349 © 2019 by the authors; licensee Asian Online Journal Publishing Group Asian Journal of Education and Training Vol. 5, No. 2, 349-361, 2019 ISSN(E) 2519-5387 DOI: 10.20448/journal.522.2019.52.349.361 © 2019 by the authors; licensee Asian Online Journal Publishing Group Which Country is More Effective in Science Teaching? Evidence from PISA 2015 as a Secondary School Assessment Tool Gökhan Ilgaz1 Menekşe Eskici2 Levent Vural3 ( Corresponding Author 1,3Faculty of Education, Educational Sciences Department, Dr. Trakya University, Turkey 2Faculty of Science and Art, Educational Sciences Department, Dr. Kırklareli University, Turkey Abstract The aim of this study is to determine how effectively different countries use educational inputs in the process of science education. The study is in the descriptive model and the data is derived from the PISA 2015 data set, which provides information to countries on secondary education programs. Data from 70 countries, including Turkey, were used in the study. The effectiveness of Turkey in the process of science teaching in secondary education has been compared with other countries and suggestions have been made to increase the effectiveness of secondary education science teaching. The data for this study are grouped into training inputs and training outputs. ―Student behaviour hindering learning”, “teacher behaviour hindering learning”, “shortage of educational material”, “shortage of educational staff”, “professional development”, “teachers participation”, “curricular development”, “total number of science teachers at school”, “index science-specific resources (sum)” were defined as the educational inputs. The science achievements of students have been taken into account in determining which countries have more effective management of educational inputs accepted in this study. Secondary scholl level education inputs while nearby countries respectively to use effectively Slovakia (97.30%), Slovenia (91.16%), Brazil (90.50%), Turkey (89.75%), Finland (84.65%), Greece (83.31%), Denmark ( 83.25%), and Czech Republic (80.61%). education outcomes in Turkey is ranked as 20th in science teaching secrets to use it effectively. Keywords: Secondary school, Curriculums, Effective science teaching, PISA 2015, Data envelopment, Training inputs and training outputs. Citation | Gökhan Ilgaz; Menekşe Eskici; Levent Vural (2019). Which Country is More Effective in Science Teaching? Evidence from PISA 2015 as a Secondary School Assessment Tool. Asian Journal of Education and Training, 5(2): 349-361. History: Received: 12 February 2019 Revised: 20 March 2019 Accepted: 29 April 2019 Published: 24 July 2019 Licensed: This work is licensed under a Creative Commons Attribution 3.0 License Publisher: Asian Online Journal Publishing Group Acknowledgement: All authors contributed to the conception and design of the study. Funding: This study received no specific financial support. Competing Interests: The authors declare that they have no conflict of interests. Transparency: The authors confirm that the manuscript is an honest, accurate, and transparent account of the study was reported; that no vital features of the study have been omitted; and that any discrepancies from the study as planned have been explained. Ethical: This study follows all ethical practices during writing. Contents 1. Introduction .................................................................................................................................................................................... 350 2. Purpose of the Study ..................................................................................................................................................................... 350 3. Method ............................................................................................................................................................................................. 351 4. Findings ........................................................................................................................................................................................... 353 5. Discussion ........................................................................................................................................................................................ 358 References ............................................................................................................................................................................................ 360 http://crossmark.crossref.org/dialog/?doi=10.20448/journal.522.2019.52.349.361&domain=pdf&date_stamp=2017-01-14 http://creativecommons.org/licenses/by/3.0/ http://creativecommons.org/licenses/by/3.0/ http://asianonlinejournals.com/index.php/EDU/article/view/872 https://orcid.org/0000-0001-8988-5279 https://orcid.org/0000-0001-6217-3853 https://orcid.org/0000-0001-9302-6143 http://asianonlinejournals.com/index.php/EDU/article/view/872 https://orcid.org/0000-0001-8988-5279 https://orcid.org/0000-0001-6217-3853 https://orcid.org/0000-0001-9302-6143 http://asianonlinejournals.com/index.php/EDU/article/view/872 https://orcid.org/0000-0001-8988-5279 https://orcid.org/0000-0001-6217-3853 https://orcid.org/0000-0001-9302-6143 http://asianonlinejournals.com/index.php/EDU/article/view/872 https://orcid.org/0000-0001-8988-5279 https://orcid.org/0000-0001-6217-3853 https://orcid.org/0000-0001-9302-6143 http://asianonlinejournals.com/index.php/EDU/article/view/872 https://orcid.org/0000-0001-8988-5279 https://orcid.org/0000-0001-6217-3853 https://orcid.org/0000-0001-9302-6143 Asian Journal of Education and Training, 2019, 5(2): 349-361 350 © 2019 by the authors; licensee Asian Online Journal Publishing Group Contribution of this paper to the literature This study contributes to the existing literature by determining how effectively different countries use educational inputs in the process of science education. 1. Introduction Science is more than merely a part of daily life; it is daily life itself. Subjects such as stirring sugar in tea to make it dissolve faster, the need to take plugs out of sockets during lightning storms, how natural events such as rain and snow occur, and the harm caused to the body by not drinking enough water, all fall under the scope of science. Considered from this point of view, not only is it true that for humans, who lead their lives as part of nature, it is impossible to live apart from science, it is also a fact that if they are not sufficiently equipped with regard to science, their lives will be made equally difficult. The importance of science stems from the fact that it encompasses not only the most basic activities in daily life, but also much more complex subjects such as the production of technology, increasing efficiency in the use of natural resources, and discovering the keys to a longer life. Developments in the field of science and reflecting them into daily life are the most significant indicators of the establishment of modern life in society and of the progress of nations. Therefore, the teaching of science plays a key role in an individual‘s personal development as well as in the advancement of nations. Scientific knowledge is a lesson that is so much a part of life that Davis et al. (2006) stress that a science education that focuses on real challenges is the main reason for success. It can be said that science and technology lessons carried out in a social context change attitudes towards science in a positive way and support the creation of the basis for a solid scientific understanding. The best way to increase students‘ interest in science is to associate the lesson subjects to daily life, since it is possible for students to perceive the benefits of science when they see its reflection in daily life (Bennett et al., 2007; Maltese et al., 2014; Sheldrake et al., 2017). If attitudes towards science become negative and if students‘ concerns about the subject of science affect their career choices and lead them in other directions, this will be a worrying situation for developments in the field of science. Therefore, endearing students to science builds the mainframe for their future achievements related to science (Osborne et al., 2003). It would not be wrong to say that attitudes towards science form the source of skills related to scientific knowledge, technological developments and science. The problem here is that although science is so important, attitudes towards science are low. Conducted studies have shown that attitudes