ABSTRACT This study investigates the factors that influence students’ purchasing intentions towards online academic courses.. The research focuses on six key influencing factors: perceiv
INTRODUCTION 122
Research items? nan
We state some questions with the aim of clarifying our stated problem Particularly:
- Perceived lecturer expertise - The lecture has a good background of education (LE1) - The lecture is experienced in his/her field (LE2) - The lecture is knowledgeable (LE3)
- The lecturer is an expert in his/her field (LE4) - Prior learning experience
- My previous experience of learning via this learning platform were great
- used to receive high-quality lectures on this platform (PRE2) - This platform was able to offer high quality (PRE3) - Personal trial experience
- Before purchasing the course, I was able to appropriately try it out (PET 1) - Before purchasing the course, I was permitted to try out the course for a limited amount of time to decide whether it was worthy or not (PET2) - Iwas able to try out the course based on my requirements (PET3)
- I believe the platform is competent to provide me with high-quality courses (T01)
- Ibelieve the platform is trustworthy (T02) - I believe the platform will provide courses that align with their introduction
(T03) -_ Ibelieve the paid courses ơn the platform are reliable (T04) - Performance expectation
- Texpect that paid courses could improve my efficiency of learning or work (PEE1)
- Texpect that paid courses could improve my ability to solve problems (PEE2) - Texpect that paid courses could improve my knowledge reserve for future use
- Ifnecessary, I will consider purchasing a paid course to study in the future (IP 1) - Itis likely that I will purchase paid courses in the future (IP2)
Research sample and SCORD€: - SH HH TH HH ng ng ry 7
Students’ who took online academic courses 145 people were asked to do the survey After the cleaning process, 135 answers were qualified for further analysis
Time: Our group carries out the project as the final exam on Customer Behavior in the first semester of 2023 to study the factors that affect the intentions of students when buying and taking online academic courses From there, understand the relationships between these factors and how they interact to affect students’ purchasing decisions, and then give suggestions to improve the quality of online education and make it more accessible to diverse students
Space: Conduct surveys with students that have taken online courses
CHAPTER II: THEORETICAL BASIS 2.1 Literature review
Online learning and paid courses Some famous knowledge payment platforms (for example, Coursera, Google Garage, LinkedIn Learning, and others) provide a variety of online courses aimed at improving professional knowledge or abilities (e.g programming and development, products and operations, language learning, vocational examination, and other fields)
Therefore, buyers struggle to evaluate the performance of the intangible course product
According to Peinkofer (2016), perceived performance might act as a confirmation of expectations and affect customer behavioral intentions According to social learning theory (Cai 2020), the process of evaluating a course product before purchase may be the expectation formation process
Mechanisms and factors influencing purchase intention toward OPCs Despite the rising relevance of the paid knowledge market, few studies have lately looked at the elements that impact consumer purchase intent for online paid knowledge, utilizing theories such as trust theory (Zhao et al., 2018), information foraging theory (Shi et al., 2020), and so on Most notably, there is currently a scarcity of research on the factors that influence the purchase intentions of OPCs
Taking into account the learner's motivation for OPCs, result expectancies as a mechanism, with performance expectation being one of the most important, have the potential to affect purchase intention Particularly, the performance of the course product impacts the learning performance, which leads to an effect on extrinsic incentive for purchase (i.e attaining personal goals that are separable from the purchase) Second, the predicted level of course product performance may provide important information about whether students believe the course is of high quality According to Cai et al (2020), an online consumer's decision to purchase a course may be impacted by personal expectations about the course's product quality ( e., product performance)
