Online Communities in Saudi Arabia: Evaluating the Impact on Culture Through Online Semi-Structured Interviews

Abstract:The aim of this article is to discuss the lessons learned from conducting semi-structured interviews online in an ethnographic study that took place in Saudi Arabia during the period 2001-2002. The purpose of the study was to explore individuals' participation in online communities in Saudi Arabia and also understand how online communities in Saudi Arabia are affecting participants' offline culture. Semi-structured online interviews were used to report the perceptions of 15 participants (8 females, 7 males) about their online community experience in Saudi Arabia.

After a brief background on Saudi culture, the article discusses the conceptual and methodological aspects of the study. It then discusses semi-structured interviews, highlighting the rationale for conducting them online. A discussion of sampling and participant selection follows. The article then discusses the recruitment of participants and the difficulties involved in this process. The ways in which the online interviews were conducted and the data were analysed are discussed next. The article concludes by offering a discussion of the lessons learned from using this technique.

Key words: online communities, web-based forums, asynchronous communication, Saudi Arabia, ethnography

Semi-structured interviews have long been conducted face-to-face in qualitative, and even in some quantitative, research because of their ability in gaining rich and in-depth experiential accounts from individuals. Computer-mediated communication (CMC) offers yet another medium through which semi-structured interviews can be conducted. CMC technology has made possible the conduct of research in offline environments such as conservative societies that were previously difficult to access. It is also now possible to study groups of people who are otherwise inaccessible¹⁾, such as females in the Saudi gender-segregated society. [1]

Semi-structured online interviews, and the lessons learned from using them in an ethnographic study, are the topics of the following discussion. The study was conducted in Saudi Arabia during the period 2001-2002. The purpose of the study was to explore individuals' participation in online communities²⁾ in Saudi Arabia and also understand how online communities in Saudi Arabia are affecting participants' offline culture. A total of fifteen semi-structured online interviews were conducted with eight female participants and seven male participants for the purpose of elucidating their perspectives on their online community experience in Saudi Arabia³⁾. [2]

Saudi society is largely conservative and religious. Islam plays a central role in defining the culture and determining the norms, values, attitudes, and practices of society (ALMUNAJJED, 1997). One of the important features that profoundly influence every aspect of public and social life in Saudi Arabia is the segregation of sexes. Segregation of the sexes is maintained physically, socially and psychologically. This segregation, which does not permit women to mix with unrelated men in Saudi Arabia, is prescribed by the Islamic religion (ALMUNAJJED, 1997; EMBER & EMBER, 1998; WHEELER, 2000). It is a general rule that applies to education, banking, public transportation and the work place. It also applies to restaurants, schools, and libraries. "The practice of segregation and confining women to their own company is an institutional mechanism designed to regulate women", to protect their chastity and to "prevent other men from encroaching on the male honour of the family" (ALMUNAJJED, 1997, p.8, p.34). [3]

One of the most striking consequences of the use of online communities in Saudi Arabia is that they have enabled males and females to communicate with each other in a way not possible before. While they remain physically segregated when communicating with each other online, their communication makes them, to some extent, overcome this gender separation. This makes the question of whether (or not) this form of communication is wrong, remain unanswered. While it is clear that face-to-face contact between women and unrelated men is wrong, it is not clear if online communication across gender lines is also wrong, particularly because the later lacks face-to-face contact. Perhaps more research on this technology will help answer this question. [4]

This was an interpretive, naturalistic study, the method of which was guided by a constructivist paradigm, which emphasises understanding the multiple meanings people make of the phenomenon under study. Social construct theory (one of two constructivist theories⁴⁾) provided the theoretical lens through which the method in this study was applied and the data were interpreted. Social constructivists recognise the effect of the social environment, culture and religion on how people construct their realities about their world (BERGER & LUCKMANN, 1967). They argue that meaning is developed through the interactions of social processes involving people, language and religion (BERGER & LUCKMANN, 1967). They see people as developing meanings for their activities together, that is, people "socially" construct reality (WILLIAMSON, 2000a, p.30). Religion and culture in Saudi Arabia not only shape people's attitudes, practices, and behaviours, but also shape the way they see and do things and perceive their lives. Similarly, the social environment, in the case of online communities "the web-based forum technology"⁵⁾, may also exact some influence on people's behaviour both online and offline. This makes social construct theory appropriate for understanding online communities in that society. Additionally, the use of this theory to investigate and understand online communities is also in line with the literature (see, for example, JONES, 1998, p.5; MARKHAM, 1998; COSTIGAN, 1999; FERNBACK,1999; DODGE & KITCHIN, 2001; MANASZEWICS, WILLIAMSON, & McKEMMISH, 2002). [5]

