Online Marketing Research: Reliability and Appropriateness of Social Network Sites Wendy Tapia

Abstract The widespread acceptance that the web 2.0 platform offers new opportunities and possibilities to meet the current marketing research challenges requires an in depth appreciation of the advantages and disadvantages involved in its implementation. The purpose of this paper is to take a critical look at the use of Social Network Sites as an Online Marketing Research tool. Due to the vast number of Social Network Sites available this paper focuses specifically on the Social Network Site, Facebook. The paper starts by giving a brief insight of how Social Network Sites are defined, it then contrasts the advantages and disadvantages of its use for research purposes based on previous studies conducted in similar fields but not exclusively in Social Network Sites. And finally, it is argued that the implementation of Social Network Sites as a research tool is suitable to some instances and that its wide range of applications would positively contribute to a better interaction with users. Social Network Sites does offer new research tools opportunities and can also function as a supporting or complimentary research tool; however, the use of Social Network Sites for marketing research purposes in instances that require in depth data, sensitive information or large world populations is not recommended.

Introduction Marketing Research is used to identify opportunities and problems. Its general purpose is to generate, evaluate and improve marketing actions, this includes; monitoring performance; and improve understanding of marketing as a process. The importance of marketing research relies on the fact that many decisions in businesses are based on results obtained through this type of research making this process a crucial activity within businesses. In the last decade, undertaking any type of research has suffered a decline in response, and marketing research is no exception (McDonald & Adam, 2003, p. 85) Savage and Burrows (2007, as cited in Beer 2008, p.619) have described a “coming crisis of empirical sociology” caused by a tension between the limited scope and depth of insight of traditional social research techniques, against the vast resources of the business sector. This cannot be denied since technological developments and the increase of Internet penetration all over the world has brought a trend wave for web 2.0 applications, which

1

are also seen as a promising instrument that could contribute to improve the level of response. Before elaborating on the contents and uses of Web 2.0 it is worth to establish a definition. Web 2.0 is a set of collaborative web applications that are similar to traditional media characterized for allowing users to utilize their computers as a broadcasting medium in which they can publish, share, discuss and be part of communities. Its increasing popularity has attracted millions of users also countless business and researchers. What is known as Web 2.0 is composed by a number of tools or applications, some examples of this are: Virtual worlds, discussion groups, blogs, gaming sites, aggregators, microblogs, and social network sites (Safko, 2009). Its highly social and participatory nature has attracted a large audience and many researchers recognize the potential that Social Network Sites could have in research as a tool. This paper aims to analyze the appropriateness and reliability of Social Network Sites as a marketing research tool. Since there are no formal studies on Social Network Sites as a research tool this analysis is based on previous studies conducted in other web 2.0 tools, as a point of reference and further documentation of the implementation of Social Network Sites as a research tool topic is suggested.

Analysis of the profile of Social Network Sites The wide variety of web applications 2.0 available offers to marketing researchers the possibility of reaching large audiences and it is believed that by using web 2.0 tools the rates of response could experience an increase (Redmond, 2010) this belief increases as we witness the active participation of users in Social Network Sites. Until today the tools that have been used for online marketing research purposes and that have been widely documented, include: online surveys, research communities, and online focus groups. In this study we pay special attention to Social Network Sites because while there has been very little research in this field its implementation is already taking place on a casual shape.

2

 Social Network Sites Description Social Network Sites term is referred in the literature review as: social networking, social networking sites or simply: networking. However, ‘Networking’ emphasizes relationship initiation, very often between strangers, and this is not necessarily the case in these sites. Therefore in this study it will be referred to as Social Network Sites (Boyd & Ellison, p. 211).