The trust mechanism has also been used in the online paid knowledge context, where trust beliefs are classified into three characteristics (ability, kindness, and integrity) that influence online users’ payment decisions (Zhao et al., 2018) Despite the lack of precise explanations of course quality, trust minimizes transaction uncertainty and risk for online learning customers, thus the course’s quality may be regarded to be high, increasing learners’ willingness to pay
Furthermore, as Shin (2021) suggests, under the dual process model, trust and performance expectations both play significant roles in comprehending a user's decision-making process
Furthermore, present theories on online paid knowledge have severe flaws when it comes to the antecedents of performance expectations and trust Nevertheless, Yuan Chen et al (2021) stated that learning-oriented signals may also be employed to reflect the course product's quality accurately Consequently, based on the criteria suggested by Yuan Chen et al., this study investigates the effects of learning-oriented-related factors on the purchase intention of online learning consumers for OPCs (2021)
Based on the model of Yuan Chen et al (2021), we decided to refer to their scale and hypothesis for this topic a) Trust Many studies have found that e-vendors and potential consumers who trust platforms and sellers are more willing to transact online (Chen et al., 2014; Escobar-Rodriguez and Bonson Fernandez, 2017; Lee et al., 2011) Establishing trust in an online learning platform should be the providers’ primary focus in the context of online learning (Bhagat and Chang, 2018)
Students are more likely to participate in online learning as their trust grows (Sun et al., 2018)
At this point, a trustworthy platform is likely to be perceived as capable of producing high- quality OPCs in order to meet client demand (e.g., an overall improvement in learning performance) As a result, buyers are more likely to pay for the courses Before determining whether to purchase OPCs, online students may assess the platform's skill, personal integrity, and generosity As a result, when online learners believe a platform is trustworthy, they are more likely to purchase OPCs Accordingly, we propose that:
H1: Trust in the platform has a significant positive effect on purchase intention b) Paid course expectations The performance expectation specifies the extent to which using a specific piece of technology will assist users when performing specific tasks (Venkatesh et al., 2012) Post-purchase enjoyment, according to Hsu and Lin (2015), is a function of expectations being satisfied and perceived performance, which influences repurchase intention Consumers may be hesitant to buy if they have low expectations about the items or service before making the purchase (Wen et al., 2011)
Learning consumers who turn to online paid learning may make a purchase in an effort to match the course product's performance with their expectations The degree to which a consumer's expectations of the performance of a course product actually indicate the quality of the course
According to previous research (Alalwan, 2018; Kazancoglu and Aydin, 2018), perceived performance expectations of a product or service enhance customers’ purchase intentions As a result, the anticipation of course product performance may act as a motivator for OPC purchases Based on this discussion, we hypothesize the following:
H2: Performance expectation towards OPCs has a significant positive effect on purchase intention c) Prior learning experience Prior learning experiences on learning platforms may help a student thrive in online learning (Wladis et al., 2014) Prior learning experiences, according to Haverila (2011), have a significant influence on how well students learn in subsequent online courses In this study, previous learning experiences include formal, paid courses as well as free trial courses For instance, prior experience is clearly shown to be the reason for purchase intention (Wang et al., 2020; Yoo and Lee, 2012)
Because information search plays such an important role in client purchasing choices, it is included in the overall online learning experience in the context of paid learning (Bhatnagar et al., 2017) Prior search history (such as searching for trainer introductions and good opinions about lecturers and courses) may affect users’ trust in the platform, allowing for early assessment of course quality, or platforms Consumers who were pleased with their prior purchasing experiences were more likely to return to the website (Liao et al., 2017) In our case, online learners’ happiness with previous learning experiences on the paid platform might help trigger positive thoughts, leading to the belief that the platform can offer them high-quality courses
