Statistical Analysis of Psychological Data: A Case Study of Reactive Attachment Disorder in Children

Download as pdf or txt
Download as pdf or txt
You are on page 1of 3

Statistical Analysis of Psychological Data: A Case Study of Reactive Attachment Disorder in Children

David Young1,2, , Joanne Reekie1 , Angelica Ronald3 & Helen Minnis2

Department of Statistics and Modelling Science University of Strathclyde Glasgow, UK


2

Greater Glasgow Health Board Yorkhill Division Glasgow, UK


3

Institute of Psychiatry Kings College London, UK

Corresponding Author:

Department of Statistics & Modelling Science University of Strathclyde 26 Richmond Street Glasgow, G1 1XH Scotland, UK email: davidy@stams.strath.ac.uk

Topic Area: Psychological Measurement / Medical Statistics Introduction There is an inherent diculty in the development of diagnostic tools that adequately identify and classify individuals with dierent psychological problems. In psychological research, statistical methods such as factor analysis and cluster analysis are often used to try to identify patterns in data which group individuals in some way. The diculties with this type of research stem mainly from the investigative nature of the work. In many situations, cases are undiagnosed so there is no benchmark from which to validate the statistical analysis. In addition, alternative statistical methods can produce dierent outcomes or diagnoses and the standard approach appears to be that the most interpretable results are reported. Objective ways of deciding on a suitable statistical analysis need to be investigated and developed for data collected in psychological research studies. Background Reactive Attachment Disorder (RAD) is a psychological disorder found in young children who have failed to form attachments with one or more primary care-giver or who have had this attachment severely disrupted [1]. These children lack crucial social and emotional skills and are unable to form and maintain meaningful relationships. In later life this can lead to social exclusion and in some cases severe anti-social behavior, making early diagnosis crucial. Research in this area would benet greatly from the development of a standard diagnostic tool. Methods The data came from the Twins Early Development Study (TEDS), a UK population study of twins born between 1994 and 1996 and contacted from birth records [2]. A questionnaire for RAD behaviors, the Relationship Problems Questionnaire (RPQ), was sent to the parents of 7687 sets of twins and 3766 were returned. The maximum score on this questionnaire was 54. Cluster analysis was used to classify individuals as RAD cases or non-cases based on the RPQ scores. The nal procedure was validated using prior knowledge of prevalence, and additional information from a Strengths and Diculties Questionnaire (SDQ) using logistic regression, while controlling for cognitive functioning, age, gender and social class [3].

Results Cluster 1 (98.8%) had a mean RPQ score of 7.5 (median 6, range 0-34) and Cluster 2 (1.2%) had a mean RPQ score of 42.4 (median 49, range 34-54). The individuals in Cluster 2 are considered likely cases of RAD. Logistic regression analysis indicated that being a RAD case was signicantly associated with increased conduct problems, peer relationship problems and emotional problems. Discussion Objective methods are required for analyzing this type of data. This involves close collaboration between the statistician and clinician in order to apply existing statistical methods in such a way that meaningful results are achieved. Methods for choosing a particular approach have to be considered rather than the blind application of several investigative techniques to achieve suitable results. Additional post-hoc methods should also be considered, where possible, to validate the results of the nal analysis. References [1] Hanson, R.F. and Spratt, E.G. (2000). Reactive Attachment Disorder: what we know about the disorder and implications for treatment. Child Maltreatment, 5(2), 137145. [2] Trouton, A., Spinath, M. and Polmin, R. (2002). Twins Early Development Study (TEDS): a multivariate, longitudinal investigation of language, cognition and behaviour problems. Twin Research, 5, 444448. [3] Goodman, R., Ford, T., Simmons, H. Gatward, R. and Meltzer, H. (2002). Using the Strengths and Diculties Questionnaire (SDQ) to screen for child psychiatric disorders in a community sample. British Journal of Psychiatry, 177, 534-539.

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy