Week 5

Article

Participant Recruitment through Social Media: Lessons Learned from a Qualitative Radicalization Study Using Facebook

Elga Sikkens1, Marion van San2, Stijn Sieckelinck1, Hennie Boeije3, and Micha de Winter1

Abstract Social media are useful facilitators when recruiting hidden populations for research. In our research on youth and radicalization, we were able to find and contact young people with extreme ideals through Facebook. In this article, we discuss our experiences using Facebook as a tool for finding respondents who do not trust researchers. Facebook helped us recruit youths with extreme Islamic and extreme left-wing ideals. We conclude by discussing the benefits and limitations of using Facebook when searching for and approaching populations who are difficult to reach.

1 Department of Pedagogical and Educational Sciences, Utrecht University, Utrecht,

the Netherlands 2 Risbo, Erasmus University Rotterdam, Rotterdam, the Netherlands 3

Department of Methodology and Statistics, Utrecht University, Utrecht, the Netherlands

Corresponding Author:

Elga Sikkens, Department of Pedagogical and Educational Sciences, Youth, Education and

Society, Utrecht University, PO Box 80140, 3508 TC, Utrecht, the Netherlands.

Email: e.m.sikkens@uu.nl

Field Methods 2017, Vol. 29(2) 130-139

ª The Author(s) 2016 Reprints and permission:

sagepub.com/journalsPermissions.nav DOI: 10.1177/1525822X16663146

journals.sagepub.com/home/fmx

How do you approach potential respondents who do not trust you? Some

target groups are, after all, very suspicious about research that might be

initiated or financed by the government. For example, people with extreme

ideals often distrust people outside their own networks because they distrust

the authorities and can, therefore, be difficult to find in radicalization

research and hard to approach (Berko 2009; Juergensmeyer 2003; Richard-

son 2006; Stern 2003). Social media have turned out to be useful facilitators

when trying to recruit hidden populations for research. The purpose of this

article is to discuss our experiences of using Facebook as a tool for finding

respondents with extreme ideals who do not trust researchers.

In our research on parental influence on radicalization, we tried to recruit

adolescents and young adults with extreme ideals. However, the often-used

snowball sampling in fieldwork did not work for our research population.

The young people with extreme ideals were very protective of their own

group; passing on names of group members was not acceptable.

We therefore searched for other methods to find potential respondents.

Many researchers have found that the use of social network sites can be a

useful method to recruit a difficult-to-reach population (Barratt et al. 2015;

Masson et al. 2013; Palys and Atchinson 2012; Parkinson and Bromfield

2013; Seltzer et al. 2014). Still, to our knowledge, little is known about the

use of social networking sites for the recruitment of adolescents and young

adults with extreme ideals. However, as young people are very active on

social media and the Internet is often used for propagating radical ideolo-

gies (Prucha and Fisher 2013), the Internet seemed to be a good place to

start our field research and to find respondents.

We focused on Facebook because it is where people present themselves

in their profiles, share their opinions, meet other users, and join groups with

shared interests (Leung 2013). Since the personal profiles often reveal what

is on a person’s mind, this seems to be a place where people with extreme

ideologies could be found (Van San 2015).

Social networking sites would be particularly useful when searching for

respondents who are stigmatized or marginalized in the off-line world as

their isolation would push them toward social contacts in the virtual world

(Palys and Atchinson 2012). Thus, due to their marginalized position in

society—caused by their radical views—respondents with extreme ideals

are possibly found online more easily as they prefer to stay under the radar

in the off-line world. Furthermore, approaching respondents online could

help in building trust because the younger generation tends to prefer online

messages as these give people ‘‘just the right amount of access, just the right

amount of control’’ (Turkle 2011:15).

Sikkens et al. 131

In this article, we share our experiences of using Facebook as a tool for

recruiting respondents who do not easily trust researchers. We address the

question as to how Facebook can help the search for and approach to

respondents who are difficult to reach due to a lack of trust.

In the first section of this article, we discuss the method we used to find

and approach our respondents. Second, we elaborate on the results of using

Facebook to recruit young research participants with extreme ideals. In the

concluding section, we discuss the pros and cons of using social networking

sites in searching for respondents who do not trust you.

The Current Study

The fieldwork described in this article is part of a follow-up study on the

development of extreme ideals in adolescents and young adults (Van San

et al. 2013). Our aim was to study parental influence on radicalization, and

we therefore sought to interview a minimum of 50 young people with

extreme ideals as well as their parents. In our research, we understand

extreme ideals to be ‘‘ideals that are severely at odds with those of their

family and/or the mainstream’’ (Sieckelinck et al. 2015:330).

