Cycle of learning, adapted from [35]. 

Context in source publication

Context 1

... coding and axial coding phases went on in parallel. During the open coding and axial coding phases, a constant comparative method and theoretical sampling [51,52] were used. The use of the constant comparative method meant that when an interesting phenomenon was found later during data collection, all the earlier data was screened again to see if the phenomenon was found there, too. Theoretical sampling was used to collect more data on phenomena where more explanation was needed. The data was collected until a theoretical saturation was achieved. The use of the constant comparative method and theoretical sampling increased the researchers' interaction with the data. The selective coding phase also started early and parallel with the axial coding phase. The saturation of the new data [51] took place early, which meant that there was no need for extra interviews on that subject. A new tool for making word clouds was used to visualize part of the data (see Figure 4 for more ...