Science teaching Methods and scientific literacy

Published:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Abstract

Regions with a strong knowledge-based economy need, now more than ever, their future generations to show a strong scientific literacy. Literature points to the importance of teaching methods in achieving strong scientific literacy; schools are essential in meeting this challenge. The current study, based on the the 2006 data of OECD's Program for International Student Assessment (PISA) focus on the effect of four different teaching methods on student scientific literacy. The study includes 3843 students from 151 schools in Flanders. Controlling for social background, gender and other student or school characteristics, the data are analyzed using multivariate multilevel regression analyses. The results show, surprisingly, that innovative classroom practices, such as class debates and the possibility for students to do their own investigations, have a negative influence on scientific literacy. Relating scientific topics to real situations and doing hands-on activities positively affected scientific literacy. Interest in science also increases when the content of the science lessons is put into a broader perspective. To conclude the results suggest that 15 year students profit the most from a teacher that focuses on models or applications in science and allows students to do hands-on activities.

Keywords: scientific literacy, attitudes towards science, multilevel analysis, teaching methods

Introduction

Knowledge-based economies, such as the Flemish region in Belgium, have a need for graduates with a strong scientific literacy. Therefore schools and teachers should embrace the most effective way of teaching science in secondary education. In 2006, PISA measured student outcomes in science. The results showed Flanders to be at the 8th place of all OECD countries for scientific literacy. Furthermore, PISA also assessed which didactical methods are used in the science lessons. These data create an opportunity to examine which teaching method results in strong student scientific literacy.

The current study is a secondary analysis of the PISA data and adds the connection between student outcome and the teaching method with control for different variables like gender, social status, language and origin, using a multivariate multilevel analysis on the student and school level.

The Flemish government of education determines final terms all students should reach. This is controlled by an inspecting committee but schools can choose the didactical method they use to reach the terms. Therefore we can expect significant differences in teaching methods.

Theory

The last two decades, a lot of research has been done about educational effectiveness (e.g. Muijs & Reynolds, 2001; Creemers & Kyriakides, 2006). Effectiveness is defined as the degree in which schools succeed in obtaining their goals. For our study that is achieving a strong student scientific literacy.

Scientific Literacy

PISA defines scientific literacy as an individual's scientific knowledge and use of that knowledge to identify scientific questions, to acquire new knowledge, to explain scientific phenomena and to draw evidence-based conclusions about science-related issues. Also included in the PISA definition is understanding the characteristic features of science as a form of human knowledge and enquiry, and being aware of how science and technology shape our material, intellectual and cultural environments. Finally, for PISA, scientific literacy also includes the willingness to engage with science-related issues, and with the ideas of science, as a reflective citizen (OECD, 2007).

Scientifically literature students are able to deal with real life situations that involve science in a personal, social and global context. This ability to face such situations requires different competencies, such as identifying scientific issues, explaining phenomena scientifically and using scientific evidence. These competencies are influenced by knowledge and attitudes. Knowledge includes both scientifically content knowledge and knowledge about science.

The two attitudes in the PISA framework are: self efficacy in learning science and general interest in science. Those are directly related to goals of the European government (European Commission 2004) namely that more students study science and choose science as a career opportunity.

Teaching method

Creemers and Kyriakides (2006) describe which factors can enhance the effectiveness of schools. Their results point towards the teacher's instructional role, that was found to be consistently related to student outcomes. The quality of teaching and learning provision, supported by strategic teacher professional development matters most in affecting students' experiences and outcomes of schooling throughout their primary and secondary years (Rowe, 2003). Research on the determinants of high quality education shows that the teacher is the decisive factor (Barber & Mourshed, 2007). Constructivist learning theory, which was first described by Von Glasersfeld (1989), sees learning not only as an addition of information to existing knowledge but also as a reconstruction of what is already known. Learners are not a blank slate, and knowledge cannot be imparted without their making sense of it through a lens of their current conceptions. According to constructivism, children learn best when they are allowed to construct personal understanding based on experiences and reflecting upon those experiences. A constructivist classroom is characterised by students working in groups and learning as being interactive and dynamic; there is an emphasis on social and communication skills and on the exchange of ideas. In a traditional, nonconstructivist classroom, students work alone, learning is primarily achieved by repetition, and textbooks guide subjects.

