A dual instructional model for computer-assisted qualitative data analysis software integrating faculty member and specialized instructor: Implementation,reflections,and recommendations

Faculty reflection

The dual instructional model was a much better way to teach CAQDAS to doctoral social work students learning qualitative data analysis than were other current models. I particularly appreciated that students had open access to an expert in the software. While I use CAQDAS in my own data analysis, the trainer was much better than I at understanding and solving the problems the students were experiencing. The students taught by this new model did much more with CAQDAS in their own data analysis than students learning the software through other methods. Another benefit to this model was that I was able to show students how I use CAQDAS in my own research. Because the projects used to introduce the program were from the social work field, the students were able to understand them and to see how NVivo had contributed to the data analysis. Further, the students were all able to complete the exercises using their own data; their investment in these exercises led them to learn both benefits and limitations of CAQDAS.

In terms of difficulties with this educational model, there was additional work in setting up the sessions, communicating with the trainer, the technical assistant between sessions, and reviewing the students’ work. In addition, the frequent technical problems made me very nervous during the class sessions. If the technology failed, I had an additional job of serving as an intermediary between the students and the trainer. In terms of the course structure, the addition of the CAQDAS component led to an interruption in the flow of the not related part of the class. Because CAQDAS sessions were the first hour of a 3 h seminar, it was sometimes difficult for the students to transition from a very technical focus to abstract and philosophical discussions. CQDAS was only one component of the seminar, so I was reluctant to extend the time allotted to the distance learning piece if I thought it would be at the expense of other educational objectives.

The issue of how best to schedule the CAQDAS sessions remains. If sessions are scheduled early in the semester, when the students are introduced to the theoretical and philosophic background of qualitative research, the introduction of CAQDAS can disrupt the flow of the class. However, scheduling them later could mean that this component would not come in time to be used for the student projects and thus for the students’ own data interpretation. The addition of CAQDAS sessions also meant that the course content had to be adapted, and that some content that could not be easily integrated into this new component had to be eliminated, such as debates on philosophical lenses of qualitative research and an emphasis on non-coding-related techniques. Although the proposed model provided better integration of CAQDAS instruction than current ways of teaching, this model needs further development. The NVivo exercises need to be better integrated with the other course assignments required of the students. It was confusing to students to have two sets of assignments: one the traditional coding and analysis assignments (graded) and another set of NVivo exercises for practice (not graded).

Throughout the process, there were differential student reactions during the sessions. Doctoral students come into the class with a variety of expectations, technological experience, and levels of interest in learning CAQDAS. Although some are excited and eager to learn CAQDAS, others do not expect to use CAQDAS in their research. In terms of learning during sessions, some students were following the trainer and even jumping ahead, but others seemed to be overwhelmed. Students who were more interested and involved would ask detailed questions. During the answers to these questions, other students stopped following the lesson. This meant that students in the class were sometimes at different points in their understanding of the software. Ultimately, these drawbacks suggest that more tinkering needs to be done, though I believe the benefits to the students using this model make the time to improve the model worthwhile.

Trainer reflection

This teaching was unique in my experience because it blended the best that workshops have to offer with the best that distance elearning models have to offer at a fraction of the cost of a two-day workshop. This method is more efficient and cost-effective than the status quo. The blending of distance and onsite teaching and learning in this new model allows faculty to focus on the academic aspects of qualitative research while simultaneously allowing CAQDAS trainers to focus purely on what they do best, each knowing that the content area less familiar to them is being covered by their counterpart. The online support available between sessions ensures that students get the benefit of one-to-one support in their CAQDAS learning.

This model does require considerable collaboration between trainer and faculty. This limits growth of this method of delivering CAQDAS as these relationships and the courses that arise from them can only be uniquely developed course by course over time in each university setting, as, at this time, they are based on individual relationships between faculty and trainers. Related to this is the fact that trainers typically like to stay in their comfort zones, with neatly prepared tutorial data that deliver a “wow” factor during sessions. Additionally, using student data requires significantly more noncontact hours preparing for each session. However, once the program has been delivered successfully, it can then be rolled out repeatedly in that location with only minor modifications. Despite the extra effort required for each course, this method opens up numerous possibilities for future CAQDAS trainings.