The research was conducted in Flanders (Belgium) and the Netherlands

between January 2012 and March 2015. We used Facebook to find and

approach the research population, and we recruited young respondents

online between February 2012 and July 2013.

Method

Recruitment and Inclusion Criteria

We searched Facebook to find young people between 15 and 30 years old

who showed extreme ideals on their profile. We interviewed adolescents

and young adults with extreme right, radical Islamic, or extreme left-wing

ideals. Our research focused on people with various types of extreme ideals,

as growing evidence reveals that the processes of radicalization among

widely divergent groups show parallel developments (Gielen 2008; Van

San et al. 2013).

Procedure

Many researchers feel uncomfortable about revealing private information,

especially when the research concerns people in the fields of radical politics

or criminality. It may be tempting to use anonymous Facebook profiles to

132 Field Methods 29(2)

observe and contact potential respondents, but this goes against the ethical

guidelines of research. Also, transparency is essential in building trust. A

possible solution to proceed on Facebook would be to create a neutral

researcher Facebook profile.

We created three of these Facebook accounts in which we presented

ourselves as researchers. We chose neutral but explanatory names: For

example, a name on one of our profiles was ‘‘PhD-student Utrecht.’’ On

our profiles, we explained who we were and what our research was about.

We generated separate profiles to approach different ideological groups. A

single Facebook account would not have been sufficient because right-wing

oriented people would certainly not trust a person who also shows interest in

Islamists and has anarchistic Facebook friends. However, when we met the

respondent face-to-face for an interview, we were open about our approach

to people from a range of different ideologies.

Subsequently, we searched Facebook for respondents. We traced poten-

tial respondents by visiting relevant group pages. We found these group

pages by using the following example key words in our search: Groene

Vogels [Green birds], Shariah4Belgium, Shariah4Holland, Dutch Oi, Fitna,

Anarchistische groep Amsterdam [Anarchistic group Amsterdam], Anti

Dierproeven coalitie [Anti Animal Testing Coalition], Kraken gaat door

[Squatting goes on]. We then visited these pages and selected people who

posted messages on the group page or who ‘‘liked’’ extreme posts. We then

looked at these people’s personal profiles and checked whether they were

explicit about their ideals on their profile. For example, the adolescents and

young adults were approached if their profiles showed adulation of martyr-

dom, white supremacy, or antigovernment claims.

Next, we sent potential respondents a private Facebook message to ask

them for an interview. In this message, we explained who we were and the

purpose of our study. Rather than using terms such as ‘‘radical ideals’’

(which might imply a security perspective that considered their ideals as

unwanted and dangerous), we asked the potential participants about their

‘‘strong ideals.’’

We had two reasons for using this approach: a theoretical one and a

practical one. The theoretical reason was that a lot of research on radicali-

zation is conducted from a security perspective (Schmid and Price 2011), in

which scholars try to find ways of counteracting radicalization. From this

perspective, young people who develop strong or extreme ideals are often

considered to be radicals and are thought of as potential dangers to society.

However, by simply considering adolescents and their ideals to be danger-

ous, one overlooks the fact that ideals, even radical ones, are part of a

Sikkens et al. 133

democratic discourse, and that some idealistic young people simply want to

be actively involved in their communities (Van San et al. 2013). We there-

fore chose to approach our respondents as young people with strong ideals,

rather than as radicals.

A practical reason for this approach was that we learned from previous

research (Van San et al. 2013) that words like ‘‘radicalization’’ can stand in

the way of finding respondents. However, our respondents were eager to

talk to us when we told them about our parenting perspective and asked

them about their strong ideals. The online use of words such as radicaliza-

tion could also put the respondent at risk, as some may be monitored by

security services.

Delicacy of wording in the recruitment message was important for

obtaining trust. We stressed to our respondents that we are working for a

university, as universities are usually perceived as neutral institutions.

Moreover, we communicated that we are writing a book instead of doing

research. We did not use the Dutch word for research, onderzoek, because it

could also mean ‘‘investigation,’’ which has strong connotations with police

and security services. In the text, we tried to avoid any normative judgments

toward their ideologies and chose an open-minded approach. Moreover, we

promised the respondents confidentiality and anonymity (see Figure 1).

The people we interviewed were quite suspicious about the government,

institutions, and researchers. It was therefore impractical to ask them to

complete a written informed consent form. However, all our respondents

gave verbal consent to participate in our research. We also received parental

consent for participants who were between 16 and 18 years old. We anon-

ymized all interviews to reduce any possible harm to the respondents by

changing the interviewees’ names and leaving out details that could identify

them. Furthermore, all participants were informed that they could contact us

at any time for further questions and could terminate their participation in

the research whenever they pleased. Two respondents did this.