towards science are low because of negative experiences occurring in science lessons (Mulholland and Wallace, 1996; Palmer, 2001). Based on this, it can be said that the science education process needs to be organised in such a way as to affect students‘ attitudes towards science in a positive way. It is argued that the approach to science education should be regarded as ―education through science‖ rather than ―science through education‖ (Holbrook and Rannikmae, 2007). The real sine qua non of science education is that students should learn science by experiencing it and not by rote. The place of the laboratory environment in science education is indisputable. The opportunity for students to learn by doing and experiencing is a great chance for those students‘ science learning (Hodson, 1988; Kirschner et al., 2006). In addition to these, the design of a science education learning environment that includes activities focusing on skills like experimentation and reasoning, that is supported by visuals, and in which primary sources or models are used, is recommended (Evagorou et al., 2015). The need to consider the problem of how science education should be taught, rather than what should be taught, is pointed out. Science education is possible not merely by learning basic knowledge, but with the application of that knowledge. Baeten et al. (2010) state that education conducted with a student-centred approach supports deep learning. Success in science education is related to experiencing scientific knowledge by doing research (Clermont et al., 1994; Bybee, 2014; Osborne, 2014). Moreover, there are indicators regarding the fact that conducting science education in a collaborative way will contribute to students‘ cognitive development as a result of their mutual interaction and sharing of ideas (Koretsky et al., 2019). It is emphasised that learning and teaching science lessons in a more collaborative way will be beneficial (Huppert et al., 2002). Controversial findings related to the factors influencing success in science are found in the literature. In a study made by Stohr‐Hunt (1996) it was found that success in science classes of students who had more experience was greater than that of students with less experience. In a study conducted by Cairns and Areepattamannil (2019) it was determined that based on the PISA data, inquiry-based science education was significantly negatively correlated with science success, whereas inquiry-based science education was significantly positively correlated with dispositions towards science such as interest in and enjoyment of learning science, instrumental and future- oriented motivation for science, and science self-esteem. Using the TIMSS 2015 data for Norway, Teig et al. (2018) concluded in their study that inquiry-based science teaching on a large scale had a negative effect on science achievement, a fact which supports the findings made by Cairns and Areepattamannil (2019). Undoubtedly, there are various factors affecting academic success. The basic factors that affect students‘ academic success are considered to be teacher competencies (Rockoff, 2004; Aaronson et al., 2007; Rothstein, 2010; Harris and Sass, 2011) undesirable student behaviours (Durán-Narucki, 2008; Maxwell, 2016) school facilities (Gislason, 2010; Rivera and Lopez, 2019) and the school‘s relationship with parents (Hill and Taylor, 2004; Jeynes, 2007). The same factors can be listed as determinants of success in the subject of science. In the PISA 2015 report, the educational inputs affecting science achievement were defined as “student behaviours preventing learning”, “teacher behaviours preventing learning”, “lack of educational materials”, “lack of teaching staff”, “professional development”, “teacher participation”, “curriculum development”, “total number of science teachers in school‖ and ―science-specific resource index (total)‖. 2. Purpose of the Study In an age in which development in science technology is taken as the basis as the most important indicator of countries‘ development, the process of scientific knowledge production in universities has also gained importance. The education of individuals who can produce scientific knowledge is also of equal importance. When considering Asian Journal of Education and Training, 2019, 5(2): 349-361 351 © 2019 by the authors; licensee Asian Online Journal Publishing Group that the most determining factor for success in the education process is readiness, it can be regarded as an undisputable fact that possessing competence in science in the secondary education period has an impact on educating individuals who will be able to produce scientific knowledge in higher education. Based on this, it can be concluded that the effectiveness of science teaching carried out during the secondary education process is very important for the development of nations. The more successfully science education is conducted during the secondary education process and the higher the level of readiness of students sent into higher education, the more the levels of countries‘ scientific knowledge production and of their technological development will increase. Therefore, it is considered important to determine the effect on success of the educational inputs that will determine the effectiveness of science teaching in secondary school curricula. The aim of this study is to determine how efficiently different countries use educational inputs in the science teaching process. 3. Method 3.1. Research Model With the aim of evaluating the relative efficiency of the inputs or efforts put in by countries, using these countries‘ PISA 2015 Science scores as outputs, this study was designed according to the survey model. 3.2. Study Group The study was conducted with data obtained from schools in 70 countries/regions that participated in the PISA 2015 study. For the analysis, in order to remove negative numbers and to facilitate interpretation, the data were converted into z scores. Since there were outliers in the input scores for Tunisia and China (Beijing-Shanghai- Jiangsu-Guangdong), these were removed. It was seen that there were no outliers in the z scores obtained for the remaining 68 countries. In the correlation analysis conducted, it was determined whether or not there was multicollinearity, and the study was carried out with 68 countries. The distributions of the countries included in the study are shown in Table 1. Table-1. Countries and numbers of schools included in the study. Countries Frequency % Countries Frequency % Albania 230 1.32 Latvia 250 1.43 Algeria 161 0.92 Lithuania 311 1.78 Australia 758 4.34 Luxembourg 44 0.25 Austria 269 1.54 China (Macao) 45 0.26 Belgium 288 1.65 Malta 59 0.34 Brazil 841 4.81 Mexico 275 1.57 Bulgaria 180 1.03 Moldova 229 1.31 Canada 759 4.34 Montenegro 64 0.37 Chile 227 1.30 Netherlands 187 1.07 Chinese-Taipei (Taiwan) 214 1.22 New Zealand 183 1.05 Colombia 372 2.13 Norway 229 1.31 Costa Rica 205 1.17 Peru 281 1.61 Croatia 160 0.92 Poland 169 0.97 Czech Republic 344 1.97 Portugal 246 1.41 Denmark 333 1.91 Qatar 167 0.96 Dominican Republic 194 1.11 Romania 182 1.04 Estonia 206 1.18 Russian Federation 210 1.20 Finland 168 0.96 Singapore 177 1.01 France 252 1.44 Slovakia 290 1.66 Georgia 262 1.50 Vietnam 188 1.08 Germany 256 1.46 Slovenia 333 1.91 Greece 211 1.21 Spain 201 1.15 China (Hong Kong) 138 0.79 Sweden 202 1.16 Hungary 245 1.40 Switzerland 227 1.30 Iceland 124 0.71 Thailand 273 1.56 Indonesia 236 1.35 Trinidad and Tobago 149 0.85 Republic of Ireland 167 0.96 United Arab Emirates 473 2.71 Israel 173 0.99 Turkey 187 1.07 Italy 474 2.71 Macedonia 106 0.61 Japan 198 1.13 United Kingdom 550 3.15 Jordan 250 1.43 United States 177 1.01 Korea 168 0.96 Uruguay 220 1.26 Kosovo 224 1.28 Spain (Regions) 976 5.59 Lebanon 270 1.55 Argentina (Buenos Aires) 58 0.33 Source: Obtained from primary data. 3.3. Data Collection Tools The data for the schools taken as the study sample were utilized by utilizing the PISA 2015 database. These data were scaled in ready form. Within this scope, the means of these scales were used as inputs, while the countries‘ PISA 2015 Science success scores were used as outputs. These inputs and the questions with which they were obtained are presented in Table 2. Asian Journal of Education and Training, 2019, 5(2): 349-361 352 © 2019 by the authors; licensee Asian Online Journal Publishing Group Table-2. The inputs and its questions. Inputs The Name of Inputs Questions of Inputs Questions Input1 Student behaviour hindering learning (WLE) SC061Q01TA Student truancy