ALL Metrics
-
Views
-
Downloads
Get PDF
Get XML
Cite
Export
Track
Research Article

Antecedents of generation Z towards digitalisation. A PLS-SEM analysis

[version 1; peer review: 1 approved with reservations, 1 not approved]
PUBLISHED 24 Sep 2021
Author details Author details
OPEN PEER REVIEW
REVIEWER STATUS

This article is included in the Artificial Intelligence and Machine Learning gateway.

This article is included in the Research Synergy Foundation gateway.

Abstract

Background: The Covid-19 pandemic has forced businesses in the direction of technology development. In particular, financial institutions have started the digital transformation, embracing the usage of artificial intelligence. In this respect,  consumers’ willingness to adopt artificial intelligence in finance, appears to have relevance to current efforts to enhance the efficiency and effectiveness of the financial system. This study aims to better comprehend the antecedents towards the intention to adopt artificial intelligence in financial services among Generation Z,  with the use of the Technology Acceptance Model. 
Methods:  In this study, questionnaires were used to collect data from 150 male and female Malaysian undergraduates. Partial Least Squares-Structural Equation Modelling was employed to analyse the data. 
Results: Perceived ease of use and attitude, positively influenced the adoption of artificial intelligence in financial services. 
Conclusion: The results have suggested the improvement in user interface, information and activities to encourage generation Z to adopt artificial intelligence in financial services.

Keywords

Technology Acceptance Model (TAM), intentions to adopt artificial intelligence in financial services (INT), perceived usefulness (PU), perceived ease of use (PEoU), attitude (ATT)

Introduction

Digital transformation is seen extensively across many industries worldwide, partly triggered by prolonged lockdowns in response to the Covid-19 pandemic. According to The Star (2020) stated that artificial intelligence (AI) is gaining even more traction in driving customers experience. To a certain extent, financial institutions, for example, have begun to embark on the AI journey, embracing AI chatbot into the delivery process of banking and financial services. Likewise, massive investments have been made in AI-driven predictive analytics to improve efficiency and enhance business forecasts in areas like sales, operations, risk management, etc. Specifically, it also aims to assist its customers in financial consultancy and financial needs (New Straits Times, 2018).

In a recent survey conducted by Alam (2018), 93% of respondents from the wealth management industry emphasise the need for AI in handling many and various financial services, as a sharp increase in AI adoption in the financial industry is anticipated. While these represent crucial insights from the perspective of industry players, equally significant is whether customers are ready to embrace the use of AI in the financial industry. This is an intriguing yet often overlooked question. Thus, this study appears to be essential to investigate the antecedents influencing consumers' willingness to adopt AI in financial services. Generation Z (Gen Z), aged between 18 to 25 years as of 2020, is the respondents for this survey. Their intention about AI adoption is important to this research as Gen Z is a true digital native, as claimed by Gaidhani et al. (2019). Among the antecedents, perceived usefulness (PU), perceived ease of use (PEoU), and attitude (ATT) are hypothesised to positively influence Gen Z's intention to adopt AI-based financial services (INT).

Literature review

Technology acceptance model

Davis et al. (1989) suggested that TAM determines the antecedents of technology and system acceptability. The antecedents, namely PU, PEoU and ATT from the mentioned model, are adapted for this study to examine their influence on consumers' intention to adopt AI-based financial services. Venkatesh and Davis (2000) also pointed out that these antecedents affect each other. For example, PEoU will affect PU of using a new technology. When the particular system is easy to use, it is more valuable to the users to improve their work execution. Besides, PU will also affect ATT and individual's intention to accept the system's technology.

Intention to adopt AI-based Financial Services

Behavioural intention defines the degree to which a person performs a particular behaviour after conscious planning or self-directed motivation (Mark & Paul, 2005). According to Chuang, Liu and Kao (2016) and Cham et al. (2018), TAM is used to investigate a person's behaviour and the results show that PU, PEoU and ATT may be used to predict a person's readiness to adopt new technology.

Perceived usefulness on intention to adopt AI-based financial services

PU is a crucial factor to determine the acceptance of using AI-based financial services. PU means the extent to which an individual perceived a system that improves their job efficiency (Davis et al., 1989). Pai and Huang (2011) found that users' willingness to accept AI is related to the system's PU. Another study also demonstrates that a more positive attitude is related to a higher willingness to use AI products (Lunney et al., 2016). Likewise, Pillai and Sivathanu (2020) studied the impact of using AI-powered customer service technology in the tourism industry. The result shows that users are willing to use AI devices to plan for their trip due to the easy and interactive booking process and service.

