A Study on Factors Influencing the Intention to Adopt New Media Platform Services Technology

2.1 Previous Research

The transformation of library media and the reconfiguration of library spaces have been extensively studied. Research by Ko (1996) and Kim et al. (2003) highlights the shift from traditional print-based media to digital media, which has reshaped the spatial configuration and roles of libraries. These studies emphasize the importance of restructuring public library spaces in a user-centered manner to enhance information accessibility. Expanding the role of libraries requires them to function not only as information providers but also as cultural and social hubs for community engagement and youth settlement. By integrating digital media with physical spaces, libraries can serve as critical platforms for fostering social participation and interaction, particularly among younger generations.

Regarding the definition and characteristics of New Media, studies by Baek (2013), Kim (2017), and Lee (2020) suggest that New Media, unlike traditional media, are characterized by interactivity and immediacy. This has fundamentally changed how libraries mediate information and knowledge. Specifically, New Media—such as the internet, social networking services (SNS), and digital TV— necessitate the evolution of libraries from one-way information providers to interactive communication platforms that foster user engagement.

The adoption of new technologies remains a critical issue in the context of technological advancement, prompting the development of theories to explain how users perceive, adopt, and utilize technology (Lee, 2021). The Unified Theory of Acceptance and Use of Technology (UTAUT) is a widely used framework for analyzing technology adoption. Developed by Venkatesh et al. (2003) as an extension of the Technology Acceptance Model (TAM), UTAUT addresses TAM’s limitations, such as its inability to account for external variables and causal relationships. TAM, initially developed by Davis (1989) and grounded in the Reasoned Action Theory (RAT) by Fishbein and Ajzen (1975), aimed to explain individual media selection and usage behavior. However, TAM's narrow focus necessitated more integrative models (Lee, Y., 2021).

Venkatesh et al. (2003) addressed these shortcomings by incorporating concepts from eight theoretical frameworks, including RAT, the Theory of Planned Behavior, Innovation Diffusion Theory, and Social Cognitive Theory. This synthesis resulted in the UTAUT model, as further elaborated by Koo (2015) and Shim (2018). Drawing on key concepts from RAT—belief, attitude, and behavior—UTAUT describes a path where beliefs influence attitudes, which shape behavioral intentions and ultimately lead to technology acceptance (Yu, J. H., et al., 2010).

Taken together, previous studies suggest that technology adoption in libraries is shaped not only by perceived usefulness and ease of use but also by organizational support, peer influence, and professional understanding of emerging media. Accordingly, this study formulates hypotheses linking performance expectancy, effort expectancy, social influence, and facilitating conditions to adoption intention and adoption behavior, while also examining whether age and New Media understanding moderate these relationships. They form the foundation for exploring how libraries can adapt to integrate New Media platform services and function as cultural and social hubs.

However, existing research has primarily focused on library space design, infrastructure improvements, user satisfaction, and the broader social and cultural roles of libraries. Relatively little attention has been paid to the collection and service of materials generated by New Media Platform Services. This gap underscores the urgent need for research on effective methods for collecting, preserving, and providing services tailored to the unique characteristics of materials generated by New Media Platform Services.

This study seeks to address this gap by exploring the factors that influence librarians, as service providers, in the collection, preservation, and use of materials generated on New Media Platform Services. Additionally, it aims to propose practical directions for the collection, preservation, and utilization of New Media platform materials in libraries.

Based on these research gaps, this study connects previous discussions on digital technologies, New Media services, and technology adoption to the development of the research hypotheses. Prior studies on UTAUT suggest that performance expectancy, effort expectancy, social influence, and facilitating conditions are key determinants of technology adoption intention and behavior. In the context of public libraries, these factors are particularly relevant because librarians’ adoption of New Media Platform Services is closely related to digital content management, preservation work, user-service innovation, and institutional support. Therefore, this study applies the UTAUT framework to examine how these factors influence librarians’ adoption intention and adoption behavior toward New Media Platform Services.

2.2 Unified Theory of Acceptance and Use of Technology (UTAUT)

The Unified Theory of Acceptance and Use of Technology (UTAUT) builds upon the Technology Acceptance Model (TAM), which was initially proposed by Davis (1989) and grounded in the Theory of Reasoned Action (TRA). TAM has been widely acknowledged for explaining user acceptance of technologies and services (Choo, R., Ryu, M., 2023). However, the initial Technology Acceptance Model (TAM), despite its contributions, was criticized for limited consideration of external variables beyond perceived usefulness and ease of use, as well as its low causal validity between these variables (Kwon, O., 2010; Jeon, S., et al., 2011).

To address these limitations, Venkatesh et al. (2003) expanded TAM by integrating 33 core variables from various theories and models, including the Theory of Reasoned Action (TRA), Theory of Planned Behavior (TPB), Innovation Diffusion Theory (IDT), and Social Cognitive Theory (SCT), among others. This resulted in the Unified Theory of Acceptance and Use of Technology (UTAUT), which identified four key factors influencing technology acceptance: Performance Expectancy, Effort Expectancy, Social Influence, and Facilitating Conditions. Additionally, UTAUT introduced four moderating variables—Gender, Age, Experience, and Voluntariness of Use—to account for individual differences, enhancing its explanatory power compared to TAM (Lee, H., 2021; Zou, R., Ryu, M., 2023).

