Did It Go Accordingly? A Case of Community-Based Learning Sulam on Stock Trading in Malaysia

Community-based learning (CBL) has become an integral part of educational methodologies, consistently being incorporated into teaching, and learning pedagogies. This approach has proven effective in broadening the scope of learning, particularly when integrated within the local community. This research examines the feedback received from members of the community engaged in the CBL stock trading project. Underpinned by Kirkpatrick’s Four-Level Model, this paper assessed (1) the community’s reactions and learning, (2) the community’s learning, (3) the community’s behavior, and (4) the results. A total of 94 participants were coached by 179 bachelor’s degree students in applied investment management. The students delivered five modules to the respective participants within three months. The findings show that the participants had a positive outlook on the project, although there was a poor response that potentially posited a loophole in CBL. This research provides more insight into the study of community-based learning at Service-Learning Malaysia-University for Society (SULAM) derived from High-Impact Educational Practices. Furthermore, it is suggested by this research that a foundational system of regulatory guidelines must be implemented across the program. This would serve to establish checks and balances, enhancing both student adherence to the program and community engagement.


Introduction
Within Malaysian Higher Learning Institutions (HLIs), community engagement stands as a pivotal component of both student experience and academic lifelong learning.The present era, characterized by 21st Century Education, introduces an avenue for transformative shifts in educational paradigms.Education now transcends traditional classroom boundaries, with numerous institutions embracing CBL as an educational strategy.The inception of CBL in the United States of America traces back to the National and Community Service Trust Act of 1993.According to Maureen et al. (2022), CBL denotes a collaborative educational approach connecting universities, students, and the community, enabling students to apply their theoretical knowledge to real-world scenarios.Furthermore, Hutter (2008) characterizes CBL as an instructional method integrating students within the community fabric, thus becoming a curriculum-based educational encounter wherein students identify community needs and translate curriculum content into practical situations.
The Malaysia Education Blueprint 2015-2025 (Higher Education) directs its educational efforts toward outcomes benefiting various key stakeholders within the higher education system, encompassing the community (Ministry of Education Malaysia, 2015).Henceforth, the direct engagement with CBL occurs through the Service-Learning Malaysia-University for Society (SULAM) initiative, specifically in the project "Komuniti Bijak Pelaburan" associated with the INV621 -Applied Investment Management course offered by the Bachelor in Investment Management, Universiti Teknologi MARA.SULAM possesses the capacity to amplify the sense of social responsibility, promote racial tolerance and religious understanding, and foster the development of skills such as teamwork, critical thinking, creativity, adaptability, and social as well as communication proficiencies (Ministry of Education Malaysia, 2019).The primary objective of the "Komuniti Bijak Pelaburan" project is to provide an educational encounter wherein students engage in organized service activities that address identified community needs.Subsequently, they reflect upon these service activities and experiences, working towards achieving specific learning outcomes aimed at deepening their comprehension of the stock market.The foundational components of SULAM, depicted in encompassing course content, critical reflection, and substantial service engagement spanning across students, educators, and the community.In this context, students undertake the role of mentors or facilitators within a coaching community, imparting their investment knowledge and skills, with a specific focus on the stock market.This endeavor potentially contributes to empowering the community to generate supplementary income.Hence, the objective of this paper is to discuss the descriptive analysis of the effectiveness of the program from the community's feedback underpinned by Kirkpatrick's Four-Level Model.

Literature Review
The literature review is further discussed in the following subsections: community-based learning in Malaysia, and Kirkpatrick's Model.