The method⁶⁾ used in this study was ethnography. Constructivism is one of a number of frameworks in which ethnography can be carried out⁷⁾. The purpose of ethnography is to produce a written description of a people's way of life, beliefs and daily activities (FETTERMAN, 1989). As this requires an in-depth understanding of the people, events, activities and social structures under study, ethnography conducted from within a constructivist paradigm appeared to be the best approach. One of the important aspects of constructivist ethnography is that it allows the eliciting of the individuals' perceptions about their experiences. Another aspect is that it allows multiple interpretation of realities and alternative interpretations of data to be presented (FETTERMAN, 1989). Again the use of ethnography to study online environments is advocated by many researchers including PACCAGNLLA (1997), HINE (1998, 2000), MARKHAM (1998), MILLER and SLATER (2000), NOCERA (2000), PREECE (2000), RHEINGOLD (2000), HAMMAN (2001) and MANN and STEWART (2000). Ethnography also provides constructivists with a variety of different techniques for data collection, one of which is semi-structured interviewing (SAULE 2000). [6]

Interviewing includes a wide variety of forms. For example, interviews can be structured, semi-structured and unstructured. In structured interviewing, the interviewer asks all respondents the same series of questions in the same order or sequence with a limited set of response categories offered (FONTANA & FREY, 2000, p.649). The interviewer records the responses according to a prepared coding scheme. In semi-structured interviewing (also called focused interviewing), the interviewer uses a guide or schedule developed around a list of issues that are central to the research question. (See Section 7.4 below for information on the schedule used.) The order of questioning is not fixed and the type of questions allows for greater flexibility than in the case of the structured interview (MINICHIELLO, ARONI, TIMWELL, & ALEXANDER, 1996, p.65). In unstructured interviewing, the interviewer dispenses with formal interview schedules and ordering of questions and relies on the social interaction between interviewer and informant to elicit information. The unstructured interview takes on the appearance of a normal everyday conversation. However, it is always a controlled conversation which is geared to the interviewer's research interests (MINICHIELLO, ARONI, TIMWELL, & ALEXANDER, 1996, p.65). [7]

Unstructured and semi-structured interviews are ways of doing in-depth interviewing. In-depth interviews have been used extensively as a major technique in qualitative research (MARKHAM, 1998; MARSHALL & ROSSMAN, 1999; NOCERA, 2000; WILLIAMSON, 2000a) due to their ability in gaining in-depth and rich accounts of individuals' perceptions about their lives (KUMAR, 1996; FONTANA & FREY, 2000). In-depth interviewing means that, rather than focusing on the researcher's perspective as the valid view, it is the informant's account which is being sought and is highly valued. Moreover, in-depth interviewing allows the researcher to elucidate the informants' world by understanding their perspectives in a language that is natural to them. [8]

For the purpose of revealing the perceptions of participants about their online community experience in Saudi Arabia and how that online experience affected them, semi-structured, in-depth interviews were considered appropriate.⁸⁾ [9]

There are many reasons why the semi-structured in-depth interviews, used in the project, were conducted online. First, it is very difficult to conduct interviews with females face-to-face or on the phone in Saudi Arabia because Saudi society, as mentioned before, is gender-segregated. It is also difficult to ask a female researcher or interviewer to help with interviewing females face-to-face, because the act of communicating with an unrelated female⁹⁾, either face-to-face or by telephone, is itself wrong according to the Saudi culture. This made conducting interviews online with females a strategic option. Second, and related to the first point, had interviews with males been conducted face-to-face and with females online, bias to the research findings may have been introduced as a result of the differences in the use of the techniques to collect data. Third, the anonymity inherent in the online medium often encourages people to disclose more about themselves (MOTLUCK, 1997 cited in HAMMAN, 1999; HAMMAN, 1999; PREECE, 2000). This made online interviewing a strategic tool to gather sensitive information in a society such as the Saudi, where individuals frown upon those inquiring into their personal affairs (ALMUNAJJED, 1997). [10]

The forum¹¹⁾ from which participants were recruited was an asynchronous public discussion web forum, hosted by one of the largest and fastest growing online service providers in Saudi Arabia.¹²⁾ The main reason for the selection of this site was the richness of information available, which is a requirement for the choice of a research site (MAXWELL, 1996). At the time of doing observation, the main page in this forum contained links to newly posted topics or messages only. Older topics resided in other pages that could be accessed through small links that were found on the main page. In addition to the title of a topic, each page contained information like the nickname of the author of the topic, the date the topic was posted, the number of times the topic was read and the number of replies to it. [11]

At the time of collecting data for this research there were approximately 10,000 members in the forum studied. But of those 10,000 only some, in the order of a few hundreds, were active in their interaction with others. The rest were either irregular or inactive or joined the forum for a short while and then left. Members discussed all types of topics in this forum. Table 1 lists some of the types of topics that were commonly discussed online. Religion was as important in the lives of these members as it is in their lives offline.

