Describe the study’s research design.
The study’s design was a relatively simple one, and the approach was experimental. During the study, officers of the Minneapolis police department were asked a series of questions regarding handling domestic violence calls (Sherman & Berk, 1984). The officers in the study were asked about three different ways to handle domestic violence calls, and were then assigned to three different manners of dealing with these calls (Sherman & Berk, 1984). Bachmann & Schutt (2013) suggest that engaging in follow-up interviews can be beneficial for both qualitative and quantitative studies. The study lasted for seventeen months (Sherman & Berk, 1984). During this time, there was a follow-up period for each case of approximately six months. This allowed the researchers to investigate what became of each case, and to report the outcome of their action.
The first group of the fifty-one involved officers were supposed to send the abuser away from his or her victim for a total of eight hours, effectively removing the abuser from the area (Sherman & Berk, 1984). The second group did not intervene and remove the abuser, but instead they acted as mediators and intervened in the dispute, offering advice (Sherman & Berk, 1984). The third method for handling the issue, for the final group, was to arrest the perpetrator (Sherman & Berk, 1984). During the course of the investigation, more than 330 cases were followed by the researchers.
Was there a specific theory or hypothesis being tested? If so, state what it was.
There was a very specific theory that was being tested by the researchers in the Minneapolis domestic violence experiment (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). This experiment’s primary purpose was to determine whether the most effective response to incidents of domestic violence was mediation, removal, or arrest. Each of these reactions led to different outcomes for the cases involved in the study. The report established a baseline for understanding which approach would be the most effective at deterring future incidents of domestic violence for individuals in abusive and violence situations (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). The research only looked at police action in misdemeanor assault cases, and even then, only at cases that fit certain criteria. The hypothesis compares action against consequence in criminal justice (Bachmann & Schutt, 2013).
The researchers also hypothesized that there would be something in the victim’s history which led them to being victimized. This is why the researchers established a follow-up interview with female staff during the course of the experiment (Sherman & Berk, 1984; Bachmann & Schutt, 2013). This allowed the staff to ask questions regarding the individual’s history and background, and determine if their history and background are related to the victimization history of the individual (Sherman & Berk, 1984).
What were the important variables of the study and how were these variables operationalized?
There were a number of very important variables in ths study. First and foremost, the study looked at four separate variables in police response. In some situations, officers would send the accused away from the scene for eight hours (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). This was the control situation to some extent: by sending the accused away, the researchers were investigating whether the time spent apart was significant in the reduction of domestic violence recidivism (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). In the second case, the officers would try to discuss the situation with both the parties involved. The officer acted as a mediator, and this mediation is the second variable. In the third situation, the officer would arrest the accused—arrest is the final variable in the experimental design.
The variables were worked into the hypothesis as specific actions that could be taken by the officers involved. The officers involved were allowed to use their judgement in each situation, so the officer’s judgment is one variable that is not controlled—every officer might judge a scene differently, after all (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). However, it can be assumed that a majority of officers would make a certain decision faced wiith a specific situation most of the time.
Identify the independent and dependent variables, and explain why they are designed this way.
The dependent variables are the outcomes for each case—recidivism rates for the couples or families in the study, more specifically (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). The researchers needed to discover a way to measure outcome for families who experienced domestic violence, and measuring the rate of recidivism is one of the primary ways to do so (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). In addition, the researchers hoped to address the complicated questions associated with domestic violence, and provide quantitative information regarding outcomes for victims.
The independent variables are the decisions that the police make: these interventions might be mediation, removal, or arrest. The researchers and the outcomes have no effect on these independent variables, and they are specifically designed to change the results of the dependent variable (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). Independent variables also include demographic information of those involved, including age, race, and social class. By and large, these independent variables are not considered by the researchers as a whole; all the cases are treated equally by the quantitative research, although a qualitative follow-up was also completed.
Critique the ethics associated with this research project.
