The Impact of Academic Motivation on the Self-Directed Learning Skills and Problem-Solving and Innovation Skills of Malaysian Secondary School Students

LEE SHI EN

2023.1.EDU02.0008

CHAPTER 4 & CHAPTER 5

CONTENT

DISCUSSION

DATA ANALYSIS

CHAPTER 4: INTRODUCTION

In this chapter, will present the results of the data analysis in a complete format which gathered by using questionnaire method. All data will be analyze by applying SPSS software. There will discuss the result of demographic profile of respondents, descriptive analysis, Pearson Correlation, multiple linear regression, ANOVA, Coefficients Table and other related analysis results.

Demographic Profile

Gender

The above results of the data showing that female respondents more than male respondents among 170 respondents in high schools which are 90 (52.9 percent) of female and 80 (47.1 percent) of male. Thus, the collected data represent that the percentage of female respondents in high schools more occupied the majority compare than male respondents.

Form Level

The data information of form level were showing that there are 19 (11.2%) of the respondents who are Form 1 students, 50 (29.4%) of the respondents are Form 2 students, 37 (21.8%) of the respondents who are Form 3 students, follow on 31 (48.2%) of the respondents who are Form 4 students and lastly is 33 (19.4%) of the respondents Form 5 students. From here the finding had been shown that most majority repondents who are Form 2 students are participating in this research.

Name of School

The research had been collected from different high schools such as SMK Tinggi Melaka, SMK Yok Bin, SMK ST Francis and others. From the above research, the majority of respondents studying in SMK Yok Bin which occupied 74 respondents (43.5%) and second top is SMK ST Francis which have 22 respondents (12.9%) out of 170 respondents.

Descriptive Analysis

independent variables

independent variables

The second part of the independent variable is problem solving and innovation skill. The highest mean from these three questions is 3.18 which bring out from question 2 and the lowest mean is 3.14 which come out from question 3. In addition, the standard deviation of this part which bring out from question 1 & 3 are 0.859 which the highest and the lowest standard deviation is 0.802 which come out from question 2 out of 3 questions.

dependent variables

The dependent variable in this research is academic motivation. The highest mean from these three questions is 3.90 which bring out from question 1 and the lowest mean is 3.11 which come out from question 2. In addition, the standard deviation of this part which bring out from question 3 is 1.008 which the highest and the lowest standard deviation is 0.939 which come out from question 2 out of 3 questions.

Normality Test

In the histogram of the data, the presence of a pronounced bell-shaped curve suggests that the dataset may approximate a normal distribution. The normal distribution is a highly significant distribution in statistics, characterized by symmetry, unimodality (a single peak), and a distinct bell-shaped appearance. The performance of the curve at the center of this chart serves as evidence that the collected data is perfect, free of issues, and exhibits high accuracy.

Histogram

Normal P-P Plot

By observing the P-P plot above, we can notice that the data points closely align along the diagonal line, indicating that the residuals of the regression model largely exhibit characteristics of a normal distribution. This approximate match along the diagonal is a positive sign, supporting the assumption of normality and the reliability of data analysis.

Scatterplot

Through the observation of the scatterplot, it is evident that the distribution of data points is relatively dispersed, suggesting a high degree of uniqueness and diversity among the respondents' provided responses. The scatterplot illustrates the distribution of answers from each respondent, showing no apparent trend or clustering, indicating that each respondent expresses individual and independent viewpoints in their responses. This uniqueness not only reflects the distinctive perspectives of the respondents but also underscores the diversity of responses. Respondents exhibit a variety of opinions and answers to the given questions, contributing to a comprehensive representation of information. Given that the majority of respondents offer distinct responses, it is reasonable to consider the data reliable. This is because the data captures a range of viewpoints, thereby enhancing the accuracy of the dataset. However, when drawing conclusions, caution is still required to ensure that the scatterplot's dispersion is not due to data quality issues or other potential factors. In summary, based on these observations from the scatterplot, we can infer that this dataset possesses a high level of uniqueness and diversity, crucial for ensuring the comprehensiveness of the analysis and the credibility of the data.

Reliability testing

According to the required Cronbach's alpha value, a score above 0.70 is considered acceptable. The results indicate that the dependent variable, Academic Motivation, has a Cronbach's alpha of 0.769, demonstrating that the data is acceptable. However, the scores for the two independent variables, Self-Directed Learning Skills and Problem Solving and Innovation Skills, fall slightly below 0.70 at 0.632 and 0.679, respectively.

