Essay: The Importance of Research for Social Work: Critique of Corcoran (2006)



A Critique of Corcoran (2006):

The Purpose, Methodology, and Implications for Social Work Practice

Part I

Importance of Research for Social Work

Before the critique of Corcoran (2006) is presented, it is important to address the role research plays in social work and why it is important for social work practice. Without research, social work practice would not exist. Research informs social justice efforts, mental health care, theory, policy reform, and many other areas of social work (Faulkner & Faulkner, 2016). For example, cognitive behavioral therapy (CBT) is commonly used to treat a large variety of mental health disorders (e.g., depression, anxiety, eating disorders, and PTSD). The field of social work has adopted CBT as one of the most commonly used treatments because it has been empirically validated to be one of the most effective forms of treatment for a variety of populations (Early & Grady, 2017).

In addition to informing evidence-based practice, research can be used to enact social policy change within different organizations. Specifically, research can be used to evaluate the effectiveness of a pilot educational program within a school setting. For instance, suicide is on the rise in adolescent populations (Kim, Moon, & Kim, 2011). Some preliminary research suggests that school-based suicide prevention interventions can help reduce suicide rates (Schilling, Lawless, Buchanan, & Aseltine, 2014). By being familiar with these studies, social workers can learn how to advocate for the addition of suicide prevention education in grade schools and use these studies to support their claims.

Social workers must understand the underlying concepts and practices in research to comprehend how it informs practice. It is especially important that social workers understand the different research designs that exist, which are exploratory, descriptive, explanatory, and evaluative designs (Faulkner & Faulkner, 2016). Also, they must understand the methodologies that were used in these studies and know if they are reliable and valid to determine the generalizability of the study. For example, if a clinical social worker wants to learn about a treatment that is effective for reducing anxiety in adolescents, then they should look up evaluative research articles that test the effectiveness of evidence-based treatments for reducing anxiety with this population. Also, the social worker must understand the reliability and validity of the measures the researchers used to determine if the approach in question would be effective in treating adolescents. It would not be appropriate to utilize a treatment if the measures had low reliability or validity because that treatment might not produce the same results in a similar population.

Personal Reflection on Importance of Research for Clinical Practice

Once I graduate with my Master of Social Work degree, I plan to become a Licensed Clinical Social Worker (LCSW). My plan is to work in a mental health setting with adolescents and adults, where I will work to reduce depression, anxiety, and stress. Specifically, I hope to specialize in suicide prevention and offer support to individuals that have been affected by suicide. I will use research to understand the different types of suicide assessments that are available and how to use them in practice. When researching these assessments, I will pay close attention to the reliability and validity scores for the target population that I intend to work with.

In addition to being familiar with different suicide assessments, I plan to stay up to date with the different prevention strategies that are available for treating suicidality. If a new treatment is being developed, I will follow those studies to determine if this treatment would be effective or not. Also, this would be useful for determining what types of treatments would be effective for individuals that are bereaved from suicide. I am not familiar with the treatments that exist for these populations, so I will rely on staying current with research to educate myself about these strategies.

While research will be beneficial for me on the micro level, I will also use it in my effort to be involved at the macro level. In addition to treating suicidality, I am interested in advocating for the implementation of suicide prevention programs in middle and high schools across the nation. I plan to research different programs that exist as an effort to prepare myself for advocating at the macro level. Also, I hope to stay involved with research once I graduate. I will work with researchers from a variety of disciplines that are interested in suicide prevention efforts and contribute to new studies to advance prevention programs. If we can continue to evaluate current or new prevention efforts, then we can work towards reducing suicide rates. Overall, I believe that research is central to social work practice. Every social worker must understand the different aspects of research to accurately stay informed and practice in the field.

Part II

The remainder of this paper will focus on a discussion of the article “A comparison group study of solution-focused therapy versus ‘treatment-as-usual’ for behavior problems in children” by Dr. Jacqueline Corcoran (2006). Within this article, Corcoran (2006) presents a study they conducted that was intended to evaluate the effectiveness of using solution-focused therapy versus a “treatment-as-usual” approach (e.g., cognitive behavioral techniques) with a group of children that have behavior problems. Trained student researchers either conducted solution-focused therapy with 139 families or cognitive-behavioral techniques with 100 families. The researchers found that there was no difference in treatment outcomes between the solution-focused group and the “treatment-as-usual” group.

Introduction

This study utilized a quasi-experimental design, which is a design that includes a comparison group, but does not use random assignment (Faulkner & Faulkner, 2016). The clinical outcomes from both the solution-focused and “treatment-as-usual” group were compared with each other. Each family that participated in this study was administered a pre-test and a post-test. However, Corcoran did not use random assignment in this study, though it would have been ideal. Random assignment was not possible due to the probability that student interns would become confused if they learned two different therapies. Therefore, the solution-focused intervention was administered before the “treatment-as-usual” condition.