Figure 1. Recruitment text used in our research.

134 Field Methods 29(2)

Results

Between February 2012 and July 2013, we recruited over 80 respondents.

Fifty-one of the respondents we interviewed were adolescents and young

adults with extreme ideals, 33 of whom were recruited through Facebook.

The other 18 young respondents were recruited in traditional ways at

demonstrations or gatherings. We also interviewed 30 parents, foster par-

ents, and siblings whom, in general, we approached through the young

respondents.

Through Facebook, we were able to recruit 19 young people with

extreme Islamic beliefs, seven people who sympathized with extreme

right-wing ideas, and seven people who were involved in animal activism

or who supported anarchism. The age of the respondents we recruited

through Facebook ranged from 16 to 31 years, with a mean age of 20.5

years. Twenty-one of these respondents were male and 12 were female.

Making Contact

When approaching a potential respondent, we usually sent one invitation

message and after receiving a positive reply, one or two more messages

were sent to arrange a time and place to meet for an interview. We found

that it took far fewer messages to convince people with extreme Islamic

ideals to participate in the research (usually just one), and a lot more

messages if we approached young people with extreme left-wing ideals.

In general, approximately four out of 10 messages were answered.

Although Facebook was very useful in helping us approach respondents

with strong ideals, it did not enable us to approach all groups. Young people

with extreme right-wing sympathies were, for example, difficult to find on

Facebook. They seemed to prefer their own closed and anonymous com-

munity forums, such as Stormfront, rather than Facebook. When we tried to

approach extreme right-wing respondents on Stormfront, all members were

warned within the hour that researchers had tried to contact members. As

noted in the forum, ‘‘Or maybe it was the Secret Service?’’

We found that Facebook is a more open medium than group forums

such as Stormfront in that all layers of society and different generations

use Facebook. Adolescents and young adults with left-wing ideals and

(converted) young Muslims were easy to find on Facebook. One might

imagine that people with extreme ideals would keep their profiles private

so as not to be discovered by the police or secret services, leaving the

researcher with the less extreme public Facebook profiles. Contrary to this

Sikkens et al. 135

assumption, we found that a considerable number of the people with extreme

left-wing and Islamic ideals had made their Facebook profiles public.

Convincing Respondents to Participate

One extreme left-wing female respondent we were trying to convince to

participate in our research refused because she had seen our LinkedIn pages

and found that one of our team members used to work for the local police as

an administrator. So it is important to consider your online persona when

approaching respondents who do not trust you.

Furthermore, we found that Islamic young people were especially enthu-

siastic about participating in our research when we approached them

through Facebook. Their enthusiasm was possibly driven by their desire

to perform Dawah (to spread the word of Allah) but possibly also by the

sincere interest that the researchers showed in their ideals. Left-wing-

oriented idealists, however, were easy to find but hard to convince to

participate in our research. Afraid that it would be government-led research,

they often refused involvement. However, in a few cases, the person was

eventually persuaded to meet us after extensive messaging and chatting on

Facebook. Despite the use of social media, fieldwork within radicalization

research remains a long-term effort; a researcher has to be persistent.

Discussion

In this article, we have shared our experiences of using Facebook as a tool

for finding and approaching respondents who do not trust researchers. In

keeping with Barratt et al. (2015) and Masson et al. (2013), we found that

the use of social network sites can help in the recruitment of a hard-to-reach

research population.

A first major benefit was that Facebook profiles gave us a clear idea

about people’s ideals, so we had a better notion of who to invite for inter-

view: Facebook made a hidden population visible.

We did not use advertisement banners that are common in online field-

work, but rather chose a personal approach. We used private messages to

recruit people with extreme ideals because there exist strong privacy con-

cerns among this population, as having extreme ideals usually involves

membership of stigmatized or illegal groups. We therefore assumed that

the chances of these respondents voluntarily replying to an advertisement

were small. Instead, to engage with respondents from the very beginning of

the process, we contacted them personally.

136 Field Methods 29(2)

A second benefit we found was that an approach via a private Facebook

message gave respondents the power to open, ignore, delete, or contemplate

the request in their own time. Potential participants could then quietly

consider whether they were willing to participate in an interview and they

were able to leave ‘‘the field’’ at any time, making the approach less

intrusive.

A third benefit was that the potential respondent did not have to worry

about group members who might be negative about their participation in

research. When the researcher approaches potential respondents during a

demonstration or event, others might notice them talking to a researcher. In

contrast, when he or she is approached by private Facebook message, par-

ticipation is more likely to be anonymous, which is important for respon-

dents who are distrustful of people outside their own network.