Table 1: Topics that dominated discussions in the studied online communities [12]

Most researchers prefer purposive sampling for interpretive studies (see, for example, LINCOLN & GUBA, 1985; CRESWELL, 1998; GLESNE, 1999; MARSHALL & ROSSMAN, 1999; CHARMAZ, 2000; LOCKE, SPIRDUSO & SILVERMAN, 2000; WILLIAMSON, 2000b). Purposive sampling allows researchers to choose cases that are representative of all sub-groups and personal characteristics which might be of interest to the study. Although there is no way of ensuring that the sample selected is representative in the probability sense, as WILLIAMSON (2000b) argues, BRYMAN (1988, p.90) says: "the issue should be couched in terms of the generalisability of cases to theoretical propositions rather than to populations or universes". That is, purposive sampling is driven by theoretical considerations rather than the need to count multiple cases of the same group. In this sense also, purposive sampling is related to theoretical sampling which is again related to the concept of theoretical saturation (the point when no new information is added) (WILLIAMSON, 2000b). Purposive sampling was adopted for the study, with the concept of saturation sampling being observed. Participants were recruited to the point when no new information was forth-coming. [13]

Participants for the interviews were selected from an asynchronous web-based forum, the interactions on which were observed by the researcher daily for twelve months before the interviews began. After receiving the ethics approval from Charles Sturt University to conduct online semi-structured interviews, a list of the nicknames of the participants in the observed community was made. The list did not include all the participants in the observed community because many participants come to the community and many others leave, and not all participants are active or regular in their interaction with others in the community, indicating their lack of commitment to the place and the group (Q. JONES, 1997). The inactive or irregular participants, although subscribers to the community, do not necessarily belong to it. The participants whose nicknames were listed were those who had been active or regular in their interaction with others in the community for a year or more. [14]

After the list was produced, some of the information known about the selected participants, noted from continuously observing the community where the participants operate, was placed in front of their nicknames. This information included age, education, gender, geographic location, marital status, time spent in the community and a comment about the regularity of their participation in the community. Most of these details were known either from participants' personal profiles¹³⁾ or from the details they provided when they introduced themselves to the community. The comment about the regularity of participation was made according to how the researcher viewed and perceived these participants, while continuously observing the community where the participants operate. This list was then divided into two main groups based on participants' gender. Since in the online medium it is difficult to be certain about the gender of a participant, gender of participants was noted according to what these participants said about themselves in the observed community, that is, how they presented themselves to others in the observed community. Moreover, by observing the community for a year, it was noted that gender switching among regular and committed participants was difficult. This is because when participants claim to be males, their male friends in the community may discover if they are not males when they seek to call them on the phone or meet them outside the community, as is often the case. The same applies when participants claim to be females. [15]

The male and female groups originally listed, were further organised into three sub-groups for each gender, according to participants' age. The first sub-group included female participants who were relatively young (15-20 years old). The second sub-group included female participants who were 21-25 years old¹⁴⁾. The third sub-group included female participants who were 26-50 years old. Similar sub-groupings to those of the female participants group were made with the male participants' group. (See Fig.1, below, for information on the process of sampling).

From each of the six sub-groups mentioned above (three sub-groups for males and three sub-groups for females), two participants were selected for online interviews, making a total of twelve interviewees. While the participants were selected in this way to allow a diverse range of characteristics to be represented, the sample is not large and "representative" of all the online community participants, nor is it intended to be. Using the logic of interpretive research, the researcher was more interested in understanding, in-depth, the experiences and perceptions of diverse individuals than in being able to generalise those experiences and perceptions to a larger population (PATTON, 1990; GLESNE, 1999; LOCKE, SPIRDUSO & SILVERMAN, 2000). When interviews were analysed and results were carefully studied, more information to further illuminate some of the interviewees' responses was needed. (For example, when the twelfth interviewee, a female, said that joining the community did not make her feel important in the society because on the forum only talk is exchanged, this new lead needed to be followed.) Thus, three more participants (a male and two females) were interviewed, making the total number of interviewees fifteen. (See Table 2 below for further information on details¹⁵⁾ of participants selected for online interviews.) Deciding to include additional participants was in harmony, not only with the concept of purposive/theoretical sampling but also, in general, with a study that is interpretive in nature and where the design of the research is flexible and emergent (LINCOLN & GUBA, 1985; PATTON, 1990; CHARMAZ, 2000). In fact, LINCOLN and GUBA (1985) caution that "the design of a naturalistic study ... cannot be given in advance; it must emerge, develop, unfold ..." (p.225). Here also the reader is reminded that the determining factor of the number of interviewees was saturation (the point when no new information is added). That is, interviews in this study were conducted until patterns emerged and started repeating. Details of all the participants, finally selected, are provided in Table 2.

As can be seen from Table 2, the six age groups were represented almost evenly with regard to gender. Marital status and, to some extent, geographic location could not be varied because the majority of participants were single and lived in the main cities. Also occupation variable could not be varied because the majority of participants in the forum were young and still involved in the educated sector. [18]

The first step was to prepare a list of the interviewees' Hotmail email addresses which the researcher recorded while conducting observation in the observed community. Participants on many occasions advertise their Hotmail addresses in the community for others to see, especially when they introduce themselves to others in the special introductory topics. [19]

The first way the researcher tried to invite selected participants to take part in online interviews was by email. This attempt failed as participants ignored messages sent to them via their email addresses in the community where they participated and email messages sent to them via their Hotmail addresses. The reason could be because they were afraid to open messages sent from an unknown sender for fear of viruses. Or it could be because participants chose not to waste time on someone they did not know. [20]

