Banca de DEFESA: ADELSON REGIS TEIXEIRA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : ADELSON REGIS TEIXEIRA
DATE: 26/05/2023
TIME: 14:00
LOCAL: Remota
TITLE:

NETWORK ANALYSIS APPLIED TO THE BRAZILIAN VERSION OF THE BECK COGNITIVE INSIGHT SCALE (BCIS)

KEY WORDS:

Psychometrics; Validation Study; Cognition

PAGES: 65
BIG AREA: Ciências Humanas
AREA: Psicologia
SUMMARY:

Cognitive or clinical insight refers to the ability to attribute meaning to the experience of symptoms present in psychotic disorders, especially those related to the schizophrenia spectrum. The Beck Cognitive Insight Scale (BCIS) was developed to provide a more comprehensive assessment of the level of insight in both clinical and non-clinical settings. This instrument was adapted for the Brazilian context and only its content validity was verified. The present study sought to evaluate the psychometric properties of the BCIS formed by a Brazilian sample (BCIS-BR), considering exploratory and confirmatory factor analysis methods based on network psychometry, namely exploratory graphic analysis (EGA) and exploratory graphic analysis with bootstrap (BootEGA), respectively. This is a development research that evaluates the psychometric properties of a measurement instrument in a Brazilian sample. The instrument is originally composed of 15 items, divided into two dimensions, self-reflection (nine items) and self-certainty (six items). A non-probabilistic sample of 249 people was used, divided into 3 groups: 92 with psychotic symptoms or disorders, 41 with other disorders but without psychotic symptoms, and 116 with no psychiatric diagnosis. Data collection was carried out through a face-to-face interview with the vast majority of people with disorders with psychotic symptoms. The rest of the sample had data collected online. All analyzes were performed using the R v software. 4.0.3. The analyzes performed were as follows: (1) Exploratory Graph Analysis (EGA), to estimate the number of dimensions in multivariate data using undirected network models; (2) Bootstrap Exploratory Graph Analysis (BootEGA), to estimate and evaluate the dimensional structure of the model; (3) redundancy analysis, used to verify whether the items are answered in the same way in relation to others, through cooperation; (4) entropy value, which indicates the stability level of the system, without, however, having a cut-off point; (5) descriptive analysis for sample characteristics and items remaining from the BCIS-BR. This research proved that the structure with two dimensions was maintained in terms of using network psychometry as a procedure for dimensional configuration of the Brazilian version of BCIS (BCIS-BR). Furthermore, based on the analyses, items 3 and 13, associated with the self-reflection dimension, were excluded. Currently, a network analysis used for psychometrics, using EGA and BootEGA techniques, reveals more consistent models in understanding the interaction between items and their groupings. This fact makes it possible to observe a dimensional configuration that reflects a latent structure.

BANKING MEMBERS:
Presidente - 3042695 - MARIO CESAR REZENDE ANDRADE
Interna - 3031506 - Mônia Aparecida da Silva
Interno - 1356905 - MARCO ANTONIO SILVA ALVARENGA
Externo à Instituição - JOAO VINICIUS SALGADO - UFMG
Externo à Instituição - ARTHUR DE ALMEIDA BERBERIAN