Social Network Sites, allow users to create a personalized profile that includes information, such as: date of birth, relationship status, hobbies, preferences, education background, interests, etc. (Zywica & Danowski, 2008, p.2) These sites were originally thought as a good way to find old classmates (Page, 2010, p. 425). However, nowadays these types of sites allow conversations between users through a simple and unique interface. The core element of Social Network Sites in the interest of marketing researchers is the great interactivity generated through the accumulation of diverse application features within one‐site (Trusov et al. 2009; Boyd & Ellison, 2008) The high level of participation, interaction, and influence that Social Network Sites provide is seen as an opportunity which adds to the ability of Social Network Sites to track all the decisions made by their users and to analyse demographic, psychographic and behavioural data increasingly attracting more advertisers, marketers and businesses to take this approach. Casteleyn et al. (2009) supports this idea by saying that social networking sites display intentions of consumers and therefore techniques specifically designed to deal with this have to be introduced. The intention of a consumer is an aspect that any marketing researcher would be interested in finding; however, the task would get lost in the impracticality of reading comment by comment in a pool of millions of users. Currently, the most popular Facebook applications used informally for marketing research purposes are: Fan pages, mobile applications, integration of Facebook to other sites, and built research applications (with Facebook developer platform). All the mentioned applications offer immediate data compilation, reports, and access to a dashboard as shown in Figure 1.1.

3

 [Figure 1.1 – Screenshot taken from Griffith College’s Facebook dashboard]

Challenges of SNS implementation as a Research Tool While technology developments sound promising at first, we must consider the difficulties that a new research tool implies. In the case of Social Network Sites the first and most important issue is data analysis, in which determining an appropriate scheme for reducing the massive quantity of data received is crucial and challenging. (Feldon & Kafai, 2008) On this issue we observe that in fact, data collected and the metrics obtained are heavily dependent on the information that the specific site, in this case Facebook, decides to share, due to privacy policies (Stuart, 2009, p. 23) While the information will be reported automatically, in most of cases if detailed information is required this implies tracking each users’ participation as well as cross‐checking information with either website traffic statistics or other dashboards depending on the objective of the research. In one hand, there is the advantage of receiving automated results based on the information provided by the Social Network Sites Facebook, but that may not include the data that is to be found, limiting the access to detailed data. Another concern of conducting research in Social Network Sites is related to the identity and veracity of the demographics in Social Network Sites. It is observed that individuals may choose to establish multiple identities to interact in their online world as different

4

personae (Feldon & Kafai, 2008) Marwick and Boyd (2010) reinforce this idea by pointing out that profile owners are attentive to audience and that bringing together commonly distinct audiences complicates the need for variable self‐presentation. This raises a question in relation to the reliability of the demographics provided by Social Network Sites, which may have an impact in some instances, and for social research is an interesting aspect to study within Social Network Sites. In the online world, and especially in Social Network Sites, individuals have the opportunity of re‐formulating their identity. (Kiesler et al., 1984; Turkle, 1999 as cited in Rodriguez & Hung, 2010, p. 201) When the individuals join a Social Network Sites they find the opportunity of creating the self‐narrative they choose, the portray that they wish to show, highlighting only those elements that they wish to make evident and hiding those that are considered irrelevant within the new persona that is being created. Therefore the individual living the online, Social Network Site, life is a reflection of the idealized subject of the offline individual and then both subjects, online and offline are complimentary and grow together. (Rodriguez &Hung, 2010, p. 201) In relation to this concern Marwick (2005, as cited in Zywica & Danowski 2008, p.6) proposes how to analyze profile categories of text, pictures, and testimonials. For this, he identifies three types of presentations:
[Types of presentation] Authentic, presenting true information about the self such as real name and location; Authentic Ironic, presenting true information but modifying it using sarcasm, irony, or satire; and Fakesters, whose profiles claim that they are celebrities, objects, places, activities, or obscure in‐jokes. Skog (2005) reported that on LunarStormä, profiles, sending messages, and indicators of authenticity such as using ‘‘real’’ photos, indicate one’s status. Boyd and Ellison (2007) point out that friendship links, or ‘‘public displays of connection,’’ are another important aspect of self‐presentation

The three types of presentation suggested, may give a contextual reference to the researcher. However, an accurate understanding of the self‐narratives presented by users would be subject to the researcher’s interpretation, increasing the volume of data to be analysed. Behavior is another element of individual’s self‐narrative that can influence in marketing research. For example, a study carried out in 2008 found that 9.6% of the younger users born in 1986 or later admitted to exaggerating or making up information while 4.5% of older users do this. (Zywica & Danowski, 2008, p. 17) Depending on the

