The effect of problem-based learning on undergraduate students' learning about solutions and their physical properties and scientific processing skills

Abstract

The aim of this study was to investigate the effect of Problem-Based Learning (PBL) on undergraduate students' learning about solutions and their physical properties, and on their scientific processing skills. The quasi experimental study was carried out through non-equivalent control and comparison groups pre-post test design. The data were collected through both quantitative and qualitative means. The sample group of the study was 84 freshmen who took the General Chemistry-II course, in two separate classrooms from the Department of Primary Science Education at a school of education in a state university established in the eastern Anatolia region of Turkey, during the spring semester in the 2010 academic year. The study was implemented over five weeks. Quantitative data were collected through an academic achievement test, a scientific processing skill test, and specific scales developed for PBL. In addition, qualitative data were collected through interviews. Pre-test data was analyzed by independent samples t-test, post-test data was analyzed through ANCOVA (Analysis of Covariance). Qualitative data was analysed in a descriptive manner and presented in tables. The findings of the study revealed that PBL is more effective than conventional instruction in improving students' learning and scientific processing skills. In addition, it was also revealed that PBL increased students' levels of accessing and using knowledge, working in a group and cooperating, autonomous learning and problem solving skills. However, it was also revealed that PBL has disadvantages, such as difficulties in assessing the skills gained through PBL, students’ difficulties in getting used to the PBL, difficulty in setting up heterogenous groups and overcoming problems encountered during establishment of cooperation in the groups, dealing with competition between the students, and limited time.

---

Introduction

While information is increasing at a very fast pace in today's world in which science and technology is experiencing rapid changes, fundamental changes are taking place in learning sciences. These improvements have also changed the problems we face, our perception of education and relations between technology and society. The constructivist approach has left its mark on educational activities and created a transformation in education today (Taber, 2011) As a result of this transformation, active learning methods, which provide student–student, student–technology and student–teacher interaction using different sources of information and technology, have been the focus of educational research. Problem-based learning (PBL) is one of these active learning methods (Tarhan et al., 2008).

Although PBL was originally developed in medical school programs, at Case Western Reserve University in the United States in the 1950s and McMaster University in Canada in the 1960s, its theoretical foundations date back to the research of John Dewey (McDonald, 2002). According to Schmidt (1983), PBL is inspired by Bruner's philosophy of education and the case studies of Fraiser (cited inYucelis-Alper, 2003). According to McDonald (2002), PBL was developed by Barrows, who was inspired by Dewey's philosophy of learning which attributes to the idea of real-life learning, and he used it in the training of doctors at McMaster University.

Initially, PBL had been implemented as the answer to the question of how to improve conventional pre-clinical science courses, and how to help doctors grow as lifelong learners and problem solvers. PBL has then been applied in other areas around the world since the 1970s. Today, PBL is adapted for use in science classes (Chin and Chia, 2004; Delisle, 1997; Gallagher et al., 1995). Serin (2009) used PBL in a science course and investigated 7th grade students' science achievement, attitude towards the science course and scientific processing skills. The analyses revealed that there was no statistically significant mean difference between groups in the scores of attitude, scientific processing skills and academic achievement. In another study, Wong and Day (2009) studied PBL in two topics, “Human Reproduction” and “Density”, with 7th grade students.

In addition, there are PBL practices used in the teaching of biology and other science courses at high schools (Uden and Beaumont, 2006; Ward and Lee, 2004). Researchers have found that PBL can be successfully applied to chemistry. Dobbs (2008) used PBL in teaching the “acid–base” topic in high school chemistry. Rivarola et al. (1997) described applications of PBL in a biochemistry course. Belt et al. (2002) developed a PBL activity covering several areas of analytical chemistry and forensic science. Ram (1999), Yuzhi (2003), Yu (2004) and Larive (2004) used PBL in analytical chemistry courses. PBL was used in instrumental analysis courses by Zhang (2002) and in electrochemistry courses by Ying (2003). Groh (2001) developed PBL problem scenarios in a general chemistry course for students to learn principles of solutions and their properties. Senocak et al. (2007) used PBL related to gases in a general chemistry course. A PBL laboratory-based module for first year undergraduate chemistry has been developed and successfully implemented (Kelly and Finlayson, 2007, 2009). Gurses et al. (2007) conducted PBL in a physical chemistry laboratory course and investigated students' attitudes towards the chemistry laboratory course, the scientific processing skills of the students, and their academic achievements. Tarhan et al. (2008) examined the effectiveness of PBL on 9th grade students’ understanding of the subject of intermolecular forces: dipole–dipole forces, London dispersion forces and hydrogen bonding. PBL students’ answers to open-ended questions in the post-test showed that the students from the experimental group were better at using scientific and critical ideas. Besides, while a variety of alternative conceptions about molecular forces were determined in control group students, there was no alternative conception found in PBL students. In another study by Tarhan and Acar (2007), PBL was used in an 11th grade chemistry class to teach factors that affect cell potential, and it was found that PBL was effective at increasing students' achievement, remedying formation of alternate conceptions and developing social skills. Also, according to the responses of students in the control group, it was found that they had more alternative conceptions related to the equilibrium constant, effect of concentration and temperature on electrochemical equilibrium than did students from the PBL group.

Because of the importance of solution chemistry in the school chemistry curriculum, studies have focused on various concepts: dissolution (Ebenezer and Erickson, 1996; Smith and Metz, 1996) solubility (Ebenezer and Erickson, 1996), energy in solution processes (Ebenezer and Fraser, 2001), effects of temperature and stirring on the dissolution of a solid in liquid (Blanco and Prieto, 1997), and types of solutions: unsaturated, saturated and supersaturated (Pinarbasi and Canpolat, 2003), as well as vapor pressure lowering, solubility of a gas in water and the relationship between vapor pressure and boiling point (Pinarbasi and Canpolat, 2003). Recent studies in the field of chemistry education have shown that students have plenty of common and critical alternative conceptions in many chemistry subjects (Atasoy et al., 2003; Calik et al., 2005; Morgil et al., 2003). The subject of solutions and their physical properties is one of the areas of chemistry in which students have alternative conceptions (Pinarbasi and Canpolat, 2003). In his study of 87 freshmen university students, Pinarbasi (2002) found the following alternative conceptions relating to the topic of solutions (Table 1).

Table 1 Students' alternative conceptions related to the topic of solutions (Pinarbasi, 2002)

Students' alternative conceptions

Each mixture is a solution

Solution volumes may be collected

Dissolved particles in solution are motionless

The weight of dissolved particles in solution does not change or decrease

Dissolution rate decreases with temperature

A saturated solution at equilibrium with a solute is a supersaturated solution

The mix affects resolution

Substances do not dissolve in each other because the molecules of these substances push each other

Temperature changes bring about a change in attractive forces between the solvent and solute molecules

In relation to the resolution of a gas dissolved in the liquid, gas solubility is associated with total pressure of the gas in the liquid, not the partial pressure of the gas in the liquid

A pure solvent and its solution at atmospheric pressure and boiling temperature have different vapor pressure values

---

In another study by Pinarbasi and Canpolat (2003) with 107 university students, the alternative conceptions shown in Table 2 were found.

Table 2 Students' alternative conceptions related to the topic of solutions (Pinarbasi and Canpolat, 2003)

Students' alternative conceptions

A solution containing undissolved solute is a supersaturated solution

Undissolved solute is a component of solution

The amount of gas dissolved in a solvent is proportional to the total pressure of gases above the solution

Boiling liquids at atmospheric pressure have different vapor pressures

Because of the attractive forces between solute and solvent particles, the vapor pressure of a solution is less than that of pure solvent

---

The teaching of solutions and solubility in Turkey and elsewhere generally focuses on algorithmic problem solving rather than conceptual understanding. Although students can solve algorithmic problems successfully, they cannot understand or they misunderstand the essential ideas behind the concepts (Pinarbasi and Canpolat, 2003; Smith and Metz, 1996). Today, the conceptual learning of chemistry by students is of great importance. Instead of learning chemical principles behind the concepts taught, they prefer to memorize the answers used to solve various problems or numerical equations (Canpolat, 2002). Not understanding or misunderstanding the concepts related to physical properties of solutions affects the understanding of other relevant subjects negatively, and thus reduces student success (Smith and Metz, 1996). There is the problem of interaction between ideas arising from daily experiences and those ideas taught in school, and these should be linked to each other for teaching and learning. In addition, new information must be linked with preliminary information in the teaching process for learning. Therefore, there is a need for a study based on a constructivist approach, that will turn teaching the concepts of solutions into an active learning process and thus make it more enjoyable. Because of the common alternative conceptions on this topic, we chose to study “solutions and their physical properties” while examining the effectiveness of PBL compared to conventional instruction within the framework of General Chemistry-II courses in an undergraduate setting.

