Abstract
The dataset examines factors influencing social commerce adoption among Vietnamese Generation Z university students, focusing on technology perceptions, social participation, and trust in platform in shaping purchase intention and actual usage behavior. A structured questionnaire, based on the Technology Acceptance Model (TAM), captured eight constructs: Perceived Ease of Use (PEU), Perceived Usefulness (PU), Familiarity with Social Commerce (FSC), Social Participant (SP), Trust in Platform (TP), Intention to Buy (IB), Actual Usage Behavior (AUB), along with demographics (Income, Gender, Purchase Frequency, Area of Residence). Data were collected over six months (October 2024–March 2025) using a mixed-mode approach. Items were measured on a 5-point Likert scale, adapted from prior studies and validated through a translation–back-translation process. Online surveys were distributed via Google Forms on social media platforms (Facebook, Zalo), while offline data were gathered through QR codes at universities. After screening, 757 valid responses were retained. This dataset enables TAM-based model testing, replication studies, cross-cultural analyses, and PLS-SEM applications in digital commerce research.
Keywords: Perceived ease of use, Perceived usefulness, Familiarity with social commerce, Social participant, Trust in platform, Intention to buy, Actual usage behavior, Vietnam Gen Z student
Specifications Table
| Subject | Social Sciences |
| Specific subject area | Digital commerce adoption, technology acceptance, and consumer behavior |
| Type of data | Table, Figure. Raw. |
| Data collection | A structured questionnaire based on validated TAM constructs was distributed via Google Forms (online) and QR codes on flyers (offline), targeting Vietnamese Gen Z students (aged 18–28) active in social commerce. Screening excluded incomplete or inattentive responses. |
| Data source location | City/Town/Region: Multiple universities across Vietnam. Country: Vietnam |
| Data accessibility | Repository name: Mendeley Data Data identification number: 10.17632/g59k85hrd5.2 Direct URL to data: https://data.mendeley.com/datasets/g59k85hrd5/2 |
| Related research article | None |
1. Value of the Data
-
•
This dataset provides original, large-scale survey data (757 valid responses) on social commerce adoption among Generation Z university students in Vietnam, a rapidly growing digital consumer segment. The dataset was collected using non-probability sampling and includes a higher proportion of female and lower-income respondents, which should be considered when interpreting or reusing the data.
-
•
The dataset enables researchers to test and extend the Technology Acceptance Model (TAM) in social commerce contexts, using constructs such as perceived ease of use, perceived usefulness, familiarity with social commerce, social participation, trust in platform, intention to buy, and actual usage behavior.
-
•
Researchers can use this dataset for replication studies, cross-cultural comparisons, predictive modeling, structural equation modeling (PLS-SEM) and multigroup-analysis (MGA) to test, validate, or extend new digital commerce adoption behavior frameworks, thereby gaining deeper insights into how cognitive and social factors jointly influence social commerce behavior.
-
•
The dataset is fully anonymized, well-structured, and openly accessible, allowing researchers to conduct secondary analyses, meta-analyses, and exploratory research without additional data collection costs.
-
•
Practitioners and policymakers can utilize this dataset to explore technology adoption patterns among young consumers, providing evidence-based insights for improving platform design, trust mechanisms, and user engagement strategies in social commerce.
2. Background
This dataset investigated factors influencing social commerce (s-commerce) adoption among Generation Z university students in Vietnam. Social commerce refers to the integration of social media and online shopping, where users interact, share experiences, and purchase products through social networking platforms [1]. Generation Z is defined as individuals born between 1997 and 2012 [2]. This cohort represents digital natives who have grown up with constant access to the internet, social media, and mobile technology, making them a key segment for studying social commerce adoption. Given the definitions, this demographic is particularly and highly active on social networking platforms and increasingly engaged in online purchasing. The motivation for compiling this dataset stems from the growing attention to digital entrepreneurship in Vietnam, supported by institutional factors such as government initiatives, educational programs, and family encouragement that foster digital market participation [[3], [4], [5]]. These developments highlight the importance of understanding technology adoption behaviors within a rapidly changing digital ecosystem. As of January 2025, Vietnam had approximately 76.2 million social media users, representing about 75 % of the total population, reflecting a 5 % year-over-year increase [6]. According to Statista [7], the number of social media users in Vietnam is projected to rise by 22.7 million between 2024 and 2029, reaching nearly 98 million users, underscoring the rapid expansion of social media engagement in the country. On the other hand, the growing use of digital platforms such as Facebook, Zalo, and TikTok have fostered the rise of social commerce. These platforms reduce marketing costs, facilitate peer-to-peer transactions, and create opportunities for small businesses and individuals to engage in digital entrepreneurship. Alongside this rapid growth, the increasing use of social commerce platforms as entrepreneurial channels has created a competitive digital marketplace where user trust, technology acceptance, and social interactions play crucial roles in shaping purchasing behaviors. Prior research has widely applied constructs such as perceived usefulness, perceived ease of use, and trust to explain social commerce adoption [1]. Furthermore, Generation Z is a critical demographic for the future of s-commerce and digital market expansion due to their high level of digital fluency, early adoption of emerging technologies, and active participation in social media-driven commerce. Despite this, empirical datasets focusing on this group in Vietnam remain limited, particularly in capturing the combined effects of social, technological, and behavioral factors influencing online purchasing. A recent meta-analysis by Jadil et al. [8]. and Dwivedi et al. [9]. also confirmed these constructs as core predictors of user intention and behavior, underscoring the relevance of our dataset. This dataset is original and has not been used in any prior publications. It was developed to support ongoing research on technology acceptance and digital entrepreneurship in emerging markets. By providing structured survey data, it addresses this gap and enables replication studies, model testing, and comparative analyses in technology adoption and social commerce.
