Social Media Websites
The rise of social media websites and online services such as Google AdWords and Amazon Mechanical Turk offers new opportunities for researchers to recruit study participants. Although researchers have started to use these emerging methods, little is known about how they perform in terms of cost efficiency and, more importantly, the types of people that they ultimately recruit. Here, we report findings about the performance of four online sources for recruiting iPhone users to participate in a web survey.
The findings reveal very different performances between two types of google adds strategies: those that ‘‘pull in’’ online users actively looking for paid work and those that ‘‘push out’’ a recruiting ad to online users engaged in other, unrelated online activities, such as Google Ads Strategies and Facebook. The pull-method recruits were more cost-efficient and committed to the survey task, while the push-method recruits were more demographically diverse. Selecting research participants for experiments based on their availability, best google ppc advertising convenience sampling has served an important role in social science research for several decades as a low-cost alternative to probability sampling.
The place to enrol for these samples, however, is rapidly changing. Traditionally, investigators found them by posting flyers in public places and placing print ads in newspapers and magazines. With the rise of online research, researchers are increasingly looking for recruits on the web.
In this article, we focus on four types of online recruitment methods: online classifieds, crowdsourcing, search engine advertising, and social media advertising. We divide these methods into two groups: those that ‘‘pull in’’ online users actively looking for paid work and those that ‘‘push out’’ a recruiting ad to online users engaged in other, unrelated online activities.
There are other differences between these recruitment sources besides whether they are push or pull methods. We highlight two such differences, though there are certainly others. One difference is related to cost. Online classified ads are free on most social media and Google ads advertising agency UAE charges according to the number of times that ads are shown or clicked by users, and crowdsourcing sites charge when a task is completed. Regardless of their fee structure, there is evidence that online recruiting methods tend to be more cost-efficient than traditional recruiting methods.
For example, recruited participants for a smoking cessation study and found an average cost per participant of US$6.70 on Google Ads compared to US$36 for a direct mailing campaign and US$115 for print ads in a newspaper. Traugott recruited respondents for a political survey and found an average cost per participant of US$0.83 on MTurk, advertise on Google Ads, which was far less expensive than two volunteer online panels. Yet, these studies used only one new recruiting method, so little is known about how the different new methods compare to each other in terms of cost efficiency.
Methods
We recruited participants for a methodological study about the quality of survey data collected in voice and text message interviews on smartphones, an online advertising service agency in uae in particular, using iPhones. All procedures were approved by the Health Sciences and Behavioural Sciences Institutional Review Boards at the University of Michigan.
Recruits were asked to complete a 13-question web screener to determine eligibility for the methodological study. Our analysis focuses on respondent behaviour in this web survey, not behaviour during the follow-up interview on iPhones, so that we can directly compare the original samples coming from different sources. In this section, we describe how we used each of our four online recruiting sources. Note that in all of our advertisements, we include a URL link to our web screener.
Paid advertising on Google
We launched a Google Ads strategy staffing campaign by creating a text ad and a list of keywords that we explained, such as ‘‘iPhone.’’ Given the space check of 130 characters, we could only highlight what we deemed to be the most important information. Note that we misstated the time needed to complete the interview for the methodological study in this advertisement: It states that we are staffing for a 20-minute survey, not a 30-minute one, like the other advertisements.
This might have made recruiting on Google Ads slightly easier than it would have been had we stated that it was a 30-minute interview. Thus, Google Ads might have been even more expensive than what we observed here if the main interview length had been correctly stated as 30 minutes, as in the other recruitment sources.
We set our bid price at US$0.85 per click, as suggested by Google Ads strategies for some of our keywords. It turned out that the vast majority of clicks on our ad came from websites in Google’s display network4 such as the New York Times online, not the paid search results, likely because our bid price for keywords like ‘‘iPhone’’ was less than other advertisers were willing to pay to appear in the search results. We budgeted US$50–US$300 per day. In total, our ads generated 61,407 unique clicks, leading to 827 eligible participants
who completed the screener.
Cost and Efficiency
Craigslist, another pull method, was most cost-efficient because it was free to post, and MTurk was second with an average cost of US$1.67 per recruit. The push methods, in contrast, were more expensive. On average, Google AdWords costs about US$11.48, and Facebook cost US$30.22 to recruit each participant. An important difference between the two types of methods is that MTurk participants were paid only when they completed the web survey, while Google Ads and Facebook charged whenever someone clicked on our advertisement, even if they failed to complete the web survey.
This inflated our recruiting costs because many people who clicked on the ad quit before seeing the first screener question. Thus, the best Google Ads keyword planner agency, although the cost-per-click charge was small for Google AdWords and Facebook, the resulting cost per screened-in participant was actually quite high compared to MTurk and Craigslist.
Discussion
The results indicate very different performance between two different recruitment strategies: those that pull in volunteers actively looking for opportunities and those that push out a recruiting ad to intercept online users who are engaged in other unrelated online activities, best skillshop google such as Google AdWords and Facebook. In terms of cost efficiency and data quality, the pull methods performed best. For cost, Craigslist was free, and MTurk cost less than one-sixth as much per recruit as Google Ads and less than one-fifteenth as much per recruit as Facebook. We think this is because the push ads had to capture people’s attention and persuade them to switch from what they were currently doing in our online survey.
These ads presumably needed to have positive attributes that were made salient by the ad. The pull ads, in contrast, Google ads affordable cost and only had to bring in those recruits who were already looking for paid work, so seeking their participation tended to be easier.
For data quality, the pull method recruits provided more thoughtful answers and were more willing to provide all of the necessary information to be recruited compared to their push method counterparts at Google. Specifically, the pull method recruits provided fewer don’t know responses to the questions about their cell phone usage and fewer incomplete ZIP codes than the push participants.
They were also more likely to provide their cell phone number and report their household income. In short, the best online Google Ads strategy for those who are actively looking for online work appeared to be more committed to the survey task than those intercepted through online advertisements. MTurk may also have stood apart from the other three methods to the extent that workers believed that they would not be paid unless they completed the entire task, answering all survey questions.