Apart from the teacher's approach to teaching science, several factors are known to influence student outcomes, like school population, curriculum, social background, language spoken at home and gender. Also self efficacy in studying science and interest in science have been demonstrated to be of influence on student outcomes (OECD, 2006).

To examine the influence of the different styles of education it is necessary to control for school and student characteristics simultaneously in, using a multilevel approach (De Maeyer, van den Bergh, Rymenans, Van Petegem, & Rijlaarsdam, 2010).

School characteristics

Schools have been demonstrated to influence student outcomes (e.g. Hirt, Nicaise, von Kopp, & Mitter, 2007; Van Damme, Van Landeghem, De Fraine, Opdenakker, & Onghena, 2004; Veenstra, 1999).

A relation between average economic, social and cultural status (ESCS) from the school and the students outcomes can be expected (Veenstra, 1999; Van Damme, Van Landeghem, De Fraine, Opdenakker, & Onghena, 2004). The 2003 PISA data demonstrated large differences in student outcomes on mathematics in Flanders (Hirt, Nicaise, von Kopp, & Mitter, 2007). These differences can be partly explained by the school, and more specific by the school's cultural capital (Franquet, De Maeyer, & Kavadias, 2010).

Not only the quality of the lessons will have an effect, also the quantity is important (Creemers & Kyriakides, 2006). The actual time spend on science lessons has to be taken into account. When students choose a more scientific orientated study they will spend more hours in science class. Because the number of hours depends on the actual chosen study subjects and not on the educational track it is necessary to add the number of hours in the research model.

Student characteristics

Every pupil is different. They can differ in background, gender, language or others. Not all specific characteristics have an influence on school results. Social status, origin, language at home and gender are predictive indicators and are often used in school effectiveness research (Muijs & Reynolds, 2001).

Previous PISA research shows that economical, social and cultural status has an influence on students results (De Meyer & Pauly, 2007). In Flanders science achievement is higher than the OECD average, both for students with a low and high socio economic status. On the other hand the impact this status is bigger in Flanders than the OECD average (De Meyer & Pauly, 2007). In Flanders, students whose parents come from another country, score significantly lower than native students. The difference between both groups is also bigger than the OECD average (De Meyer & Pauly, 2007). Off course, the immigrant status is not the same has the language spoken at home. Therefore it is also necessary to take into account that students may speak another language at home than the language used to give instructions at school.

The difference between boys and girls in Flanders is not significant. Boys choose more scientific study options than girls, who prefer social studies. Because of this difference in choice some results hide a difference in science outcomes. The differences are usually bigger on school level than on region level (OECD, 2007).

research questionS

Which constructivist science teaching methods have an effect on students science outcomes

What is the effect of a science teaching methods on students attitude towards science?

What is the effect of a science teaching methods style on students self-efficacy in science?

methodology

Participants

This study reanalyzes the 2006 PISA data. PISA is initiated by the Organization for Economic Co-operation and Development (OECD) to study and compare student achievement. After a representative set of schools had been sampled in all countries, a random sample of 15-year-old students was drawn from each school. This study focuses on the 3843 students Flemish students from 151 schools who are studying general (ASO) and technical (TSO) education because these groups have a similar list of final terms in science and they had the same main curriculum at the first grade in secondary education. Both ASO and TSO are tracks which intend to lead to tertiary education; therefore the focus of this research is on this group of students.

Instruments

The PISA 2006 study was conducted as the third of the program's 3-yearly assessments of student knowledge and skills in mathematics, language and science. In the 2006 cycle, the focus of the assessment was on scientific literacy. In addition, students were administered a questionnaire assessing their background, learning habits, perceptions of their learning environment, and attitude towards science.

