Assessment of criteria
The Methodology Assessment Scale will be used to conduct a comparative analysis of the evaluation studies in this evaluation research. Explicitness, appropriateness, clarity and completeness of the evaluation have been elicited (Thornton, 2009). The clear and direct presentation of the methodology is explicitness. The suitability of the elements of the methodology determines the appropriateness. The evaluation is described as clear if the presentation is concise and to the point. Completeness is the inclusion of all possible components in the methodology (Thornton, 2009).
The assessment criteria can now be graded into defective, substandard, standard and superior using a score from one to four. The poorest rating would be a score of one when the evaluation methodology is considered defective. A score of four would denote the superior rating. The variable used will be denoted as of good quality if the rating is high. The figure below shows the Methodology Assessment Scale and the scoring key. Variables of paradigm and theory have been identified as usually having defective scores due to being indefinite and inexplicit.
Methodology Assessment Scale
Scoring Key
Methods Matrices
Research problem formulation
Design and Data Collection
Data analysis and Interpretation
Comparative analysis for methodology
Comparative analysis of six evaluation studies using the marital stress therapy for couples has been done. The efficacy of this couples’ therapy in various situations of marital stress has been evaluated. Adoption of these results is possible only if the overall quality of the methodologies is ascertained. The comparison of the variables of methodology is achieved through the research problem formulation, design and data collection, analysis and interpretation. This helps us gauge the significance of the ratings.
Research problem formulation
Variables of research problem formulation are evident in the first matrix. These are derived from the first part of the evaluation study paper. They indicate the main themes of the study and provide a beginning for the reader to understand the problem that is being researched and the reason for it being investigated.
The literature review is the first variable in the matrix. The reader gets an idea of the background information on the relevant topic and the current status on the various researches pertaining to it. All the evaluation studies selected scored high with a superior rating. The literature review in each was dealt with in a thorough and completely relevant manner. The topic was summarized precisely and concisely. Beginning with the history of the problem, the review went on to mention the many researches which clarified the different parts of the topic that came to be proven along the way. The limitations in these studies and the recommendations for future studies were also understood. The changes in research design and hypothesis over the various studies could be gauged and the researcher would be guided into an idea on the design and hypothesis that is adopted in the new research. The six evaluations studies were all equally good and comparable to each other.
The reader understands the purpose for the study from the objective of the evaluation. In this context, four of the studies were rated superior (Greenberg et al, 2010; MacIntosh and Johnson, 2008; Hollist and Miller, 2005) while one was standard (Johnson and Talitman, 1997) and one substandard (Cloutier, 2002). The superior ones clearly outlined the reasons for the studies with specific purpose statements. The standard rated study provided a clear purpose statement but they could have been more specific and so was rated standard. Cloutier’s study was a follow-up of an earlier trial to note the efficacy of EFT in terms of improvement and its statement of purpose was not explicitly written and so was rated substandard. The reader had to gauge from the reading of the whole study about the purpose of the study. The intervention method and its purpose are usually presented in the literature review. However an objective statement adds to the quality of the evaluation study, providing justification for the conduct of the study.
Paradigm and theory are the next variables in the research problem formulation. Philosophical assumptions providing guidance to thinking and action are included in paradigms (Mertens, 2009). The research decision and actions are assisted by the paradigm which forms a framework. Theories on the other hand suggest models for understanding the human behavior, how it occurs and the reason for it occurring in that manner. The paradigms in all the six evaluation studies were found deficient. The authors may have followed major paradigms but this was not sufficiently explained as to how they were used. The information for the reader would have fared better had the philosophical basis been explained more elaborately. The theories were supposed to be derived following thematic analysis after the intervention of EFT and based on its effectiveness. The EFT involved the facilitation of forgiveness with emotional, cognitive and behavioral components for improving the couple relationship after being emotionally distant from their partner following an emotional injury (Greenberg et al, 2010). Better outcomes were expected than if no intervention had been instituted. The attachment theory and the four category model were used in the MacIntosh and Johnson study (2008) for couples where one partner had been traumatised through childhood sexual abuse. The attachment styles that affected the quality of marriage especially in mid-life was also based on the attachment theory (Hollist and Miller, 2005) and EFT could help the involved move from insecure to secure attachments.
