The team chose to interview the parents of 32 children and a clinician trained in the practice of the treatment manual. The children would be treated through a variation of group and individual therapy to test out all the possibilities of outcome ie; a child may respond better with one to one therapy rather than in a group of say 8 adults and children, especially if they have a social anxiety.
Also therapy conducted without their parents present may make a difference to the outcome as a parent may display anxiety disorder of their own which exaggerates the child’s condition. Also a child may feel afraid to disclose any contributing factors in the home life that has impacted on their condition if their parents are present in the therapy room (Gosh 2006). In all cases all the variations of treatment must adhere to professional codes of ethics and practice in order to safe guard the child. There was no indication in the research paper that this had been done.
All children met criteria for at least one DMS-1V anxiety disorder (2010) based on a clinical interview and semi-structured interview (ADIS-P, Silverman 1996). It was noticed that the team chose children from different demographic origins (27 white, 4 Asian, 1 African descent) but did not comment on the ethical significance of different cultural and social beliefs, values and customs that could affect the parent’s answers during interviews and in the questionnaires. For example, what one culture sees as ‘aggressive’ another may think is just ‘spoilt’!
In the case of social desirability of an answer, saying ‘no’ to a question because you consider a ‘yes’ to have social prejudicial outcome towards the family as a whole (Edwards 1957, Vernon 1963 as cited in McLeod 2003 p68). It was also noticed that there were 20 children who presented with a co-morbidity of anxiety disorders, It was not entirely clear who had what and how many disorders they where diagnosed as having, but it could be said that multiple disorders could affect the answers to the parental questionnaires as one symptom could apply to one anxiety disorder and not to another.
It has been suggested in previous research similar to Monga et al (2009) that pseudo educating the parents on the differences between the various anxiety disorders, recognising symptoms and managing the condition would make a significant difference to the outcome of an answer (Suveg 2006). What a parent initially thought was a high scoring symptom could now have a re-educated opinion as a result of pseudo education and upon re-taking the test, after treatment scores it lower.
To omit some kind of pseudo education may have a dramatic effect on results i. e. from the uncertainty of the parents answer, the researcher is now in danger of making a type II error (beta) from the results, in other words the researcher may have accepted a null hypothesis that the theory works from the results but in fact the results may not have been confirmed by the questioning method (Duerden 2010). Data Analysis The clinician’s rating of the child’s anxiety disorder was tested using the Children’s Global Assessment Scale – CGAS (Shaffer 1983).
The conclusions where that treatment had an effect on improving symptoms of anxiety and the CGAS score on the scatter diagram concluded a statistical significance in the chance of a null hypothesis (p = 0. 001). From this statistic it was said that pre treatment and post treatment assessment using 5 different methods of assessment produced a 1 in 1000 chance that the results could be dis-proven if the research was to be done again.
The reasons why they may have chosen 5 different research questionnaires/ interviews and 2 forms of data analysis is that the combination of tests enabled a sufficient variation when reporting on multiple comparisons. This is referred to as a triangulation of research methods (McLeod 2003). The findings from one type of study such as a qualitative approach to developing a treatment manual, and then testing the effectiveness of the manual using a quantitative research study would encompass many perspectives and would hopefully clarify a statistical significant result in the research conclusions (McLeod 2003).
Conclusions It was noticed that Dr. Young and her teams’ manual had not yet been published. This could be down to numerous possibilities but perhaps the results of the pilot test, although statistically significant, highlighted areas for improvement. It was noticed that 2 months later after the research carried out with Dr Young and Mary Owens (2009), Suneeta Monga participated in a research that looked more specifically at the quantitative measurement of feelings in a computerised feelings assessment instrument called MACC (Manassis 2009).
These findings highlighted the importance of ‘fine tuning’ testing methods, adapting theories and re-testing hypothesis. Constantly reassessing treatment methods would validate outcomes and enhance the probability of cure or symptom management. A triangulation of different research methods would further enrich a Psychotherapeutic theory and support the hope for a statistical significance.
References
Beitchman, J. H. , Adlaf, E. M. , Douglas, L. , Atkinson, L. , Young, A. , Johnson, C. J. , & … Wilson, B. (2001) Co-morbidity of psychiatric and substance use disorder in late adolescence: A cluster analaytic.
American Journal of Drug & Alcohol Abuse, 27(3), 421. Birmaher B, Khetarpal S, Brent D et al (1997) The Screen for Child Anxiety Related Emotional Disorders (SCARED): scale construction and psychometric characteristics. J Am Acad Child Adolesc Psychiatry;36:545-553. Blatt et al (1996) Characteristic of effective therapists: further analysis of data from the National Institute of Mental health treatment of depression collaborative research program. Journal of Consulting and Clinical Psychology, 64 (6):1276 -84 Burton (2006) Psychiatry (pp. 161 ). Malden, MA: Blackwell.