3. Data Description
A structured survey was conducted between October 2024 and March 2025 to collect data on factors influencing social commerce adoption among Generation Z university students in Vietnam. A non-probability convenience sampling approach was used to recruit participants across multiple universities, student cafés, and online social media communities. The survey was distributed via Google Forms on platforms such as Facebook and Zalo and supported offline with QR-coded flyers in public student areas. This approach ensured broad accessibility to the target population actively using social commerce platforms.
The dataset is hosted in Mendeley Data and consists of three files [10]. The first file, S-COMMERCE_GenZ_UniversityStudent_757_DIB.csv, contains the final cleaned and anonymized dataset with 757 valid responses out of an initial 1000 collected responses. A total of 243 responses were excluded during screening due to incomplete submissions or straight-lining behavior, where the same option was selected for all questions, indicating a lack of genuine engagement. All participants were informed about the purpose of the study and provided digital informed consent before starting the survey. Participation was entirely voluntary, with the right to withdraw at any time, and no personally identifiable information was collected, ensuring anonymity and no harm to respondents. Each row represents an individual respondent, and columns correspond to demographic variables and survey items measuring seven constructs: Perceived Usefulness (PU), Perceived Ease of Use (PEU), Familiarity with Social Commerce (FSC), Social Participation (SP), Trust in Platform (TP), Social Commerce Intention (IB), and Social Commerce Behavior (AUB) (illustrated in Fig. 1). The second file, SURVEY ON SOCIAL COMMERCE BEHAVIORS.docx, contains the complete English version of the questionnaire, including all item wording, response options, and instructions. The third file provides the Vietnamese version of the survey used in the field to ensure cultural and linguistic appropriateness. These files offer transparent documentation of the data collection instrument and process, enabling replication studies, cross-cultural comparisons, and further methodological research in technology adoption and social commerce.
Fig. 1.
Conceptual model.
The questionnaire was developed based on established literature in social commerce, technology adoption, and consumer behavior. It incorporated multiple validated constructs to capture the behavioral dynamics of social commerce adoption among Vietnamese Generation Z university students. Perceived Usefulness (PU) was measured using a four-item scale adapted from Venkatesh et al. [11]. and Jain et al. [12], evaluating how social commerce platforms enhance online purchasing performance and effectiveness. Perceived Ease of Use (PEU) employed a three-item scale derived from Venkatesh et al. [11]. and Jain et al. [12], assessing the simplicity and effort required to use these platforms. Familiarity with Social Commerce (FSC) was measured with a four-item scale adapted from Shao et al. [13]. and Lu et al. [14], focusing on prior experience with platform features. Social Participant (SP) used a three-item scale from Srivastava and Panigrahi [15]. and Cornwell and Waite [16]. to capture interaction frequency with friends and relatives via social commerce. Trust in Platform (TP) was assessed with a three-item scale adapted from Shao et al. [13], reflecting platform competence, fairness, and reliability perceptions. Social Commerce Intention (IB) and Social Commerce Behavior (AUB) were employed from Venkatesh et al. [11,17,18], measuring users’ likelihood of purchasing or recommending products on these platforms and their actual behavior. The questionnaire was initially developed in English and then translated into Vietnamese using a translation–back-translation method. Three academic experts reviewed the translated version to ensure semantic equivalence and cultural appropriateness [19]. All items were rated on a five-point Likert scale (1 = “Strongly Disagree,” 5 = “Strongly Agree”). The final dataset was cleaned, coded, anonymized, and prepared for replication studies, cross-cultural comparisons, and advanced statistical analyses, ensuring transparency and high usability for future research. Table 1 presents the detailed measurement items, codes, descriptions, and sources used to operationalize each construct in this study.