Purpose of the research

The purpose of the research study is to examine the effectiveness of PBL on undergraduate students' understanding of the subject of “solutions and their physical properties: types of solution; some terms; concentration units; molecular point of view to formation of the solution; factors affecting solubility and dissolution rate; and colligative properties and colloidal mixtures”. In the context of this study, the following research questions are investigated:

1. Is there a statistically significant difference between students' performances when learning the concepts of solutions and their physical properties through PBL compared to a conventional lecture based instruction?

2. When compared to a teacher-centred conventional approach, how effective is PBL in preventing students' alternative conceptions of solutions and their physical properties?

3. Is there a statistically significant effect of PBL compared to a conventional lecture based instruction on students' scientific processing skills?

Assumption and limitations

The external validity of this study is limited in that the study was conducted in two separate classrooms in only one school. The application of PBL is restricted to the subject of solutions and their physical properties. The implementation of the study is limited to 5 weeks (20 course hours). After the study is complete, the courses in experimental and control groups are continued by the lecturers of the courses. It is assumed that this would not have any negative impacts on students. During the construction of the groups, it was assumed that the tutor was aware of students' characteristics and backgrounds. It was assumed that students followed directions in the worksheets, and the teaching was performed by the researchers.

Materials and methods

Sample

The sample of the study was a total of 84 freshmen who took the General Chemistry-II course in two different classrooms from the Department of Primary Science Education at a school of education in a state university established in the Eastern Anatolia Region in Turkey. Students in the experimental group were instructed with PBL (the nature of instruction is discussed in detail below), while students in the control group received conventional instruction. In conventional instruction, the lecturer mostly transmits the knowledge in front of the classroom and rarely encourages students to join the discussions. Each of the experimental (PBL) and control groups had 42 students assigned. However, during the study, some participants missed the classes or taking the tests, so the figures in the tables may vary. The participants were selected through convenience sampling (Johnson and Christensen, 2004).

The intervention took five weeks during the spring semester of 2010. Before determining the control and experimental groups, the Solutions and their Physical Properties Academic Achievement Test (SPPAAT) and Scientific Processing Skill Test (SPST) were applied as pre-tests. An independent sample t-test showed no statistically significant difference between the two classrooms according to SPPAAT (t (69) = −1.753; p > .05) and SPST (t (65.235) = −1.419; p > .05). According to this finding, two groups were randomly assigned into experimental and control groups.

Data collection instruments

Solutions and their Physical Properties Academic Achievement Test (SPPAAT)

In developing SPPAAT, firstly a list of concepts covered and the cognitive learning objectives according to Bloom's revised taxonomy (Tosun and Taskesenligil, 2011) were established. In order to achieve validity of the SPPAAT, the help of five experts was sought both in terms of content validity and suitability of the cognitive learning objectives in each question. Table 3 shows the breakdown of the questions in the academic achievement test according to Bloom's revised taxonomy.

Table 3 The breakdown of achievement test questions according to the learning levels of Bloom's revised taxonomy (Amer, 2006; Krathwohl, 2002)

Levels of cognition	Questions
Remembering	1, 2, 6, 19, 20
Understanding	3, 12
Applying	4, 7–9, 14, 17, 18
Analysing	10, 11, 13, 15, 16, 23
Evaluating	5, 21, 22, 24
Creating	—

---

The academic achievement test on solutions and their physical properties was composed of a total of 24 questions, which were multiple choice, open-ended, short answers and true–false items; and it was developed in light of the students' learning difficulties and alternative conceptions identified in the literature (Pinarbasi, 2002; Pinarbasi and Canpolat, 2003). This achievement test by Tosun and Taskesenligil (2011) covered all aspects of solutions and their physical properties including: types of solution; some terms; concentration units; molecular point of view of formation of the solution; factors affecting solubility and dissolution rate; and colligative properties and colloidal mixtures. The test on solutions and their physical properties was piloted with 160 students who took the General Chemistry-II course. The validity of the test was achieved by consulting chemistry professors and the researchers. Some modifications were made in terms of language and in the design of the test. Items analyses were calculated for the test. Based on the data, the average test difficulty index and test discrimination index were calculated as 0.41 and 0.40 respectively. Test reliability (KR-20) was determined as 0.77. This level of reliability coefficient for an achievement test indicates that the test could be considered satisfactorily reliable (McMillan and Schumacher, 2006, p.243). These questions were applied before and after treatment and some example items are shown in Table 4.

Table 4 Sample questions from the post-test

Sample questions

5. Which of the following(s) change both the solubility and the dissolution rate of the material for all solid, liquid and gas states?

I. Stirring II. Changing the pressure III. Dividing into small pieces IV. Changing the temperature V. Changing the amount of dissolvent

(a) Only V (b) I, III and IV (c) Only IV (d) II, IV and V (e) I, II, III, IV and V

10. Some salt is added into a bowl of water and then mixed. Which of the following (s) occur during the solution of salt in the water?

I. The ions in the salt take the place of air molecules in the water.

II. The solid salt turns into liquid salt.

III. The attraction force between ions and dissolvent molecules results in dissolution of salt.

IV. A new chemical substance is formed.

V. Na⁺ ve Cl⁻ ions surrounded with water molecules are hydrated ions.

(a) II and V (b) Only III (c) I and II (d) III and V (e) III, IV and V

19. When the water concentration in pure water is more than the water concentration in the solution, there occurs a clear movement from pure water to solution. This movement is called…

20. Solutions with a certain amount of undissolved matter are called unsaturated solutions.

21. What do you think is the reason behind the increase in the number of gas bubbles when some salt and granulated sugar is added into a glass of coke?

23. Please explain what do the small bubbles occurring in tap water after it is left in room temperature consist of and how they are formed.

---

Scientific Process Skill Test (SPST)

The test used in this study was originally developed by Okey, Wise and Burns (1982), translated and adapted into Turkish, and the internal consistency (Cronbach alpha coefficient) of the test was found to be 0.81 by Geban et al., (1992). In this test, there are thirty-six items with four choices. These items cover all science areas. The test also has five subtests: identifying variables; identifying and stating hypotheses; operationally defining; designing investigations; and graphing and interpreting data (Geban et al., 1992).

Scale of determining students' views about PBL

A scale has been developed for comprising advantages and disadvantages of PBL, which contains students' statements about the evaluation process after they worked with PBL in order to determine their views on PBL by the researchers. The aim of preparing this scale is to determine students' perceptions about the PBL process and to resolve these shortcomings where students' opinions may be negative. In addition, attempts were made to determine which skills students in the process of the PBL process developed. The scale contains five options (strongly agree, agree, undecided, disagree, totally disagree) of Likert types, and contains twenty-one items. The scale was developed by consulting instructors who are experts in the field. Each item in the scale is included in data analysis section.

Interviews

Interviews were conducted to determine information about the students' learning levels on solutions and their physical properties in the experimental and control groups. At the end of the application, a comfortable environment was created where students could freely express their thoughts. The behavior of a person in a group and alone differs. Groups can produce more alternatives compared to individuals, they solve problems faster and produce more options in shorter periods (Yildirim and Simsek, 2006). Therefore, group interviews were conducted with a total of 19 students from the experimental group (three students from five groups, and four students from one group), and a total of eight students in the control group. Students were selected as volunteers for semi-structured interviews and were selected according to the students' success in the SPPAAT. There were students from achievers, non-achievers and averages. All the 27 students that participated in the interviews agreed to audio recording during the interview. During the interviews of both experimental and control group students, their conceptual knowledge was examined. Student understanding of such basic concepts as solutions, colloidal compositions, heterogeneous compositions, dissolution, solubility, factors affecting solubility, enthalpy of solubility and colligative properties was probed.