The researcher therefore devised another technique to approach them: he added selected participants to his list in MSN Messenger, which is a synchronous or real-time-chat-software available for download free from Microsoft to any Hotmail or MSN account holder. (Note that the Hotmail email address of anyone is the same address for the MSN Messenger.) How this worked was that, once interviewees were online, the software alerted the researcher to their online status. They were then approached and invited to take part in the interviews. [21]

The schedule consisted mostly of broadly open-ended questions with in-depth, probing sub-questions. The interviews addressed the same issues with mostly the same questions but the order of questions depended on participants' responses. Also, participants were asked different probing questions depending upon the need to build on previous supplied information or in order to follow particular leads. Interviewees were asked to talk about:

Before participants were asked if they agreed to be interviewed, a translated version in Arabic of the informed consent sheet was sent to them as a file through the MSN Messenger. After they were given a couple of minutes to read the informed consent sheet, if they agreed to the contents of the informed consent sheet, the interview began. None of the participants declined to take part in an online interview after reading the informed consent sheet. [23]

The questions to the interviewees were not typed in classical Arabic, but rather in the local Saudi dialect with which both the interviewees and the interviewer were familiar. This was so that the interviewees feel comfortable with the interviewer and the questions of the interview. There are many dialects in the Arabic language as each Arabic country has its own dialect. In the online community, participants can type the spoken Arabic according to their dialect¹⁶⁾. For example, a sentence like "Do you want to call home?" can be written in Arabic according to the country's dialect as:

During the process of the interview, the researcher purposely made it clear to the interviewees that he knew them¹⁷⁾ and knew a bit about their community. This was so that the interviewees feel comfortable to talk to him and be assured that he was a genuine researcher. Also during the process of the interview, the researcher answered questions raised by the interviewees regarding the researcher or the research. The researcher, throughout the interviews, assured interviewees that he was listening to them because they did most of the talking. The researcher achieved that by saying for example "Yes, go ahead, I am listening, ok, I see". [25]

Data were analysed as they were collected. First of all, data were translated in two steps. First, the researcher performed the translation of the passage based on his knowledge of the English language. Second, he double-checked his translations by cutting and pasting the passages into a web English dictionary, which offers translation from Arabic to English. The web English dictionary he used is located at http://www.ajeeb.com/. The translation covered the whole dialogue of the interview, except for a few of the things that were trivial or not useful, resulting from the attempts to encourage participants and to keep the interview going. After completion of the translation, the interview passages were arranged under headings to form sections. This was done to facilitate the analysis. The translated interview passages were then inserted as Rich Text Format into NVIVO for further analysis. [26]

After editing the documents in the NVIVO editor, they were read more than once so that a feel for the data was gained. Next, within each of these documents, keywords were identified after careful study of each word and each line in the document passages. Themes were then developed based on these keywords. To illustrate these last two steps, an example is used. Consider the following passage taken from an online interview with a female participant:

In the beginning, I was naïve. I looked at the members with good intentions. I treated them like my own brothers. Some of them did not appreciate that. Others took it in the wrong way. If I replied to a member through a poem, it meant I was in love with him. But that was in the beginning only. Now, things are different. Everyone is treated appropriately.

For the purpose of ensuring the appropriateness of these themes, each was defined and evaluated. The process of evaluation involved ensuring that certain phrases (or similar to them) are found in a participant's statement for a theme to be assigned to a participant's statement. (Table 3 shows an example of this process.)

The next step was assigning nodes in the NVIVO Coder representing the developed themes. Once these nodes were assigned, the node browser in NVIVO was invoked to view all segments of text, from all the interview documents, for the purpose of printing them out. Very importantly, the nodes stored all the data related to a particular theme from all these documents. That is, these nodes became buckets or baskets into which segments of text, from all the documents, related to a particular theme were placed. [29]

Following the printing out of the segments of text pertaining to a particular node, came careful reading and studying of these segments of text off screen to make sure that the themes developed from these passages were appropriate. Then came two important steps: first, all nodes were further structured or organised into groups based on the general concepts of the research they addressed. Second, all nodes were compared and contrasted to ensure that they were assigned appropriately. These methods of data reduction and inductive processing allowed the researcher to develop a theoretical description of online communities while at the same time grounding the description in the data (LINCOLN & GUBA, 1985; STRAUSS & CORBIN, 1998). [30]

Finally, for the purpose of producing a summary of the data, statements that encompassed the developed nodes (themes) were written. These statements were combined to form paragraphs that formulated the results. [31]

The following are some of the lessons learned from using online semi-structured interviewing in researching two online communities in Saudi Arabia. [32]

One of the important lessons learned is that online interviewing enabled the researcher to access a wider geographical area from which to select participants. The researcher was able to interview participants from different cities within the country, including Madinah, Jeddah and Riyadh, without moving from his chair. This would have been very difficult if interviews were to be conducted face-to-face. Also online interviewing enabled the researcher to interview groups of people, such as females, who are difficult to interview face-to-face, given the strict regulations against communication across gender lines in Saudi society. This made conducting interviews with females online the best technique available. [33]

