Construct validity of the attitude diagnostic inventory

Straipsnyje apibūdinamas mokinių požiūrio į mokymąsi diagnostikos aprašo, aprobuoto užsienyje, konstrukto validacijos Lietuvos bendro ugdymo mokyklos VIII–X klasėse procesas. Sprendžiant šį uždavinį, vadovautasi reikalavimu tikrinti užsienyje parengtų ugdymo diagnostikos priemonių tinkamumą kitokiomis, nacionalinėmis, sąlygomis, visų pirma, atskleisti aprašo konstrukto validumą. Šiam uždaviniui išspręsti pasitelkti pakankamai patikimos imties duomenų faktorių analizės ir užduočių skiriamosios gebos vertinimo rezultatai.The article reflects a part of the research, which aims at designing an inventory intended for the evaluation of the attitude towards learning by the students of Lithuanian comprehensive school. Since the experience of preparing this kind of diagnostic tools in Lithuania is minimal, “SMALSI” (School motivation and learning strategies inventory; compiled by K. C. Stroud, C. R. Reynolds, 2006), an inventory designed in the USA, has been referred to. The quality of the tool has been tested under the condition of the USA; however, the application of the diagnostic tool for the social and education reality of Lithuania invokes both a scientific and a practical problem as the students’ answers directly depend on their understanding of the question. Therefore, the adaptation of culture-dependent means falls under certain requirements, analogical to the requirements raised for the original diagnostic tool. The inventory has been modified with reference to the results of the analysis of content validity. It consisted of 48 statements classified into three scales. It was filled in by 648 students of 8–10th classes of nine schools. The data required for the description of construct validity were analyzed using computer software: alpha factor analysis for the verification of the hypothesis of contents mono-dimensionality; method of correlation pleiads for a detailed insight into the internal structure of each scale; relations of estimates, which students attributed to the statements, with the charts presenting the scale; weighted coefficients, which were intended for the quantitative assessment of the resolution of the initial estimates of the statements; scale reliability coefficients obtained by the method of half-splitting the mass of initial estimates; indices of statement popularity, which defined the differences in the frequency of approving and disapproving statements. The research resulted in the empirical verification of the scales intended for the measurement of students’ attitude towards learning: low academic motivation, learned helplessness and concentration / attention difficulties, which, depending on the application needs (education, management, research, etc.) can be transformed as quantitative or ordinal measurement scales while interpreting the data of each individual scale or general estimates of the attitude towards learning