The reader gets an idea of the expectations of outcomes of the researcher’s study when he goes through the hypothesis. He also becomes aware of the philosophy and frame of thinking behind it. Greenberg’s (2010) and MacIntosh’s (2008) studies have mentioned hypotheses and maintained a “standard” rating. The latter predicted that the levels of relationship satisfaction would change from pre-test to post-test and survivors would show a reduction in traumatic symptoms between the two tests. Greenberg’s study (2010) aimed at evaluating the effectiveness of EFT in couples when one partner had a history of emotional trauma which could not be forgotten. Three other studies have been rated substandard because hypotheses have not been well-defined or explicitly delineated (Hollister and Miller, 2005, Cloutier, 2002 and Dandeneau, 1994). Johnson and Talitman’s study (1997) has not mentioned any hypothesis and has been rated defective.
The programs of EFT are the independent variables. They have been well-detailed in Greenberg’s (2010) and MacIntosh’s (2008) studies and therefore marked superior. Hollist and Miller (2005) have just followed up participants of earlier interventions of EFT through a questionnaire. Cloutier’s study (2002) also is a follow-up after two years. Both these cannot be considered of good standard and have been rated substandard. Johnson and Talitman, 1997, and Dandeneau (1994) have both delineated the programs well and are marked superior.
The dependent variables are the various values obtained by using standardized scales and observational measures. A number of scales were used for assessing changes that occurred in specific domains. Greenberg (2010) used many valuable instruments for measurements and has been rated superior. MacIntosh (2008) has used some good measures sufficient for his study and has been rated standard. Similarly Johnson and Talitman (1997) have used many scales in a superior manner and have done well in investigating the predictors of success which no researcher had attempted to find till then. Hollist and Miller (2005) and Cloutier (2002) have used only a minimal number of scales for their studies which have minimized the quality of the studies and have been rated substandard. Dandeneau (1994) has been rated standard; the superior evaluations operationalized the dependent variables so that the reader was able to make sense out of the study. Greenberg(2010) has taken pains to explain in detail the various measures used and how the criteria operationalizes the variable. Merely citing the scale and providing the results would not have convinced the reader about the quality of study and outcomes as Greenberg (2010) has done now. Dandeneau also has done a good job. MacIntosh (2008) selected quantitative outcome measures which had psychometric validity, were sensitive clinically and could be compared to other studies of EFT: the Dyadic Adjustment Scale and trauma measures.
Design and Data Collection
The design and data collection constituted the second matrix and contained variables which detailed the research design and implementation. The quality of the research and whether the methodology was flawed in any manner was understood from the analysis of these variables. It is necessary to decide the validity of the outcomes of the research. Design, the first variable, describes the methods which the researcher has adopted to conduct the evaluations. The structure of the program, the staff involved and the method of data collection all come under the design. All the studies had either superior or standard designs. Greenberg used a qualitative methodology to decide the effectiveness of the EFT-for couples. The research strategy is a case study replication in MacIntosh’s study (2008). Hollist and Miller (2005) used a qualitative design with structured questionnaires. The research designs and implementation of the intervention could be aptly explained by the evaluation studies. In four studies there were therapists and in some of these, supervisors were also present and they could have conducted the interventions and collected the data. Two studies did not mention therapists as they were follow-up studies (Hollist and Miller, 2005, Cloutier, 2002) and it was indefinite as to who collected the data or how it was done.
The instruments constitute the next variable. The ones used by Greenberg (2010) and Talitman and Johnson (1997) and Dandeneau (1997) were of superior quality. The wide variety of measures used by Talitman and Johnson, Greenberg and Dandeneau provided a superior quality to the instrumentation. They were of standard quality in MacIntosh’s study (2008). The other two were also of standard quality. The Dyadic Adjustment Scale which was used in all the studies had confirmed reliability and internal validity obtained by Spanier, 1976. Measurement of marital qualities were through a self-report using the Revised Dyadic Adjustment Scale and attachment qualities through the Measure of Attachment Qualities Scale in Hollist and Miller’s study (2005). Cloutier’s study (2002) used the social intimacy scale apart from the dyadic adjustment scale and the reliability and validity of the instrument Dyadic Adjustment Scale has been mentioned giving the rating “standard”. Hollist and Miller (2005) could also be deemed to have used standard measures like the Dyadic Adjustment Scale. The measures were mainly surveys in the form of standardized scales and provided qualitative and quantitative outcomes as required. Some studies quantified the data based on observational measures through the method of coding. Each measure used was described in detail specially referring to internal validity and reliability. The superior rated studies were thorough in this. More confidence is instilled in the reader about the reliability of the measures and the validity of the outcomes or results.
