Introduction
The article ‘knowledge and attitudes of nurses regarding domestic violence and their effect on the identification of battered women’ is authored by Merav Ben Natan and Idit Rais and published in the 2010 publication of Journal of Trauma Nursing, volume 17 issue 2 from page 112 to 117.
Natan and Rais (2010) address question of identifying battered women by examining how knowledgeable nurses are in the issue and their attitudes towards women who present with cases of domestic violence.
The authors conducted their study among 100 nurses in Israel and they concluded that most nurses have knowledge about domestic violence and the need to screen for such cases but unfortunately this is not always reflected in practice.
Ethics
The article by Natan and Rais (2010) is a quantitative article based on its research design, including the aspect of collecting first-hand data and use of statistical procedures (Pearson and Spearman correlation) in analysis of the data before making various conclusions.
The article is described as descriptive quantitative since the data that is collected is more descriptive than quantitative, whether demographic data or data on attitude and knowledge.
In addition, it is clear that the study seeks to establish existence of relationships between independent variables (“nurse knowledge, department routines, and attitudes, Natan & Rais 2010, p. 112”) and the dependent variable (identifying battered women) and quantifying the same.
The conduction of this research was upon approval by an institutional review board (IRB) as indicated by the authors that questionnaires were only administered upon a go ahead from an institutional ethics committee.
In conducting this research, the authors sought the consent of the prospective participants (nurses) where the purpose of the study was explained to participants and confidentiality of information to be collected was reassured. Moreover, the nurses were left free to make a decision of not participating or even dropping off from the study if they deemed good not to continue with the study.
It is possible, although not stated, that some nurses may have questioned a lot regarding what the information gathered from them was intended for with fears that it may be used against them. However the researchers were able to convince the nurses of confidentiality and purpose of the study.
Beneficence is all about doing something with the aim of benefiting others. This principle is clearly demonstrated in this study with the authors focusing on how nurses handle battered women and even their awareness on how to handle battered women, all in the aim of helping nurses to address the problems of battered women more effectively.
There is no doubt that the biggest gainers in this study are women (victims of domestic violence) since nurses would utilize the results of this study in equipping themselves better in screening and attending to battered women more effectively.
Justice in medical ethics entails being fair and making sure that every human being remains dignified and his/her rights respected. This is best demonstrated by attending to the needs of those who are more vulnerable with the same magnitude as those who are less vulnerable would be handled.
Natan and Rais (2010) essentially address the woes of a generally vulnerable group, battered women. It is evident that women hardly report domestic violence voluntarily but they do so if a nurse assesses them. This makes them more vulnerable to continued battering and its consequences when it is no addressed.
This study however addresses this issue by bringing into light the competence of nurses in assessing and handling battered women. The findings of this study can be used to put more effective systems of addressing cases of battered women, which is essentially an act of justice.
The principle of ‘respect for persons’ appears to be adhered to in this study as there is no instance of coercion either at commencement or in the course of the study. Moreover, the fact that the researchers sought a go ahead from an IRB is evidence that there was willingness to adhere to respect for informed decision making from the participants, which is central in research dealing with human subjects.
Overall, it is clear that there are more benefits in this study than any potential risk. In fact there are almost no risks since the study is merely seeking opinions and attitudes and not taking biological samples from the subjects.
The researchers can be said to respect participant’s autonomy and they adhere to the principle of non-maleficence as no harm is seen to be inflicted on the participants. The study is also lawful and in fact it addresses issues that are in line with the Act for Preventing Domestic Violence in Israel.
Research Problem
The research problem in this study is that it is impossible to tell how widespread cases of domestic violence are in Israel due to difficulty in identifying battered women. That not withstanding, nurses have been identified as among the best avenues for identifying battered women since such women tend to be more open if a nurse assesses them.
It therefore becomes convincing to assess how knowledgeable nurses are in the area of assessing and helping battered as well as the attitudes of the nurses on the same. I believe that this conducting this study would shed light into the areas where nurses need to be equipped in positive identification of battered women and subsequently addressing their needs appropriately.
The hypothesis is not directional since the authors do not seem to speculate that the number of domestic violence cases is high or low due to problems in identifying battered women. Instead, the authors state that the “the prevalence is unknown.”
Perhaps the hypothesis is more of causal-directional than directional i.e. the unknown prevalence of domestic violence in Israel is hypothetically attributed to hardships in identifying battered women.
The independent variables in this research include nurse knowledge i.e. training on violence, departmental routines, nurse attitudes towards battered women and intentions and actual behavior of nurses whereas the dependent variable is ‘identifying battered women’ since it is affected by nurse knowledge, departmental norms as well as attitudes and intentions/behaviors of nurses.
One of the extraneous variables in this study is the experience (more so in years of practice) of nurses since this may affect attitudes of the nurses towards battered women as well as their ability to identify a battered woman.
The definition of attitudes of nurses towards battered women is a good example of conceptual definition with an operational definition. These are not measurable in the study but they are described and their relationship with the dependent variable established.
Review of Literature
The authors of this study reviewed 17 articles as part of their literature review. It is unfortunate that a good majority of the literature was not recent since most (10 references) were more than five years old. In fact some references (7 references) were dated as old as 1990s with only a few articles being published in or later than 2005 (7 references). Apparently, 17 references can be considered few for a research worth publication.