Another important lesson learned is the benefit of observing the research site and collecting specific information about the potential interviewees before conducting the interviews. Interviewing people, whom the researcher was already aware of, from observing the site where they operate, gave confirmation about, for example, the true gender and age of the persons interviewed. This occurred because it was possible to ask the interviewees specific questions to verify that those interviewed over the MSN Messenger were the same participants identified in the observed community. Additionally, as mentioned earlier, interviewing participants who had been observed for a while, meant knowing something about them, which helped the researcher in making them feel more comfortable to talk to him and be more likely to help him. Related to this point also is the importance of observing the site in the identification of the most appropriate sample. Given that sampling was purposive, observing the site for a while allowed the researcher to choose cases that were representative of all sub-groups and personal characteristics. [34]

Another important lesson is that inviting participants in the studied online communities to take part in interviews through email did not turn out to be effective. All participants in this study ignored messages sent to them via their email accounts. As mentioned above, the reason could be that participants were worried that someone unknown might send a virus. It is also possible that participants chose not to respond to the emails because they thought of it as another Spam email. On the other hand, inviting participants to take part in interviews through the MSN Messenger appeared to be very effective. Approaching the interviewees through the MSN Messenger, when they were online, was similar to approaching people in the street and asking them for directions to a shopping mall. Participants found it hard to reject the idea of being of assistance to someone when they could. Their culture instructs them to help others if they can. In addition, the fact that the request that asked participants to take part in interviews was sent synchronously, that is, in real time, is quite important. Sending the request synchronously was similar to talking with the interviewees face-to-face. This could be the main reason participants agreed to be interviewed. This, however, does not mean that participants agreed to be interviewed because they felt embarrassed to reject a request the researcher sent them. Participants could have ignored the request completely as after all it came from a total stranger, or apologised by saying they were busy or not interested. They could have also simply and politely said "NO" to the request. It should also be noted that, before participants were asked if they agreed to be interviewed, a translated version in Arabic of the informed consent sheet was sent to them as a file through the MSN Messenger. Interviews began only if they agreed to interviewed and agreed with the contents of the informed consent sheet. [35]

The anonymity inherent in the communication medium no doubt made online interviewing effective and successful. First, anonymity allowed for better contact between the researcher and participants, in contrast to the suggestions of some researchers that an interviewer-interviewee "relationship" is difficult to establish online due to a lack of oral cues (FONTANA & FREY, 2000; MANN & STEWART, 2000). One reason for this could be because both the interviewer and interviewees were experienced users of this technology. Another reason could be because there are a number of features provided in the online medium, such as smiley faces, that compensated for the lack of social cues. Second, the CMC medium did not seem to result in a lack of social presence and its narrow bandwidth did not make communication over it less deep or less rich, which is again contrary to what some researchers, such as FONTANA and FREY (2000), claim to be the case. An example that illustrates this, is the following comment in which a female participant describes how she felt when she learned about the death of a male participant:

"I was in one of the family gatherings when my dear sister Igbal called me with a lot of pain hidden in her voice and asked where I was. I asked what is the matter, she said Naif died. Neither words nor feelings, not even expressions, rescued me. A tear choked in my throat. I do not know how I managed to leave that place and return home. I called Muna and told her [what happened]. At the same time I was crying all over." [36]

Third, anonymity reduced hierarchical features. Age, race, gender, wealth, physical appearance and status characteristics, which are very significant is Saudi Arabia were blurred online or became hard to notice. This may have encouraged participants to disclose more about themselves and their experiences. Given that hierarchy in Saudi society is significant, the observation that anonymity reduces hierarchical features was noticed even between the interviewees and the researcher. This may mean that CMC is an ideal medium for conducting unbiased research as suggested by MANN and STEWART (2000). Fourth, anonymity encouraged participants to be honest in their responses as they were not worried about their reputation and there was no social pressure on them to lie. Asking individuals in Saudi society sensitive questions usually makes them frown upon those inquiring into their lives. Also, anonymity made participants, especially females, feel less embarrassed or less uncomfortable to discuss some aspects of their online experience. The fact that they are largely unknown to others and their faces are absent online, could really save the participants from feeling uncomfortable or embarrassed in discussing any sensitive issues. [37]

One of the advantages of conducting interviews online is that the interviewer had time to phrase follow up questions, while the interviewees were composing their answers to earlier questions. The interviewer also used that time to quickly scan the information about the interviewees collected during observation. Knowing about the interviewees was an added advantage because it helped the interviewer "break the ice" with interviewees more easily. This in turn made the interviews smooth and more like conversations. Main questions can be typed, for example, in MSWord beforehand so that they are made ready thus saving time on typing. A checklist of questions was also used and a tick was made on each question when an answer for it was received. This avoided the interviewer forgetting to ask the interview questions. Also, the initial questions were typed in the MSN Messenger text area and sent in a "one-send" operation, so that questions received at the interviewee's MSN Messenger text area were complete. This way the interviewees were more likely to understand the question. [38]

Obtaining information related to a lived experience can be difficult if participants are asked the question only once. People, who are not naturally spontaneous, tend not to give adequate responses when they are suddenly asked a difficult or sensitive question. For this reason interviewees in this study were asked the main question twice. The first time at the beginning of the interview, the second before the end of it. [39]

