Practical Guidelines for Qualitative Research Using Online Forums

Abstract

With an increasing number of Internet research in general, the number of qualitative Internet studies has recently increased. Online forums are one of the most frequently used qualitative Internet research methods. Despite an increasing number of online forum studies, very few articles have been written to provide practical guidelines to conduct an online forum as a qualitative research method. In this paper, practical guidelines in using an online forum as a qualitative research method are proposed based on three previous online forum studies. First, the three studies are concisely described. Practical guidelines are proposed based on nine idea categories related to issues in the three studies: (a) a fit with research purpose and questions; (b) logistics; (c) electronic versus conventional informed consent process; (d) structure and functionality of online forums; (e) interdisciplinary team; (f) screening methods; (g) languages; (h) data analysis methods; and (i) getting participants’ feedback.

Qualitative Online Forums: Practical Guidelines

With advances in computer and Internet technologies, the number of Internet studies both quantitative and qualitative has recently increased in nursing.¹ However, because of the short history of Internet research, there exist few guidelines for Internet research in general. Furthermore, most existing guidelines for Internet research tend to focus on Internet quantitative studies in the forms of Internet surveys and/or questionnaires, and very few guidelines for Internet qualitative research exist in the literature.^{2, 3} Because of paradigmatic differences between quantitative and qualitative research, guidelines for Internet qualitative research should be different from those for Internet quantitative studies. For example, quantitative research emphasizes researchers’ objective stance while qualitative research is based on close relationships between researchers and research participants.^{4, 5} Also, quantitative research aims to find explanatory laws while qualitative research aims at in-depth descriptions of phenomenon.^{4, 5} In addition, while quantitative research operationalizes what is a truth to develop universal laws ^{4, 5} qualitative research explores what is a truth within the dynamic contexts of reality.^{4, 5} Thus, it is imperative to develop independent guidelines for Internet qualitative research.

Considering that research methods used by qualitative researchers are mainly for interviews, observation, and/or document analyses,⁴ Internet qualitative research methods can be easily defined as Internet methods that are used to collect qualitative data for interviews, observation, and/or document analyses. For instance, Web-pages could be easily used to conduct a quantitative Internet survey, but they could be also used to observe group dynamics with or without interacting with the group members as a qualitative research method. Qualitative researchers could login to the Web-pages, read and analyze the posted messages, and track the postings by different users to explore interactions among the members, while posting or not posting messages to the Web-pages. Thus, Internet methods that are used to collect qualitative data through interviews, observations, and/or document analyses can be Internet qualitative research methods.

In a recent review of Internet qualitative research,⁵ it was found that online forums were one of the most frequently used qualitative Internet research methods. Online forums refer to Internet sites for discussion where users can post messages on specific topics and interact with other users to discuss the topics. Online forums are known to be commonly used by diverse groups of users for elective emotional and informational support.^{6, 7} Online forums are also reported to supply sheltered environment for some to discuss delicate private health issues,⁷ and have been suggested as an excellent option for a qualitative study.⁸

Despite an increasing number of online forum studies, very few articles have been written to provide guidelines to conduct an online forum as a qualitative research method.^{2, 3} In this paper, practical guidelines in using an online forum in qualitative studies are proposed based on three previous online forum studies. First, characteristics of online forums as a qualitative research method are concisely described. Then, the three studies that are the bases for the guidelines for an online forum as a qualitative research method are described. Finally, included are the guidelines for future research using an online forum as a qualitative research method.

Characteristics of Online Forums

A researcher-led online forum is a discussion site on the Internet where users can discuss specific topics through posting a series of messages and a researcher moderate the discussion.^8a The message posted on the discussion site could be used as the data for analysis and the interactions made among the users could also be used as the observation data. The online forum is not just analyzing an existing online forum, but rather it is a forum intentionally established by a research team as a method to study a topic. Usually, participants are recruited through Internet communities/groups and/or through community settings (e.g., hospitals, community health centers, clinics). Typically, when participants visit the online forum site, they are assigned with user IDs and passwords through which they could access the online forum site. Researchers usually post discussion topics and/or prompts so that the participants could initiate their discussion on the specific topics and/or prompts.