However, it is evident that there is paucity of literature on this subject and in fact the available literature is somewhat outdated. It is therefore reasonable to have 17 references for this research. Some of the primary sources included in this study are the study by Moore, Zoccaro and Parsons (1998), Kim and Motsei (2002) and the study by Chrisofides and Silo (2005).
Secondary sources include Nelson, Nygren, Mclnerey and Klein (2004), The Act of Preventing Domestic Violence (1991) and the study by Ben David L (2008).
It is clear that the populations and types of studies in the studies are compatible with this study since they cover nurses and social workers knowledge in assessing domestic violence, barriers in screening for domestic violence as well as perceptions of nurses regarding patients who have been battered among other relevant issues. None of the sources provided an opposing idea but instead all of them were in support of the study.
A commendable thing is that the review of literature flowed well such that the gap in the issue of problems in assessing domestic violence (battered women) is brought out well and the need to conduct the study justified.
Framework
The theory/framework of this study is based on nursing theory whereby a client-nurse relationship is deemed as important in effective nurse practice. As such, ability of nurses to assess battered women is a step forward in addressing the problems of this group of patients. This therefore makes the conceptual model very relevant not only to nursing but also to nursing research.
As earlier noted, the concepts of the model are making a solid nurse-client relationship as a way of establishing a therapeutic relationship. In addition, the aspect of empowering nurses as a tool to positive diagnosis and treatment is emphasized.
The provided theory clearly explains the phenomenon as eventually it becomes possible to acknowledge the gap in knowledge and need for research in nurses’ knowledge on identifying and handling domestic violence cases.
Design
This study followed a descriptive quantitative research design. This is because the study seeks to identify relationships between the dependent variable and the independent variables with no need for the direction of causality. Moreover, the participants (nurses) are only assessed once. The sample is also quite large (120), although not very large, and this is characteristic of a descriptive research.
These characteristics prove important in answering the research question since first-hand information on how nurses perceive and handle the problem of battered women and relationships are identified. The main threat to internal validity is nurses’ experience in nursing practice and assessment of battered women since this is an extraneous variable.
Non-random sampling on the other hand introduces bias and compromises external validity. The fact that participants were aware of what the researchers were looking for in the study may have influenced their reaction/responses thus compromising external validity. The authors’ decision to use a non-parametric statistical analysis was an effort to deal with non-normally distributed data due to non-random sampling.
Sampling
Natan and Rais (2010) used convenience sampling which is a non-random sampling technique as evidenced by issuance of questionnaires from a predetermined group of nurses from selected hospitals and communities. It would have been more appropriate to use random sampling as it would have eliminated bias and strengthened internal validity.
Sample characteristics are well described and such characteristics include gender, age, education level, and ethnicity/nationality. This generalizability of this study is limited due to use of a non-random sampling method and a relatively small sample, considering this is a descriptive study which requires large samples.
No attrition, retention or strategies are mentioned in the sampling process while recruitment was done for central Israel community and hospital nurses only.
Data collection
The instrument used for this study is clearly stated as questionnaires but the rationale for the selection of questionnaires is not provided. The data collection method is however appropriate for this study since questionnaires ere effective in obtaining perceptions, attitudes and beliefs of participants concerning a particular issue.
All the data collection procedures were similar for all subjects and both closed and open-ended questions were used with interval scales (Likert scale of strongly disagree to strongly agree), nominal scale (Yes/No questions) as well as ratio scales (“number of violence incidences”) being used.
Control
To test the validity of the content, experts (nurses) in the area of domestic violence were utilized whereas reliability was tested using Cronbach alpha during the pilot study. The Chronbach’s alpha was reported as.85 which is a good reliability value since it was above the recommended value of.80. These were the best options of testing validity (best done by experts in the field) and reliability.
The authors did not mention the strengths and weakness of the instrument reliability in the “limitations” section. The validity of the questionnaires is however questioned as exemplified by inability to differentiate attitudes from beliefs among participants. The questionnaires may have had a weakness of having leading questions, especially the close-ended questions thus compromising reliability.
Statistical analysis
Natan and Rais (2010) have used descriptive analysis of data, Pearson correlation as well as Spearman correlation tests. While Pearson correlation is a parametric test, Spearman correlation is the non-parametric counterpart of Pearson correlation. The authors used these two tests since they are very helpful in establishing relationships between variables, which is central to this study.
Moreover, a combination of a parametric and a non-parametric test was just in case the data was not normally distributed due to non-random sampling. The null hypothesis for this study may have been that “there is no relationship between problems in identifying battered women and nurses’ knowledge, departmental routines and attitudes on the identification of battered women.”
The null hypothesis is however rejection since significant relationships between the independent variables and the dependent variables were established. The fact that the correlations (rp or rs) were always found to have a p value less than.05 is a clear evidence of statistical significance of the relationships.
For instance, there is a statistically significant relationship between the nurses’ having an intention to ask the patient about domestic violence and the actual questioning, rs =.42, p <.01.
Reference
Natan, M. B. and Rais, I. (2010). Knowledge and attitudes of nurses regarding domestic violence and their effect on the identification of battered women. Journal of Trauma Nursing, 17(2): 112-117.