Asking questions online was in some cases difficult. One of the difficulties arose when the interviewer received a response to an earlier question or comment after he had already sent the interviewee a new question. The time lag between the sending and receiving of messages appeared in some cases to complicate the taking of turns. This further resulted in interrupting the flow of the interviews. Knowing this, the researcher, however, tried to reduce its impact by being patient and understanding. [40]

Online semi-structured interviews have proven to be successful in researching online communities in Saudi Arabia. In addition to their low cost and speed of return and the benefits discussed in the lessons above, it was clear from the findings of the study that they were effective in providing in-depth and rich accounts of individuals' experiences. They enabled the researcher to engage the participants in reflective conversations that generated profound and detailed qualitative data. [41]

The interpretive-qualitative research methodology used in the study proved to be appropriate. The choice of the constructivist paradigm to study online communities in Saudi Arabia enabled the deep exploration of the online communities in their cultural context to be undertaken. The lens of social construct theory was useful for understanding how the members collectively behaved online and how they developed together the meanings about their online experiences. The theory was particularly useful for understanding the ways in which the web-forum technology affected members' behaviour online and offline. [42]

As shown above, the benefits of using online semi-structured interviews are numerous. Most notably is the fact that they allowed the researcher the opportunity to select participants from a wider geographical area and from groups of people who are difficult to interview otherwise. They also provided a rich medium for conducting in-depth interviews, contrary to what some researchers suggested that the CMC medium lacks social presence and its narrow bandwidth may make communication over it superficial. One of the interesting conclusions drawn from this research also is that inviting participants to take part in interviews through the MSN Messenger appeared to be more effective than inviting them through email. [43]

3) For information about the results of this research see (AL-SAGGAF, WECKERT, & WILLIAMSON, 2002a, 2002b, 2002c; AL-SAGGAF, 2004). <back>

4) Another constructivist theory is the personal construction of reality. <back>

5) A forum is a specific type of software that supports the facilitation and implementation of discussion groups through the exchange of messages via computer networks (CSU, 2004). <back>

6) This researcher takes the position of distinguishing between the terms "technique" and "method". The former should mean the specific procedures that are used to collect or analyse data. The latter should mean the general rules that govern the implementation of these procedures. <back>

7) Other frameworks are critical theory and post-modern ethnography (SAULE, 2000). <back>

8) Other techniques used in the study were silent observation of an online community; a participant role by the researcher in another similar online community and face-to-face semi-structured interviews with key informants. <back>

9) While it is wrong, from a religious point of view, to communicate with an unrelated female on this matter, it is possible to communicate with a blood related female, but such a person was not available to this researcher. <back>

10) The term "online forum" should not be thought of as a synonym for the term "online community". The former should be understood as a "platform", "place", or "environment". The latter should be understood here as "social aggregations that emerge from the Net when enough people carry on those public discussions long enough, with sufficient human feeling, to form webs of personal relationships in cyberspace" (RHEINGOLD, 2000). This means the forum is "platform" of which a community exist around. <back>

11) For ethical considerations, the name of the forum under study and its URL were omitted as KING (1996) and EYSENBACH and TILL (2001) suggest should be the case when carrying out research on Internet communities. <back>

12) The researcher, Dr Yeslam AL-SAGGAF, originally comes from Yemen. He is a permanent resident of Saudi Arabia and a 29 year old unmarried man. He holds a degree in Engineering from Malaysia and a Masters and a PhD in Information Technology from Charles Sturt University, Australia, where he now works as a lecturer. Having spent many years in Saudi Arabia, he is very familiar with the Saudi society and its culture. In addition, Yemen and Saudi Arabia not only share the same language and religion but also most of the cultural characteristics. <back>

13) At the time of registration, members are expected to record in their personal profiles some of their details, such as, age, gender, and geographic location. <back>

14) The majority of participants in the online community were under thirty years old, which is consistent with research elsewhere—see, for example, WHEELER (2002). For this reason the age span (26-50) was appropriate. <back>

15) To preserve anonymity, the names given here are fictitious, although other details are actual, except for city names. While city names are real, they are swapped across interviewees. <back>

16) Despite the use of local dialects, people in most of the Arabic countries can still understand each other. <back>

17) The researcher did not ask the interviewees about their age or occupation for example, because these details he already obtained from continuously observing the community where they operated. <back>

Al-Saggaf, Yeslam (2004). The Effect of Online Community on Offline Community in Saudi Arabia. Electronic Journal of Information Systems in Developing Countries (EJISDC), 16, 1-16. Retrieved August 21, 2004, from http://www.ejisdc.org/.

Al-Saggaf, Yeslam; Weckert, John & Williamson, Kirsty (2002a). What Do Individuals in Saudi Arabia Say About their Participation in Online Communities. WWW/Internet 2002 International Conference, 13-15 November, Lisbon, Portugal.

Al-Saggaf, Yeslam; Williamson, Kirsty & Weckert, John (2002b). Online Communities in Saudi Arabia: An Ethnographic Study. The Thirteenth Australasian Conference on Information Systems ACIS, 4-6 December, Melbourne, Australia.