5

sample that our research aims to reach, this percentage could have an impact on the results. Researchers, accepting this limitation should probably consider a percentage of users who may respond as an alter persona. However, not every study has reported this as an issue, Yee (2007 as cited in Feldon & Kafai, 2008) highlights that in the results of a quantitative study conducted in virtual worlds to 30,000 online gamers, most users reported interacting with others in very much the same way as in the real‐world environment. Nevertheless, there seems to be different results on this matter depending on the research being quantitative or qualitative. Another aspect to be considered behavior is that individuals participating of online research marketing are interacting in both: the online and offline world. Acting simultaneously is believed to have an impact, conversations between two users who are present in the same physical environment can prompt changes in online action (Feldon &Kafai, 2008) In the specific case of Facebook a very recently launched tool, called: ‘Question’ opens a debate in relation the influence that one user can have over others. The question tool offers such a simple platform to ask questions to users that it could be useful to conduct a express survey (only one question at a time allowed) with the weakness of potentially affecting other users’ responses due to the format in which it is delivered by being published in the individual’s profile along with the answer chosen as observed in Figure 1.2. [Figure 1.2 – Screenshot taken from Griffith College’s Facebook Account ‐ Question tool]

Marketing practitioner Poynter (2008, p.11) believe that the concerns of strict researchers about representativeness, sampling methodology and validity of online methodology are unnecessary considering the falling response rates and claims that we

6

can no longer assume conventional research meets any of these criteria. On this matter, Beer (2008) also stresses that in a context where data are routinely gathered and the difficulty of attracting participants increases, the traditional methods seem insignificant or irrelevant. The urgent need of increasing participation is evident, however, it is also critical to understand that methodological rigor is what builds reliable data and that emerging tools should be embraced and tested but without excluding important elements from traditional research, such as: Representativeness, reliability and validity which are pillars of research, which should not be undermined. Until 2011, Facebook had 500 million users, which is a unique figure taking into consideration that indirect competitors of the Social Network Sites have in average 150 million registered users. Still, this figure excludes a significant proportion of the world’s population. While this argument may not be relevant to those conducting research in social grades: A, B, C1 and C2 which comprehend: upper middle class, middle class, lower middle class and skilled working class; it becomes an issue to those who intend to conduct research in social grades D, E and F, which depending on the country, these groups may live under radically different conditions and access to a computer is not part of their profile. A good example of this is the absence of Facebook in China, while there are other Social Network Sites as powerful as Facebook in China these may not be representative of specific groups. An important aspect of conducting research in SNS would be that these sites, in itself, are already segmented and that they do not represent every demographic. In the cases in which the population that is to be researched does belong to Social Network Sites there is another element considered a challenge, and that is the power of users to choose what they wish to consume. Within the Facebook platform now, more than ever, users have been empowered to read, watch and participate of only those things that they wish to further explore. This is a great aspect of Social Network Sites but it also represents a challenge to any researcher willing to enter this atmosphere. Marketers must develop a very close relationship with their audience before assuming that users will respond to their questions. Poynter (2008) gives two examples of brands who are successfully getting responses and feedback from users whenever they are surveyed, both brands enjoy a privileged positioning within their corresponding industries and are market leaders, which pronounces the challenging position in which

7

smaller businesses and researchers will face in trying to influence users to participate of research. The main challenges of conducting marketing research in Social Network Sites have been pointed out. In the next section the advantages of Social Network Sites is explored. Advantages of Social Network Sites as a Research Tool Based on research conducted in other online channels, a number of advantages have been identified. It is believed that the here mentioned advantages are the driving forces of many marketers to implement research in Facebook. Among the most popular advantages of online resources it is frequently found in literature review that web‐based applications offer greater speed, this means that data is gathered and reported in real‐time (Cooke et al. 2009), which avoids time consuming manual inputs and reduces human errors to a minimum. Facebook is not exception, through its dashboard of data collected and a very comprehensive interface, the results are not only recorded and displayed but it also creates a report on a monthly basis which is sent by email to the administrator of the account. Another advantage of using Social Network Sites as a marketing research tool is its flexibility in design and the potential to improve as technology advances. For example, some web‐based applications may offer the ability to show visual material through real‐ time conversations with the support of a camera, this is currently available on Facebook but it is a foreseeable facility. The open source technology in which Facebook is based also allows researchers to team up with programmers to find new ways of enhancing the functionality and experience of users when participating of research. In this regard it is imperative that researchers and application developers spend enough time in designing tools that are informative, and appealing in order to encourage future participation. Cooke et al. (2009) also mentions the convenient element to conducting online research as self‐administering tools as it represents that participants can respond at their own pace and schedule, rather than at the interviewers’. It has been identified that when there is a migration from offline to online methodology data often appear more critical than the offline on satisfaction measures. This could point out to the fact that without an