Table 1.
Measurement Items.
| Construct | Item | Item Description | Source(s) |
|---|---|---|---|
| Perceived Usefulness (PU) | PU1 | Using social commerce platforms improves my online shopping performance | Venkatesh et al. [11]. and Jain et al. [12]. |
| PU2 | Using social commerce platforms increases my productivity in finding products | ||
| PU3 | Using social commerce platforms enhances my effectiveness in making online purchases | ||
| PU4 | I find social commerce platforms useful for my online transactions | ||
| Perceived Ease of Use (PEU) | PEU1 | My interaction with the social commerce platform is clear and understandable | Venkatesh et al. [11]. and Jain et al. [12]. |
| PEU2 | Interacting with the social commerce platform does not require a lot of mental effort | ||
| PEU3 | It is easy for me to get the social commerce platform to do what I want it to do | ||
| Familiarity with Social Commerce (FSC) | FSC1 | I am familiar with social commerce platforms through experiencing related social commerce applications (e.g., using platforms like Facebook Marketplace, Amazon, or eBay) | Shao et al. [13]. and Lu et al. [14]. |
| FSC2 | I am familiar with social commerce platforms through reading news, blogs, or other materials about social commerce | ||
| FSC3 | I am familiar with social commerce platforms through communicating with others about social commerce, such as discussing experiences, reviews, or best practices | ||
| FSC4 | Communicating with others about social commerce, such as discussing experiences, reviews, or best practices | ||
| Social Participant (SP) | SP1 | Frequency of attending online group meetings or events organized through social commerce platforms | Panigrahi [15]. and Cornwell and Waite [16]. |
| SP2 | Frequency of socializing with friends via social commerce platforms (e.g., commenting, sharing posts, or direct messaging) | ||
| SP3 | Frequency of socializing with relatives through interactions on social commerce platforms | ||
| Trust in Platform (TP) | TP1 | I think social commerce platforms are competent and effective in facilitating online shopping and transactions | Shao et al. [13]. |
| TP2 | I think social commerce platforms act in the user's best interests | ||
| TP3 | I believe that social commerce platforms can be trusted at all times | ||
| S-Commerce Intention (IB) | IB1 | I am very likely to provide the online vendor with the information it needs to better serve my needs. | Venkatesh et al. [11,17,18]. |
| IB2 | I am willing to provide my experiences and suggestions when my friends on social commerce platforms (e.g., Facebook, Zalo, and TikTok) want my advice on buying something. | ||
| IB3 | I am willing to buy the products recommended by my friends on social commerce platforms (e.g., Facebook, Zalo, and TikTok). | ||
| IB4 | I will consider the shopping experiences of my friends on social commerce platforms (e.g., Facebook, Zalo, and TikTok) when I want to shop. | ||
| S-Commerce Behavior (AUB) | AUB1 | Using social commerce platforms is a pleasant experience. | Venkatesh et al. [11,17,18]. |
| AUB2 | I really use social commerce platforms to shop safely online. | ||
| AUB3 | I spend a lot of time on social commerce platforms. | ||
| AUB4 | I use social commerce platforms on a regular basis. |
Source: Author’s work.
Table 2 summarizes the descriptive statistics for all measurement items included in the dataset. Each construct is represented by multiple items rated on a five-point Likert scale (1 to 5). The table reports each item's mean, median, minimum, and maximum values, alongside variability (standard deviation) and distribution shape (excess kurtosis and skewness). These statistics provide a preliminary understanding of the dataset, showing central tendencies and response dispersion across all constructs. Overall, the item means ranged from 3.45 to 3.79, indicating moderately positive responses to statements related to perceived usefulness, ease of use, familiarity with social commerce, social participation, trust in platform, purchase intentions, and actual usage behavior. Standard deviations were generally below 0.90, suggesting a relatively consistent pattern of responses among participants. Skewness values were negative across most items, indicating a slight tendency toward agreement, while excess kurtosis values suggest that response distributions were generally close to normal. These results demonstrate the suitability of the dataset for subsequent structural modeling and advanced statistical analyses.
Table 2.