Community-based Learning in Malaysia
Community-based learning (CBL) is gaining traction in Malaysia to provide students with practical experience and soft skills.The direction given by the Ministry of Education through the SULAM initiative will significantly amplify the practice of CBL further as a method of learning among students.Several studies have explored CBL programs in Malaysia.Since the CBL can be implemented through multitudes of approaches, the results between the initiatives may vary but carry the same narrative of bringing positive outcomes to the students overall.Hamzah (2023) proposed a framework for effective CBL that incorporates six elements: learning outcomes, activities, assessments, student reflection, community partnerships, and program evaluation.This framework aims to guide Malaysian universities in developing highquality CBL courses.Kamin (2010) evaluated a work-based learning (WBL) program for automotive engineering students at two community colleges.Students, lecturers, and industry partners praised the program for improving students' technical and soft skills.However, issues like student accommodation and program focus needed improvement.Haslip (2013) described a community-based character education program for children that aims to empower students through promoting positive traits and knowledge.The program could be adapted for different cultural contexts.Meanwhile, Siah (2020) studied the challenges faced by refugee teachers in community-based learning centers, which provide education for refugee children in Malaysia.Using a resource and demands framework, the study found that teachers struggled with a lack of resources, poor infrastructure, and heavy workloads.More support from NGOs and communities could help address these issues.Ali (2017) found that WBL improved students' soft skills, including communication, problem-solving, and leadership.Ahmad (2021) studied how "tarbiyah"-intellectual discourses on Islam-provided organizational learning for an Islamic NGO.Tarbiyah boosted members' passion for volunteering and self-esteem, benefitting both the organization and the community.Ramlie (2021) proposed developing an online education community service (OECS) model based on SULAM.The model would allow community service to continue during COVID-19.A design and development research approach with multiple phases and methods was suggested to construct the model.The model could fulfill the Ministry of Higher Education's goal of holistic student development and the national vision of prosperity and sustainable development.
In summary, research shows that CBL is valuable in Malaysian higher education.Programs aim to cultivate students' soft skills, character, and community spirit through practical experience.However, programs face issues like a lack of resources and framework that future research could address.Johan et al (2022) posited that the dynamics of CBL should be further enhanced with a greater self-regulated framework in SULAM, especially on the project execution to negate any potential setbacks.However, with continuous improvement, CBL can achieve its goal of empowering both students and communities in Malaysia.

Kirkpatrick's Four Stages Model
The Kirkpatrick Model is a widely used framework for evaluating the effectiveness of training programs.Created by Donald Kirkpatrick in the 1950s, this model has undergone iterations over time and consists of four levels: reaction, learning, behavior, and results.It serves as a comprehensive guide for assessing the impact and success of training initiatives within organizations.Golfiroozi et. al (2023) highlight the beneficial impact of in-service CPR training across all dimensions delineated within the Kirkpatrick model for nurses.These findings play a pivotal role in enhancing the operational efficiency of healthcare personnel.The critical aspect contributing to this efficiency is the rigorous assessment and evaluation of training course quality, which has demonstrably amplified the effectiveness of these programs.The results from the cited study indicated the favorable effectiveness of incorporating the CPR training program across the entirety of the Kirkpatrick model's three levels.Additionally, Omar et al. (2009) affirmed the value of assessments conducted on healthcare middle managers in various Iranian provinces, aligning with the Kirkpatrick model's framework.Anita Cuteanu et. al (2023) in response to this educational gap, curriculums for microsurgery have emerged, primarily emphasizing technical skills.This study aimed to present an assessment of a longstanding curriculum utilizing the Kirkpatrick model.The thorough evaluation showcased a favorable influence across all four levels outlined within the Kirkpatrick model.In a similar vein, prior studies in the research field Kirkpatrick model by Wenlan Zhao et al. (2023) on the competence and confidence of participants as EMI instructors at levels 1-3 of the Kirkpatrick evaluation model.They found English as a Media of Instruction (EMI) training has an effective impact on the competence and confidence of participants as EMI instructors at levels 1-3 of the Kirkpatrick evaluation model.This training may have a potentially beneficial effect on the teaching quality of Bachelor of Medicine and Bachelor of Surgery (MBBS) education.
Neethu and Abhishek (2023) analyzed the perceived effectiveness of virtual training among outgoing undergraduate and postgraduate students under the Kirkpatrick model during the postpandemic period.The researcher found that their reaction toward virtual training was also positive because they liked online sessions.Regarding the learning aspect of virtual training, they agreed that they have gained the expected knowledge and skills.In addition, they strongly agreed that the trainer was knowledgeable in all the topics covered in virtual training.This outcome supports the arguments made by (Tores et al., 2021), (Gannon and Elsa, 2007), and (Stricker et al. (2011).In consideration of the behavior parameter, trainees reached a consensus on their willingness to engage in training sessions alongside their teammates following the conclusion of virtual training.Additionally, they collectively acknowledged the effective utilization of pertinent skills and knowledge by the trainer throughout these virtual sessions.Addressing the result parameter, the trainees expressed a strong collective agreement regarding their observation of the advantages derived from virtual training.Furthermore, they unanimously shared high expectations and satisfaction with the outcomes generated from this mode of training.