Although UTAUT2 later extended the original model by adding hedonic motivation, price value, and habit, this study retained the original UTAUT structure because the research context is an organizational and professional public-library environment rather than a consumer market. In this context, performance expectancy, effort expectancy, social influence, and facilitating conditions remain theoretically appropriate for explaining librarians’ professional adoption of New Media Platform Services. The model has been widely applied in fields such as education, healthcare, and library services, demonstrating its robustness in predicting user acceptance and use behaviors across various contexts (Ko, J., Eom, G., 2022; Feng, S., Ryu, M., 2023).

3.1 Research Procedure

Figure 1 visually presents the research procedure and content in a step-by-step manner. This study aims to explore the factors affecting the adoption behavior of New Media Platform Services in public libraries and analyze the structural relationships. The research is divided into six main stages.

Fig. 1.

Research Procedure and Content

First, the Problem Identification stage involves setting the basic direction for understanding the structural relationships between variables and the influence of factors on the adoption behavior of public library librarians regarding New Media Platform Services. This includes examining the effects of performance expectancy, effort expectancy, social influence, and facilitating conditions in the adoption and use of New Media Platform Services, as well as the moderating effects of new media understanding and age.

Second, the Literature Review and Conceptualization stage focuses on comprehensively reviewing relevant previous studies to establish the theoretical background. The main research is based on the Unified Theory of Acceptance and Use of Technology (UTAUT), and the review includes literature on the status of data collection, utilization, and preservation of New Media Platform Services. Additionally, similar studies and specific cases of new media usage in public libraries were investigated to strengthen the theoretical foundation.

Third, the Research Model Setting stage is where the research model and hypotheses are established based on the literature review and theoretical background investigation. The hypotheses primarily deal with the direct and indirect effects of independent variables such as performance expectancy, effort expectancy, social influence, and facilitating conditions on the dependent variables of adoption intention and adoption behavior, including the moderating effects of age and new media understanding.

Fourth, the Analysis and Hypothesis Testing stage involves applying various statistical analysis techniques to data collected through surveys. The primary analytical methods include frequency analysis, descriptive statistics, exploratory factor analysis, construct validity, and bivariate correlation analysis. Reliability verification, correlation analysis, and structural equation modeling (SEM) are conducted. To test the moderating effects, multi-group analysis is applied.

Fifth, the Discussion stage provides empirical evidence on the adoption and utilization of New Media Platform Services in public libraries and draws policy implications.

Finally, discussions are made on legal and institutional foundations for the preservation of new media data and strategies for effective service provision through public library data collection and utilization methods.

3.2 Survey Content and Survey Structure

Table 1 presents the survey structure used in this study, systematically outlining the design of each item in alignment with the research objectives. The survey investigates librarians’ perceptions and behaviors regarding the adoption, utilization, and preservation of New Media Platform Services in public libraries. Each item is designed to address key aspects required to achieve the research goals.

Table 1.

Survey Structure

Basic respondent information was collected to identify demographic characteristics and ensure the reliability and validity of the survey. Perception items related to New Media Platform Services measure how librarians perceive the definition and functions of New Media Platform Services, forming the basis for analyzing differences in perception.

The section on data collection and preservation evaluates the importance of these tasks and explores the factors and conditions librarians consider when managing new media materials. The current utilization section examines how libraries are presently using New Media Platform Services, reflecting the growing importance of digital materials in library operations.

The future utilization and preservation plans section investigates librarians’ perceptions of strategies and policy directions for effectively using and preserving New Media Platform Services, offering insights for practical implementation. The section on adoption intention analyzes the influence of performance expectancy, effort expectancy, social influence, and facilitating conditions on the intention to adopt New Media Platform Services, grounded in the Unified Theory of Acceptance and Use of Technology (UTAUT).

Lastly, the "Other Comments" section invites respondents to provide detailed, open-ended feedback on the utilization and preservation of New Media Platform Services, enriching the study’s qualitative analysis.

The development of the survey items followed a systematic process to ensure alignment with the research objectives and goals. First, a thorough review of relevant literature was conducted to define the research objectives and construct the survey items. Studies related to the Unified Theory of Acceptance and Use of Technology (UTAUT) were referenced, focusing on key factors such as performance expectancy, effort expectancy, social influence, and facilitating conditions. This literature review helped enhance the validity of the research by grounding the survey items in established frameworks tested for reliability and validity.

Second, survey items were designed to reflect the practical work of librarians in public libraries, particularly their adoption and utilization of New Media Platform Services. These items incorporated aspects of the workflow in public libraries and critical elements related to the preservation and utilization of digital data, ensuring their relevance to the respondents' professional experiences.

Third, a pilot test was conducted to assess the clarity and consistency of the survey items. The pilot involved a small group of librarians, whose feedback was used to evaluate the appropriateness and effectiveness of the questions. This process helped identify and revise unclear or ambiguous items, resulting in a more refined and precise final survey.