8

interviewer respondents are free to criticize if they wish. Cooke et al., (2009) also explain that in research conducted to heavy internet users aged 16 – 65 years old produced richer data than those sourced from face‐to‐face contacts. In the specific case of Facebook platform, it is a possibility that users feel more comfortable in a familiar environment and therefore more open to provide in‐depth criticism. Other advantages of the implementation of online marketing research are costs and time, which are considerably lower compared to traditional methods. In the case of Facebook its open source nature will also offer significant budgetary reductions, although, a very sophisticated application may require a considerable investment. The implementation of Social Network Sites can be seen as an expansion of the researcher’s toolbox, where it is the researcher’s duty to explore, experience and improve the new tools to avail. A few advantages of Social Network Sites have been identified based on the features of Facebook as well as previous research conducted in other online resources. In order to contrast and reach a conclusion, the disadvantages of Social Network Sites are now presented. Disadvantages of Social Network Sites as a Research Tool The advantages presented earlier provided with a good reference of the possible aspects to be contrasted in order to find the disadvantages of implementing Social Network Sites as a marketing research tool. Firstly, the evident enthusiasm to embrace new technologies in order to improve research participation has blurred the fact that cooperation rates are heavily dependent on the persons’ interests more than the channel through which individuals are being reached. (McDonald & Adam, 2003, p.85) In this aspect if a brand conducting research does not have a good positioning in the market, or enjoy some level of reputation, users may struggle to identify it and therefore opt for not joining a discussion. Lampe et al. (2006, as cited in Zywica & Danowski 2008, p.8) also found a similar result and pointed out that Facebook users engage in searching for people with whom they have an offline connection more than they browse for complete strangers to meet. So, this opens the question if a brand, business or researching agency should have a presence on Facebook before venturing to launch any research tool, this in order to gain users and to build a

9

relationship. In traditional research, the surveyor could randomly approach people in the street and conduct the survey, while in Social Network Sites users do not show the same level of openness and even many are protective of their accounts by exclusively adding friends and people they know in the offline world. Another disadvantage comes from the decline in response in marketing research, which may impact on the representativeness of a sample and so the validity of the results. This also links to the issue that reaching audiences, is limited to the number of registered users of, in this instance, Facebook. A reality is that Internet penetration varies from country to country and some demographic groups will not be online in the foreseeable future in a number of countries. (Cooke et al., 2009) in terms of Internet penetration in Ireland, specifically, a report published by Datamonitor (2008, p. 70) reveals that
… it is estimated that in 2007 around 46% of the Irish population had a PC and around 60% of the population was connected to the Internet. Broadband connection rates were at 16 broadband subscribers per 100 inhabitants, compared to 30 for countries like Denmark, the Netherlands and Switzerland.

This data may seem positive at a first glance, however, as pointed out in Datamonitor’s report this does not fulfill the expected performance and it contrasts these figures with the 30 broadband connections for every 100 in countries like Denmark, Netherlands and Switzerland.

Another element to be thought through is the Computer Mediated Communication limitations. While online focus groups and virtual communities may find a way to read non‐verbal communication, Social Network Sites such as Facebook does not offer applications that facilitate video conferencing or a virtual community as such. Games developer, Zynga, is the closer attempt to developing communities in Facebook by offering similar interaction through its different games however; this is another type of online marketing research and will not be further discussed in this paper. Lastly, opting to use Social Network Sites may offer more freedom to users, but it still to be considered that the self‐completion methodology may lead to vague answers if the respondent is unsure of the question and there is nobody there to clarify. (Cooke et al., 2009) on this matter researchers are required to design supporting tools, such as help chats or an online user that can answer specific questions. The size and reach of the tool will be defined on the basis of the type of research aspired, evaluating the suitability of

10

Social Network Sites takes us into our next analysis which focuses on the appropriateness of SNS as a research tool.

