Part I: Network Analysis
The intial part of the food blog study was a network crawl of food blogs based on the blog roll or in-text links to other food blogs. The web crawler was seeded with the names of 59 top blogs as listed by a food blog aggregator site. All of the links from these blogs were then crawled if the content on those links had a high matching score with the food word dictionary developed to sort out non-food related websites. Many but not all non-food sites were successfully weeded out. Additional sites were blacklisted and included sites such as food retailer websites, food magazine websites, and food blog aggregator sites which had high matching scores with the food word dictionary but were not blogs and therefore were rejected from the study. Addtionally, blogs whose content was based on beer, wine, and/or cocktails were also excluded from the network to maintain the focus on food blogs.
The network crawl generated a list of 32,000 food blogs in 20+ languages. A random sample of 2500 blogs were selected from the 32,000 sites in the network. Each one was visited to ensure that it was written in English, that 60% of its content included food in some way, and to obtain active contact information. Only 793 blogs met all of these criteria. These 793 blogs received invitations to participate in the web-based survey.
Part II: Blogger Motivation Survey
Invitations to participate in the survey were delivered via email and contact forms embedded in the food blogs themselves (when email addresses were unavailable) to the 800 subjects. There were 303 respondents yielding a response rate of 38%. Some bloggers informed our research team that the invitations to participate had been diverted to their spam folders. It is not possible to know how often that occurred.
The content covered in the survey included:
- Demographics of the blogger: age, race, gender, educational attainment, marital status, income, etc.
- Demographics of the blog: lifespan of blog, content characteristics, frequency of posting, etc.
- Time Investment in blogging
- Audience: page views, comments
- Economics of blogging
- Satisfaction associated with blogging
The survey results are being analyzed for publication. Preliminary findings are now available.
Part III: Qualitative interviews with food bloggers
To strengthen the quantitative results from the survey, 26 interviews with 29 food bloggers were conducted. Subjects were identified through their food blogs which emerged from internet searches and links from other blogs, taking the physical location of the blogger into account for half of the subjects. The interviews were stratified to include roughly half from the Boston area (N=12) with the remaining food bloggers located across the US (N=13) with one from the UK (N=1). Interviews were conducted in English and face-to-face where possible. The sample was stratified by geography in order to help elucidate the relationship, if any, between physical proximity and community-making practices in the food blogosphere. The demographic characteristics of the interview subjects largely mirror the demographics of the food bloggers in the survey.
Analysis of the qualitative interviews was presented at the 2012 Crossroads Culture Conference in Paris. These findings focused on the intersection of gender and class in three typoologies of food blogging practice that were similar theoretically to the serious leisure framework developed by Douglas Stebbins. The interviews also provided evidence for extending the communities of practice framework (Lave and Wegner) to include class-influenced orientations towards the skills-based communities.