the previous RAU study in Goromonzi (RAU.2014), and the participants were asked to comment on each story in turn with two focused questions: “what was happening in this story?”, and “was this common in your community?” The individual interviews ranged in duration from 30 minutes to an hour, whilst the FGDs took around three hours. All interviews and focus groups were audio-recorded, as well as noted being taken, and then transcribed in Shona and later translated into English. The transcripts were analysed through Nvivo, a qualitative data analysis was used as a tool to conduct the analysis. The Nvivo analysis was done by the RAU staff as team, with the support of two other researchers, highly experienced in both qualitative research and the use of Nvivo. The analysis revealed 13 key variables (see Appendix 2), which were all linked to specific analytical questions. In addition, 21 “attributes” were derived for the analysis, these key demographic features providing the basis for both understanding marriage as a whole as well as the basis for various cross-tabulations and comparisons. The “attributes” are summarized as follows:              Age: current age; age when married; husband’s age when married Length of marriage: year of marriage Child marriage: married under 18 years Current marital status: married; separated; widowed Number of marriages Place of residence: rural; urban; per-urban; combinations of these Employment: informal sector; housewife; farming; employed; none Education: primary school; secondary school; did not finish school; none Religion BEFORE marriage: formal Christian; alternative Christian; Apostolic; none Religion AFTER marriage: formal Christian; alternative Christian; Apostolic; none Family of origin: size of family; birth order in family; number of children Type of marriage: UCLU; Eloped; Chapter 5 marriage; Co-habiting Current mental health: SRQ-8 score; suicidal ideation Some of these “attributes” deserve some brief comment. The husband’s age at marriage was deemed important as previous research had suggested that there were cases where young girls were married off to older men to relieve family pressure or to conform to religious practices. This is linked to several other attributes; level of education (early marriage results in girls leaving school), and religion, both before and after marriage. These latter were included to examine the possible effects of apostolic religious practices, and to determine whether there might be voluntarism amongst women in joining an apostolic church as opposed to being brought up in such a religion, marrying within that religion (whether early or not), and remaining in the religion. Finally, the questions about the family of origin were included to examine whether large families (and economic pressures) are contributory factors in girls being forced into or choosing early marriage. 10