Psychometric Properties of the Violence Scale (EV) in Colombian Men

Response to Reviewer 1 Comments

Questions

Response and Revisions

Is the content succinctly described and contextualized with respect to previous and present theoretical background and empirical research (if applicable) on the topic? Must be improved

The adjustments were made according to the comments.

All revisions are blue on the paper.

Are the research design, questions, hypotheses and methods clearly stated? Must be improved

The adjustments were made according to the comments.

Are the arguments and discussion of findings coherent, balanced and compelling? Must be improved

The adjustments were made according to the comments.

For empirical research, are the results clearly presented? Must be improved

The adjustments were made according to the comments.

Is the article adequately referenced? Must be improved

The adjustments were made according to the comments.

Are the conclusions thoroughly supported by the results presented in the article or referenced in secondary literature? Must be improved

The adjustments were made according to the comments.

Comments and Suggestions for Authors

Response and Revisions

In its present form, the article is confusing with a range of different terms used that seem to be used interchangeably but are not the same. The article commences by discussing Intimate Partner Violence. It then discusses domestic violence. Different types of violence are briefly mentioned. There is also mention of spouse abuse, which falls under the term family violence in many countries. There is also discussion on violence by men against women and by women against men. It was confusing what was being discussed and what was understood to be known about each type because the type of family violence perpetrated by each gender tends to differ. 

Thank you for this suggestion. Since three studies that used the same intimate partner violence scale were compared—the study by Valdez-Santiago et al. (2006) involving women, the study by Acuña-Arango (in press) involving women, and the present study involving men—these results are compared by analyzing the phenomenon of intimate partner violence among men and women. However, several adjustments were made in the discussion to clarify that the study analyzes intimate partner violence against men, even though historically women have been the victims.

There was not enough information about the method to be clear on how the data were collected and what processes were employed. 

This section has been revised to provide more detail: the data were collected using a snowball sampling strategy by distributing a Google Forms to the researchers’ contacts, who in turn forwarded it to other people. This has been included in the manuscript.

The results tables were unclear. It is also unclear why the scale used as a foundation was selected.

We appreciate the suggestion regarding the scale: we used this scale as a basis because this article is part of a larger project that seeks to study violence against men and women in a country like Colombia, which has a high level of violence, and because the scale is a tool that covers the various types of violence referenced in the literature (physical, severe physical, sexual, psychological, and economic). This clarification has been included in the introduction.

Regarding the tables, we acknowledge that Table 3 may be confusing due to the inclusion of the items. We included them so that the reader can clearly identify which item corresponds to each dimension; however, when performing AFC with rival models, this can be difficult to understand. Nevertheless, we appreciate the opportunity to present the three models side by side to demonstrate that the loadings are indeed better in model 3.

it is crucial to determine what the study is about. Is is IPV? Is it family violence? Is it domestic violence? These are not the same things.

The terms were consolidated into a single one: Intimate Partner Violence (IPV).

I would like to encourage the authors to revise the article. It does require a lot of work though as it is not at all clear in its current format. 

A thorough review of the entire article was conducted in accordance with the suggestions.

I would like to thank the authors again and wish them the very best with their research in this very important field.

Thank you for your feedback, which helps us improve our article

4. Response to Comments on the Quality of English Language

The English could be improved to more clearly express the research.

A native English speaker reviewed the entire article once the recommended changes had been made.

Response to Reviewer 2 Comments

Questions

Response and Revisions

Is the content succinctly described and contextualized with respect to previous and present theoretical background and empirical research (if applicable) on the topic? Can be improved

The adjustments were made according to the comments.

All revisions are blue on the paper.

Are the research design, questions, hypotheses and methods clearly stated? Can be improved

The adjustments were made according to the comments.

Are the arguments and discussion of findings coherent, balanced and compelling? Yes

The adjustments were made according to the comments.

For empirical research, are the results clearly presented? Can be improved

The adjustments were made according to the comments.

Is the article adequately referenced? Can be improved

The adjustments were made according to the comments.

Are the conclusions thoroughly supported by the results presented in the article or referenced in secondary literature? Can be improved

The adjustments were made according to the comments.

