=Paper=
{{Paper
|id=Vol-2216/healthRecSys18_paper_11
|storemode=property
|title=What Drives the Perceived Credibility of Health Apps: Classical or Expressive Aesthetics?
|pdfUrl=https://ceur-ws.org/Vol-2216/healthRecSys18_paper_11.pdf
|volume=Vol-2216
|authors=Kiemute Oyibo,Ifeoma Adaji,Julita Vassileva
|dblpUrl=https://dblp.org/rec/conf/recsys/OyiboAV18
}}
==What Drives the Perceived Credibility of Health Apps: Classical or Expressive Aesthetics?==
https://ceur-ws.org/Vol-2216/healthRecSys18_paper_11.pdf
What Drives the Perceived Credibility
of Health Apps: Classical or Expressive
Aesthetics?
Kiemute Oyibo Ifeoma Adaji Julita Vassileva
University of Saskatchewan University of Saskatchewan University of Saskatchewan
S7N 5C9, Saskatoon, Canada S7N 5C9, Saskatoon, Canada S7N 5C9, Saskatoon, Canada
kiemute.oyibo@usask.ca ifeoma.adaji@usask.ca jiv@cs.usask.ca
ABSTRACT
1 INTRODUCTION
Research has shown that visual aesthetics is a key determinant of
the perceived credibility of human-computer-interaction systems. Credibility is an important design attribute that determines the
However, there is limited research on which of the two dimen- adoption and usage of persuasive systems that are designed to
sions of visual aesthetics (classical and expressive) has a stronger promote behavior change [1], [2]. Generally, in human-computer
impact on the perceived credibility of persuasive health apps. Con- interaction (HCI) systems design, research [3]–[5] has shown that
sequently, we conducted an empirical study among 669 subjects visual aesthetics plays a very important role in the judgment of
to investigate: (1) which of the two dimensions has a stronger in- credibility. While there has been a substantial amount of research
fluence on the perceived credibility of fitness apps modeling exer- on the influence of visual aesthetics as a higher-order construct on
cise behavior; and (2) the moderating effect of gender-based per- (perceived) credibility (e.g., [3]–[7]), in the health domain, research
sonalization. Our results show that, regardless of gender-based investigating which of its two key dimensions (classical and ex-
personalization, it is the perception of classical aesthetics that de- pressive) has a stronger impact on credibility is scarce. Uncovering
termines the judgment of the credibility of a persuasive health app. the more important dimension in the relationship between visual
Expressive aesthetics has no significant influence on perceived cred- aesthetics and perceived credibility will help designers know which
ibility. Our findings underscore the need for designers of persua- to focus more on when designing persuasive systems in the health
sive systems in the health domain to focus more on the classical domain, which users can trust and use. As Robins and Holmes ob-
dimension of aesthetics (orderliness, clarity and simplicity) when served, “If a web site is not perceived as credible, it is unlikely it will
designing health applications for behavior change in order to en- be used” (p386) [7].
hance their perceived credibility. Consequently, we carried out an empirical study in the fitness
domain among 669 participants, using exercise behavior modeling
CCS CONCEPTS as a case study. Behavior modeling is a persuasive strategy used
in persuasive systems (such as fitness apps) to motivate behavior
• Human-centered computing → Human computer Interaction
change. In behavior modeling, “an expert shows [a] person how to
→ HCI design and evaluation methods → User models
correctly perform a behavior, for example, in class or on video” (p.
382) [8]. The expert, also known as a social agent or coach, can be
KEYWORDS
real or virtual. However, in computerized systems, for users to
Classical aesthetics; expressive aesthetics, credibility; fitness app; imitate the modeled behavior demonstrated by the virtual coach,
persuasive technology; gender, personalization; path model there is need for the users to have trust in its visual design. This
has become important due to the fact that a badly modeled exer-
ACM Reference format: cise behavior can lead to a body injury when the user tries to rep-
licate it. Thus, the behavior models have to be designed in a such
K. Oyibo, I. Adaji, and J. Vassileva. 2018. What Drives the Per-
a way that enhances their perceived credibility.
