Scholarly article on topic 'Technology Acceptance and Hand Anthropometry (TAHA) Model: Insights from Somatosensory Technology'

Technology Acceptance and Hand Anthropometry (TAHA) Model: Insights from Somatosensory Technology Academic research paper on "Economics and business"

CC BY-NC-ND
0
0
Share paper
Academic journal
Procedia Manufacturing
OECD Field of science
Keywords
{"Hand anthropometry" / "Somatosensory technology" / "Technology acceptance model" / "Touchless system"}

Abstract of research paper on Economics and business, author of scientific article — Pei-Lee Teh, Pervaiz K. Ahmed, Ravindra S. Goonetilleke, Emily Yim Lee Au, Soon-Nyean Cheong, et al.

Abstract Somatosensory technology is an emerging technology used in applications such as home entertainment, medical and healthcare. It can track human hand movement and enable users to interact with digital devices or physical environment by using hand gesture(s). Effectiveness of somatosensory technology is determined by the compatibility between the technology's operational features and the anthropometric characteristics of the user's hand. Currently, only limited guidance is provided for designers and manufacturers in their development of consumer somatosensory technology products. Motivated by the shortfalls in the extant hand anthropometry literature, this study examines how hand anthropometry influences user technology acceptance. Drawing from Technology Acceptance Model (TAM), we develop Technology Acceptance and Hand Anthropometry (TAHA) model to investigate the impact of hand-size on user's somatosensory technology acceptance. A product trial experiment of 60 participants was conducted to test TAHA model. The results show that hand-size influences the relationship between perceived ease of use and behavioral intention to use somatosensory technology. Our findings have significant implications for hand anthropometry research and practice.

Academic research paper on topic "Technology Acceptance and Hand Anthropometry (TAHA) Model: Insights from Somatosensory Technology"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia Manufacturing 3 (2015) 4197 - 4204

6th International Conference on Applied Human Factors and Ergonomics (AHFE 2015) and the

Affiliated Conferences, AHFE 2015

Technology Acceptance and Hand Anthropometry (TAHA) Model: Insights from somatosensory technology

Pei-Lee Teha*, Pervaiz K. Ahmeda, Ravindra S. Goonetillekeb, Emily Yim Lee Aub,

Soon-Nyean Cheongc, Wen-Jiun Yapc

aMonash University, Jalan Lagoon Selatan, Bandar Sunway, 47500, Malaysia bHong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong cMultimedia University, Persiaran Multimedia, Cyberhaya, 63100, Malaysia

Abstract

Somatosensory technology is an emerging technology used in applications such as home entertainment, medical and healthcare. It can track human hand movement and enable users to interact with digital devices or physical environment by using hand gesture(s). Effectiveness of somatosensory technology is determined by the compatibility between the technology's operational features and the anthropometric characteristics of the user's hand. Currently, only limited guidance is provided for designers and manufacturers in their development of consumer somatosensory technology products. Motivated by the shortfalls in the extant hand anthropometry literature, this study examines how hand anthropometry influences user technology acceptance. Drawing from Technology Acceptance Model (TAM), we develop Technology Acceptance and Hand Anthropometry (TAHA) model to investigate the impact of hand-size on user's somatosensory technology acceptance. A product trial experiment of 60 participants was conducted to test TAHA model. The results show that hand-size influences the relationship between perceived ease of use and behavioral intention to use somatosensory technology. Our findings have significant implications for hand anthropometry research and practice.

© 2015 The Authors.PublishedbyElsevierB.V.This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of AHFE Conference

Keywords: Hand anthropometry; Somatosensory technology; Technology acceptance model; Touchless system

* Corresponding author. Tel.: +603-55144971; fax: +603-55146192. E-mail address: teh.pei.lee@monash.edu

2351-9789 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.Org/licenses/by-nc-nd/4.0/).