Descriptive Statistics of the Constructs' Items.
| Item | Mean | Median | Min | Max | Standard Deviation | Excess Kurtosis | Skewness |
|---|---|---|---|---|---|---|---|
| PU1 | 3.753 | 4.000 | 1.000 | 5.000 | 0.756 | 1.370 | −0.621 |
| PU2 | 3.770 | 4.000 | 1.000 | 5.000 | 0.778 | 1.391 | −0.706 |
| PU3 | 3.779 | 4.000 | 1.000 | 5.000 | 0.764 | 0.886 | −0.511 |
| PU4 | 3.791 | 4.000 | 1.000 | 5.000 | 0.818 | 1.195 | −0.714 |
| PEU1 | 3.674 | 4.000 | 1.000 | 5.000 | 0.793 | 1.444 | −0.701 |
| PEU2 | 3.687 | 4.000 | 1.000 | 5.000 | 0.777 | 1.305 | −0.697 |
| PEU3 | 3.740 | 4.000 | 1.000 | 5.000 | 0.791 | 1.213 | −0.577 |
| PEU4 | 3.720 | 4.000 | 1.000 | 5.000 | 0.785 | 1.005 | −0.512 |
| FSC1 | 3.741 | 4.000 | 1.000 | 5.000 | 0.841 | 1.013 | −0.684 |
| FSC2 | 3.740 | 4.000 | 1.000 | 5.000 | 0.796 | 0.625 | −0.459 |
| FSC3 | 3.712 | 4.000 | 1.000 | 5.000 | 0.815 | 0.854 | −0.555 |
| SP1 | 3.499 | 4.000 | 1.000 | 5.000 | 0.838 | 0.395 | −0.396 |
| SP2 | 3.690 | 4.000 | 1.000 | 5.000 | 0.809 | 0.823 | −0.500 |
| SP3 | 3.672 | 4.000 | 1.000 | 5.000 | 0.825 | 0.531 | −0.490 |
| SP4 | 3.546 | 4.000 | 1.000 | 5.000 | 0.847 | 0.537 | −0.456 |
| TP1 | 3.613 | 4.000 | 1.000 | 5.000 | 0.830 | 1.184 | −0.674 |
| TP2 | 3.581 | 4.000 | 1.000 | 5.000 | 0.771 | 1.113 | −0.594 |
| TP3 | 3.477 | 4.000 | 1.000 | 5.000 | 0.862 | 0.454 | −0.442 |
| IB1 | 3.454 | 4.000 | 1.000 | 5.000 | 0.880 | 0.803 | −0.660 |
| IB2 | 3.695 | 4.000 | 1.000 | 5.000 | 0.823 | 0.931 | −0.652 |
| IB3 | 3.572 | 4.000 | 1.000 | 5.000 | 0.827 | 0.595 | −0.371 |
| IB4 | 3.703 | 4.000 | 1.000 | 5.000 | 0.826 | 0.841 | −0.571 |
| AUB1 | 3.761 | 4.000 | 1.000 | 5.000 | 0.793 | 1.665 | −0.790 |
| AUB2 | 3.740 | 4.000 | 1.000 | 5.000 | 0.783 | 1.324 | −0.667 |
| AUB3 | 3.523 | 4.000 | 1.000 | 5.000 | 0.874 | 0.228 | −0.433 |
| AUB4 | 3.683 | 4.000 | 1.000 | 5.000 | 0.846 | 0.681 | −0.539 |
Source: Author’s work.
Table 3 summarizes the reliability and convergent validity results for all constructs measured in this dataset. Following the guidelines of Hair et al. (2021) for Partial Least Squares Structural Equation Modeling (PLS-SEM), each construct was assessed using multiple items, and the table reports factor loadings, variance inflation factors (VIF), Cronbach’s alpha (CA), composite reliability (CR), and average variance extracted (AVE). These metrics provide evidence of the measurement model's internal consistency and construct validity. All factor loadings exceeded the recommended threshold of 0.70, indicating that each item contributed strongly to its underlying construct. The CA and CR values for all constructs were above 0.85, surpassing the commonly accepted reliability standard of 0.70. The AVE values for each construct were above 0.70, suggesting that the latent construct explained a substantial proportion of variance compared to measurement error. VIF scores were below 3.5, confirming that multicollinearity was not a concern in this dataset. These results demonstrate that the measurement model possesses high internal consistency, convergent validity, and reliability, supporting its use for subsequent structural equation modelling and hypothesis testing.
Table 3.
Result of reliability and convergent validity.
