186 The Professional Counselor | Volume 14, Issue 2 Sand Tray Activity Hays and Singh (2023) stated that “visual methods in general provide participants the opportunity to express themselves in a nonverbal manner that may access deeper aspects of their understanding and/or experience of a phenomenon” (p. 332). After the semi-structured interview, Bevly invited participants to create their personal therapy experience in a sand tray using the figures and materials provided. This method is consistent with Measham and Rousseau (2010), who used sand trays as a method of data collection for understanding the experiences of children with trauma. The sand trays were documented by digital photos (see Appendix), and participants’ discussions about their creations are part of the audio recordings. Data Analysis We sent the audio recordings to a professional transcriptionist for transcription of each interview and sand tray session. We reviewed transcripts while listening to the recordings for participants’ tone and to verify accuracy. Consistent with phenomenological procedures, the research team conducted data analysis according to an adaptation of classic analysis (Miles et al., 2019), in which three main activities take place: data reduction, data presentation, and conclusion or verification. Prior to initial coding, the research team completed several tasks in order to develop the preliminary coding manual: taking notes, summarizing notes, playing with words, and making comparisons (Miles et al., 2019). Taking notes involved the research team as well as Bevly’s own independent analysis of a subset of the first three interviews and sand tray explanation transcripts. We divided the transcripts into 10-line segments and wrote notes in the margins. The research team noted our initial reactions to the material. Summarizing notes involved discussion between the team regarding our reactions to the interview material. We compared and contrasted our margin notes and highlighted shared perspectives and inconsistent viewpoints in a summary sheet. To play with words, we generated metaphors based on our summary sheet. We developed phrases that represented our interpretation of the participants’ interview responses. During the making comparisons task, we compared and contrasted the key phrases developed in the previous step and grouped them into categories. The team then facilitated reduction of the data as we combined similar phrases and merged overlapping categories. Hays and Singh (2023) asserted the importance of sieving the data to eliminate redundancy. We continued to merge categories and reformat the category headings. From this process, we developed preliminary themes based on the data. To develop initial codes, we established agreement by independently applying the preliminary codes to a subset of three interviews. The research team met weekly to discuss inconsistencies and points of agreement, adjust the preliminary codes, and reapply them to the data subset. We continued to discuss any remaining discrepancies and concerns until we reached a mean agreement of 86% to 90% (Creswell & Creswell, 2017). We reached a mean agreement of 95.1% and then finalized the codes to use in our coding manual. It is important to note that the research team sensed that we had reached saturation during the final coding process once we began to read the same comments repeatedly in the participant transcripts. In final coding, we applied the final coding manual to each of the interviews and sand
RkJQdWJsaXNoZXIy NDU5MTM1