Perceived ease of use on intention to adopt AI-based financial services

To master or use a system or technology without much effort is identified as perceived ease of use (Davis et al., 1989). Hamid et al. (2016) found that PEoU is positively related to the adoption of e-government system. It indicates that the respondents are keen on using the new system if the technology seemed easily understood. Furthermore, Patil and Mugdha (2019) found a significant positive relationship between PeoU and adoption of chatbot services in financial services. This is because the system is a simple drag-and-drop interface and hence respondents are more willing to use it.

Attitude on intention to adopt AI-based financial services

Azjen (1991) defined attitude as the degree of a user's assessment and expression of the behaviour's execution. Favourable or unfavourable attitude emerges from the personnel's beliefs on behavioural outcomes (Cordano & Frieze, 2000). The user's attitude toward AI will affect their willingness to accept AI in financial services (Schepman & Rodway, 2020). Previous studies discovered a positive relationship between attitudes and intentions to use the system (Chang, Vitell, & Lu, 2019; Le & Kieu, 2019; Prendergast & Tsang, 2019; Tweneboah-Koduah, Adams, & Nyarku, 2019).

Methods

Research framework

Research framework (Figure 1) is shown in the proposed research framework.

c35b8d8c-ca20-4c05-8403-6c821ef19ae2_figure1.gif

Figure 1. Proposed research framework.

Data collection

A set of self-administered questionnaires is distributed online via Google Form (See underlying data) (Chan et al., 2021). Purposive sampling is used in this study because the respondents must first be knowledgeable about the AI system so they can fully comprehend the questions in the survey (Tongco, 2007). It means that the target respondents have to be technology savvy and well versed with financial terminology.

The cover page of the questionnaire stated the disclaimers, and the respondents would have to agree to participate in the survey voluntarily before they respond to the questionnaire. They also had given the written consent to publish their responses. The sample size is determined via G*Power analysis. Since there are only three predictors in this study, the minimum acceptable sample size to surpass the sufficient statistical power of 0.8 is 74 responses. This study has collected 150 usable responses from undergraduate students with the age range of 18 to 25 years to represent Gen Z in Malaysia (See underlying data) (Chan et al., 2021). The Partial Least Squares Structural Equation Modelling (PLS-SEM) method was employed to analyse the collected data. The software used is Smart PLS 3.0 and SPSS version 23.

Results

Pilot test

The Cronbach's alpha test is used to determine the questionnaire's reliability. Additionally, the SPSS software is used to evaluate the reliability analysis with 30 respondents. Results shown in Table 1 indicates that Cronbach's alpha's value is greater than 0.7 for all the analysed variables, and thus, the questionnaire is reliable.

Table 1. Reliability test.

Cronbach’s alphaNumber of items (N)
PU0.7535
PEoU0.7775
ATT0.7735
INT0.8175

Descriptive analysis

Table 2 tabulates the descriptive analysis. Among 150 respondents, 44% are male, while 56% are female respondents. According to Dolot (2018), individuals born after 1995 are referred to as Gen Z. As a result, respondents are between the ages of 20 and 22 (55.3%), between 23 and 25 (22.0%), under 20 (14.0%), and above 25 (8.7%) years.

Table 2. Descriptive analysis.

FrequencyPercent (%)
GenderFemale8456.0
Male6644.0
AgeBelow 20 years2114.0
20-22 years8355.3
23-25 years3322.0
Above 25 years138.7

Assessment of measurement model

This study performs a measuring model to evaluate the elements of a path model, which contains indicators and their relationships with the constructs. The minimum threshold of 0.708, 0.70, and 0.50 are assessed on factor loadings, composite reliability and average variance, respectively. They are evaluated on the validity and reliability of the constructs (Hair et al., 2019). Hence, PU1, PU2, PEOU1, ATT1, ATT3, INT1 and INT5 were removed because their respective loadings were less than 0.708. The factor loadings, composite reliability, and average variance extracted for the final measurement model met the minimum threshold, which the values are tabulated in Table 3. The results also suggested that the measurement model was reliable and had adequate convergent validity. Subsequently, a discriminant validity assessment was performed because the content and substance of the constructs had to be identified. Therefore, this study employed the Heterotrait-Monotrait Ratio (HTMT) (Henseler et al., 2015). Firstly, the construct - the intention had failed to meet the HTMT 0.9 criteria. Upon checking with the cross-loading, INT3 was removed. Table 4 shows that all the values are lower than 0.90, which is the required threshold suggested by Gold et al. (2001). Hence, the findings shows that the proposed hypotheses are accepted and verified by the value of the discriminant validity.