Peer-review under responsibility of AHFE Conference

doi:10.1016/j.promfg.2015.07.395

1. Introduction

There is a growing interest in the use of somatosensory technology in today's applications such as home entertainment, medical and healthcare. Somatosensory technology differs from other technology as it can track human hand movement and enable users to interact with digital devices or physical environment by using hand gesture(s). Effectiveness of somatosensory technology is determined by the compatibility between the technology's operational features and the anthropometric characteristics of the user's hand. Past research (see: [1], [2], [3], [4]) has examined hand anthropometry of different occupation groups including pianists, garment factory workers, agricultural and manufacturing employees. Some researchers (e.g., [5], [6], [7]) have studied hand anthropometry of different ethnic groups such as Hong Kong, United States, England and Japan females, Indian women and Bangladesh adults. Human factors research (see: [8], [9], [10]) has published on different hand models such as 3D-models of hand tool, optimization-based model and 3D hand gesture taxonomy. However, limited guidance is provided for the designers and manufacturers in the development of somatosensory consumer products. Motivated by the shortfalls in the extant hand anthropometry literature, this study examines how hand anthropometry influences user's somatosensory technology acceptance.

Technology Acceptance Model (TAM) has been long applied in research predicting technology usage. However, TAM research has not examined the impact of hand anthropometry on user's technology acceptance. In this study, we draw upon TAM to develop Technology Acceptance and Hand Anthropometry (TAHA) model to investigate the impact of hand-size on user's somatosensory technology acceptance.

2. Literature review

The study of anthropometry is the study of human body measurements to improve understanding human physical variations and assists in anthropological taxonomy [11]. Fransson and Winkel [12] and Blackwell et al. [13] highlight the importance of the relationship between hand anthropometry and handle size/shape in affecting individual's hand posture or grip strength. Other than a few select studies such as, hand anthropometry study on mobile phone keypad [14] and 3D hand gesture taxonomy [10], little empirical research on how hand-size influences user's somatosensory technology acceptance.

TAM posits that technology adoption behavior is determined by one's intention to use a specific technology, which is influenced by individuals' perceived ease of use and perceived usefulness of the system [15]. Typically, many researchers have restricted their study to investigate perceived ease of use and perceived usefulness in new technology acceptance research, thereby overlooking other controllable variables such as hand anthropometry and human anatomy. With increasing adoption of somatosensory technology consumer products in today's world, it is important to further our understanding of the impact of hand anthropometry on user's somatosensory technology acceptance. Given that hand-size variations exist between individuals, we expect that the relationships between perceived ease of use and behavioral intention, perceived usefulness and behavioral intention, and usability and behavioral intention, will be moderated by hand-size. Therefore, we propose:

• H1: The relationship between perceived ease of use and behavioral intention will be different depending on hand-size (i.e., small as compared to medium).

• H2: The relationship between perceived usefulness and behavioral intention will be different depending on hand-size (i.e., small as compared to medium).

• H3: The relationship between usability and behavioral intention will be different depending on hand-size (i.e., small as compared to medium).

3. Research methodology

3.1. Measures

Survey instrument used in this study is presented in Appendix A. Perceived ease of use and perceived usefulness were measured using items adapted from Davis [16], Venkatesh [17] and Chau [18]. Usability was operationalized using items modified from the instrument developed by Ryu et al. [19] and Bhuiyan and Picking [20]. Survey items of behavior intention were measured using one new item introduced in this study and two items from Venkatesh [17]. All survey items were assessed using a seven-point Likert scale from 1 (strongly disagree/very difficult) to 7 (strongly agree/very easy).

3.2. Sample

To test the hypotheses, we conducted an experiment with 60 students (30 females and 30 males) from a university in Malaysia. Participation was voluntary and anonymous. The participants read and signed an informed consent agreement prior to the experiment. The examination of hand characteristics was carried out based on the guidelines of Martin and Saller [21]. The relation between the length and the breadth of the hand is regarded as Hand index (characteristic 3). Hand length was measured by means of a ruler. All measurements were taken with the hand stretched out and laid flat, with the fingers straight. Participants with hand length below eight centimeters were categorized as small hand-size, while participants with hand length above eight centimeters were regarded as medium hand-size.

