Adverse Consequences of Crime Victimization for School-Aged Girls in South Korea

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Social Sciences

Adverse consequences of crime victimization for school-aged girls: A case study of South Korea

Reviewer #2

The study is based on a national sample of South Korean female adolescents who were interviewed a decade ago. The main objective of the analysis is to determine if those who have been victimized in the past are more likely to experience (re)victimization. Additionally, the authors examined the effect of prior victimization on life satisfaction, emotional well-being, fear of crime, and delinquent behavior. The paper is clearly written, and the data analysis appears to be correctly executed, but the paper is not particularly well conceptualized and the contribution of the study to the literature is not apparent. Some of the issues the authors should address are listed below.

Lit review

Although several theoretical concepts (e.g., low self-control, parental attachment, school attachment, negative emotions, strain, etc.) are used, the analysis is not informed by a theory and no hypotheses have been formulated. The authors mentioned Agnew’s strain theory, which could have been used to frame a more focused analysis. For instance, it can be assumed that past victimization as well as other sources of strain will produce negative emotions, which in turn would mediate the effect of prior victimization on more recent victimization and delinquent behavior. A mediation analysis would have made this paper more interesting.

Methods

The authors should note that their study is based on a secondary data analysis (if this is the case). The authors should also indicate how they obtained access to the data.

What was the response rate? How did the authors deal with missing cases?

Measures

The description of the measures should be improved, and more details should be added.

Life satisfaction – it is not clear how many questions have been used to create the composite measure. It appears that 3 items have been used and the authors should list these three questions.

Negative emotions – the authors should list the scale components. Are these emotions recent?

Fear of crime – the authors used 21 questions to create the measure, but they didn’t offer any examples of the questions used. Is this measure unidimensional? Does it have construct validity?

Stress – “Four items assessed the level of stress respondents experienced related to academic performance, relationships with parents, parenting, and peer relationships.” What exact questions have been used? What was the time frame?

Past and current victimization – The authors noted that 14 questions were used for each index, but some examples should be provided. Did these measures include verbal and emotional victimization? Why wasn’t the severity of victimization considered? The authors could have focused only on violent victimization.

The authors explored the effect of prior victimization on recent victimization and delinquent behavior, but they didn’t examine the effect of prior delinquency on recent delinquency. The revised analysis should control for prior offending as well.

Results

It is customary to report the descriptive statistics (see Table 1) in the “results” section of the paper and not before the variables of interest are described.

Although the paper is about victimization and its detrimental effects, the authors did not indicate what proportion of girls did experience at least one type of victimization or what type of victimization was more common. Based on the descriptive stats, it appears that most girls were never victimized in the past or recently. This should be noted.

The results of the univariate analysis should be briefly presented. For instance, what percentage of girls engaged in delinquent behavior? What delinquent acts were more frequent? On average, what were the levels of fear of crime, life satisfaction, and emotional well-being? The authors should not reiterate what’s presented in the table; the reported values should be interpreted (e.g., what does a mean of .32 for recent victimization indicate?).

The authors should indicate if collinearity diagnostics were obtained. Results of the bivariate analyses (correlation matrix) should be included.

Discussion

The authors should compare their results with those presented in prior research. Findings should be more cautiously interpreted. For instance, several studies showed that offending and victimization often overlap, but the analysis didn’t take this into account. On p. 8, the authors noted: “…the results of the model for delinquent behavior show that previous victimization increases the likelihood of students committing delinquent behavior (IRR=1.13, p<.001”. But those who reported prior victimization could have been offenders as well. So, we should not assume that current offending/delinquent behavior is solely the result of prior victimization when prior offending was not considered.

The study has more limitations than those presented (see for instance, the construction of the measures and the lack of variation of the focal independent variable). And what are the policy implications of the results?

Minor issues

p.1 -> Abstract: “The results of regression analyses indicate that previous victimization leads to (…) a higher chance of victimization (…).” The sentence should be reformulated (e.g., prior victimization appears to increase the risk of revictimization and delinquent behavior).

