Glenda D. Peralta
Julian J. Meimban III
Jachelle Anne D. Terrago
New Era University
How to Cite:
Peralta, G. D., Meimban, J. J., III, & Terrago, J. A. D. (2022). Teachers’ competence and students’ achievement goals as predictors of students’ motivation towards science. NEU Likha Journal: A Refereed Journal of the New Era University School of Graduate Studies, 1(1), 37–52.
Abstract
This study aimed to determine if teachers’ competency level and students’ achievement goal were predictors of Grade 10 students’ motivation (SM) in studying science in public and private schools in the School District of Quezon City, Philippines using multiple linear regression. The study also aimed to determine if there is significant difference in science motivation of students in public school and in private school using Mann-Whitney U test. Teachers’ competency level was a latent variable manifested by three factors or sub-scales, namely: classroom management (CM), instructional materials and facilities (IMF), and instructional method (IM). Students’ achievement goal was also a latent variable manifested by four indicator variables: mastery-approach (MAP), mastery-avoidance (MAV), performance-approach (PAP), and performance-avoidance (PAV). For multiple linear regression, the sample size used was 108, computed using Cohen’s (1997) procedure with the following specifications: α = .05, effect size (ES) (R2) = 0.15 (medium), and power = .80. For Mann-Whitney U test, the sample size used was 128, computed with α =.05, ES (d) = .50 (medium), and power = .80. Respondents were randomly selected from target schools. The model derived by forward regression was cross-validated using new set of randomly selected sample. The linear composite of MAP, MAV, and IM best predicted SM (Adjusted R2 = .495). IM (β = .38) contributed most in the prediction of SM, followed by MAV (β = .32) and MAP (β = .23). On the average, the SM in the Private School was significantly higher than that in the Public School, p = .024, ES (r) = .20 (small to medium).
Keywords: science motivation, teachers’ competence, students’ achievement, multiple linear regression
The need to develop a country’s science and technology has generally been known as one of the imperatives of socio-economic advancement in the modern world hence this becomes a prevalent concern of the government. The vision for science and technology by 2020 (National Science and Technology Plan 2002-2020) is to have developed a wide range of globally competitive products and services which have a high technology content. It means that by this year, the Philippines should have world-class capabilities in information and communication technologies (ICT), technological leadership in Association of Southeast Asian Nation (ASEAN) in the fields of biotechnology, material science and microelectronics, highly-developed culture of innovation and science and technology consciousness. Science, technology, and research are among the aspects now being focused on by the administration to contribute to the achievement of ASEAN’s vision 2025 of sustainable and dynamic community.
Science plays an important role to ensure Filipinos are ready to face challenges for innovation. Through science education, students have the capability to understand scientific knowledge, identify important scientific questions, draw evidence-based conclusions and make decisions about how human activity affects the natural world [Organization for Economic Cooperation and Development (OECD), 2007]. In addition, students who are scientifically literate can easily grasp essential science concept, understand the nature of science, realize the relevance of science and technology in their lives and willing to continue their science study in school, or beyond school (National Research Council, 2000). It is important for all students to become scientifically literate (Feinstein, 2011; Roberts, 2007) to produce a society that is scientifically oriented, progressive, knowledgeable, having a high capacity for change, forward-looking, innovative and a contributor to scientific and technological developments in the future. Due to greater importance of science and technology, schools have been encouraging students to learn science and other science related subjects. It is thought that one of the greatest challenges of this century is to motivate students for maintaining their learning and success in science. However, students’ motivation and performance in science classes in secondary education was not found adequate and did not improve in the last decade (Tella, 2007; Program for International Student Assessment, 2019). Hence, knowing the predictors of students’ motivation towards learning science is an evitable need in today’s world.
The low interest of students in science has been a big problem not only in the Philippines but also in other countries. The mounting evidence of declining rank of the Philippines in terms of science and technology and innovation (STI) index (Global Innovation Index, 2018; PISA, 2019), and the interest of young people in pursuing scientific careers has been alarming issue for the past years that is why science educators need to do something to address the problems and increase the motivation of students towards learning science.
Motivation is considered an important factor in science learning (Koballa & Glynn, 2007 as cited by Fernandez & Garcia, 2019). The promotion of favorable motivation towards science and science learning which has always been a component of science education is increasingly a matter of concern. It argues that continuing decline in numbers choosing to study science at the point of choice requires a research focus on students’ motivation towards science if nature of problem is to be understood and remediated. The concept of teachers’ competence and students’ achievement goals as predictors of science motivation in the Philippines is somewhat nebulous, often poorly articulated and not well understood hence, this study.