In terms of Cronbach's alpha, scores within the range of 0.6 to 0.7 suggest a moderate level of internal consistency reliability for the measurement tools. This implies that there is some correlation among the items, although it may not be as strong as desired for research or assessment purposes.

Pearson Correlation

Through Pearson correlation analysis, the research obtained the correlation coefficients and p-values for two independent variables (Self Directed Learning and Problem-Solving and Innovation Skills) with a dependent variable (Academic Motivation). The correlation coefficient falls between 0.4 and 0.6, indicating a moderate level of correlation, suggesting a certain linear relationship between these two independent variables and the dependent variable. Furthermore, with a p-value less than 0.05, the observed correlation is statistically significant. It has shown sufficient evidence to reject the null hypothesis, indicating that this correlation is not due to random chance but holds statistical significance.

Multiple Linear Regression Analysis

R, is the multiple correlation coefficients among all of the estimate variables and dependent variable and the value of R that shown in the diagram below is 0.522 which presents that it is a best output of variance result by the independent variables and dependent variable. Through the analysis of the regression model's performance, the following results were obtained: R-squared (R²) is 0.272, indicating that the model can explain 27.2% of the variance in the dependent variable. The Adjusted R-squared is 0.264, slightly lower than R-squared, suggesting that the added predictor variables may have limited improvement on the model's explanatory power or indicate potential overfitting. The standard error is 0.61054, signifying a relatively low average prediction error, implying reasonably accurate predictions from the model.

In overall assessment, although R-square is not high, it still indicates the model's effectiveness in explaining a portion of the dependent variable's variability. The lower Adjusted R-squared suggests a need to consider variable selection and potential overfitting. The relatively low standard error indicates accurate predictions.

ANOVA Table

For the Anova table which result show that there are display model fit between independent variables which are self-directed learning and problem solving and innovation skills, with dependent variable which is academic motivation because it p-value is less than 0.05 which represent that all independent variables has a great relationship with dependent variables during this research.

In addition, F-value of Multiple Linear Regression (MLR) analysis, it is 31.263 and its significant value of 0.000 which lower than 0.05 confidence level. Hence, the linear relationship which be explore for this research will not be forsaken.

Coefficients Table

The finding that shown in below is about the coefficients of two independent variables which are self-directed learning and problem solving and innovation skills to the dependent variables which is academic motivation. For the significant value, if the value more than 0.05, which represent that significant will effect to dependent variable. Therefore, all two variable has significant relationship to dependent variables which is academic motivation.

Hypothesis

02.

H2 is recorded as supported and accepted

It represent that there is a relationship between problem solving and innovation skills and academic motivation as p-value less than 0.05 in its confidence level.

DISCUSSION &

CONCLUSION

CHAPTER 5: INTRODUCTION

In this chapter will be conclude all results and provide some relevant information based on the results of data analysis such as discuss about the research objectives, implication, limitation, future study and conclusion of the research.

Research Findings

1. The Academic Motivation variable has a reliable Cronbach's alpha of 0.769. However, the two independent variables, Self-Directed Learning Skills (0.632) and Problem Solving and Innovation Skills (0.679), are slightly below the 0.70 threshold. These scores, within the 0.6 to 0.7 range, suggest a moderate level of internal consistency reliability, indicating some correlation among items but not ideal for research. Future steps could involve refining the scale through item modification or additional data collection to enhance reliability, and conducting analyses with larger sample sizes for more robust results.

​

• The Pearson correlation analysis revealed a moderate correlation (0.4-0.6) between the independent variables (Self Directed Learning and Problem-Solving and Innovation Skills) and the dependent variable (Academic Motivation). The statistically significant p-value (< 0.05) indicates a reliable relationship. To enhance understanding, future studies could explore specific mechanisms influencing this correlation. Longitudinal studies would provide insights into changes over time. Introducing additional variables, considering contextual factors, and conducting validation studies would contribute to a more nuanced interpretation. Practical applications may involve designing interventions to boost Self Directed Learning and Problem-Solving and Innovation Skills, positively impacting Academic Motivation.