Comments and Suggestions for Authors

Response and Revisions

The main concerns relate to the sample, the framing of male victimisation, the psychometric analysis, the interpretation of low factor loadings, the handling of missing data, and the need for more cautious conclusions.

We appreciate this comment: these issues were corrected throughout the paper.

Regarding the absence of data on occupation or marital status, these variables are presented as part of the participants’ characteristics, but they are not included in the statistical analysis nor are they analyzed variables. However, we accept the recommendation, note this as a limitation, and highlight the need for factor analyses that include these variables as indicators of factor invariance.

Regarding the SRMR values, which fall just within the acceptable range, this may be due to some low factor loadings; however, the other AFC values are adequate and high, which supports the model’s suitability.

Finally, regarding the suitability of certain items with low loadings, we accept the recommendation: indeed, the loadings of some items—which, while still significant, are low in some cases—are explained by the low frequency of certain events (the most severe violent behaviors occur less frequently), which results in lower loadings; but their factor consistency and conceptual congruence support their inclusion, and other indicators—such as the AFC values themselves, reliability values, and item-item and item-test correlations—support their retention. We have included this in the manuscript.

The sample needs a clearer description and justification. The authors state that participants were Colombian men aged 18 to 65, but the mean age was 24.94 with a standard deviation of 4.29, suggesting a predominantly young sample. The manuscript should provide a clearer age distribution and explain how representative this sample is of Colombian men in relationships.

We accept the reviewer’s recommendation: there is indeed an age imbalance, and most participants were young adults; therefore, we recommend replicating these findings with participants of other age groups. This was included in the “Limitations” section of our study.

The inclusion criteria require clarification. The authors state that participants were men in a relationship or cohabiting for more than two years and “not considered victims of IPV.” However, the study aims to validate a tool for assessing violent behaviors experienced by men. The authors should explain whether this means participants had never reported IPV, did not self-identify as victims, or were not registered in official victim records. This distinction is important because it affects how the scale can be interpreted and applied.

We have improved the description of the inclusion criteria, and we have also added the following text: “The participants had never reported intimate partner violence, did not identify themselves as victims, and were not listed in official victim registries”.

The sampling method limits generalisability. Participants were recruited through non-probability and snowball sampling. This is acceptable for an initial validation study, but the authors should be more cautious when making claims about Colombian men generally. The limitation is acknowledged briefly, but it should be integrated more strongly into the interpretation of the findings.

We agree with the comment and accept it: we have expanded the section on limitations to include more details about the limitations of our sample, as well as limitations related to imbalances in age, region, educational level, and family type, which make it difficult to truly generalize these findings, and we recommend replication studies that carefully control for these factors.

There is substantial missing sociodemographic information. Table 1 shows high levels of missing data for occupation and marital status. The authors should explain why these data are missing and how missingness was handled in the analysis. If the missingness is systematic, it may affect the interpretation of the sample and findings.

Regarding the lack of data on employment status or marital status, these variables are presented as part of the participants’ profile but are not included in the statistical analysis or analyzed as variables. However, we accept the recommendation, note this as a limitation, and suggest the need for factor analyses that include these variables as indicators of factor invariance.

The content validity process needs more detail. The authors state that eight expert judges participated, including four IPV professionals and four IPV victims. However, the manuscript should provide more information about how these judges were selected, what criteria they used, whether a formal content validity index was calculated, and what changes were made to the items after this stage.

This part was explained in greater detail in the document, in the “Method and Procedure” section.

The adaptation of the instrument for men needs a clearer explanation. Because the original scale was developed in México with women, the authors should explain how item wording was adapted for men, whether cognitive interviews were conducted, and whether the items were tested for clarity, acceptability, and gender relevance in the Colombian context.

We appreciate the recommendation: we have adjusted the procedure to provide greater clarity about how the wording of the items was adapted for men (expert judges and former victims of IPV), and how the clarity, pertinence, relevance, and adequacy of each item were reviewed.

The statistical methods are generally appropriate, but they need more detail. The authors state that CFA was performed using a weighted least squares method suitable for ordinal items and that reliability and convergent validity were assessed. However, the manuscript should clarify whether polychoric correlations were used, how missing item data were handled, and how model identification was achieved.