ceived Credibility of Health Apps: Classical or Expressive Aes-
Based on our structural equation model (SEM), our results
thetics? In Proceedings of the Third International Workshop on
show that it is solely the perception of the classical dimension of
Health Recommender Systems co-located with Twelfth ACM Confer-
aesthetics that determines the perceived credibility of the health
ence on Recommender Systems (HealthRecSys’18), Vancouver, BC,
app. Expressive aesthetics has no significant influence on perceived
Canada, October 6, 2018, 6 pages.
credibility. Our finding underscores the need for designers of per-
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada.
suasive health systems, such as behavior models in fitness apps,
© 2018 Copyright for the individual papers remains with the authors. Copying to focus more on classical aesthetics (cleanness, orderliness, clarity
permitted for private and academic purposes. This volume is published and copy- and pleasantness) than expressive aesthetics as a means to increase
righted by its editors. the perceived credibility of their persuasive applications.
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada K. Oyibo et al.
The rest of this paper is organized as follows. Section 2 focuses Oyibo and Vassileva [4], [5] investigated which of perceived
on background and related work. Section 3 focuses on the re- aesthetics and perceived usability has a stronger effect on perceived
search method. Section 4 focuses on the result. Section 5 focuses credibility in the tourism domain. They found that perceived aes-
on the discussion. Finally, Section 6 focuses on the conclusion. thetics has a stronger effect than perceived usability on perceived
credibility. Furthermore, Oyibo et al. [6] investigated which of the
2 BACKGROUND AND RELATED WORK two dimensions of perceived aesthetics has a stronger effect on per-
This section provides a brief overview of the concepts of aesthetics ceived credibility in the same domain. They found that, irrespec-
and credibility, their relationships, and a review of related work. tive of age and gender, the classical dimension has a stronger ef-
fect than the expressive dimension on perceived credibility.
2.1 Aesthetics However, they are yet to be studies in the health domain fo-
cused on uncovering the more influential dimension of visual aes-
Aesthetics refers to sensory pleasure and delight derived from the
thetics on the perceived credibility of persuasive systems. The re-
perception of an object or artifact. It can be either a property of
viewed studies in the first paragraph did not address this research
an object or perceived, or both [9]. Specifically, in this paper, we
question. Moreover, Oyibo et al.’s [6] findings in the second par-
are concerned with the subjective judgment of aesthetics, which is
agraph are in the tourism domain, which may not generalize to
in the eye of the beholder (i.e., the user). In HCI, visual aesthetics—
the health domain. Thus, the main thrust of this paper is to inves-
defined as the pleasing appearance of HCI artifacts [10], user in-
tigate which of the two dimensions of visual aesthetics has a
terfaces and systems—is regarded as a multi-dimensional con-
stronger influence on perceived credibility. The secondary objec-
struct composed of classical aesthetics and expressive aesthetics
tive of this paper is to investigate whether the previous finding in
[11]. Classical aesthetics entails the historic notion of beauty: “sim-
the tourism domain (classical aesthetics has a stronger influence
plicity,” “orderliness,” proportion, “symmetry,” etc. These attributes
than expressive aesthetics on perceived credibility) generalizes to
of classical aesthetics highly correlate with perceived usability [11],
the health domain.
[12]. Thus, it is described by words such as “well-organized,”
“clear,” “clean,” etc. On the other hand, expressive aesthetics has to
3 METHOD
do with the expressive power of the HCI system designer, which
reflects his/her creativity and originality. Thus, it is described by This section focuses on our research question, research model,
words such as “original,” “creative,” “fascinating,” “sophisticated,” measurement scales and participants’ demographics.
etc. [11].
3.1 Research Objective
2.2 Credibility The aim of this study is to answer the following research ques-
Credibility refers to the believability of a HCI system [1]. Just like tions on the relationship between the two aesthetics dimensions
the subjective notion of aesthetics, credibility is perceived by the and perceived credibility in the health domain:
user and thus is not a property of the system. According to Fogg 1. What drives the perceived credibility of fitness apps fea-
[13], the perceived credibility of a HCI system is based on the sim- turing exercise behavior models: classical aesthetics or
ultaneous judgment of the perceived trustworthiness of the system expressive aesthetics, or both?
