A mixed methods approach to mediumship research

Elizabeth C Roxburgh, Chris A Roe

Research output: Contribution to Book/ReportChapterpeer-review


There is a movement gaining ground within the social and behavioral sciences to introduce a new research method paradigm that combines qualitative and quantitative elements so as to capitalize on the strengths and compensate for the weaknesses of each. This approach, known as mixed methods research ( Johnson, Onwuegbuzie, & Turner, 2007), has, in our view, great potential benefits for the study of mediumship, which naturally combines a requirement to evaluate evidential claims in a balanced and objective manner with a need to pay careful attention to the subjective meaning of mediumistic experiences in the context in which they occur. In this essay we will outline some of the essential features of the mixed methods approach, and draw attention to the procedural developments that have taken place over the past two decades, focusing particularly on design issues such as how data collection might be implemented (i.e., whether concurrent or sequentially) and prioritized (i.e., whether qualitative and quantitative aspects will have equal or unequal emphasis), and at what point data will be integrated or mixed (i.e., at the data collection, analysis or interpretation stage). The second part of this essay will demonstrate the value of adopting a mixed methods approach in mediumship research using illustrations from our own work with Spiritualist mental mediums in the UK (Roxburgh, 2010, 2011; Roxburgh & Roe, 2011a).
Original languageEnglish
Title of host publicationThe Survival Hypothesis: Essays on Mediumship
EditorsAdam J Rock
Place of PublicationJefferson, NC
PublisherMcFarland & Company
Number of pages15
ISBN (Electronic)9781476614427
ISBN (Print)9780786472208
Publication statusPublished - 16 Apr 2014


  • Mediumship
  • Mixed methods
  • Parapsychology
  • Psychology


Dive into the research topics of 'A mixed methods approach to mediumship research'. Together they form a unique fingerprint.

Cite this