Compared with traditional focus groups, online forums have unique characteristics that influence the relationships between researchers and research participants.^{3, 9} First of all, an easily noticeable characteristic is their non-face-to-face interactions. Due to non-face-to-face interactions in online forums, researchers and research participants cannot see each other in person. Thus, researchers cannot even determine if the research participants are actually those whom they target, especially when the inclusion criteria depend on physical characteristics such as race, ethnicity, and gender. Indeed, researchers pointed out that Internet interactions made corporeal characteristics vague so that researchers could not affirm social groups of research participants.³ Subsequently, concerns on authenticity of research participants have frequently been raised. However, some researchers emphasized that research participants’ corporeal characteristics were so deeply articulated to their opinions and perspectives that their identities could be easily divulged in their Internet interactions.¹⁰ Also, others pointed out that the use of cameras and microphones could contribute to eliminating this concern on authenticity in some ways.⁹ Although many traditional research studies without face to face contacts with participants (e.g., surveys for example) exist, this characteristics is particularly important for qualitative online forums because qualitative research is based on trust-building and intense interactions with participants and because it aims to get in-depth data on the participants’ experiences and opinions that frequently require an understanding of the contexts that are circumscribed by corporeal characteristics.

Asynchronous interactions could be another unique characteristic of online forums. In qualitative research, close interactions and short distances between researchers and research participants are essential to establish trust between them and to get credible rich data.¹¹ In online forums, it is difficult for researchers to get prompt and immediate responses because the interactions are asynchronous. However, this characteristic could be beneficial for qualitative researchers because the participants could directly communicate with researchers without restrictions in time zones and geographic distance.^{12, 13} Asynchronous Internet interactions can make research participants on different schedules or in distant time zones communicate easily.¹² Indeed, researchers have reported that research participants could more easily communicate with researchers by directly asking questions and getting answers.^{14, 15}

Another unique characteristic of online forums could be natural selection of specific participants because all the participants need to have access to the Internet and can use the Internet. Despite recent changes in demographics of Internet users, Internet users still tend to be a selected group of people who tend to be well-educated, literate, and skilled users of computers and have access to the Internet.¹⁶ Fallows¹⁷ reported that more than a half of the Internet users are women and online populations still tend to be highly educated, predominantly white, and younger participants. Furthermore, compared with simple Internet survey questionnaires, online forums require more advanced computer skills because online forums usually require the users to register and login using usernames and passwords.

Finally, online forums have unpredictable security issues as in other Internet research methods.^{14, 18} Researchers pointed out that no Internet interactions can be 100% secure; Internet research including online forums frequently has an issue of intrusion by outsiders.⁹ If a research participant registers with a website of an online forum, then the webmaster of the online forum site could monitor and track what Internet users view or spend online, which means that the information could be easily passed on to third parties if the webmaster wants.

The Studies

The three studies that provided the basis for the guidelines that are proposed in this paper include: (a) one study on gender and ethnic differences in cancer pain (Study 1); (b) another study on ethnic differences in menopausal symptoms (Study 2); and (c) the third study on ethnic differences in midlife women’s attitudes toward physical activity (Study 3). Each study is concisely described as follows.

Study 1 aimed to develop a standardized decision-support computer program (DSCP) that can be used by nurses as an assessment support tool for dealing effectively with gender and ethnic differences in cancer pain experiences based on cancer patients’ own views and experiences.^{19, 20} The study included two phases: (a) data collection, and (b) development of the DSCP for cancer pain management. The data collection phase included an Internet survey of 480 cancer patients (148 Whites, 105 Hispanics, 109 African Americans, and 118 Asians) and four ethnic-specific online forums of 82 cancer patients (29 Whites, 15 Hispanics, 11 African Americans, and 27 Asians). The Internet survey included multiple instruments to measure cancer pain, symptoms accompanying with pain, and functional status. For the online forums, nine online forum topics related to cancer pain experiences were used to guide the 6-month online forum. The quantitative data were analyzed using descriptive and inferential statistics, including ANOVA and hierarchical multiple regression analyses; and the qualitative data were analyzed using thematic analysis. Then, based on the data, the DSCP-CA was developed, and an online forum for oncology nurses was also conducted to preliminarily evaluate the DSCP. For this paper, only the findings from the qualitative part of the study (qualitative online forums) were used. More detailed information on the study can be found in other publications by the authors.^{19, 20}