This study aimed to determine if teachers’ competency level and students’ achievement goals were predictors of Grade 10 students’ motivation (SM) to study science in public school and in private school. Teachers’ competency level was a latent variable manifested by three factors or sub-scales, namely: classroom management (CM), instructional materials and facilities (IMF), and instructional method (IM). Students’ achievement goals was also a latent variable manifested by four indicator variables: mastery-approach (MAP), mastery-avoidance (MAV), performance-approach (PAP), and performance-avoidance (PAV).
Specifically, the study addressed the following research questions:
- How well does the linear combination of CM, IMF, IM, MAP, MAV, PAP, PAV, and ST (School Type: Public or Private) predict SM?
Ho: CM, IMF, IM, MAP, MAV, PAP, PAV, and ST are not predictors of SM.
Ha: at least one of them is a predictor of SM. - Is there a significant difference in students’ motivation in public school and in private school?
Ho: μ1 = μ2
Ha: μ1 ≠ μ2
Where:
μ1 = population mean of SM of Grade 10 students in public schools
μ2 = population mean of SM of Grade 10 students in private schools
Method
Research Design
This study employed a correlational design and used survey questionnaire to collect data.
Population and Sample
The target population was Grade 10 students from three (3) public schools and three (3) private schools in the District of Quezon City for the academic year 2019-2020. The six schools had three sections each, with an aggregate total of about 800 students. For Research Question 1, the appropriate sample size was computed using Cohen’s (1997) procedure with the following specifications: alpha = .05, effect size = .15 (medium), and power = .80. These specifications require a total sample size of 107, which was rounded up to 108 in this study. For Research Question 2, the sample size was 128, computed with α = .05, effect size (d) = .50 (medium), and power = .80.
The plan for random selection of respondents from the Public School and Private School followed a three-stage selection scheme. For Research Question 1, the 108 required sample was collected ensuring that there were equal number of respondents from the Public School and Private School, or 54 respondents per School type. In the first stage, in the Public School, the 54 respondents were equally divided randomly among the three randomly selected Schools, or 18 respondents per School. In the second stage, the 18 respondents were equally divided randomly among the three sections in each School, or six (6) respondents per section. Finally, in the third stage, the six respondents were randomly selected from each section by lottery method. The same scheme of apportioning and randomly selecting respondents was followed in the Private School.
Similarly, for Research Question 2, the 128 required sample was collected ensuring that there were equal number of respondents from the Public School and Private School, or 64 respondents per School type. In the first stage, in the Public School, the 64 respondents were divided among the three randomly selected Public Schools, 21 respondents were to be drawn from each of the two Schools and 22 respondents from the third school. In the second stage, the respondents were apportioned among the three sections in each School. In the two Schools with 21 respondents each, the respondents were equally divided among three sections, or seven (7) respondents per section. In the other School with 22 respondents, the respondents were apportioned among three sections where seven (7) respondents were to be drawn from each of the two sections and eight (8) respondents from the third section. Finally, in the third stage, the respondents were randomly selected from each section by lottery. The same scheme of apportioning and randomly selecting respondents was followed in the Private School.
Instrument
Three survey instruments were used, namely: students’ science motivation (Science Motivation II by Glynn, 2011), teachers’ science competency level (Students Evaluation for Teachers by Family School of Quezon City Inc., 2018), and students’ achievement goals (Achievement Goal Questionnaire by Elliot and Murayama, 2008). Each of these instruments underwent face and content validity assessment by three professional survey instrument developers who were asked to critique the accuracy of the items in relation to the objective of the study and to suggest necessary improvements.
Furthermore, the instruments were tested for internal consistency reliability (Cronbach’s Alpha) using sample of 40 students (20 per ST) from the target population. For the Teachers’ Competency scale, the Cronbach’s Alpha of its three sub-scales were as follows: classroom management, α = .83; instructional tools and materials, α = .87; and instructional method, α = .87. For the Achievement Goal Questionnaire, the Alpha of its four sub-scales were as follows: mastery-approach, α = .84; mastery-avoidance, α = .88; performance-approach, α = .92; and performance-avoidance, α = .94. The Alpha of the Science Motivation Questionnaire was .84. All Alphas met the generally acceptable minimum standard value of .70 for a reliable survey instrument.
Administration and Retrieval
The researcher tapped Science teachers of the different schools in Quezon City to administer the questionnaire to the Grade 10 students. A schedule was obtained for administering the questionnaire to Grade 10 respondents. The researcher personally met the science teachers who administered the survey. The survey was preceded by an instruction in which the science teacher explained to the students that collection of information on their achievement goals and their assessment on the performance of their science teacher would be able to help them discover what motivates them. They were assured that their answers would in no way affect their grades but rather would be useful in helping them understand themselves better. After which, the questionnaire was collected and handed to the researcher.