P. 2 – > Backgrounds [it should be Background]

p. 3 -> “Andrews and colleagues (2003) also found that women victims (…) are more responsive to negative responses from family and friends.” Please reformulate! It is not clear what being “responsive” means.

p. 3, line 140 -> “South Korea is a well-known with its great academic performance but low psychological well-being of youth the well-being” / The sentence should be revised.

Tables 1 & 3 include a variable labeled “violence”. If this refers to delinquent behavior, the authors should make the necessary corrections.

 

Author Response

We appreciate the valuable comments. We have revised the paper accordingly, with changes marked in yellow in the manuscript.

Lit review

Although several theoretical concepts (e.g., low self-control, parental attachment, school attachment, negative emotions, strain, etc.) are used, the analysis is not informed by a theory and no hypotheses have been formulated. The authors mentioned Agnew’s strain theory, which could have been used to frame a more focused analysis. For instance, it can be assumed that past victimization as well as other sources of strain will produce negative emotions, which in turn would mediate the effect of prior victimization on more recent victimization and delinquent behavior. A mediation analysis would have made this paper more interesting.

• We fully agree with R2 that it can strengthen our paper. However, this study aims to explore the various consequences of crime victimization, and we have identified significant links between victimization and adverse outcomes. We would like to reserve the exploration of mechanisms for future studies. We have added a statement clarifying that the purpose of this study is to examine the various adverse consequences of crime victimization.

Methods

The authors should note that their study is based on a secondary data analysis (if this is the case). The authors should also indicate how they obtained access to the data.

• We have indicated that the current study analyzes secondary data and identified the sources of that data

What was the response rate? How did the authors deal with missing cases?

• The technical report of the Korean Youth Victimization Survey indicated that, considering the characteristics of schools (strata), 1,546 out of 2,336 schools in Korea were contacted, and 158 schools (10.2%), including 77 middle schools and 81 high schools, participated in the survey. Additionally, students from two selected classes responded to the survey under the guidance of their home-class teachers. In the analysis, no missing cases were reported. We have added more details about the sampling and data collection in the data section

Measures

The description of the measures should be improved, and more details should be added.

Life satisfaction – it is not clear how many questions have been used to create the composite measure. It appears that 3 items have been used and the authors should list these three questions.

• We have included an example question along with the factor loading values

Negative emotions – the authors should list the scale components. Are these emotions recent?

• Negative emotions are assessed based on feelings respondents experienced in the past month. The items asked how often respondents felt these emotions during that time.
• We have included an example question along with the factor loading values

 

Fear of crime – the authors used 21 questions to create the measure, but they didn’t offer any examples of the questions used. Is this measure unidimensional? Does it have construct validity?

• We have included an example question along with the factor loading values
• Items asking specific fear of crime

Stress – “Four items assessed the level of stress respondents experienced related to academic performance, relationships with parents, parenting, and peer relationships.” What exact questions have been used? What was the time frame?

• The items inquire about the general levels of stress that respondents experience. We have included an example question.

Past and current victimization – The authors noted that 14 questions were used for each index, but some examples should be provided. Did these measures include verbal and emotional victimization? Why wasn’t the severity of victimization considered? The authors could have focused only on violent victimization.

• Examples of victimization have been included. However, this study focuses on exploring the consequences of victimization and does not differentiate between types of victimization based on severity.

The authors explored the effect of prior victimization on recent victimization and delinquent behavior, but they didn’t examine the effect of prior delinquency on recent delinquency. The revised analysis should control for prior offending as well.

• The data did not include information about prior offending behavior, as it is not a longitudinal survey. We have addressed this limitation regarding causality and the lack of consideration of other important compounding effects in the discussion

Results

It is customary to report the descriptive statistics (see Table 1) in the “results” section of the paper and not before the variables of interest are described.

• The results of the descriptive statistics are now included in the results section, with a more detailed description provided

Although the paper is about victimization and its detrimental effects, the authors did not indicate what proportion of girls did experience at least one type of victimization or what type of victimization was more common. Based on the descriptive stats, it appears that most girls were never victimized in the past or recently. This should be noted.