Research Findings

3. The multiple correlation coefficient (R) is strong at 0.522, indicating significant variance explained by the variables. However, the model's R-squared is 0.272, suggesting 27.2% variance explained, with an Adjusted R-squared of 0.264, indicating potential overfitting. The standard error of 0.61054 implies reasonably accurate predictions. Consider refining the model for improved explanatory power and avoiding overfitting.

​

• The ANOVA table indicates a significant model fit, as the p-value is less than 0.05, suggesting a strong relationship between independent variables (self-directed learning and problem-solving and innovation skills) and the dependent variable (academic motivation). Additionally, the F-value in the Multiple Linear Regression analysis is 31.263, with a significant value of 0.000 below the 0.05 confidence level, reinforcing the robustness of the linear relationship explored in the research. No specific suggestions are provided; consider emphasizing the significance of the identified relationships for future research or practical applications.

Discussion of the Research

For this research, initially prepared structured survey forms for the targeted respondents, requiring them to provide personal opinions. The questionnaire format comprised closed-ended questions to ensure more consistent and useful data for establishing significant relationships between independent and dependent variables. In conclusion, the research indicates that self-directed learning and problem-solving and innovation skills have the most significant impact on the academic motivation of secondary school students in Malaysia. Therefore, stakeholders in Malaysian secondary schools should prioritize academic motivation, as it directly influences students' self-directed learning and problem-solving and innovation skills.

This study set three main objectives:

• firstly, to investigate the impact of self-directed learning on academic motivation;
• secondly, to explore the role of problem-solving and innovation skills in academic motivation;
• thirdly, to examine how these factors manifest among secondary school students in Malaysia.

The results indicate that both self-directed learning and problem-solving and innovation skills significantly influence academic motivation. This provides substantive insights for educational practices and highlights directions for further research. While the study successfully addressed its objectives, we acknowledge limitations in sample selection and data collection, presenting opportunities for improvement in future research. Overall, this study has made positive strides in exploring factors influencing academic motivation.

Limitation of the Research

Firstly, the sample size, standing at 170, although statistically acceptable, could benefit from an expansion to 200 or more for enhanced accuracy and credibility.

Secondly, the focus on two primary independent variables, namely self-directed learning and problem-solving and innovation skills, might limit the comprehensiveness of the results, suggesting the potential value of introducing additional relevant independent variables.

Additionally, time constraints have impacted the study's ability to thoroughly explore all dimensions and related factors.

Lastly, despite diligent data collection efforts, the study received only 170 responses, potentially affecting the overall comprehensiveness and representativeness of the results.

These limitations serve as avenues for future improvements. To deepen the understanding of academic motivation among secondary school students, it is recommended that future research considers enlarging the sample size, introducing more independent variables, dedicating ample time for in-depth exploration, and employing targeted data collection methods to ensure a more robust and representative dataset. These enhancements are poised to elevate the scientific and practical value of future research endeavors.

Recommendation

Firstly, concerning the sample size, it is suggested that future studies make efforts to expand the sample size to enhance statistical power. Setting a more challenging target, such as reaching or surpassing 200 valid responses, will render the study results more representative and credible.

​

Secondly, to obtain more comprehensive data outcomes, future research is advised to consider introducing additional relevant independent variables. Careful selection of these variables ensures their substantive contribution to answering the research questions.

​

Considering the constraint of time, it is recommended that future research efficiently utilizes time during the study design and execution processes. A well-planned research schedule ensures ample time for in-depth exploration and data analysis.

​

Lastly, to address the current issue of data collection yielding only 170 survey responses, future research may adopt more proactive data collection strategies. Enhancements such as diversifying survey channels and refining questionnaire design can be implemented to improve response rates and ensure more thorough and representative data collection.

​

These recommended improvements aim to guide future research in overcoming the current limitations, enabling a more comprehensive and reliable exploration of the research questions.

Conclusion

This study explores the connections between self-directed learning, problem-solving, innovation skills, and academic motivation, revealing significant insights. Both self-directed learning and problem-solving skills notably impact academic motivation among Malaysian secondary school students, offering practical implications for administrators and policymakers.

Despite positive outcomes, acknowledging limitations such as sample size and time constraints is crucial. Future research should address these by expanding samples and refining data collection.

In conclusion, this study provides valuable insights into students' academic motivation, laying a foundation for future research and educational practices. It serves as a reference for education professionals and decision-makers to enhance the education system's responsiveness to students' academic needs.

THANK YOU