NPPA EyeTrack Research Methodology: Gathering Photographs, Perspectives, Opinions, Comments, And Data

Part Four of a four-part weekly series. Read Part One here, Part Two here, and Part Three here.




ATHENS, GA (Februrary 21, 2015) – We had big questions as we started our research on photojournalism funded by the National Press Photographers Association back early in 2014: What are people drawn to in a photograph? How long do they look? Do they read captions? How do they perceive quality? What do they value, share and remember?

But first, we needed a simple system for gathering data:

* A parallel testing experience for participants as they viewed a large number of news photographs.
* A system for tracking and recording what they viewed and in what order.
* A way to record ratings for the quality and the likelihood that people might share each photograph.

We started by gathering 200 photographs from thousands published among more than two-dozen U.S. news organizations over the first four months of 2014. Half of the photographs were made by professional photojournalists, half had been contributed by the general public. All photographs fell roughly evenly into categories we dubbed news, feature and sports.

Developer and researcher Dave Stanton built a testing site, so that research participants could look through a random sequence of photographs in an even number between professional and UGC. The site tabulated all of the ratings at the end of each hour-long testing session.

Each participant was given the exact same instruction as they came in for the testing session, read from a script. The initial screen on the site:

“Thanks for taking part in our research to see how people engage with photojournalism. Please look through the photographs presented on the screen and answer the questions beneath the image. Spend as much time with each photo as you would like before going on to the next one.” 

Want to see what the participants looked at? Here’s a link to the testing site. Enter your name to begin. Your answers will not be recorded.

There were five elements to the testing, including eye tracking with an invisible, infrared camera that captures the gaze of the eyes, ratings for quality and the likelihood that a photograph might be shared, a test asking participants to guess the source of a photograph and an exit interview.

Here’s a small sample of the EyeTracking video produced by the device.

So that we might look for clear differences in the study, we recruited people in two distinct age sets: 18-30 year-olds — a group we have been calling “digital natives” because they are among the first adults who don’t have strong recollection of life before digital  —and 45-60 year-olds, or “printnets” referring to one foot in the print world, one foot in the “’Net” world. The 52 participants in the study came from a variety of backgrounds—a cement worker, housewives, a flight attendant, a school nurse, a statistics grad student, students, office workers, etc.

WEBINAR: Want more detail? Watch a rebroadcast of a full presentation on Poynter’s NewsU. To register go online here. 

NPPA members will receive a 50 percent discount with this code: 15SARA50



Funded by the National Press Photographers Association, this study follows a long line of research, looking at how people consume news content. Other studies developed by this research team include Poynter’s Eyetracking the News, a comparison of print and online news reading habits (2007) and Poynter EyeTrack: Tablet News Experiences (2012).

This project began when then-NPPA president Mike Borland called early in 2014 to ask what eyetracking had shown about photographs in previous research. There were a few, specific findings, which still hold:

- Documentary photos get more attention than static or posed photographs.

- Faces attract a lot of attention — often they are where eyes go first.

- Mug shots get only a fleeting glance — purely informational, unless there is strong, supporting information in the presentation.

 We realized the need to analyze photojournalism in much greater detail, particularly with the changing dynamic of digital and social media.

So that we might look for clear differences in the study, we recruited people in two distinct age sets: 18-30 year olds (a group we have been calling “digital natives” because they are among the first adults who don’t have strong recollection of life before digital) and 45-60 year olds, or “printnets,” referring to those who have one foot in the print world and one foot in the “’Net” world.

NPPA has offered to present specific results and testing methodology with news organizations and universities to help make sense of digital sharing and determine the best forms for storytelling.