SC061Q02TA Students skipping classes SC061Q03TA Students lacking respect for teachers SC061Q04TA Student use of alcohol or illegal drugs SC061Q05TA Students intimidating or bullying other students Input2 Teacher behaviour hindering learning (WLE) SC061Q06TA Teachers not meeting individual students‘ needs SC061Q07TA Teacher absenteeism SC061Q08TA Staff resisting change SC061Q09TA Teachers being too strict with students SC061Q010TA Teachers not being well prepared for classes Input3 Shortage of educational material (WLE) SC017Q05NA A lack of educational material (e.g. textbooks, IT equipment, library or laboratory material). SC017Q06NA Inadequate or poor quality educational material (e.g. textbooks, IT equipment, library or laboratory material). SC017Q07NA A lack of physical infrastructure (e.g. building, grounds, heating/cooling, lighting and acoustic systems). SC017Q08NA Inadequate or poor quality physical infrastructure (e.g. building, grounds, heating/cooling, lighting and acoustic systems) Input4 Shortage of educational staff (WLE) SC017Q01NA A lack of teaching staff SC017Q02NA Inadequate or poorly qualified teaching staff SC017Q03NA A lack of assisting staff. SC017Q04NA Inadequate or poorly qualified assisting staff Input5 Professional development (WLE) SC009Q06TA When a teacher has problems in his/her classroom, I take the initiative to discuss matters. SC009Q08TA I pay attention to disruptive behaviour in classrooms SC009Q12TA When a teacher brings up a classroom problem, we solve the problem together. SC009Q06TA When a teacher has problems in his/her classroom, I take the initiative to discuss matters. Input6 Teachers participation (WLE) SC009Q09TA I provide staff with opportunities to participate in school decision-making SC009Q010TA I engage teachers to help build a school culture of continuous improvement SC009Q011TA I ask teachers to participate in reviewing management practices. Input7 Curricular development (WLE) SC009Q01TA I use student performance results to develop the school‘s educational goals. SC009Q02TA I make sure that the professional development activities of teachers are in accordance with the teaching goals of the school. SC009Q03TA I ensure that teachers work according to the school‘s educational goals Input8 Total number of science teachers at school Input9 Index science specific resources (Sum) SC059Q01NA Compared to other departments, our school's science department is well equipped. SC059Q02NA If we ever have some extra funding, a big share goes into improvement of our science teaching. SC059Q03NA Science teachers are among our best educated staff members. SC059Q04NA Compared to similar schools, we have a well-equipped laboratory. SC059Q05NA The material for hands-on activities in science is in good shape. SC059Q06NA We have enough laboratory material that all courses can regularly use. SC059Q07NA We have extra laboratory staff who help support science teaching. SC059Q08NA Our school spends extra money on up-to-date science equipment. Source: Obtained from primary data. 3.4. Data Analysis The aim of this study was to investigate whether or not the inputs or efforts considered to affect the PISA 2015 Science success scores were used efficiently by the countries. With this aim, to determine the relative efficiency, data envelopment analysis was used. The analysis was performed in line with the views of Lorcu (2008): 1. Firstly, the Decision Making Units (DMU) were determined. According to Dyson et al. (2001)) form inputs and s outputs, the number of decision making units should be 2 mXs. In this study, 70 countries (68 were taken for analysis; the reason for this is presented below) were sufficient as DMUs for the 9 inputs and 1 output. 2. According to Cooper et al. (2001) form inputs and s outputs, there should be an N number of DMUs, where N≥{m x s; 3 X (m + s). Accordingly, since 70 (68) ≥{9 X 1; 3 X (9 + 1)}=70(68) ≥{9; 30}, the number of DMUs was sufficient. 3. z distribution should be controlled. In the first z distribution in the study, since there were outliers in the input scores for Tunisia and China (Beijing-Shanghai-Jiangsu-Guangdong), these were removed. It was seen that there were no outliers in the z scores recalculated for the remaining 68 countries. Both z distributions are shown in Table 3. To facilitate interpretation of the scores and to remove negative values, conversion to T scores was performed. Asian Journal of Education and Training, 2019, 5(2): 349-361 353 © 2019 by the authors; licensee Asian Online Journal Publishing Group Table-3. Distribution of inputs for 68 countries. Inputs N Minimum Maximum N Minimum Maximum Student behaviour hindering learning (WLE) 70 -1,87272 2,06593 68 -1,85571 2,05420 Teacher behaviour hindering learning (WLE) 70 -2,07622 2,43954 68 -2,07918 2,52854 Shortage of educational material (WLE) 70 -1,96928 2,12009 68 -1,96773 2,21297 Shortage of educational staff (WLE) 70 -2,12386 3,58366 68 -2,26788 2,37804 Professional development (WLE) 70 -2,35916 2,90058 68 -2,47614 2,14454 Teachers participation (WLE) 70 -2,09584 2,42160 68 -2,06065 2,40872 Curricular development (WLE) 70 -2,30067 2,18401 68 -2,45245 2,21827 Total number of science teachers at school 70 -2,95748 1,84587 68 -3,08392 1,85672 Index science specific resources (Sum) 70 -1,13323 5,53576 68 -1,43348 2,86126 Source: Obtained from primary data. 4. Correlation control was performed for the 68 countries. No high correlation was determined. The correlation results are presented in Table 4. Table-4. Correlations between inputs. No Inputs Input 2 Input 3 Input 4 Input 5 Input 6 Input 7 Input 8 Input 9 PISA Score Input1 Student behaviour hindering learning (WLE) -,107 ,171 -,43** -,038 -,013 -,113 ,104 ,415** ,331** Input2 Teacher behaviour hindering learning (WLE) ,475** ,163 ,227 ,011 ,079 -,021 -,190 ,086 Input3 Shortage of educational material (WLE) ,058 ,495** -,099 -,154 -,107 ,025 ,187 Input4 Shortage of educational staff (WLE) ,534** ,168 ,120 -,004 -,189 -,403** Input5 Professional development (WLE) -,058 -,067 -,201 ,032 ,004 Input6 Teachers participation (WLE) ,633** ,654** ,033 -,592** Input7 Curricular development (WLE) ,671** -,224 -,491** Input8 Total number of science teachers at school ,038 -,481** Input9 Index science specific resources (Sum) ,268* Source: Obtained from primary data. **. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed). 5. Starting with the hypothesis that as decision-makers, we would have an effect on the inputs, analysis of the inputs was made. Moreover, under the hypotheses of constant returns and returns to scale used in approaches of this type, an attempt was made to obtain the countries‘ total, technical and scale efficiency values. Total efficiency was calculated with the input-based CCR model developed by Charnes et al. (1978) while technical efficiency was calculated with Banker et al. (1984) BCC model and scale efficiency was calculated with the ratio of these. Furthermore, returns to scale situations were also calculated. 4. Findings In the study, the research results for total efficiency calculated with the CCR model, technical efficiency with the BCC model, scale efficiency with the CCR/BCC ratio, and returns to scale with total lambdas of the selected countries are presented in Table 5. According to the analyses made, a total of 15 countries/regions (Canada, Switzerland, Estonia, Finland, Hong Kong (China), Hungary, Iceland, Japan, Korea, Lithuania, Macao (China), Poland, Singapore, Slovenia and Vietnam) out of the 68 countries/regions were totally efficient. Of the countries that were totally efficient, the top three countries shown as references were Japan (37), Poland (35) and Estonia (34). Vietnam, which was a totally efficient country, was not shown as a reference for any other country. On the other hand, the totally efficient Slovenia was shown as a reference for Latvia and Sweden. Latvia‘s reference coefficient was extremely low, and, due to rounding down, was evaluated as zero. When the general distribution of totally efficient countries is examined, it is seen that they mostly consist of continental European and Asia-Pacific countries. 