Problem scenarios

In this study, five problem scenarios were designed to be used in the PBL process by researchers. Four aspects were considered in designing every problem scenario. These were: every problem scenario covered different aspects of the solutions and their physical properties; each problem was related to or linked with a real-life context; the problems had multiple solution paths; and also the problems were designed to link to common student alternative conceptions (Pinarbasi, 2002; Pinarbasi and Canpolat, 2003). All problem scenarios were supported with a title, a picture, a text and keywords. Each scenario, and whether the problem scenarios prepared by the researchers have the required properties of PBL, was evaluated by nine academicians working in chemistry teacher training using the problem evaluation scale developed by Tatar (2007). The problem evaluation scale was a Likert type scale consisting of three levels: sufficient, average and poor. Through this scale, attempts were made to identify whether the problem situations have the required complex structure to attract students' attention and arouse their interest; to push students to do research to comprehend the meaning of the concepts and to use scientific process skills; to make assumptions and decisions and to produce alternative solution options, by asking 9 experts' opinions. In addition, we tried to determine whether these cases helped students to interrelate their existing knowledge with new information, whether the problem cases cover the acquisitions of the course or not, whether more than one solution is found or not, whether the presentation is supported with visual elements or not, and whether these cases have the properties to help students to go from the knowledge level of Bloom's taxonomy to the upper analysis, synthesis and evaluation levels. Necessary amendments were made according to the suggestions and opinions of the instructors and the final version of problem scenarios were given. A sample problem scenario can be seen in Appendix 1.

Procedure

Treatment was completed in classes (each class lasting 50 minutes) in five weeks (four classes per week). Both of the groups were taught by the same researcher, who was experienced in active learning and PBL. In the control group, conventional instruction methods were used. These methods were mainly dominated by lecturing and included in-class discussions and PowerPoint slides and problem solving. In the control group, students mostly took notes and asked questions when they had difficulty in understanding.

The lesson began with researcher explanation about resolution, factors affecting resolution, factors affecting solubility of gases, solubility of gases and the pressure effect.

Solubility is defined as “the maximum amount of substance that can dissolve in any solvent or solution at a certain temperature or pressure”. Later, students were asked the factors that affect the solubility of gases in liquids.

How temperature and pressure affect the solubility of gases in liquids is explained.

Depth intoxication is given as an example of the effect of pressure on the solubility of gases in liquids. Depth intoxication: “Under high-pressure air dissolves more in blood and in body fluids compared to normal pressure. When the diver returns back to the surface that pressure is removed and dissolved N₂ (g) forms small bubbles. These bubbles might cause “depth intoxication” which might result in paralysis or hemiplegia or even in death”. Regarding the effect of pressure on the solubility of gases in liquids, students were asked why gaseous drinks are recommended to be drunk cold. It was mentioned that the solubility of gases in gaseous drinks increases under high pressure and decreases when the pressure is removed. It was also mentioned that the solubility of gases in liquids decreased as the temperature increased. Thermal pollution was given as an example. Additionally, the presence of the small bubbles that occur in tap water at room temperature was explained. Thus, some concepts were discussed by instructor-directed questions. Towards the end of the course, students were informed about Henry’s law and were given sample problems. Volunteer students solved some problems which required algorithms and formulas to arrive at correct answers.

42 students in the experimental group were grouped into seven groups (each group formed of 6 students). While forming the groups, students' pre-test results, General Chemistry-I and General Chemistry Laboratory-I grades, and their social abilities were taken into account. Students' pre-test scores and last semester chemistry grades were classified as high, medium and low and their social abilities were categorized according to communicating, working in a group, active listening, managing time, using technology and library. Students in the groups were encouraged to decide who would be the leader, recorder, timekeeper and reflector.

In the experimental group, firstly students were informed about PBL and its proceedings.

A manual named: “Student and Teacher Manual on PBL” was prepared in order to inform them about the PBL. This manual includes information about the definition of PBL, scientific process steps to be followed and activities to be realized during the implementation phase (team work, preparation of research report, evaluation criteria of group works, leaf tests, student and teacher roles). Under the following headings in the content of the manual: “What is PBL?”, “What are the roles of the moderator in PBL?”, “What are the roles of students in PBL?” “How a good research report can be prepared?”, students were informed before the implementation and attempts were made to clarify issues that were not clear.

The courses in the experimental group were taught in a laboratory which is designed in such a way that the students can sit together with the members of their group to discuss and provide alternative solutions in case of a problem.

The students went through and followed consecutive stages in the experimental group during the PBL process. In stage 1, the students were given a problem scenario in class and told to carefully read it and were encouraged to write their ideas about the problem scenario as shown in appendix 1.

In stage 2, the students identified learning issues related to the problem scenario and organized them around four focus questions (Gallagher et al., 1995) using a “need-to-know” worksheet.

The questions were as follows: (a) What do we know about the problem? (b) What do we need to learn to solve the problem? (c) How can we have sources of knowledge? and (d) Hypothesis on problem solving.

At this step the students read and discussed the problem and developed a possible solution of the problem by using existing knowledge. Following the group discussions, additional questions were identified to solve problem cases.

The researcher also stimulated the students to gain more information on topics such as Henry’s law, depth intoxication, thermal pollution, factors affecting the solubility and dissolution rate of solids, liquids and gases in water, the occurrence of bubbles when tap water is kept at room temperature and how they are formed, the reason behind recommendation to drink fizzy drinks cold as well as the reason behind the fact that most fish prefer to live in cold waters, and what could be the least dangerous thing when a diver dives to depth: using oxygen or helium in his tube.

Hypotheses were identified using brain storming and students planned their research strategy. In addition, towards the end of this phase, task distribution was made among group members. Groups were asked to fill in their worksheet, decide on their research plan and distribution of tasks and to gather together outside the classroom to share the data they obtained individually or as a group (being responsible for the learning of others).

During this time, the tutor aroused the interest of the students to guide the group work. He visited all the groups one by one. The group members recorded their ideas and questions onto this worksheet regularly. In this way, worksheets served as “a central focus point for the unit and problem definition, information gathering, analysis and synthesis of information and problem redefinition” (Senocak et al., 2007). Stage 3 covers the working process outside the classroom. The students gathered data to answer their own questions in the worksheet. Group members collaborated to solve the problem scenarios. They collected data from different resources. For example, they visited the science laboratory and library to do experiments and to pick up printed and electronic resources, and consulted domain-specific experts to better understand some concepts. After this stage, the students, who gathered in a classroom environment, discussed what they had learned during independent study process in their groups, and students, under the direction of their group leader, shared their information and knowledge acquired from various resources.

In stage 4, the students reported on what they had achieved and prepared a report for presentation to the classrooms. Each group gave a 7–8 min oral presentation on what they had learned about their problem scenario and how they reached the problem solution. In this stage, the tutor asked all groups to ask their questions about other groups' presentations if they had any. Then, some students asked questions under the tutor’s guidance and they discussed for a while. The tutor then gave explanations of the solution of the problem scenarios and answered the students' questions on the problem scenario. The students also submitted a group report which documented the group's findings and details of the inquiry process.

Data analysis

The data were analyzed using SPSS 12.0 (Statistical Package for Social Sciences). The significance level was set to 0.05 since it is the most used value in educational studies. While pre-test data were analyzed by the independent samples t-test, post-test data were analyzed by ANCOVA. Although there was not a significant difference in academic achievement pre-test results of the students in both experimental and control group students before the implementation, ANCOVA was used to eliminate the existing difference between groups. Qualitative data were analyzed descriptively and presented in categories into tables.

Results

Comparisons of SPPAAT mean scores

Descriptive statistics related to total mean scores of pre-test and post-test and corrected means for post-test after ANCOVA are presented in Table 5. The control group pre-test total mean scores (X̄ = 37.34, SD = 12.022) were higher than that of the experimental group (X̄ = 33.14, SD = 7.798) while the experimental group post-test total mean scores (X̄ = 73.86, SD = 10.232) were higher than that of the control group (X̄ = 62.23, SD = 15.051). This result showed that students performed better in the experimental group compared to the control group. An independent sample t-test results also confirms that there is a statistically significant difference between students' total mean scores in the post test (t (69) = 3.818; p < .05), although there was no statistically significant difference between the pre-test total mean scores of experimental and control groups (t (69) = −1.753; p > .05).

Table 5 The average values for the levels of the academic achievement

Group	Number	Pre-test		Post-test		Corrected means of post-test after ANCOVA
		Mean	SD	Mean	SD
Experimental (PBL)	36	33.14	7.798	73.86	10.232	74.729
Control (conventional )	35	37.34	12.022	62.23	15.051	61.336

---

ANCOVA results (see Table 6) also confirm the independent sample t-test result recommendation that there is a statistically significant difference between the corrected total mean scores of students in the experimental group (X̄ = 74.729) where PBL was trialed compared to the control group students (X̄ = 61.336) where a conventional teaching approach was used [F(1.70) = 20.450; p < .05; η² = .231)]. These results provide an indicator of difference in students’ understanding of solutions and related concepts.

