Page 154 - GHES-2-4

Global Health Economics and
Sustainability
Impact of dating violence on mental health

Items are scored on a seven-point Likert scale, with higher variable. IBM Statistical Package for the Social Sciences
scores indicating more traditional gender role attitudes. Statistics for Windows, version 22.0, was used for statistical
The internal consistency for this sample was 0.87. analyses and graphs.

2.2.6. Feminine/expressive and masculine/ 3. Results
instrumental and traits
3.1. Dating violence victimization and perpetration
To assess participants’ self-attribution of personality traits When analyzing the frequency of violence, we found the
stereotypically associated with masculinity and femininity, following percentages for victimization: 60.8% for men
the reduced and Spanish-translated version of Bem’s Sex with a same-sex partner, 59% for women with a same-
Role Inventory (Bem, 1981) was used. The masculinity sex partner, 67.5% for men with an opposite-sex partner,
scale consists of 10 items describing attributes traditionally
considered masculine, such as “independent,” “assertive,” and 48.6% for women with an opposite-sex partner. The
“dominant,” and “aggressive,” assessing masculine/ differences in these percentages were statistically significant,
2
instrumental traits. The femininity scale includes 10 items χ (3, n = 528) = 1.09, p = 0.018. For psychological violence
victimization, the percentages were 55.8%, 56.3%, 60.8%,
with attributes traditionally considered feminine, such as and 45.1%, respectively, and these differences were not
“compassionate,” “warm,” “tender,” and “kind,” assessing statistically significant, χ (3, n = 528) = 7.26, p = 0.064.
2
feminine/expressive traits (Matud, 2018). A seven-point The percentages of individuals who experienced physical
Likert-type scale was used for responses, with higher scores violence victimization were 21.7%, 11.1%, 25.8%, and 9.7%,
indicating greater self-attribution of these traits. In this respectively; these differences were statistically significant,
study’s sample, the internal consistency of the masculinity χ (3, n = 528) = 17.77, P < 0.001. The percentages of
2
scale was 0.77 and that of the femininity scale was 0.88. sexual violence victimization were 10%, 9%, 7.5%, and
2.2.7. Demographic data 6.3%, respectively; these differences were not statistically
significant, χ (3, n = 528) = 1.46, p = 0.692.
2
Participants’ sociodemographic characteristics were
collected using a sociodemographic data sheet, which The percentage of individuals reporting perpetration of
included gender (woman, man, other), age, marital status, dating violence was 51.7% for men with a same-sex partner,
occupation, and education. Since very few people (<1%) 52.1% for women with a same-sex partner, 56.7% for
selected the “other” category, only those identifying as men with an opposite-sex partner, and 45.8% for women
either a woman or a man were included in the study. with an opposite-sex partner. The differences were not
statistically significant, χ (3, n = 528) = 3.15, p = 0.369. The
2
2.3. Statistical analysis percentages of psychological violence perpetration were
Internal consistency was measured using Cronbach’s 50%, 48.6%, 52.5%, and 44.4%, respectively, which were not
2
alpha. Comparisons of sociodemographic characteristics significantly different, χ (3, n = 528) = 1.82, p = 0.611. The
between opposite-sex and same-sex partners and between percentages of physical violence perpetration were 10.8%,
men and women were made using student’s t-tests for age 9%, 20%, and 4.5%, respectively; these differences were
2
and Pearson’s Chi-squared test for education level, marital statistically significant, χ (3, n = 528) = 16.29, p = 0.001.
status, and occupation. To determine differences in dating The percentages of sexual violence perpetration were 5.8%,
violence perpetration and victimization between same-sex 2.1%, 3.3%, and 2.1%, respectively; these differences were
2
and opposite-sex partners and between women and men, not statistically significant, χ (3, n = 528) = 3.83, p = 0.281.
six 2 × 2 between-subjects factorial analyses of variance Table 3 presents the ranges of scores for victimization
(ANOVA) were conducted. The factor variables were and perpetration for each type of violence in the four
partner type (opposite-sex partner, same-sex partner) and groups. For both victimization and perpetration, the score
gender (women, men), with the dependent variables being range allowed by the scale was from 0 to 42 for total dating
psychological, physical, and sexual violence victimization violence, from 0 to 18 for psychological violence, from
in the first set of analyses and psychological, physical, 0 to 20 for physical violence, and from 0 to 4 for sexual
and sexual violence perpetration in the second set. violence. As shown in Table 3, for none of the types of
Correlational analyses were conducted to examine the violence victimization did the range of scores reach the
associations between victimization and perpetration of maximum allowed by the scale, indicating that the violence
dating violence, as well as the rest of the variables of the victimization was not severe or very frequent, particularly
study. All correlations were performed using Pearson’s r for total and physical violence. The range of scores for
correlation coefficient, except for education, which was total violence victimization was greater for men than
calculated using Spearman’s rho because it is an ordinal for women, and the range of scores for physical violence

Volume 2 Issue 4 (2024) 5 https://doi.org/10.36922/ghes.3300