The variable of research sample will now be discussed. The sample is a representation of the larger population and refers to a group of people who are selected for a study due to some common characteristics or features which are to be investigated using planned instrumentations and chosen design. Greenberg (2010) had a small sample of 20 couples who had emotional injuries with their partner, the injuries not having been resolved yet. The partners were more than 18 years old, had cohabited for more than two years and the trauma having occurred more than two years ago. MacIntosh(2008) had 10 couples who had lived together for 1 year with one partner giving the history of child abuse. They had 19 sessions of EFT between the pre and post-treatment sessions. The CSA partners were the females. In general all the studies had a standard quality of participants ranging from 10 couples to 429 married people. All the participants had some form of marital stress ranging from stress due to emotional injury, child sex abuse, attachment style, having chronically ill children or stress in midlife. The criteria for selection of the samples were definite. Only two studies were follow-ups and therefore included the sample of the original studies (Hollister and Miller, 2005; Cloutier, 2002). Though these studies aimed at looking at the long-term benefit of the patients who had undergone EFT couples therapy for some form of marital stress, it was difficult to tell the criteria on which they were selected. The other four studies provided detailed descriptions of the married couples selected including the demographics and were given standard ratings.
Data collection is the next variable in the matrix. The excellence with which data has been collected, the tools adopted and the selection of method of data collection all need to be looked into. All the studies used different methods of data collection using different measures or standardized scales and quantification of observational data. It is noticed that the four studies that had therapists who provided the EFT had the data collection by these trained staff. Details of people who collected data are not given in all the evaluation studies. Three studies have been rated superior due to the good methods of data collection and a fairly detailed description of the collection. Comprehensive data collection was noted in these 3 studies (Greenberg, 2010; MacIntosh, 2008 and Dandeneau, 1994). These studies also enabled the reader to gain confidence. The other three also had standard ratings as they had ample explanation of data collection but they failed to provide details (Cloutier, 2002, Hollister and Miller, 2005, Talitman and Johnson, 1997). A complete understanding of the collection is possible by the reader only if clearly and explicitly done. In such circumstances, even the results would be of good standard.
The effects of extraneous variables are reduced if a control group is used in a study. Comparison of an intervention group to a control group provides better results than if the intervention group alone were present. A larger sample is inevitable also if the control group is made with similar demographics. The larger sample is also not always feasible. The difficulty lies in obtaining a control group which is similar to the intervention group in its composition. The sample in Greenberg’s study (2010) and Taliman and Johnson’s study (1997) formed the waitlist controls till they were subjected to the intervention of EFT. Dandeneau (1994) also had a control sample. MacIntosh’s study (2008) and Hollist and Miller’s study (2005) did not have a control hence the difficulty to generalize arose.
The time span of observation or in which studies were conducted is the final variable under design and data collection matrix. The period of intervention therapy and the follow-up both together constitute the period of observation. MacIntosh (2008) and Hollist and Miller (2005) do not have a time span indicated. All the other four studies were rated standard as they included the time span of the treatment sessions with the follow-up. The two without the time span could be demarcated as substandard as they did not have the period of observation. The length and duration of treatment sessions were also included. The long term maintenance of programs can be performed only through a follow-up measurement. Maintenance of program outcomes is useful in that findings need not be repeated. Valuable information would be obtained from the maintenance of results. This would be especially useful in the studies on the chronically ill children.
