2. Methods and procedures

The sample consisted of 457 adolescents studying at secondary schools in the Slovak Republic (268 girls, 189 boys) aged from 15 to 18 years (mean age 16.24). We provide data collected at the end of the 2nd grade in public high schools throughout the nation (end of school year 2015/ 2016). Students' participation in the research was voluntary (the research battery was administrated personally in 18 high schools). Before administration of research methods, every group of students were given a complete explanation of the research study.

Academic self-efficacy was measured by The Morgan-Jinks Student Efficacy Scale [17], which was designed to acquire information about student efficacy beliefs that is related to school success. The scale consists of 30 items with 3 subscales: talent (15 items), context (9 items) and effort (6 items). All the items are designed using a four-interval scale (from 1 = really agree, 2 = kind of agree, 3 = kind of disagree and 4 = really disagree.). The score for academic selfefficacy was calculated by summing up the scores for the 30 items, after reversing the scores for 9 items (4, 6, 15, 17, 19, 21, 23, 25 and 28). Scores for general academic self-efficacy range from minimum 30 to maximum 120, scores for the talent subscale ranges from minimum 15 to maximum 60, scores for the context subscale ranges from minimum 9 to maximum 36 and scores for the effort subscale range from minimum 6 to maximum 24. The reliability of the scale was Cronbach's alpha = 0.687. The reliability of subscales was: talent: Cronbach's alpha = 0.720, context: Cronbach's alpha = 0.601 and effort: Cronbach's alpha = 0.402.

Heteronomous evaluation of academic achievement (HAA) we measured by GPA in whole study subjects in the end of school year 2015/2016. In the Slovak Republic a student's achievements in individual subjects are measured using the following grades: 1 = excellent, 2 = very good, 3 = good, 4 = sufficient and 5 = poor/unsatisfactory, while the final evaluation in a given subject included in the school report at the end of each school year results from an average of many grades given regularly in relation with achieved results (a teacher can take/takes into account also attributes other than only marks given during the school year). The resulting GPA reflects an average academic achievement calculated on the basis of all end-year grades at a school report (minimum 1, maximum 5).

Autonomous evaluation of academic achievement (AAA) we measured by perceived selfevaluation of academic achievement (expressed by self-perceived quality level of knowledge and skill in whole study subjects in end of school year 2015/2016). The adolescents subjectively rated their success at school at 6-point scale: from excellent to absolutely unsatisfactory. The score for AAA ranges from minimum 1 to maximum 6. The scale was developed by the authors of research study.

Approach to learning (two main scales: deep – DA and surface – SA) we measured by The Revised Study Process Questionnaire (R-SPQ-2F [47], Slovak version [43], four subscales: deep motive (DM), deep strategy (DS), surface motive (SM) and surface strategy (SS). Each of the subscales consisted of five items. The final version of the questionnaire therefore has two main scales deep approach (DA) and surface approach (SA) with four subscales. It is rated on a 5-point Likert scale (1 = never, 2 = occasionally, 3 = sometimes, 4 = usually and 5 = always). The score for the deep approach to learning ranges from minimum 10 to maximum 50, the score for the deep motive subscale ranges from minimum 5 to maximum 25, the score for the deep strategy subscale ranges from minimum 5 to maximum 25, the score for the surface approach to learning ranges from minimum 10 to maximum 50, the score for the surface motive subscale ranges from minimum 5 to maximum 25 and the score for the surface strategy subscale ranges from minimum 5 to maximum 25. The reliability of main scales was: deep approach: Cronbach's alpha = 0.658, surface approach to learning: Cronbach's alpha = 0.683. The reliability of subscales was: deep motive: Cronbach's alpha = 0.616, deep strategy: Cronbach's alpha = 0.567, surface motive: Cronbach's alpha = 0.549 and surface strategy: Cronbach's alpha = 0.548.

For identification of significance of the relationship between research variables, we used Pearson's correlation coefficient. For identification of significant relationship between a dependent variable (HAA, AAA) and independent variables (predictors), we used linear regression analysis. The statistical significance criterion for our research study is 5% (p ≤ 0.05). For all data analysis, we used Statistical Package for the Social Sciences (SPSS 20.0).