Al-Saggaf, Yeslam; Weckert, John & Williamson, Kirsty (2002c). The Effect of Online Community Experience on Individuals' Offline Lives: A Saudi Arabian Perspective. Electronic Networking 2002 Building Community Conference, 3-5 July, Melbourne, Australia.

Berger, Peter L. & Luckmann, Thomas (1967). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. New York: Anchor Press.

Bryman, Alan (1988). Quantity and Quality in Social Research (1st ed.). (Contemporary Social Research). London :Unwin Hyman.

Charmaz, Kathy (2000). Grounded Theory: Objectivist and Constructivist Methods. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of Qualitative Research (2nd ed.) (pp.509-535). Thousand Oaks, California: Sage.

Costigan, James T. (1999). Introduction: Forests, Trees, and Internet Research. In Steven G. Jones (Ed.), Doing Internet Research: Critical Issues and Methods for Examining the Net (pp.xvii-xxiv). Thousand Oaks, California: Sage.

Creswell, John W. (1998). Qualitative Inquiry and Research Design: Choosing Among Five Traditions. Thousand Oaks, California: Sage.

Ember, Carol R. & Ember, Melvin (1988). Anthropology. New Jersey: Prentice Hall.

Eysenbach, Gunther & Till, James E. (2001). Ethical Issues in Qualitative Research on Internet Communities. British Medical Journal, 323(7321), 1103-1105.

Fernback, Jan (1999). There Is a Their There: Notes Toward a Definition of Cybercommunity. In Steven G. Jones (Ed.), Doing Internet Research: Critical Issues and Methods for Examining the Net (pp.203-220). Thousand Oaks, California: Sage.

Fetterman, David M. (1989). Ethnography: Step by Step (Applied Social Research Methods Series). Thousand Oaks, California: Sage.

Fontana, Andera & Frey, James H. (2000). The Interview: From Structured Questions to Negotiated Text. In Norman K. Denzin & Yvonna S. Lincoln (Eds.), Handbook of Qualitative Research (2nd ed.) (pp.645-672). Thousand Oaks, California: Sage.

Glesne, Corrine (1999). Becoming Qualitative Researchers: An Introduction (2nd ed.). New York: Longman.

Hamman, Robin (1999). The Online/Offline Dichotomy: Debunking Some Myths about AOL Users and the Effects of Their Being Online Upon Offline Friendships and Offline Community (MPhil thesis, Liverpool University, 1999).

Hamman, Robin (2001). Computer Networks Linking Network Communities. In Chris Werry & Miranda Mowbray (Eds.), Online Communities: Commerce, Community Action, and the Virtual University (pp.71-95). New Jersey: Hewlett-Packard.

Hine, Christine (1998). Virtual Ethnography. IRISS '98 (pp.1-7). 25-27 March. Bristol, UK.

Jones, Quentin (1997). Virtual -Communities, Virtual Settlements & Cyber-Archaeology: A Theoretical Outline, JCMC, 3(3). Retrieved August 24 2004, from http://www.ascusc.org/jcmc/vol3/issue3/jones.html.

Jones, Steven G. (1998). Information, Internet, and Community: Notes Toward an Understanding of Community in the Information Age. In Steven G. Jones (Ed.), CyberSociety 2.0: Revisiting Computer-Mediated Communication and Community (pp.1-35). Thousand Oaks, California: Sage.

Jones, Steven G. & Kucker, Stephanie (2001). Computers, the Internet, and Virtual Cultures. In James Lull (Ed.), Culture in the Communication Age (pp.212-225). London: Routledge.

King, Storm A. (1996). Researching Internet Communities: Proposed Ethical Guidelines for the Reporting of Results. The Information Society, 12(2), 119-128.

Kumar, Ranjit (1996). Research Methodology: Step-By-Step Guide for Beginners. Melbourne: Addison Wesley Longman Australia.

Lincoln, Yvonna S. & Guba, Egon G. (1985). Naturalistic Inquiry. Newbury Park, California: Sage.

Locke, Lawrence F.; Spirduso, Waneen W. & Silverman, Stephen J. (2000). Proposals That Work (4th ed.). Thousand Oaks, California: Sage.

Manaszewicz, Rosetta; Williamson, Kirsty & McKemmish, Sue (2002). Breast Cancer Knowledge Online: Towards Meeting the Diverse Information Needs of the Breast Cancer Community, Electronic Networking 2002 Building Community Conference, 3-7 July, Melbourne, Australia.

Mann, Chris & Stewart, Fiona (2000). Internet Communication and Qualitative Research: A Handbook for Researching Online. London: Sage.

Markham, Annette N. (1998). Life Online: Researching Real Experience in Virtual Space. Walnut Creek, California: AltaMira.

Marshall, Catherine & Rossman, Gretchen B. (1999). Designing Qualitative Research (3rd ed.). Thousand Oaks, California: Sage.

Maxwell Joseph A. (1996). Qualitative Research Design: An Interactive Approach (Applied Social Research Methods series). Thousand Oaks, California: Sage.

Miller, Daniel & Slater, Don (2000). The Internet: An Ethnographic Approach. Oxford: Berg.