Data Analysis and Interpretation
The data analysis and interpretation were all either rated superior or standard. Analysis and interpretation involve the final matrix where the evaluations are validated and data presented. How well a researcher explains his methods and whether he is able to do it in a reader-friendly manner is what decides the criteria for data analysis. The studies that have been rated superior in data analysis are the Greenberg study (2010), MacIntosh study (2008) and the Dandeneau study (1994). The Greenberg study (2010) used the views of the therapists and supervisors to analyze whether the treatment manual was adhered to, using a 5-point scale and whether forgiveness was dealt with. The therapeutic alliance was found satisfactory. The effect of treatment was assessed using a repeated measure 2×3 ANOVA. This was also used for the 3-month follow-up. Thematic analysis is the method used in the research by MacIntosh (2008) to organize the data into different themes. Initially a coding framework is developed based on the theory or concept adopted and the hypothetical questions. A set of lower order themes are then derived. In the third step, the lower order themes are reorganized into global themes. Emotional flooding, emotional numbing, dissociation, constricted range of affect and affect dysregulation, hypervigilance to attachment figures and sexuality were the global themes derived in this study. From these, the researchers built theories and concepts of change. Thematic analysis conveyed the information that affect regulation was the area of challenge in the application of EFT in CSA trauma patients (MacIntosh, 2008).
Hollist and Miller (2005) used a structural equation modeling for analysis. It was a two stage analysis. The confirmatory factor analysis was first done to ensure The factor structure, reliability and applicability of the instrument was determined using the analysis of the confirmatory factor. The model was then fit into the data. The two variables, attachment style and marital quality were compared using the regression coefficients. It was found that the Revised Dyadic Scale was appropriate for the analysis. Modifications to the model were made by adding affect regulation and sexual relations.
Cloutier (2002) used the analysis of the Dyadic Adjustment Scale to interpret results. Reliability of this instrument has been found to be a coefficient alpha of 0.96 and its criterion validity has been demonstrated in earlier studies. The studies superior in interpretation have been the Greenberg study (2010) and the Dandeneau study (1994). They have explained the data analysis methods explicitly and have not indulged in the usage of technical language.
Talitman and Johnson (1997) used analysis of covariance on the outcome variables; the covariates being pre-treatment scores. Planned comparisons were done with t tests at posttest and follow-up for each hypothesis. Couple mean scores were used to perform analysis. The self-reports provided the measures of intimacy as the primary therapeutic effects in both the EFT and Cognitive Marital Therapy (CMT). Both the therapies showed increased self-reported intimacy when compared to the wait-list controls. Differential effects were in favor of EFT when compared to CMT.
Dandeneau’s study (1994) had preliminary analyses to see that the variables satisfied assumptions in the statistical analyses. Reliability analyses from literature review ensured the validity of results. EFT was found to be well instituted. Inadequate interventions were rooted out. Analyses were corrected using the Bonferroni method.
Results analysis
Results Matrix
Comparative analysis of results
Each of the six evaluations were compared with paradigms, expected results and obtained results. All the paradigms had been rated defective as there was hardly any discussion in the paradigm or framework of guidance to research decisions and actions (Mertens, 2009). It is a sad situation where researches do not bother to include clear and explicit discussions on these subjects. Only one point is clear; all the evaluations had a postpositivist paradigm.
The quality of summarisation of the results has been placed under the heading of expected results. Ratings of the selected studies have ranged from defective to superior. Greenberg (2010) explained the results that could be expected but glossed over them though his obtained results fared much better and were rated superior due to the intense discussion. Many results could be referenced following the discussion. All the other expected results of the remaining studies were rated substandard as the expected results were not obvious from the paper.
The obtained results section ratings provided were similar to the interpretation of data section as there is a relationship between the two variables. The obtained results depended on the manner in which or the explicitness with which the intervention outcomes were presented. Greenberg (2010) and Dandeneau (1994) received superior ratings while all the other four received standard ratings. The studies with standard ratings had long and explanatory results but they were not too reader friendly. The two superior quality evaluation studies were rated thus because the data were presented well, with explanations, clarity and conciseness.
Ethical considerations, Conclusion and Discussion
Ethical considerations
Ethical issues have always been a part of social science research. Moral guidelines have been set out by the National Association of Workers (NASW) Code of Ethics to ensure that researchers know the limits they may cross in the name of research. The well-being of the vulnerable population has to be protected and the possible ill-effects of the research prevented through awareness. The moral integrity of a researcher must be adequate enough to ensure a trustworthy and valid research process. Beneficence, respect and justice must be an integral part of the research. The research subject has every right to safe-guard his integrity and privacy. The study should leave minimal impact on the physical and mental health of the subject. His personality should not be disturbed.