Table 1 shows the participants' demographic. Our research participants consisted of 30 males and 30 females. Seventy percent (n=42) of the participants were between 18 and 24 years old, and thirty percent (n=18) were between 25 and 34 years old. Forty-two participants were small hand-size and 28 were medium. The breakdown of technology adopter was as follows: 11.7 percent of innovators, 11.7 percent of early adopter, 33.3 percent of early majority, 23.3 percent of late majority and 20 percent of laggards.

Table 1. Participants' Demographic.

Variable Classification Frequency (n=60) Percent (%)

Gender Male 30 50

Female 30 50

Age 18-24 42 70

25-34 18 30

Hand-size Small 32 53.3

Medium 28 46.7

Consumer Types Laggard 12 20

Late Majority 14 23.3

Early Majority 20 33.3

Early Adopter 7 11.7

Innovator 7 11.7

3.3. Data collection

The somatosensory technology/touchless system tested in this study was an in-house developed system, enabling users to interact with five home applications using hand gestures. The five home applications include accessing living room lights, movie on demand on the television, edutainment (games), kitchen e-cook book and digital photo albums. Each participants was facilitated by a research assistant who demonstrated the hand gestures (i.e., point-and-wait and palm-and-close) to generate commands for all five home applications. All participants were given five minutes for self-practice. To reduce fatigue, participants were given another two minutes rest time before running

the test session. During the test session, each participant was required to complete five tasks involving the home applications using their hand gestures. After completing the experiment, each participant was asked to answer a survey questionnaire.

4. Results

4.1. Reliability, validity and factor analyses

Data analysis was performed using IBM SPSS. Reliability and validity of the scales were assessed using Cronbach's alpha (CA), composite reliability (CR) and the average variance extracted (AVE). As presented in Tables 2 and 3, all CA and CR values were greater than 0.80, meeting the value suggested by Nunnally and Bernstein [22] and Bagozzi and Yi [23]. For convergent validity (see Tables 2 and 3), the values of AVE for all variables were greater than the minimum value of 0.50 recommended by Fornell and Larcker [24]. Discriminant validity was established as all the square roots of AVE values were greater than the off diagonal values in the correlation table.

Factor analysis was conducted to check the construct validity. As presented in Tables 4 and 5, each items had good factor loadings and the values of Kaiser-Meyer-Olkin (KMO) were above 0.50. The numbers for the Bartlett test of sphericity were significant for all variables, with values ranging from 41.735 (UsabilityModel 2) to 140.703 (Behavior intention Model 2). All variables had eigenvalues of greater than 1. As a result, all variables (i.e., perceived ease of use, perceived usefulness, usability and behavioral intention) in Models 1 and 2 were significant to be analyzed in this study.

Table 2. Results of reliability and validity for Model 1. Bold values in the diagonal row are square roots of the AVE.

Perceived Ease of Use Perceived of Usefulness Usability Behavioral Intention

Perceived Ease of Use 0.789

Perceived of Usefulness 0.686** 0.869

Usability 0.621** 0.276 0.926

Behavioral Intention 0.771** 0.779** 0.305 0.938

Mean 4.771 4.789 4.063 4.219

Standard Deviation 1.020 1.145 1.294 1.243

Cronbach's Alpha 0.876 0.891 0.914 0.932

Composite Reliability 0.908 0.925 0.948 0.957

Average Variance Extracted 0.623 0.755 0.858 0.880

Table 3. Results of reliability and validity for Model 2. Bold values in the diagonal row are square roots of the AVE.

Perceived Ease of Use Perceived of Usefulness Usability Behavioral Intention

Perceived Ease of Use 0.825

Perceived of Usefulness 0.741** 0.839

Usability 0.418* 0.453* 0.882

Behavioral Intention 0.586** 0.750** 0.232 0.987

Mean 4.286 4.429 3.833 4.071

Standard Deviation 1.280 1.591 1.353 1.844

Cronbach's Alpha 0.904 0.859 0.851 0.987

Composite Reliability 0.927 0.905 0.913 0.992

Average Variance Extracted 0.680 0.704 0.778 0.975

Table 4. Results of factor analysis for Model 1.