Appropriateness It is understood that researchers should develop their research tools to the best form to obtaining data required for a particular study and an essential duty is to weigh all options available and choose the tool/s that will potentially gather the most representative and reliable source of data. The importance of this process is highlighted in this analysis as the rushed trend to implementing Social Network Sites as a source of data may have been overlooked. As previously mentioned, the understanding of interaction with users of Social Network Sites is necessary to any implementer. Approaching a user to answer a question does not have the same nature as surveying randomly in the street, where the surveyor has a chance to introduce and identify him/herself. Therefore the introduction process in Social Network Sites preferably should occur in advance, by establishing a reason why the user should participate. This will lead many marketing researchers to wonder if they should have a permanent presence in Social Network Sites; businesses and researchers considering using this platform as a research tool should reflect on this point. The adaptability of the tool is another element to consider. It has been discussed that, in this aspect, Facebook does offer great flexibility due to its open source platform and the customization ability. Re‐designing a tool to migrate from offline to online must consider aspects such as: length efficiency, and interaction. (Cooke et al., 2009) and these elements are crucial to an adaptable tool. An advised approach is to use mixed methods; this is in order to secure relevant data from different sources. Facebook offers the ability to customize applications, which gives the researcher the possibility of adding close questions, open ended questions, multiple option questions as well as traditional commentary on a particular topic, video (only uploading, not real‐time broadcasting). Feldon & Kafai (2008) support this idea by recognizing the value of using mixed methods to address methodological and analytical challenges. Researchers using Social Network Sites as a research tool should have a good understanding of the context, atmosphere and its users, especially if the research is qualitative. A recommended method to the implementation of Social Network Sites

11

tested by Cooke and Macfarlane (2007, p.341) is training the researcher on applying these methodologies through experimenting in order to gather empirical knowledge. Testing Social Network Sites as a tool will contribute to find the weaknesses of its implementation and therefore the possibility of finding solutions. Marketing researchers are advised to revisit the traditional methodology and to adapt them to the new requirements of Social Network Sites in order to increase its efficiency as measurement instruments, moreover, to improve users’ experience. Cooke et al., (2009) found that in a financial research it was inappropriate to replace the traditional face‐to‐face methodology to an online‐based solution. In similar instances, organisations and researchers are encouraged to closely observe the circumstances and objectives of the research to determine the appropriateness of the tool. It is the researcher’s duty to determine the appropriateness of a tool and it is of great importance that practitioners reflect on the suitability of such a tool to their studies, and that the current trend of using this platform is thoroughly analysed, also considering ethical implications.

Conclusions Web 2.0 applications are seen as a great opportunity for developing new research tools. It is of special interests to marketing researchers to understand consumers, and in this effort, many have turned to join and conduct casual research in Social Network Sites and especially in Facebook. Some of the advantages of implementing Social Network Sites as a research tool are: lower costs, speed in response, and the participatory nature of the channel as well as customizable tools within the site. The identified advantages show that Social Network Sites are a potential platform to be embraced as a research tool, however, the researcher depending on the nature of the study can only determine the appropriateness of its application. Once that the researcher has determined that Social Network Sites are an appropriate platform to a particular study, Facebook’s interface and its wide range of applications could contribute to the tool’s design by mixing different methods. A single method research may lead to conflicting findings or an overwhelming amount of data, and it is therefore not advised.

12

New technologies should be embraced for marketing research purposes as long as they are thoroughly understood, Cooke and Macfarlane (2007) suggestion of pilot programmes is a good solution to finding weaknesses that could potentially impact on final results. In the other hand, it cannot be overlooked the fact that some instances still require the human element in their research by studying non‐verbal communication or responses to a specific environment, especially within qualitative research sourcing some type of data may not be feasible through Social Network Sites. It is therefore believed that the implementation of Social Network Sites as a research tool is suitable to some instances and that its wide range of applications would positively contribute to a better interaction with users. Social Network Sites does offer new research tools opportunities and can also function as a supporting or complimentary research tool; however, the use of Social Network Sites for marketing research purposes in instances that require in depth data, sensitive information or large world populations is not recommended. Those instances that require full accessed to detailed data are advised to design private platforms or research communities that allows them to gather information ethically by informing the participant of the purpose of the platform, and avoiding privacy issues involved in the usage of Facebook.