and the perceived expertise of the designer of the system. 2. Are the interrelationships among all three design con-
structs moderated by the gender of the behavior model
2.3 Aesthetics and Credibility Relationship and the gender of the observer?
A substantial amount of research has been carried out with re- 3. Does the finding that “classical aesthetics has a stronger
spect to the relationship between perceived aesthetics and per- influence than expressive aesthetics on perceived credi-
ceived credibility in the evaluation of HCI systems. Robins and bility in the tourism domain” [6] generalize to the health
Holmes [7] examined the link between the perceived aesthetics and domain?
the perceived credibility of websites. They found that a high-aes-
thetic website is more likely to be perceived credible than a low- 3.2 Research Design
aesthetic website. Alsudani and Casey [14] conducted a study on To address our research questions, we designed a fitness app pro-
the judgment of the credibility of recruitment agency websites. totype featuring exercise behavior models. We considered race
They found that aesthetics characteristics such as unity in design, (black/white), gender (male/female) and exercise-type (push-
balance and harmony significantly influence perceived credibility. up/squat) in the design of the behavior models (Figure 1).1 This
Fogg et al. [15] conducted a large scale study on the judgment of resulted in eight versions. Push-up and squat were chosen because
web credibility. They found that about 46% of participants’ com- they are among the most commonly featured exercise-types in the
ments on the credibility of various websites cutting across differ- fitness apps market. We randomly presented one of the eight de-
ent domains were about design look—which is related to visual aes- signs to each participant and asked them to answer questions on
thetics [4]. classical and expressive aesthetics, and perceived credibility.
1 In this paper, we are specifically investigating the moderating effect of the gender
of the behavior model as well as the gender of the observer.
What Drives the Perceived Credibility of Health Apps: Classical
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada
or Expressive Aesthetics?
an exploratory approach to investigate which of the two aesthetics
dimensions has a stronger influence on perceived credibility. We
also used an exploratory approach to investigate whether the
three interrelationships in Figure 2 are moderated by the gender
of the behavior model and the gender of the observer.
Figure 2. Research model
3.5 Participants
Figure 1. Behavior models performing push-up/squat [16]
The survey was approved by the Behavioral Research Ethics
3.3 Measurement Instruments Board of the University of Saskatchewan. Thereafter, it was
posted on Amazon Mechanical Turk to recruit participants resi-
To measure the three constructs of interest, we used existing val- dent in Canada and United States. Each of the participants who
idated instruments. For classical and expressive aesthetics, we used completed the survey was compensated with $0.6 for their time.
the scale proposed by Lavie and Tractinsky [11], as adapted by Table 2 shows the demographics of the valid participants after
van Schaik and Ling [17]. For credibility, we used a single-scale cleaning: 48.9% females and 51.1% males.
item scale, which research has shown is as reliable as multi-item
scales [18]. All three scales ranged from “Strongly Disagree (1)” to Table 2: demographics of participants (n = 669)
“Strongly Agree (7).” Table 1 shows their respective items. In the Criterion (Female, Male) = (327, 342)
questionnaire, the overarching question preceding all of the items 18-24 (56, 70); 25-34 (139, 157); 35-34 (79, 76);
(randomized) is, “Please rate the visual design above on the follow- Age
45-54 (38, 22); 54+ (15, 17)
ing criteria based on your first impression.” Technical/Trade School (47, 39); High School
Education (66, 70); BSc (154, 162); MSc (42, 54); PhD (9,
Table 1: Visual design construct’s scales and items 6); Others (9, 11)
Measure Items in each Scale Country Canada (104, 111); United States (194, 183);
(1) The design is visual. of Origin Others (26, 51)
Classical Aesthetics
(2) The design is clean. Years on 0-3 (2, 2); 4-6 (18, 13); 7-9 (20, 40); 10+ (287,
(3) The design is pleasant. the Internet 287)
(1) The design is creative.
Expressive Aesthetics (2) The design is sophisticated. Behavior
Male behavior models (177, 175);
(3) The design is fascinating. Model
The design is credible Female behavior models (150, 167)
Perceived Credibility Distribution
Please kindly tell us the impression the vis-
Participant’s Comment
ual design above had on you.