No. of Items KMO BTS EV Factor Loadings

Item1 Item2 Item3 Item4 Item5 Item6

Perceived Ease of Use 6 0.766 109.380*** 3.741 0.873 0.818 0.708 0.693 0.809 0.820

Perceived of Usefulness 4 0.695 90.586*** 3.020 0.946 0.865 0.837 0.823 Nil Nil

Usability 3 0.747 64.627*** 2.573 0.938 0.937 0.903 Nil Nil Nil

Behavioral Intention 3 0.749 75.283*** 2.640 0.952 0.948 0.914 Nil Nil Nil

Note: ***p < 0.001; KMO=Kaiser-Meyer-Olkin; BTS=Barlett's Test of Sphericity; and EV=Eigen-values.

Table 5. Results of factor analysis for Model 2.

No. of Items KMO BTS EV Factor Loadings

Item1 Item2 Item3 Item4 Item5 Item6

Perceived Ease of Use 6 0.853 97.664*** 4.082 0.787 0.787 0.907 0.886 0.776 0.795

Perceived of Usefulness 4 0.581 108.299*** 2.817 0.881 0.848 0.771 0.853 Nil Nil

Usability 3 0.654 41.735*** 2.333 0.901 0.938 0.801 Nil Nil Nil

Behavioral Intention 3 0.788 140.703*** 2.927 0.986 0.987 0.990 Nil Nil Nil

Note: *** p < 0.001; KMO=Kaiser-Meyer-Olkin; BTS=Barlett's Test of Sphericity; and EV=Eigen-values.

4.2. Hypotheses testing

Stepwise regression analysis was performed to test Models 1 and 2. Tables 6, 7 and Figure 1 show the results of regression analysis. The predictors explained 71.3 percent and 56.2 percent of behavioral intention's variance in Models 1 and 2. For small hand-size participants, intention to use the touchless system was determined by perceived usefulness ($/=0.473; p-value<0.01) and perceived ease of use (/^=0.447; p-value<0.01). For medium hand-size participants, perceived usefulness (/2=0.750; p-value<0.001) was the dominant factor predicting individuals' behavioral intention. Our findings showed non-significant usability-intention relationships (/1=-0.179; p-value>0.05; /2=-0.135; p-value>0.05) in Models 1 and 2. These results lend support to hypothesis H1 but not H2 and H3.

Table 6. Summary of the Models.

Model R2 Adjusted R2 F-statistic p-value

1 0.713a 0.694a 36.096 0.000***

2 0.562a 0.545a 33.332 0.000***

Note: *** p < 0.001; aDependent Variable=Behavioural Intention; Model 1=Small Hand-size; Model

2=Medium Hand-size.

Table 7. Regression Analysis for Small Hand-size as Compared to Medium Hand-size.

Model Non- standardized coefficients Standardized coefficients t-value Sig.

1 Predictor variables B SE B

(Constant) -0.840 0.613 -1.369 0.182

Perceived Usefulness 0.513 0.148 0.473 3.463 0.002**

Perceived Ease of Use 0.545 0.166 B in 0.447 3.274 0.003**

Eliminated variables

Usability -0.179 -1.387 0.176

Dependent variable: Behavioral Intention

2 Predictor variables

(Constant) 0.224 0.707 0.316 0.754

Perceived Usefulness 0.869 0.150 B in 0.750 5.773 0.000***

Eliminated variables

Perceived Ease of Use 0.068 0.343 0.734

Usability -0.135 -0.922 0.365

Dependent variable: Behavioral Intention

Note: ** p < 0.01; *** p < 0.001; 5=Beta Coefficient; SE=Standard Error; Model 1=Small Hand-size; Model 2=Medium Hand-size.

(a) (b)

Fig. 1. (a) Results of Model 1; (b) Results of Model 2.