12 of the countries (which countries) that were above the OECD average in the PISA 2015 science scores listing were also totally efficient. It was determined that the five countries that were closest to total efficiency (Taiwan (China), the Netherlands, Austria, Denmark and France) were also among the countries above the OECD average. The analysis revealed that 53 countries were inefficient. The five countries furthest from total efficiency were Jordan, Kosovo, Trinidad and Tobago, Algeria and the Dominican Republic. Again, it is seen that these countries occupied the last places in the PISA 2015 science success listing, and that in fact, the Dominican Republic was in last place for both total efficiency and academic success. In other words, these countries‘ efforts were not sufficient. Examining the countries‘ returns to scale situations, it is seen that apart from New Zealand, all countries had increasing returns to scale. That is to say, an input of 1 unit in their inputs resulted in an output greater than 1 in their outputs. Examining the recommendations made for countries that were not totally efficient in order for them to become efficient, it was determined that there was no reference indicator in regional terms. In most reference indicators, there was a mixture of Europe and Asia-Pacific. It was determined that especially Canada and totally efficient countries in Europe were mostly references for their neighbours. Turkey was one of the countries that were not totally efficient. Turkey‘s total efficiency ratio was 71.92%. With this score, Turkey was in 51st place in the list of 68 countries. Estonia and Japan were references for Turkey. That is, for Turkey to be totally efficient, it needed to use its inputs with the input ratios of those countries. Asian Journal of Education and Training, 2019, 5(2): 349-361 354 © 2019 by the authors; licensee Asian Online Journal Publishing Group Table-5. CCR results. NO DMU CCR Country1 λ1 Country2 λ2 Country 3 λ3 Country4 λ4 Country 5 λ5 Country 6 λ6 Total λ Returns to Scale 1 Albania 0,8635 Hungary 0,37 Japan 0,08 Korea 0,08 Singapore 0,19 0,72 Increasing 2 United Arab Emirates 0,7140 Japan 0,15 Korea 0,12 Singapore 0,4 0,67 Increasing 3 Australia 0,9040 Canada 0,24 Estonia 0,24 Iceland 0,01 Poland 0,12 Singapore 0,31 0,92 Increasing 4 Austria 0,9666 Switzerland 0,21 Estonia 0,29 Finland 0,32 Japan 0,01 Singapore 0,07 0,9 Increasing 5 Belgium 0,8640 Switzerland 0,16 Japan 0,2 China (Macao) 0,11 Poland 0,49 0,96 Increasing 6 Bulgaria 0,8621 Japan 0,04 Poland 0,76 0,8 Increasing 7 Brazil 0,6502 Estonia 0,03 Finland 0,21 Japan 0,3 Poland 0,04 0,58 Increasing 8 Canada 1,0000 Canada 1 1 9 Switzerland 1,0000 Switzerland 1 1 10 Chile 0,8566 Canada 0,11 Iceland 0,59 Poland 0,08 Singapore 0,06 0,84 Increasing 11 Colombia 0,5974 Estonia 0,59 Japan 0,03 0,62 Increasing 12 Costa Rica 0,6396 Estonia 0,46 Japan 0,17 0,63 Increasing 13 Czech Republic 0,9004 Estonia 0,34 Finland 0,16 Japan 0,02 Poland 0,39 0,91 Increasing 14 Germany 0,8775 Switzerland 0,15 Estonia 0,17 Finland 0,38 Japan 0,1 Singapore 0,13 0,93 Increasing 15 Denmark 0,9596 Switzerland 0,14 Estonia 0,26 Finland 0,01 Japan 0,02 Poland 0,5 Singapore 0,04 0,97 Increasing 16 Dominican Republic 0,3872 Estonia 0,27 Japan 0,04 Poland 0,05 0,36 Increasing 17 Algeria 0,5164 Estonia 0,49 0,49 Increasing 18 Spain 0,8661 Estonia 0,18 Hungary 0,1 Japan 0,18 Korea 0,14 Poland 0,19 Singapore 0,12 0,91 Increasing 19 Estonia 1,0000 Estonia 1 1 20 Finland 1,0000 Finland 1 1 21 France 0,9446 Switzerland 0,56 Japan 0,01 China (Macao) 0,36 0,93 Increasing 22 United Kingdom 0,9280 China (Hong Kong) 0,46 Japan 0,03 Korea 0,17 Poland 0,21 Singapore 0,08 0,95 Increasing 23 Georgia 0,7821 Estonia 0,04 Hungary 0,18 Korea 0,33 Lithuania 0,03 Poland 0,08 0,66 Increasing 24 Greece 0,8704 Estonia 0,16 Japan 0,03 Korea 0,59 0,78 Increasing 25 China (Hong Kong) 1,0000 China (Hong Kong) 1 1 26 Croatia 0,7852 Japan 0,49 Poland 0,35 0,84 Increasing 27 Hungary 1,0000 Hungary 1 1 Increasing 28 Indonesia 0,8515 Hungary 0,05 Korea 0,35 Singapore 0,21 0,61 Increasing 29 Republic of Ireland 0,8509 Estonia 0,63 Japan 0,13 Poland 0,14 0,9 Increasing 30 Iceland 1,0000 Iceland 1 1 31 Israel 0,7019 Estonia 0,39 China (Hong Kong) 0,08 Japan 0,19 Korea 0,01 Poland 0,11 0,78 Increasing 32 Italy 0,7210 Estonia 0,17 Finland 0,08 Japan 0,28 Poland 0,07 Singapore 0,21 0,81 Increasing 33 Jordan 0,5733 Estonia 0,58 Japan 0,01 0,59 Increasing 34 Japan 1,0000 Japan 1 1 35 Korea 1,0000 Korea 1 1 36 Kosovo 0,5411 Japan 0,37 Poland 0,15 0,52 Increasing 37 Lebanon 0,5872 Japan 0,13 Korea 0,09 Singapore 0,29 0,51 Increasing 38 Lithuania 1,0000 Lithuania 1 1 39 Luxembourg 0,9271 Switzerland 0,44 Japan 0,17 China (Macao) 0,27 0,88 Increasing 40 Latvia 0,9270 Estonia 0,32 Poland 0,6 Slovenia 0 0,92 Increasing Asian Journal of Education and Training, 2019, 5(2): 349-361 355 © 2019 by the authors; licensee Asian Online Journal Publishing Group 41 China (Macao) 1,0000 China (Macao) 1 42 Moldova 0,6850 Estonia 0,46 Poland 0,23 0,69 Increasing 43 Mexico 0,6251 Estonia 0,09 Japan 0,39 Poland 0,15 0,63 Increasing 44 Macedonia 0,6588 Estonia 0,04 Hungary 0,13 Poland 0,41 0,58 Increasing 45 Malta 0,8411 China (Hong Kong) 0,63 Korea 0,15 Poland 0,01 Singapore 0,02 0,81 Increasing 46 Montenegro 0,7586 Japan 0,22 Poland 0,43 0,65 Increasing 47 Netherlands 0,9765 Switzerland 0,28 Estonia 0,25 China (Hong Kong) 0,41 Japan 0,03 0,97 Increasing 48 Norway 0,8980 Estonia 0,27 Finland 0,24 Japan 0,15 Poland 0,09 Singapore 0,13 0,88 Increasing 49 New Zealand 0,8794 Estonia 0,05 China (Hong Kong) 0,29 Poland 0,66 Singapore 0,02 1,02 Decreasing 50 Peru 0,5846 Estonia 0,16 Japan 0,4 0,56 Increasing 51 Poland 1,0000 Poland 1 1 52 Portugal 0,7663 Switzerland 0,14 Estonia 0,31 Japan 0,09 Poland 0,3 Singapore 0,1 0,94 Increasing 53 Argentina (Buenos Aires) 0,6985 Estonia 0,08 Japan 0,15 Poland 0,22 Singapore 0,36 0,81 Increasing 54 Qatar 0,8148 Korea 0,22 Poland 0,02 Singapore 0,37 0,61 Increasing 55 Spain (Regions) 0,8800 Hungary 0,19 Japan 0,19 Korea 0,04 Poland 0,41 Singapore 0,11 0,94 Increasing 56 Romania 0,8624 Lithuania 0,57 Poland 0,24 0,81 Increasing 57 Russian Federation 0,7899 Estonia 0,41 Japan 0,29 Poland 0,16 0,86 Increasing 58 Singapore 1,0000 Singapore 0 59 Slovakia 0,8867 Estonia 0,18 Hungary 0,16 Poland 0,51 0,85 Increasing 60 Slovenia 1,0000 Slovenia 0 61 Sweden 0,9194 Canada 0,06 Estonia 0,53 Singapore 0,2 Slovenia 0,06 0,85 Increasing 62 Chinese-Taipei (Taiwan) 0,9959 Japan 0,33 Singapore 0,62 0,95 Increasing 63 Thailand 0,7457 Hungary 0,25 Korea 0,23 Lithuania 0,12 Poland 0,13 0,73 Increasing 64 Trinidad and Tobago 0,5390 Estonia 0,57 Japan 0,05 Poland 0,04 0,66 Increasing 65 Turkey 0,7192 Estonia 0,21 Japan 0,44 0,65 Increasing 66 Uruguay 0,6042 Estonia 0,46 China (Hong Kong) 0,18 Japan 0,04 Poland 0,01 0,69 Increasing 67 United States 0,8614 Canada 0,23 Estonia 0,25 Poland 0,15 Singapore 0,26 0,89 Increasing 68 Vietnam 1,0000 Vietnam 1 Source: Obtained from primary data. Asian Journal of Education and Training, 2019, 5(2): 349-361 356 © 2019 by the authors; licensee Asian Online Journal Publishing Group Table-6. Total, Technical and Scale Efficiencies of Countries and their Returns to Scale. NO Countries CCR BCC Scale Efficient Returns to Scale 8 Canada 1 1 1 9 Switzerland 1 1 1 19 Estonia 1 1 1 20 Finland 1 1 1 25 China (Hong Kong) 1 1 1 27 Hungary 1 1 1 30 Iceland 1 1 1 34 Japan 1 1 1 35 Korea 1 1 1 38 Lithuania 1 1 1 41 China (Macao) 1 1 1 51 Poland 1 1 1 58 Singapore 1 1 1 60 Slovenia 1 1 1 68 Vietnam 1 1 1 62 Chinese-Taipei (Taiwan) 0,9959 1 0,9959 Increasing 47 Netherlands 0,9765 1 0,9765 Increasing 4 Austria 0,9666 1 0,9666 Increasing 21 France 0,9446 1 0,9446 Increasing 39 Luxembourg 0,9271 1 0,9271 Increasing 1 Albania 0,8635 1 0,8635 Increasing 56 Romania 0,8624 1 0,8624 Increasing 6 Bulgaria 0,8621 1 0,8621 Increasing 10 Chile 0,8566 1 0,8566 Increasing 28 Indonesia 0,8515 1 0,8515 Increasing 54 Qatar 0,8148 1 0,8148 Increasing 23 Georgia 0,7821 1 0,7821 Increasing 46 Montenegro 0,7586 1 0,7586 Increasing 65 Turkey 0,7192 1 0,7192 Increasing 44 Macedonia 0,6588 1 0,6588 Increasing 7 Brazil 0,6502 1 0,6502 Increasing 37 Lebanon 0,5872 1 0,5872 Increasing 50 Peru 0,5846 1 0,5846 Increasing 59 Slovakia 0,8867 0,9954 0,8908 Increasing 24 Greece 0,8704 0,9886 0,8804 Increasing 36 Kosovo 0,5411 0,981 0,5516 Increasing 40 Latvia 0,927 0,9802 0,9457 Increasing 17 Algeria 0,5164 0,9794 0,5273 Increasing 45 Malta 0,8411 0,9777 0,8603 Increasing 16 Dominican Republic 0,3872 0,9688 0,3997 Increasing 15 Denmark 0,9596 0,9641 0,9953 Increasing 61 Sweden 0,9194 0,9535 0,9642 Increasing 63 Thailand 0,7457 0,9519 0,7834 Increasing 13 Czech Republic 0,9004 0,9421 0,9557 Increasing 22 United Kingdom 0,928 0,9387 0,9886 Increasing 42 Moldova 0,685 0,9342 0,7332 Increasing 12 Costa Rica 0,6396 0,9255 0,6911 Increasing 48 Norway 0,898 0,9219 0,9741 Increasing 43 Mexico 0,6251 0,919 0,6802 Increasing 3 Australia 0,904 0,9182 0,9845 Increasing 55 Spain (Regions) 0,88 0,9134 0,9634 Increasing 2 United Arab Emirates 0,714 0,9107 0,784 Increasing 18 Spain 0,8661 0,9022 0,96 Increasing 67 United States 0,8614 0,8986 0,9586 Increasing 5 Belgium 0,864 0,8977 0,9625 Increasing 11 Colombia 0,5974 0,8942 0,6681 Increasing 29 Republic of Ireland 0,8509 0,8933 0,9525 Increasing 14 Germany 0,8775 0,8901 0,9858 Increasing 26 Croatia 0,7852 0,8828 0,8894 Increasing 49 New Zealand 0,8794 0,8796 0,9998 Decreasing 57 Russian Federation 0,7899 0,8772 0,9005 Increasing 33 Jordan 0,5733 0,8626 0,6646 Increasing 31 Israel 0,7019 0,8317 0,8439 Increasing 53 Argentina (Buenos Aires) 0,6985 0,8303 0,8413 Increasing 32 Italy 0,721 0,8065 0,894 Increasing 52 Portugal 0,7663 0,7824 0,9794 Increasing 64 Trinidad and Tobago 0,539 0,7692 0,7007 Increasing 66 Uruguay 0,6042 0,7688 0,7859 Increasing Source: Obtained from primary data. As shown the Table 6, when technical efficiency was evaluated, it was determined that 33 countries were technically efficient, while 35 countries were not technically efficient. This proportion is almost fifty fifty. 