Academic Journals Database
Disseminating quality controlled scientific knowledge

Optimal scaling for early life stress measurement

ADD TO MY LIST
 
Author(s): Olena Iakunchykova | Tatiana I Andreeva | Zoreslava Shkiryak-Nizhnyk | Yuri Antipkin | Daniel Hryhorczuk | Alexander Zvinchuk | Natalia Chislovska

Journal: Tobacco Control and Public Health in Eastern Europe
ISSN 2222-2693

Volume: 3;
Issue: 1;
Start page: 17;
Date: 2013;
Original page

Keywords: psychological stress | questionnaires | mathematical models | algorithms | scaling | statistics | computation | optimal scaling | correspondence analysis

ABSTRACT
BACKGROUND: Early life stress is associated with high risk for both negative psychological and physical health outcomes. However, many of the stressful life events inventories that have been used in epidemiological research have not been validated or checked for reliability or consistency. The aim of our study is to use optimal scaling and correspondence analysis that employ categorical principal components analysis (CATPCA) algorithm to consider the internal structure and the geometry of the space of variables obtained through the questions measuring early life stress. This approach was chosen because it allows quantification of categorical (both nominal and ordinal) scales and reduction of initial number of variables with interval quantification of the resulting dimensions. METHODS: A questionnaire for measuring early life stress was applied to the participants of the Ukrainian component of European Longitudinal Study of Pregnancy and Childhood at the age of three and seven years. CATPCA algorithm was used to elaborate a tool for estimating related integral quantified characteristics.RESULTS: Application of quantification and dimension reduction techniques to the categorical variables measuring stress in three- and seven-year-old children resulted in two dimensions. The first dimension that accounts for a major part of initial variance and is associated with all the collected variables can be interpreted as the overall measure of stress. The second dimension accounts for smaller but still considerable part of variance and can be related to child’s attachment to mother and acquiring new experience as the route of development.CONCLUSIONS: Application of optimal scaling to the empirical data of early life stress measurement resulted in construction of two integral indicators – first measuring overall stress and second contrasting security related to child’s attachment to mother and new experiences – consistent across age groups of three and seven-year-old children.
RPA Switzerland

Robotic Process Automation Switzerland

    

Tango Jona
Tangokurs Rapperswil-Jona