5. Discussion

The purpose of this study is to examine how hand anthropometry influences user's somatosensory technology acceptance. We found a significant difference in the relationship between perceived ease of use and behavioral intention in small hand-size as compared to medium hand-size. Perceived ease of use was a significant predictor of an individual's behavioral intention to use the somatosensory technology for small hand-size, while its relationship to usage intention was insignificant for medium hand-size. Our findings indicate that as the direct interaction between human and technology takes place human anthropometry/human anatomy will come to feature more and more as a key feature in the technology adoption decision. For individuals with small hand-size, they faced incompatibility to technology (i.e., lack of sensitivity or inability of the technology to read the signal) leads

frustration to fatigue. When people with small hand-size are unable to access the digital appliance using hand gesture, they can break into frustration and fatigue.

Notably, for both small and medium hand-size, perceived usefulness was found to be a significant predictor of user's behavior intention to use the somatosensory technology. This indicates that once users begin to use a somatosensory technology, usefulness appears to be more significant overall in predicting behavioral intention.

In this study, usability was found to be non-significant in both small and medium hand-size. A possible explanation for this result is somatosensory technology is an enabler of smart home applications. Therefore, a user does not experience usability of somatosensory technology itself, but usability relates to what is being consumed (e.g., browsing digital photo album and playing edutainment).

6. Conclusion

Our study has important implications for theory and practice. From the theoretical aspect, this study expands our understanding of factors influencing somatosensory technology acceptance. Over the years, TAM research has led to numerous replications. However, research has not focused on understanding the influence of hand anthropometry on somatosensory technology acceptance. Given that somatosensory technology is operated by hand gesture, individuals cannot be considered as a homogeneous group with regard to somatosensory technology adoption. This study proposed and tested the TAHA Model, which presents an exposition of how hand anthropometry (i.e., hand-size) influences perceived ease of use, and thereby provides researchers and practitioners with a wider understanding of the dynamics underlying the change of perceived ease of use of somatosensory technology.

From a practical viewpoint, this study suggests hand anthropometry should be considered as one of the factors in designing applications for somatosensory technology. Technology system designers and developers need to formulate anthropometric guides with higher latitude to accommodate different hand anthropometry in the ergonomic design of somatosensory technology and applications.

This study has two research limitations. First, the subjects recruited in this study were Asians, and the results could differ with different ethnic groups such as Caucasians. In future, our model should be replicated across different ethics groups. Second, cross-sectional data was used in this study. Future work should conduct longitudinal analysis in order to strengthen the direction of causality.

Acknowledgements

The authors would like to thank Chloe Lu, Koh Jet Yang, Shama Mathew, Martina Rachel Heinecke, Rory Evan Mcdade, Foundation Umaa Kundiona, Tey Eng Xin, Andy Chua Jia Luen and Chong Chin Jou in facilitating the data collection. Our sincere thanks go to the participants for sharing their insights on this project.

Appendix A. Survey Instrument

1. Perceived ease of use (1=strongly disagree; 7=strongly agree)

Item 1: Learning to operate the touchless system will be easy for the users.

Item 2: User will find it easy to get the touchless system to do what they want it to do.

Item 3: Users' interaction with the touchless system will be clear and understandable.

Item 4: User will find the touchless system to be flexible to interact with.

Item 5: It will be easy for users to become skillful at using the touchless system.

Item 6: User will find the touchless system easy to use.

2. Perceived usefulness (1=strongly disagree; 7=strongly agree)

Item 1: Using the touchless system would enable users to access home edutainment more quickly. Item 2: Using the touchless system would enable users to access home control system more quickly. Item 3: Using the touchless system would improve users' life.

Item 4: Users would find the touchless system useful in their life.

3. Usability (1=very difficult; 7=very easy)

Based on your experience in using the touchless system, what do you think about your experience with respect to:

Item 1: Accurate pointing

Item 2: Target selection

Item 3: Gesture-based interaction

4. Behavioral intention (1=strongly disagree; 7=strongly agree) If the touchless system becomes available:

Item 1: I intend to use a touchless system at home.

Item 2: It is likely that the touchless system will be the medium I use at home. Item 3: I predict I would use a touchless system at home.

References

[1] C.H. Wagner, The Pianist's Hand: Anthropometry and Biomechanics, Ergonomics. 31 (1988) 97-131.