Minichiello, Victor; Aroni, Rosalie; Timwell, Eric & Alexander, Loris (1996). In-Depth Interviewing (2nd ed.). Melbourne: Longman.

Nocera, Jose L.A. (2000). Ethnography and Hermeneutics in Cybercultural Research: Accessing IRC virtual communities. In Fay Sudweeks & Charles Ess (Eds.), Cultural Attitudes Towards Communication and Technology (pp.163-173). Murdoch University, Australia.

Paccagnella, Luciano (1997). Getting the Seats of Your Pants Dirty: Strategies for Ethnographic Research on Virtual Communities, JCMC, 3(1). Retrieved August 24 2004, from http://www.ascusc.org/jcmc/vol3/issue1/paccagnella.html.

Patton, Michael Q. (1990). Qualitative Evaluation and Research Methods (2nd ed.). Thousand Oaks, California: Sage.

Rheingold, Howard (2000). The Virtual Community: Homesteading on the Electronic Frontier (Revised ed.) Cambridge: MIT Press.

Saule, Solveiga (2000). Ethnography. In Kirsty Williamson (Ed.), Research Methods for Students and Professionals: Information Management and Systems (pp.159-176). Wagga Wagga: Centre for Information Studies, CSU.

Strauss, Anselm & Corbin, Juliet (1998). Basics of Qualitative Research: Techniques and Procedures for Developing Grounded Theory. Thousand Oaks, California: Sage.

Wheeler, Debora (2000). New Media, Globalization and Kuwaiti National Identity. Middle East Journal, 54(3), 432-448.

Wheeler, Debora (2002). Islam, Community and the Internet: New Possibilities in the Digital Age, Journal of Education, Community and Values, 3. Retrieved April 23 2003, from http://bcis.pacificu.edu/journal/2002/03/islam.php.

Williamson, Kirsty (2000a). The Two Major Traditions of Research. In Kirsty Williamson (Ed.). Research Methods for Students and Professionals: Information Management and Systems (pp.25-48). Wagga Wagga: Centre for Information Studies, CSU.

Williamson, Kirsty (2000b). Sampling. In Kirsty Williamson (Ed.). Research Methods for Students and Professionals: Information Management and Systems. Wagga Wagga: Centre for Information Studies, CSU.

Dr. Yeslam AL-SAGGAF, Lecturer in Information Technology, Associate Course Coordinator, Master of Information Technology, School of Information Studies, Charles Sturt University

School of Information Studies
Charles Sturt University
Wagga Wagga
NSW Australia 2678

Dr. Kirsty WILLIAMSON, Director, Information and Telecommunications Needs Research Monash University and Charles Sturt University

Monash University
Level 7 26 Sir John Monash Dve PO Box 197 Caulfield, VIC. 3145
Australia

Al-Saggaf, Yeslam & Williamson, Kirsty (2004). Online Communities in Saudi Arabia: Evaluating the Impact on Culture Through Online Semi-Structured Interviews [43 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 5(3), Art. 24, http://nbn-resolving.de/urn:nbn:de:0114-fqs0403247.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License

Participants' Names	Gender	Age	Occupation	Geographic location	Martial status	Time spent on the forum
Hanan	F	18	University	Jeddah	Single	Two years
Wafa	F	15	School	Riyadh	Single	Two years
Arwa	F	19	University	Madinah	Single	Two years
Safa	F	18	School	Madinah	Single	Two years
Nadia	F	21	University	Riyadh	Single	Two years
Reem	F	21	University	Riyadh	Single	Two years
Nada	F	26	University	Riyadh	Single	Two years
Asrar	F	45	Housewife	Riyadh	Married	Two years
Ali	M	20	No data	Riyadh	Married	18 months
Faisal	M	20	University	Riyadh	Single	Two years
Nawaf	M	24	University	Riyadh	Single	Two years
Ibrahim	M	25	Employed	Tabook	Single	18 months
Ahmed	M	26	Employed	Taif	Single	Two years
Samy	M	26	University	Jeddah	Married	18 months
Mohamed	M	26	University	Jeddah	Single	Two years

Theme	Definition (from within the context of this study)	Phrases containing key words that must be found in participants' statements
Increased awareness	When participants indicate that they gained knowledge or realisation of something in lives after they joined online communities.	"I realised that", "I learned something new", "I used to think that", "now I am different", "I have now changed", "I did not know before that", etc.
Open-mindedness	When participants indicate that they showed flexibility in their views, ideas and beliefs after their online experience.	"everyone has a different point of view", "I now listen to everyone", "I used to force my ideas", "I now listen to both sides of the story", "I now consider different views" , I am now open to new ideas", etc.
Self-confidence	When participants indicate that they felt more confident in themselves and their abilities after their online experience.	"I became more confident", "my confidence in myself increased", "before I was afraid", "it gave my ego a boost", etc.
Intellectual confusion	When participants indicate that they felt confused about their ideals and beliefs after their online experience.	"Exposed to different beliefs", "influence on my culture", "whose writings against my religion", "spread unacceptable values", "people with different ideologies"