Araştırmada fen bilgisi öğretmen adaylarının fen okuryazarlıklarının, epistemolojik inançları ve fen bilgisi öğretimi öz yeterlik inançlarıyla ne düzeyde açıklandığını bir model aracılığıyla incelemek amaçlanmıştır. Araştırmanın bağımlı değişkeni üzerindeki bağımsız değişkenlerin doğrudan ve dolaylı etkilerini belirleyecek bir model oluşturulmuş ve test edilmiştir. Araştırmada nedensel karşılaştırmalı desen kullanılmıştır, araştırmanın çalışma grubunu farklı üniversitelerden 415 bayan ve 181 erkek olmak üzere 596 üniversite öğrencisi oluşturmaktadır. Araştırmada veri toplama aracı olarak Laugksch ve Spargo (1996), tarafından geliştirilen ve Yetişir (2007), tarafından Türkçe'ye uyarlanan temel fen ve teknoloji okuryazarlık ölçeği, Riggs ve Enochs (1990), tarafından geliştirilen ve Özkan, Tekkayave Çakıroğlu (2002) tarafından Türkçe'ye uyarlanan fen bilgisi öğretimi öz yeterlik inancı ölçeği ve Schommer (1990), tarafından geliştirilen ve Deryakulu ve Büyüköztürk (2002) tarafından Türkçe'ye uyarlanan Epistemolojik inanç ölçeği kullanılmıştır. Araştırmadan elde edilen veriler SPSS 21 ve AMOS programlarıyla analiz edilmiştir. Araştırmada model testi için path analizi yapılmıştır. Araştırma sonucunda fen bilgisi öğretmen adaylarının fen okuryazarlıklarının, epistemolojik inançları bakımından doğrudan, fen öğretimi öz yeterlik inançları bakımından ise dolaylı olarak açıklandığı belirlenmiştir. Çalışmada değişkenler (fen bilgisi öğretmen adaylarının fen okur yazarlığı, epistemolojik inançları ve fen öğretimi alanındaki öz yeterlik inançları) arasındaki ilişki için kavramsal bir model önerilmiştir. Tek bir doğru olduğuna dair inanç değişkeninin bağımsız ve diğer değişkenlerle etkileşimsiz olduğu görülmüş ve bu değişken modele alınmamıştır

The aim of the research is to investigate to what extent the science literacy of science teacher candidates is explained by their epistemological beliefs and self-efficacy beliefs of science teaching through a model. A model was created and tested to determine the direct and indirect effects of the independent variables on the dependent variables of the research. The causal-comparative design was used in the research, and the study group of the research is composed of 596 university students, 415 of which are female and 181 of which are male. The basic science and technology literacy scale developed by Laugksch and Spargo (1996) and adapted into Turkish language by Yetişir (2007); the science teaching self-efficacy belief scale developed by Riggs and Enochs (1990) and adapted into Turkish language by Özkan, Tekkaya and Çakıroğlu (2002); and the Epistemological beliefs inventory developed by Schommer (1990) and adapted into Turkish language by Deryakulu and Büyüköztürk (2002) were used in the research as data collecting tools. The data obtained from the research were analyzed with SPSS 21 and AMOS software packages. The path analysis was performed for testing the model in the research.It was consequently determined in the research that scientific literacy of science teacher candidates was explained by their epistemological beliefs directly, their science teaching self-efficacy beliefs indirectly. In the study, a conceptual model was suggested for the relationship between the variables (science literacy of science teacher candidates, their epistemological beliefs and their selfefficacy beliefs in science teaching). It was observed that the variable of the belief that there is only one correct was correlation inefficient with independent and the other variables, and this variable wasn’t taken into the model.The main aim of science teaching is to grow scientific literate individuals. Growing individuals who can decide when faced with a problem on science, who can make suggestions on scientific discussions, know the basic concepts on science and can use these concepts for the problems in their lives, is rather important for developing societies (Çepni, Bacanak and Küçük, 2003). Although scientists make a lot of definitions about science literacy, a common definition can’t be developed. Yet, in general terms, a common and basic concept can be mentioned for the aims in science teaching in the schools (Deboer, 2000; Robert, 1983; as cited in Laugksch, 2000). The other variable beside science literacy that effects the scientific development of individuals is epistemological beliefs. These beliefs are about the nature of knowledge and learning but not a strict philosophical perception (Schommer, 1998). An extended definition of epistemological belief is the subjective beliefs about what knowledge is and how knowing and learning ocur. There is a consistent relation between epistemological beliefs and learning and studying strategies of the students. The students who have improved beliefs in learning process use more qualified learning strategies and can be more successful in terms of academic than the students who have unimproved epistemological beliefs (Deryakulu, 2004). The characteristics that science teachers must have are science literacy and developed epistemological and self-efficacy beliefs. Selfefficacy is one of the concepts that came to forefront in Bandura’s social learning theory. Bandura firstly mentioned about this concept in his work "Self-efficacy: Toward a Unifying Theory of Behavioral Change" in 1977 (Bıkmaz, 2004; Pajares, 2002). Self-efficacy is an important qualification for forming behaviours and individual’s judgement about himself for making necessary organisations and sustaining successfully (Bandura, 1997). The self-efficacy in science education can be defined as the judgment about the students’ own skills for participation, motivation and as a result the success of them and effective and fertile science teaching of teachers (Küçükyılmaz and Duban, 2006). So, the low or high selfefficacy of the teachers may affect the applications of the teachers in science education (Schriver and Czerniak, 1999). In the study, the self-efficacy belief in science education is very important in terms of science teacher candidates. The teacher candidates with high self-efficacy beliefs will be more determined, knowing what he wants and what he must do. In this study, the science literacy of the teacher candidates were explained in terms of the variables, epistemological belief and self-efficacy belief for science teaching, and the reasons and results were discussed comparatively CONCEPTUAL FRAMEWORK It can be said that, as well as teachers’ personal characteristics, their educational process at faculty of education has a great role for their scientific literacy. That is why many research has been done about the level of scientific literacy on science teacher candidates. It