Analysis

We used IBM SPSS Statistics version 20.0 to calculate frequencies, means, and standard deviations (SDs) for the socio-demographic data. A thematic analysis approach based on grounded theory was used to analyse the focus group data, which allows themes to emerge from the data (Strauss & Corbin, 2007). All coding was iterative and collaborative. To establish codes, categories, and sub-categories, two of the investigators manually (paper and pen) and independently reviewed, abstracted, and coded a segment (four pages) of the de-identified transcript from the first focus group with mothers. The researchers then met to discuss their codes, categories, subcategories, and rationale. After the initial meeting, the two investigators continued to manually and independently review, abstract, and code an additional four pages. The two researchers independently developed provisional codebooks (code name and description, categories, subcategories, and examples for each). Then the researchers met to discuss their codes (agreements and discrepancies of assigned codes) and merged their provisional codebooks. This consensus meeting process was repeated four more times until the entire transcript was coded. These codification and abstraction processes were done iteratively to test, revise, and refine the thematic classifications and to develop a final codebook. Themes identified were further compared and contrasted for similarity and/or difference in responses by a third investigator.

At each consensus meeting, to assess coding agreement, we calculated the percentage of agreement for the most frequently coded sections of the transcript. Coding agreement, inter-coder reliability, was when both coders assigned the same code to the main idea of a segment from a transcript (Burla et al., 2008). The overall inter-coder reliability for the transcript was 91.4%. Using the final codebook, we used NVivo to code the other two transcripts.

Consensus between the three investigators regarding codification and abstraction resulted in five identified themes related to the socio-cultural environment of the participants. Categories and themes are presented and appropriate quotes were selected (Strauss & Corbin, 2007).