• The percentage and actual number of respondents who experienced victimization are presented, along with the most common type of victimization

The results of the univariate analysis should be briefly presented. For instance, what percentage of girls engaged in delinquent behavior? What delinquent acts were more frequent? On average, what were the levels of fear of crime, life satisfaction, and emotional well-being? The authors should not reiterate what’s presented in the table; the reported values should be interpreted (e.g., what does a mean of .32 for recent victimization indicate?).

• We have added more details about the characteristics of the population

The authors should indicate if collinearity diagnostics were obtained. Results of the bivariate analyses (correlation matrix) should be included.

• Bivariate correlation analysis was conducted, and the results are presented in Table 2.

Discussion

The authors should compare their results with those presented in prior research. Findings should be more cautiously interpreted. For instance, several studies showed that offending and victimization often overlap, but the analysis didn’t take this into account. On p. 8, the authors noted: “…the results of the model for delinquent behavior show that previous victimization increases the likelihood of students committing delinquent behavior (IRR=1.13, p<.001”. But those who reported prior victimization could have been offenders as well. So, we should not assume that current offending/delinquent behavior is solely the result of prior victimization when prior offending was not considered.

• We have interpreted the results more carefully and added a limitation regarding the interpretation of causal relationships between factors in the discussion

The study has more limitations than those presented (see for instance, the construction of the measures and the lack of variation of the focal independent variable). And what are the policy implications of the results?

• We acknowledge additional limitations and have included possible policy implications

Minor issues

p.1 -> Abstract: “The results of regression analyses indicate that previous victimization leads to (…) a higher chance of victimization (…).” The sentence should be reformulated (e.g., prior victimization appears to increase the risk of revictimization and delinquent behavior).

• Abstract has been revised

1. 2 – > Backgrounds [it should be Background]

• Corrected

1. 3 -> “Andrews and colleagues (2003) also found that women victims (…) are more responsive to negative responses from family and friends.” Please reformulate! It is not clear what being “responsive” means.

• We revised the statement and argued that women are more sensitive to negative response

1. 3, line 140 -> “South Korea is a well-known with its great academic performance but low psychological well-being of youth the well-being” / The sentence should be revised.

• The sentence has been revised
• “South Korea is renowned for its exceptional academic performance; however, this is often accompanied by low psychological well-being among its youth, primarily due to significant academic stress and pressure”

Tables 1 & 3 include a variable labeled “violence”. If this refers to delinquent behavior, the authors should make the necessary corrections.

• The term “violence” changed into “delinquent behavior”

Reviewer 2 Report

Comments and Suggestions for Authors

Please see attached comments.

Comments for author File: Comments.pdf

Author Response

We appreciate the valuable comments. We have revised the paper accordingly, with changes marked in yellow in the manuscript

 

Line 92: The relationship with risky lifestyles is more cyclical. We don’t want to say that victimization causes risky lifestyles. Perhaps reword this piece, adding in something about negative coping mechanisms.

• We agree with R1. We have revised the statement and argued that victimization tends to lead to the development of negative coping mechanisms for dealing with strain or stress
• “For instance, individuals who have experienced crime victimization tend to develop negative coping mechanisms to deal with strain or stress, making them more prone to further victimization. Examples include juveniles running away from home or frequent substance use among those with traumatic victimization. (Kim et al. 2009; Ruback et al. 2014; Tyler and Johnson 2006).”

 

• Line 112: I’m not sure if women are more prone to physical assault than men. They are definitely more prone to intimate partner assault, but not assault generally.
• We have revised the statement and argued that women are more prone to physical assault by intimate partners

 

• Line 117: “intimate partner violence and rape” would be more appropriate here. It seems weird to say that intimate partner violence leads to unplanned pregnancies. It makes it sound like you’re suggesting that IPV rape is not rape.
• We have revised the statement

 

• Line 129: Women who experience ANY form of crime victimization are less likely or women who experience IPV and rape?
• The paper examined the impact of violent and property victimization, suggesting that women who experience such victimization are less likely to participate in social activities.
• Is there any research on consequences of teen victimization? Perhaps some of the studies you cited are about this, but it is not clear. This entire introduction seems to be about adults, when your study is on adolescents.
• We have added additional statements highlighting the importance of studying the impact of crime victimization on girls
• Given that children and juveniles are the most vulnerable segments of the population to crime victimization, and that the impact is severe and critical for these groups (Finkelhor 2008; Macmillan 2001; Turanovic 2019), it is imperative to study the adverse consequences of crime victimization specifically for girls
• What about consequences of victimization in Eastern cultures? Is there anything specifically on this? Your literature review is good, but it doesn’t provide much in relation to the actual population are looking at.
• Thank you for bringing this point to our attention. We have added a section on crime victimization and its consequences for Korean youth on page 4

 

• Lines 140-141: Proofread and edit
• we reviewed these lines and edited them.