can be thought that scientific literacy of science teacher candidates is influenced by epistemological beliefs and self-efficacy beliefs. Yet, there is no research in literature. Searching how science candidate teacher’s scientific literacy is predicted by their epistemological beliefs and self efficacy beliefs on science teaching has a crucial role on teacher’s education, correcting deficiencies in science teaching field and enlighting for further studies. METHOD In this study, casual-comparative research method, one of the scanning models, was used. Casual-comparative researches are for determining the reasons of a natural situation or event and the variables affecting these reasons or the results of an effect This method was used for explaining the science literacy of teacher candidates depending on the epistemological belief and the variables of self-efficacy belief for science teaching. Data Collection Tools The Scale of Basic Science and Technology Literacy: This scale used in the study was developed by Laugksch and Spargo (1996) and adapted to Turkish by Yetişir (2007). The Scale of Self-Efficacy Belief on Science Teaching: This scale was developed by Riggs and Enochs (1990) and adopted to Turkish by Özkan, Tekkaya and Çakıroğlu (2002). The Scale of Epistemological Belief: This scale was developed by Schommer (1990) and adopted to Turkish by Deryakulu and Büyüköztürk (2002). FINDINGS In the study, a conceptual model was suggested for the relationship between the variables (science literacy of science teacher candidates, their epistemological beliefs and their self-efficacy beliefs in science teaching). It was observed that the variable of the belief that there is only one correct was correlation inefficient with independent and the other variables, and this variable wasn’t taken into the model. The coefficient of concordance for ultimate model was quite above the accepted limit. As CFI, IFI, NFI, TLI and RFI values are over .90 and RMSA is under .05, the model can be evaluated as good fitted (Şimşek, 2007). The fit indices are high in the study, so it can be said that the model has high fit.Discussion for Direct Effects In generally, it can be stated that the epistemological beliefs of science teacher candidates influences the science literacy directly As a result of the study, it was found that the belief that learning depends on effort, the sub-dimension of epistemological belief, influenced the nature of science, the sub-dimension of science literacy, directly. Accordingly, it can be said that the more the level of the belief that learning depends on effort in terms of science teacher candidates increases, the more the level of science literacy for the nature of science increases. The other dimension that explains the nature of science is the belief that learning depends on skill. It was observed that the more the belief that learning depends on skill increased, the more science literacy level for the nature of science increased. But this effect is lower than the dimension of the belief that learning depends on effort. The nature of science includes the concepts of reason and imagination, too. When it is thought that the people who have learning skills can use their reason and imagination more effectively, the relationship between the belief level that learning depends on skill and the perception of the nature of science can be explained. The other independent variable of the study is the impact of science and technology on society, the sub-dimension of science literacy. The first variable influencing it directly is the belief that learning depends on effort. The more the belief that learning depends on effort increases, the more the perception level for the impact of science and technology on society increases The other variable that influences the impact of science and technology on society, the sub-dimension of science literacy, directly is the belief that learning depends on skill. But the relationship between them isn’t a powerful relation. The reason may be higher beliefs of individuals that learning depends on effort than the belief depends on skill. The last of independent variable in the model is the science content knowledge, the sub-dimension of science literacy. The first variable that influences the science content knowledge is the belief that learning depends on effort. It is understood that the more the belief that learning depends on effort increases, the more science content knowledge increases. This relation is one of the highest relations in the model. The reason may be that the students with developed beliefs that learning depends on effort are generally the students who study systematically and are successful, and the science content knowledge of these students is developed The other variable that influences the science content knowledge is the belief that learning depends on skill, but like the other independent variable, its degree of influence is lower than the belief that learning depends on effort. Still, there is a relationship between the belief that learning depends on skill and the science content knowledge. The reason of this may be that the individuals link their success to learning effects or intelligence. Discussion for Indirect Effects At the end of the research, it was observed that the self-efficacy beliefs of science teacher candidates influenced the science literacy indirectly from epistemological beliefs. When the search model was examined, it was noticed that the personal self-efficacy belief in science teaching influenced the nature of science from the belief that learning depends on effort. It is understood from the model that the personal self-efficacy belief in science teaching influences the impact of science and technology on society through the belief that learning depends on effort. Science can be said to be related to environment, daily life and nature, so the individuals with high personal self-efficacy belief are expected to have high perception on the impact of science and technology on society. Also, it is noticed from the model that the personal self-efficacy belief in science teaching influences the science content knowledge from the belief that learning depends on effort Personal self-efficacy belief influences the nature of science from the belief that learning depends on skill. From the model, it can be understood that personal self-efficacy belief on science teaching influences the impact of science and technology on society through the belief that learning depends on skill. But this effect is weaker than the effect of belief that learning depends on effort. The reason of this may be that the teacher candidates’ having