4 RESULT
3.4 Research Model and Hypotheses In this section, we present our measurement models, structural
models, and multigroup analysis results obtained from the SEM
Based on prior literature, we formulated three hypotheses, which analysis using “PLSPM” package in R [19].
are represented in the research model shown in Figure 2. All three
hypotheses (H1, H2 and H3) are based on the findings by Oyibo et 4.1 Measurement Models
al. [6] in the tourism domain. Specifically, they found that classi-
cal aesthetics has a stronger influence than expressive aesthetics on We evaluated the inner models to ensure the reliability of our
perceived credibility. However, in the health domain, we adopted structural models [20]. Specifically, the indicator reliability metric
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada K. Oyibo et al.
(outer loadings) and internal consistency reliability metric
(DG.rho) were greater than 0.7. Similarly, the convergent validity
(Average Variance Extracted) was greater than 0.5. Finally, with
respect to discriminant validity, no construct’s indicator loaded
higher on any other construct than itself.
4.2 Structural Models
Figure 3 shows the global models for the male and female observ-
ers. The goodness of fit (GOF) represents how well the model fits
the data, which is moderate: 56% for males and 66% for females.
The coefficient of determination (R2) represents the amount of
variance of perceived credibility explained by the model: 43% for
males and 56% for females—also moderate. The path coefficient
(β) represents the direct effect of one construct on another. Over-
all, the direct effect of classical aesthetics on perceived credibility is
significant for males (β = 0.58, p < 0.001) and for females (β = 0.65,
p < 0.001). However, the direct effect of expressive aesthetics on
perceived credibility is not significant for both males (β = 0.12, p =
n.s) and females (β = 0.13, p = n.s). Moreover, we built subgroup
models based on the following criteria (dyads): Figure 4. Path models based on gender of observer (left) and
gender of behavior model (right) – M = Male, F = Female
1. Males evaluating male behavior models (MM)
2. Males evaluating female behavior models (MF) 4.3 Multigroup Analysis
3. Females evaluating female behavior models (FF) To uncover how the gender of the observer and that of the behav-
4. Females evaluating male behavior models (FM) ior model moderates the interrelationships among the three de-
sign constructs, we conducted a multigroup analysis (MGA) as
MM and FF represent gender-based personalization, while MF and shown in Table 3.
FM represent contra-tailoring. As shown in Figure 4, the βs, GOFs
and R2s, to a large extent, confirm the findings at the global level. Table 3: Multigroup analysis based on gender
For example, in the global model, we see that the relationship be- Relationship M F p
tween classical aesthetics and perceived credibility is significant (p Classical Aesthetics→Credibility .58 .65 n.s
< 0.001), while that between expressive aesthetics and perceived Expressive Aesthetics→Credibility .12 .13 n.s
credibility is not significant. This is perfectly confirmed at the sub- Classical Aesthetics→Expressive Aesthetics .63 .74 n.s
group level, with the relationship between classical aesthetics and
perceived credibility for FM being the strongest (β = 0.82, p < 0.001) Relationship MM MF p
and that for MM being the weakest (β = 0.46, p < 0.001).
Classical Aesthetics→Credibility .46 .67 .05
Expressive Aesthetics→Credibility .15 .12 n.s
Classical Aesthetics→Expressive Aesthetics .65 .65 n.s
Relationship FF FM p
Classical Aesthetics→Credibility .56 .82 n.s
Expressive Aesthetics→Credibility .20 .00 n.s
Classical Aesthetics→Expressive Aesthetics .75 .79 n.s
Relationship MM FF p
Classical Aesthetics→Credibility .46 .56 n.s
Expressive Aesthetics→Credibility .15 .20 n.s
Classical Aesthetics→Expressive Aesthetics .65 .75 n.s
Relationship MF FM p
Classical Aesthetics→Credibility .67 .82 n.s
Expressive Aesthetics→Credibility .12 .00 n.s
Figure 3. Global path model for male and female observers Classical Aesthetics→Expressive Aesthetics .65 .79 .05
What Drives the Perceived Credibility of Health Apps: Classical
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada
or Expressive Aesthetics?