The majority of the time, result expectations are based on what others have experienced or acquired from product reviews or word-of-mouth (Wen et al., 2011) Consumers’ previous online purchasing experiences will influence how much effort they anticipate needing to put in in the future (Yeo et al., 2017) Thus, we propose the following hypotheses:
H3a: Prior learning experience through the platform has a significant positive effect on trust in the platform
H3b: Prior learning experience through the platform has a significant positive effect on performance expectation towards OPCs (Online Paid Courses)
Proposed researchi riOel . 5-5 <+s + + k+s +3 +4 9 1313151131111 H1 1x 1 14
Perceived lecturer expertise Purchase intention
CHAPTER III: THE METHODOLOGY 3.1 Research methods of the topic
3.1.1 Methods of analysis and synthesis To analyze the factors that lead to students’ decisions to purchase online courses, we have identified and examined several elements These elements include Perceived lecturer expertise;
Prior learning experience; Personal trial experience; Trust; Performance expectation; Purchase intention For each element, the research team has further divided the questions to gain a better understanding of its characteristics and nature
Having understood the nature of the factors, the research team then conducted a synthesis, using SPSS 20, to test their influence on students’ purchase intentions toward online academic courses This synthetic approach aimed to integrate the insights gained from examining each element and reveal their combined impact on students’ decision-making By combining the findings from the various factors, the research team was able to draw a more comprehensive picture of the factors that affect students’ purchasing decisions toward online courses
3.1.2 The method of data collection This research follows a widely accepted method of data collection in scientific research We have collected data by referencing and synthesizing relevant research articles that discuss the factors that influence students’ purchasing decisions toward online courses In addition to drawing insights from previous research, the research team also created a questionnaire and directly collected students’ opinions and evaluation levels
By combining these two approaches, the authors were able to obtain a diverse range of data and insights from various sources This hybrid method allowed the authors to validate and complement their findings and gain a more nuanced understanding of the factors that influence students’ decision-making processes The resulting research paper presents a comprehensive analysis of the various factors that affect students’ purchasing intention toward online courses
Theoretical basics Scale reference from Propose research other prior reseaches model
Cronbach's Regression Correlation EFA Alpha
3.3.1 Research subject All students who have learned online academic courses and may have the intention to buy online courses in the future
3.3.2 Sample size The group took a survey sample of 135 students
3.3.3 Choose a research sample In this study, two methods of non-probability sampling will be applied:
- Convenience sampling method: The authors create a questionnaire about “Factors Influencing Purchase Intention Of Students Toward Online Academic Courses” and post it on Facebook groups, classes, and fan pages with a large number of UEH students such as UEH Study Group, UEH K46 Official
- Snowball method: The authors shared the questionnaire for the friends and asked them to survey and share with their friends
3.3.4 Data collection method To ensure the sample size of N = 135 students, we used Internet survey tools, shared on social networking sites, and surveyed groups with a large number of students, group classes, and friends
Since we have the result from the Google Form, we recheck the data and upload it to the SPSS.20 for further processing and data analysis
CHAPTER IV: RESEARCH RESULT 4.1 Descriptive statistics of the survey sample
We have conducted a survey asking about 145 students through a Google Forms questionnaire survey At the end of the survey, the research team checked and eliminated unsatisfactory or duplicate answer samples, and we finally obtained 135 complete answer samples Our team’s statistical results are displayed in charts with the following details:
Figure 4.1 details information about the gender of students who have answered our survey
Among the respondents, 44.8% were male (65 subjects) and 55.2% were female (80 subjects)
Figure 4.2 provides information about the age of the respondents Overall, the proportion of people who age from 18 to 22 accounts for more than half of the 135 answers, which is 56.6%
Students who are under 18 make up 26.2% of the statistics (38 subjects) The proportions of
