engleski

The Structure of Self-Competence Beliefs of Croatian Primary School Students in STEM area

Proposal Information (e.g. topic, research question, and objective, conceptual or theoretical framework …) (600 words) Many contemporary policies and current research evidence suggest how the interest of European youth for STEM school subjects is declining, the number of students choosing to study STEM university programs is low and especially low is the number of students choosing carriers in STEM professional area. This is a matter of concern amongst policy makers in many particular countries, (Roberts, 2002 ; National Academy of Sciences 2005) and across Europe (European Commission, 2004). Existing evidence highlighting the issue of self-competence beliefs, shows this constructs as especially important in understanding and explaining students’ school performance in STEM area. Self-competence beliefs represent “children's cognitive representations of how good they are at a given activity” (Freiberger, Steinmayr, & Spinath, 2012, p. 518). They encompass information important for one’s identity and information important for one’s performance. Information about one’s efficacy, or academic self-efficacy beliefs (ASE) represent individuals' convictions that they can successfully perform given academic tasks at designated levels (Bandura, 1997 ; Schunk, 1991). They refer to students' perceptions of their ability to master given tasks or develop specific competences (Bong, 2001) or judgments that reflect task-specific performance expectations (Zimmerman, 2000). On the other hand, self-evaluations of specific abilities or qualities are sometimes termed as domain-specific self-concepts, or more broadly as an academic self-concept (ASC) (Marsh 1990, Marsh, Byrne, & Shavelson, 1988 ; Marsh and O' Mara, 2008 ; Schavelson, Hubner, & Stanton, 1976). These constructs have sometimes been used interchangeably despite their conceptual and operational differences (Bong & Skaalvik, 2003). This paper aims to further explain some theoretically driven research question. The main objectives of the present paper are to examine structure of Croatian primary school students’ self-competence beliefs in STEM school area, where beliefs related to mathematics are especially important issue of consideration. Existing studies do not provide a definite answer to the question whether students structure their achievements in this broad area in a one-dimensional academic self-concept or whether it is reasonable to assume the existence of separate self-perceptions, in different STEM areas. The debates in the STEM field are implicitly based on the assumption that the STEM academic self-concept is one-dimensional (Scherer, 2013), while contemporary models of academic self-concept, such as the Marsh/Shavelson model, (Marsh, 1990 ; Marsh & Shavelson, 1985 ; Marsh et al., 1988) are based on the distinction between several academic areas. Jansen, Schroeders and Lüdtke (2014) have shown that it is possible to find confirmation for both one-dimensional and a subject specific model of self-concept in the STEM field, with somewhat stronger support for the subject specific model. In their study, a strong effect of the structure of curriculum was observed. This issue is intriguing in Croatian primary schools, since the curriculum contains several school subjects that cover STEM fields. We expected that related self-competence beliefs facets can be recognized within the STEM field as indicators of academic self-concept, which form a clear structure of self-competence beliefs in the STEM school achievement area. At the same time, we expected that the data suggest that Croatian primary school students mostly structure their beliefs about themselves in relation to specific school subject. Methodology, Methods, Research Instruments or Sources Used (400 words) The data for this paper arise from the JOBSTEM project (www.jobstem.eu), which is a comprehensive longitudinal-sequential (cross-sequential) research project with experimental STEM intervention intended to increase the interest of primary school students for STEM school subjects, help them develop positive attitudes towards STEM, and give them more information about possible future STEM educational and career choices. A total of 1920 primary school students and their parents/guardians from 16 primary schools participate in the study. Students are grouped in three age cohorts, which are followed longitudinally through three measurement waves. We used several research constructs to answer the research questions related to self-competence beliefs. They are intended to measure school subjects ‘Self-Concept of Ability’. In the Croatian school system and school curriculum the STEM area is covered by the following school subjects, depending on the grade: mathematics, nature and society, biology, chemistry, physics, ICT, technical education and geography. For these school subjects we developed Likert scales for each school subject (e.g. ‘Math Self-Concept of Ability Scale’, ‘Chemistry Self-Concept of Ability Scale’ etc.) and each scale consisted of 5 items. Items we used are derived from Jacqueline Eccles headed longitudinal “The Childhood And Beyond (CAB)” research project, from which we adapted items for usein our JOBSTEM research project. (e.g. “How good are you at math? /1=not very good, 7=very good/, “Compared to other subjects how good are you at math? /1= a lot worse, 7= a lot better/” etc.). Conclusions, Expected Outcomes or Findings (300 words) In order to determine the basic number of dimensions that define the self-competence STEM area we used an items correlation matrix. We used the principal axing factoring, where the number of significant latent dimensions that will remain is determined based on the Kaiser-Guttman criterion and Cronbach’s alpha analysis to determine structure and internal consistency of expected composite in STEM area. We obtained several one factor structures of used items where on the Math scale of self-concept of ability first component explained 77, 37% of the variance, on the Nature and Society scale 75, 67% of the variance, on the Geography scale 82, 42%, on the Technical education 78, 11%, on ICT scale 77, 8% and on the Computing scale 73, 82%. Cronbach’s alpha scores for each for all of the six scales were shown to be high: .925 ; .918 ; .946 ; .928 ; .927 ; and .91, respectively. The interrelations of composite yields interesting picture – the highest correlations are those between factors explaining ICT and Computing scale (.-64), Math and Geography scale (.42) and Geography and Nature and society scale (.36). Results suggest that self-concept of the ability in STEM area in Croatian primary school system should be mostly regarded as subject- specific, multidimensional construct with different factors explaining differences among students’ self-evaluations in specific STEM school subjects. Results show the existence of a substantial relationship between nominally different aspects of STEM school area and indicate that covered STEM space could reasonably be well described by a smaller number of underlying factors which are more in directions of existence of school subject’s specific self-concepts. This outcome, will be discussed in the context of current self-competences research.

STEM, self-competence, STEM education, Mathematics education

nije evidentirano