The MGA for male vs. female observers shows that that there is expected (external) benefit is the favorable outcomes in engaging
no significant difference between both genders of observers in in physical activity, which could be physical or social. On the con-
their evaluation of the behavior models. However, at the level of trary, hedonic systems provide entertainment or enjoyment to the
personalization, the MGA for MM vs. MF shows that there is a users. In other words, they satisfy the intrinsic needs of the user.
significant difference between both groups in the relationship be- This explains why expressive aesthetics (manifested in fascinating
tween classical aesthetics and perceived credibility (p < 0.05). More- design and use of special effects [11]) does not influence perceived
over, there is a significant difference between MF and FM with credibility. However, classical aesthetics (which is related to the
respect to the relationship between classical aesthetics and expres- traditional notion of usability such as orderliness, cleanness, clar-
sive aesthetics (p < 0.05). ity, etc. [11]) does influence perceived credibility, regardless of the
gender of the observer and behavior model. Table 4 shows a cross-
5 DISCUSSION section of participants’ comments, which reflect their concern
We presented a path model to investigate: (1) which of the two with the classical dimension of aesthetics in the evaluation of fit-
dimensions of visual aesthetics has a stronger influence on the per- ness apps. For example, P212 commented that s/he is satisfied with
ceived credibility of the visual design of behavior models aimed at the squat behavior model design because it is simple and clear
changing behavior; and (2) the moderating effect of gender-based (nothing distracting) and easy to follow.
personalization. In this section, we discuss the validations of two
of our main hypotheses and their implications, our contributions Table 4: participants’ comments on classical aesthetics
and the limitations of our study. Participants’ Comment
5.1 Validation of Hypotheses “The visual design is very simple and clear. The steps are
1 easy to follow and the image of the person is a great illus-
Our path analysis (Figures 3 and 4) shows that, regardless of the
tration of how to properly do the squat. There's nothing dis-
gender of the observer, two of our three hypotheses are sup- tracting on the screen and the steps are simple” [P212,
ported: classical aesthetics has a significant influence on perceived Squat].
credibility (H1); (2) classical aesthetics has a significant influence
“I like the visual design. It's uncluttered and simple. It does
on expressive aesthetics (H3); but (3) expressive aesthetics has no
a good job demonstrating the exercise in an easy to follow
significant influence on perceived credibility (H2). The validation way and I like that it also illustrates the muscle groups be-
of H1 suggests that users rely on the classical dimension—and not 2 ing used. Although the visuals are nothing over-the top or
the expressive dimension—of visual aesthetics in the evaluation of groundbreaking, they are effective at accomplishing their
the perceived credibility of behavior model design. This finding goal of demonstrating an exercise” [P476, Push-Up].
replicates that of Oyibo et al. [6] in the tourism domain, in which “The design is simple and i like that. The instructions are as
the authors found that the classical dimension of aesthetics has a well and the figure's movements are clear and easy to fol-
3
stronger influence than the expressive dimension on perceived low” [P30, Squat].
credibility. The confirmation of H1 in the health domain suggests
that visual aesthetics attributes, such as cleanness, clarity and 4 “I think that it is cleanly and clearly designed. I think that
it is easy to follow and understand” [P403, Push-Up].
pleasantness (encompassed in simplicity and orderliness), have
more impact on the perceived credibility of the behavior model de- 5 “I really like the design of the app, I think it's clear and easy
sign than complex attributes (e.g., fascination, sophistication, cre- to follow…” [P263, Squat].
ativity, etc.). This is well reflected in the following participant’s
comment: “It's really clear, it is very credible because it's an actual 5.2 Findings based on Multigroup Analysis
exercise and it shows which muscles are working. It's not creative With respect to the personalization of the gender of the behavior
because it just shows a woman doing squats” [P116, squat]. Despite models to the user (Table 3), the significant difference we found
commenting that the squat behavior model design is not creative between the group that evaluated behavior models of the same
(low in expressive aesthetics), participant 116 still considered the gender and the group that evaluated behavior models of the op-
design “very credible” owing to its perceived clarity (classical aes- posite gender is with respect to MM vs. MF. The effect of classical
thetics). Another comment that reflects participants’ judgment of aesthetics on perceived credibility is significantly stronger for the
credibility based on mainly classical aesthetics is: “The squat video opposite-gender MF dyad (β = 0.67, p < 0.001) than for the same-
is self explanatory and clean. Perfect for my needs” [P96, squat]. gender MM dyad (β = 0.46, p < 0.001). This finding seems to be
A possible explanation for why classical aesthetics has a replicated in the MGA for FF (β = 0.56, p < 0.001) vs. FM (β = 0.82,
stronger influence on perceived credibility than expressive aesthet- p < 0.001)—only that the numerical difference between both
ics does is that fitness apps modeling behavior are more of utili- groups is not significant. Despite the difference between FF and
tarian than hedonic systems. Utilitarian systems provide instru- FM is not statistically significant, the significant difference be-
mental value to the user. They “address tasks and activities where tween MM and FM seems to suggest that users rely on classical
user motivation in using the system is driven by the expectation of a aesthetics more in the judgment of credibility of the behavior
reward or benefit external to his interaction with the system” (p. model design when the behavior models are contra-tailored.