Unfortunately, this experiment has a number of questionably-ethical structures underlying the experiment as a whole. First and foremosst, the individual participants in the study were not ever informed—before they began their interaction with the police officer—that they would be participating in a study (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). Informed consent is a fundamental aspect of ethical research, and these individuals, many of whom were in very dangerous situations, were not informed of their participation in the study until much later (Sherman & Berk, 1984). The lack of informed consent is one of the most concerning aspects of the study as a whole, and could lead to significant questions regarding validity, and lack of consent must be consideres (Bachmann & Schutt, 2013).
The study also potentially affected the outcome for these victims. Depending on the way the officer reacted in a particular situation, the victim’s potential outcome might have been better (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). If a researcher’s action, even by policeman proxy, leads to the harm of another individual, the action is undoubtedly unethical. While there is no real evidence of this harm, it is entirely possible and perhaps even probable, given the statistics available regarding the reoccurance of domestic violence (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992).
Were there major threats to validity in this study?
The most significant threat to the validity of this study is the ethical concern regarding whether the participants were ethicaly required to give consent to be part of the study (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). Because the researchers required specific actions on the part of the police based on certain scenarios, people were undoubtedly affected by this study, and possibly in a negative manner. This kind of ethical questionability often leads to questions of validity, although not always. In addition, the replicability of the study is in question (Sherman, Schmidt, & Rogan, 1992). Other researchers have tried to re-create the study to no avail; some have even expanded the study and controlled for demographic factors and had very different results. This is, of course, a very concerning thing regarding the potential validity of the study. The lack of concern regarding social and demographic factors leads to questions about the validity of the study as well, as demographic factors tend to play a very specific role in the sociological strata of a community and an individual’s experience within that community. Ethics must be heavily considered when designing any experiment in criminal justice, as the long-term implications for participants can be significant (Bachmann & Schutt, 2013).
What conclusions can be drawn from this experiment?
If the results are to be believed—and the validity of the experiment is certainly in question—there are a number of potential conclusions that can be drawn from this particular experiment. First and foremost, the conclusion that the researchers drew was that the removal of the abusive individual to jail was the most effective way to respond to situations of domestic violence (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). The study suggests that when an individual is arrested, he or she is much less likely to re-offend against their partner with more domestic violence or domestic violence at a later date (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992).
However, the research also found that this difference accounted only in the case of relatively minor incidents of domestic violence, not in the serious, life-threatening incidents of domestic violence (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). These life-threatening, serious incidents are not discussed by the researchers, and there are few ways to ethically design an experiment that can functionally test the hypothesis that arrest is the best solution in these cases (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). The researchers also conclude that further research is needed to determine when arrest should be necessary and when it should not be.
What are some of the policy implications for this experiment?
The most important policy implication for this experiment is, of course, the evolution of behavior in police departments regarding the way domestic violence calls are handled (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). In the past, incidents of domestic violence were often passed off as something that was between the partners; many police officers were hesitant to get involved in these kinds of family disputes (Sherman & Berk, 1984). However, this study demonstrates that this policy is not only ineffective, it has the potential to be actively harmful to those in the abusive household (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992). This undoubtedly had long-term impacts for criminal justice systems around the country, as today, arrest is the primary response to incidents of domestic violence (Bachmann & Schutt, 2013).
Indeed, there was a significant response in terms of policy shift from this study. Many police departments began to require arrests when there was probable cause for abuse in the household, and today, this has become one of the most common policies for police departments in domestic violence cases (Sherman, Schmidt, & Rogan, 1992). In addition, this shifted the policy of questioning and blame from the victim to the perpetator. There are still problems regarding domestic violence and the handling of these cases in the United States, but the situation is not exacerbated nearly as badly by law enforcement today (Sherman & Berk, 1984; Sherman, Schmidt, & Rogan, 1992).
References
Bachman, R., & Schutt, R. K. (2013). The practice of research in criminology and criminal justice. Sage.
Sherman, W., & Berk, R. A. (1984). The Minneapolis domestic violence experiment.
Sherman, L. W., Schmidt, J. D., & Rogan, D. P. (1992). Policing domestic violence: Experiments and dilemmas. Free Press.