Study 2 aimed to determine the differences in the menopausal symptom experience reported by women in the four most common ethnic groups in the U.S. (Hispanic, White, African American, and Asian) and explore the ethnic-specific contexts of women’s daily lives that influence their self-reported menopausal symptom experience.^{21, 22} The study had two components: (a) an Internet survey of 512 midlife women from four major ethnic groups in the U.S. (120 Hispanics, 160 Whites, 121 African Americans, and 111 Asians); and (b) four ethnic-specific online forums involving 87 midlife women (27 Hispanics, 23 Whites, 20 African Americans, and 17 Asians). The instruments for the Internet survey included questions on background, self-reported ethnic identity, health and menopausal status and menopausal symptoms, and an interview protocol on perceived causes, meanings, and management strategies for menopausal symptoms. In the online forums, seven topics related to the menopausal symptom experience were discussed. The quantitative data were analyzed using descriptive and inferential statistics, including ANOVA and multiple regression analyses; the qualitative online forum data were analyzed using thematic analysis. For this paper, only the findings from the qualitative part of the study (qualitative online forums) were used. More detailed information on the study can be found in other publications by the authors.^{21, 22}

Study 3 aimed to explore attitudes of midlife women from four ethnic groups [Hispanic, Non-Hispanic (N-H) White, N-H African Americans, and N-H Asians] toward physical activity.^{23, 24} This study included: (a) a quantitative Internet survey and (b) four ethnic-specific online forums. The Internet survey was conducted among 552 midlife women (160 Whites, 129 Hispanics, 138 African Americans, and 125 Asians), and four ethnic-specific online forums involved 90 women (29 Whites, 23 Hispanics, 21 African Americans, and 17 Asians) recruited from the Internet survey participants. The instruments for the Internet survey included questions on background characteristics and health status, the Physical Activity Assessment Inventory, the Modified Barriers to Health Activities Scale, the Questions on Attitudes toward Physical Activity, Subjective Norms, Perceived Behavioral Control, and Behavioral Intention, and the Kaiser Physical Activity Survey. For the online forums, ten topics on women’s attitudes toward physical activity and 7 topics on ethnic specific contexts influencing women’s participation in physical activity were used. The quantitative data were analyzed using descriptive and inferential statistics, including ANOVA, correlation analyses, and hierarchical multiple regression analyses; and the qualitative data were analyzed using thematic analysis. For this paper, only the findings from the qualitative part of the study (qualitative online forums) were used. More detailed information on this study can be found in other publications by the authors.^{23, 24}

Methods

Throughout the research process of the three studies, research staff recorded issues as they arose and wrote memos regarding the issues in conducting the online forums. These memos, as well as written records of discussions during weekly group meetings, were reviewed and analyzed using the content analysis methodology suggested by Weber.²⁵ The unit of analysis was individual words, and the memos and written records were analyzed using line-by-line coding. Then, the codes were categorized according to the contents while trying to identify issues related to online forum administration as a qualitative research method. Finally, idea categories were developed from the categorization process. During the analysis process, nine idea categories were extracted: (a) a fit with research purpose and questions; (b) electronic versus conventional informed consent process; (c) structure and functionality of online forums; (d) interdisciplinary team; (e) screening methods; (f) languages; (g) logistics; (h) data analysis methods; and (i) getting participants’ feedback.

Practical Guidelines for Qualitative Online Forum Studies

Specific guidelines, based on the extracted nine idea categories related to issues in administration of online forms as a qualitative research method, are proposed as follows.

