Friday, 15 March 2013

Sampling issues of using Twitter to recruit participants for research on sensitive topics

Vicki McDermott and Billy Mcfarlane are PhD Researchers at UEA. They can be contacted at and

This blog post is aimed at highlighting the sampling issues we encountered in our respective PhD research. We are both researching sensitive topics; Vicki’s survey was about young people’s experiences of partner violence in their intimate relationships, and Billy surveyed young people about bullying. Both studies used online surveys and social media was used to facilitate data collection. Surveys were targeted at young people aged between 16 and 19. Comparing some of the demographic characteristics of the participants of our research we intend to briefly highlight some of the sampling issues that we encountered in using social media to collect survey data in our research on sensitive topics. 

In relation to the gender of our survey samples, more than three quarters of participants in the partner violence study were female (78%) whereas in the bullying study 58% of participants were female. Bullying may be perceived as a more gender neutral issue than teenage intimate partner violence, which is more likely to be experienced by females. A further issue to consider is that the individuals and groups sharing or retweeting the partner violence survey on social networking sites might also have been women focused and/or have an emphasis on intimate partner violence/domestic abuse.

It is also interesting to compare the ages of participants who completed our surveys. A third of the partner violence study participants were 18 years old (33%), slightly over a quarter were 17 (27%) and 19 (28%). Sixteen year olds made up the smallest proportion at 12%. In comparison, the bullying study just over a third of participants were 17 (34%) whilst under a third were 18 (31%), 16 year olds made up 19% of this sample, with 19 year olds being the smallest group at 16% of participants. The slightly older skew in Vicki’s study may point towards IPV not being identified by young people until they reach early adulthood. In contrast bullying is emphasised from a young age and so younger participants may be more likely to take part in this survey.


  1. That's pretty much what I've used oѵer thе paѕt few
    dayѕ, but іѕ alѕo due to the prt-a-pοrter collесtіon.
    But it is nоt the fіrѕt timе offeгіng a
    full catωalk shоω in Lοndon Tom Foгd.
    Princess Deena Abdulaziz At the Ohne Titel fаll 2013 runwау ѕhοw during New
    Үoгk fashіon Week іn Ѕеptember, Brozzetі
    ѕaid. Nо doubt thіs collectіon, and otheгѕ.

    Mу hοmepage; Thoi trang nam