Statistical Analysis
Multiple linear regression was used to predict students’ motivation (SM) using a set of independent variables (CM, IMF, IM, MAP, MAV, PAP, PAV, and ST). Mann-Whitney U test was also used to compare SM of students from private and public schools. The variables, except ST, were calculated as average of their corresponding indicators from questionnaires with 5-point Likert scale, with 5 being the highest score. The descriptive statistics and correlations for the different variables are shown in Table 1.
Before performing multiple linear regression, linearity between the dependent variable (DV) and each independent variable (IV) was checked. Linearity assumption was met as indicated by significant correlation between DV and each IV (Table 1). As a result, all IV were considered in building the regression model.
Forward multiple regression method was used to select variables. The final model (Model 3) is the linear combination of IM, MAP, and MAV which produced the highest Adjusted R2.
Before Model 3 was used to predict SM, other assumptions were checked first if they were met. The tolerance values of MAP (.610), IM (.805), and MAV (.711) are well above the tolerance threshold value of .50 which indicates that there is no multicollinearity problem among the three remaining variables. The histogram of the standardized residuals is a bell-shape curve (Figure 1). Also, the dots in normal p-p plot approximately follow a straight line. Thus, the assumption of normality of residuals is met. The scatter plot of the standardized residuals against the standardized predicted value of SM showing that most of the points are concentrated at point 0 and form a rectangular shape (Figure 2) indicates that the assumption of homoscedasticity is met. The model satisfied the assumption of no autocorrelation among the residuals as shown by the value of Durbin-Watson (2.042) which lies within the range 1.5 to 2.5.
Since all assumptions of multiple linear regression were met, Model 3 is considered final or “best” regression prediction model for SM and is subjected to model validation. The cross validation using new set of 60 randomly selected students (30 students per ST), resulted in a significant Pearson r of .786 (p = .01, 2-sided). This value indicates “high relationship” and implies that the derived model can be generalized to the target population.
Results and Discussion
The best fitting model is given by (SM) = 0.658 + 0.273(IM) + 0.196(MAV) + 0.273(MAP). This model is highly significant, F(3,104) = 35.95, p < .001. Of the eight original potential predictors, only MAP (t = 2.060, p = .011), MAV (t = 3.889, p < .001), and IM (t = 4.929, p < .001) were found to be the significant predictors of SM. MAP (Mastery-Approach) and MAV (Mastery-Avoidance) were students’ achievement goals manifest variables while IM (Instructional Method) was a teachers’ competency level manifest variable. The model’s Adjusted R2 of .495 indicates that 49.5% of the variance in SM can be predicted by the linear combination of the three predictors.
Of the three best predictors, IM (β = .38) contributed most in the prediction of SM, followed by MAV (β = .32) and MAP (β = .23). IM explained 18.9% of the variance of SM beyond the variance explained by MAV and MAP. Likewise, MAV and MAP explained 12.7% and 6.2%, respectively, of the variance of SM that was unexplained by other variables. Removing IM in the best fitting model lowered the R2 by 11.5%; that of MAV and MAP by 7.1% and 3.2%, respectively (Table 2).
Result of Mann-Whitney U test shows that the SM in the Private School (mean rank = 71.91) was statistically significantly higher than that in the Public School (mean rank = 57.09), U = 1574.00, p = .024 (2-tailed). The ES (r) was computed to be .20, which according to Cohen (1988) is small to medium.
Seemingly, the science motivation of students (SM) in the Private School was significantly greater than that in the Public School, despite of the fact that ST did not contribute significantly in predicting SM as shown in the discussion of Research Question 1. This apparent inconsistency can be explained by the fact that, among the eight candidate predictors, ST had the weakest correlation with SM (Table 1). Although ST was not highly correlated with the rest of the variables and so collinearity problem was not an issue, it did not significantly contribute in explaining the variance of SM. However, if ST were considered alone, i.e., disregarding the potential correlational effects of the other predictors, and used as a grouping variable to compare SM in Private School and in Public School, significant difference between schools would have been observed. In short, multiple regression modeling considers the impact of each predictor in linear combination with other predictors in explaining the variance in the outcome variable. On the other hand, difference research questions, such as Research Question 2, are addressed by specifying a particular categorical predictor as grouping variable in determining difference in the outcome variable, ignoring the effects of other possible predictors which were not selected by the researcher.
References
Cornell University, INSEAD, WIPO. (2018). Global Innovation Index 2018: Energizing the World with Innovation. WIPO
Council, N. R., Center for Science, M. E. E., & Inquiry, C. D. A. N. S. E. S. S. (2000). Inquiry and the National Science Education Standards: A Guide for Teaching and Learning. National Academies Press.
Elliot, A. J., & Murayama, K. (2008). On the measurement of achievement goals: Critique, illustration, and application. Journal of Educational Psychology, 100(3), 613–628. https://doi.org/10.1037/0022-0663.100.3.613
Family School of Quezon City Inc. (2018). Student Evaluation of Teacher’s Competence.