| Construct | Item | Factor Loading | VIF | CA | CR | AVE |
|---|---|---|---|---|---|---|
| S-Commerce Behavior (AUB) | AUB1 | 0.861 | 2.387 | 0.880 | 0.918 | 0.736 |
| AUB2 | 0.870 | 2.561 | ||||
| AUB3 | 0.831 | 2.042 | ||||
| AUB4 | 0.869 | 2.379 | ||||
| Familiarity with Social Commerce (FSC) | FSC1 | 0.920 | 3.034 | 0.905 | 0.940 | 0.840 |
| FSC2 | 0.909 | 2.694 | ||||
| FSC3 | 0.921 | 3.093 | ||||
| S-Commerce Intention (IB) | IB1 | 0.794 | 1.774 | 0.860 | 0.905 | 0.704 |
| IB2 | 0.855 | 2.167 | ||||
| IB3 | 0.857 | 2.162 | ||||
| IB4 | 0.849 | 2.146 | ||||
| Perceived Ease of Use (PEU) | PEU1 | 0.886 | 2.741 | 0.912 | 0.938 | 0.791 |
| PEU2 | 0.877 | 2.583 | ||||
| PEU3 | 0.908 | 3.214 | ||||
| PEU4 | 0.887 | 2.849 | ||||
| Perceived Usefulness (PU) | PU1 | 0.893 | 3.021 | 0.922 | 0.945 | 0.811 |
| PU2 | 0.903 | 3.277 | ||||
| PU3 | 0.903 | 3.276 | ||||
| PU4 | 0.902 | 3.285 | ||||
| Social Participant (SP) | SP1 | 0.819 | 1.878 | 0.866 | 0.909 | 0.714 |
| SP2 | 0.865 | 2.277 | ||||
| SP3 | 0.857 | 2.273 | ||||
| SP4 | 0.838 | 1.989 | ||||
| Trust in Platform (TP) | TP1 | 0.886 | 2.214 | 0.868 | 0.919 | 0.791 |
| TP2 | 0.890 | 2.271 | ||||
| TP3 | 0.892 | 2.321 |
Source: Author’s work.
Table 4 presents the results of the discriminant validity assessment using the Fornell–Larcker criterion and the HTMT ratio, by guidelines proposed by Hair et al. [20]. for PLS-SEM. According to the Fornell–Larcker criterion, the square root of the AVE for each construct (shown on the diagonal) should be greater than its correlation with any other construct. This condition was met for all constructs, indicating satisfactory discriminant validity. In addition, HTMT values were computed to assess the degree of discriminant separation among constructs. All HTMT values remained well below the conservative threshold of 0.9, confirming that the constructs are empirically distinct. These findings provide robust evidence of discriminant validity, ensuring that the constructs in the measurement model represent conceptually and statistically distinct variables, which supports the integrity of the subsequent structural model analysis.
Table 4.
Discriminant Validity with Fornell–Larcker Criterion and HTMT Ratio.
| Fornell-Larcker criterion |
|||||||
|---|---|---|---|---|---|---|---|
| AUB | FSC | IB | PEU | PU | SP | TP | |
| AUB | 0.858 | ||||||
| FSC | 0.747 | 0.917 | |||||
| IB | 0.685 | 0.659 | 0.839 | ||||
| PEU | 0.805 | 0.780 | 0.695 | 0.889 | |||
| PU | 0.774 | 0.804 | 0.658 | 0.793 | 0.900 | ||
| SP | 0.693 | 0.768 | 0.643 | 0.722 | 0.720 | 0.845 | |
| TP | 0.662 | 0.711 | 0.663 | 0.685 | 0.678 | 0.734 | 0.889 |
| Heterotrait-Monotrait Ratio (HTMT) | |||||||
| AUB | FSC | IB | PEU | PU | SP | TP | |
| AUB | |||||||
| FSC | 0.836 | ||||||
| IB | 0.786 | 0.746 | |||||
| PEU | 0.898 | 0.858 | 0.783 | ||||
| PU | 0.859 | 0.880 | 0.737 | 0.865 | |||
| SP | 0.792 | 0.867 | 0.746 | 0.811 | 0.804 | ||
| TP | 0.757 | 0.802 | 0.768 | 0.770 | 0.757 | 0.847 | |
Source: Author’s work.
Table 5 summarizes the structural path coefficients estimated through PLS-SEM, including direct and indirect effects between the model constructs and control variables. The table reports the original sample estimates, standard deviations, t-statistics, and p-values obtained from a non-parametric bootstrapping procedure with 5000 resamples, following the guidelines of Hair et al. [20]. The first section of the table displays direct effects, showing how the independent constructs influence intention, and how intention and demographic factors (area, frequency, gender, income) predict actual usage behavior. The second section presents the indirect effects, specifically capturing mediation effects through intention on usage behavior. Reported coefficients include statistically significant and non-significant results, providing a transparent view of the estimated relationships. This table enables researchers to assess all hypothesized paths' magnitude, direction, and statistical relevance for potential model replication or extension.
Table 5.