Finally, to enhance the survey's reliability and validity, all items were structured using a Likert scale. This approach allowed the quantification of respondents’ perceptions and attitudes while ensuring that the questions enabled clear and consistent responses.

3.3 Research Subjects and Data Collection

The purpose of this study is to investigate the perceptions of public library librarians regarding the use and preservation of New Media Platform Services, to understand the current status of New Media Platform Services in public libraries, and to establish future directions for their collection, preservation, and utilization. Additionally, the study seeks to identify the factors influencing the adoption of New Media Platform Services by public libraries. To achieve this, a survey was conducted targeting public library librarians, focusing on the current status of new media platform operations, their perceptions, and adoption behavior.

To enhance methodological rigor, procedural controls were applied to reduce common method bias. The questionnaire assured anonymity, used neutral wording, and separated key constructs across different sections. In addition, Harman’s single-factor test was conducted as a diagnostic check. Possible non-response bias was also assessed by comparing early and late respondents on major demographic variables and key constructs, and no substantial differences were identified. These procedures suggest that common method bias and non-response bias were unlikely to seriously distort the interpretation of the results.

A pilot test was conducted to ensure the validity of the survey instrument. The pilot test was carried out from August 1 to August 5, 2024, with 10 librarians participating. The collected data were analyzed to confirm the reliability and validity of the measurement tool, and the reliability coefficient exceeded the required threshold, allowing the main survey to proceed. The main survey was conducted from September 19 to September 30, 2024. The survey was distributed to public librarians through an online survey platform. A total of 255 responses were returned, and 232 valid responses were used for the final analysis after excluding incomplete or missing responses. Possible non-response bias was assessed by comparing early and late respondents on major demographic characteristics and key variables. No substantial differences were identified, suggesting that non-response bias was unlikely to seriously affect the interpretation of the results. Of the 1,271 public libraries contacted, 437 agreed to participate, 255 responses were returned, and 232 valid responses were used for analysis. The effective response rate was therefore 18.3% based on all contacted libraries and 53.1% based on libraries that agreed to participate. To assess possible non-response bias, early and late respondents were compared on major demographic variables and key constructs. No substantial differences were identified, suggesting that non-response bias is unlikely to seriously affect the interpretation of the results. The survey links were sent to librarians from these institutions, and 255 surveys were returned. After excluding incomplete responses and missing data, 232 valid surveys were analyzed.

The demographic characteristics of the respondents showed that the highest number of librarians were from Seoul, with 33 participants (14.2%), followed by 31 (13.4%) from Gyeonggi-do, 22 (9.5%) from Jeollanam-do, and 20 (8.6%) each from Gyeongsangbuk-do and Gyeongsangnam-do. In terms of age, the largest group of respondents was in their 30s, comprising 111 librarians (47.8%), followed by 64 in their 40s (27.6%) and 57 in their 20s (24.6%). Regarding gender, 167 respondents were female and 85 were male. Based on the valid sample of 232 respondents, the proportions were 72.0% and 36.6%, respectively. With respect to library experience, 112 respondents (46.3%) worked for 1-5 years, which was the largest proportion, followed by 26 respondents (10.7%) with 6-10 years of experience, and 26 respondents (11.2%) with either 11-15 years or more than 16 years of experience. Regarding their current duties, the largest group of respondents worked in cultural programs (16.4%), followed by reading and lending services (12.9%), acquisitions (11%), and reading room management (8.8%).

This study adhered to ethical standards and obtained approval from [Insert Institution Name]. Participants were informed about the study's objectives and provided with detailed consent forms before participation. The research complied with international standards for human subject research, including data privacy and confidentiality guidelines outlined in the Declaration of Helsinki. These measures ensured that the study was conducted ethically and in accordance with institutional and national regulations.

To improve methodological transparency, this study clarified the sampling procedure and valid-response process. The survey was distributed to public librarians through an online survey platform, and incomplete or missing responses were excluded from the final analysis. Possible non-response bias was assessed by comparing early and late respondents on major demographic characteristics and key variables. No substantial differences were identified, suggesting that non-response bias was unlikely to seriously affect the interpretation of the results.

3.4 Sample Size Adequacy

To ensure an adequate sample size for this study, a sample size calculator based on structural equation modeling (SEM) was utilized. According to Kline (2015), a minimum sample size of 200 is required for reliable results when analyzing models with multiple latent variables. Additionally, the commonly accepted ratio of 5–10 respondents per parameter estimate, as suggested by Bentler and Chou (1987), was applied.

Given the study's research model, which included 12 hypotheses and multiple variables, a minimum of 220 participants was determined to be necessary. The final sample size of 232 valid responses exceeds this threshold, ensuring the robustness of the analysis and the validity of the findings. Furthermore, the effective sample size aligns with statistical power recommendations (power = 0.80, alpha = 0.05) for detecting medium to large effect sizes, as outlined by Cohen (1992).

This validation confirms that the collected data is sufficient to ensure the reliability and generalizability of the study’s results.