18 of the countries (Taiwan (China), the Netherlands, Austria, France, Luxembourg, Albania, Romania, Bulgaria, Chile, Indonesia, Asian Journal of Education and Training, 2019, 5(2): 349-361 357 © 2019 by the authors; licensee Asian Online Journal Publishing Group Qatar, Georgia, Montenegro, Turkey, Macedonia, Brazil, Lebanon and Peru) that were not totally efficient appeared as technically efficient. Another type of efficiency examined in the study was that of scale efficiency. The results of the analysis reveal that only 15 of the totally efficient countries (Canada, Switzerland, Estonia, Finland, Hong Kong (China), Hungary, Iceland, Japan, Korea, Lithuania, Macao (China), Poland, Singapore, Slovenia and Vietnam) appeared as scale efficient. None of the 18 countries that were technically efficient (Taiwan (China), the Netherlands, Austria, France, Luxembourg, Albania, Romania, Bulgaria, Chile, Indonesia, Qatar, Georgia, Montenegro, Turkey, Macedonia, Brazil, Lebanon and Peru) appeared as scale efficient. A total of 35 countries (Slovakia, Greece, Kosovo, Latvia, Algeria, Malta, the Dominican Republic, Denmark, Sweden, Thailand, the Czech Republic, Great Britain, Moldova, Costa Rica, Norway, Mexico, Australia, Spain (Regions), the United Arab Emirates, Spain, the United States, Belgium, Colombia, the Republic of Ireland, Germany, Croatia, New Zealand, Russia, Jordan, Israel, Buenos Aires (Argentina), Italy, Portugal, Trinidad and Tobago and Uruguay) were neither technically nor scale efficient. These countries could not produce on a suitable scale and used their resources incorrectly, leading to waste (Lorcu, 2008). The analysis results revealed that Turkey was not scale efficient. With a ratio of 71.92%, Turkey was in 56th place in the list of countries. The countries below Turkey were Middle Eastern (Jordan, Lebanon and Algeria), Balkan (Kosovo and Macedonia), and especially Latin American (Trinidad and Tobago, Costa Rica, Mexico, Colombia, Brazil, Peru and the Dominican Republic) countries. Turkey and the other countries need to reorganize their inputs in order to become totally efficient. This situation is expressed as hypothetical input. For these inputs, the percentage reduction rates for the inputs of these countries are presented in Table 7. Table-7. Percentage reduction rates for countries‘ hypothetical inputs. NO DMU Input1 Input2 Input3 Input4 Input5 Input6 Input7 Input8 Input9 62 Chinese- Taipei (Taiwan) 0,41 0,41 1,179666 13,7512 19,3463 6,1131 18,7791 18,1933 8,6619 47 Netherlands 2,35 34,3886 31,68467 2,3500 11,9847 2,3500 4,1952 27,0037 2,3500 4 Australia 3,34 17,21739 11,32106 3,3400 20,6639 3,3400 3,3400 3,3400 12,6728 15 Denmark 4,04 4,04 18,85167 4,0400 4,0400 27,9978 27,6577 4,0400 4,0400 21 France 5,54 16,34248 16,20656 5,5400 17,1621 31,7544 5,5400 28,0071 17,9762 22 United Kingdom 7,2 7,2 7,2 9,7808 7,2000 37,2942 26,9896 41,0381 7,2000 39 Luxembourg 31,120877 20,99233 7,29 7,2900 24,1121 38,5281 7,2900 36,5538 24,9321 40 Latvia 23,622925 13,14401 7,3 7,3000 31,0314 24,0729 22,0382 23,2473 7,3000 61 Sweden 8,06 21,56784 18,4718 8,0600 28,7500 8,0600 23,3917 11,0841 8,0600 3 Australia 21,029468 9,6 23,64318 9,6000 9,6000 9,6000 25,6593 27,4727 9,6000 13 Czech Republic 9,96 15,23478 9,96 9,9600 9,9600 17,4456 10,0054 23,8044 26,0640 48 Norway 10,2 10,2 31,56588 10,2000 10,2000 16,0262 21,6251 32,7815 10,2000 59 Slovakia 11,33 22,87238 11,33 21,5541 11,3300 33,7618 28,8305 43,0068 18,9558 55 Spain (Regions) 12 12 12 12,0000 30,1987 26,7861 22,7905 12,0000 12,3412 49 New Zealand 12,06 17,3037 22,60662 12,0600 22,2703 12,0600 24,7238 37,4955 12,0600 14 Germany 12,25 12,79305 20,84747 12,2500 26,2408 12,2500 12,2500 12,2500 21,6741 24 Greece 16,51095 12,96 12,96 23,4581 34,2359 30,0037 29,4406 25,6073 12,9600 18 Spain 13,39 13,39 13,39 13,3900 23,4020 28,3511 18,7795 13,3900 13,3900 5 Belgium 13,6 18,96346 28,59455 13,6000 32,0794 32,2257 13,6000 24,6390 13,6000 1 Albania 13,65 13,65 13,65 43,8646 27,0993 56,0900 49,6056 44,2956 13,6500 56 Romania 34,344399 24,94393 13,76 13,7600 28,3417 33,8643 44,5774 40,9121 46,3502 6 Bulgaria 13,79 22,57853 23,63463 22,0770 13,7900 52,0827 48,0497 54,5018 21,8415 67 United States 13,86 25,50405 20,2285 13,8600 13,8600 33,8979 34,4570 37,5156 13,8600 10 Chile 24,697078 14,34 34,68483 14,3400 14,3400 14,8210 20,8200 34,3366 14,3400 28 Indonesia 14,85 14,85 14,85 58,7276 37,2498 45,6712 48,2556 43,9182 18,5493 29 Republic of Ireland 36,729693 14,91 22,19725 15,4192 14,9100 24,5992 29,7712 15,2819 14,9100 45 Malta 25,562443 15,89 15,89 20,7528 15,8900 61,7133 36,0633 29,7510 15,8900 54 Qatar 43,972102 18,52 18,52 30,2472 18,5200 52,2928 42,7348 52,7342 49,5123 57 Russian Federation 37,134488 41,40629 26,24585 21,0100 21,2443 39,1383 21,0100 44,8810 21,0100 26 Croatia 21,48 51,59208 30,13419 34,9890 21,4800 42,3832 50,2355 57,3564 29,1336 23 Georgia 21,79 21,79 21,79 34,9955 21,7900 56,4775 51,4114 45,0387 21,7900 52 Portugal 23,37 23,37 23,37 23,3700 43,9348 38,4243 33,0523 23,3700 41,5322 46 Montenegro 24,14 46,49518 37,35317 43,5811 24,1400 63,1785 62,7513 68,1030 44,1156 63 Thailand 25,43 25,43 25,43 25,4300 38,4761 35,5938 48,9402 40,7981 37,6314 32 Italy 27,9 31,54264 29,13209 27,9000 27,9000 34,2268 27,9000 27,9000 38,3123 65 Turkey 28,08 52,33819 34,74174 28,0800 41,4593 64,3816 69,6966 67,3763 37,6027 2 United Arab Emirates 46,31994 28,6 28,6 39,2833 42,5411 45,6664 47,7577 51,8846 28,6000 31 Israel 39,215089 29,81 29,81 31,3629 29,8100 45,6293 29,8319 41,3017 29,8100 53 Argentina (Buenos Aires) 30,15 33,06685 31,08413 30,1500 30,1500 32,6883 30,1500 34,9841 31,0367 42 Moldova 34,810714 40,1603 37,51735 42,6619 31,5000 52,3699 47,1707 50,7961 31,5000 44 Macedonia 34,12 34,49138 34,12 42,2547 34,1200 62,2931 64,6883 61,5516 55,9165 Asian Journal of Education and Training, 2019, 5(2): 349-361 358 © 2019 by the authors; licensee Asian Online Journal Publishing Group 7 Brazil 34,98 57,70405 45,2428 34,9800 34,9800 71,8150 62,6351 76,6881 34,9800 12 Costa Rica 36,04 55,87323 48,22179 54,5544 51,1341 50,2251 46,0711 49,6859 36,0400 43 Mexico 37,49 51,12929 37,49 40,3785 37,4900 56,8647 60,7133 71,5741 37,9340 66 Uruguay 45,460843 39,58 46,42855 39,5800 39,5800 55,2151 46,1444 47,0768 39,5800 11 Colombia 43,833594 40,26 43,6059 48,8245 43,7563 50,9612 50,1059 44,2320 40,2600 37 Lebanon 51,632562 41,28 41,28 53,3953 47,9188 62,6416 48,3802 60,0132 41,2800 50 Peru 41,54 44,93124 46,2811 43,7699 43,9571 49,1171 59,1633 69,6916 41,5400 33 Jordan 57,27877 42,67 54,89973 56,4964 55,9529 64,2589 53,8926 53,6599 42,6700 36 Kosovo 45,89 55,39768 47,9673 54,1936 45,8900 73,1962 73,1858 79,7636 55,9755 64 Trinidad and Tobago 46,256612 53,31624 58,46661 53,2495 46,1000 58,7632 53,4892 46,1000 46,1000 17 Algeria 60,995029 51,69957 63,98131 54,9547 57,1371 62,8556 49,7818 49,2289 48,3600 16 Dominican Republic 61,28 62,53561 65,5459 65,8370 61,2800 78,9322 78,4211 78,3243 61,2800 Mean 26,341275 28,34484 27,78411 27,5067 28,8968 40,0066 37,12907 40,36946 26,92325 Standard Division 15,801207 16,03278 15,04331 17,38627 13,73561 19,49362 18,68829 18,98823 15,199 Minimum 0,4100 0,4100 1,1797 2,3500 4,0400 2,3500 3,3400 3,3400 2,3500 Maximum 61,28 62,53561 65,5459 65,83699 61,28 78,9322 78,42111 79,7636 61,28 Source: Obtained from primary data. For hypothetical input 1, the lowest percentage was recommended for Taiwan (China), which was the country closest to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 26.341275% was recommended for