Path analysis.
| Path | Original Sample (O) | f-squared | Standard Deviation (STDEV) | T Statistics (|O/STDEV|) | P Values | |
|---|---|---|---|---|---|---|
| Indirect effect | PU -> IB | 0.119 | 0.009 | 0.057 | 2.080 | 0.038 |
| PEU -> IB | 0.297 | 0.060 | 0.071 | 4.200 | 0.000 | |
| FSC -> IB | 0.075 | 0.003 | 0.059 | 1.279 | 0.201 | |
| SP -> IB | 0.101 | 0.007 | 0.053 | 1.906 | 0.057 | |
| TP -> IB | 0.252 | 0.057 | 0.046 | 5.473 | 0.000 | |
| IB -> AUB | 0.683 | 0.878 | 0.029 | 23.386 | 0.000 | |
| AREA -> AUB | −0.040 | 0.003 | 0.026 | 1.512 | 0.131 | |
| FREQ -> AUB | −0.014 | 0.000 | 0.027 | 0.522 | 0.602 | |
| GENDER -> AUB | 0.012 | 0.000 | 0.028 | 0.411 | 0.681 | |
| INCOME -> AUB | 0.003 | 0.000 | 0.027 | 0.119 | 0.905 | |
| Indirect effect | ||||||
| FSC -> IB -> AUB | 0.051 | 0.040 | 1.277 | 0.202 | ||
| PEU -> IB -> AUB | 0.203 | 0.052 | 3.924 | 0.000 | ||
| PU -> IB -> AUB | 0.081 | 0.039 | 2.067 | 0.039 | ||
| SP -> IB -> AUB | 0.069 | 0.036 | 1.903 | 0.057 | ||
| TP -> IB -> AUB | 0.172 | 0.032 | 5.311 | 0.000 | ||
Source: Author’s work.
Table 6 presents the structural model's explanatory power (R²) and predictive relevance (Q²). R² and adjusted R² values indicate the proportion of variance in the endogenous constructs explained by their predictors. The results show that the proposed model explains 47.1 % of the variance in actual usage behavior and 56.6 % of the variance in intention. Q² values, derived using the blindfolding procedure (Q² = 1 – SSE/SSO), assess the model's predictive accuracy, with values above zero indicating predictive relevance as Hair et al. [20]. suggested. Both constructs exhibit substantial predictive relevance, supporting the model’s ability to predict behavioral outcomes within the dataset. These results provide baseline explanatory and predictive metrics that future research can use for model testing and comparison.
Table 6.
Structural Model Explanatory and Predictive Power.
| R Square | R Square Adjusted | Q² (=1-SSE/SSO) | |
|---|---|---|---|
| AUB | 0.471 | 0.467 | 0.342 |
| IB | 0.566 | 0.563 | 0.393 |
Source: Author’s work.
4. Experimental Design, Materials, and Methods
This research adopted a quantitative approach to investigate the behavioral factors influencing social commerce adoption among Generation Z university students in Vietnam. Drawing upon the Technology Acceptance Model (TAM) as the core theoretical foundation, the study extended the model by incorporating social participation and trust-related factors to better capture the behavioral characteristics of young consumers engaging with social commerce platforms. The constructs were selected based on established theoretical models and empirical evidence in technology adoption and social commerce research. Specifically, perceived usefulness and perceived ease of use were derived from TAM to represent cognitive evaluations of platform functionality, while trust and social participation reflected social and relational influences emphasized in social commerce literature. Intention to buy and actual usage behavior were included to capture the behavioral outcomes of these perceptions. The development of the questionnaire was guided by a comprehensive review of relevant literature, ensuring that the constructs were theoretically grounded and contextually appropriate. Items were adapted from well-established scales and refined through expert consultations. This preliminary phase helped ensure the survey items' clarity, semantic accuracy, and cultural relevance before full-scale data collection commenced.
Data was collected between October 2024 and March 2025, combining online and offline channels to reach a diverse student population. Online surveys were distributed through social networking platforms such as Facebook and Zalo, by posting survey links in verified university groups, student association pages, and community channels, while offline surveys were administered in universities and public student areas using QR-coded flyers. Participants were informed about the study’s objectives, voluntary participation, confidentiality measures, and the anonymized handling of responses. Digital informed consent was obtained prior to participation, and no personally identifiable information was collected, ensuring complete adherence to ethical research standards. An initial 1000 responses were gathered, of which 757 valid cases remained after removing incomplete entries and responses showing straight-lining behavior.