Yes, the matrices used for the AFC were indeed polychoric matrices, given the ordinal nature of the items (this is already the default setting in the software if items with ordinal categories are selected and the described method is used); however, this was not explicitly stated in the text, and it has now been included.

The choice of the five-factor model is plausible, but the interpretation needs caution. The model has good global fit indicators, including CFI and TLI above 0.95, and lower RMSEA than the other models. However, the SRMR is exactly 0.080, which is at the threshold rather than clearly below it. The authors should avoid overstating this as an unequivocally excellent fit.

Regarding the SRMR values, which fall right at the acceptable threshold, this may be due to some low factor loadings (this occurs in items that address severe violent behaviors, which are infrequent and load asymmetrically). However, the other AFC values are adequate and high, which supports the model’s suitability. In any case, these implications are discussed in the discussion section.

Several factor loadings are very low, especially for severe physical violence items. In Table 3, some items have loadings below conventional thresholds, even if they are statistically significant. Statistical significance is likely influenced by the large sample size. The authors should discuss whether these items are substantively and psychometrically strong enough to retain, rather than relying only on p-values.

Regarding the suitability of certain items with low loadings, we accept the recommendation: indeed, the loadings of some items—which, while still significant, are low in some cases—can be explained by the low frequency of certain events (the most severe violent behaviors occur less frequently), and this results in lower loadings; however, their consistency with the factor and their

Conceptual consistency supports its inclusion, and other indicators—such as the AFC’s own values or reliability values, item-item correlations, and item-test correlations—support its retention. We have included this in the manuscript.

The manuscript states that all items load on their factor and are significant, but this needs to be checked carefully. Some loadings appear extremely low, and at least one item in one model appears non-significant. The authors should revise this statement and provide a more nuanced interpretation.

Thank you for reviewing the table: we have verified it, and all loadings are significant even though the factor loadings are low. And to be extra cautious, we reviewed reliability indicators (which are adequate for all items), simulations of removing each item (in all cases, reliability decreases, which speaks to the internal consistency of the factor and the contribution of each item), item-item correlations within each factor, and item-test correlations (significant correlations greater than 0.30), which allows us to affirm that each item belongs to the dimension in which it is classified. In any case, we have adjusted the discussion to note that these factor loadings were low and to provide an explanation for this.

The severe physical violence factor requires special attention. The authors explain that severe events had a low frequency, which may explain the low loadings. This is reasonable, but it also means the factor may be unstable in a community sample of men who do not self-identify as IPV victims. The authors should discuss this more directly.

We agree on the attention that severe physical violence and the items in this dimension require: we expand the discussion along these lines, noting that despite these values, the items are effective for assessing severe physical violence; however, this is a serious issue: it points to a normalization of violence, and we must focus on measurement processes in community samples, rather than waiting for reports of abuse, and work on prevention programs, given the severity of these violent behaviors. This has been included in the manuscript.

The use of AVE as evidence of convergent validity is not sufficient on its own. The authors should avoid presenting AVE as full evidence of convergent validity unless external related constructs were also examined. If no external measures were included, the manuscript should state that convergent validity evidence is limited.

Agreed: We clarify this limitation by noting that AVEs are measures of internal convergence, but that we do not conduct convergence analyses with other measures, and that this should be addressed in future research.

The manuscript should consider additional validity evidence. For example, known-groups validity, criterion validity, test-retest reliability, or measurement invariance across age, socioeconomic level, relationship status or region would strengthen the study. If these analyses are not possible, this should be acknowledged as a limitation.

We agree with this recommendation, and we have included these aspects, such as intergroup comparisons, test-retest reliability, and factor invariance tests, among other analyses, in the limitations of our study; we also highlight the need for these lines of research in future studies.

The discussion should be more balanced in its treatment of gender and IPV. It is important to recognise male victimisation and underreporting, but this should not be framed in a way that minimises the gendered burden of violence against women. The authors should clearly state that recognising male victimisation can complement, rather than displace, gender-sensitive IPV research and services.