If not done in person, the dual instructional model leaves the trainer dependent on the technological setup being correct in the receiving institution. If there are problems at the receiving institution, there is little the trainer can do to help in advance. Still to be resolved is the issue of student use of the trainer’s support between sessions. In the traditional workshop format there is time for the trainer to walk around and see what students are doing and help them as they run into problems. This was not possible in our sessions, due to both time and location concerns, so the practice time needs to take place between the sessions. Posttraining support between and after sessions is a vital component in this intervention because sessions were short and hands off. Although postsession support was included by the trainer as part of the program package, students were reluctant to use it; in-person learning offers an increased ability to build student–trainer rapport, and, in turn, helps to encourage students’ interest. Overall, however, the advantages of delivering CAQDAS training via the blended onsite and distance learning model considerably outweigh the disadvantages many of which could be resolved in future courses. In this instance, the financial cost to the learning instituting was over 75% less expensive than traditional workshops. Additionally, the students appeared to grasp the material more easily and their work showed more mastery of the software than in traditional workshops I have taught, indicating that this format was more beneficial to their learning.

Overall student evaluations

Of the six enrolled students, five completed the course evaluation at the end of the semester. In an effort to address potential bias, these evaluations collected no identifying information from students and are not made available to faculty until after grades were submitted. The evaluation question most relevant to the CAQDAS component of the course asked: “To what extent did the course meet the objective to understand the potential uses of qualitative data analysis software (NVivo 10) in analyzing qualitative data, as well as its limitations, and be able to code documents using this software?” Responses were open ended. All students were positive in terms of their comments, with one student simply answering “yes” and another saying she had “gained a greater understanding and was able to use it for coding project.” For the two other responses provided by students, one described the objective as being met “extremely well” and the other thought this particular course component was a “tremendous help.”

On the more general questions asked on the evaluation, some of the student responses referred to the CAQDAS session in the course. One appreciated the ability to tie their data to their learning: “I loved the NVivo training and process of using our own data and having incremental practice linked with assignments.” The same student answered the question “What, if anything, did the [faculty] trainer do that supported your learning?” by referring to the CAQDAS component, writing “integrating NVivo with course content.” Another student mentioned “learning a new software” in her list of the most helpful elements of the course while another answered with “NVIVO lessons with ‘trainer.”

In terms of more critical feedback, two of the students made suggestions regarding the CAQDAS training. Both suggested some lab time for practice with supervision, or a lab format (e.g. holding the CAQDAS sessions in a computer facility whereby all students could work along with the trainer) for the NVivo-focused session. One student suggested adding more exposure to other CAQDAS programs in addition to the NVivo training.

Individual student reflection

As a doctoral student, one of the most frightening things you can hear from a professor is “Let’s try something new.” Although I think there is a lot of fear from students in learning software packages, I do feel as though it is a marketable skill and one I would rather learn prior to rather than postdissertation. Thus, I reasoned that even if I did not fully grasp how to use CAQDAS at the end of the course, I would get some familiarity with the software. Ultimately, however, I became a proponent of teaching CAQDAS this way, in part due to the proficiency I was able to develop with it during this short time. What I have learned has made me more comfortable with using CAQDAS for a wider variety of projects, and I have more familiarity with features beyond coding even if I am not an expert in using them.

In spite of ultimately thinking this method is better for doctoral social work students to learn CAQDAS, there were several disadvantages to the method. We had the option of seeking help outside of class from our trainer. Although I occasionally experienced problems with the software, I never took advantage of this opportunity. Part of this was due to the desire for help at the very moment when there was a problem and an inability to explain the problem beyond “I think the screen that I should be seeing isn’t there.” Although our trainer offered to make plans so that he could meet at a certain time, I felt guilty him go out of his way to reteach me something we had just learned in class. Similarly, when I did encounter problems using the software, I was never sure if the error was mine or was an issue with the software. Finally, the experience of learning CAQDAS in combination with what we were learning in class was overwhelming. At times, it seemed like there was little overlap, especially in the early class session. This meant that the class consisted of 1 h of technological teaching in combination with 2 h instruction on general philosophy, history, notable figures in, and methodology of qualitative research; these two didn’t blend together logically in my head until later instruction.

The advantages to learning this way far outweigh the problems I experienced. The seminar setting made it easy and comfortable to ask questions, and the questions were able to be focused on social work research specifically. The combination of our trainer and our professor meant that things could be translated back and forth from social work to CAQDAS applicable terms. It also meant that when we asked questions, we got answers that were both technologically and qualitatively fulfilling to us. The experience led me to be more curious about CAQDAS than I would have been otherwise because I developed specific questions and curiosities based on my data. It also meant that my understanding of CAQDAS progressed in a natural manner over the course, rather than feeling forced as might be possible in a time-limited workshop format. Although there were concerns about integrating CAQDAS lessons into our seminar as a whole, the pacing of learning CAQDAS, practicing, and talking together about what we knew very successfully integrated the use of CAQDAS into my understanding of working with qualitative data.