Abstract:

Self-regulated learning is a competence that allows students to apply cognitive, evaluative and metacognitive strategies in an autonomous and motivated way in their own learning process, in order to achieve better results. For this reason, the relevance that the educational systems of the different countries, including Mexico, provide to the development of this competence has increased in recent years, in order to achieve higher educational quality.

The objectives of this work are to design an instrument to measure SFL (Self-Regulated Learning) in upper secondary level in the context of Mexico, and to do an exploratory analysis about its psychometric quality in terms of reliability and construct validity.

The Scale of Self-Regulated Learning (ESAA for its acronym in Spanish: “Escala de Autorregulación del Aprendizaje”) was designed considering different factors associated with SFL development that were identified in the literature. On the first version it had 46 items in a five level Likert-scale, corresponding to three main constitutive dimensions: (1) motivation an attitude towards learning, (2) cognitive strategies, and (3) autoevaluation and metacognitive regulation. After the first application and psychometric analysis, the ESAA-2 version was designed with 40 items. The focus of this study was towards this new version.

In order to achieve the objectives, a quantitative, non-experimental, transactional study, with a descriptive and correlational scope, was developed. The sample included 149 upper secondary students of a private school in Aguascalientes (Mexico) who freely participated after being informed about the purpose of the study and the confidentiality in the use of the gathered data. The instrument was applied online by G-Forms (Google Suites), including additional items related to contrast variables such as sex, school grade, academic average, among others.

The database was processed in SPSS 26 (IBM). The analysis considered two perspectives: (1) the internal consistency, using the Cronbach Alpha coefficient to interpret the reliability of each subscale, and (2) the construct validity, using the item-domain method as well as an exploratory factorial analysis for each subscale.

The data analysis gives evidence of a good level of reliability in the instrument (Alpha(D1) = 0.87, Alpha(D2) = 0.87, Alpha(D3) = 0.85). Also an adequate construct validity was observed in terms of the moderate or high correlation forces among the items and their corresponding domains (r >= 0.43), and the correct fit between almost all the items (38 from 40) and their corresponding indicators in the factorial analysis models elaborated for each of the subscales, in which good levels of explained variance were identified (D1 = 69.39, D2 = 61.49, D3 = 60.34).

The main strength of this study is the design of the instrument itself, which has shown adequate psychometric quality. The foregoing benefits the Mexican educational system, considering that no instruments were identified to measure SFL in the national context, in which the official education approach strongly encourages the development of this competence. The continuity of this study will be oriented to an application of the instrument to a larger sample with local, regional and national representation, in order to achieve a confirmatory validation.

Keywords:

Educational innovation, Self-regulated learning, Test, Reliability, Construct Validity, Upper secondary level, Mexico.