Methodology
This research examines the feedback received from members of the community engaged in the CBL stock trading project.Underpinned by Kirkpatrick's Four-Level Model, this paper assessed (1) the community's reactions level, (2) the community's learning level, (3) the community's behavioral level, and (4) the community's results level.Numerous scholars have customized Kirkpatrick's model for application in academic settings by devising metrics and evaluations tailored to specific learning environments, underscoring its potential utility in this domain (Praslova, 2010;Cahapay, 2021).Ruiz and Snoeck (2018) elaborate on the versatility of Kirkpatrick's model, asserting its relevance to diverse educational programs within both national and multinational contexts.As Bewley and O'Neil (2013) posit, the Kirkpatrick model has demonstrated efficacy in evaluating a wide range of training and educational scenarios.Likewise, Heydari et al. (2019) affirm that, despite inherent limitations in other models, the Kirkpatrick model stands out as suitable and exhibits a commendable track record in assessing educational programs.
A total of 94 participants were purposively selected and were coached by 179 bachelor's degree students in applied investment management.The students delivered five modules of stock trading ranging from fundamental analysis and technical analysis to the respective participants within three months with a minimum of twenty training hours.The instruments were adapted from Alsalamah and Callinan (2021) who have adopted a similar model in an education setting.The questionnaire was immediately given after the completion of training to measure level 1 to level 3: reaction, learning, and behavior using five questions using a fivepoint Likert scale ranging from "strongly disagree" (1) to "strongly agree" (5).As for the results, a questionnaire that included both open-ended and closed-ended questions (similar five-point scale) was given a month after the training completion to gather feedback.The quantitative data from close-ended questions were interpreted through SmartPLS 4 to confirm its reliability, meanwhile, the close-ended questions were interpreted thematically.Based on Table 2, we can see that the majority of the participants in the study are students with a value of 45, followed by private sector employees and government employees.Housewives, unemployed individuals, and pensioners make up a small percentage of the participants.The majority of the respondents are in the 20-29 years old age group with a percentage of 61%, followed by the 30-39 years old age group which contributes to 29%.The 40 years old and above age group has the lowest number of people.4 shows all the Cronbach's Alpha values provided for different constructs (L1 through L4) are high, ranging from 0.92 to 0.965, indicating strong internal consistency among the items within each construct.Composite Reliability (rho_a and rho_c) measures how well a set of indicators (items) reflects a latent construct.Both rho_a and rho_c are similar to Cronbach's Alpha but are estimated differently.Similar to Cronbach's Alpha, values closer to 1 indicate higher reliability.In the provided data in Table 4, the values for rho_a and rho_c are consistently high across different constructs, ranging from 0.924 to 0.971.These high values suggest the strong reliability of the measurement scales.Average Variance Extracted (AVE) assesses the convergent validity of a construct.It represents the average amount of variance captured by the construct's items relative to the amount of measurement error.Table 4 shows that AVE values above 0.5 are generally considered acceptable.In the given data, AVE values range from 0.695 to 0.886, indicating that the variance captured by the items within each construct is relatively high, demonstrating good convergent validity.At this initial level of evaluation, the focus is on participants' immediate reactions and responses to the trainer.Kirkpatrick Model Level 1 seeks to evaluate the participants' satisfaction, engagement, and attitudes toward the trainer.This level measures how the participants felt toward the trainer regarding the trainer's communication skills, the trainer's preparation in term material and training activities, the trainer's method conducted the training, the trainer gives opportunity to participation to discuss and ask questions, and the trainer's achieved the goals of the program.In the context of evaluations or reactions to the trainer within the Kirkpatrick Model, Table 5 presents the results which involved participants responding to five questions using a five-point Likert scale ranging from "strongly disagree" (1) to "strongly agree" (5).The analysis shows that the highest mean score was when the trainer used appropriate training methods that were compatible with the program objectives with a value of 4.723, signifying a high level of reaction to the trainer.