Table 6 ANCOVA results for levels of academic achievement

Source	Df	Mean square	F	p	η ^2
R ^2 = .264, Adjusted R^2 = .242.a Significant at .05 level.
Corrected model	2	1817.716	12.199	.000	.264
Intercept	1	14913.235	100.085	.000	.595
Group (experimental-control)	1	3047.218	20.450	.000^a	.231
Pre-test	1	1234.050	8.282	.005	.109
Error	68	149.006
Total	71
Corrected total	70

---

The rate of correct answers that students gave in the achievement test applied before and after the implementation on both experimental group students, with whom PBL was used, and control group students with whom conventional instruction was used, was analyzed. The rate of correct answers of both experimental and control group students' to multiple choice, true–false, short answer and open-ended questions in the academic achievement test in both pre-test and post-test are given in Table 7.

Table 7 Correct answering rates of the questions in the academic achievement test

Question types	Control		Experimental
	Pre-test	Post-test	Pre-test	Post-test
Multiple choice questions	40.5	64.1	35.3	73.5
True–false questions	67.0	77.8	58.3	84.5
Short-answer questions	39.0	82.8	39.8	90.3
Open-ended questions
Sound understanding		29.9		43.7
Partial understanding		5.2		16.0
Weak understanding		7.7		4.2
No understanding or leaving blank	84.5	57.0	87.2	36.0

---

According to Table 7, the rate of correct answers given in the multiple choice questions in the achievement test is 35.3% in the pre-test and 73.5% in the post-test in the experimental group; and the result of the same test for the control group is 40.5% in the pre-test and 64.1% in the post-test. Considering these results, the rate of success of the experimental group students in the post-test increased by 38.2% compared with their pre-test results, and also a 23.6% increase is seen in the control group students' success in their post-test results.

The average of the correct answer rate of experimental group students in the true–false questions in the pre-test is 58.3% while the same average is 84.5% in the post-test; the average of the correct answer rate of control group students is 67.0% in the pre-test and 77.8% in the post-test. In addition, the average of the correct answer rate of the experimental group students in short-answer questions is 39.8% in the pre-test and 90.3% in the post-test; the average of the correct answer rate of the control group students is 39.0% in the pre-test and 82.8% in the post-test. When these results are taken into consideration, it is seen that the success rates of the students in the experimental group increased 50.5% post-test, compared to pre-test, in short-answer questions; and 26.2% in true–false questions; while the same results were 43.8% in short-answer questions and 10.8% in true–false questions in the control group students.

There are four open-ended questions in the achievement test. The students' answers to these questions were assessed on a four point scale ranging from sound understanding (3), partial understanding (2), weak understanding (1) and no understanding or leaving blank (0). Each of the answers was evaluated by researchers, after that scores were compared and discussed until an agreement was reached. According to this classification (see Table 7), the rate of answers at a sound understanding level to the open ended questions of the achievement test in the post-test was 43.7% in the experimental group and 29.9% in the control group; while the rate of correct answers at a partial understanding level was 16% in the experimental group and 5.2% in the control group, and the rate of correct answers at a weak understanding level was 4.2% in the experimental group and 7.7% in the control group; and the rate of no understanding or leaving blank was 36% in the experimental group and 57% in the control group. Besides, the average of no understanding or leaving blank in the experimental group students in open-ended questions in the pre-test is 87.2% while the post-test result is 36.0%; the same average in the control group is found to be 84.5% in the pre-test and 57.0% in the post-test. This reveals that there is a 51.2% increase in students' correct answers in open ended questions in the experimental group and a 27.5% increase in the control group students' correct answer rates.

Findings obtained from achievement test and interviews on conceptual learning

This section includes the correct answer rates of some questions in the achievement test as well as data obtained from interviews on conceptual learning. The findings obtained from the interviews and answers in the achievement test applied as a post-test in the experimental and control group are examined under the headings given below.

Types of solutions: Findings on concepts under some terms

Question 1 is intended to measure students knowledge of saturated, unsaturated, supersaturated, dilute and concentrated solutions and it is prepared according to Bloom's revised taxonomy (Krathwohl, 2002). The rate of correct answers to the 1st question in the control group is higher (80%) than the rate in the experimental group (67%). In the semi-structured interviews conducted after the implementation, a similar question was asked to both experimental and control group students. Based on students’ answers, students were asked such probe questions as: “How would you turn a saturated solution into a supersaturated solution?”. The aim in doing so was to collect in-depth data. The results were analyzed and are presented in Table 8.

Table 8 Students' opinions about types of solutions

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group; CG: control group; f = frequency.
Defined unsaturated solution as “a solution which dissolves a lot less than it can dissolve.”	5	2
Defined saturated solution as “a solution which dissolved the amount of substance it can dissolve.”	3	1
Defined saturated solution as “a solution with a certain amount of undissolved substance in it.”	2	—
Defined saturated solution as “a solution in which there is a small amount of solute”.	—	2
Defined supersaturated solution as “solution which dissolved more substance than it can dissolve”	3	1
Defined supersaturated solution as “if the solvent dissolved a lot more than it can dissolve and if there is no available space or sank to the bottom”	1	—
Defined supersaturated solution as “if the solvent dissolves as much substance as it can and sank to the bottom.”	—	2
A saturated solution can be turned into a supersaturated solution by decreasing the temperature.	3	1
When we prepare a saturated solution at a certain temperature and then we decrease the temperature to a degree where solubility is less, generally the excess solute sinks and sometimes no precipitation occurs and the amount of dissolved matter in the solution is more than the necessary amount to dissolve at that temperature, and this solution is called supersaturated solution.	1	—
A saturated solution can be turned into a supersaturated solution by decreasing the amount of solvent in it.	1	—
By increasing the amount of solute, a solution can be turned into supersaturated.	1	—
A saturated solution can be turned to supersaturated by increasing the temperature and pressure.	—	1

---

Findings on concepts under a molecular viewpoint to solution formation

The rate of correct answers to Question 11, which is asked to question if the given statements about a solution composed of two different liquids are true or false, is higher in the experimental group (72%) than the control group (37%). When this question is analyzed it is seen that 20% of the students in the experimental group and 52% of the students in the control group have such an alternative conception: “the volume of a solution composed of two different liquids is equal to the total volume of the liquids”. Besides, another alternative conception in 3% of the control group students is that “solute has no volume”. Regarding the Question 12 on insolubility of oil in water, it is seen that the rate of correct answers of the control group students (57%) is lower than the correct answer rates of the experimental group (61%). In this question, it is seen that there is an alternative conception in 28% of the experimental group students and in 37% of the control group students. This alternative conception is “water and oil molecules do not repel each other”. 52% of the students in the experimental group and 55% of the students in the control group are determined to reply to the open-ended Question 22, intending to identify students’ sound understanding of the concept of insolubility of oil in water, the diffusion of oil in water when detergent is added and the role of detergent in this process, and 31% of the students in the experimental group and 9% in the control group are found to reply at a partial understanding level. The same question is asked to the students who participated in the interview. The data of the interview was analyzed, and the results are presented in Table 9.

Table 9 Students' opinions about the diffusion of oil in water when detergent is added and the role of the detergent in this process

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
Detergent is a surface-active substance	5	—
Detergent decreases surface tension	1	—
Detergent has two poles, which are hydrophilic and hydrophobic poles and the hydrophilic pole of the detergent interacts with water while the hydrophobic pole interacts with oil	3	2
Water is a polar substance while oil is an apolar substance.	1	1
Hydrophilic pole of the detergent is polar and hydrophobic pole is apolar	5	1
Since detergent dissolves oil in water it becomes a homogeneous mixture	—	1

---

Question 10 intends to identify what kind of interactions occur between molecules during the dissolution of salt in water. The rate of correct answers to this question was higher in the experimental group (89%) compared to the control group (83%). In this question, 1 student from both the experimental and control groups thought that “the ions of salt will take the place of air molecules in water”. 6% of the students in the experimental group and 14% of the students in the control group expressed that “a new substance is formed with the dissolution of salt in water”. The same question was asked again to the students who participated in the interview. The data of the interview was analyzed, and the results are presented in Table 10.