This comment is very interesting; in the discussion section, it was revised to provide a more balanced analysis.

The manuscript should be cautious with claims of being a “pioneer” study in Latin America unless this is fully supported by a systematic search. It may be safer to state that the study contributes to a limited evidence base on IPV measurement among men in Colombia and Latin America.

We have adjusted this terminology to be more cautious with this statement.

Ethical reporting should be expanded. The manuscript states that informed consent was obtained and that ethics approval was granted, which is appropriate. However, given the sensitivity of IPV questions, the authors should also describe whether participants were provided with support resources, referral information, or a distress protocol if participation raised emotional discomfort.

We agree with the reviewer: these ethical considerations were not included in the manuscript because they were part of the approval granted by our ethics committee, however we have included the following statement: “Although participation was voluntary, if the questions prompted reflection, discomfort, or a desire to seek psychological services, email addresses were provided for participants to contact us.”

The manuscript may also benefit from engaging with recent work showing how intimate partner violence intersects with socioeconomic vulnerability and barriers to healthcare access. This is relevant because the study argues for better identification and assessment of IPV to inform care and prevention. One relevant source is 10.1016/j.socscimed.2025.117785  This could help strengthen the discussion of why measuring multiple dimensions of IPV, including economic and sexual violence, matters for health and social support responses.

The suggested reference was included and cited in the body of the document at the end of the discussion

The writing and terminology require careful editing. Terms such as “domestic violence,” “intimate partner violence,” “partner violence,” “IP,” and “VP” should be used consistently. The authors should also correct wording such as “economic violence,” awkward translations, repeated phrases, and formatting issues in the tables.

The terms used were consolidated into a single expression (IPV), which is: intimate partner violence.
We have reviewed the entire text to reduce awkward translations and repeated phrases, and we have adjusted formatting elements throughout the text. In addition, the manuscript was reviewed by a native English speaker. Regarding “economic violence,” this concept refers to actions that limit a partner’s access to, control over, or use of financial resources and assets, thereby reinforcing dependency and unequal power dynamics within the relationship, and it is the term to which our study aims to contribute (upon review, this is indeed the correct term). 

Lastly, this is a valuable and relevant manuscript, but it requires major revision before publication. The authors should strengthen the sample description, clarify inclusion criteria, provide more detail on content validation and item adaptation, interpret low factor loadings more cautiously, expand validity evidence or acknowledge its absence, strengthen ethical reporting, and moderate claims about the scale’s applicability.

All of these changes were made throughout the document and highlighted in blue so the reviewer could see them.

Thank you for your suggestions; they helped improve the article.

4. Response to Comments on the Quality of English Language

A native English speaker reviewed the entire article once the recommended changes had been made.

Response to Reviewer 1 Comments

Comments and Suggestions for Authors

Response and Revisions

1) The statement "However, IPV perpetrated by women against men 33 is still not recognized as a social problem" is incorrect. It is not widely recognised, and this is a problem and can be a barrier for men coming forward or getting support. However, it is not that it is not recognised and research in this area does exist which should be explored. 

The comment has been accepted. The revision has been made. Upon reviewing the referenced literature, we found that it emphasizes access to services than on recognizing the phenomenon as a social problem, and we have clarified that the historical view of intimate partner violence as a phenomenon that predominantly affects women has contributed to the marginalization of male victims, and that this is precisely why it is not widely recognized. This is exactly the problem.

2) A key strength of the article is linked to the previous point. There is insufficient understanding of violence perpetrated by women towards their male partners. This is a strength of this research. I think the authors need to be more proud in explaining this, as it seems almost buried throughout the article. 

We appreciate your comment and your acknowledgment of our study’s contribution: the adjustment has been made to the discussion section.

3) The conclusion refers to "domestic violence" rather than IPV which needs to be corrected. The conclusion should also be made stronger. I think the phrase that men are in second place is a little awkward and instead, the conclusion should state that violence by women against men is not well understood or researched and this can result in a lack of assistance and support for one gender, and more needs to be known to ensure the entire cycle of family violence is managed, which means understanding and supporting it for either gender.