A Fit with Research Purpose and Questions

When planning a qualitative online forum study, the first thing to do would be to check if the research purposes and questions actually require the study to be an online forum study. Although the use of online forums could be viewed as innovative, the method has some inherent limitations due to its unique characteristics that were described above. Also, some research purpose and questions could be adequately addressed with conventional research methods. If a conventional research method works well for the research purpose and questions, there is no need to conduct an online forum study that has a number of shortcomings such as lack of theoretical saturation and difficulties in trust building. In Studies 1 to 3, we chose the online forums as a qualitative component of the study because we needed a national approach to recruit an adequate number of ethnic minority participants and because the participants resided in nationally dispersed areas. Thus, without the use of online forums, the researchers or participants would have made long-distance trips to generate the qualitative data, which would have increased the study cost. Also, some participants of Study 1 (Asian Americans) had stigma attached to cancer, which might have inhibited their participation in in-depth discussion on cancer pain experience in face-to-face interactions. Some participants of Study 2 (Asian Americans and Hispanics) tended not to discuss their menopausal symptom experience with strangers, including researchers, which might have prevented their participation in face-to-face discussions on menopausal symptom experience. However, if a study could be conducted in a local area and there is no stigma attached to the specific condition of interest, traditional focus groups would likely work better than online forums.

Logistics

Researchers need to decide detailed logistics of online forums in the planning stage. In all three studies, detailed procedures for the online for online forums were pre-set and followed. When any issue in following the pre-set procedures came up during the actual online forum process, the research team discussed and resolved the issues through their research team meetings. The detailed logistics for the online forums in the three studies were as follows. First, in all three studies, the participants were recruited from among the Internet survey takers while asking if they would be interested in participating in additional online forums. About 80% of the Internet survey takers agreed to participate in the online forums, and we did not have any issues in recruiting the online forum participants mainly because they were recruited among the existing participants of the Internet surveys. When a participant agreed to participate in the online forums, the participants were asked to be registered, and IDs and initial passwords were emailed to them with a registration confirmation message. They were asked to choose pseudonyms for the online forum discussions so that their real names could not be identified by other participants. The IDs and passwords must be used whenever participants log in to the online forum sites. Also, their visits at the online forum sites were recorded, monitored and controlled. Only those who register were allowed to enter the online forums to ensure confidentiality and protect privacy. In all three studies, when adequate numbers of participants (a minimum of 30 per forum) were recruited, one of the researchers sent an e-mail to registered participants to tell them the forums had begun. The opening page of the online forums showed the introductory questions so that participants could introduce themselves with their pseudonyms when they first visited. These questions, along with discussions topics, were posted serially on the forum sites (3–4 topics per month) and remained for the whole 6 months. Participants could post messages about the topics at their convenience in any form they wanted (e.g., stories, conversations, responses to others’ messages, etc.). The researcher in charge of online forums steered discussions about the online topics throughout the 6-month period using related prompts as needed, always considering the content and flow of discussion. The number and length of messages were not limited, yet at least one message per topic was required for reimbursement of participation. In all three studies, we provided a $50 gift card per participant at the completion of the study when they posted one message per topic for all topics. Although there was an initial concern of this monetary reimbursement (e.g., attracting those in financial needs or unauthentic cases), the reimbursement itself was not a big issue in the online forums mainly because the participants should stay for the 6-month period to get reimbursed. This was a big contrast to the reimbursement of Internet surveys (a $10 gift card per participant for an one-time Internet survey) that attracted several unauthentic cases (e.g., six midlife women living in one mailing address). Probably, the 6-month period was long enough to not attract unauthentic cases. During the 5th month, the participants were asked to add topics that they wanted to discuss with other participants, and the added topics were available on the online forum sites for the remainder of the 6 months; 1–2 additional topics were added by the participants. The discussions from online forums were summarized monthly and posted on the online forum sites so that participants could see and provide feedback.

