The Gender Violence - Implicit Association Test to measure attitudes toward intimate partner violence against women

Abstract

Attitudes play a central role in intimate partner violence against women and are related to its origin, to the responses of women who suffer violence, and to the settings where it occurs. In fact, these attitudes are recognized as one of the risk factors linked to violent perpetration and to public, professional, and victim responses to this type of violence. However, even though available research generally shows a broad rejection of this violence, it remains a serious social and health problem that has reached epidemic proportions. This suggests that the information available about these attitudes (obtained through explicit and direct measures, i.e., self-reports) may be distorted or influenced by factors such as social desirability. In this context, the overall objective of our research project is to provide multi-method measures (explicit and implicit) of attitudes toward intimate partner violence against women, and the main goal of this paper is to propose an instrument for the implicit measurement of these attitudes. In this regard, the Implicit Association Test (IAT) is the most common procedure used, providing a superior predictive validity compared to explicit measures for socially sensitive topics. We will present an exploratory study that describes its adaptation for our purposes, and the development of the Gender Violence - Implicit Association Test (GV-IAT) to use among Spanish-speaking populations, and discuss the strengths and limitations of this proposal.

Psychosocial Implications of Supportive Attitudes towards Intimate Partner Violence against Women throughout the Lifecycle

“…The Gender Violence Implicit Association Test (GV-IAT [ 64 , 65 , 66 ]) is a form of personalized IAT used as an implicit measure of attitudes towards IPVAW with two target categories—Gender Violence vs. Non-Gender Violence—and two attribute categories—Good vs. Bad. Related to the target category, it is important to note that, under Spanish law [ 67 ], IPVAW is known as gender violence (see [ 25 ]), which is why GV-IAT uses the term “Gender Violence” to refer to IPVAW.…”

“…The fundamental principle of IAT is that, where two concepts are strongly associated (compatible phase), the response latency (RL) is less than when this is not the case (incompatible phase). The IAT effect (D-scores) is calculated using the difference between incompatible-compatible critical phase latency-based responses [ 54 , 64 ]. The typical IAT procedure with feedback was used [ 64 , 66 ], and the participants completed the GV-IAT task in seven blocks, of which three were considered practice trials (B1–B2 including 24 trials and B5 including 48 trials) and four as critical blocks (B3–B6 including 24 trials in the compatible phase and B4–B7 including 48 trials in the incompatible phase) [ 54 ].…”

“…The IAT effect (D-scores) is calculated using the difference between incompatible-compatible critical phase latency-based responses [ 54 , 64 ]. The typical IAT procedure with feedback was used [ 64 , 66 ], and the participants completed the GV-IAT task in seven blocks, of which three were considered practice trials (B1–B2 including 24 trials and B5 including 48 trials) and four as critical blocks (B3–B6 including 24 trials in the compatible phase and B4–B7 including 48 trials in the incompatible phase) [ 54 ]. Additionally, the first trials of each critical block were considered “practice trials” and only the next trials of each block were computed in the D score (in the end, 20 trials were computed in B3–B6 and 40 trials in B4–B7).…”

“…In summary, the higher is the D-score value, the higher is the difference between response latencies in incompatible versus compatible phases and, therefore, the stronger is the implicit rejection of IPVAW. The internal consistency by the split-half reliability method [ 54 ] recommended by Kurdi et al [ 70 ] yielded an average estimate of 0.73 [ 64 ], indicating an adequate reliability [ 49 , 71 ].…”

O Que os Olhos Não Veem, o Coração Não Sente? Desenvolvimento de um instrumento brasileiro para avaliar a violência psicológica contra a mulher

“…Assim, ela pode estar presente em uma parcela significativa dos relacionamentos, mas não ser percebida como motivo para intervenção, uma vez que parece existir um certo grau de naturalização, além de crenças e atitudes tolerantes em relação a essa violência nos relacionamentos íntimos (Ferrer-Perez, Sánchez-Prada, Delgado-Álvarez & Bosch-Fiol, 2020), o que gera impactos negativos para a mulher e suas relações. Pode ocorrer uma percepção de que os comportamentos violentos são "um modo de ser casal", não sendo identificados como uma violência de fato.…”