[2] J.E. Fernandez, D.E. Malzahn, O.K. Eyada, C.H. Kim, Anthropometry of Korean Female Industrial Workers, Ergonomics. 32 (1989) 491-

[3] W.S. Marras, J.Y. Kim, Anthropometry of Industrial Populations, Ergonomics. 36 (1993) 371-378.

[4] H. Hsiao, D. Long, K. Snyder, Anthropometric Differences among Occupational Groups, Ergonomics. 45 (2002) 136-152.

[5] A.J. Courtney, Hand Anthropometry of Hong Kong Chinese Females Compared to Other Ethnic Groups, Ergonomics. 27 (1984) 1169-1180.

[6] A. Nag, P.K. Nag, H. Desai, Hand Anthropometry of Indian Women, Indian Journal of Medical Research. 117 (2003) 260-269.

[7] S.N. Imrhan, M.D. Sarder, N. Mandahawi, Hand Anthropometry in Bangladeshis Living in America and Comparisons with Other

Populations, Ergonomics. 52 (2009) 987-998.

[8] S.W. Lee, X. Zhang, Development and Evaluation of an Optimization-based Model for Power-grip Posture Prediction, Journal of

Biomechanics. 38 (2005) 1591-1597.

[9] M.S. Rogers, A.B. Barr, B. Kasemsontitum, D.M. Rempel, A Three-dimensional Anthropometric Solid Model of the Hand based on

Landmark Measurements, Ergonomics. 51 (2008) 511-526.

[10] E. Choi, H. Kim, M.K. Chung, A taxonomy and notation method for three-dimnsional hand gestures, International Journal of Industrial Ergonomics. 44 (2014) 171-188.

[11] C. Adams, Anthropometry, 2015. [Online] http://ergonomics.about.com/od/glossary/g7anthropometry.htm (Accessed on 30 March 2015)

[12] C. Fransson, J. Winkel, Hand Strength: The Influence of Grip Span and Grip Type, Ergonomics. 34 (1991) 881-892.

[13] J.R. Blackwell, K.W. Kornatz, E.M. Heath, Effect of grip span on maximal grip force and fatigue of flexor digitorum superficialis, Applied Ergonomics. 30 (1999) 401-405.

[14] V. Balakrishnan, P.H.P. Yeow, A Study of the Effect of Thumb Sizes on Mobile Phone Texting Satisfaction, Journal of Usability Studies. 3 (2008) 118-128.

[15] W. Hong, J.Y.L. Thong, W.M. Wong, K.Y. Tam, Determinants of User Acceptance of Digital Libraries: An Empirical Examination of Individual Differences and System Characteristics, Journal of Management Information Systems. 18 (2001) 97-124.

[16] F.D. Davis, Perceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology, MIS Quarterly. 13 (1989) 319340.

[17] V. Venkatesh, Determinants of Perceived Ease of Use: Integrating Control, Intrinsic Motivation, and Emotion into the Technology Acceptance Model, Information Systems Research. 11 (2000) 342-365.

[18] P.Y.K. Chau, An Empirical Assessment of A Modified Technology Acceptance Model, Journal of Management Information Systems. 13 (1996) 185-204.

[19] Y.S. Ryu, D.H. Koh, D. Ryu, D. Um, Usability Evaluation of Touchless Mouse Based on Infrared Proximity Sensing, Journal of Usability Studies. 7 (2011) 31-39.

[20] M. Bhuiyan, R. Picking, A Gesture Controlled User Interface for Inclusive Design and Evaluative Study of Its Usability, Journal of Software Engineering and Applications. 4 (2011) 513-521.

[21] R. Martin, K. Saller, Lehrbuch der anthropologie. Gustav Fischer Verlag, Stuttgart, 1957.

[22] J.C. Nunnally, I.H. Bernstein, Psychometric Theory, 3rd ed. United States of America, McGraw-Hill Inc, 1994.

[23] R. Bagozzi, Y. Yi, On the Evaluation of Structural Equation Models, Journal of the Academy of Marketing Science, 16 (1988) 74-94.

[24] C. Fornell, D.F. Larcker, Evaluating Structural Equation Models with Unobservable Variables and Measurement Error, Journal of Marketing Research, 18 (1981) 39-50.