The research problem was relatively easy to identify in the introduction section. Corcoran provided an organized argument for why they attempted to conduct this study. This study was intended to address the existing gaps in research that evaluates the effectiveness of solution-focused therapy with children. Specifically, Corcoran identified several problems that exist in prior research. Some of these issues include a lack of utilizing pre-tests as a baseline measure, lack of inclusion of standardized measures, smaller sample sizes, including a comparison group, and identifying predictors of high dropout rates in previous studies. These components would be addressed by Corcoran as an effort to determine if solution-focused therapy would be more effective than “treatment-as-usual” for treating children with behavior issues. Also, the author intended to address predictors of dropout within these treatments.

Corcoran identified several objectives and hypotheses that are central to this study. The first objective was to evaluate the effectiveness of using solution-focused therapy with a group of children that had behavior problems. A comparison group (i.e., “treatment-as-usual”) was included within this study to test the effect size of the solution-focused outcomes. As an effort to address previous concerns in the literature, Corcoran included a pre-test as a baseline measure, used standardized self-report measures, and employed a larger sample of families for this study. The second objective that the author mentioned was to examine the dropout rates of the families in both treatment conditions. This question was included to address the gap in the literature that fails to address the high dropout rates of families in solution-focused treatments. The final objective that Corcoran identified was to examine these dropout predictors and test if the factors identified in the previous literature would be related to dropout within the sample in question.

There are four hypotheses that were tested within this study to address the objectives. Corcoran first predicted that children who received the solution-focused treatment would have improved behaviors compared to the control group. The second hypothesis anticipated that the solution-focused group would have lower dropout rates compared to the “treatment-as-usual” group. The final two hypotheses predicted that single, minority race parents with a low income would be more likely to drop out of the treatment. Also, if the child had a more severe behavioral problem or was of an older age, they would be more likely to drop out of treatment.

Method

Corcoran recruited 239 families who had children with behavior problems. These problems were specifically identified by the author (e.g., aggression, defiance, and conduct problems in school). The children and parents were recruited in this study with a convenience and purposive sampling technique. Corcoran selected families with children with behavior problems that were referred to the clinic by the neighboring school district. The sample had a diverse range of ages, including children between age 5 and 17. However, the sample was not racially diverse, which could limit the generalizability of the findings. The majority of the children in the sample identified as white (n= 108; 78%), while 11 participants identified as African American (n= 11; 8%), 12 identified as Hispanic (n=12; 9%), and 8 identified as another race (n= 8; 6%).

This nonprobability sampling method was appropriate for the nature of this study. It was convenient to sample from the neighboring school district and the referrals allowed Corcoran to recruit the target population with ease. However, obtaining a sample from one location can reduce the diversity characteristics of the sample. Corcoran presented a demographics table that highlights the race, age, income, marital status, and other characteristics of the sample. The majority of this sample identified as white and most of the parents reported that they made less than 40,000 dollars a year. This sample could be more diverse if participants were recruited from surrounding towns and cities. Also, the final sample consisted of 85 participants because the majority dropped out of treatment. It would be more ideal to obtain a larger sample, like the one in the beginning, to reduce sampling error. While a sample size of 85 participants is decent for this study, the findings are less generalizable because less of the population is represented.

Corcoran provided detailed descriptions of the various methods and measures that were employed in this study. Before the families received the treatment, they completed a pre-test. This consisted of two self-report measures: the Conners’ Parent Rating Scale and the Feelings, Attitudes, and Behaviors Scale for Children (FAB-C). Corcoran included citations of articles that have validated these measures. Also, the purpose of each scale along with how many items are included in the scales were discussed. Once these questionnaires were administered, trained graduate students delivered solution-focused therapy to the families. A control group then received cognitive-behavioral interventions once data was collected from the solution-focused group. After both groups received treatment, they completed a post-test that consisted of the Conners’ Parent Rating Scale and the FAB-C.

Reliability and Validity Considerations

While this study presented great and appropriate methodology, some of the measures had issues with reliability. The FAB-C has reliability values ranging from .61 to .78. While these reliability values are somewhat strong, the Conners’ Parent Rating Scale was reported to have a reliability score that ranges between .46 to .57. These values are average, meaning that this scale is not incredibly reliable. While the author included this scale because they deemed it as an appropriate measure of parents’ ratings of child behavior, they could have searched for other scales that are more reliable in measuring this. The low reliability of the Parent Rating Scale could be one of the reasons why some of the hypotheses were not supported in this study.