Level 1: Reaction
Conversely, the lowest mean score, indicating slightly lower satisfaction towards the trainer pertained to the trainer achieving the goals of the program with a value of 4.489.Among the evaluated factors, the trainer was able to use available training equipment signifying greater variability in participant responses, while the trainer used appropriate training methods that were compatible with the program objectives and showed the lowest standard deviation, suggesting more consistent feedback.This discrepancy implies that the trainer prepared training activities appropriately and by the objectives of the training program the highest degree of dispersion compared to the other factors.The trainer used appropriate training methods that were compatible with the program objectives has a relatively low standard deviation of 0.515.This suggests that participants generally had a more uniform perception regarding the appropriateness of the training methods employed by the trainer.Conversely, higher standard deviations, such as those observed in the trainer achieved the goals of the program of 0.805, imply more significant variability in participant responses.This indicates that opinions regarding the trainer's success in achieving the program's objectives varied considerably among respondents.Table 6 illustrates various aspects related to reactions to the training delivery, offering insights into participants' perceptions regarding different elements of the program.The mean represents the average score given by participants for each item, the standard deviation indicates the degree of variability or spread of responses around the mean for each item.The means and standard deviations serve as crucial statistical measures to understand the overall perception and variability of responses in the training program evaluation.
For the means, Table 6 shows the evidence that participants generally perceived the training program positively across all aspects evaluated.The mean scores range from approximately 4.302 to 4.681, indicating that participants, on average, found the program favorable.Moreover, examining the standard deviations alongside the means offers valuable insights into the consistency of opinions among participants.Lower standard deviations, such as those around 0.5 to 0.6, suggest more agreement among respondents regarding specific aspects of the training program.For instance, items like the material were presented in a manner appropriate to my training needs and relevance of subject content, combining theory and practice had relatively lower standard deviations, indicating more consistent opinions among participants.Analyzing the means reveals that participants generally held favorable views about these specific aspects of the training environment.The mean scores, ranging from approximately 4.5 to 4.7, indicate a positive perception overall.Additionally, examining the standard deviations alongside the means provides insights into the consistency or variability of opinions among participants.Lower standard deviations, such as those around 0.5 to 0.6, suggest more agreement among respondents regarding specific aspects of the training environment.For example, items like the appropriateness of training techniques and the suitability of services provided had relatively lower standard deviations, indicating a more consistent perception among participants.On the other hand, higher standard deviations, as observed in the suitability of facilities item of 0.805, suggest a greater variability in participant responses.This implies that opinions on the adequacy of facilities, including the platform, notes, and live training, varied considerably among respondents.Table 8 presents various aspects related to learning outcomes and experiences within a training program.Across all items related to learning experiences, the mean scores range from approximately 4.5 to 4.7, indicating a generally positive perception of the program's impact on knowledge acquisition, skill development, motivation, and attitude change.The Standard Deviation values alongside the means demonstrate the degree of variability or spread of responses around the mean for each criterion.Lower Standard Deviation values, such as those around 0.5 to 0.6, indicate more agreement or consensus among respondents regarding specific aspects of the learning outcomes.For instance, the criterion "through the training program, I learned about some theory, practices, and learned information I did not know before" has a relatively low standard deviation of 0.527, suggesting more uniformity in participants' agreement about gaining new knowledge.Conversely, higher standard deviation values, like the one for "The training program has helped to change my attitude towards the topic and training area of 0.656 suggest a greater variability in participant responses.This implies that opinions about the program's impact on attitude change towards the training topic and area varied considerably among respondents.Table 9 outlines various aspects related to behavior under Level 3 of the Kirkpatrick Model that changes perceived by participants after engaging in a training program.Mean scores represent the average evaluation by participants for each behavioral aspect.Across the different items related to behavior, the Mean scores range from approximately 4.4 to 4.7, indicating a generally positive perception of the program's impact on behavioral changes related to trading roles and performance improvement.Standard Deviation values, when considered alongside the means, illustrate the degree of variability, or spread of responses around the mean for each criterion.Lower Standard deviation values, such as those around 0.5 to 0.6, suggest more agreement or consensus among respondents regarding specific behavioral changes resulting from the program.For instance, the criterion the training program helped me to develop trading behavior has a relatively low standard deviation of 0.549, indicating more uniformity in participants' agreement about the program's helped me to develop trading behavior.Conversely, higher Standard Deviation values, such as the one for my trading behavior changed after completing this program 0.805, suggest a greater variability in participant responses.This implies that opinions about the program's impact on actual behavioral changes varied considerably among respondents.Next, the open-ended responses were assessed thematically as in Figure 1: Feedback on Training.Though the majority are somewhat on a positive note about the program, there were three outlier responses indicating areas for improvement.For instance, poor facilitators (N=1), the comment from the community is as below: "Classes were conducted three times only.The facilitators provided a Chapter 2 video with very poor quality in the last class.There was no more class after the video.This group of facilitators were dishonest and looked like they wanted to cheat their lecturers".This has posited a different view regarding the program outcome.Although the program's objective was set to educate the community on stock trading, such a response would undermine the community's confidence in the future of the program and of the overall institution.Nonetheless, this indicates an opportunity for improvement and the need to have better governance to serve as a check and balance.