We have revised the term “domestic violence” and replaced it with “intimate partner violence,” and we have done so throughout the document. 

About the conclusion, we have improved this section by including the points recommended by the evaluator

(violence by women against men is not well understood or researched, and this can result in a lack of assistance and support for one gender), supplementing them with the need for a comprehensive and balanced perspective for both women and men, and concluding with the usefulness of this instrument for assessing violence by women against men.

Response to Reviewer 2 Comments

Comments and Suggestions for Authors

Response and Revisions

First, the authors should avoid overstating the validity evidence. The manuscript still sometimes suggests that the scale is fully valid and reliable for assessing IPV against Colombian men. A more cautious phrasing would be preferable, such as “the scale provides initial evidence of validity and reliability in this sample.” This is important because the study does not include external validity measures, test-retest reliability, measurement invariance, or known-groups comparisons.

We appreciate the feedback, and we agree: we acknowledge that this study provides a scale with valuable data on its validity and reliability, but that elements such as measures of external validity, test-retest reliability, measurement invariance, or known-groups comparisons are missing. For this reason, we have refined the level of our assertions and indicated that there is adequate initial evidence of its validity and reliability, and that further complementary studies are needed to fully confirm its metric quality.

Second, the low factor loadings for some severe violence items should remain clearly acknowledged. The explanation that these items represent low-frequency events is reasonable, but very low loadings still require cautious interpretation. The authors should avoid implying that statistical significance alone is sufficient evidence of strong item performance.

In response to the editor’s suggestion, these items should be treated with greater caution. While we decided to retain them due to their relevance not only to the significance of the charges but also to the adjustment of the factorial model, their theoretical–conceptual contribution to measuring severe physical violence requires further scrutiny. Accordingly, we recommend that future research examine these items more carefully to ensure both methodological rigor and conceptual clarity.

Third, the ethical support statement should be strengthened slightly. The manuscript states that email addresses were provided if participants experienced discomfort or wished to seek psychological services. Given the sensitivity of IPV research, the authors should clarify whether participants were also provided with independent referral information or support services. If not, this should be acknowledged as a limitation.

The suggestion is accepted, and the adjustment is included in the limitations. Participants were informed that psychological support was available via email upon questionnaire completion. However, one limitation was the lack of localized information on accessible care centers, especially in the country’s major urban centers.

Fourth, Table 1 should be made clearer. The educational level categories are coded numerically but not explained. Occupation and marital status also have substantial missing data. The authors acknowledge this limitation, but the table itself should be easier for readers to interpret.

The commentary was reviewed, accepted, and revised in accordance with the reviewer's instructions. Specifically, in Table 1, the numerical code for each educational level was replaced with the precise name of each category. We have also included a note in the table regarding missing data, and that this demographic data is provided to identify the study participants.

Fifth, the manuscript still needs careful copy-editing. There are some remaining language issues, including “validate scale” instead of “validated scale,” “polychromic matrices” instead of “polychoric matrices,” “JASAP” where the intended software may be “JASP,” and occasional inconsistencies between “domestic violence,” “intimate partner violence,” and “VP.” The authors should ensure that terminology is consistent throughout.

We appreciate these comments: we have revised these terms (validate scale, JASP, polychoric matrices) and standardized the term “intimate partner violence” or “IPV” throughout the text.

Sixth, some item wording in Table 3 should be checked carefully. There are repeated or awkward phrases such as “knife, machete, or knife,” and some pronoun wording remains inconsistent. Since this is a psychometric paper, item wording must be precise and professionally presented.

The comment has been accepted. The corresponding adjustment has been made in Table 3 for all three models. These linguistic difficulties may be because explained the scale was validated in a Spanish-speaking Latin American context.

Lastly, the authors have substantially addressed the main concerns raised in the previous review. The manuscript now presents a clearer and more cautious account of the adaptation and psychometric testing of the Violence Scale-EV among Colombian men. The remaining issues are minor but important for clarity, ethical completeness, and precision. I support acceptance after these minor revisions are completed.

We appreciate these comments in our article: your initial feedback was very valuable in improving its scientific quality, and we have also taken your new recommendations into account. Thank you.