620) [21]. In the case of fitness apps modeling behavior, the user’s In contrast, users seem to rely on expressive aesthetics more
HealthRecSys’18, October 6, 2018, Vancouver, BC, Canada K. Oyibo et al.
in the judgment of credibility of the behavior model design 6 CONCLUSION
when the behavior models are personalized. For example, with
This paper presents the results of an empirical study among 669
respect to MM vs MF, we see that the effect of classical aesthetics
participants to investigate which of the two dimensions of visual
on perceived credibility is significantly stronger for MF (β = 0.67, p
aesthetics (classical and expressive) has a stronger impact on the
< 0.001) than for MM (β = 0.46, p < 0.001), while the effect of ex-
perceived credibility of exercise behavior model design in fitness
pressive aesthetics on perceived credibility is non-significantly
apps. Our path analysis shows that, regardless of the gender-based
stronger for MM (β = 0.15, p = n.s) than for MF (β = 0.12, p = n.s).
personalization of the behavior model to the user, classical aes-
Similarly, we see that the effect of classical aesthetics on perceived
thetics has a stronger influence than expressive aesthetics on per-
credibility is non-significantly stronger for FM (β = 0.82, p < 0.001)
ceived credibility. Our finding underscores the need for designers
than FF (β = 0.56, p < 0.001), while the effect of expressive aesthetics
of persuasive systems with utilitarian benefits, such as fitness
on perceived credibility is non-significantly stronger for FF (β =
apps, to emphasize classical aesthetics (cleanness, orderliness, clar-
0.20, p = n.s) than MF (β = 0.00, p = n.s). However, given the non-
ity and simplicity) more than expressive aesthetics (enrichment and
replication of the significant difference in the classical aesthetics-
complexity). This will go a long way in enhancing the perceived
perceived credibility relationship between MM and MF with that
credibility of health apps aimed at changing behaviors.
between FF and FM, there is a need for further research into the
above highlighted proposition.
REFERENCES
[1] B. J. Fogg, Persuasive Technology: Using Computers to Change What We
5.3 Main Findings and Contributions Think and Do. Morgan Kaufmann, 2003.
Based on the significant relationship between classical aesthetics [2] H. Oinas-kukkonen and M. Harjumaa, “Persuasive Systems Design Key
Issues, Process Model, and System Features,” Commun. Assoc. Inf. Syst.,
and perceived credibility (p < 0.001), we recommend the following vol. 24, no. 28, pp. 485–500, 2009.
general design guideline: [3] B. J. Fogg, C. Soohoo, D. Danielson, L. Marable, J. Stanford, and E. R.
Tauber, “How Do People Evaluate a Web Site ’ s Credibility ? Results from
When designing persuasive systems with utilitarian value a Large Study,” 2002.
[4] K. Oyibo and J. Vassileva, “The Interplay of Aesthetics, Usability and
or benefit, such as fitness applications featuring behavior Credibility in Mobile Website Design and the Moderation Effect of
models (virtual coach) aimed at changing behavior, HCI Gender,” SBC J. Interact. Syst., vol. 7, no. 1, 2017.
[5] K. Oyibo and J. Vassileva, “The Interplay of Aesthetics, Usability and
designers should focus on enhancing classical aesthetics Credibility in Mobile Websites and the Moderation by Culture,” 2016.
(cleanness, orderliness, clarity and simplicity) to enhance [6] K. Oyibo, A. Ifeoma, R. Orji, and J. Vassileva, “What Drives the Perceived
the perceived credibility of the fitness applications. Credibility of Mobile Websites: Classical or Expressive Aesthetics?,” in
20th Int’l Conference on Human-Computer Interaction, 2018, pp. 576–594.
