Cost and Effectiveness of Using Facebook Advertising to Recruit Young Women for Research: PREFER (Contraceptive Preferences Study) Experience.
Social media is a popular and convenient method for communicating on the Web. The most commonly used social networking website, Facebook, is increasingly being used as a tool for recruiting research participants because of its large user base and its ability to target advertisements on the basis of Facebook users' information.
We evaluated the cost and effectiveness of using Facebook to recruit young women into a Web-based intervention study (PREFER). The PREFER study aimed to determine whether an educational video could increase preference for and uptake of long-acting reversible contraception (LARC).
We placed an advertisement on Facebook over a 19-day period from December 2017 to January 2018, inviting 16- to 25-year-old women from Australia to participate in a Web-based study about contraception. Those who clicked on the advertisement were directed to project information, and their eligibility was determined by using a screening survey.
Our Facebook advertisement delivered 130,129 impressions, resulting in over 2000 clicks at an overall cost of Aus $918 (Aus $0.44 per click). Web-based project information was accessed by 493 women. Of these, 462 women completed the screening survey, and 437 (437/463, 95%) women were eligible. A total of 322 young women participated in Surveys 1 and 2 (74% response rate), and 284 women participated in Survey 3 (88% retention rate), with an advertising cost of Aus $2.85 per consenting participant.
Facebook proved to be a quick, effective, and cost-efficient tool for recruiting young Australian women into a study that was investigating contraceptive preferences. However, Web-based recruitment may result in sociodemographic biases. Further research is required to evaluate whether Facebook is suitable for recruiting older study populations.
McCarthy E
，Mazza D
《JOURNAL OF MEDICAL INTERNET RESEARCH》
Social Media Recruitment Strategies to Recruit Pregnant Women Into a Longitudinal Observational Cohort Study: Usability Study.
Use of social media for study recruitment is becoming increasingly common. Previous studies have typically focused on using Facebook; however, there are limited data to support the use of other social media platforms for participant recruitment, notably in the context of a pregnancy study.
Our study aimed to evaluate the effectiveness of Facebook, Twitter, and Instagram in recruiting a representative sample of pregnant women in a longitudinal pregnancy cohort study in Calgary, Alberta, between September 27, 2021, and April 24, 2022.
Paid advertisements were targeted at 18- to 50-year-old women in Calgary, with interests in pregnancy. Data regarding reach, link clicks, and costs were collected through Facebook Ads Manager (Meta Platforms, Inc) and Twitter Analytics (Twitter, Inc). The feasibility of each platform for recruitment was assessed based on the recruitment rate and cost-effectiveness. The demographic characteristics of the participants recruited through each source were compared using the chi-square test.
Paid advertisements reached 159,778 social media users, resulting in 2390 link clicks and 324 participants being recruited. Facebook reached and recruited the highest number of participants (153/324, 47.2%), whereas Instagram saw the highest number of link clicks relative to the number of users who saw the advertisement (418/19,764, 2.11%). Facebook and Instagram advertisements were cost-effective, with an average cost-per-click of CAD $0.65 (US $0.84; SD $0.27, US $0.35) and cost-per-completer of CAD $7.89 (US $10.25; SD CAD $4.08, US $5.30). Twitter advertisements were less successful in terms of recruitment and costs. Demographic characteristics of participants did not differ based on recruitment source, except for education and income, where more highly educated and higher-income participants were recruited through Instagram or Twitter. Many issues related to fraudulent responses were encountered throughout the recruitment period.
Paid social media advertisements (especially Facebook and Instagram) are feasible and cost-effective methods for recruiting a large sample of pregnant women for survey-based research. However, future research should be aware of the potential for fraudulent responses when using social media for recruitment and consider strategies to mitigate this problem.
Pekarsky C
，Skiffington J
，Leijser LM
，Slater D
，Metcalfe A
... -
《JOURNAL OF MEDICAL INTERNET RESEARCH》
Sampling Sexual and Gender Minority Youth With UnACoRN (Understanding Affirming Communities, Relationships, and Networks): Lessons From a Web-Based Survey.
Periodic surveys of sexual and gender minority (SGM) populations are essential for monitoring and investigating health inequities. Recent legislative efforts to ban so-called conversion therapy make it necessary to adapt youth surveys to reach a wider range of SGM populations, including those <18 years of age and those who may not adopt an explicit two-spirit, lesbian, gay, bisexual, transgender, and queer (2S/LGBTQ) identity.
We aimed to share our experiences in recruiting SGM youth through multiple in-person and online channels and to share lessons learned for future researchers.
The Understanding Affirming Communities, Relationships, and Networks (UnACoRN) web-based survey collected anonymous data in English and French from 9679 mostly SGM respondents in the United States and Canada. Respondents were recruited from March 2022 to August 2022 using word-of-mouth referrals, leaflet distribution, bus advertisements, and paid and unpaid campaigns on social media and a pornography website. We analyzed the metadata provided by these and other online resources we used for recruitment (eg, Bitly and Qualtrics) and describe the campaign's effectiveness by recruitment venue based on calculating the cost per completed survey and other secondary metrics.
Most participants were recruited through Meta (13,741/16,533, 83.1%), mainly through Instagram; 88.96% (visitors: 14,888/18,179) of our sample reached the survey through paid advertisements. Overall, the cost per survey was lower for Meta than Pornhub or the bus advertisements. Similarly, the proportion of visitors who started the survey was higher for Meta (8492/18,179, 46.7%) than Pornhub (58/18,179, 1.02%). Our subsample of 7037 residents of Canada had a similar geographic distribution to the general population, with an average absolute difference in proportion by province or territory of 1.4% compared to the Canadian census. Our US subsample included 2521 participants from all US states and the District of Columbia. A total of CAD $8571.58 (the currency exchange rate was US $1=CAD $1.25) was spent across 4 paid recruitment channels (Facebook, Instagram, PornHub, and bus advertisements). The most cost-effective tool of recruitment was Instagram, with an average cost per completed survey of CAD $1.48.
UnACoRN recruited nearly 10,000 SGM youth in the United States and Canada, and the cost per survey was CAD $1.48. Researchers using online recruitment strategies should be aware of the differences in campaign management each website or social media platform offers and be prepared to engage with their framing (content selection and delivery) to correct any imbalances derived from it. Those who focus on SGM populations should consider how 2S/LGBTQ-oriented campaigns might deter participation from cisgender or heterosexual people or SGM people not identifying as 2S/LGBTQ, if relevant to their research design. Finally, those with limited resources may select fewer venues with lower cost per completed survey or that appeal more to their specific audience, if needed.
Delgado-Ron JA
，Jeyabalan T
，Watt S
，Black S
，Gumprich M
，Salway T
... -
《JOURNAL OF MEDICAL INTERNET RESEARCH》
ResearchGate
(全网免费下载)
钛学术
(全网免费下载)