In all three studies, the roles of research team members in the online forums were clearly set at the beginning stage. During the online forum process, two researchers were intensely involved in each of the forums; one engineering researcher oversaw daily monitoring and control of access to the online forum sites, and the other researchers oversaw electronic communications with participants, building trust, steering discussions about the online forum topics, and carefully reviewing messages twice a week. In posting feedback messages, researchers encouraged the discussion to flow naturally with a minimum of interruptions, subtly guided the proceedings when necessary, were comfortable with displays of emotion, welcomed diversity of opinion, and remained nonauthoritarian and nonjudgmental.

Electronic versus Conventional Informed Consent

When planning a qualitative online forum study, one of the major issues to consider is how to get informed consent from their participants. In all the three studies, researchers used electronic informed consent because the participants resided in different geographical areas throughout the nation. The electronic informed consent was obtained by asking potential participants to visit the project website, click into the informed consent sheet on the web-site, review the informed consent sheet, and click a button saying “I agree to participate.” The process was automatic by programming the website. Although we expected that the study could be easily approved by the IRB committees, the reality was not easy in Study 1. The research team needed to attend a full board IRB meeting and explained that there would not be any risks other than those experienced in daily lives. The IRB office asked the researchers to place several additional security statements on the project website such as “If, during the online forum process, the researchers find a participant at risk for harm (e.g., suicide, threats to kill, harm to their own and others’ health), at the decision of the researchers, the data may be disclosed to the police” and “If incidents of this sort happen, they will be reported to the IRB and the NIH.” Unlike an Internet survey (quantitative research method), an online forum involves human interactions that could place a participant at risks for harm as in focus group discussions. Thus, compared with an Internet quantitative survey, stricter and more rigorous measures to ensure safety of participants needed to be made for the online forums.

In Study 3, the online forums involved discussion on physical activity that added an additional dimension to safety issues of the online forums. Because the participants could increase their physical activity during the online forums (by the influences of their discussions on benefits of physical activity), researchers needed to add two cautionary measures in the online forums. First of all, all the participants were asked to consent to keep their usual level of physical activity because a sudden increase of physical activity could increase some cardiovascular and musculoskeletal risks. Also, to prevent the potential risks, the messages posted on the online forums were monitored daily by a registered nurse under the supervision of a nurse practitioner, and any potential and actual adverse reactions (e.g., uncontrolled blood pressure, angina symptoms, etc.) were identified. Therefore, before starting the online forums, the participants were asked if they had a regular source of health care so that the researcher could refer them if there were to be an adverse event identified during the online forum. If the participants did not have a regular source of health care, the researcher got on the web and found a clinic or health care centers near the participants’ locations that they could go to for any health care, if needed. Also, at the end of the study, the participants were provided with information on what they would need to do prior to becoming more physically active.

Structure and Functionality of Online Forums

Researchers need to decide the structure and function of the online forums during the planning process. In all three studies, interruption by the researchers was minimized to make the participants interact in a more natural way, and the researchers took a role of facilitator of discussion rather than moderating the discussion. However, in online forums, researchers could be more active in the discussion by acting in a moderator role so that researchers could draw out the in-depth data that they aim at. In all three studies, using a feminist perspective, researchers aimed at having the research participants disclose their own experience rather than targeting specific information related to the researchers’ interests. If a researcher had taken a different perspective, he/she could have taken a different role of moderator (e.g., actively interacting with the participants’ discussion and drawing the participants’ attention to the topics and/or questions that the researcher was interested in). Thus, in the planning process, a researcher needs to decide the structure of the online forums and the role of the researchers. In all three studies, online forums were used as a data collection method rather than a data collection setting (e.g., a hospital, a community center) because the online forums aimed at getting descriptive data on the specific research topics. However, online forums could also provide a setting for observation of group dynamics or an intervention administration. Depending on the purpose of the study, the functionality of the online forums needs to be set appropriately.

