TY - JOUR
T1 - The stated preferences of community-based volunteers for roles in the prevention of violence against women and girls in Ghana
T2 - A discrete choice analysis
AU - Arora, Nikita
AU - dit Sourd, Romain Crastes
AU - Quaife, Matthew
AU - Vassall, Anna
AU - Ferrari, Giulia
AU - Alangea, Deda Ogum
AU - Tawiah, Theresa
AU - Dwommoh Prah, Rebecca Kyerewaa
AU - Jewkes, Rachel
AU - Hanson, Kara
AU - Torres Rueda, Sergio
N1 - Publisher Copyright:
© 2023 The Authors
PY - 2023/5
Y1 - 2023/5
N2 - Violence against women and girls (VAWG) is a human rights violation with substantial health-related consequences. Interventions to prevent VAWG, often implemented at the community level by volunteers, have been proven effective and cost-effective. One such intervention is the Rural Response System in Ghana, a volunteer-run program which hires community based action teams (COMBATs) to sensitise the community about VAWG and to provide counselling services in rural areas. To increase programmatic impact and maximise the retention of these volunteers, it is important to understand their preferences for incentives. We conducted a discrete choice experiment (DCE) among 107 COMBAT volunteers, in two Ghanaian districts in 2018, to examine their stated preferences for financial and non-financial incentives that could be offered in their roles. Each respondent answered 12 choice tasks, and each task comprised four hypothetical volunteering positions. The first three positions included different levels of five role attributes. The fourth option was to cease volunteering as a COMBAT volunteer (opt-out). We found that, overall, COMBAT volunteers cared most for receiving training in volunteering skills and three-monthly supervisions. These results were consistent between multinomial logit, and mixed multinomial logit models. A three-class latent class model fitted our data best, identifying subgroups of COMBAT workers with distinct preferences for incentives: The younger ‘go getters’; older ‘veterans’, and the ‘balanced bunch’ encompassing the majority of the sample. The opt-out was chosen only 4 (0.3%) times. Only one other study quantitatively examined the preferences for incentives of VAWG-prevention volunteers using a DCE (Kasteng et al., 2016). Understanding preferences and how they vary between sub-groups can be leveraged by programme managers to improve volunteer motivation and retention. As effective VAWG-prevention programmes are scaled up from small pilots to the national level, data on volunteer preferences may be useful in improving volunteer retention.
AB - Violence against women and girls (VAWG) is a human rights violation with substantial health-related consequences. Interventions to prevent VAWG, often implemented at the community level by volunteers, have been proven effective and cost-effective. One such intervention is the Rural Response System in Ghana, a volunteer-run program which hires community based action teams (COMBATs) to sensitise the community about VAWG and to provide counselling services in rural areas. To increase programmatic impact and maximise the retention of these volunteers, it is important to understand their preferences for incentives. We conducted a discrete choice experiment (DCE) among 107 COMBAT volunteers, in two Ghanaian districts in 2018, to examine their stated preferences for financial and non-financial incentives that could be offered in their roles. Each respondent answered 12 choice tasks, and each task comprised four hypothetical volunteering positions. The first three positions included different levels of five role attributes. The fourth option was to cease volunteering as a COMBAT volunteer (opt-out). We found that, overall, COMBAT volunteers cared most for receiving training in volunteering skills and three-monthly supervisions. These results were consistent between multinomial logit, and mixed multinomial logit models. A three-class latent class model fitted our data best, identifying subgroups of COMBAT workers with distinct preferences for incentives: The younger ‘go getters’; older ‘veterans’, and the ‘balanced bunch’ encompassing the majority of the sample. The opt-out was chosen only 4 (0.3%) times. Only one other study quantitatively examined the preferences for incentives of VAWG-prevention volunteers using a DCE (Kasteng et al., 2016). Understanding preferences and how they vary between sub-groups can be leveraged by programme managers to improve volunteer motivation and retention. As effective VAWG-prevention programmes are scaled up from small pilots to the national level, data on volunteer preferences may be useful in improving volunteer retention.
KW - Discrete choice experiments
KW - Latent class analysis
KW - Preferences
KW - Violence against women and girls
UR - http://www.scopus.com/inward/record.url?scp=85151251817&partnerID=8YFLogxK
U2 - 10.1016/j.socscimed.2023.115870
DO - 10.1016/j.socscimed.2023.115870
M3 - Article
C2 - 37012185
AN - SCOPUS:85151251817
SN - 0277-9536
VL - 324
JO - Social Science and Medicine
JF - Social Science and Medicine
M1 - 115870
ER -