3.5 Research Model and Research Hypotheses

The primary objective of this study is to identify the factors influencing the adoption behavior of New Media Platform Services in public libraries and analyze the structural relationships between variables. To achieve this, the study designates performance expectancy, effort expectancy, social influence, and facilitating conditions as independent variables. adoption intention serves as the mediating variable, while adoption behavior is the dependent variable. Additionally, the moderating effects of new media understanding and age are examined. The research model is depicted in <Fig. 2>.

Fig. 2.

Research Model

The research hypotheses (H) established to explore the factors influencing adoption behavior and analyze the structural relationships between variables are as follows:

Table 2.

Research Hypothesis

3.6 Measurement Tools

The survey instrument for this study was structured into two main sections: one focusing on the perception of New Media Platform Services and the other on the adoption behavior of New Media Platform Services. The measurement tools for performance expectancy, effort expectancy, social influence, facilitating conditions, behavioral intention, and adoption behavior were developed based on operational definitions from prior research. These tools were modified to align with the objectives of this study and were informed by both domestic and international studies.

The survey included five demographic questions regarding library staff. Perception-related items were divided into the following categories: 5 items on the perception of New Media Platform Services, 10 items on the preservation of New Media Platform Services, 3 items on platform operation, 6 items on the status of new media platform ownership and usage, 2 items on challenges in platform operation, 19 items on platform collection, 3 items on considerations when collecting platforms, 3 items on strategies for platform service activation, and 4 items on improvements to platform policies.

Regarding the adoption behavior of New Media Platform Services, as summarized in Table 3, the survey included 4 items each for performance expectancy, effort expectancy, social influence, facilitating conditions, behavioral intention, and adoption behavior.

Table 3.

Operational Definitions of Variables in the Unified Technology Acceptance Model for New Media Platform Adoption

All measurement items related to adoption behavior were rated using a 5-point Likert scale, where 1 = strongly disagree and 5 = strongly agree. Statistical analysis of the survey data was conducted with a significance level of p < 0.05, ensuring the robustness and reliability of the findings.

4.1 Exploratory Factor Analysis and Reliability Analysis

4.2 Confirmatory Factor Analysis

Confirmatory factor analysis was conducted to assess the validity of the measurement model. The measurement model included six latent variables: performance expectancy, effort expectancy, social influence, facilitating conditions, adoption intention, and adoption behavior. Each latent variable was measured using its corresponding observed indicators.

The final measurement model consisted of four items for performance expectancy, two items for effort expectancy, four items for social influence, four items for facilitating conditions, three items for adoption intention, and four items for adoption behavior. For adoption intention, the third item was removed because its squared multiple correlation (SMC) value fell below the acceptable threshold.

The standardized factor loadings of the retained items ranged from 0.61 to 0.97 and were statistically significant, indicating that the observed variables adequately represented their corresponding latent variables. These results support the convergent validity of the measurement model and provide a basis for subsequent structural equation modeling.

The measurement model and the structural model were evaluated separately. Confirmatory factor analysis was conducted to assess the validity of the measurement model, including the relationships between latent variables and their observed indicators. Subsequently, structural equation modeling was performed to test the hypothesized relationships among performance expectancy, effort expectancy, social influence, facilitating conditions, adoption intention, and adoption behavior. The model-fit indices for each model are reported separately to ensure transparency and consistency in the interpretation of the results.

The structural equation model (SEM) employed in this study was designed to examine the relationships between the independent variables—performance expectancy, effort expectancy, social influence, and facilitating conditions—and the dependent variables, intention to adopt and adoption behavior. The model also evaluated the mediating effect of intention to adopt on adoption behavior. Furthermore, the SEM incorporated multiple indicators for each independent variable, enabling effective estimation of the latent variables and providing a comprehensive analysis of the relationships within the framework.

4.3 Discriminant Validity

The discriminant validity of the measurement tools used in this study—performance expectancy, effort expectancy, social influence, facilitating conditions, intention to adopt, and adoption behavior—was evaluated using Average Variance Extracted (AVE) and the correlation coefficients between the constructs. According to Fornell and Larcker (1981), discriminant validity is established if the squared correlation coefficient between constructs does not exceed the AVE.

The AVE was calculated based on the formula proposed by Fornell and Larcker (1981), with an AVE value of 0.5 or higher indicating convergent validity. This implies that at least half of the variance in the items is explained by the construct. Composite Reliability (CR) was also assessed, with a threshold of 0.7 or higher indicating good convergent validity (Anderson and Gerbing, 1986).

The correlation analysis results, summarized in Table 4, revealed significant relationships among the constructs. Performance expectancy exhibited the highest positive correlation with effort expectancy at .625 (p < .01) and with adoption behavior at .622 (p < .01). Effort expectancy showed significant positive correlations with social influence at .570 (p < .01), facilitating conditions at .454 (p < .01), and intention to adopt at .384 (p < .01). Social influence demonstrated significant positive correlations with intention to adopt at .629 (p < .01), adoption behavior at .533 (p < .01), and facilitating conditions at .591 (p < .01). Facilitating conditions showed a positive correlation of .576 (p < .01) with intention to adopt and .822 (p < .01) with adoption behavior.