Several issues of internal and external validity were present within study. In terms of internal validity, the threat of selection was present in this study because participants were not randomly assigned to comparison groups. Also, the testing threat was present because participants completed a pre-test and post-test. The participants’ post-test scores could have been influenced by their previous experience with the pre-test. Attrition is also a huge threat to internal validity in this study. Corcoran reported that 154 families dropped out of the study. This threat was reduced, however, through the inclusion of a pre-test. Corcoran included this to test their hypothesis about the factors that would predict dropout rates and characteristics.

In addition to internal validity threats, external threats are also present. Corcoran had to obtain consent from the parents and assent for the children, so the Hawthorne Effect was present in this study. Also, the researchers were faced with the interaction of selection and treatment threat because an initial selection threat was present. An initial testing threat to internal validity was also present, so the reactive effects of testing threat is present. All of these threats could result in the generalizability of the findings being reduced.

While several threats were present within this study, there are additional threats to reliability and validity that could be present, but were not reported. Corcoran reported that the study took place between 1997 and 1999. A threat to contemporary history could have also been present, but the location of the study was not reported. In addition to this, intrasession history threats could have occurred. For example, some of the participants might have witnessed a shooting or a robbery. In terms of performance, Corcoran did not report if any participants had an outlying score on the pre-test or post-test. It is very possible that several participants could have been outliers. Also, the students that administered therapy in the “treatment-as-usual” group were not audio or video recorded. Corcoran explained that these sessions were not recorded because the cognitive-behavioral techniques were more eclectic than solution-focused interventions. As a result, a threat to interrater reliability is present. Each supervisor was not able to equally observe the sessions with each participant, so there could be differences in treatment that are unknown to the audience.

Results and discussion

Overall, Corcoran found that participants who received and completed the solution-focused intervention had no differences in behavior than the children in the “treatment-as-usual” group. Also, the differences in dropout rates were difficult to explain due to the children primarily being referred for school-related problems, not problems at home. Corcoran also presents several limitations that were present in this study, including the lack of school observations, lack of inclusion of multiple questionnaires, unexperienced student therapists administered the treatment, similarity between treatments, and a lack of long-term follow-up procedures. The conclusions and limitations that Corcoran presented are consistent with the findings. For example, if a licensed professional that has years of experience in solution-focused and cognitive-behavioral techniques administered these treatments, then the participants could have had different outcomes. Different questions would have been asked that could have led to different outcomes.

These results should be carefully considered before generalizing them to different populations. As mentioned earlier, the final sample size was relatively small in comparison to the sample at the beginning of the study. Sampling error and a lack of diversity are an issue with this sample. In addition to sample size, several issues of internal and external validity are present within this study. With the high attrition rate, these results must be taken with caution and future studies should replicate these methods with a large sample. Also, future research should carefully consider the recommendations that Corcoran identified for designing future studies.

There are additional limitations that exist in the study. Corcoran only included two measurements of behavior ratings in this study, which are self-report. Also, these measures had lower reliability scores that can limit the generalizability of these findings. The author should have considered researching additional measures that have higher reliability and can be observed in a session to eliminate parental biases. Also, Corcoran reported that the field supervisors in this study tended to prefer cognitive-behavioral approaches over solution-focused treatments. They were also advised to not teach solution-focused approaches to the students. The supervisors could have been biased when they were critiquing the students in the solution-focused treatment.

Even though this study has several limitations, the strengths should also be considered. Corcoran had a great method for recruiting participants with behavior problems. It can be difficult to find these participants through random sampling and it is convenient to recruit from a school, where conduct and aggressive behaviors can be prevalent. Also, Corcoran explained that each student that administered the treatments were closely supervised, especially within the solution-focused group. Students in this group were audio and videotaped, and their sessions were watched by the supervisor before a one-hour session with the student. This is one of the most effective ways that students can learn how to administer a treatment.

Despite the hypotheses not being supported in this study, Corcoran highlighted important considerations that are important for social work practitioners and researchers. For instance, dropout rates were higher in the “treatment-as-usual” group compared to the solution-focused treatment group. These findings could suggest that there is something unique about solution-focused treatment for behavior problems compared to cognitive-behavioral. Perhaps practitioners that are using a cognitive-behavioral approach with a family experiencing these issues can consider incorporating solution-focused techniques in their practice. This could increase the likelihood that the family will stay involved in therapy.

In addition to addressing the dropout rates, Corcoran highlighted that there was some observable change in both treatment groups within the four to six sessions they did complete. This finding could be useful for social workers that are working with children with behavior problems. If they are only observing small changes at first, they could be informed from this study that longer lengths of treatment could promote positive change in the client. They should continue trying to use a solution-focused approach for the next few sessions and then if they are not observing change, they can try another form of treatment. Overall, this study has important implications for social work and future research should continue to explore this area to reduce the gap in the literature.

...(download the rest of the essay above)