A Logit Regression Approach to Identifying Predictors of Intimate Partner Violence in Tanzania

Abstract

Introduction: Intimate Partner Violence (IPV) is among the common forms of gender-based violence worldwide. The IPV magnitude in Tanzania is approximated to be 44%, in which 4 in 10 married women experienced spousal violence. The present study zeroed into the uses of logit regression to identify the predictors of IPV in Tanzania.

Methods: The data was obtained from the Tanzania Demographic and Health Survey 2022. The unity of analysis was currently married women and men. Logit (logistic) regression analysis was performed to determine the predictors of IPV in Tanzania. IPV was measured into three categories: emotional violence, physical violence, and sexual violence.

Results: Women’s age, partners’ education level, partners’ occupation, and frequency of husband/partner being drunk are factors that predict emotional violence significantly (P<0.05). Age of household head, age of women at first birth, partners’ occupation, women’s current age, frequency of husband/partner's drinking habit, women’s education level, and literacy level are factors that predict less and more severe physical violence significantly (P<0.05). In addition, the wealth index, women’s age at first birth, and women’s occupation explain the sexual violence significantly (P<0.05). While residing in Dodoma, Arusha, Tanga, Tabora, Kagera, Mwanza, Mara, Manyara, Kusini Unguja, Tanga, Songwe, Kilimanjaro, and Katavi has a high likelihood of experiencing IPV significantly (P<0.05).

Conclusion: The intervention to address IPV should first focus on changing men's and women's attitudes and beliefs. Change of attitude and belief should be to both men and women who believe that wife beating is okay, and that perpetuates the increase of IPV. This study recommends an emphasis on archiving employment, especially for men who reside in regions where IPV has been expected, to reduce the likelihood of men committing violence.