Table 4.

Correlations Among Key Variables and Evidence of Discriminant Validity

The analysis of AVE and composite reliability confirmed that the thresholds were met for all constructs. Performance expectancy had an AVE of .753 and a CR of .924; effort expectancy had an AVE of .905 and a CR of .950; social influence had an AVE of .658 and a CR of .883; facilitating conditions had an AVE of .661 and a CR of .886; intention to adopt had an AVE of .810 and a CR of .928; and The revised table reports values that meet the logical and statistical requirements for convergent validity, and the explanatory text has been revised accordingly. These results indicate that the constructs demonstrated both sufficient convergent validity and discriminant validity.

4.4 Hypothesis Testing of the Structural Model (Direct Effects)

After confirming the validity of the measurement model, structural equation modeling was conducted to test the hypothesized relationships among performance expectancy, effort expectancy, social influence, facilitating conditions, adoption intention, and adoption behavior. The structural model fit indices were as follows: χ²(df = 174, n = 232) = 518.501, p < .001, χ²/df = 2.980, GFI = .826, CFI = .909, NFI = .870, TLI = .890, and RMSEA = .093. Although GFI and NFI were slightly below the commonly recommended threshold and RMSEA was somewhat high, the overall fit indices were considered acceptable for testing the hypothesized structural relationships.

4.4.2 Direct Effects on Use Behavior

Regarding adoption behavior, performance expectancy demonstrated a significant positive effect (β = 0.294, p < 0.001), while effort expectancy showed a significant negative effect (β = −0.161, p < 0.01). The negative effect of effort expectancy on adoption behavior may indicate that librarians who are highly aware of the effort required to operate New Media Platform Services also recognize the additional workload, role ambiguity, copyright responsibility, and technical maintenance burden involved. Thus, ease of use alone may not translate into actual adoption when the technology introduces new professional responsibilities. This finding suggests that public libraries should address not only usability but also workload allocation, institutional support, and role clarification. Adoption intention exhibited a significant positive effect on adoption behavior (β = 0.753, p < 0.001), indicating that stronger adoption intention leads to higher adoption behavior.

Based on these findings, hypothesis 5 (performance expectancy affects use behavior of New Media Platform Services), hypothesis 6 (effort expectancy affects use behavior of New Media Platform Services), and hypothesis 9 (intention to use affects use behavior of New Media Platform Services) were supported. However, hypothesis 7 (social influence affects use behavior of New Media Platform Services) and hypothesis 8 (facilitating conditions affect use behavior of New Media Platform Services) were rejected.

Table 5.

Direct Effect Analysis Using Path Coefficients

4.5 Indirect Effects

In this study, intention to accept New Media Platform Services was selected as a mediating variable. To evaluate the indirect effects of intention to accept, bootstrapping (500 iterations, 𝑝 < .05) was conducted. The analysis revealed that the indirect effect of performance expectancy on adoption behavior, mediated by adoption intention, was significant, with a 95% confidence interval (CI) of (.029 ~ .420), which does not include zero. Similarly, the indirect effect of social influence on adoption behavior, mediated by adoption intention, was significant, with a 95% CI of (.110 ~ .540). Additionally, the indirect effect of facilitating conditions on adoption behavior, mediated by adoption intention, was significant, with a 95% CI of (.077 ~ .401). These findings confirm the presence of significant indirect effects, as detailed in <Table 6>.

Table 6.

Indirect Effects Analysis Results

The effect decomposition table, which outlines the direct, indirect, and total effects within the research model, is presented in <Table 7>.

Table 7.

Decomposition of Direct, Indirect, and Total Effects

In summary, performance expectancy, social influence, and facilitating conditions demonstrated significant indirect effects on adoption behavior through adoption intention. However, no significant indirect effect was observed for effort expectancy. Based on these results, hypothesis 10 was partially supported.

4.6 Moderating Effect

4.6.1 Moderating Effect by Age

This study examined age as one of two moderating variables (the other being librarians' understanding of new media). The moderating effect was analyzed using Multi-Group Analysis (MGA) within the structural model to assess differences in influence among age groups. Age was categorized into three groups: 20s (57 participants, 24.6%), 30s (111 participants, 47.8%), and 40s and older (64 participants, 27.6%).

The results of the analysis showed significant moderating effects by age, with df = 16, CMIN = 30.560, 𝑝 = .015 (𝑝 < .05), and NFI Delta-1 = .007. For the 20s group, facilitating conditions significantly influenced adoption intention, and acceptance intention had a strong positive effect on adoption behavior. In the 30s group, performance expectancy and social influence significantly influenced acceptance intention, while performance expectancy positively and effort expectancy negatively influenced adoption behavior. In the 40s and older group, effort expectancy, social influence, and facilitating conditions significantly affected acceptance intention, with effort expectancy showing a negative effect.

