Page 86 - IJPS-11-4

International Journal of
Population Studies Droughts and intimate partner violence

of control issues faced by the respondents decreased control issues and the incidence of SV in states exposed to
over time. However, the magnitude of this correlation is drought. Conversely, a distinct decline is observed in such
not substantively significant. Conversely, a positive and cases for the comparison group (Figure 1B). As shown in
substantive, as well as statistically significant point-biserial Figure 1C, the incidence of EV for the number of control
correlation coefficient (0.1382; n = 4938; p < 0.01) for issues provides a clearer picture: the incidence was higher
states exposed to drought indicates that respondents in for the treatment group post-exposure and lower for the
these states faced a statistically higher number of control comparison group. One common thread among all three
issues from their intimate partners. Our results are similar types of IPV is that the more controlling the partner, the
to those of Cooper et al. (2021), who found a positive higher the incidence of IPV. The incidence of IPVs has
relationship between exposure to drought and controlling consistently increased with the number of control issues.
behavior by partners among women from Asian and Latin Dhanaraj and Mahambare (2021) found that working
American countries, even though they could not find such women who faced controlling behavior from their partners
an association between drought and other forms of IPV. also experienced a higher incidence of IPV through the male
Our results also align with Epstein et al. (2020), who found backlash channel. We found a strong, positive association
strong associations between exposure to severe drought between the working status of women and IPV, even after
and the likelihood of having a controlling partner. accounting for the controlling behavior of their partners.
It is important to note that the question regarding the The pathway through which the controlling behavior of
controlling partner does not have a recall period that is partners increases IPV requires further research.
limited to the past 12 months. Thus, a woman who has We found that the difference in the incidence of IPV
ever faced any control issues with a partner will report to between the states exposed to the NEM drought and those
a controlling partner. However, exposure to drought has not exposed widened over time. The DID estimates in
been shown to increase controlling behavior. models with a complete set of control variables show that
Controlling behavior by intimate partners is often linked the odds of PV increase by approximately 48% among
to IPV in women (Antai, 2011; Dhanaraj & Mahambare, respondents exposed to the NEM drought. Similarly, the
2021). We further explored whether respondents with odds of EV increase by 44%. These results are significant
controlling partners also report a higher incidence of PV at the 1% level of significance. Although Rai et al. (2021)
by calculating the point-biserial correlation between the found a positive relationship between drought and PV,
number of control issues and the incidence of PV in the their results were not statistically significant. The main
treatment and comparison groups separately for the period reason for this may be that the data recall period during
before and after exposure to the NEM drought. The results data collection did not coincide with the drought period,
are presented in Table S3. We observed a strong association which may have led to an insignificant result. Our study
between having a controlling partner and experiencing improves upon Rai et al. (2021) by accurately capturing
PV. Although the size of the point-biserial correlation the effect of drought on IPV, which is one of the reasons
coefficient decreased from 2015 – 2016 and 2019 – 2021 for obtaining highly significant results in our study. Our
for the comparison group, it increased for the treatment findings are consistent with Epstein et al. (2020), who
group, further strengthening the finding that drought reported an increased risk of PV in women exposed to
exposure intensifies partners’ controlling behavior. both mild and severe droughts.
Having a controlling partner is a risk factor for IPV. We suspect that the causal link between drought
Figure 1A-C were generated by plotting the percentage of and IPV is driven by increased stress following reduced
women reporting a particular form of IPV for each number income. As discussed earlier, the NEM drought adversely
of control issues faced. A clear positive association between affected agricultural output. Decreased agricultural output
the number of control issues reported and the PV reporting translates into economic stress, which may intensify IPV.
is evident in Figure 1A. We also observed that respondents However, we cannot test this statistically, as the NFHS
in states not exposed to drought reported a lower incidence dataset does not contain information about income or
of PV in 2019 – 2021 for more than two control issues. In agricultural production. This could serve as a future
contrast, for the states exposed to drought, the incidence research agenda.
of PV was higher for almost every control issue category. Following the logic that drought affects agricultural
It is possible that decreasing income due to drought led output, we suspected a heterogeneous effect of exposure
to strife between partners, and the more controlling the to NEM drought between rural and urban areas. We
partner, the more likely this strife resulted in PV. We did estimated the models separately for rural and urban
not find any clear relationship between the number of samples. The results are interesting: the rural areas of

Volume 11 Issue 4 (2025) 80 https://doi.org/10.36922/ijps.3065