The dataset includes detailed demographic information to contextualize behavioral patterns in social commerce usage. Most respondents were female (77.7 %), with male participants accounting for 22.1 % and a small proportion identifying as other. Most participants reported a monthly income below $100 (88.8 %), reflecting the typical financial status of university students in Vietnam. Urban residents comprised 60.9 % of the sample, while 29.3 % were from rural areas and 9.8 % from suburban areas. Daily use of social commerce platforms was reported by 95.8 % of respondents, highlighting their high level of engagement with online shopping activities (Table 7). The second part of the dataset captures the primary research constructs, which include perceived usefulness, perceived ease of use, familiarity with social commerce platforms, social participation, trust in platform, intention to buy, and actual usage behavior. Each construct was operationalized with three to four items adapted from previous research in technology adoption and e-commerce, measured on a five-point Likert scale (1 = “Strongly Disagree,” 5 = “Strongly Agree”). The questionnaire was initially developed in English and translated into Vietnamese following the translation–back-translation procedure (Brislin, 1970), with reviews by three academic experts to ensure semantic equivalence and cultural appropriateness (Van de Vijver & Leung, 2021).
Table 7.
Sample Characteristics.
| Characteristics | Distribution | Frequency | Percentage (%) |
|---|---|---|---|
| Gender | Male | 167 | 22.1 |
| Female | 588 | 77.7 | |
| Other | 2 | 0.3 | |
| Income | Below $100 | 672 | 88.8 |
| $100-$200 | 68 | 9 | |
| $200-$300 | 12 | 1.6 | |
| $300-$400 | 2 | 0.3 | |
| Above $400 | 3 | 0.4 | |
| Area | Urban | 461 | 60.9 |
| Suburban | 74 | 9.8 | |
| Rural | 222 | 29.3 | |
| Frequently | Daily | 725 | 95.8 |
| A few times a week | 14 | 1.8 | |
| A few times a month | 7 | 0.9 | |
| Rarely | 11 | 1.5 |
Note: 1 USD, approximately 25,000 VND during the survey period.
Source: Author’s work.
Data analysis was performed using SmartPLS 3.2.9 under the methodological guidelines of Hair et al. (2021). The process involved data screening to address missing values, outliers, and low-quality responses, followed by a rigorous assessment of the measurement model. Reliability and validity were examined using Cronbach’s alpha, composite reliability, average variance extracted, variance inflation factors, Fornell-Larcker criterion, and HTMT ratio to confirm convergent and discriminant validity. The structural model was then evaluated to estimate path coefficients, test statistical significance via bootstrapping with 5000 resamples, and assess the model's explanatory (R²) and predictive (Q²) power. This methodological approach ensured robust testing of the hypothesized relationships and produced a high-quality dataset suitable for replication studies, cross-cultural comparisons, and advanced statistical modeling.
Limitations
Several limitations should be noted regarding the dataset. First, the study employed a non-probability convenience sampling approach and the overrepresentation of women and low-income students, which may limit the generalizability of findings to the entire Generation Z population in Vietnam. Although the sample size of 757 valid responses is robust for PLS-SEM, it is not nationally representative due to practical constraints in accessing a full sampling frame. Second, the data rely on self-reported responses, which may introduce common method bias or social desirability bias, as participants could have over- or under-reported their social commerce usage behavior and intentions. Third, as the dataset is cross-sectional, it captures perceptions and behavior at a single point in time. As such, it cannot capture evolving trends or longitudinal dynamics in social commerce adoption. Future studies could build upon this dataset by employing a longitudinal or panel design to examine how users’ attitudes, trust, and engagement with social commerce platforms change over time and across different market conditions. Lastly, while rigorous data screening was performed to remove incomplete and low-quality responses (e.g., straight-lining behavior), it is possible that non-engaged respondents or external contextual factors (e.g., marketing campaigns during data collection) could have influenced some responses. When reusing this dataset for replication or comparative studies, these limitations should be considered.
Ethics Statement
All participants were thoroughly briefed on the purpose, objectives, and scope of the research before engaging in the survey, and their informed consent was obtained before data collection. Respondent anonymity was strictly maintained, with no means of re-identifying or contacting participants after submission. Formal approval from an Institutional Review Board (IRB) was not mandated for this study. Even though, the study was reviewed and approved by the Thai Nguyen University of Economics and Business Administration (Letter of Confirmation, October 1, 2024). Participation was entirely voluntary, and the introductory statement of the survey clearly outlined confidentiality assurances, indicating that completing the questionnaire constituted informed consent. The study did not include minors and was conducted by the ethical principles outlined in the Declaration of Helsinki. To safeguard privacy and data security, all responses were fully anonymized, personally identifiable information was excluded, and unique non-traceable codes were assigned to each entry. These measures ensured strict adherence to recognized ethical standards for research involving human subjects.
CRediT Author Statement
Cheng-Kun Wang: Supervision, Methodology, Validation, Writing – review & editing. Chieh-Yu Lin: Project administration, Resources, Supervision, Writing – review & editing. Le Bao Han: Conceptualization, Methodology, Data curation, Formal analysis, Writing – original draft, Visualization.