Table 10 Students' opinions about the molecular viewpoint of solution formation

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
(+) ions interact with (−) part of the H2O while (−) ions interact with (+) part.	5	2
Crystal lattice in an ionic molecule begins to fragment and the ions that become free are surrounded with water molecules and this is called hydration.	4	1
The interaction between solvent–solvent, solute–solute molecules is broken and instead solvent–solute interaction occurs.	3	1

---

Findings on the concepts of colloidal mixtures

In Question 4, students are asked if the given samples are solutions or not. In addition, in Question 14, some colloid types are given and students are asked if the given samples are matched correctly according to the dispersed phase, dispersion environment and type. The rate of correct answers to Question 4 in the experimental group is higher (81%) than the rate of correct answers in the control group (54%). Additionally, the rate of correct answers to Question 14, a multiple-choice question, in the control group is higher (71%) than the rate of correct answers in the experimental group (69%). In another question (Question 15), student knowledge on the statements given on the insolubility of sand in water and on the fact that silica does not sink to the bottom for years is questioned. The rate of correct answers to this question in the experimental group is higher (42%) than the rate of correct answers in the control group (37%). In this question 3% of the students in the experimental group and 9% of the students in the control group do not agree that “the reason behind the insolubility of sand in water is because of the interaction between sand particles is stronger than the interaction between the molecules in water”. Additionally, 22% of the students in the experimental group and 11% of the students in the control group stated that: “it is not the particle surface absorption effect or ion holding effect in the mixture that is the most important factor suspending colloid silica particles.” A question like: “how would you understand if a mixture is a solution or a colloidal mixture” is also asked to the students in the interview. The data of the interview was analyzed, and the results are presented in Table 11.

Table 11 Students' opinions about colloidal mixtures

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
Tyndall effect is the method used to determine if a mixture is a solution or a colloidal mixture.	5	2
When light comes from a real solution, if the observer looks directly at the point where light comes from, s/he can not see the light.	5	2
In colloidal diffusion, however, light can be easily seen since it is diffused in all directions.	5	2
In order for a substance to be classified as colloidal either one or two dimensions of the material should be almost 1–1000 nm	2	—
It is the particle surface absorption effect or ion holding effect in the mixture that suspends colloidal silica particles.	—	2

---

Findings on the concepts of factors that affect solubility and dissolution rate

Question 3 intends to identify whether students know the factors that affect the solubility of gaseous substances. The rate of correct answers to this question in the experimental group is higher (83%) than the rate of correct answers in the control group (71%). In addition, students are asked the reason behind the increase in gas bubbles in a glass of cola when some salt or sugar is added. Question 21, an open-ended question, is answered at a sound understanding level by 70% of the students in the experimental group and 31% of the students in the control group, while it is answered at partial understanding level by 8% of the students in the experimental group and 6% of the students in the control group. The same question is also asked to the students who participated in the interview. The data of the interview was analyzed, and the results are presented in Table 12.

Table 12 Students' opinions about factors that affect solubility and dissolution rate

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
When sugar or salt, whose solubility in water is better than carbon dioxide, is added into cola, water molecules leave some carbon dioxide molecules and interact with salt and sugar molecules, since sugar and salt molecules can be dissolved better.	8	2
Although small, there is an interaction between carbon dioxide molecules inside cola and the water molecules.	5	—
Stated that the reason behind this is the “pressure”	—	1
The reason is the rise of the gas up in bubbles because of the increase in amount.	—	1
Since the density of the added substances is high, they substitute gas bubbles.	—	1
Since the solubility of a gas in water is directly proportional to the pressure of the gas on liquid, the number of bubbles increases.	—	1
The dissolution of the added substances in water is endothermic. Since gases will dissolve better in low temperatures when the temperature of the medium decreases, there will be an increase in bubbles.	—	1

---

Question 5 aims to determine student knowledge of factors that affect the solubility and dissolution speed of solids, liquids and gases in water. The rate of correct answers to Question 5 in the experimental group is higher (72%) than the rate of correct answers in the control group (66%). In addition, Question 23, an open-ended question, questioning what the small bubbles consist of when tap water is kept at room temperature, is answered at a sound understanding level by 50% of the students in the experimental group and 28% of the students in the control group, while it is answered at partial understanding level by 6% of the students in the experimental group and 3% of the students in the control group. The students that participated in the interview are also asked the factors that affect the solubility and dissolution speed of solids, liquids and gases in water. With various probe-style questions, the factors that affect solubility of three states of matter and how they affect it are questioned. The data of the interview was analyzed, and the results are presented in Table 13.

Table 13 Students' opinions about factors that affect the solubility and dissolution speed of solids, liquids and gases in water

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
Among all states of matter, only temperature affects solubility and dissolution rate.	4	2
As the pressure increases for the gaseous state of the substance, the solubility of gases in liquids increases as well.	3	—
The bubbles that occur when tap water is kept under room temperature for some time are formed of already dissolved air when the water was cold.	4	—
The solubility of gases in liquids decreases as the temperature increases.	7	—
Regarding the recommendation to drink carbonated drinks cold, the reason behind drinking them cold is the solubility of gases in liquids at low temperatures is higher.	4	2
Regarding the depth intoxication, since air dissolves more in blood and bodily fluids under high pressure compared to normal pressure, when the diver rises to the surface, the pressure disappears and dissolved N2 gas forms small bubbles.	6	2
The bubbles that occur when tap water is kept at room temperature are due to evaporation.	—	2
Since the solubility of gases such as chlorine, minerals etc. in tap water will decrease, they will show up as bubbles.	—	3

---

Findings on concepts under colligative properties

The rate of correct answers to Question 18, which requires mathematical operation regarding the preparation of water + ethylene alcohol mixture, is lower in the experimental group (64%) than in the control group (66%). In addition, in Question 24, an open-ended question, it is questioned whether students know the reason behind the fact that the freezing point of the solution that occurs when a solute is added to pure solvent is lower than that of the pure solvent at molecular level. It is found that the answer to this question is known at a sound understanding level by 3% of the experimental group students and by 6% of the control group students and at a partial understanding level by 19% of the experimental group students and 3% of the control group students. The same question is asked again to students who participated in the interview. The data of the interview was analyzed, and the results are presented in Table 14.

Table 14 Students' opinions about concepts of colligative properties

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
Since antifreeze is more irregular compared to pure solvent it requires more energy to turn regular, thus, its freezing point decreases.	4	—
Since the amount of solute in orange is more than in lemon, orange has a lower freezing point.	5	2
The liquid sprayed on airplanes on cold winter days is propylene glycol and it is used to lower freezing point.	5	2
The reason behind pouring salt on the road in winter is to lower the freezing point.	4	2

---

The rate of correct answers to Question 9, which asks students to calculate partial vapor pressures of hexane and pentane by knowing the mole fraction of a solution which is formed with the vapor pressures of pure hexane and pentane at a certain temperature, is higher in the experimental group (72%) than in the control group (54%). Besides, students who participated in the interview are asked to explain the reason behind the fact that the vapor pressure of a sugar–water solution is lower than the vapor pressure of pure water at the same temperature. The data of the interview was analyzed, and the results are presented in Table 15.

Table 15 Students' opinions about the reason behind the fact that the vapor pressure of a sugar–water solution is lower than the vapor pressure of pure water at the same temperature

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
The molecules of the solute blocks part of the surface and decreases the escaping speed of the solvent molecules from the solution.	3	1
The vapor pressure of a solution of non-volatile matter is in direct proportion with the mole fraction of the solvent.	2	1
Since the number of particles that hits the surface decreases, so does the vapor pressure.	1	—
The vapor pressure of a sugar–water solution is higher than the vapor pressure of pure water of the same temperature and the reason is the increase in impurities.	—	3
The vapor pressure of a sugar–water solution is equal to the vapor pressure of pure water of the same temperature and without a change in temperature pressure does not change.	—	2

---

The rate of correct answers to Question 6, which aims to identify whether osmosis, reverse osmosis, osmotic pressure and hydrostatic pressure has been learned, is higher in the experimental group (69%) than in the control group (66%). In addition, the rate of correct answers to Question 8, which requires mathematical calculation of the molar mass of solute in a solution whose osmotic pressure is given, is lower in the experimental group (81%) than the rate of correct answers in the control group (83%). Students who participated in the interview are also asked some questions on osmosis, reverse osmosis and probe-type questions. The data of the interview was analyzed, and the results are presented in Table 16.

