Use Automated Content Analysis for Your Qualitative Research Projects October 17, 2016 - Blogs on Text Analytics

 

When managing international qualitative research projects, one of the major challenges is getting answers to our key research questions, and the relevant quotations to support them, in a standardized format, from all our field researchers across target cities and countries.

My first international qualitative project, nearly twenty years ago, covered four Caribbean countries; four CPG categories: soap, toothpaste, household cleaning products and laundry detergent;with women of three socio-economic groups.Add to that video footage for the ethnographic research part.The objective was to roll out a region-wide Marketing campaign which reflected local realities, as well as to optimize the product portfolio and positioning, to have an offering to cater both to the more affluent consumer looking for more sophistication and the less-affluent consumer looking for affordable products more efficient than traditional methods.

Enter text analytics software! The end of fieldwork coincided with the launch of a new text analytics software program. With a fellow consultant, a university researcher, we plugged in all of the transcripts, eager to try it out. The software was not particularly user-friendly or intuitive. Moreover, it required so much manual programming, that given our tight timeline we decided it’d be more efficient to not incorporate any computer-assisted analysis.

Fortunately, the text analytics’ software landscape has evolved tremendously since then.

Given clients’ tight deadlines, we try to provide a quick turnaround of key take-aways, followed by an in-depth report which captures the actionable insights to respond to their strategic questions. We want them to have the feeling of having been at every interview. We may want to feel like we have conducted every interview ourselves, even where language or logistical barriers have prevented us from doing so.

I have always found that along with the review of transcripts, recordings, notes, there is immense value in an oral or written topline from each field interviewer, highlighting their findings in their own words. Complement this with an out-of-the-box content analysis tool such as Provalis Research’s WordStat , that allows you to zero in on answers to specific research questions, and to segment by country, consumer segment, demographic, or any other relevant variable. This analysis is achieved by visual modeling tools which can be integrated into your reporting as well as give you rapid understanding into key concepts that emerge and how these differ across your target populations.Moreover, the keyword-in-context feature displays the verbatims so you can choose your relevant quotations easily.

You can achieve this initial high-level content analysis quickly, without coding a single piece of text, and dispensing of the use of manual content analysis procedures in spreadsheets or text documents.

In my next post, I will look at how to prepare your transcripts to optimize the use of an analysis tool such as WordStat for content analysis.

Have an experience you’d like to share with us? Interested in a demonstration using your own transcripts? We’d love to hear from you!