The research model:

The model used for this research is based upon the CIPO-model in combination with the PISA model for scientific literacy. The CIPO-model (Scheerens, 1990) refers to the importance of context, input, process and output. The output variables are knowledge and attitudes based on the PISA structure of scientific literacy. The teaching method is a process variable. School and students characteristics are the input and context items we integrate in our model as control variables.

Figure 1: the research model

Variables

Three types of variables are included in our analysis: output measures on the one hand, and school and student characteristics as explanatory variables on the other. Unless mentioned otherwise, the variables described were used as they were constructed by PISA. All information on scale construction can be found in the PISA 2006 technical report (OECD, 2009). All non-categorical variables are standardized scores.

Scientific knowledge

Science abilities were measured by a 30-minute science assessment on knowledge of physical systems, living systems, earth and space, technology systems, scientific enquiry, and scientific explanations. The questions were distributed over items identifying scientific issues, explaining phenomena scientifically, and using scientific evidence (OECD, 2007). PISA 2006 reported five plausible values for science abilities, which were highly correlated (all r(4,997) > 0.93, p < .001). We selected plausible value 1 as a measure for science ability and computed a z-score for this scale using only the Flemish students. 

Attitudes towards science

The index of general interest in science was derived from students' level of interest in learning eight different scientific topics. Examples are human biology, astronomy, requirements for scientific explanations, etc. A four-point scale with response categories from "high interest" to 'no interest" was used for the eight items scale (Cronbachs' alpha 0.82).

Self-efficacy was derived from the students' beliefs in their own ability to perform eight different tasks. Students could answer on a four point scale from "I could do this easily to "I couldn't do this" on questions like "Describe the role of antibiotics in the treatment of disease". These eight items have a high Cronbachs' alpha (0.80) for our analysis. None of the items needs to be excluded.

Explanatory variables, teaching method

PISA 2006 describes 4 science teaching methods: interaction, hands-on activities, student investigations and focus on model or application. These variables are based on 17 different questions. Students were asked to what degree four constructivist science teaching methods occurred. "When learning science topics at school, how often do the following activities occur?" was answered on a four range scale: 'in all lessons', 'in most lessons', 'in some lessons' and 'never or hardly ever' (OECD, 2007). The first variable on constructivist teaching methods is the scale 'interaction in science teaching and learning' (INTACT), indicating the frequency with which different elements of interactive teaching occur in their classroom. The INTACT scale has an acceptable reliability (alpha: 0.76). 'Occurrence of opportunities for scientific investigation' (INVEST, alpha: 0.73) and the 'Hands-on learning' (HANDS, alpha: 0.69) are constructivist teaching methods that are measured by the second and third scale respectively. The last scale for a constructivist science teaching method focuses the application of science to real life situations (APPLY) and shows acceptable reliability as well (alpha: 0.71).

Examples of different items according to this scales are:

INTACT: Interactions:

"The lessons involve students' opinions about topics."

HANDS: Hands-on activities:

"Students spend time in the laboratory doing practical experiments."

INVEST: Investigations:

"Students are given the chance to choose their own investigations."

APPLY: Focus on model or applications:

"The teacher uses examples of technological applications to show how science is relevant to society."

We included the constructivist teaching methods at the aggregated school level. Snijders and Bosker (1999) state that a scale's measurement error is reduced when a large number of respondents are included at the lowest level (as in our study n=3843). The large number increases the reliability of the estimate at school level. Moreover, as underlined by Griffith (2002), the appropriate level of analysis (at student or school level) is variable-dependent. When concepts measured at the individual level are nearly identical to concepts aggregated at the higher level (like 'Quality of instruction') the reliability can be assumed greater (Griffith, 2002). Therefore, the mean on each of the four scales for constructivist teaching was calculated for each school. Afterwards, each of the four scales were standardised at the Flemish school level. A positive score on for example 'hands-on learning' indicates that, in that school and according to the students, science is taught in a more hands-on way than in the average school.

