Introduction Biggs (1987) defines approaches to learning as a concept measured by inventories that expresses how students lead off the learning process. Other definitions state that approaches to learning involve a number of strategies used to fulfill the task and create motivation for learning (Prosser & Trigwell, 1999; Kember, Biggs & Leung, 2004). Studies about students’ approaches to learning started at Gothenburg University in Sweden in the 1970s. This pioneer study used the qualitative research method to determine approaches to learning by giving students the task of reading a scientific article and then evaluating their comprehension levels (Marton, 1975; Marton & Saljo, 1976). According to this research, students’ approaches to learning are divided in two according to their comprehension levels. If students have understood the reading at a high level, they have a deep approach to learning. If students have understood at a low level, they have a surface approach to learning. In other research, Ramsden (1979) found a third approach to learning, which he named strategic approach. A deep approach to learning is based on meaning search and making of student while handling learning procedure. Students who prefer the deep approach to learning pursue a goal of comprehension, deal with the structure of learning a task, correlate theoretical ideas with daily experiences, structure content by converting it into a consistent whole, actively participate in the learning process and use their metacognition skills (Biggs, 1987; Ramsden 2000). The deep approach to learning also includes the effort to correlate previous knowledge and experiences with new knowledge and the analysis of accuracy of knowledge. Someone who adopts the deep approach to learning analyses the meaning of the thing he studies more deeply (Entwistle & Ramsden, 1983). On the other hand, the surface approach to learning is based on exterior motivation or fear of failure. According to students who use the surface approach to learning, the aim of learning is to get a profession, make their family happy or to prevent boredom. Minimum time and effort are spent to meet their basic needs. Repeating and memorizing information without meaning are most common strategies in the surface approach to learning (Entwistle & Ramsden, 1983; Biggs, 1991, 1993). The priority of students who adopt the strategic approach is to get the highest grade; therefore they use both deep and surface approaches to learning and they have competitor and vocational motivation (Ramsden, 1979). The purpose of this study is to determine the students’ preference levels of approaches to learning and whether there is a significant difference in preference levels regarding the department, gender and class rank of students in preschool and classroom teaching environments.

Method The research uses a survey model. The universe of this research comprises students studying preschool and primary school teaching at the Kazım Karabekir Education Faculty of Atatürk University in spring of the 2011-2012 academic year. The research sample includes 330 candidate teachers. The Approaches and Study Skills Inventory for Students (ASSIST) was used to obtain research data. The instrument was developed by Tait, Enwistle and McCune (1998) and Senemoğlu (2011) carried out its adoption to Turkish, and validity and reliability studies. The inventory is made up of four chapters and 67 items on a five-point Likert-type scale. The first chapter contains 52 items composed of three dimensions of deep, strategic and surface learning. This research used the first chapter of the inventory. The Cronbach’s Alpha reliability coefficient of measurement based on the original form of the inventory varies between 0.71 and 0.81 for the whole inventory and sub-inventories. The Cronbach’s Alpha reliability coefficient is calculated as 0.73 for the deep approach to learning, 0.85 for the strategic approach to learning, 0.74 for the surface approach to learning and 0.84 for the whole inventory. The required critical alpha value is generally .70 or more for a study’s internal consistency level to be accepted as ideal (Özdamar, 2004). In this case, it can be inferred that measurements are reliable. The total grade averages of deep and surface approaches to learning have been complemented to 100 to compare the obtained data. Afterward, arithmetic average and standard deviation values were calculated based on total grades to determine the students’ preference levels of approaches to learning. A one-way MANOVA was carried out for each independent variable to determine whether there was a significant difference between students’ approach to learning preference levels and their department; gender and class rank independent variables. MANOVA was used because there is a theoretical connection between the approaches to learning (deep, strategic, surface) as dependent variables. However, the use of MANOVA requires meeting some basic assumptions. Therefore, before the data analysis and comment on the findings, researchers analyzed whether these assumptions were met. The findings proved that questions whose answers are sought can be answered using MANOVA, a multivariate parametric test. In situations in which a significant difference was found, approach to learning was analyzed separately to determine in which approach or approaches to learning there is a difference and for the purpose of control. Data had a (0.05/3) .017 significance level using the Bonferroni correction. Findings According to the results, the primary and preschool candidate teachers preferred deeper and strategic approaches to learning close to the level and the surface approach to learning on a lower level. No significant difference was found between approach to learning preferences of preschool