Table 16 Students' opinions about osmosis and reverse osmosis

Student opinions on the concepts	EG (f)	CG (f)
EG: experimental group, CG: control group; f = frequency.
The reason behind the shrinkage of cucumber after it is kept in concentrated salty water for some time is reverse osmosis.	3	1
Sea water can be separated from salt via reverse osmosis method.	2	1
If a pressure higher than the osmotic pressure is applied to the solution inside a semi-permeable membrane, this pressure reverses the flow of the solvent and directs it to flow from solution to pure solvent and this is called reverse osmosis.	4	—
The process in which water moves from an area in which it is present in high concentrations to another area with low concentration with the help of semi-permeable membrane.	3	—
Osmosis is the passing of water molecules from a dilute sugar solution to a concentrated sugar solution.	—	2
If the concentration of the water is higher than the concentration of ions in the solution, the transfer from water to solution is called osmosis.	—	1
The pressure that should be applied to a solution in order to stop osmotic passing is called osmotic pressure.	1	—
Explained reverse osmosis as the reason behind the shrinkage of cucumber after it is kept in concentrated salty water after some time.	—	1

---

Comparisons of SPST mean scores

The total mean scores of SPST for pre-intervention were X̄ = 20.62 and X̄ = 21.79 for the experimental and control groups respectively. An independent sample t-test indicates that there was no statistical difference between the students' scientific process skills before the intervention (t = −1.419; p > .05). After the intervention, the SPST was re-administered and the total mean scores were found as X̄ = 23.59, (SD = 4.091), and X̄ = 22.65, (SD = 4.227), for experimental and control groups respectively (Table 17).

Table 17 The average values for the scientific process skill levels

Group	Number	Pre-test		Post-test		Corrected means of post-test after ANCOVA
		Mean	SD	Mean	SD
Experimental (PBL)	34	20.62	3.229	23.59	4.091	24.004
Control (conventional )	34	21.79	3.599	22.65	4.227	22.231

---

ANCOVA results (see Table 18) also confirm that there is a statistically significant difference between the corrected total mean scores of students in the experimental group (X̄ = 23.59) where PBL was trialed compared to the control group students (X̄ = 22.65) where conventional instruction was used [F(1.67) = 4.451; p < .05; η² = .064)]. These results provide an indicator of difference in students' scientific process skills.

Table 18 ANCOVA results for the levels of scientific process skills

Source	Df	Mean square	F	p	η ^2
R ^2 = .346. Adjusted R^2 = .326. *Significant at .05 level.
Corrected model	2	400.261	17.189	.000	.346
Intercept	1	110.737	9.511	.003	.128
Group (experimental-control)	1	51.825	4.451	.039	.064
Pre-test	1	385.202	33.084	.000	.337
Error	65	756.798
Total	68	37 498.000
Corrected total	67	1157.059

---

Students performed better in the experimental group compared to the control group. Because of the students’ use of scientific process skills at almost every stage in PBL, this result was expected.

Students' views about the PBL

Students’ perceptions were determined about the PBL process with the scale which was developed by researchers. In the scale there are statements covering the advantages and disadvantages of the PBL. With the statements in the scale, the aim was to identify if there was a change in students’ attitude and motivation towards chemistry classes and what kind of skills they acquired during this process. Table 19 is formed in line with the answers that 40 students in the experimental group gave to the statements in this scale.

Table 19 Students' views about the PBL

At the end of this study	%
My skill in accessing and using sources developed.	90.0
My individual learning skills developed.	90.0
My critical thinking skills developed.	92.5
My problem solving skills developed.	75.0
My scientific transaction skills (making observations, classification, measurement, forecast, deduction) developed.	82.5
My skills to work in a group and in collaboration with a group developed.	72.5
My communication skills developed.	80.0
I developed high motivation for chemistry classes.	75.0
I developed a positive attitude for chemistry classes.	87.5
I believe that the information I learned on solutions and their physical properties is permanent.	92.5
I had the opportunity to participate actively in the class.	90.0
This method served as a model to us as teacher candidates.	82.5
My science literacy developed.	80.0
I could not get used to the method.	12.5
I had problem with assessment.	45.0
I had a fear about having imperfect knowledge.	17.5
I had a problem with the structure of the groups and with inadequate cooperation.	57.5
I had a problem with limited time.	62.5
I had hesitations about the quality of the knowledge I learned.	5.00
I had the difficulty of not having a course book or a curriculum.	22.5
I had problem with accessing resources.	10.0

---

Conclusions and discussion

Since it is not possible to create an active learning environment for students in teacher-centered instruction methods, it is not possible to train students to be equipped with the required scientific skills of today's society. The analysis results of the tests revealed that there is no significant difference (p > .05) between experimental and control group students in terms of success in solutions and their physical properties concepts and in scientific process skills. Academic success tests on solutions and their physical properties were applied on both groups after the subject on comparing two learning/teaching methods of learning concepts of solutions was discussed.

Post-test results showed a statistically significant difference (p < .05) in terms of academic achievement between the experimental and the control groups; and the experimental group post-test total (corrected) mean scores (X̄ = 74.729) were higher than that of the control group (X̄ = 61.336). According to this result, it is said that PBL is more effective than conventional instruction on the subject of solutions and their physical properties. This result is in parallel with the results of experimental research which was based on the PBL within the framework of science courses (Diggs, 1997; Dobbs, 2008; Kelly and Finlayson, 2007, 2009; Tarhan and Acar, 2007; Tarhan et al., 2008 and Tatar, 2007).

In addition, findings obtained from interviews on conceptual learning showed that students have better and more accurate conceptual understanding in the experimental group compared to the control group. Conceptual learning was more detailed in the experimental group compared to the control group. The results showed that the students in the PBL class were also better in terms of sound understanding on the open-ended questions. This can also be deduced from the higher rate of correct answers in the experimental group to questions in the achievement test, which are of higher level according to Bloom's revised taxonomy.

The analysis of students' responses showed that, while the control group students had many alternate conceptions about solutions and their physical properties, the experimental group students presented fewer alternate conceptions. It is found that they have alternate conceptions regarding the reason behind the increase in the number of gas bubbles in a glass of cola after some salt and sugar is added. The students in the control group ascribed it to pressure, increase in amount and to the excess density of the added substance. In addition, one student from the control group said: “since the added solid substances’ dissolution in water is endothermic, when the temperature in the setting decreases, an increase in bubbles occurs since gases dissolve better at low temperatures”, and showed he has an alternative conception. Besides, the students in the control group think that the degassing occurring in tap water when it is kept in room temperature is because of evaporation and of the decrease in solubility of such gases as chlorine, minerals etc., at room temperature.

Some students from the control group think that the vapor pressure of a sugar–water solution is higher than the vapor pressure of pure water at the same temperature, and the reason behind this is thought to be the increase in impurities. Another group of students from the control group gave a different response: “the vapor pressure of a sugar–water solution is equal to the vapor pressure of pure water at the same temperature and since temperature does not change so does the pressure”.

It is seen that both experimental and control group students have a correct understanding of basic concepts when they are asked such questions as “What is the role of detergent in the diffusion of oil in water? How do you understand if a mixture is a solution or a colloidal mixture?” and “some salt is added and mixed into a bowl of water. What kind of an interaction occurs between the particles in the dissolution of salt in water?”.

Regarding supersaturated solutions, on the other hand, it is seen that there are alternate conceptions in both experimental and control group students. It is also determined that some students both from experimental and control groups have alternate conceptions such as: “in order to turn a saturated solution into a supersaturated solution, the amount of solvent is increased/the amount of solute is increased and it can be done by the effect of the pressure.”

The concepts that are more frequently seen in the control group students than in the experimental group students can be listed as: the volume of a solution composed of two different liquids is equal to the total volume of the liquids, the reason behind the insolubility of sand in water is because the interaction between the particles in sand is stronger than the interaction between the molecules in water, water and oil molecules do not push each other and as a result of the dissolution of salt in water, a new chemical material occurs.

The rate of students who stated that “it is not the particle surface absorption effect or ion holding effect in the mixture that is the most important factor suspending colloidal silica particles” in the experimental group is higher than the rate of students who stated the same in the control group. The rate of students who think that the ions of salt will take the place of air molecules in water is equal in the experimental and control group students. Besides, the alternate conception that solute does not have a volume is only seen in the control group students.

Most of the students (92.5%) in the experimental group expressed the opinion that PBL increased the retention of knowledge in their minds. Among the reasons for this situation are such factors as beginning the subject with an attention-grabbing scenario from everyday life, providing an opportunity for students to express their opinions in discussions in the classroom, providing access for students to reach information about the subject, sharing the information that each group found in the classroom and moderator's help in making the necessary explanations regarding the problematic situation can be listed. In view of the above, greater retention of learning in the mind is expected. Yucelis-Alper (2003), Tatar (2007), Wong and Day (2009) obtained the same test results.