all countries. A slightly above average reduction of 28.08% was recommended for Turkey. For hypothetical input 2, the lowest percentage was recommended for Taiwan (China), which was the country closest to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 28.34484% was recommended for all countries. A reduction of 52.33819%, which was almost twice the average, was recommended for Turkey. For hypothetical input 3, the lowest percentage was recommended for Taiwan (China), which was the country closest to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 27.78411% was recommended for all countries. An above average reduction of 34.74174% was recommended for Turkey. For hypothetical input 4, the lowest percentage was recommended for the Netherlands, which was the second closest country to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 27.5067% was recommended for all countries. A slightly above average reduction of 28.08% was recommended for Turkey. For hypothetical input 5, the lowest percentage was recommended for Denmark, which was the fourth closest country to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 28.8968% was recommended for all countries. An above average reduction of 41.4593% was recommended for Turkey. For hypothetical input 6, the lowest percentage was recommended for the Netherlands, which was the second closest country to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 40.0066% was recommended for all countries. A well above average reduction of 64.3816% was recommended for Turkey. For hypothetical input 7, the lowest percentage was recommended for Austria, which was the third closest country to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 37.12907% was recommended for all countries. A well above average reduction of 69.6966% was recommended for Turkey. For hypothetical input 8, the lowest percentage was recommended for Austria, which was the third closest country to total efficiency, while the highest percentage was recommended for the Kosovo, which was in fourth from last place for total efficiency. For this input, a mean percentage reduction of 40.36946% was recommended for all countries. A well above average reduction of 67.3763% was recommended for Turkey. For hypothetical input 9, the lowest percentage was recommended for the Netherlands, which was the second closest country to total efficiency, while the highest percentage was recommended for the Dominican Republic, which was in last place for success and total efficiency. For this input, a mean percentage reduction of 26.92325% was recommended for all countries. An above average reduction of 37.6027% was recommended for Turkey. 5. Discussion In this study, based on the data of the PISA 2015 report, the science achievement of students in 68 countries was evaluated with reference to educational inputs, which are considered to affect their success. 15 of the countries included in the study (Canada, Switzerland, Estonia, Finland, Hong Kong (China), Hungary, Iceland, Japan, Korea, Lithuania, Macao (China), Poland, Singapore, Slovenia and Vietnam) were totally efficient. Based on this, it can be said that these countries used their inputs and efforts efficiently and conducted their activities on a suitable scale (Lorcu, 2008). They were able to successfully manage the factors regarded as educational inputs in the PISA 2015 report, namely ―student behaviours preventing learning‖, ―teacher behaviours preventing learning‖, ―lack of educational materials‖, ―lack of teaching staff‖, ―professional development‖, ―teacher participation‖, ―curriculum development‖, ―total number of science teachers in school‖ and ―science-specific resource index (total)‖. In this way, they increased achievement in science. Examining the literature, studies have been conducted that support this finding. In a study conducted by Lavonen and Laaksonen (2009) based on the PISA 2006 data, it was concluded that the source of Finnish students‘ success in science lessons was practical work in class and the students‘ deductions. A knowledge-based society, development of educational equality and decisive power on a local level, Asian Journal of Education and Training, 2019, 5(2): 349-361 359 © 2019 by the authors; licensee Asian Online Journal Publishing Group and teacher education were regarded as the most important education policy issues underlying students‘ high performance in science lessons in the PISA 2006 report. Similarly, in a study conducted by McConney et al. (2014) based on the PISA 2006 data, it was stated that science achievement of students in Australia, Canada and New Zealand was related to how much they experienced inquiry-based teaching and learning. In Tang (2015) study, the results of the analyses made based on the PISA 2006 data for 5,995 students in the USA and Taiwan revealed that with their science teaching-learning activities, students in the USA had more opportunities to learn science than students in Taiwan. It was revealed that in the USA, students who had a higher socioeconomic level had more opportunities with regard to reform-based science learning activities. Moreover, it was determined that students‘ higher socioeconomic level bore no relation to reform-based science learning in Taiwan. The data in the PISA 2015 report were also evaluated for different countries according to science teachers‘ perceptions school principals‘ perceptions and implementation of the teaching programme. In their study, Chi et al. (2018) used the science performance of students in Beijing, Shanghai, Jiangsu and Guangdong, and background research data for PISA 2015, and the results revealed that rather than teacher support, disciplinary climate had a positive effect on the relationship between inquiry-based science activities and students‘ science achievement for both genders. In Oztürk (2018) study based on the PISA 2015 data, it was determined that for different socioeconomic levels there were significant relationships between scientific literacy and environmental awareness, and between scientific literacy and environmental optimism. In a study by Susongko and Afrizal (2018) using the PISA 2015 data for Indonesian students, it was concluded that environmental awareness was related to the students‘ enjoyment of learning science, their self-efficacy in science, their instrumental motivation, inquiry-based instruction in science learning, and their science proficiency and epistemic beliefs. In Lau and Lam (2017) study, according to the PISA 2015 data, adaptive instruction, teacher-directed instruction and interactive application were positively correlated with science performance in all regions. The regions apart from Japan and Korea tended to have a high frequency of teacher-directed instruction facilitated by more or less authoritative discussion in class. It was seen that of the countries having total efficiency, the three countries mostly shown as references were Japan, Poland and Estonia. In Lewis (1995) study, it is argued that Japanese schools support their students‘ academic success because they meet the students‘ needs, thereby supporting strong, positive emotional ties between students and their school, and this supports the finding obtained in this study that Japan was one of the country‘s most often taken as an example. Zawistowska (2014) considers that the greatest impact on the increase in Poland‘s PISA scores was the educational reform made in 1999. In this reform, it is considered that the most important change affecting PISA data was the introduction of assessments other than multiple choice. It may be considered that these changes in the education system in Poland are a reason why Poland was shown as a reference for countries that were not totally efficient. It was revealed that the countries furthest from being totally efficient were Jordan, Kosovo, Trinidad and Tobago, Algeria and the Dominican Republic. In a study conducted by Al-Amoush et al. (2011) it was concluded that science teachers and preservice science teachers in Jordan tended to use traditional teaching methods, based on which, it can be said that the fact that the country was among the most unsuccessful in the PISA data may be related to procedures in the teaching process. In a study made by Al-Amoush et al. (2014) teaching-learning beliefs of German, Turkish and Jordanian preservice teachers were determined and compared. These findings correspond to the research findings made by Al-Amoush et al. (2011). The preference of the Jordanian teacher candidates was the general educational approach, which