Control variables

ESCS. The economic, social and cultural status (ESCS) of the student is composed by PISA from aspects like home possessions, highest occupational status and educational level of the parents. This variable is included both at the student level and aggregated at the school level.

Curriculum. An indicator for how much time students actually spent on regular science lessons. We reduced the original 5-point Likert scale from PISA to a 3-point Likert scale. The categories "none" and "less than 2h" are rescaled in a new category "few". The category "2 to 4h" now is called "average" and the categories "4 to 6h" and "more than 6h" are composed to "much". When these categories were reduced, we controlled that each category had an approximate equal number of respondents.

Origin. Native students are born in Belgium, and so is at least one of their parents, Second generation immigrants are born in Belgium but their parents aren't, and first generation immigrants are born in another country and so are their parents.

Gender. A dummy variable singling out the girls in the data

Language. Language is an individual variable because the terms used at school are often different than the register used at home. This is particularly difficult for students with an different language spoken at home (Trudgill, 2003). We only use the categories 'same language spoken at home as during class' and 'different language spoken at home'. We don't make a separation based on languages.

Analysis

This study can be divided in three main steps. First we will examine the intra class correlation between the two levels, school and student. This will be done with a basic model. Secondly we add the explanatory variables in a gross model to see what their impact is on the three outcome variables. And finally we add all control variables with school and student characteristics to control which factors at the school and student level explain variation in the outcome variables.

RESULTS

As explained in the methodology we will describe the results step by step starting with the basic model.

From the first model of the multivariate multilevel analysis (table 1) we can conclude that the variance for self-efficacy can be found for 7% at school level. Nearly the same can be said about interest in science where we find 8% at school level. This may not come as a surprise because both these variables are attitudes which are often personal aspects. So it is clear that variance can be found mostly at student level. For scientific knowledge the analysis shows that school level describes 38% of the variance. All variances are significant. We can conclude that schools matter a great deal when it comes to knowledge about science and scientific knowledge.

Self-efficacy and knowledge correlate rather strong and significant on school level (.760). So apparently schools where pupils score high on average for self-efficacy are more likely to score well on average for scientific knowledge too. Self-efficacy and interest in science have a medium correlation of .55 also at school level. All other correlations at school or student level are smaller.

On student level, high scores on one of the three variables don't necessarily go along with high scores on the other variables. A student can be highly interested in science, without having a large scientific knowledge. This means that the three dependent variables all measure some independent characteristic from scientific literacy.

Table 1: Estimated parameters first model (correlations are italic; variances bold)

School level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.068

0.553

0.760

Interest

0.077

0.378

Knowledge

0.392

Student level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.935

0.354

0.394

Interest

0.925

0.156

Knowledge

0.644

In our second model we add the four different science teaching methods (table 2). The results show 'investigations' has a significant negative relationship with self-efficacy and knowledge. This means that students from a school where there are more than average possibilities to do their own investigations in science, design their own experiments and test out their own ideas they score less than average on the topic knowledge. The same effect can be found when students have the opportunity to interact during science lessons. Discussions, class debates, and explaining their ideas too are significant negative for their science knowledge.

Hands-on activities like doing experiments or drawing conclusions from an experiment on the other hand are significantly positive for both self-efficacy and knowledge and can be seen a good style of teaching for achieving more scientific literacy. The same conclusion can be drawn for schools in which teachers who explain phenomena, use science to help students to understand the world and explain the relevance for science to our lives. Schools in which teachers that focus on models or applications in science not only have students with more self-efficacy and knowledge but also interest in science. This teaching method seems to be the only one who has a significant positive effect on interest in science.

Table 2: Estimated parameters and standard errors (S.E.) from the fixed part in the gross model

Self Efficacy

Interest

Knowledge

 

Estimated

S.E.

Estimated

S.E.

Estimated

S.E.

