Skip to main content



Child-Maltreatment-Research-L (CMRL) List Serve

Database of Past CMRL Messages

Welcome to the database of past Child-Maltreatment-Research-L (CMRL) list serve messages. The table below contains all past CMRL messages (text only, no attachments) from Nov. 20, 1996 - March 6, 2018 and is updated quarterly.

Instructions: Postings are listed for browsing with the newest messages first. Click on the linked ID number to see a message. You can search the author, subject, message ID, and message content fields by entering your criteria into this search box:

Message ID: 8064
Date: 2009-03-04

Author:Chaffin, Mark J. (HSC)

Subject:perceptions of child abuse

An interesting survey study was cited in which the public was asked to rate whether a behavior was "abusive" on a 1-5 scale. People rated abusiveness at an average of 2.5 when the parent's behavior was in response to public child misbehavior, as compared to 3.0 when it was not. I think how you might interpret this sort of difference depends among other things on what the midpoint (i.e. 3) of the 1-5 scale was. If it's a neutral midpoint (e.g. "neither abusive nor non-abusive" or something like that), then you have to be careful to interpret with an eye toward what questionnaire researchers call the midpoint issue. That is, the item may not reflect one dimension or construct, but two. For example, on the classic strongly disagree to strongly agree 5-point scale, a score of 2.5 vs. 2.0 does not reflect "more agreement," but "weaker disagreement." It sounds semantic, but it's not. Someone who disagrees is different from someone who agrees, and disagreement is not just less agreement and vice versa. Same with perceptions of abusive vs. non-abusive parenting. One classic example of the midpoint issue in the child abuse research field was a survey of college males done long ago that asked something along the lines of "if you knew you could get away with it, would you commit rape?" on a strongly disagree-strongly agree scale. The finding that people over the years cited from the study was that a substantial number of college males indicated "some willingness" to commit rape. Willingness here was defined as any response other than "strongly disagree." The problem, of course, was that virtually everybody disagreed with the idea of committing rape, just to different strengths of disagreement....which is not the same thing as saying that you have "some willingness."



Similarly, rating a behavior as neutral for abusiveness in one context (3.0) and not abusive in another (2.5) is not necessarily the same thing as saying that an abusive behavior was viewed as less abusive depending on context. Many people may have felt the behavior was non-abusive in both contexts, but felt slightly more confident in their assessment of non-abusiveness when the context was public child misbehavior. It might be informative to check the pattern of responses on the items, not just the mean scores, and if possible to do within-subjects comparisons in order to clarify this. Comparing mean scores makes assumptions that do not account for the issues described above, especially when standard statistical procedures are employed. An interesting and potentially more informative look at responses to the item might ask, "how many people judged the behavior to be abusive in one context, but shifted to thinking it was non-abusive in the other?" My hypothesis is that this number would be quite small.



MC



Mark Chaffin, Ph.D.

Professor of Pediatrics

University of Oklahoma Health Sciences Center

P.O. Box 26901; CSC 225

Oklahoma City, OK 73190

(405) 271-8858





Confidentiality Notice

This email, including any attachments, contains information from [insert name of College/Department/Clinic], which may be confidential or privileged. The information is intended to be for the use of the individual or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this email in error, please notify the sender immediately by a "reply to sender only" message and destroy all electronic and hard copies of the communication, including attachments.









An interesting survey study was cited in which the public was asked to rate whether a behavior was "abusive" on a 1-5 scale. People rated abusiveness at an average of 2.5 when the parent's behavior was in response to public child misbehavior, as compared to 3.0 when it was not. I think how you might interpret this sort of difference depends among other things on what the midpoint (i.e. 3) of the 1-5 scale was. If it's a neutral midpoint (e.g. "neither abusive nor non-abusive" or something like that), then you have to be careful to interpret with an eye toward what questionnaire researchers call the midpoint issue. That is, the item may not reflect one dimension or construct, but two. For example, on the classic strongly disagree to strongly agree 5-point scale, a score of 2.5 vs. 2.0 does not reflect "more agreement," but "weaker disagreement." It sounds semantic, but it's not. Someone who disagrees is different from someone who agrees, and disagreement is not just less agreement and vice versa. Same with perceptions of abusive vs. non-abusive parenting. One classic example of the midpoint issue in the child abuse research field was a survey of college males done long ago that asked something along the lines of "if you knew you could get away with it, would you commit rape?" on a strongly disagree-strongly agree scale. The finding that people over the years cited from the study was that a substantial number of college males indicated "some willingness" to commit rape. Willingness here was defined as any response other than "strongly disagree." The problem, of course, was that virtually everybody disagreed with the idea of committing rape, just to different strengths of disagreement....which is not the same thing as saying that you have "some willingness."



Similarly, rating a behavior as neutral for abusiveness in one context (3.0) and not abusive in another (2.5) is not necessarily the same thing as saying that an abusive behavior was viewed as less abusive depending on context. Many people may have felt the behavior was non-abusive in both contexts, but felt slightly more confident in their assessment of non-abusiveness when the context was public child misbehavior. It might be informative to check the pattern of responses on the items, not just the mean scores, and if possible to do within-subjects comparisons in order to clarify this. Comparing mean scores makes assumptions that do not account for the issues described above, especially when standard statistical procedures are employed. An interesting and potentially more informative look at responses to the item might ask, "how many people judged the behavior to be abusive in one context, but shifted to thinking it was non-abusive in the other?" My hypothesis is that this number would be quite small.



MC



Mark Chaffin, Ph.D.

Professor of Pediatrics

University of Oklahoma Health Sciences Center

P.O. Box 26901; CSC 225

Oklahoma City, OK 73190

(405) 271-8858





Confidentiality Notice

This email, including any attachments, contains information from [insert name of College/Department/Clinic], which may be confidential or privileged. The information is intended to be for the use of the individual or entity named above. If you are not the intended recipient, be aware that any disclosure, copying, distribution or use of the contents of this information is prohibited. If you have received this email in error, please notify the sender immediately by a "reply to sender only" message and destroy all electronic and hard copies of the communication, including attachments.