 

• A small paragraph on self-control should be added since that is a variable in your study and the research is abundant.
• Since self-control is not a primary focus of the study, we have added statements in the measurement section explaining why it is necessary to include it as a control variable.
• “Individuals with low self-control tend to prioritize immediate gratification, which can lead to higher engagement in criminal behavior (Gottfredson and Hirschi, 1990). Studies have also identified self-control as a significant predictor of life satisfaction (e.g., Hofmann et al., 2014) and fear of crime (e.g., Higgins et al., 2008).”

 

Methods/Results

• Satisfaction with life: Please provide an example question.
• We’ve added an example question

 

• All variables: Did you use previously validated scales/items or did you create your own? Please specify and cite scales that are validated.
• The current study analyzed secondary data, utilizing items related to the concept of the variables. While we did not use validated scales, we believe that the face validity and statistical support (e.g., factor loadings and Cronbach’s alpha value) for these items lend credibility to our measurement strategies

 

• Was there any overlap between the types of crime? For instance, would hitting someone at school be a school crime, violent crime, or both?
• The items clearly identify the location or behavior associated with criminal activity. While respondents may have been confused by some questions, the organization of the questions minimizes the likelihood of overlap between different types of crime.
• Please explain the socioeconomic variable. I suggest using South Korean currency and the appropriate symbol. Right now, it reads as 1000 USD, which would barely let someone survive one month.
• We provided amounts in both Korean currency and USD.

 

• Please include an “N” for each dependent variable model so we can see exactly how many respondents were in each model.
• We’ve included “N” for each dependent variable.

 

• Current victimization is a scale, but in the results, you refer to things as leading to current victimization or not. Did you recode current victimization into two categories for the analysis? If it is still a summative scale, then results would suggest “more types of recent victimization” not just any current victimization or not.
• We have changed the term 'more victimization' to 'more types of recent victimization.'

 

• Delinquent behavior: Same comments as the current victimization variable Discussion
• We have also changed into ‘more types of delinquent behavior’

 

• Policy implications needs to be significantly expanded on, especially since you are examining a less-studied population. Please relate these specifically to the population studied and support them with research on either Western-teen populations or Eastern population generally.
• We have added policy implications for the global teen population in the discussion section

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors addressed most of my concerns and I found the manuscript improved. They also highlighted the limitations of their analysis. There are, however, a few things that require the authors’ attention.

p.3 -> “South Korea is renowned for its exceptional academic performance.” South Korea or South Korean students? Please revise.

The paper is about the effects of victimization; are the authors suggesting that parental or societal pressure to perform well in school is a form of victimization? It is not clear how the strain caused by the above-mentioned pressure relates to this paper. The authors should clarify what they intend to convey. Are they suggesting that those who perform well in school are more often victims of bullying? The newly introduced paragraph on page 3 should be revised [starting with line 143]. Maybe, this discussion about academic performance should be placed later to justify the inclusion of one of the control variables.

p.3, line 174 ->  Specifically2023 [delete 2023]

line 177 -> “Based on findings from previous literature and theoretical frameworks such as strain theory…” Who formulated the strain theory and when? References should be included. What is the strain theory stating? The authors should not assume that the reader is familiar with the theory. Hypotheses should be listed in accordance with the theoretical assertions. More needs to be added here.

In table 2 (correlation matrix) the authors should number the variables listed in the first column. Prior victimization is the focal independent variable and it should be the first variable in the list; this way, the reader can see clearly how it relates to the  other variables in the model.

