Limits...
Development of a computerized adaptive test for Schizotypy assessment.

Fonseca-Pedrero E, Menéndez LF, Paino M, Lemos-Giráldez S, Muñiz J - PLoS ONE (2013)

Bottom Line: The Information Function estimated using the Gradual Response Model indicated that the item pool effectively assesses schizotypy at the high end of the latent trait.The correlation between the CAT total scores and the paper-and-pencil test was 0.92.The mean number of presented items in the CAT with the standard error fixed at ≤ 0.30 was of 34 items.

View Article: PubMed Central - PubMed

Affiliation: Department of Educational Sciences, University of La Rioja, La Rioja, Spain ; Center for Biomedical Research in the Mental Health Network (CIBERSAM), Madrid, Spain.

ABSTRACT

Background: Schizotypal traits in adolescents from the general population represent the behavioral expression of liability for psychotic disorders. Schizotypy assessment in this sector of population has advanced considerably in the last few years; however, it is necessary to incorporate recent advances in psychological and educational measurement.

Objective: The main goal of this study was to develop a Computerized Adaptive Test (CAT) to evaluate schizotypy through "The Oviedo Questionnaire for Schizotypy Assessment" (ESQUIZO-Q), in non-clinical adolescents.

Methods: The final sample consisted of 3,056 participants, 1,469 males, with a mean age of 15.9 years (SD=1.2).

Results: The results indicated that the ESQUIZO-Q scores presented adequate psychometric properties under both Classical Test Theory and Item Response Theory. The Information Function estimated using the Gradual Response Model indicated that the item pool effectively assesses schizotypy at the high end of the latent trait. The correlation between the CAT total scores and the paper-and-pencil test was 0.92. The mean number of presented items in the CAT with the standard error fixed at ≤ 0.30 was of 34 items.

Conclusion: The CAT showed adequate psychometric properties for schizotypy assessment in the general adolescent population. The ESQUIZO-Q adaptive version could be used as a screening method for the detection of adolescents at risk for psychosis in both educational and mental health settings.

Show MeSH

Related in: MedlinePlus

Information Functions of the item pool with polytomous and dichotomous data and number of items utilized in each CAT for the θ estimation.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3760882&req=5

pone-0073201-g001: Information Functions of the item pool with polytomous and dichotomous data and number of items utilized in each CAT for the θ estimation.

Mentions: The random subsample of 500 participants obtained from the initial sample of 3,056 participants was used to calibrate the 51 items of the CATs. The θ for each participant was estimated both for the polytomous and dichotomous CAT. In Figure 1, the Information Function that corresponds to both models used, as well as the estimated θ values in relation to the number of items presented to reach the cut-off criterion of SE <0.30 (stopping rule of the CAT) is shown. As can be seen in this Figure, the Information Function of the polytomous model provides more information for a greater value range than the dichotomous model. Both functions provide optimal estimations in those adolescents with high latent-trait values (schizotypy).


Development of a computerized adaptive test for Schizotypy assessment.

Fonseca-Pedrero E, Menéndez LF, Paino M, Lemos-Giráldez S, Muñiz J - PLoS ONE (2013)

Information Functions of the item pool with polytomous and dichotomous data and number of items utilized in each CAT for the θ estimation.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3760882&req=5

pone-0073201-g001: Information Functions of the item pool with polytomous and dichotomous data and number of items utilized in each CAT for the θ estimation.
Mentions: The random subsample of 500 participants obtained from the initial sample of 3,056 participants was used to calibrate the 51 items of the CATs. The θ for each participant was estimated both for the polytomous and dichotomous CAT. In Figure 1, the Information Function that corresponds to both models used, as well as the estimated θ values in relation to the number of items presented to reach the cut-off criterion of SE <0.30 (stopping rule of the CAT) is shown. As can be seen in this Figure, the Information Function of the polytomous model provides more information for a greater value range than the dichotomous model. Both functions provide optimal estimations in those adolescents with high latent-trait values (schizotypy).

Bottom Line: The Information Function estimated using the Gradual Response Model indicated that the item pool effectively assesses schizotypy at the high end of the latent trait.The correlation between the CAT total scores and the paper-and-pencil test was 0.92.The mean number of presented items in the CAT with the standard error fixed at ≤ 0.30 was of 34 items.

View Article: PubMed Central - PubMed

Affiliation: Department of Educational Sciences, University of La Rioja, La Rioja, Spain ; Center for Biomedical Research in the Mental Health Network (CIBERSAM), Madrid, Spain.

ABSTRACT

Background: Schizotypal traits in adolescents from the general population represent the behavioral expression of liability for psychotic disorders. Schizotypy assessment in this sector of population has advanced considerably in the last few years; however, it is necessary to incorporate recent advances in psychological and educational measurement.

Objective: The main goal of this study was to develop a Computerized Adaptive Test (CAT) to evaluate schizotypy through "The Oviedo Questionnaire for Schizotypy Assessment" (ESQUIZO-Q), in non-clinical adolescents.

Methods: The final sample consisted of 3,056 participants, 1,469 males, with a mean age of 15.9 years (SD=1.2).

Results: The results indicated that the ESQUIZO-Q scores presented adequate psychometric properties under both Classical Test Theory and Item Response Theory. The Information Function estimated using the Gradual Response Model indicated that the item pool effectively assesses schizotypy at the high end of the latent trait. The correlation between the CAT total scores and the paper-and-pencil test was 0.92. The mean number of presented items in the CAT with the standard error fixed at ≤ 0.30 was of 34 items.

Conclusion: The CAT showed adequate psychometric properties for schizotypy assessment in the general adolescent population. The ESQUIZO-Q adaptive version could be used as a screening method for the detection of adolescents at risk for psychosis in both educational and mental health settings.

Show MeSH
Related in: MedlinePlus