Table 3. Final measurement model.

ConstructsItemsLoadingsComposite reliability (CR)Average variance extracted (AVE)
1. PUPU30.8660.8430.644
PU40.812
PU50.722
2. PEoUPEOU20.7560.8570.600
PEOU30.811
PEOU40.791
PEOU50.738
3. ATTATT20.8860.8650.683
ATT40.853
ATT50.732
4. INTINT20.8980.8660.763
INT40.848

Table 4. Discriminant validity (HTMT Assessment).

1.2.3.4.
1. Perceived Usefulness
2. Perceived Ease of Use0.79
3. Attitude0.7950.825
4. Intention0.5660.7510.89

Assessment of structural model

Table 5 presents the hypotheses statements in this study. The structural model is presented in Figure 2. These findings revealed that the Coefficient of Determination (R2) value of the model is 0.467, which suggested a moderate predictive model (Chin, 1998). It also means that 46.70% of the INT can be explained by PU, PEoU and ATT.

Table 5. Hypotheses statements.

HypothesisStatement
Hypothesis 1 (H1)Perceived usefulness (PU) has positively influenced the intention to adopt AI-based financial services (INT)
Hypothesis 2 (H2)Perceived ease of use (PEoU) has positively influenced the intention to adopt AI-based financial services (INT)
Hypothesis 3 (H3)Attitude (ATT) has positively influenced the intention to adopt AI-based financial services (INT).
c35b8d8c-ca20-4c05-8403-6c821ef19ae2_figure2.gif

Figure 2. Structural model.

Bootstrapping was applied to determine the significant level of the constructs. The path coefficient and corresponding t-values were gathered with 5,000 resamplings. As a result, PEoU positively influenced the INT, and ATT positively influenced the INT. Hence, H2 (t-value = 2.683) and H3 (t-value = 5.408) were supported. Nonetheless, PU and INT were insignificant in this study. Thus, H1 (t-value = 0.552) was not supported. Figure 2 presented the structural model, and Table 6 summarises the hypotheses testing.

Table 6. Summary of hypotheses testing.

HypothesisRelationshipt-valueDecision
H1PU -> INT0.552Not supported
H2PEOU -> INT2.683***Supported
H3ATT -> INT5.408***Supported

Discussion

These findings suggest that both PEoU and ATT are positively related to INT. In other words, H2 and H3 are supported in this study which is consistent with previous studies (Schepman & Rodway, 2020; Patil & Mugdha, 2019; Hamid et al., 2016).

Individuals will intend to use the AI-based financial services when they believe that it would eliminate obstacles such as making the right investment decision or process the handling of their financial matters much faster. (Hong et al., 2021). The Edge Markets (2017) further elaborated that AI technology allows financial service providers to engage customers more intuitively and offers real-time support, thereby transforming complicated banking processes into more simple tasks. Additionally, ATT also plays a vital role in influencing an individual's intention to use AI-based financial services. TAM model states that PEOU is one of the motivators of consumer attitude towards using new technology (Venkatesh & Davis, 2000). Individuals will develop a positive attitude when they perceive that using AI is effortless in the AI-based financial service context.

Meanwhile, PU is not significant in this study. Thus, H1 is not supported. Jahangir and Begum (2008) mentioned that it would not be enough by just offering AI-based financial services to the general public and expect it to utilised. Financial service providers must establish the conviction that AI technology can improve the efficiency, security, and ease of use in financial services.

Conclusion

The main purpose of this study was to identify the factors influencing Gen Z's intention to use AI-based financial services. The results confirm that PEoU and ATT positively influenced the INT among Gen Z in Malaysia. This indicates that Gen Z is ready for the digital transformation, and they are willing to accept and use AI-based financial services.

This study draws a few implications for financial institutions/financial service providers: Financial institutions are suggested to focus on designing a user-friendly interface that provides quick access to the standard features. The AI-based financial services would be an easy tool for the consumers to use, and the system would not require much training and technical skills.