In all age groups, acceptance intention positively influenced adoption behavior, with the strongest effect observed in the 20s group. This indicates that younger participants, particularly those in their 20s, are more adept at adopting new technologies and integrating them into their work. In contrast, participants in their 30s and 40s and older demonstrated lower confidence in the ease of learning and using New Media Platform Services. Social influence had a stronger impact on acceptance intention in the 30s and 40s groups, whereas it did not significantly affect the 20s group. The 30s group also exhibited higher performance expectancy, influencing both acceptance intention and behavior, suggesting that expectations of technological outcomes play a stronger role for this group.

Based on these findings, hypothesis 11 was supported.

Table 8.

Moderating Effect by Age

4.6.2 Moderating Effect by Level of New Media Understanding

The moderating effect based on librarians' level of understanding of new media was examined in this study. Participants were categorized into three groups: high (56 participants, 24.1%), medium (131 participants, 56.5%), and low (45 participants, 19.4%). When these groups were applied to the research model, the results indicated df = 16, CMIN = 33.820, 𝑝 = .006 (𝑝 < .01), and NFI Delta-1 = .007. These findings confirm a statistically significant moderating effect based on the level of new media understanding.

Table 9.

Research Model Verification Results (Standardized Coefficients)

The group with the lowest level of new media understanding showed a significant positive effect on acceptance intention. Similarly, the group with medium new media understanding also demonstrated a significant positive effect on acceptance intention. All groups exhibited a significant positive impact of facilitating conditions on acceptance intention, with the medium understanding group showing the highest influence.

For adoption behavior, only the group with the highest level of new media understanding showed significant results. Performance expectancy positively influenced adoption behavior, while effort expectancy and facilitating conditions negatively influenced adoption behavior. These findings suggest that higher levels of new media understanding lead to greater acceptance of New Media Platform Services. However, as experience and understanding increase, the need for strong technical support and ease of learning (effort expectancy) becomes less critical. Based on these results, hypothesis 12 was supported

4.7 Hypothesis Testing Results and Model Validation Results

4.7.1 Hypothesis Testing Results

The results of the hypothesis testing conducted in this study are summarized in <Table 10>. The analysis identified performance expectancy, social influence, and facilitating conditions as significant factors influencing acceptance intention. For adoption behavior, significant factors included performance expectancy, effort expectancy, and acceptance intention. The indirect effect analysis further confirmed that performance expectancy, social influence, and facilitating conditions had significant indirect effects on adoption behavior through acceptance intention, whereas effort expectancy did not demonstrate a significant indirect effect. Additionally, the analysis of moderating effects revealed statistically significant influences based on age and new media understanding, confirming their role as moderating variables in the relationships examined.

Table 10.

Hypothesis Testing Results Table

4.7.2 Results of the Research Model Verification

This study examined the factors influencing the adoption behavior of New Media Platform Services in public libraries using the Unified Theory of Acceptance and Use of Technology (UTAUT). It also explored differences in librarians' perceptions of the utilization and preservation of New Media Platform Services. To verify the 12 hypotheses established, structural equation modeling (SEM) was employed for analysis.

The goodness-of-fit of the structural equation model was assessed using the following indices: chi-square statistic (χ²), degrees of freedom (df), Comparative Fit Index (CFI), Incremental Fit Index (IFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). A CFI or IFI value above 0.90 indicates a good model fit, while a RMSEA value below 0.08 is considered acceptable.

The model fit analysis yielded the following results: χ² = 362.147, df = 218, CFI = 0.917, IFI = 0.920, RMSEA = 0.067, and SRMR = 0.045. These values indicate that the model is generally well-fitted and has explanatory power in analyzing the adoption behavior of New Media Platform Services.

Figure 3 presents the results of hypothesis verification for the structural model, encompassing direct effects, indirect effects, and moderating effects, as applied to the research model.

Fig. 3.

Confirmatory Factor Analysis Results (Standardized Estimates

Fig. 4.

Research Model Verification Results (Standardized Coefficients)

Fig. 5.

Research Model Reflecting Hypothesis Testing

This chapter analyzes the impact of various factors on the acceptance of New Media Platform Services and identifies the key drivers influencing users' acceptance intention and adoption behavior.

First, the analysis of technical statistics and normality confirmed that the measurement tools used in the study were statistically appropriate. Mean and standard deviation values revealed that performance expectancy had the highest mean score of 3.36, indicating high expectations for performance when utilizing New Media Platform Services. Both skewness and kurtosis values were within acceptable limits, confirming the normality of the data.

Second, the exploratory factor analysis demonstrated clear distinctions among the independent variables—performance expectancy, effort expectancy, social influence, and facilitating conditions. A cumulative variance of 75.135% indicated that the variables were well explained, and Cronbach’s Alpha values exceeding 0.6 for all factors ensured their reliability.

Third, the confirmatory factor analysis confirmed the structural model’s suitability, as the fit indices were statistically significant. This verified that the constructs used in this study were valid and aligned with the research objectives.

Fourth, discriminant validity was established using the Average Variance Extracted (AVE) and construct reliability measures. Correlation analysis revealed that performance expectancy had a strong positive correlation with effort expectancy and adoption behavior, while social influence and facilitating conditions significantly correlated with acceptance intention. These results clarified the interactions between the study variables.