was the teacher-centred approach. The preference of the German teacher candidates was the modern educational approach. The situation of the Turkish preservice teachers was between these two extremes. Although they had an approach close to the traditional educational approach, they had more modern educational beliefs than those of the Jordanian teacher candidates. It was concluded that apart from New Zealand, all countries had increasing returns to scale. In other words, an input of 1 unit in the inputs resulted in an output greater than 1 in the outputs. Fernandez et al. (2008) emphasised that New Zealand needed to update its physics curriculum. The justification for this was shown to be the lack of communication between teachers and curriculum designers. In support of the study findings, criticisms were made by Buabeng (2015) to the effect that physics teachers in New Zealand were not trained to be adequately equipped, and that they were not sufficiently qualified to be able to use investigative teaching methods or techniques, or to make correct assessments, in their lessons. In this study, the results also revealed that there was no relationship between being a totally effective country and geographic proximity. In a study by Taht and Must (2010) the PISA data of five neighbouring countries, namely Russia, Latvia, Estonia, Sweden and Finland were evaluated. Differences were determined among scores. It was considered that these differences may be due to differences in cultural influences on personality or to national educational and social policies. Similarly, in a study carried out by Herbst and Wojciuk (2017) it was stressed that the neighbouring countries of the Czech Republic, Hungary, Poland and Slovakia had institutional differences in their education systems and that differences between national approaches towards educational reforms were reflected especially in education centralisation, school autonomy, accountability and financial mechanisms. It was seen that Turkey lagged behind many countries in terms of total efficiency. This situation shows that there are problems related to the use of educational inputs. In a study by Anil (2011) in which the 2006 PISA data were used, ―time‖ was the variable that best predicted science success and the factor most clearly determining success. The other factors that predicted science achievement were ―environment‖, ―education‖ and ―attitudes‖, respectively. Bakir et al. (2015) determined that success rates in PISA data were related to countries‘ socioeconomic statuses. It was stated that Turkey‘s low achievement levels were due to the fact that national income is also low. In a study by Kahraman and Celik (2017) an attempt was made to determine Turkish students‘ success and sources in the PISA 2012 data. It was found that attendance at nursery school, school starting age, mother‘s employment status, parents‘ educational level and numbers of computers and books at home affected student success. Having a working mother and high school starting age had a negative effect on student success. On the other hand, attendance at nursery school, parents with a high educational level and a high number of books and computers in the home had a Asian Journal of Education and Training, 2019, 5(2): 349-361 360 © 2019 by the authors; licensee Asian Online Journal Publishing Group positive impact on students‘ achievement. Moreover, it was determined that the sociocultural and environmental locations of the students affected their success. Based on these findings, it can be recommended that in order for countries whose science success was determined as low in the PISA report to increase their success, they should manage their educational inputs and take reference countries that are suitable for themselves as examples. It can be recommended that when organising and implementing their education programmes, they should take their educational inputs and successful countries into consideration. It can also be recommended to researchers that they undertake similar studies using the PISA data, and that they examine the reasons for the findings of this study with different methods. References Aaronson, D., L. Barrow and W. Sander, 2007. Teachers and student achievement in the Chicago public high schools. Journal of Labor Economics, 25(1): 95-135.Available at: https://doi.org/10.1086/508733. Al-Amoush, S., S. Markic, M. Usak, M. Erdogan and I. Eilks, 2014. Beliefs about chemistry teaching and learning—a comparison of teachers‘and student teachers‘beliefs from jordan, turkey and germany. International Journal of Science and Mathematics Education, 12(4): 767-792.Available at: https://doi.org/10.1007/s10763-013-9435-7. Al-Amoush, S.A., S. Markic, I. Abu-Hola and I. Eilks, 2011. Jordanian prospective and experienced chemistry teachers' beliefs about teaching and learning and their potential role for educational reform. Science Education International, 22(3): 185-201. Anil, D., 2011. Investigation of factors influencing Turkey's PISA 2006 science achievement with structural equation modelling. Educational Sciences: Theory and Practice, 11(3): 1261-1266. Baeten, M., E. Kyndt, K. Struyven and F. Dochy, 2010. Using student-centred learning environments to stimulate deep approaches to learning: Factors encouraging or discouraging their effectiveness. Educational Research Review, 5(3): 243-260.Available at: https://doi.org/10.1016/j.edurev.2010.06.001. Bakir, S., H. Demirel and Y.E. Yilmaz, 2015. PISA scores from 2003 to 2012: A comparison of Turkey with the three countries which have been successful in each term in field of science. Procedia-Social and Behavioral Sciences, 174: 2733-2742.Available at: https://doi.org/10.1016/j.sbspro.2015.01.960. Banker, R.D., A. Charnes and W.W. Cooper, 1984. Some models for estimating technical and scale inefficiencies in data envelopment analysis. Management Science, 30(9): 1078-1092.Available at: https://doi.org/10.1287/mnsc.30.9.1078. Bennett, J., F. Lubben and S. Hogarth, 2007. Bringing science to life: A synthesis of the research evidence on the effects of context-based and STS approaches to science teaching. Science Education, 91(3): 347-370.Available at: https://doi.org/10.1002/sce.20186. Buabeng, I., 2015. Teaching and learning of Physics in New Zealand high schools. PhD Thesis, University of Canterbury, New Zealand. Bybee, R.W., 2014. NGSS and the next generation of science teachers. Journal of Science Teacher Education, 25(2): 211-221.Available at: https://doi.org/10.1007/s10972-014-9381-4. Cairns, D. and S. Areepattamannil, 2019. Exploring the relations of inquiry-based teaching to science achievement and dispositions in 54 countries. Research in Science Education, 49(1): 1-23.Available at: https://doi.org/10.1007/s11165-017-9639-x. Charnes, A., W.W. Cooper and E. Rhodes, 1978. Measuring the efficiency of decision making units. European Journal of Operational Research, 2(6): 429-444.Available at: https://doi.org/10.1016/0377-2217(78)90138-8. Chi, S., X. Liu, Z. Wang and S. Won Han, 2018. Moderation of the effects of scientific inquiry activities on low SES students‘ PISA 2015 science achievement by school teacher support and disciplinary climate in science classroom across gender. International Journal of Science Education, 40(11): 1284-1304.Available at: https://doi.org/10.1080/09500693.2018.1476742. Clermont, C.P., H. Borko and J.S. Krajcik, 1994. Comparative study of the pedagogical content knowledge of experienced and novice chemical demonstrators. Journal of Research in Science Teaching, 31(4): 419-441.Available at: https://doi.org/10.1002/tea.3660310409. Cooper, W.W., S. Li, L.M. Seiford, K. Tone, R.M. Thrall and J. Zhu, 2001. Sensitivity and stability analysis in DEA: Some recent developments. Journal of Productivity Analysis, 15(3): 217-246. Davis, E.A., D. Petish and J. Smithey, 2006. Challenges new science teachers face. Review of Educational Research, 76(4): 607-651.Available at: https://doi.org/10.3102/00346543076004607. Durán-Narucki, V., 2008. School building condition, school attendance, and academic achievement in New York City public schools: A mediation model. Journal of Environmental Psychology, 28(3): 278-286.Available at: https://doi.org/10.1016/j.jenvp.2008.02.008. Dyson, R.G., R. Allen, A.S. Camanho, V.V. Podinovski, C.S. Sarrico and E.A. Shale, 2001. Pitfalls and protocols in DEA. European Journal of Operational Research, 132(2): 245-259.Available at: https://doi.org/10.1016/s0377-2217(00)00149-1. Evagorou, M., S. Erduran and T. Mäntylä, 2015. The role of visual