CONS

-0,012

0,023

-0,007

0,027

-0,042

0,037

INVEST_mean_school

-0,100

0,031

*

-0,028

0,035

-0,271

0,047

*

INTACT_mean_school

-0,048

0,028

-0,014

0,032

-0,210

0,043

*

APPLY_mean_school

0,076

0,027

*

0,101

0,031

*

0,175

0,042

*

HANDS_mean_school

0,085

0,030

*

-0,013

0,034

0,161

0,047

** = significant at the p<0.05 level

When we look at the random part (table 3) we see that the correlations between the dependent variables are smaller in this model compared with the first one. There is still a strong positive correlation between self-efficacy and interest and between self-efficacy and knowledge. At student level these correlations are medium positive.

Table 3: Estimated parameters gross model (correlations are italic; variances bold)

School level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.039

0.511

0.573

Interest

0.066

0.298

Knowledge

0.168

Student level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.935

0.354

0.395

Interest

0.925

0.156

Knowledge

0.644

In the last model we add all student and school characteristics. The curriculum matters for all three dependent variables. Overall we can say that the number of hours of science has a positive impact on self-efficacy, interest and knowledge. This seems to be a logical conclusion because students who like doing science more often choose scientific studies than others and they feel more confident in studying science which will lead to a bigger self-efficacy.

The economic and social status of the student also has a positive influence on self-efficacy and knowledge but is not a significant predictor for interest in science. Schools with higher than average ESCS-score also have a positive effect on self-efficacy and knowledge, but not on interest. Interest is a very personal aspect which has no relationship with social background or school.

Second generation immigrants have a lower score in knowledge than first generation and native students. First generation students seem also to have more self-efficacy than both other groups. For interest in science there are no significant differences.

Students that speak another language at home than in the classroom, show a large and significantly lower score on knowledge. For interest and self-efficacy, the language spoken at home has no significant effect.

Girls score significantly worse than boys on all three dependent variables but the effect is twice as big on self-efficacy and knowledge as on interest. Probably self-efficacy is more related to gender than the style of teaching as we can see that girls have a significant lower score (-0.272 SD) than boys. Girls also seem to be less interested in science than boys (-0.125 SD) and they score 0.263 standard deviations lower on knowledge.

Table 4: Estimated parameters and standard errors (S.E.) from the fixed part in the net model

Self Efficacy

Interest

Knowledge

 

Estimated

S.E.

Estimated

S.E.

Estimated

S.E.

CONS

0,189

0,034

*

0,055

0,039

0,243

0,037

*

INVEST_mean_school

-0,055

0,029

-0,014

0,036

-0,189

0,039

*

INTACT_mean_school

0,000

0,026

0,015

0,033

-0,094

0,036

*

APPLY_mean_school

0,044

0,024

0,083

0,030

*

0,117

0,033

*

HANDS_mean_school

0,045

0,027

-0,033

0,034

0,094

0,037

*

CuSmall

-0,263

0,039

*

-0,178

0,040

*

-0,351

0,032

*

CuLarge

0,175

0,040

*

0,201

0,042

*

0,213

0,033

*

ESCS_school

0,168

0,060

*

0,052

0,076

0,420

0,082

*

ESCS

0,071

0,017

*

0,032

0,018

0,084

0,014

*

Or2nd

0,191

0,119

0,220

0,123

-0,391

0,098

*

Or1nd

0,291

0,135

*

0,123

0,139

-0,209

0,110

Language

-0,085

0,088

0,020

0,091

-0,383

0,073

*

Girl

-0,272

0,033

*

-0,125

0,035

*

-0,263

0,028

*

* = significant at the p<0.05 level

After having controlled for student and school characteristics the correlation between self-efficacy and knowledge is only half of what is was before at school level. The curriculum has a strong effect on the self-efficacy, this explains why the correlation between self-efficacy and knowledge becomes smaller in the more advanced model. The correlation at school level between self-efficacy and interest remains more or less unchanged. At student level all correlations are the same as before.