Fifth, path analysis indicated that performance expectancy, social influence, and facilitating conditions significantly influenced acceptance intention. Performance expectancy and acceptance intention positively affected adoption behavior, whereas effort expectancy had a negative impact on adoption behavior. This suggests that as effort expectancy increases, the likelihood of adoption behavior decreases, potentially reflecting challenges specific to librarians’ roles and familiarity with New Media Platform Services.

Sixth, the analysis of indirect effects through bootstrapping revealed that performance expectancy, social influence, and facilitating conditions had significant indirect effects on adoption behavior. acceptance intention played a crucial mediating role in these relationships.

Seventh, the moderating effects of age and new media understanding were significant for acceptance intention and behavior. Younger respondents in their 20s showed the most positive responses to new media acceptance, while effort expectancy negatively influenced acceptance intention among those in their 30s and 40s. Regarding new media understanding, higher levels of understanding amplified the effect of performance expectancy on adoption behavior.

In conclusion, this study explored how factors related to new media platform acceptance influence acceptance intention and behavior. Performance expectancy, social influence, and facilitating conditions had positive effects on acceptance intention, with facilitating conditions being the most impactful. Performance expectancy also positively influenced both acceptance intention and behavior, whereas social influence and facilitating conditions did not significantly affect adoption behavior. Notably, effort expectancy exhibited a negative impact on adoption behavior, a result that contrasts with traditional expectations.

The significant moderating effects of age and new media understanding highlighted the importance of demographic and experiential factors. Younger respondents showed higher acceptance of new technologies, while greater understanding of new media enhanced the influence of performance expectancy on behavior.

These findings suggest that public libraries should move beyond general technology training and establish role-specific New Media Platform Service strategies. First, librarians should be supported through job redesign that clarifies responsibilities for platform-based material collection, copyright review, metadata creation, preservation, and user service. Second, libraries should develop internal guidelines for collecting and preserving social media, immersive content, and metaverse-based materials. Third, training should be differentiated by age group and level of New Media understanding, with introductory, applied, and preservation-focused modules.

5.1 Factors Affecting Technology acceptance intention

This study found that performance expectancy significantly influences the intention to adopt New Media Platform Services (β = 0.267, p = 0.005) (Hypothesis 1). This aligns with findings by Venkatesh (2012) in the Unified Theory of Acceptance and Use of Technology (UTAUT), which identifies performance expectancy as a major determinant of usage intention. Similarly, Zuiderwijk et al. (2015) reported that performance expectancy positively impacts user intentions in the adoption of open data technologies, and Wong et al. (2015) presented comparable results in mobile advertising contexts. Son et al. (2014) also emphasized the critical role of performance expectancy in shaping the adoption intention for wearable devices. This study reinforces that performance expectancy is a key driver influencing users' intentions to adopt New Media Platform Services.

Conversely, effort expectancy was not found to have a significant effect on adoption intention (β = -0.132, p = 0.081) (Hypothesis 2). This contrasts with findings from Venkatesh et al. (2003) and Zhou et al. (2010), which identified effort expectancy as a core factor in shaping technology adoption intentions. The absence of significance in this study may be attributed to the intuitive and user-friendly design of New Media Platform Services, which likely reduces the perceived effort required. Chen et al. (2014) previously emphasized effort expectancy as crucial for the adoption of age-friendly technologies, but the present study suggests that the high digital literacy of users involved in this research diminishes the impact of effort expectancy.

Social influence was found to have a significant positive effect on adoption intention (β = 0.384, p < 0.01) (Hypothesis 3). This finding is consistent with Venkatesh (2012), who highlighted social influence as a key driver in increasing the technology adoption intentions of early users. Park Ilsun (2012) reported that social influence enhances the adoption intention of mobile credit card services. Similar findings were reported by Martins et al. (2014) in internet banking and Lai & Lai (2014) in mobile commerce, where social influence increased user intentions. The results of this study confirm that the initial adoption of New Media Platform Services is significantly influenced by peer or supervisor recommendations.

Additionally, facilitating conditions were found to significantly affect adoption intention (β = 0.333, p < 0.01) (Hypothesis 4). This is consistent with Venkatesh et al. (2003), who identified facilitating conditions—such as technical support and environmental factors—as essential in enhancing technology adoption. Wang et al. (2021) similarly emphasized the role of facilitating conditions in influencing user intentions in community-based remote education technologies. This study demonstrates that sufficient resources and support significantly increase librarians' intentions to adopt New Media Platform Services.

5.2 Factors Affecting adoption behavior

First, performance expectancy was found to have a significant impact on adoption behavior (β = 0.294, p < 0.01) (Hypothesis 5). This finding is consistent with Zuiderwijk et al. (2015), who reported that performance expectancy plays a crucial role in translating technology use into actual behavior. Similarly, Hoque & Sorwar (2017) found that performance expectancy significantly influences both behavioral intention and actual behavior in the context of mHealth services. This study confirms that when users anticipate positive performance outcomes from a technology, they are more likely to engage with and use the platform.

