Research In Action

For each and every one of our social media and blog posts, our Outreach team spends time discussing appropriate accompanying images. At times it can be challenging to find ones that strike the right tone, as well as appropriately convey our message. A new study suggests that this investment in time is well worth the effort. It explored whether mismatched safety-related visuals on social media impact a parent’s knowledge of safety recommendations.

The study included 150 eligible participants, over age 18 with a child age 7 or younger who lives with them at least 50% of the time, who have used personal social media account within the last 30 days, and had no history of certain eye conditions that would interfere with eye tracking equipment.

What They Did

On a computer with eye tracking software, participants were shown injury prevention social media posts presented in random order, with either matching images (where the image depicted the recommendations in the text of the post) or mismatched images (where the image did not depict the recommendations in the text of the post). Injury topics included infant safe sleep, poison prevention, and bike safety. Multiple choice survey questions were administered after each post to assess knowledge of the recommendations from that post.

Primary outcome measures included the proportion of correct answers on knowledge items to each of the matched and mismatched posts. Predictor measures included whether there was a matching image and eye tracking time on the imagery. Health literacy was also measured.

What They Found

• Participants were about 50% white, mostly female (71%), and mostly with at least some college education (84%).
• 15% had less than adequate health literacy.
• Of statistically significant importance: More time was spent viewing the text (6.5 versus 5.5 seconds), imagery (5.3 versus 3.3 seconds), and the overall entire post (13.5 versus 12.3 seconds) for the matched social posts. These significant differences persisted after accounting for differences in overall viewing times among participants.
• The proportion of visual attention was associated with a 3% greater knowledge score among matched images.
• The proportion of visual attention was negatively associated with a 1.1% lower knowledge score among mismatched images.

An important limitation to note is that pre-existing knowledge about safety topics was not measured among participants, and it is unclear whether this could have impacted knowledge scores. More research is needed to explore findings in different populations, including among fathers.

Implications

Our social media strategy initially did not always include an image with our blog or social media posts. With experience, we have found that posts with images receive more engagement than those without images. This study suggests that it’s not just the presence of an image, but that of a matching image which is most important when trying to effectively convey health messages.