References • • “Alexa Top 500 Global Sites,” [Online] Available at http://www.alexa.com/topsites. Accessed on 15th April 2011. “Free demographics classifications, lifestyles and social grades listings, acorn 2003 profiles, UK population percentages,” [Online] Available at http://www.businessballs.com/demographicsclassifications.htm Accessed on 15th April 2011. “Griffith College Dublin Facebook account” [Online] Available at http://www.facebook.com/griffithcollege Accessed on 20th April 2011. “My Data Mining Weblog » Top 50 Search Competition of Social Networking Sites in Google, Yahoo & Bing,” [Online] Available at http://mydatamine.com/?p=541 Accessed on 15th April 2011. “Statistics | Facebook,” [Online] Available at http://www.facebook.com/press/info.php?statistics Accessed on 20th April 2011. “TECHNOLOGICAL LANDSCAPE.,” Ireland Country Profile (January 2008): 68‐ 73.

• •

•

•

13

•

•

•

• •

•

•

•

•

•

• •

•

•

• • •

Colin C. Cheng, Dennis Krumwiede, and Chwen Sheu, “Online audio group discussions.,” International Journal of Market Research 51, no. 2 (March 2009): 219‐241. Cristina Maria Băltereţu and Delia Cristina Balaban, “Motivation in Using Social Network Sites by Romanian Students. A Qualitative Approach.,” Journal of Media Research 3, no. 1 (January 2010): 67‐74. Danah M. Boyd and Nicole B. Ellison, “Social Network Sites: Definition, History, and Scholarship.,” Journal of Computer‐Mediated Communication 13, no. 1 (November 2007): 210‐230. David Beer, “Researching a confessional society.,” International Journal of Market Research 50, no. 5 (2008): 619‐629. David F. Feldon and Yasmin B. Kafai, “Mixed Methods for Mixed Reality: Understanding Users’ Avatar Activities in Virtual Worlds,” Educational Technology Research and Development 56, no. 5‐6 (December 1, 2008): 575‐ 593. Fiona Redmond, “Social Networking Sites: Evaluating and Investigating Their Use in Academic Research.” (presented at the International Conference of Education, Research and Innovation, Madrid, 2010), http://arrow.dit.ie/cgi/viewcontent.cgi?article=1003&context=aaschlancon. Heath McDonald and Stewart Adam, “A comparison of online and postal data collection methods in marketing research,” Marketing Intelligence & Planning 21, no. 2 (2003): 85‐95. Jolene Zywica and James Danowski, “The Faces of Facebookers: Investigating Social Enhancement and Social Compensation Hypotheses; Predicting Facebook™ and Offline Popularity from Sociability and Self‐Esteem, and Mapping the Meanings of Popularity with Semantic Networks.,” Journal of Computer‐Mediated Communication 14, no. 1 (November 2008): 1‐34. Jordi Casteleyn, André Mottart, and Kris Rutten, “How to use Facebook in your market research.,” International Journal of Market Research 51, no. 4 (July 2009): 439‐447. Jyun‐Cheng Wang and Chi‐Lu Chen, “An Automated Tool for Managing Interactions in Virtual Communities ‐‐ Using Social Network Analysis Approach.,” Journal of Organizational Computing & Electronic Commerce 14, no. 1 (March 2004): 1‐26. Lon Safko, The Social Media Bible: Tactics, Tools, and Strategies for Business Success, 2nd ed. (John Wiley & Sons, 2010). Michael Trusov, Randolph E Bucklin, and Koen Pauwels, “Effects of Word‐of‐ Mouth Versus Traditional Marketing: Findings from an Internet Social Networking Site.,” Journal of Marketing 73, no. 5 (2009): 90‐102. Mike Cooke, Nick Watkins, and Corrine Moy, “A hybrid online and offline approach to market measurement studies.,” International Journal of Market Research 51, no. 1 (January 2009): 29‐48. Mike Cooke, Nick Watkins, and Corrine Moy, “A hybrid online and offline approach to market measurement studies.,” International Journal of Market Research 51, no. 1 (January 2009): 29‐48. Ray Poynter, “Facebook: the future of networking with customers.,” International Journal of Market Research 50, no. 1 (January 2008): 11‐12. RUTH PAGE, “Re‐examining narrativity: small stories in status updates.,” Text & Talk 30, no. 4 (July 2010): 423‐444. Tom Smith, “The social media revolution.,” International Journal of Market Research 51, no. 4 (July 2009): 559‐561.

14

 Appendix Facebook statistics as reported in the website.

15