The IRBs watch over all proposals with humans and animals. Any institution obtaining Federal Funds for research on human or animal subjects had to constitute an Institutional Review Board (IRB). A lapse on this rule could mean a forfeiture of the Funds. Projects had to be scrutinized for predictable risks as compared to the benefits. The effort to increase positive outcomes while avoiding undue harm or risk is the essence behind the principle of beneficence. Courtesy to all cultural populations demonstrates respect. The principle of justice insists on benefit to any participant (Thornton, 2009). These principles are accepted by all social science researchers as basic morals to be adopted in the quest for knowledge for ensuring their accountability.
Consent must be freely available from the participant who has a right to know what he is offering himself for. No group of individuals must be left out of the study due to his race, sex, religion or class. Debates have raised the question of whether Informed consent procedures and IRBs are sufficient to protect human participants. The evaluation studies that address the ethical concerns need to have recognized standards. Research designs would be implemented by researchers who are competent and qualified. Voluntary consent is taken, risks are informed and then the expected results of the research are imparted to the participant.
Associations now advertise and discuss many ethics issues on the internet on their website. Ethical categories of conduct, general principles of professional conduct dealing with issues of integrity, justice, privacy and confidentiality are outlined. Guidelines on record keeping, fees, therapeutic issues in practice and sexual intimacy with clients are some of the published information.
Issues concerning the evaluation studies in this paper may be different from the usual ones. The study with married couples with chronically ill children may be mentioned in this context (Cloutier, 2002). Children as a whole need to have their rights protected being a vulnerable population, more so when they are chronically ill. When the parents are being involved in the evaluation study, the welfare of the children must be taken care of; they should not suffer in the process. The chronic illnesses could include disruptive behaviors which could mean that they need attention.
Care must also be taken with adults who give a history of child sex abuse (MacIntosh, 2008). These adults could be having “anxiety, anger, depression, re-victimization, self-mutilation, sexual problems, suicidal ideation, impaired self-concept, obsessions and compulsions, PTSD and somatization (MacIntosh, 2008).
The implementation of the research design needs to be examined. The evaluation includes the structure of the intervention, the staff and the manner of collection of data. Competent and qualified researchers must conduct the research (Thornton, 2009). The intervention programs in the studies selected were administered by therapists and supervisors who were trained for the study. The ethical issue here is that a therapeutic intervention like EFT may have disastrous effects if not managed efficiently. The evaluations included in this study did not appear to have any major ethical issues. Though ratings varied for the various parts of the research, ethics appears to be well above standard rating.
Conclusion
Social programs and therapeutic interventions have to be evaluated for effectiveness. This project aims at comparing the effectiveness of the therapeutic Emotionally- Focussed Therapy in couples with marital stress of various origins. The evaluation studies for EFT have been selected from existing literature. How the inferences made in this project can be used to contribute to real practice will be determined also. Marital stress could occur due to emotional injuries acquired from the partner and which could be corrected through forgiveness by the partner (Greenberg, 2010). The effectiveness of EFT for partner’s history of child sex abuse was the second study evaluated (MacIntosh, 2008). The studies indicated that EFT was effective in addressing marital stress and did not need any changes caused by the diversity of origins. The evaluations of the six studies indicated that the quality of the research is high and has several implications for practice and the society. Reduction of the maladaptive interactions between spouses improves the good interaction between the spouses.
EFT is a combination of experiential and family systems approach to psychotherapy. The treatment of relationship discord appears to be most effective through EFT. The cycles of interaction are identified initially and the underlying emotional experience recognized for each part of the cycle. The problem is reframed by the therapist who then facilitates its emotional acceptance by each partner. The interaction between the couple brings forth more issues which are then tackled by the partners themselves. Goal attainment, marital adjustment and marital intimacy becomes evident in the course of therapy lasting for about 8-12 sessions. The outcomes of EFT have been indicated to be better than cognitive behavioral therapy (Dandeneau, 1994).
There could be some limitations to EFT on a long term basis. The evaluation studies had some limitations though a two year follow-up was done in some. Tracking participants after a few years is almost impossible. Multiple studies in diverse populations could produce better outcomes. Further research would enhance the knowledge on the practice of EFT.