“over 25” and “22 to 25” are 6.9% and 10.3%, respectively
Figure 4.3 Monthly income of respondents
Figure 4.3 shows the respondent's monthly income Most of them have under 2 million VND to spend in one month, accounting for a relatively high percentage of 37.9% of the 135 answer samples (55 subjects) Only 15 students seem to have a monthly income of over 10 million
VND, a high amount of money for a month proportions of “2M to 5M” and “5M to 10M” are 25.5% and 26.2%, respectively, accounting for more than 50% of the answer samples
4.1.4 Platforms used by students Figure 4.4 Platforms that students used to take online courses
This figure demonstrates the platforms that provide online courses that are used by students In general, the percentages of Coursera and Google answers are the two highest proportions, at 22.3% and 28.2%, respectively LinkedIn and Udemy seem to be used less, 12.6% and 8.4% in order The highest proportion is "Other’, accounting for 28.6%
4.2 Analytical data from key questions
In order to verify the model proposed, we decided to use SPSS 20 to process the data acquired The process includes
- Reliability test: Cronbach Alpha - Exploratory Factor Analysis - Correlation Analysis - Regression Analysis 4.3 Reliability test: Cronbach Alpha
To start our analytical process, we first take a look at the Reliability test through the Cronbach Alpha The result is expressed as follows
Scale Cronbach’s Alpha N of items
Therefore, the scale is reliable, all the observed variables have good explanations for the factors
From the proposed model mentioned in the 2.4 section, the factor analysis includes 2 results:
4.4.1 EFA for independent variables LE, PRE, and PET
At first, the variables PRE2 had 2 factor loading at the same time Hence, we decided to eliminate it and rerun the EFA
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization a Rotation converged in 3 iterations
Matt C Howard (2015) said that if an observed variable upload in two factors but the difference in load factor is less than 0.2, the observed variable should be considered to be removed In this case, no observed variables were eliminated
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .878 Bartlett's Test of Approx Chi-Square 608.892
Total Variance Explained Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component Total % of Variance | Cumulative % Total % of Variance | Cumulative % Total % of Variance | Cumulative % 1 4.907 49.065 49.065 4.907 49.065 49.065 3.840 38.403 38.403 2 1.383 13.832 62.897 1.383 13.832 62.897 2.449 24.494 62.897 3 744 7.445 70.342
Extraction Method: Principal Component Analysis
Component Plot in Rotated Space
4.4.2 EFA for the intermediate variable T(Trust) and PEE (Performance Expectation)
KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure of Sampling Adequacy .868 Bartlett's Test of Approx Chi-Square 508.840
Initial Eigenvalues Extraction Sums of Squared Loadings Rotation Sums of Squared Loadings Component | Total % of Variance | Cumulative % Total % of Variance | Cumulative % Total % of Variance | Cumulative % 1 4.157 §9.382 59.382 4.157 59.382 59.382 2.972 42.458 42.458 2 1.117 15.957 75.339 1.117 15.957 75.339 2.302 32.881 75.339 3 443 6.322 81.661
Extraction Method: Principal Component Analysis
Extraction Method: Principal Component Analysis
Rotation Method: Varimax with Kaiser Normalization a Rotation converged in 3 iterations
Component Plot in Rotated Space
After 2 times of running EFA, we came to the decision to combine 2 factors LE and PRE for further analysis
Correlations mean_pre_le | mean_pet | mean_t | mean_pee | mean_ip mean_pre_le Pearson Correlation 1 514 779° 703 534
™ Correlation is significant at the 0.01 level (2-tailed)
All factors are appropriate, showing a positive correlation relationship with each other
After considering the linear correlation between the four independent variables and the dependent variable Students’ engagement, the author continues to perform a regression analysis to draw conclusions about the influence of independent variables on the intermediate variables, and the influence of the intermediate variables on the dependent variable Purchase Intention
To test the suitability of the regression model, we ran multivariate 3 regression analysis data on SPSS between the variables:
4.6.1 The regression model between the LE+PRE (Lecturer Expertise + Prior Learning Experience) and PET (Personal Trial Experience) to T (Trust)
Model Squares df Mean Square F Sig
Total 91.062 134 a Dependent Variable: mean_t b Predictors: (Constant), mean_pet, mean_pre_le The regression model is statistically significant
Adjusted R Std Error of Durbin- Model R R Square Square the Estimate Watson