Effort expectancy also significantly influenced adoption behavior (β = -0.161, p = 0.006) (Hypothesis 6). This result aligns with Venkatesh et al. (2003), who identified ease of use as a critical factor in translating technology use into actual behavior. However, this study suggests that the impact of effort expectancy might be more limited depending on how it interacts with performance expectancy, rather than its standalone importance.

Social influence was found not to have a significant impact on adoption behavior (β = 0.096, p = 0.204) (Hypothesis 7). This result partially contradicts the findings of Venkatesh (2012), indicating that after the initial formation of intention, adoption behavior may be more strongly influenced by personal motivation rather than external social factors.

Facilitating conditions were also found not to significantly influence adoption behavior (β = -0.72, p = 0.297) (Hypothesis 8). This suggests that while facilitating conditions play a significant role in shaping technology acceptance intention, their influence may diminish once the actual behavior phase is reached.

Finally, acceptance intention had a significant impact on adoption behavior (β = 0.753, p < 0.001) (Hypothesis 9). This finding aligns with Venkatesh (2012), who emphasized that intention is a key variable in translating intention into actual behavior. This study reaffirms this relationship in the context of New Media Platform Services.

5.3 Mediating Effect of acceptance intention

Hypothesis 10 proposed that acceptance intention mediates the influence of performance expectancy, effort expectancy, social influence, and facilitating conditions on adoption behavior. The results confirmed that acceptance intention served as a mediator for performance expectancy, social influence, and facilitating conditions. However, no mediating effect was observed for effort expectancy. This finding is consistent with previous studies that have explored the mediating role in technology acceptance models, and it reaffirms the significance of acceptance intention in the context of new media platform adoption.

5.4 Moderating Effects of Age and New Media Understanding

A significant moderating effect based on age was observed (Hypothesis 11), with users in their 20s showing the highest levels of adoption behavior. This finding aligns with previous research by Venkatesh et al. (2003) and Lee and Han (2019), which suggested that age moderates the relationship between acceptance intention and behavior.

New media understanding was found to strengthen the relationship between facilitating conditions and adoption behavior (Hypothesis 12). This result is consistent with the findings of Kim et al. (2018) and Wang et al. (2021), who emphasized the positive impact of understanding on the technology acceptance process. The study empirically confirms that users with higher levels of understanding are better able to leverage technological support, thus enhancing their acceptance of New Media Platform Services.

5.5 Comparison with Previous Studies

The findings of this study align with and expand on previous research in several important ways.

First, the significant influence of performance expectancy on adoption intention (β = 0.267, p = 0.005) confirmed in this study corroborates earlier findings by Venkatesh et al. (2003) and Zuiderwijk et al. (2015), who reported similar results in various technology adoption contexts. These studies emphasized that users are more likely to adopt technologies they perceive as beneficial to their performance, which is consistent with the behavior observed among librarians in this study.

Second, the negative influence of effort expectancy on adoption behavior (β = -0.161, p = 0.006) contrasts with the generally positive impact reported by Zhou et al. (2010) and Awwad & Al-Majali (2015). This discrepancy may arise from the high level of digital literacy among librarians, which likely reduces the relevance of ease of use. Additionally, the specific characteristics of New Media Platform Services, which require complex, evolving technical skills, might contribute to this difference.

Finally, the significant role of facilitating conditions in shaping adoption intention (β = 0.333, p < 0.01) is consistent with findings by Wang et al. (2021), who emphasized the importance of supportive environments in technology adoption. However, this study diverges from earlier research in its finding that facilitating conditions did not significantly impact actual usage behavior (β = -0.072, p = 0.297). This suggests that organizational support may be more crucial during the initial adoption stages rather than during the continued usage phase.

5.6 Practical and Theoretical Implications

The findings of this study offer actionable insights for public libraries and policymakers. First, the strong influence of performance expectancy emphasizes the need for libraries to effectively communicate the benefits of New Media Platform Services to staff, such as improved efficiency and enhanced user engagement. Second, the moderating effect of age suggests that training programs should be tailored to different age groups, addressing varying levels of digital literacy and readiness for adoption. Finally, the critical role of facilitating conditions highlights the importance of a robust technical infrastructure and continuous organizational support to ensure the successful implementation and sustained use of New Media Platform Services.

The study contributes to the Unified Theory of Acceptance and Use of Technology (UTAUT) by demonstrating that adoption behaviors in public library contexts are influenced by unique factors, such as professional roles and organizational settings. The negative impact of effort expectancy on actual usage behavior challenges traditional assumptions, suggesting that perceived ease of use may not always facilitate adoption, especially among digitally literate users. Additionally, the significant moderating role of new media understanding expands the theoretical framework, offering new opportunities to explore how domain-specific knowledge influences technology adoption.

In conclusion, this study explored the factors influencing the adoption behavior of New Media Platform Services in public libraries, applying the UTAUT model to analyze structural relationships. Key findings include the significant roles of performance expectancy, social influence, and facilitating conditions in shaping adoption intention, as well as the mediating effect of adoption intention on actual usage behavior