Discussion
Partners who had promised to stay together during the EFT showed particular changes during and following the therapy (Greenberg, 2010). This was not seen during the waitlist period which is the control in this study. It was found that the partner who had been emotionally injured exhibited improvement. They were able to excuse discord and trusted their partner more. Increased ability to forgive was a major change. EFT was found effective in treating an injured partner’s marital distress by promoting the attitude of forgiveness. The other partner or the injurer also promoted forgiveness in the injured partner but the injurer did not change in forgiveness or trust. Follow-up at three months showed the persistence of improvement in forgiveness in the injured but trust appeared to have weakened. The injurer also showed the improvement but here again trust was less. The EFT had an enduring effect. Though the partners were more forgiving after the EFT, many of their symptoms still remained. Forgiveness could lead to a persistence of marital satisfaction even after two years. The therapy being conducted for partners living together produces better results than for individual cases. The emotional injuries were gradually amended and a “better more positive relational narrative” assisted in the recovery from emotional trauma (Greenberg, 2010).
MacIntosh’ study (2008) indicated that half of the couples where a partner was a victim of child sex abuse showed an increase in mean relationship satisfaction following EFT. Those who reported deterioration were separated by the end of therapy. Most of the couples actually expressed a desire to continue therapy after the study. The trauma symptoms were reduced in five of the 10 couples. The instrument CAPS showed a bigger improvement in trauma symptoms than from the TSI. Affect regulation was found to be the main issue for CSA survivors in EFT. “Shame, anger, hypervigilance and inability to trust and take risks” were the areas to work upon for the therapists and participants (MacIntosh, 2008). Thematic analysis has provided rich data for future research. During the EFT, the participants were having difficulty in controlling their affect regulation.
It was indicated that couples who had been married for more than ten years could not find a positive relationship between attachment styles and quality of relationship that is usually seen in young people (Hollist and Miller (2005). The relational security if established as in the long years of marriage allows increased resilience to marital problems. The stability and security of attachments increases with time. Once the patterns of relating and perceiving remain for a long period and become established, they remain stable throughout the rest of marriage. This study also supported the four category model of attachment. EFT may therefore not be a suitable method for treating midlife marriage.
EFT produces secure bonding even though there is a continuous stressor like chronically ill children (Cloutier, 2002). The trust, intimacy and affiliation with the partner allow them to face problems with more ease. These features of a close relationship can be assisted along by EFT. Both the EFT and CMT were found to produce equal levels of self-reported intimacy in married couples (Talitman and Johnson, 1997). Marriage can be enriched with improvement of communication skills. EFT therapist needs to focus on building a strong alliance between married partners (Dandeneau, 1994). Marital disharmony requires psychological interventions like the EFT to produce changes in behavior which could overcome the stress and facilitate harmony. More research needs to be done for evolving therapy in natural settings, developing better interventions and aiming at better outcomes. Therapy should be able to lead a highly dissatisfied person to shift to become moderately satisfied group. The dysfunctional person should move to a functional one. Marital distress not being a psychiatric disorder but creating distress all around needs to be carefully managed so that the resultant problems of children, the family members, friends and associates are reduced to a minimum.
The therapist and the psychiatrist together should be able to manage cases of marital distress. The key goal of therapy in EFT is to change insecure attachments to increase marital satisfaction. Midlife couples who have secure attachments will not benefit from EFT but those who have insecure attachments may do so. EFT therapists should focus on insecure attachments for their practice. The perception of relational quality would improve.
Couples requiring marital therapy are at the ends of their tether due to painful conflicts, blaming their spouse for most of it. The concept of couple therapy has changed over the past few years from behavioral and cognitive approaches to therapies focusing on emotions, empathy, acceptance and vulnerability. This therapy has become more research-based in recent days and the adult attachment theory has become part of it. Assumptions, interventions and strategies are well-defined. The short duration of the therapy, lasting from 8 to 20 sessions is appealing to the couples. Various researchers have suggested methods for improving the results and ensuring a lasting marital satisfaction among couples, reducing the number of separations and divorces and improving the psychosocial functioning of the involved children. Practice research contributes greatly to social interventions.