1 790° 625 619 50883 1.718 a Predictors: (Constant), mean_pet, mean_pre_le b Dependent Variable: mean_t
In this case, the independent variables have 61.9% confidence to influence the dependent variable
Model B Std Error Beta t Sig
1 (Constant) 022 273 079 937 mean_pre_le 863 077 697 11.215 000 mean_pet 129 050 159 2.557 012 a Dependent Variable: mean_t
Both 2 factors have the appropriate sigma stat
The combined factor of LE and PRE (Lecturer Expertise and Prior Learning Experience) has a regression relation to the factor Trust (T)
The factor PET (Personal Trial Experience) also has a regression relation to the factor Trust
4.6.2 The regression relation between LE-PRE ((Lecturer Expertise + Prior Learning Experience) and PET (Personal Trial Experience) to PEE (Performance Expectation)
Adjusted R Std Error of Durbin-
Model R R Square Square the Estimate Watson
1 703° 494 486 57586 2.030 a Predictors: (Constant), mean_pet, mean_pre_le b Dependent Variable: mean_pee
In this case, the independent variables have 48.6% confidence to influence the dependent variable
Model Squares df Mean Square F Sig
Total 86.466 134 a Dependent Variable: mean_pee b Predictors: (Constant), mean_pet, mean_pre_le
The regression model is statistically significant, generally
Model B Std Error Beta t Sig
1 (Constant) 701 309 2.272 025 mean_pre_le B57 087 710 9.841 000 mean_pet -.012 057 -.015 -.209 834 a Dependent Variable: mean_pee
Noticeably, the PET factor has a surprisingly high sigma stat (0.834 > 0.05) Therefore, this factor is eliminated
The factor PET (Personal Trial Experience) is eliminated The combined factor of LE and PRE (Lecturer Expertise and Prior Learning Experience) has a regression related to the factor PEE (Performance Expectation)
4.6.3 The regression relationship between T (Trust), PEE (Performance Expectation) to IP (Purchase Intention)
Adjusted R Std Error of Durbin-
Model R R Square Square the Estimate Watson
1 606" 367 357 78485 1.727 a Predictors: (Constant), mean_pee, mean_t b Dependent Variable: mean_ip
In this case, the independent variables have 35.7% confidence to influence the dependent variable
Model Squares df Mean Square F Sig
Total 128.444 134 a Dependent Variable: mean_ip b Predictors: (Constant), mean_pee, mean_t
The regression model is statistically significant, generally
Model B Std Error Beta t Sig
1 (Constant) 787 389 2.024 045 mean_t 636 100 536 6.343 000 mean_pee 135 103 110 1.308 193 a Dependent Variable: mean_ip
Noticeably, the PEE factor has a surprisingly high sigma stat (0.193 > 0.05) Therefore, this factor is eliminated
The factor PEE (Performance Expectation) is eliminated due to sig > 0.05 The factor T (Trust) could affect the dependent variable IP (Purchase Intention) (Bang coefficient)
The result is illustrated as follows
Personal Trial Experience v Purchase Intention
We conducted a comprehensive study on the dynamics driving consumer purchase intentions towards OPCs, focusing on the context in which online learners transition from free to paid knowledge platforms This research disclosed several major results
From the result of the data analysis, the research team has come to the following conclusion:
- The notion that Trust (T) and Purchase Intention (IP) are positively correlated suggests that trust is essential to the paid online knowledge market For instance, courses offered by the online learning platform are considered to be more valuable to purchase if students have trust in the platform's potential
THE METHODOLOOY ĐÀ 2À S41 S2S2 211111111121 81c 15 3.1 Research methods of the tODIC - 5-25 << 2 +2 1+3 112112323191 21111111111 1 15 3.1.1 Methods of analysis and syntẽieS1S - 5-25 SĂ se 15 3.1.2 The method of dafa €olleCfIOTN 5 S5 S4 S11 HH He, 15 Kon nan
Choose a research saImpIÌe +5 + x + Se k1 KH HH TH HH HH ri, 16 3.3.4 Data collection mmethOd - - 5 << sa x22 112 1 11121111 11H ri, 17 3.4 Data analysis method 00 eee
- Convenience sampling method: The authors create a questionnaire about “Factors Influencing Purchase Intention Of Students Toward Online Academic Courses” and post it on Facebook groups, classes, and fan pages with a large number of UEH students such as UEH Study Group, UEH K46 Official
- Snowball method: The authors shared the questionnaire for the friends and asked them to survey and share with their friends
3.3.4 Data collection method To ensure the sample size of N = 135 students, we used Internet survey tools, shared on social networking sites, and surveyed groups with a large number of students, group classes, and friends
Since we have the result from the Google Form, we recheck the data and upload it to the SPSS.20 for further processing and data analysis