The Necessity of an Interdisciplinary Team

In all three studies, the research teams included both a nursing research team and an engineering research team. The nursing research team focused on the administration of the recruitment process and the data collection while the engineering team focused on technical aspects of the studies (e.g., computer server, project website, security measures to protect Internet data, etc.). In these days, there exist many ready-to-use computer programs that researchers could use to conduct an online forum. Even when a researcher uses a ready-to-use computer program for her/his study, having an interdisciplinary team who could identify and accommodate technical issues would be helpful in conducting the study. Sometimes, commercial companies just want to sell their products, and a researcher would not have adequate technical knowledge to evaluate the products.

Screening Methods

In Studies 1 and 2, researchers used automatic computer programs to screen potential participants whether they met the inclusion criteria by automatically linking their Internet survey data to the online forums. In Study 3, because there were a number of screening questions related to cardiovascular risks that the participants should meet, researchers used emails to identify if each participant met the inclusion criteria. Depending on the screening methods that a researcher plans to use, the screening process in the online forums could be easily automated as in Studies 1 and 2. However, if the screening methods that a researcher plans to use are more complicated and sophisticated, individual screening through emails, phone calls, or face-to-face meetings would be necessary. Thus, in the planning stage, researchers need to think about the screening methods in advance, consult with an engineering team regarding necessary technical supports, and plan the screening process accordingly.

Languages

During the planning process, researchers need to consider the language that they want to use in their online forums. In Studies 2 and 3, only English was used because the online forum discussion in English among the participants went smoothly without a major language issue. However, in Study 1, researchers needed to use three languages including English, Spanish, and Mandarin Chinese. At the beginning, researchers started the online forums only in English. As the online forums proceeded, researchers recognized that the discussion only in English would not work for Hispanic and Asian participants. The Hispanic and Asian participants tended to be foreign born and older. Although the inclusion criteria included English literacy, researchers found that these participants needed to use their languages of origin to discuss the topics in-depth. Otherwise, their discussion stayed meager and simple. Thus, researchers adopted two additional languages for these two ethnic groups of participants. After adopting the two languages, the online forums went smoothly while having research participants actively discuss the forum topics.

Another aspect related to the use of language(s) in online forums is how to translate the messages in different languages and allow those who use different languages interact without a problem. Adopting 2–3 languages in an online forum is an easy task since most computer programming allows displays of multiple languages. However, the translation of each message posted by participants was not an easy task. For the Hispanic online forum, two Spanish bilingual moderators/facilitators intervened. Then, whenever a message was posted, one of the moderators/facilitators translated the message in English into Spanish and the message in Spanish into English. The accuracy of the translation was checked by the other bilingual moderator/facilitator. In this way, the Hispanic participants in the online forum could interact each other.

An additional aspect of online forums related to language was the actual language of data that would be analyzed. In the online forums using multiple languages, researchers need to choose the language version that they will use in analyzing the data in advance. In Study 1, we used the original language versions of the messages. For example, if a participant posted her/his message in Spanish, her/his data in Spanish were coded in Spanish to keep the original meanings of her/his messages. If a participant posted her/his message in English, her/his data in English were coded in English. Then, codes in other languages were translated into English using the standard-back translation strategies. Subsequently, categorization and thematic extraction were conducted in English.

Data Analysis Methods

To analyze the online forum data, researchers could easily use a content or thematic analysis. However, considering difficulties in getting theoretical saturation due to non-face-to-face interactions on the Internet,²⁶ a qualitative research method that requires theoretical saturation would not be a good choice. With advances in computer technologies to supplement non-face-to-face interactions (e.g., camera, chatting, multimedia functions), it would be possible for future researchers to use a different qualitative research method. Yet, until a more sophisticated method is available, the analysis method for the online forum data would be limited. Thus, researchers need to consider possible options for the data analysis, and determine the method at the planning stage.

Researchers need to consider first if computer software to analyze the qualitative data from the online forums is available. In Study 1, because three languages were used in the online forums, the analysis of the data tended to be complicated by difficulties in finding computer software that could be used to code the transcripts in Spanish and Mandarin Chinese. Because of the lack of computer software for the analysis of the data in Spanish and Mandarin Chinese, the MS word program and its “find” function were used to supplement the line-by-line coding of the transcripts in Spanish and Mandarin Chinese.

