Table 1 Design table.

1. Does participation in the training increase participants’ capacity to take action to create inclusive, safe field environments?

(a) Post-training prevention self-efficacy, collective efficacy and knowledge will increase significantly in the intervention group compared to the control

(b) Changes in primary outcomes will not be sustained over time

Our power analysis suggests that with roughly 250 participants in each experimental condition (intervention and control) and assuming a standard deviation of 1 (which is consistent with our pilot data), we are powered to detect an effect size of 0.32 for continuous outcomes with an alpha or 0.05 and power 0.95. Previous research indicates that this is within the range of effect sizes detected in other studies of sexual assault prevention knowledge, self-efficacy and collective efficacy in non-scientific settings^33,44,45

(a) We will use adjusted and unadjusted linear regressions for responses regarding respondent knowledge, self-efficacy, prevention behavioural intention, and self-reported prevention behaviour

(b) We will use multi-level models to compare change in response scores for data nested within an individual from pre-surveys (Time 1) compared to immediately after training (Time 2) and several weeks after training (Time 3)

(a) If the response scores for questions related to self-efficacy, collective efficacy, and knowledge are significantly higher for the intervention group compared to the control (Time 3 survey), we will conclude finding support for Hypothesis 1a

(b) If the response scores for questions related to self-efficacy collective efficacy, and knowledge are significantly greater immediately after the training compared to the later time period, we will conclude finding support for Hypothesis 1b

2. Does participation in the training increase participants’ actions to create inclusive, safe field environments?

(a) Post-training prevention behaviour (self-reported) and behavioural intention will increase significantly in the intervention group compared to the control

(b) Changes in primary outcomes will not be sustained over time

Same as above

(a) We will use adjusted and unadjusted linear regressions for responses regarding respondent knowledge, self-efficacy, prevention behavioural intention, and self-reported prevention behaviour. We will use repeated-measures ANOVA tests to compare responses at three time points (Times 1, 2, and 3)

(b) We will use multi-level models to compare responses for date nested within an individual from pre-surveys (Time 1) compared to immediately after training (Time 2) and several weeks after training (Time 3)

(a) If the response scores for questions related to behaviour (self-reported) and behavioural intention are significantly higher for the intervention group compared to the control, we will conclude finding support for Hypothesis 2a

(b) If response changes in scores for questions related to behaviour (self-reported) and behavioural intention are significantly greater immediately after the training compared to the later time period, we will conclude finding support for Hypothesis 2b

3. Does the training work equally well for all demographic groups?

(a) Increases in post-training knowledge, self-efficacy, and behavioural intention and self-reported behaviour will be higher for women and gender minorities compared to men

(b) When controlling for gender, no significant differences will be observed in post-training outcomes based on gender, age, race/ethnicity, role, region, or time at the agency

(c) Increases in post-training behavioural intention will be higher for staff who reported higher levels of pre-training prevention behaviour and prevention personal norms compared to their less engaged and committed peers

Same as above

(a) We will use linear regressions to compare the effectiveness of the trainings for different groups with CDFW (Times 1 and 2)

(b) We will use linear regressions to compare the effectiveness of the trainings for different groups with CDFW controlling for gender (Times 1 and 2)

(c) We will use moderation analyses to determine if pre-training prevention behaviour and beliefs moderated the impact of the training on post-training behavioural intention and self-reported behaviour (Times 1 and 2)

(a) If there is significant difference among change in response scores between women and gender minorities and men, we will conclude finding support for Hypothesis 3a

(b) If there is no significant difference among change in response scores for participants of different genders, ages, race/ ethnicities, roles, regions, or duration of time at the agency, we will conclude finding support for Hypothesis 3b

(c) If there is no significant effect of pre-training behaviour and beliefs on post-training behavioural intention and self-reported behaviour, we will conclude finding support for Hypothesis 3c

4. Do reporting rates increase after participants receive information about sexual harassment and assault?

(a) Post-training confidence in reporting and likelihood to report an incident of sexual harassment and assault will be higher in post-training surveys than pre-training

Same as above

(a) We will use adjusted and unadjusted linear regressions and multilevel models for responses regarding confidence in reporting and likelihood to report. We will use multilevel models repeated-measures ANOVA tests to compare responses at three time points (Times 1,2, and 3)

If the change in response scores (from pre- to post-surveys) for questions related reporting are significantly higher for the intervention group compared to the control and within subjects over time, we will conclude finding support for Hypothesis 4a