higher beliefs that learning depends on effort It is occurred that personal self-efficacy belief in science teaching influences the science content knowledge through the belief that learning depends on skill. This effect is weaker than the effect that occurs through the belief that learning depends on effort because it may be that individuals think that effort is more effective than skill in term of reaching the knowledge. The other variable in the model is the outcome expectation in science teaching. It is the weakest effective variable for science literacy. The outcome expectation in science teaching influences the impact of science and technology on society through the beliefs that learning depends on effort and skill. Although this effect is higher than affecting the nature of science, it is still weak. Also, the outcome expectation in science teaching influences the science content knowledge through the beliefs that learning depends on effort and skill, and it is again a weak effect. This may be because the effects of science content knowledge and science and technology on society are better perceived by the teacher candidates than the nature of science In the study carried out for the science teacher candidates, the following conclusions some of which are compatible to the literature and some are different were arrived: The Results for Direct Effects As a result of the study, the belief that learning depends on the effort is identified as the variable that directly influences the nature of science. According to this result, the more the teacher candidates have increasing level of beliefs that learning depends on the effort, the higher their perceptions related to the nature of science are. Besides, the other variable that influences the nature of science is the belief that learning depends on skill. Accordingly, the more the teacher candidates have increasing level of beliefs that learning depends on the skill, the higher their perceptions related to the nature of science are. But its effectiveness is weaker than the belief that learning depends on effort. The other independent variable is the impact of science and technology on society, and the belief that learning depends on effort influences this variable directly. Accordingly, the more the belief that learning depends on effort increases, the more perception levels of the teacher candidates on the impact of science and technology on society increases. Another variable that influences the impact of science and technology on society directly is the belief that learning depend on skill. Accordingly, the more the belief that learning depends on skill increases, the more the perception levels of individuals on the impact of science and technology on society increases, but this effect is weaker than the effect of belief that learning depends on effort. The other independent variable of the study is the science content knowledge. The belief that learning depend on effort influences this variable directly, so the more the belief that learning depends on effort increases, the more the science content knowledge of the teacher candidates increases. This direct effect is the most powerful effect, too. The other variable that influences the science content knowledge directly is the belief that learning depends on skill. Accordingly, the more the belief that learning depends on skill increases, the more the science content knowledge of the teacher candidates increases. The Results for Indirect Effects As a result of the study, personal self-efficacy belief in science teaching influences indirectly the nature of science, the impact of science and technology on society and the science content knowledge through the belief that learning depends on effort. In other words, personal selfefficacy belief in science teaching influences the belief that learning depends on effort directly, and the belief that learning depends on effort influences the nature of science, the impact of science and technology on society and science content knowledge directly. As a result of the study, personal self-efficacy belief in science teaching influences indirectly the nature of science, the impact of science and technology on society and the science content knowledge through the belief that learning depends on skill. In other words, personal self-efficacy belief in science teaching influences the belief that learning depends on skill directly, and the belief that learning depends on skill influences the nature of science, the impact of science and technology on society and science content knowledge directly. But this effect is weaker than the mediator variable, the belief that learning depends on effort. In the study the other indirect effect is the outcome expectation. It influences the nature of science, the impact of science and technology on society and science content knowledge indirectly through the belief that learning depends on effort and the belief that learning depends on skill. This effect is the weakest in the model. This is because there is a weak relationship between the belief that learning depends on skill and the outcome expectation in science teaching. In the study, a model was formed, which tries to explain the science literacy through epistemological beliefs and self-efficacy beliefs in science teaching. It is concluded that science literacy can be explained by epistemological beliefs and self-efficacy beliefs. The study is limited to the study group. The repeat of the study in other universities can contribute to the generalization of results. Also, epistemological beliefs and selfefficacy belief was used as the descriptors of science literacy in the study. This may be the constraint of the study. In this context, to do the study again by adding variables like motivation and self-regulation may be helpful for healthier knowledge on this subject. Casual-comparative research model, one of scanning models, was used in this study, but it may contribute to use different experimental and qualitative studies for the generalization of the results. Also, the study was conducted only for the science teacher candidates. If the same study was conducted for class teachers, it may contribute to the results. It is concluded that the science literacy of science teacher candidates can be explained by epistemological beliefs and self-efficacy beliefs in science teaching. In consideration of these results, some measures may be taken and the science literacy of teacher candidates may be increased. Besides, some studies to increase the epistemological beliefs and self-efficacy beliefs in science teaching can be made for raising the science literacy level of students