Our main contribution to the body of knowledge is that we repli- [7] D. Robins and J. Holmes, “Aesthetics and credibility in web site design,”
Inf. Process. Manag., vol. 44, no. 1, pp. 386–399, 2008.
cated the above prior finding in the tourism domain [6] in the [8] C. Abraham and S. Michie, “A taxonomy of behavior change techniques
health domain, indicating that the hypothesis—classical aesthetics used in interventions.,” Health Psychol., vol. 27, no. 3, pp. 379–87, 2008.
has a stronger effect than expressive aesthetics on the perceived cred- [9] M. Moshagen and M. T. Thielsch, “Facets of visual aesthetics,” Int. J. Hum.
Comput. Stud., vol. 68, no. 10, pp. 689–709, 2010.
ibility of utilitarian systems—is likely to cut across domains. Aside [10] N. Tractinsky, “Visual Aesthetics,” Interaction-Design, pp. 1–71, 2002.
from the relationship between classical aesthetics and perceived [11] T. Lavie and N. Tractinsky, “Assessing dimensions of perceived visual
aesthetics of web sites,” Int. J. Hum. Comput. Stud., vol. 60, no. 3, pp. 269–
credibility, we showed that there is a significant relationship be- 298, 2004.
tween the two dimensions of visual aesthetics, which confirms our [12] K. Oyibo and J. Vassileva, “What Drives Perceived Usability in Mobile
Web Design: Classical or Expressive Aesthetics ?,” in 19th International
third hypothesis (H3). This suggests that the higher users’ percep- Conference on Human-Computer Interaction, 2017, pp. 445–462.
tion of a fitness app to be classically aesthetic, the higher will be [13] B. J. Fogg and H. Tseng, “The elements of computer credibility,” in
their perception of the app as expressively aesthetic. However, a Proceedings of the SIGCHI conference on Human factors in computing
systems the CHI is the limit - CHI ’99, 1999, pp. 80–87.
high perception of the app as expressively aesthetic does not result [14] F. Alsudani and M. Casey, “The Effect of Aesthetics on Web Credibility,”
in a high perception of the app as credible, indicating that expres- 23rd Br. HCI Gr. Annu. Conf. People Comput. Celebr. People Technol., pp.
512–519, 2009.
sive aesthetics does not play a mediating role between the effect of [15] B. J. Fogg, C. Soohoo, D. R. Danielson, L. Marable, J. Stanford, and E. R.
classical aesthetics on perceived credibility. Tauber, “How do users evaluate the credibility of Web sites?: a study with
over 2,500 participants,” in Proceedings of the 2003 conference on Designing
for user experiences, 2003, pp. 1–15.
5.4 Limitations and Future Work [16] K. Oyibo, R. Orji, I. Adaji, O. Babatunde, M. Azizi, and J. Vassileva,
Our study has some limitations. The main limitation is that our “Persuasive Effect of Behavior Model Design in Fitness Apps,” in
Proceedings of the 2018 Conference on User Modeling Adaptation and
findings are based on users’ perception (i.e., self-report). As such, Personalization - UMAP ’18, 2018, pp. 219–228.
we cannot guarantee that they will generalize to an actual context [17] P. van Schaik and J. Ling, “The role of context in perceptions of the
in which the users have to use the fitness app and answer the aesthetics of web pages over time,” Int. J. Hum. Comput. Stud., vol. 67, no.
1, pp. 79–89, 2009.
posed questions thereafter. The second limitation of our study is [18] L. Bergkvist and J. R. Rossiter, “The Predictive Validity of Multiple-Item
that most of the participants are from Canada and the United Versus Single-Item Measures of the Same Constructs,” J. Mark. Res., vol.
44, no. 2, pp. 175–184, 2007.
States. This may threaten the generalization of our findings to [19] G. Sanchez, “PLS Path Modeling with R,” Berkley Trowchez Ed., 2013.
other demographics. Thus, in future work, we look forward to in- [20] K. K. Wong, “Partial Least Squares Structural Equation Modeling (PLS-
vestigating our findings among other demographics as well as by SEM) Techniques Using SmartPLS,” Mark. Bull., vol. 24, no. 1, pp. 1–32,
2013.
using an actual fitness app. [21] H. Wigelius and H. Väätäjä, “Dimensions of context affecting user
experience in mobile work,” in IFIP Conference on Human-Computer
Interaction, 2009, pp. 604–617.