Factor structure of the convergence insufficiency symptom survey questionnaire

Abstract

The purpose of this study is to analyze the factorial structure of the Convergence Insufficiency Symptom Survey questionnaire with 15 items, in order to identify latent dimensions that can contribute to a more focused implementation. The questionnaire was self-administered in paper by 183 university students, in the age span of 18 to 30. Both Kaiser-Meyer-Olkin measure and Bartlett's sphericity test were performed to ensure the validity of the factorization. In order to analyze the principal components, factors which obtained eigenvalues greater than 1 were chosen. The extraction of factors was performed after computing a Promax rotation and a Kaiser criterion. In each extracted factor, the internal consistency was used to prove its reliability. Bartlett's sphericity test was statistically significant (p <0.001), and the Both Kaiser-Meyer-Olkin test was 0.89, confirming the factorization of this instrument. Exploratory factor analysis followed by a Promax rotation and scree plots graphic, extracted three factors that explained 62.1% of the total variance. The composition of each factor suggests the following meanings: Factor 1 (somatic sensation) includes 8 of 15 items; Factor 2 (impaired vision) includes 3 of 15 items; Factor 3 (cognitive performance) includes 4 of 15 items. Cronbach's alpha coefficient demonstrated good internal consistency (α> 0.75) in three dimensions. The multivariate statistical analysis of the Convergence Insufficiency Symptom Survey revealed a three-factor structure, so new forms of questionnaire analysis can be designed in order to obtain gains in the management of a symptomatic patient, using 3 subscores, one for each factor, instead of a total score. The factorial structure of the questionnaire can be used with a high level of confidence in future investigations, which aim to assess which dimensions are most affected in each type of visual changes and develop more targeted therapeutic performances.

Fig 1. Steps followed in Exploratory Factor Analysis.

Fig 2. Scree plot showing the variance explained by each factors to extract.

It is recommended to extract factors with a factorial load greater than 1 that are located above the "elbow" of the graph, that is, the factors above the dashed line.