Additionally, financial service providers should make sure that their websites portray clear, concise, and intelligible information. This helps customers understand that AI-based financial services are always simple to use.

Finally, financial institutions are encouraged to get involved in campaigning and frequent collaboration with the universities to create awareness among the Gen z community regarding the importance and ease of using AI tools in financial services. This will help Gen Z build a positive attitude towards AI-based financial services.

Ethical approval

This study is in accordance with the university's ethical standard, and ethical approval (number EA0432021) was obtained.

Data availability

Underlying data

Data archiving and networked services (DANS): Survay of Malaysian view on the roles of artificial intelegence for the future changes in the financial services industries

DOI: https://doi.org/10.17026/dans-2xr-x6wm (Chan et al., 2021).

This project contains the following underlying data:

File 1: The blank questionnaire used in this study.

File 2: Responses to the survey of the Malaysians’ view on the roles of artificial intelligence for the future changes in the financial services industry.

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

Author contributions

Kar Hoong Chan and Chiu Yu Teh presented the idea and background of this manuscript. Tuan Hock Ng developed the framework and the adapted theory. Chiu Yu Teh also collected the data for this study. Lee Ying Tay performed the data analysis and discussion of the study. All authors discussed and contributed to the final manuscript.

Comments on this article Comments (0)

Version 1
VERSION 1 PUBLISHED 24 Sep 2021
Comment
Author details Author details
Competing interests
Grant information
Copyright
Download
 
Export To
metrics
Views Downloads
F1000Research - -
PubMed Central
Data from PMC are received and updated monthly.
- -
Citations
CITE
how to cite this article
Chan KH, Ng TH, Tay LY and Teh CY. Antecedents of generation Z towards digitalisation. A PLS-SEM analysis [version 1; peer review: 1 approved with reservations, 1 not approved]. F1000Research 2021, 10:963 (https://doi.org/10.12688/f1000research.73081.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
track
receive updates on this article
Track an article to receive email alerts on any updates to this article.

Open Peer Review

Current Reviewer Status: ?
Key to Reviewer Statuses VIEW
ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested
Approved with reservations A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.
Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions
Version 1
VERSION 1
PUBLISHED 24 Sep 2021
Views
1
Cite
Reviewer Report 05 Mar 2024
Sayed Fayaz Ahmad, College of Engineering Sciences, Institute of Business Management, Karachi, Pakistan 
Approved with Reservations
VIEWS 1
Thanks a lot for giving me a chance to review the paper. Overall the work is good, yet i have some suggestions for the improvement of the paper.

1. The literature review is very short. Please make ... Continue reading
CITE
CITE
HOW TO CITE THIS REPORT
Ahmad SF. Reviewer Report For: Antecedents of generation Z towards digitalisation. A PLS-SEM analysis [version 1; peer review: 1 approved with reservations, 1 not approved]. F1000Research 2021, 10:963 (https://doi.org/10.5256/f1000research.76704.r245496)
NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article.
Views
13
Cite
Reviewer Report 20 Dec 2021
Darmawan Napitupulu, Research Center for Science, Technology and Innovation Policy and Management, Indonesian Institute of Sciences, Jakarta, Indonesia 
Not Approved
VIEWS 13
Some recommendations:
  1. The problem of this research was not clear. What is the matter with Gen Z? If they are true digital natives, they should not have a problem with adopting technology, including AI. Besides, the
... Continue reading
CITE
CITE
HOW TO CITE THIS REPORT
Napitupulu D. Reviewer Report For: Antecedents of generation Z towards digitalisation. A PLS-SEM analysis [version 1; peer review: 1 approved with reservations, 1 not approved]. F1000Research 2021, 10:963 (https://doi.org/10.5256/f1000research.76704.r100547)
NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article.

Comments on this article Comments (0)

Version 1
VERSION 1 PUBLISHED 24 Sep 2021
Comment
Alongside their report, reviewers assign a status to the article:
Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested
Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.
Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions
Sign In
If you've forgotten your password, please enter your email address below and we'll send you instructions on how to reset your password.

The email address should be the one you originally registered with F1000.

Email address not valid, please try again

You registered with F1000 via Google, so we cannot reset your password.

To sign in, please click here.

If you still need help with your Google account password, please click here.

You registered with F1000 via Facebook, so we cannot reset your password.

To sign in, please click here.

If you still need help with your Facebook account password, please click here.

Code not correct, please try again
Email us for further assistance.
Server error, please try again.