Acknowledgments
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
We would like to extend our appreciation to all those who took the time to complete the survey. We highly value the insightful critiques provided by the editors and reviewers.
While preparing this work, the authors used ChatGPT and Grammarly to improve readability. After using this tool/service, the authors reviewed and edited the content as needed and took full responsibility for the publication's content.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Data Availability
References
- 1.Hajli N. Social commerce constructs and consumer’s intention to buy. Int. J. Inf. Manage. 2015;35:183–191. [Google Scholar]
- 2.Dimock M. Defining generations: where Millennials end and Generation Z begins. Pew. Res. Center. 2019 [Google Scholar]
- 3.Nguyen P.N.D., Nguyen H.H. Vol. 3. Springer Nature Singapore; Singapore: 2025. Future of digital entrepreneurship: the expanded role of government in the entrepreneurial event model; pp. 1–14. (Information Systems Research in Vietnam). [DOI] [Google Scholar]
- 4.Nguyen P.N.D., Nguyen H.H. Examining the role of Family in shaping digital entrepreneurial intentions in emerging markets. Sage. Open. 2024;14 doi: 10.1177/21582440241239493. [DOI] [Google Scholar]
- 5.Nguyen P.N.D., Nguyen H.H. Unveiling the link between digital entrepreneurship education and intention among university students in an emerging economy. Technol. Forecast Soc. Change. 2024;203 doi: 10.1016/j.techfore.2024.123330. [DOI] [Google Scholar]
- 6.Simon Kemp, Digital 2025: vietnam, DataReportal – Global Digital Insights (2025). https://datareportal.com/reports/digital-2025-vietnam (accessed October 19, 2025).
- 7.Statista, Social media users in Vietnam 2020–2029, 2024. Retrieved from https://www.statista.com/forecasts/1147065/social-media-users-in-vietnam (accessed August 4, 2025).
- 8.Jadil Y., Jeyaraj A., Dwivedi Y.K., Rana N.P., Sarker P. A meta-analysis of the factors associated with s-commerce intention: hofstede’s cultural dimensions as moderators. Internet Res. 2022;33:2013–2057. [Google Scholar]
- 9.Dwivedi Y.K., Ismagilova E., Sarker P., Jeyaraj A., Jadil Y., Hughes L. A meta-analytic structural equation model for understanding social commerce adoption. Inform. Syst. Front. 2023:1–17. [Google Scholar]
- 10.B.H. Le, Social commerce adoption factors among Gen Z University students in Vietnam, Mendeley Data V1 (2025). 10.17632/G59K85HRD5.1. [DOI]
- 11.Venkatesh Morris, Davis Davis. User acceptance of information technology: toward a unified view. MIS Quarter. 2003;27:425–478. doi: 10.2307/30036540. [DOI] [Google Scholar]
- 12.Jain G., Singh H., Chaturvedi K.R., Rakesh S. Blockchain in logistics industry: in fizz customer trust or not. J. Enterprise Inform. Manage. 2020;33:541–558. [Google Scholar]
- 13.Shao Z., Zhang L., Brown S.A., Zhao T. Understanding users’ trust transfer mechanism in a blockchain-enabled platform: a mixed methods study. Decis. Support Syst. 2022;155 [Google Scholar]
- 14.Lu B., Zeng Q., Fan W. Examining macro-sources of institution-based trust in social commerce marketplaces: an empirical study. Electron. Commer. Res. Appl. 2016;20:116–131. [Google Scholar]
- 15.Srivastava S.K., Panigrahi P.K. Social participation among the elderly: moderated mediation model of information and communication technology (ICT) Commun. Associat. Inform. Syst. 2019;44:33. [Google Scholar]
- 16.Cornwell E.Y., Waite L.J. Social disconnectedness, perceived isolation, and health among older adults. J. Health Soc. Behav. 2009;50:31–48. doi: 10.1177/002214650905000103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Venkatesh, Thong Xu. Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology. MIS Quarter. 2012;36:157–178. doi: 10.2307/41410412. [DOI] [Google Scholar]
- 18.Venkatesh V., Davis F.D. Theoretical extension of the technology acceptance model: four longitudinal field studies. Manage. Sci. 2000;46:186–204. doi: 10.1287/mnsc.46.2.186.11926. [DOI] [Google Scholar]
- 19.Van de Vijver F.J.R., Leung K. Cambridge University Press; 2021. Methods and Data Analysis For Cross-Cultural Research. [Google Scholar]
- 20.Hair J.F., Jr, Hult G.T.M., Ringle C.M., Sarstedt M., Danks N.P., Ray S. Springer Nature; 2021. Partial Least Squares Structural Equation Modeling (PLS-SEM) Using R: A workbook. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.