representations in scientific practices: From conceptual understanding and knowledge generation to ‗seeing‘how science works. International Journal of STEM Education, 1(2): 1-13.Available at: https://doi.org/10.1186/s40594-015-0024-x. Fernandez, T., G. Ritchie and M. Barker, 2008. A sociocultural analysis of mandated curriculum change: The implementation of a new senior physics curriculum in New Zealand schools. Journal of Curriculum Studies, 40(2): 187-213.Available at: https://doi.org/10.1080/00220270701313978. Gislason, N., 2010. Architectural design and the learning environment: A framework for school design research. Learning Environments Research, 13(2): 127-145.Available at: https://doi.org/10.1007/s10984-010-9071-x. Harris, D.N. and T.R. Sass, 2011. Teacher training, teacher quality and student achievement. Journal of Public Economics, 95(7-8): 798- 812.Available at: https://doi.org/10.1016/j.jpubeco.2010.11.009. Herbst, M. and A. Wojciuk, 2017. Common legacy, different paths: The transformation of educational systems in the Czech Republic, Slovakia, Hungary and Poland. Compare: A Journal of Comparative and International Education, 47(1): 118-132.Available at: https://doi.org/10.1080/03057925.2016.1153410. Hill, N.E. and L.C. Taylor, 2004. Parental school involvement and children's academic achievement: Pragmatics and issues. Current Directions in Psychological Science, 13(4): 161-164.Available at: https://doi.org/10.1111/j.0963-7214.2004.00298.x. Hodson, D., 1988. Experiments in science and science teaching. Educational Philosophy and Theory, 20(2): 53-66.Available at: https://doi.org/10.1111/j.1469-5812.1988.tb00144.x. Holbrook, J. and M. Rannikmae, 2007. The nature of science education for enhancing scientific literacy. International Journal of Science Education, 29(11): 1347-1362.Available at: https://doi.org/10.1080/09500690601007549. Huppert, J., S.M. Lomask and R. Lazarowitz, 2002. Computer simulations in the high school: Students' cognitive stages, science process skills and academic achievement in microbiology. International Journal of Science Education, 24(8): 803-821.Available at: https://doi.org/10.1080/09500690110049150. Jeynes, W.H., 2007. The relationship between parental involvement and urban secondary school student academic achievement: A meta- analysis. Urban Education, 42(1): 82-110.Available at: https://doi.org/10.1177/0042085906293818. Kahraman, Ü. and K. Celik, 2017. Analysis of PISA 2012 results in terms of some variables. Journal of Human Sciences, 14(4): 4797- 4808.Available at: https://doi.org/10.14687/jhs.v14i4.5136. Kirschner, P.A., J. Sweller and R.E. Clark, 2006. Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educational Psychologist, 41(2): 75-86.Available at: https://doi.org/10.1207/s15326985ep4102_1. Koretsky, M.D., M. Vauras, C. Jones, T. Iiskala and S. Volet, 2019. Productive disciplinary engagement in high- and low-outcome student groups: Observations from three collaborative science learning contexts. Research in Science Education, 49(138): 1-24.Available at: https://doi.org/10.1007/s11165-019-9838-8. Lau, K.-c. and T.Y.-p. Lam, 2017. Instructional practices and science performance of 10 top-performing regions in PISA 2015. International Journal of Science Education, 39(15): 2128-2149.Available at: https://doi.org/10.1080/09500693.2017.1387947. Asian Journal of Education and Training, 2019, 5(2): 349-361 361 © 2019 by the authors; licensee Asian Online Journal Publishing Group Lavonen, J. and S. Laaksonen, 2009. Context of teaching and learning school science in Finland: Reflections on PISA 2006 results. Journal of Research in Science Teaching, 46(8): 922-944.Available at: https://doi.org/10.1002/tea.20339. Lewis, C.C., 1995. The roots of Japanese educational achievement: Helping children develop bonds to school. Educational Policy, 9(2): 129- 151.Available at: https://doi.org/10.1177/0895904895009002003. Lorcu, F., 2008. Data envelopment analysis (DEA) to evaluate the effectiveness of Turkey and the European Union countries in the health field. Unpublished PhD Thesis, Istanbul University, Institute of Social Sciences, Istanbul. Maltese, A.V., C.S. Melki and H.L. Wiebke, 2014. The nature of experiences responsible for the generation and maintenance of interest in STEM. Science Education, 98(6): 937-962.Available at: https://doi.org/10.1002/sce.21132. Maxwell, L.E., 2016. School building condition, social climate, student attendance and academic achievement: A mediation model. Journal of Environmental Psychology, 46: 206-216.Available at: https://doi.org/10.1016/j.jenvp.2016.04.009. McConney, A., M.C. Oliver, A. WOODS‐McCONNEY, R. Schibeci and D. Maor, 2014. Inquiry, engagement, and literacy in science: A retrospective, cross-national analysis using PISA 2006. Science Education, 98(6): 963-980.Available at: https://doi.org/10.1002/sce.21135. Mulholland, J. and J. Wallace, 1996. Breaking the cycle: Preparing elementary teachers to teach science. Journal of Elementary Science Education, 8(1): 17-38.Available at: https://doi.org/10.1007/bf03173739. Osborne, J., 2014. Teaching scientific practices: Meeting the challenge of change. Journal of Science Teacher Education, 25(2): 177- 196.Available at: https://doi.org/10.1007/s10972-014-9384-1. Osborne, J., S. Simon and S. Collins, 2003. Attitudes towards science: A review of the literature and its implications. International Journal of Science Education, 25(9): 1049-1079.Available at: https://doi.org/10.1080/0950069032000032199. Oztürk, Ö., 2018. Using PISA 2015 data to analyze how the scientific literacy of students from different socioeconomic levels can be predicted by environmental awareness and by environmental optimism. (Doctoral Dissertation, Bilkent University). Palmer, D.H., 2001. Factors contributing to attitude exchange amongst preservice elementary teachers. Science Education, 86: 122– 138.Available at: https://doi.org/10.1002/sce.10007. Rivera, M.D. and S.R. Lopez, 2019. Some pennies are more equal than others: Inequitable school facilities investment in San Antonio, Texas. Education Policy Analysis Archives, 27(16): 1-34.Available at: http://dx.doi.org/10.14507/epaa.27.4191. Rockoff, J.E., 2004. The impact of individual teachers on student achievement: Evidence from panel data. American Economic Review, 94(2): 247-252.Available at: https://doi.org/10.1257/0002828041302244. Rothstein, J., 2010. Teacher quality in educational production: Tracking, decay, and student achievement. The Quarterly Journal of Economics, 125(1): 175-214.Available at: https://doi.org/10.1162/qjec.2010.125.1.175. Sheldrake, R., T. Mujtaba and M.J. Reiss, 2017. Science teaching and students‘ attitudes and aspirations: The importance of conveying the applications and relevance of science. International Journal of Educational Research, 85(1): 167-183.Available at: https://doi.org/10.1016/j.ijer.2017.08.002. Stohr‐Hunt, P.M., 1996. An analysis of frequency of hands-on experience and science achievement. Journal of Research in Science Teaching, 33(1): 101-109.Available at: https://doi.org/10.1002/(sici)1098-2736(199601)33:1<101::aid-tea6>3.0.co;2-z. Susongko, P. and T. Afrizal, 2018. The determinant factors Analysis of Indonesian students‘ environmental awareness in PISA 2015. Indonesian Science Education Journal, 7(4): 407-419.Available at: https://doi.org/10.15294/jpii.v7i4.10684. Taht, K. and O. Must, 2010. Are the links between academic achievement and learning motivation similar in five neighbouring countries? Trames Journal of the Humanities and Social Sciences, 14(3): 271–281.Available at: https://doi.org/10.3176/tr.2010.3.04. Tang, N.E., 2015. The effect of reform-based science teaching on SES-associated achievement gap on Pisa 2006: A comparative study of the United States and Taiwan. (Doctoral Dissertation, University of Missouri--Columbia). Teig, N., R. Scherer and T. Nilsen, 2018. More isn't always better: The curvilinear relationship between inquiry-based teaching and student achievement in science. Learning and Instruction, 56(1): 20-29.Available at: https://doi.org/10.1016/j.learninstruc.2018.02.006. Zawistowska, A., 2014. The black box of the educational reforms in Poland: What caused the improvement in the PISA scores of polish students? Polish Sociological Review, 187(3): 333-350. Asian Online Journal Publishing Group is not responsible or answerable for any loss, damage or liability, etc. caused in relation to/arising out of the use of the content. Any queries should be directed to the corresponding author of the article. http://dx.doi.org/10.14507/epaa.27.4191