Author Response

We appreciate the valuable comments of Reviewer 1. Here are our responses and the revised manuscript

The authors addressed most of my concerns and I found the manuscript improved. They also highlighted the limitations of their analysis. There are, however, a few things that require the authors’ attention.

p.3 -> “South Korea is renowned for its exceptional academic performance.” South Korea or South Korean students? Please revise.

• As another reviewer pointed out, the sentence is not closely relevant to the topic and it has been removed

The paper is about the effects of victimization; are the authors suggesting that parental or societal pressure to perform well in school is a form of victimization? It is not clear how the strain caused by the above-mentioned pressure relates to this paper. The authors should clarify what they intend to convey. Are they suggesting that those who perform well in school are more often victims of bullying? The newly introduced paragraph on page 3 should be revised [starting with line 143]. Maybe, this discussion about academic performance should be placed later to justify the inclusion of one of the control variables.

• Thank you for pointing this out. We've tried to introduce the unique characteristics of Korean youth and the educational environment that can be related to crime. However, these sentences were not relevant to the scope of the paper and have been removed from the section. Now, the section solely focuses on crime victimization and its consequences among Korean youth. 

p.3, line 174 -> Specifically2023 [delete 2023]

• 2023 is deleted.

line 177 -> “Based on findings from previous literature and theoretical frameworks such as strain theory…” Who formulated the strain theory and when? References should be included. What is the strain theory stating? The authors should not assume that the reader is familiar with the theory. Hypotheses should be listed in accordance with the theoretical assertions. More needs to be added here.

• Thank you for pointing this out. The core concept of general strain theory was articulated in the victimization section on page 3. We have added more content about general strain theory to page 3. 

In table 2 (correlation matrix) the authors should number the variables listed in the first column. Prior victimization is the focal independent variable and it should be the first variable in the list; this way, the reader can see clearly how it relates to the  other variables in the model.

• The previous victimization variable is now listed as the first variable, and numbers have been assigned to each variable

Reviewer 2 Report

Comments and Suggestions for Authors

·         Introduction (first three paragraphs) could be condensed. They are repetitive.

·         Line 59: Remove. This is the 3rd time this sentence is used in the first 2 pages.

·         Lines 62-69: Condense to one sentence. Very repetitive.

·         The studies you mention in lines 155-167 seem to seem to counteract your statement about this area being less documented. What is your study adding that the previous studies did not? An expansion on adolescent research may help this as well.

·         Household income: Weekly? Monthly? Annual?

·         Discussion does not relate the findings to previous research. At a minimum, the results of this study should be related to the previous research done on Korean youth.

Author Response

We appreciate the valuable comments of Reviewer 2. Here are our responses and the revised manuscript

Introduction (first three paragraphs) could be condensed. They are repetitive.

• We have condensed the introduction by deleting some repetitive sentences

Line 59: Remove. This is the 3rd time this sentence is used in the first 2 pages.

• The sentence is removed.

Lines 62-69: Condense to one sentence. Very repetitive.

• This section has been condensed.

The studies you mention in lines 155-167 seem to seem to counteract your statement about this area being less documented. What is your study adding that the previous studies did not? An expansion on adolescent research may help this as well.

• Even though several studies have examined the effects of victimization on Korean youth, they focused on limited aspects of adverse outcomes. The current study aims to expand the previous literature by focusing on various adverse outcomes and the school-aged girl population. This point is highlighted in our research.
• Taken together, studies have reported that victimization can create adverse consequences for Korean youth. However, a comprehensive approach to exploring these impacts has rarely been utilized. Specifically, focusing on female adolescents, who often perceive higher levels of fear of crime in Korea (Lee and Park 2022), will be beneficial for expanding the understanding of the various adverse impacts of crime victimization among school-aged girls in South Korea. 

Household income: Weekly? Monthly? Annual?

• Household income captures monthly income. We've clarified it in the measurement section. 

Discussion does not relate the findings to previous research. At a minimum, the results of this study should be related to the previous research done on Korean youth.

• The discussion now relates the findings to previous research conducted in Korea

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript has been improved greatly since the first submission. Great work to the authors.

Comments on the Quality of English Language

Some minor issues with tense (i.e. have versus has, is versus are) - Mainly on page 1.