Stepping Stones and Creating Futures Intervention Trial

Physical intimate partner violence is assessed using five items based on the WHO VAW scale. A positive response to any item leads to a person being classified as perpetrating (men) and experiencing (women) in the past year. With 0=none, 1=yes.

Sexual intimate partner violence is assessed using three items based on the WHO VAW scale. A positive response to any item leads to a person being classified as perpetrating (men) and experiencing (women) in the past year. With 0=none, 1=yes.

Severe sexual and/or physical intimate partner violence perpetration (men) and experience (women) is assessed using 8 items. Past year severe sexual and/or physical IPV is assessed as positive if a person responds to two (or more) items as once, or one item as few (or more), essentially creating a more than once categorization.

Controlling behaviours are assessed using a modified Sexual Relationship Power (SRP) scale. Men's control of female sexual partner's and women's experience of controlling behaviours from a male partner. A mean score is calculated with higher scores indicating more controlling (bad). Range: 0-24

A single item question asks "Considering all the money you earned from jobs or selling things (excluding grants), how much did you earn last month?" Responses are in Rands and a continuous scale, with no upper limit.

Modified gender equitable men's scale (GEMS) assess participant's gender attitudes. A mean score is calculated with higher scores indicative of more gender inequitable attitudes (range: 0-60)

Past week depressive symptomology is assessed by the Centre for Epidemiological Studies Depression (CESD) scale, with the full twenty items (range 0-60). A mean score is calculated, with higher scores indicating more depression.

Past four week suicidal ideation was assessed using a single item question. Responses are either 0=no, or 1=yes.

Four items modified from the Satisfaction with Life Scale (SWLS) by Denier et al. (1985), with items on a five point Likert Scale (range 0-20) higher scores indicative of greater life satisfaction.

Problem drinking in the past year is assessed using the AUDIT (alcohol use disorders identification test) scale. With a range of 0-40, and recoded with scores of 8 or more classifing a participant as having potentially problematic alcohol use.

A single item binary response question asks whether participants have argued about alcohol with their sexual partner in the past year (0=no; 1=yes). Only asked to those who reported alcohol use in the past year.

A single item assess who the participant last had sex with, with possible responses being: "main partner", "casual partner", "once-off sex partner" or "ex-partner". As per the protocol, we recategorised this into either "1=main partner" or "0=other". Positive change is towards main partner. And the reported number (proportion) is those reporting Main partner.

Five item scale asks about transactional sex with casual or once-off sexual partners in the past year based on the scale developed and tested by Dunkle et al (2004). Each item has a no/yes response. A response of yes to any is classified as having engaged in transactional sex.

Four items assess participants' feelings of shame about not enough work, based on a scale developed for use in the IMAGES study, with . Mean score calculated, with higher scores indicative of more shame (range: 4-16)

Four items assess participants' feelings of stress about not enough work, based on a scale developed for the IMAGES study. Mean score calculated, with higher scores indicating more stress (range: 4-20)

A single item asks participants how hard it would be to mobilise R200 (~US$15) in an emergency. Those responding it would be "very difficult, or somewhat difficult" are classified as finding it hard (=1). Those responding "fairly easy or easy" are classified as not finding it hard (=0). Reported number (proportion) is of those finding it hard to mobilise money in an emergency. A positive change is towards a lower proportion.

Single item question asks about stealing because of lack of food or money in the past month with responses "never", "once", "every week" or "every day". Responses are recoded, as per protocol, into "0=never" and "1=once or more". The number/proportion of those stealing in the past month is reported. A lower proportion is a positive change.