Throughout the research process of the three studies, researchers wrote memos on what they observed during the online forum process. Researchers found that the information from the memos was helpful in the interpretation of the study findings. In Study 3, researchers found one participant was very active in her discussion until the last three topics. Then, the participant faded out in the discussion. Later, during the reimbursement process, the researchers found out that she was out of town for a business trip. Thus, in the inferences of the reasons for dropouts, the researchers could easily include “business trips” as a reason.

Getting Participants’ Feedback

Online forums provide an excellent way to get participants’ feedback on preliminary findings of the online forums. Researchers need to consider using this unique characteristic to increase credibility of their study findings. However, researchers also need to be aware that getting participants’ feedback on the study findings would need additional efforts and costs. If the findings could be posted on the online forums immediately after the discussion on each topic, the feedback could be more promptly obtained compared with when the findings are posted at the completion of the study. In all the three studies, preliminary findings of the online forums were posted at the completion of the online forums because the analysis of the online forum data took a long time because of the huge amount of data for each topic. Thus, researchers posted their questions from the data analysis process in the online forums whenever they identified unclear aspects of the participants’ answers. However, the participants’ feedback took awhile because they tended to visit the online forum sites only when a new email was sent to remind them of visiting the online forum sites for new topics. Also, because only one message per topic was required for participant reimbursement, the participants were not obligated to answer any other questions. Thus, in future studies, researchers need to think about creative strategies to promote the participants’ responses to the posted study findings or questions. Maybe, researchers could use new functionality of email notifications/reminders to increase the response rate of the participants to the questions that the researchers post.

Conclusions

In this paper, based on three previous studies that used online forums as a qualitative research method, practical guidelines in nine idea categories for future qualitative research using online forums are suggested. First, researchers need to check if an online forum would be essential to achieve their research purposes and/or to address their research questions. Also, researchers need to consider how online forums would be structured (e.g., moderated or facilitated) and function (e.g., a setting for observation, focus group discussion, an intervention). Then, researchers need to plan an appropriate informed consent process. Researchers also need to consider having an interdisciplinary team with an engineering team member who could provide expertise in technical aspects of online forum administration. Researchers need to think about the screening method of the participants and the language that will be used in the online forums. Subsequently, they need to plan the data analysis methods and consider having a plan for getting participants’ feedback on the study findings. The critical questions that researchers need to ask in their selection of an online forum for their qualitative research are summarized in Table 1. Yet, these guidelines have several limitations mainly because they are proposed based on only three previous studies. Also, the limitations came from the fact that online forums were used as a data collection method only among Internet populations. Thus, for researchers who plan to use an online forum for a different purpose such as intervention administration, participant observation, or document analyses, care needs to be taken when adopting and using these guidelines.

Table 1.

Critical questions to ask in planning an online forum for a qualitative study

Steps of Planning	Questions to ask
Research Purpose and/or Questions?	Is it essential to have an online forum to achieve research purpose and/or address research questions?
Logistics	What are the specific steps that you want to take in administering an online forum? Do you want to use user IDs and passwords? Facilitator-moderated? The format of the messages?
Informed Consent	Is it feasible to get conventional informed Consent or electronic informed Consent?
Structure and Functionality	What are the structure and functionality required to achieve the research purpose and to address research questions?
Expertise in Computer and Internet Technology	Could an interdisciplinary research team be built? Does the study design require specific computer and Internet technologies?
Screening Methods	How do you want to check if potential participants meet the inclusion criteria? Electronically (through a website or by email), by phone, or by face-to-face meeting?
Languages	What are the languages that will be used in the online forum? Single or multiple languages? Does the forum site support multi-language functionality? How do you want to translate the message in languages other than the original language used in the online forum?
Data Analysis Methods	Do you want to use computer software to manage and analyze the data? Do you want to print out the data as transcripts directly from the database? What specific qualitative data analysis method do you want to use to analyze the online forum data?
Getting Participants’ Feedback	When and how do you want to get participants’ feedback on your study findings?