Limitations
There are several limitations concerning our sampling technique and methodology. We conducted four FGs in each community with only one per gender/age group. In a larger study, we would have conducted more FG per age/gender group to increase our sample size. In addition, not all FGs were conducted by native language speaker moderators. This resulted in longer sessions for one Mexican and two Cambodian FGs, contributing potentially to loss in time to go deeper into the different topics. In addition, where language concordance was not achieved, the moderators may have been less familiar with the respective communities, which may have influenced the follow up questions. Further, not all participants contribute equally in all FGs, potentially leaving shy or withdrawn participants out. To avoid this bias, a longer study time frame would have allowed us to conduct individual interviews in addition to FGs. Finally, this study analyzed self-reported food preferences and eating practices. The larger study timeframe did not allow sufficient time for gathering observational data of actual health behaviors prior to developing the intervention program. Future studies of observable immigrant family eating practices are needed to deepen our understanding of the eating preferences and self-reported eating patterns in these immigrant and refugee populations.