Table 5: Estimated parameters net model (correlations are italic; variances bold)

School level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.021

0.515

0.267

Interest

0.057

0.307

Knowledge

0.092

Student level

Self-efficacy

Interest

Knowledge

Self-Efficacy

0.858

0.335

0.365

Interest

0.893

0.137

Knowledge

0.554

Conclusion and discussion

Although this research revealed some interesting topics, some points of discussion can be made. Investigations need to be done to explore if this results are equal for all secondary students. It is likely to think that students who have more scientific background will be able to discuss some topics and design their own experiments. Different topics within science education can also be investigated on their own. Geography and physics might need an other approach than biology and chemistry.

The results show none of the science teaching method to have a significant effect on self-efficacy after control for student and school characteristics. Student interest in science, is after control for school and student characteristics, positively influenced by the teaching method 'focus on models or applications'. Schools in which teachers explain how a scientific idea or subject can be applied to a number of different phenomena or they explain the relevance of scientific concepts for our daily lives, have students that show a greater interest in science. Students are more interested in a subject when they know where it is useful for.

All four science teaching methods have a significant effect on the students' knowledge about science and scientific knowledge, which is in line with previous research about teacher effect on student outcome (e.g. Creemers & Kyriakides, 2006; Barber & Mourshed, 2007).

When students can explain their ideas, discuss their opinions or have class debates they will score 0.1 standard deviations less than students who have an average amount of interaction during science lessons. The largest effect is the negative impact of student investigations. When students are allowed to design their own experiments, they can chose their own investigations or they can test out their own ideas valuable study time is wasted when we want to add knowledge to our children. The common factor in both these teaching methods is the big input students can have on the topic of the lessons.

When students can do more than average hands-on activities like practical experiments either with the help of the teacher or not, but always by his instruction, they have more chances to score high on knowledge. We can say that they learn more about science when something is actually done by the students. The largest positive effect is a teaching method that helps students to understand the connection between science in class and the world outside the school. When teachers explain the relevance of the different scientific topics this not only has a positive impact on their interest, as explained above, but also on their knowledge.

These findings support Van de Werf (2005) who concluded that children benefit from a structured education and a hierarchical structure of the syllabus. When students themselves have to set goals and find what they are supposed to learn this may not be as efficient as when the teacher does.

Our results suggest that 15 year old students can benefit from somebody explaining different scientific phenomena and their relation to daily practices, the results also suggest that they might gain more science knowledge by following an instruction and doing an experiment which is set up by the teacher. Probably the scope of possibilities in science of 15 year olds is not broad enough to explore the scientific jungle by themselves. They need a guide to lead them to the different interesting spots and keep them on the right path. When they start exploring alone they will do it superficially and not as accurate as with their teacher. Teaching methods do make a difference when science has to be taught.

Writing Services

Essay Writing
Service

Find out how the very best essay writing service can help you accomplish more and achieve higher marks today.

Assignment Writing Service

From complicated assignments to tricky tasks, our experts can tackle virtually any question thrown at them.

Dissertation Writing Service

A dissertation (also known as a thesis or research project) is probably the most important piece of work for any student! From full dissertations to individual chapters, we’re on hand to support you.

Coursework Writing Service

Our expert qualified writers can help you get your coursework right first time, every time.

Dissertation Proposal Service

The first step to completing a dissertation is to create a proposal that talks about what you wish to do. Our experts can design suitable methodologies - perfect to help you get started with a dissertation.

Report Writing
Service

Reports for any audience. Perfectly structured, professionally written, and tailored to suit your exact requirements.

Essay Skeleton Answer Service

If you’re just looking for some help to get started on an essay, our outline service provides you with a perfect essay plan.

Marking & Proofreading Service

Not sure if your work is hitting the mark? Struggling to get feedback from your lecturer? Our premium marking service was created just for you - get the feedback you deserve now.

Exam Revision
Service

Exams can be one of the most stressful experiences you’ll ever have! Revision is key, and we’re here to help. With custom created revision notes and exam answers, you’ll never feel underprepared again.