and primary school candidate teachers. That there is no significant difference between approaches to learning preference levels is an expected result because the curriculum and characteristics in preschool and primary school teaching departments were similar. When the candidate teachers’ approach to learning preferences were analyzed according to gender variable, a significant different was found in surface learning. No significant difference was observed between female and male candidate teachers in deeper and strategic approaches to learning. Male students preferred the surface approach to learning at a more significant level than female candidate teachers. According to the class rank variable, only in surface learning, a significant difference was found only in surface learning. No significant difference was observed in the deep and strategic approaches to learning in terms of the class rank variable. In the surface approach to learning, a significant difference was observed in favor of freshman and sophomore between freshman and sophomore and senior. Discussion, Conclusions and Recommendations Students in preschool and classroom teaching departments prefer deep and strategic approaches to learning within close to the level and surface approach to learning in a low level. Each of the three approaches to learning is preferred within a percentage of more than six out of a hundred. These results overlap with the research carried out by Senemoğlu (2011) and in which the approaches to learning of Turkish and American students are determined and the same inventory is used. In her study in which she determined university students’ approaches to learning, Ekinci (2009) also found that grade averages related to students’ approaches do not seem high or low enough to qualify them as dominant in terms of any approach to learning. She examined this case because universities have features directing both deep and surface learning, but no dominant directive for deep learning. No significant difference was found between the approach to learning preferences of students in preschool and primary school teaching. That there is no significant difference between approaches to learning preference levels is an expected result because students’ curriculum and characteristics in preschool and primary school teaching departments, which are both in primary school departments, are similar. A significant difference was found between the the surface approach to learning and the gender variable. No significant difference was found between girls and boys in the deep and strategic approaches to learning. Male students preferred the surface approach to learning more than female students at a significant level. Senemoğlu (2011) observed a significant difference between male and female students in the strategic and surface approaches. She found that female students preferred the surface and strategic approach to learning more than male students at a significant level. Senemoğlu (2011) has not been able to reach a significant difference in any approach to learning between American students in terms of gender. Selçuk, Çalışkan and Erol (2007) also have not been able to observe a significant difference in terms of gender in their studies in which they analyze approaches to the learning of physics teacher candidates. Differences between learning-teaching environments can arise from other variables. According to these results, it can be inferred that the gender variable is not very important in approach to learning preferences of students, but female students are more focused on getting high grades than male students, and female students frame their approaches to learning in accordance with this aim, even though there is no significant difference. Only in the surface approach to learning has a significant difference been observed in the preference of approach to learning according to class rank. The difference is in favor of freshmen and sophomores between freshman and sophomore and senior. Therefore, the higher their class rank, the less students prefer surface approaches to learning. Selçuk, Çalışkan and Erol (2007) reached the same results. This case shows that students move away from memorizing as their class rank gets higher. However significant this result, the fact that there is no significant increase in the deep approach to learning shows that students do not use learning based on comprehension, giving the meaning more significance. Ekinci (2009) also revealed that there is no statistically significant difference between the average grades of freshmen and seniors when the average related to learning. When approach is compared to class rank, these results show that the teaching-learning environment in universities does not influence an increase of deep learning orientation over time. However, the deep learning features of universities are required to be activated and students’ orientation is expected to increase as long as they are in an academic environment. These results overlap with the results of other studies carried out around the country (Ekinci & Ekinci, 2007; Senemoğlu, 2011). According to these results, the following recommendations can be given: 1) in teachers’ education, more learning environments that will persuade students to choose deep learning environments can be created; 2) repetition of the study in different samples can be useful when it is thought that great numbers of variables and cultural factors are effective in approaching students’ learning; and 3) research can be carried to determine why students in the education faculty do not prefer the deep approach to learning more as their class ranks increase; and 4) by forming different learning environments, experimental studies analyzing their effects on approaches to learning can be carried out.