One of the aims of this study is to examine the effect of PBL and conventional instruction on the scientific process skills of the students. According to the data obtained, post-test results showed a statistically significant difference (p < .05) in terms of scientific process between the experimental and the control groups. It was also found that experimental group post-test total (corrected) mean scores (X̄ = 24.004) were higher than that of control group (X̄ = 22.231). As PBL promotes the use of scientific process skills at almost every stage of student work, this result is an expected one.

Scientific process skills test is used to determine the changes in students' scientific skills. Also, a scale to determine student opinions on PBL is utilized to determine how much students use these skills. According to the data obtained from student views; more than 70% of the students stated that PBL has a positive contribution to students' access to and utilization of the resource, their self-learning ability, critical thinking skills, scientific process skills (making observations, classification, measurement, forecast, and deduction), ability to work in a group cooperatively, communication skills, research skills, problem-solving skills. During the PBL process students are asked to follow a series of procedures including defining the problem, producing a hypothesis, testing the hypothesis, designing research, planning the research, distribution of tasks among the group members, accessing sources of information, analysing the data obtained, preparing possible solutions, reporting and presenting these solutions. Since students realize the process by themselves in each of the scenarios, they will learn which steps to follow in solving the problem when faced with it in everyday life. These results comply with the findings of the study by Diggs (1997), Ram (1999) and Tatar (2007).

Although PBL offers many advantages, its application seems to be difficult in Turkey as in most countries. In this study, more than half of the students complained about limited time (62.5%) during the implementation of PBL while 57.5% complained about the group structure and inadequate cooperation and 45% about the assessment problems. Also, 20% of the students are indecisive about these situations. It is seen that some students have some problems relating to distribution of tasks in groups and that instead of struggling to solve the problem everyone prefers individual study. When the groups were formed, the researcher forming the groups considered students' grade point averages from General Chemistry-I and General Chemistry Laboratory-I, and thus a heterogenous structure was created inside the group while a homogenous structure was created outside the group. This might be a complaint of the students who could not take part in the same group although they wanted to be team mates. In the first weeks of the implementation there was a negative competition environment in the classroom in which relative evaluation system was applied. Students were not used to making objective assessments so they caused problems with the low grades they gave to their friends in the beginning of the implementation. However, towards the last weeks of the implementation, this unnecessary competition environment was removed. This result complies with the finding of Ram (1999) in which he said that while students avoided using negative statements while assessing themselves and their friends, towards the end of the term they turned out to be more objective and eager with the support of the lecturer.

The biggest problem of the students during PBL implementations was found to be the problem with time limits. PBL is not limited to classroom activities. It requires students to gather together outside the classroom, prepare reports, discuss with their friends and prepare presentations. Therefore, in PBL time for outside classroom activities is required as well as classroom activities. Besides, time limitation is not only a problem for students in PBL but also a problem for teachers. First of all, preparing the problem situations, controlling the students during the whole process and assessing the students also requires a long time for teachers. Therefore, the fact that there are limited number of class hours and there is an intense curriculum are negative elements for the applicability of PBL.

A small group of students (12.5%) stated that they could not get used to the method and 10% of them said they had problems in accessing the resources; and the rate of students who were undecided was 15% and 7.5% respectively. On the other hand, 17.5% of the students stated that they were afraid of having imperfect knowledge, while 5% of the students stated that they had hesitations about the quality of the information they learned, and 22.5% of them stated that they had experienced the difficulty of not having a certain coursebook and a curriculum. This shows that students who are used to conventional instruction are having difficulty in adapting to PBL practices. In interviews, some students were complaining about the inconsistency between the problem scenarios and the questions used in SPPAAT, as the scenarios mostly focussed on conceptual understanding while the SPPAAT contained more algoritmic questions. Compared to the control group, the rate of correct answers to Questions 8, 13 and 18 was lower in the experimental group. It is seen that these questions are at an application level according to Bloom's revised taxonomy (Krathwohl, 2002) and solving these questions require some mathematical procedures. Contrary to conventional classes in which mathematical problems are successfully solved, in PBL implementations not many sample problem situations are solved, which may explain such a result. These results are consistent with the results of the studies of Smith and Metz (1996), Pinarbasi and Canpolat (2003).

In order to overcome these important difficulties in the transition to PBL from teacher-centred learning, the curriculum should be revised, PBL activities should be developed and validated, teachers should be trained and rewarded for PBL participation, and the infrastructure of schools should be developed for PBL. Besides, while forming groups in PBL sessions, the number of students in each group has to be kept to a minimum in order to increase active participation. Since it is not possible to practice PBL in every course or in every subject, it is better to implement it in certain cases where the subject is related to daily-life activities.

Appendix 1. A sample problem scenario used in this study

“SCUBA DIVER”

Engineers working at the project development department of an international research center are working on a new project. They decide to design a closed-cycle regulator system in order to minimize the number of depth intoxication cases that divers face in deep dives and to make longer dives and to carry less diving tube and gas.

Engineers first examine the situations divers experience in deep waters and summarize it as:

Under high-pressure air dissolves more in blood and in bodily fluids compared to at normal pressure. When the diver returns back to the surface that pressure is removed and dissolved N_2(g) forms small bubbles. These bubbles cause intense pain in joints and vessels and affect the nervous system. Most important of all, these bubbles might cause “depth intoxication” which might result in paralysis or hemiplegia or even in death. In order to minimize this danger in diving the engineers began to work to add the following properties to the previous version of the open-circuit regulator system.

The most significant difference between closed-circuit regulators and open-circuit regulators is that in closed-circuits gas is flushed to the breathing cycle from two gas tubes. One of them is a pure oxygen tube and the other is a composite gas cylinder. In the composite gas cylinder there is either oxygen + nitrogen (nitrox), oxygen + helium (heliox) or oxygen + helium + nitrogen (trimix). The gas that will be used in the dive changes according to the maximum depth of the dive and the time of the depth time.

(a) If you were to make scuba diving using the new invention of the engineers, the closed-circuit regulator system, which gas composite (oxygen + nitrogen, oxygen + helium or oxygen + helium + nitrogen) do you think would be the least dangerous?

(b) If we accept that you have oxygen + helium + nitrogen (trimix) in your composite gas cylinder, what should be the composition rates of the gases in your diving tube in order to reach a 100 meter depth without depth intoxication? (Air consists of %78-nitrogen, %21-oxygen and %1-other gases).

Note: Please consider that if the partial pressure of O₂ in the gas breathed is over 1.6 atm, it will increase the risk of oxygen toxicity and N₂ will have a narcotic effect as of 30 meters.

Acknowledgements

The authors are indebted to the Department of Chemistry Education (Kazim Karabekir Education Faculty) and Ataturk University for financial support. We also thank Dr E. Senocak, Dr K. Doymus and Dr M. Sozbilir for their useful suggestions and discussions.