Conclusion & Recommendations
This study provides insights into community-based learning on Malaysian stock market trading.All instruments (L1 through L4) indicate a good reliability index.Level 1 of the Kirkpatrick Model shows an overall positive reaction to the trainer, reaction to the training delivery, and reaction to the training environment.Positive reactions often correlate with increased engagement and willingness to learn.Understanding participants' reactions helps trainers gauge the training's initial impact.The 2 nd level Kirkpatrick Model assesses the actual knowledge, skills, and competencies gained by participants because of the training.It examines whether participants acquired the intended learning outcomes and the extent to which they can apply the newly acquired knowledge and skills.Evaluation at this level verifies if the training achieved its intended learning outcomes.Assessing what participants have learned helps in identifying areas where further reinforcement or support might be needed.Level 3 of the Kirkpatrick Model focuses shifts to observing changes in participants' behavior or performance in their actual roles because of the training.It assesses whether they apply the newly acquired knowledge and skills in their environment.The results evaluating behavioral changes help in understanding the impact of training.The final level assesses the result or broader impact of the training program on the training program on the community.The shows a major positive impact on how the training impacts towards community and achieves the objective of the training.Nonetheless, hiccups were coming from the small negative response to the program that may offer new insight into the best practice in governing the process of CBL among the students and the communities.
This highlights the potential of community-based methods in enhancing comprehension and involvement with intricate financial systems, potentially promoting financial literacy and socio-economic awareness among communities, and ultimately enabling them to engage in real-world trading.While indicating the promise of community-based learning as an educational tool in the financial market's domain, a thorough evaluation of its efficacy, ethical considerations, and real-world applicability is imperative.This also corroborates with the past studies that highlighted multiple symbiosis among students and communities.For instance, it can enhance professional development, critical thinking, and civic engagement (Flinders 2013;Rock 2021).It also provides students with hands-on learning experiences and a deeper understanding of diversity and social justice (Nicholas 2011).Furthermore, it can improve student success and retention, and strengthen university-community relations (Rock 2021).Overall, community-based learning is a valuable tool for both students and the communities they serve.
Ultimately, this SULAM-CBL initiative aligns with the Sustainable Development Goal (SDG) on quality education, specifically SDG 4: Quality Education.SDG 4, is one of the 17 Sustainable Development Goals established by the United Nations in 2015 as part of the 2030 Agenda for Sustainable Development.Through the program, communities have acquired knowledge, skills, experience, and confidence before entering the realm of actual trading.As this study only analyzed the descriptive outcome of the program, future studies with different depths of analysis are imperative to be conducted to explore more on the effectiveness of community-based learning.

Figure 1 :
Figure 1: Feedback on Training, Level 4 Results

Table 4 .
Instrument Reliability Alpha measures the internal consistency reliability of a scale or set of items within a construct.It ranges between 0 and 1.Higher values of Cronbach's Alpha indicate greater reliability.Typically, values above 0.7 are considered acceptable, and values above 0.8 are desirable.Table

Table 5 .
Reactions to the trainer

Table 6 .
Reactions to the training delivery

Table 7 .
Reactions to the training environment

Table 10 .
Results Results were collected a month after the training program through open-ended and closed-ended questions.Table 10 highlights various aspects regarding the perceived impact of a training program on the community.Mean scores represent the average evaluation by participants for each aspect related to the training program's impact on the community after the program ended.Across all items related to community impact, the Mean scores range from approximately 4.6 to 4.7, indicating a generally positive outlook of the program's effect on the community.Lower Standard Deviation values, such as those around 0.55 to 0.56, suggest more agreement or consensus among respondents regarding the specific impacts of the training program on the community.