Questions about belief: an analysis of Yahoo! Answers queries regarding ‘belief’ in Islam and Christianity06/06/2014
This paper will present a content analysis of questions related to religious belief in Islam and in Christianity, as posed to social media Q&A sites. We choose Yahoo! Answers as a representative social media site because we wish to investigate religious information behavior by laypeople. We will conduct a concept analysis of the questions, to create a categorization of questions by intent (e.g., a factual information need, self-expression of personal viewpoint, etc.), and for factual questions, we will further categorize the types of conventional metadata expressed in the question.
everyday, authentic information seeking. To that end, we investigate the motivations for posting questions on religion on social media websites, and explore the match between conventional document metadata and the metadata offered by users in their questions to describe their factual information needs.
We choose the concept of ‘belief’ as the focus for our investigation of religious information behavior. This concept is selected because it is central to both Islam and Christianity (whereas other concepts such as tithing, religious laws, etc. are more specific to practice rather than to ideology). We examine queries pertaining to both Islam and Christianity so that we can compare the emergent question categorizations across the two.
The questions are sampled from the “answered” queries posed, that are retrieved by a search on the Society & Culture > Religion category by searches on “Christian belief” and “Islam belief” (as well as alternatives such as “Christianity believe”, “Muslim belief”, etc.). We will sample recent Yahoo! Answers until we identify 400 questions for each of Islam and Christianity (after eliminating duplicates, mis-categorized questions, and other noise in the dataset).
Content analysis will be employed to categorize the questions by intent of the questioner. Previous research suggests possible categories, but we anticipate that additional / different categories will emerge from the analysis. We will further categorize the types of conventional metadata that users specify in factual questions; we will seek to match these metadata types to the metadata offered in digital libraries with a focus on religion.
To our knowledge, there has been no previous study of religion-related information seeking behavior of laypeople; results of our concept analysis will provide the first evidence-based insights into the motivations of laypeople for posting questions regarding religion, and for seeking religious information through social media. Moreover, this study is based on authentic information behavior (rather than post-hoc recollections of behavior obtained through surveys or questionnaires, or artificial behavior exhibited in IR laboratory experiments).
We anticipate that our analysis of the Yahoo! Answers data will produce:
• a taxonomy of question types likely including emotional / affective, task-based, and self-expressive categories, as well as more conventional factual or evidence-based question categories. We base this assumption on the findings of previous analyses of queries to social Q&A sites (e.g., Bowler et al, 2012; Cunningham and Bainbridge, 2013),
• a categorization of metadata presented in the question and type of document or response requested as a successful answer. This type of categorization can inform the design of information systems / digital libraries to support information behavior in the target field (e.g., Cunningham and Bainbridge, 2013; Hinze, 2010).
Cunningham, S. J., and Bainbridge, D. (2013). An analysis of cooking queries: Implications for supporting leisure cooking. iConference 2013 Proceedings (pp. 112-123). doi:10.9776/13160
Bowler, L., Oh, J. S., He, D., Mattern, E., & Jeng, W. 2012 Eating disorder questions in Yahoo! Answers: Information, conversation, or reflection? Proceedings of the American Society for Information Science and Technology, 49(1), 1-11.
Hinze, A.M., Chang, C. & Nichols, D.M. Contextual queries express mobile information needs, Proceedings of the 12th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI 2010), ACM. 327-336.
Dr. Sally Jo Cunningham has an extensive research background in information behavior, including studies that are interview-based; ethnographic studies of music and ‘serious hobby’ document searching and browsing; and studies of queries posed to online community systems for music, images, and cooking information. She primarily studies everyday information behavior in its ‘native’ context. She has over 150 peer-reviewed publications, primarily in information behavior, digital libraries, and human-computer interaction.
Dr Annika Hinze has an established research track record on information seeking, contextual information delivery, and personalization, as well as substantial experience in data analysis regarding information seeking behavior (exploration of information needs based on location and context). She has been involved in a number of projects on human-centered search and contextual information delivery. She has almost 100 peer-reviewed publications on contextual information delivery, digital libraries, and information behavior.