References

1. Amer A., (2006), Reflections on Bloom's Revised Taxonomy, Elec. J. Res. Educ. Psychol., 8, 4(1), 213–230.
2. Atasoy, B., Kadayifci, H. and Akkus, H. (2003). Lise 3.sınıftaki öğrencilerin kimyasal bağlar konusundaki yanlış kavramaları ve bunların giderilmesi üzerine yapılandırmacı yaklaşımın etkisi [The misconception of students in the 11th grade of high schools as regards chemical bonds and the influence of the constructive approach on the elimination of IT]. Türk Eğitim Bilimleri Dergisi,1(1), 61–79.
3. Belt, S. T., Evans, E. H., McCreedy, T., Overton, T. L. and Summerfield, S. (2002). A problem-based learning approach to analytical and applied chemistry. Univ. Chem. Educ.,6, 65–72.
4. Blanco, A. and Prieto, T. (1997). Pupils' views on how stirring and temperature affect the dissolution of a solid in a liquid: A cross-age study (12 to 18). Int. J. Sci. Educ., 19(3), 303–315.
5. Calik, M., Ayas, A. and Ebenezer, J. V., (2005). A review of solution chemistry studies: Insights into students' conceptions. J. Sci. Educ. Technol., 14(1), 29–50.
6. Canpolat, N. (2002). Kimyasal Denge ile İlgili Kavramların Anlaşılmasında Kavramsal Değişim Yaklaşımının Etkinliğinin İncelenmesi [Investigation of Effectiveness of Conceptual Change Approach on Understanding of Chemical Equilibrium Concepts]. Unpublished Doctoral Dissertation, Ataturk University, Erzurum.
7. Chin, C. and Chia, L. (2004). Problem-based learning: Using students' questions to drive knowledge construction. Sci. Educ.,88(5), 707–727.
8. Delisle, R. (1997). How to use problem-based learning in the classroom. Alexandria, VA: Association for Supervision and Curricuum Development.
9. Diggs, L. L. (1997). Student Attitude Towards and Achievement in Science in a Problem Based Learning Educational Experience. PhD Thesis, University of Missouri-Columbia.
10. Dobbs, V. (2008). Comparing Student Achievement in the Problem-Based Learning Classroom and Traditional Teaching Methods Classroom. PhD Thesis, Walden University.
11. Ebenezer, J. V. and Erickson, L. G. (1996). Chemistry students' conception of solubility: A phenomenograpy. Sci. Educ.,80(2), 181–201.
12. Ebenezer, J. V. and Fraser, M. D. (2001). First year chemical engineering students' conception of energy in solution processes: Phenomenographic categories for common knowledge construction. Sci. Educ.,85, 509–535.
13. Gallagher, S. A., Stepien, W. J., Sher, B. T. and Workman, D. (1995). Implementing problem-based learning in science classrooms. Sch. Sci. Math.,95(3), 136–146.
14. Geban, O., Aksar, P. and Ozkan, I. (1992). Effects of computer simulations and problem-solving approaches on high school students. J. Educ. Res.,86(1), 5–10.
15. Groh, S. E. (2001). Using problem-based learning in general chemistry. In Duch, B.J, Groh, S.E., and Allen, D.E. (Eds.), A practical “how to” for teaching undergraduate courses in any discipline: The power of problem-based learning, pp. 121–131, Sterling, V.A: Stylus Publishing.
16. Gurses, A., Acikyildiz, M., Dogar, C., and Sozbilir, M. (2007). An investigation of effectiveness of problem-based learning at physical chemistry laboratory. Res. Sci. Technol. Educ.,25, 99–111.
17. Johnson, B. and Christensen, L., (2004). Educational Research: Quantitative, Qualitative and Mixed Approaches. Pearson Education, Inc., Second edn, 562 p, Boston.
18. Kelly, O. and Finlayson, O. (2009). A hurdle too high? Students' experience of a PBL laboratory module. Chem. Educ. Res. Prac.,10, 42–52.
19. Kelly, O. C., and Finlayson, O. E. (2007). Providing solutions through problem based learning for the undergraduate 1st year chemistry laboratory. Chem. Educ. Res. Prac.,8(3), 347–361.
20. Krathwohl, D. R. (2002). A revision of Bloom's Taxonomy: An Overview. Theory into Practice,41(4), 212–218.
21. Larive, C. K. (2004). Problem-based learning in the analytical chemistry laboratory course. Anal. Bioanal. Chem.,380, 357–359.
22. McDonald, J. T. (2002). Using problem-based learning in a science methods course. Proceedings of the Annual International Conference of the Association for the Education of Teachers in Science, ERİC ED 465 621, Charlotte.
23. Mcmillan, J. H. and Schumacher, S. (2006). Research in Education: Evidence-Based Inquiry, 6th edn, 517 p, London, UK.
24. Morgil, I., Erdem, E. and Yılmaz, A. (2003). Kimya eğitiminde kavram yanılgıları [Misconceptions in chemistry education]. Hacettepe Üniversitesi Eğitim Fakültesi Dergisi,25, 246–255.
25. Okey, J. R., Wise, K. C. and Burns, J. C. (1982). Integrated process skills test. Unpublished manuscript, Department of Science Education, The University of Georgia, Athens, Georgia, USA..
26. Pınarbasi, T., (2002). Çözünürlükle İlgili Kavramların Anlaşılmasında Kavramsal Değişim Yaklaşımının Etkinliğinin İncelenmesi [Investigations of Effectiveness of Conceptual Change Approach on Understanding of Solubility Concepts]. Unpublished Doctoral Dissertation, Ataturk University, Erzurum.
27. Pinarbasi, T. and Canpolat, N. (2003). Students' understanding of solution chemistry concepts. J. Chem. Educ.,80(11), 1328–1332.
28. Ram, P. (1999). Problem-based learning in undergraduate education: A sophomore chemistry laboratory. J. Chem. Educ.,76(8), 1122–1126.
29. Rivarola, V. A., Bergesse, J. R., Garcia, M. B. and Fernandez, A. C. (1997). A problem-based learning approach to a biological chemistry laboratory class for students of veterinary science. Biochem. Educ.,25(1), 22–23.
30. Schmidt, H. G. (1983). Problem based learning: Rationale and description. Med. Educ., 17(1), 11–16.
31. Senocak, E., Taskesenligil, Y., and Sozbilir, M. (2007). A study on teaching gases to prospective primary science teachers through problem-based learning. Res. Sci. Educ.,37, 279–290.
32. Serin, G. (2009). Probleme Dayalı Öğrenme Öğretiminin 7. Sınıf Öğrencilerinin Fen Başarısına, Fene Karşı Tutumuna ve Bilimsel Süreç Becerisine Etkisi [The Effect of Problem Based Learning Instruction on 7th Grade Students' Science Achievement, Attitude Toward Science and Scientific Process Skills]. Unpublished Doctoral Dissertation, Middle East Technical University, Ankara.
33. Smith, K. J. and Metz, P. A. (1996). Evaluating student understanding of solution chemistry through microscopic representations. J. Chem. Educ., 73(3), 233–235.
34. Taber, K. S. (2011). Constructivism as Educational Theory, Contingency in Learning, and Optimally Guided Instruction. Educ. Theory, pp. 39–61.
35. Tarhan, L., Ayar-Kayali, H., Ozturk-Urek, R. and Acar B. (2008). Problem-based learning in 9th grade chemistry class: Intermolecular forces. Res. Sci. Educ.,38(3), 285–300.
36. Tarhan, L., and Acar, B. (2007). Problem based learning in an eleventh grade chemistry class: factors affecting cell potential. Res. Sci. Technol. Educ.,25(3), 351–369.
37. Tatar, E. (2007). Probleme Dayalı Öğrenme Yaklaşımının Termodinamiğin Birinci Kanununu Anlamaya Etkisi [Effect of Problem Based Learning Approach on Understanding of the First Law of Thermodynamics]. Unpublished Doctoral Dissertation, Ataturk University, Erzurum.
38. Tosun, C. and Taskesenligil, Y. (2011). Revize edilmiş Bloom'un taksonomisine göre çözeltiler ve fiziksel özellikleri konusunda başarı testinin geliştirilmesi; geçerlik ve güvenirlik çalışması [Development of an achievement test about solutions and their physical properties based on Bloom's revised taxonomy: validity and reliability]. Kastamonu Eğitim Dergisi,19(2), 499–522.
39. Uden, L. and Beaumont, C. (2006). Techonology and Problem-Based Learning. Information Science Publishing, 344 p, London, UK.
40. Ward, J. D. and Lee, C. L. (2004). Teaching strategies for FCS: Student achievement in problem-based learning versus lecture-based instruction. J. Fam. Consum. Sci.,96(1), 73–76.
41. Wong, K. K. H. and Day, J. R. (2009). A comparative study of problem-based and lecture based learning in junior secondary school science. Res. Sci. Educ.,39, 625–642.
42. Yildirim, A. and Simsek, H. (2006). Sosyal Bilimlerde Nitel Aratırma Yntemleri [Qualitative Research Methods in Social Sciences]. 5th edn, Seckin Publishing, 366 p, Ankara.
43. Ying, Y. (2003). Using problem based teaching and problem based learning to improve the teaching of electrochemistry. The China Papers, July, 42–47.
44. Yu, L. (2004). Using a problem-based learning approach to improve the teaching quality of analytical chemistry. The China Papers, July, 28–31.
45. Yuzhi, W. (2003). Using problem based learning in teaching analytical chemistry. The China Papers, July, 28–33.
46. Yucelis-Alper, A. (2003). Web Ortamlı Probleme Dayalı Öğrenmede Bilişsel Esneklik Düzeyinin Öğrenci Başarısı ve Tutumları Üzerindeki Etkisi [The Effect of Cognitive Flexibility on Students' Achievement and Attitudes in Web Mediated Problem Based Learning]. Unpublished Doctoral Dissertation, Ankara University, Ankara.
47. Zhang, G. (2002). Using problem-based learning and cooperative group learning in teaching instrumental analysis. The China Papers, October, 4–8.

---

Footnote

† This study is a part of the first author’s doctoral dissertation.

---