Assessing Conflict Tactics Scale Validity by Examining Intimate Partner Violence Overreporting
View/ Open
File version
Accepted Manuscript (AM)
Author(s)
Ackerman, Jeffrey
Griffith University Author(s)
Year published
2018
Metadata
Show full item recordAbstract
Objective: Although many scholars have questioned, on a logical basis, the validity of the Conflict Tactics Scale (CTS) to adequately measure intimate partner violence (IPV), relatively few critiques have used extensive empirical data gathered specifically for this purpose. This research analyzed these types of data to investigate an important mechanism of potential validity problems, IPV misreporting, by adding context questions to determine whether participants endorsed (as if actual IPV) accidents or other acts that neither party took seriously. The objective was to determine not only the extent to which this form of ...
View more >Objective: Although many scholars have questioned, on a logical basis, the validity of the Conflict Tactics Scale (CTS) to adequately measure intimate partner violence (IPV), relatively few critiques have used extensive empirical data gathered specifically for this purpose. This research analyzed these types of data to investigate an important mechanism of potential validity problems, IPV misreporting, by adding context questions to determine whether participants endorsed (as if actual IPV) accidents or other acts that neither party took seriously. The objective was to determine not only the extent to which this form of overreporting occurs but also how males and females differed in misreporting patterns. Method: Students from 1 U.S. and 1 Australian university (Total N = 1,758) completed a computer-administered survey. Multilevel logistic regression subsequently assessed the degree to which several factors predicted whether participants overreported CTS items. Results: Of the 1,174 event endorsements, 22.1% were classified as overreports. Whether males or females were more prone to overreporting, however, differed across event type, sample, age, relationship status, perpetration versus victimization, and current versus former partnerships. There were statistically significant interactions between gender and many of these factors. Among the most important of the findings was that males were more likely to overreport victimizations by female partners, whereas females were more likely to overreport perpetrations against male partners. Conclusions: The magnitude and intricate gendered nature of the overreporting problem imply that overreporting is a substantial problem, having the potential to negatively affect scale validity and thus the testing of partner-violence theories.
View less >
View more >Objective: Although many scholars have questioned, on a logical basis, the validity of the Conflict Tactics Scale (CTS) to adequately measure intimate partner violence (IPV), relatively few critiques have used extensive empirical data gathered specifically for this purpose. This research analyzed these types of data to investigate an important mechanism of potential validity problems, IPV misreporting, by adding context questions to determine whether participants endorsed (as if actual IPV) accidents or other acts that neither party took seriously. The objective was to determine not only the extent to which this form of overreporting occurs but also how males and females differed in misreporting patterns. Method: Students from 1 U.S. and 1 Australian university (Total N = 1,758) completed a computer-administered survey. Multilevel logistic regression subsequently assessed the degree to which several factors predicted whether participants overreported CTS items. Results: Of the 1,174 event endorsements, 22.1% were classified as overreports. Whether males or females were more prone to overreporting, however, differed across event type, sample, age, relationship status, perpetration versus victimization, and current versus former partnerships. There were statistically significant interactions between gender and many of these factors. Among the most important of the findings was that males were more likely to overreport victimizations by female partners, whereas females were more likely to overreport perpetrations against male partners. Conclusions: The magnitude and intricate gendered nature of the overreporting problem imply that overreporting is a substantial problem, having the potential to negatively affect scale validity and thus the testing of partner-violence theories.
View less >
Journal Title
Psychology of Voilence
Volume
8
Issue
2
Copyright Statement
© 2018 American Psycological Association. This article may not exactly replicate the final version published in the APA journal. It is not the copy of record. Reproduced here in accordance with publisher policy. Please refer to the journal link for access to the definitive, published version.
Subject
Health services and systems
Public health
Psychology
Social Sciences
Psychology, Clinical
Criminology & Penology
Family Studies
Psychology