decision to add content is based in part on whether the content will generate traffic.

As a result of this study, is clear that some assumptions, for example having an event associated with a celebrity may attract more traffic than the typical aviation safety event, appear to be justified. On the other hand, given that only around half of the spike events were associated with events that took place during the time frame covered by this study, there is room for improvement when it comes to predicting what events will lead to high levels of traffic.

Additional research
AirSafe.com invites anyone who is interested an opportunity to review this study, and to use data from the study to conduct additional research. Below are links to the full study, including:
AirSafe.com raw traffic data 2006 - 2015 (sessions.csv)
Spike days details (spike_days.csv)
R code used for the analysis
Completed analysis (HTML)
Completed analysis (PDF)
Completed analysis (on RPubs)

Follow on study with heat maps
The spike study had a follow on study that focused on displaying the same key outcome data from the first study, but in the form of a heat maps. These heat maps are a visual representation of the same general temporal patterns seen earlier, particularly the observation that a disproportionate portion of all spike days occur on Monday, Tuesday, and Wednesday.
R code used to create the heat maps
Completed heat maps analysis (HTML)
Completed heat map analysis (PDF)
Completed heat map analysis (on RPubs)