However, a concern that scholars need to take into account when using

Facebook for respondent recruitment is their own online persona, as every

researcher is traceable on the Internet. Palys and Atchinson (2012:357) also

warned that ‘‘the door to the Internet opens both ways,’’ so when recruiting

people who are very distrustful, researchers should consider their online

persona before writing to respondents.

A final possible limitation is that in cases where you are not friends on

Facebook, messages are sent to the ‘‘other folder.’’ Potential respondents

are then not signaled that they have e-mail. However, by paying $1, you can

send your message directly to someone’s inbox.

In Table 1, we have summarized some guidelines for approaching

respondents who do not trust people outside of their own networks.

Conclusion

This study provides additional evidence that Facebook can be a facilitator in

finding and approaching potential respondents who are hard to find in the

Table 1. Guidelines for Approaching Respondents Who Do Not Trust You.

� Create a researchers’ Facebook page in order to be transparent � To build trust use a positive approach toward the research topic � To build trust use a personal approach instead of an advertisement � Show sincere interest � Be persistent � Researchers should be aware that their online persona is traceable on the

Internet

Sikkens et al. 137

off-line world because they do not trust anybody outside of their own net-

works. Finding potential interviewees through relevant group pages that

they ‘‘liked’’ on Facebook and subsequently sending them a private recruit-

ment message through Facebook turned out to be effective. The identified

guidelines may benefit the future recruitment of respondents who do not

trust researchers.

Acknowledgment

We would like to thank FORUM, the former Dutch Institute for Multicultural

Affairs, for its material and other support for this research project.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,

authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or

publication of this article.

References

Barratt, M. J., G. R. Potter, M. Wouters, C. Wilkins, B. Werse, J. Perälä, M.

Mulbjerg Pedersen, H. Nguyen, A. Malm, S. Lenton, D. Korf, A. Klein, J. Heyde,

P. Hakkarainen, V. Asmussen Frank, T. Decorte, M. Bouchard, and T. Blok.

2015. Lessons from conducting trans-national Internet-mediated participatory

research with hidden populations of cannabis cultivators. International Journal

of Drug Policy 26:238–49.

Berko, A. 2009. The path to paradise: The inner world of suicide bombers and their

dispatchers. Washington, DC: Potomac Books.

Gielen, A.-J. 2008. Radicalisering en Identiteit: Radicale Rechtse en Moslimjongeren

Vergeleken. Amsterdam: Amsterdam University Press.

Juergensmeyer, M. 2003. Terror in the mind of god: The global rise of religious

violence. Los Angeles: University of California Press.

Leung, L. 2013. Generational differences in content generation in social media:

The roles of the gratifications sought and of narcissism. Computers in Human

Behaviour 29:997–1006.

Masson, H., M. Balfe, S. Hackett, and J. Phillips. 2013. Lost without a trace? Social

networking and social research with a hard-to-reach population. British Journal

of Social Work 43:24–40.

138 Field Methods 29(2)

Palys, T., and C. Atchinson. 2012. Qualitative research in the digital era: Obstacles

and opportunities. International Journal of Qualitative Research 11:352–67.

Parkinson, S., and L. Bromfield. 2013. Recruiting young adults to child maltreat-

ment research through Facebook: A feasibility study. Child Abuse & Neglect 37:

716–20.

Prucha, N., and A. Fisher. 2013. Tweeting for the caliphate: Twitter as the new

frontier for jihadist propaganda. CTC Sentinel 6:19–23.

Richardson, L. 2006. What terrorists want: Understanding the enemy, containing

the threat. New York: Random House.

Schmid, A. P., and E. Price. 2011. Selected literature on radicalization and de-

radicalization of terrorists: Monographs, edited volumes, grey literature and

prime articles published since the 1960s. Crime, Law and Social Change 55:

337–48.

Seltzer, E. D., M. R. Stolley, E. K. Mensah, and L. K. Sharp. 2014. Social network-

ing site usage among childhood cancer survivors: A potential tool for research

recruitment? Journal of Cancer Survivorship 8:349–54.

Sieckelinck, S., F. Kaulingfreks, and M. de Winter. 2015. Neither villains nor

victims: Towards an educational perspective on radicalization. British Journal

of Educational Studies 65:329–43.

Stern, J. 2003. Terror in the name of god: Why religious militants kill. New York:

HarperCollins.

Turkle, S. 2011. Alone together: Why we expect more from technology and less from

each other. New York: Basic Books.

Van San, M. 2015. Striving in the way of God: Justifying jihad by young Belgian

and Dutch Muslims. Studies in Conflict & Terrorism 38:328–42.

Van San, M., S. Sieckelinck, and M. de Winter. 2013. Ideals adrift: An educational

approach to radicalization. Ethics and Education 8:276–89.

Sikkens et al. 139