We designed and distributed an online survey to query participants about their use of social networks. The survey was mainly, but not only, addressed to radiologists and resident radiologist. The multiple-choice survey was created using free online survey software (Google Forms, Google LLC, Mountain View, California, US). Participants were invited to take part in the survey between November 10, 2018 and February 28, 2019 by e-mail solicitations from author’s personal radiologist networks, by posts on social media platforms, Twitter, Facebook, Instagram, Weibo, Reddit. The email networks included approximately 60 trainees and faculty in the author’s radiology department and 20 radiology colleagues from other academic and private hospitals in Italy. A series of posts were submitted on Facebook and Twitter in order to promote and invite to participate in the survey. The post contained a message with a description of the survey, the promoter info (name, surname, email, affiliation), and time of completion (3minutes). The survey was anonymous and the project research didn’t receive any funding.
The survey questionnaire focused on four main sections: demographic data of the participant (Age groups, country, job position), type of social networks user, content quality and daily practice utility perception of radiology learning materials shared on the different social networks (Facebook, Instagram, Twitter, Linkedin, Figure1, Weibo), user activity on social networks.
The survey was designed with the use of the redirection option, so, on the basis of the different answers, some sections were not displayed to the participants.
Age groups were categorized by three different age-range (18-34,35-54,>55). Participants selected their job position from various options (Radiologist, Radiologist in training, Medical Doctor, medicine student, Radiographer, other).
To estimate the quality of the radiology content shared on social networks we used semantic differential rating scale. The respondents were asked to choose where their position lies, on a 5-factors scale between two polar conditions (very high quality – very low quality).
Descriptive analysis and ANOVA test were used for comparisons, as appropriate. A p<0.05 indicated a statistically significant difference.