Abstract:

In contemporary educational approaches, focused primarily on the development of competencies, the training of students capable of leading their learning process is a relevant factor, since it enables the achievement of learning outcomes. Self-regulated learning (SRL) is the ability of a student to apply cognitive, metacognitive, and evaluative skills in an autonomous and motivated way to achieve efficient and effective learning.

Given the relevance of SRL it is important to have scientific instruments to measure it in students to generate improvement strategies. There are different instruments that have been designed in this regard, but not specifically contextualized in Guatemala. Therefore, an adaptation to Guatemala context of the Learning Self-Regulation Scale (ESAA-2), an instrument designed and validated in Mexico, has been carried out. It includes 40 Likert-type scale items, divided into three dimensions:
(1) motivation an attitude towards learning,
(2) cognitive strategies, and
(3) autoevaluation and metacognitive regulation.

The objective of this work is to assess the psychometric quality of the adapted instrument (ESAA-G) in terms of content and construct validity, as well as reliability. To achieve it, a study with instrumental design was developed. The sample included 517 middle school students (basic and diversified levels) of 6 schools in Guatemala who freely participated after being informed about the purpose of the study and the confidentiality of the gathered data. The instrument was applied online by G-Forms (Google Suites). The whole investigation process was developed considering the ethical principles promoted by the Committee on Publication Ethics (COPE).

The database was processed in SPSS Statistics 27 and AMOS (IBM). The analysis included three aspects:
(1) the content validity developed through the expert judging method,
(2) the internal consistency using the Cronbach Alpha coefficient to interpret the reliability of the subscales (D1, D2, D3) and the general scale (GS), and
(3) the construct validity applying a confirmatory factor analysis (CFA) performed with structural equations.

The data analysis gives evidence of an adequate content validity of ESAA-G, due to the favourable perspective of the judges regarding the item-domain congruence, the sufficiency of items to measure SRL, and the pertinent wording of items. A good level of reliability was observed (Alpha(D1) = 0.85, Alpha(D2) = 0.87, Alpha(D3) = 0.91, Alpha(GS) = 0.95). Finally, the CFA model reflected an adequate fit (RMESEA = 0.05, CFI = 0.90, TLI = 0.89, NFI = 0.84) demonstrating construct validity.

The main strength of this study is the adequate psychometric quality of the adapted instrument (ESAA-G). The foregoing benefits the Guatemalan educational system, considering that no instruments were identified to measure SRL in the national context. The continuity of this study will be oriented to an application of the instrument to a larger sample with national representation, to achieve a confirmatory validation for the entire country.

Keywords:

Educational innovation, Self-regulated learning, Test, Reliability, Construct Validity, Guatemala.