=Paper=
{{Paper
|id=Vol-1153/Paper_5
|storemode=property
|title=Exploring the Persuasiveness of Behavior Change Support Strategies and Possible Gender Differences
|pdfUrl=https://ceur-ws.org/Vol-1153/Paper_5.pdf
|volume=Vol-1153
|dblpUrl=https://dblp.org/rec/conf/persuasive/Orji14
}}
==Exploring the Persuasiveness of Behavior Change Support Strategies and Possible Gender Differences==
https://ceur-ws.org/Vol-1153/Paper_5.pdf
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
41
Exploring the Persuasiveness of Behavior Change Support
Strategies and Possible Gender Differences
Rita Orji
Computer Science Department
University of Saskatchewan
Saskatoon, SK. S7N 5C9, Canada
{rita.orji@usask.ca}
Abstract. There is need to investigate the persuasiveness of various health
behavior promoting strategies that are commonly employed in behavior
change interventions design with respect to possible gender effect. Behavior
change researchers have advocated the need to adapt persuasive approaches to
various user characteristics. Gender has been identified to influence behavior
in many domains. Therefore, this paper presents a comparative study
investigating the perceived persuasiveness of health behavior promotion
applications depicting ten commonly employed behavior change strategies.
The population of interest are males and females and the purpose of the study
is to investigate differences in persuadability and the perceived persuasiveness
of behavior change strategies overall. To achieve this, we conducted a large-
scale study on 1108 participants (575 males and 533 females) to examine the
persuasiveness of ten strategies that are commonly employed in health
behavior change intervention design. We also examined possible gender
effects on the persuasiveness of various strategies. The results of the analysis
show that some of the strategies studied are highly persuasive overall, while
others were rated low in persuasiveness. The results also suggest that males
and females differ significantly in persuadability – with females being more
receptive to most of the behavior change strategies. Some strategies are more
suitable for persuading one gender than the other. We therefore conclude that
gender-dependent approaches would generally be more appropriate for
designing behavior change support systems that will effectively promote
health behavior change than the one-size-fits all approach.
Keywords: Persuasive Technology, Behavior Change, Gender, Persuasive
Strategies, Persuasiveness, Health Behavior, PSD, health intervention,
mhealth, health.
1 Introduction
Recent years have witnessed an increasing number of lifestyle-related health
problems. Research has shown that adoption of healthy behavior can prevent or at
least reduce the risk of many diseases, including obesity, heart disease, and type 2
diabetes [34]. It is, therefore, not surprising that interventions aimed at modifying
health behavior have been identified as a major solution to these health conditions
42
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
[16]. As a result, research on how to design technology to motivate behavior
change is a key area of inquiry of Behavior Change Support Systems (BCSSs)
research within the Persuasive Technology (PT) community. Research has shown
the potential of behavior change support systems to motivate healthy behavior –
help people achieve personal wellness, manage diseases, and engage in
preventive behaviors [6,11,14,22] using several persuasive strategies.
Over the years, several persuasive strategies have been developed [9,20].
However, many of these strategies are conjecture and their effectiveness have not
been validated on a large-scale study while few of them have only been
qualitatively evaluated – with systematic validation. As a result, most of the
BCSSs assume a one-size-fits-all approach with respect to their choice of
behavior change strategies to employ in their intervention design. This is based
on the assumption that the strategies are equally persuasive and would similarly
motivate people to change their behavior. However, people differ in motivation; a
strategy that motivates one type of person to change her behavior may actually
deter behavior change for another type of person [15]. Therefore, designing a
technology that will inspire a positive user experience and effectively motivate
health behavior change requires adapting the strategies based on the knowledge
of their persuasiveness. Research has shown that tailoring behavior change
strategies would increase the effectiveness of behavior change support systems in
the domain of health [15]. According to Berkovsky et al. [2], tailoring persuasive
strategies has a “huge untapped potential to maximize the impact of persuasive
applications”. The success of different BCSSs will be partly dependent on the
persuasiveness of the strategies employed in their design and the appropriateness
of the strategies for the target users or user group. However, research on tailoring
behavior change strategies based on the knowledge of their persuasiveness is just
beginning.
In choosing approaches for group-based tailoring, research has shown that
gender is a reliable approach [26]. Research has also established gender
differences in many areas including the perception of different behavior
determinants [26], gameplay, and health behavior [7]. However, whether or not
gender influences the persuasiveness of various behavior change strategies as
highlighted by the Persuasive System Design (PSD) framework [20] has not been
examined. Investigating the persuasiveness of these strategies and how they are
perceived by different gender group is necessary to aid tailoring BCSS to the
various gender groups to increase their effectiveness at achieving their intended
objective of motivating behavior change.
Therefore, this paper investigates the persuasiveness of various behavior change
strategies and possible gender differences in the persuasiveness of the strategies.
We achieve this by comparing the effectiveness of ten PT strategies –
competition, comparison, cooperation, customization, personalization, praise,
simulation, Self-monitoring and Feedback, suggestion, and reward (from Fogg
[9] and Oinas-Kukkonen [20]) – within and across the gender groups. The results
of a large-scale study of 1108 participants (575 males and 533 females) suggest that
males and females differ significantly in persuadability – with females being
more receptive to most of the PT strategies. The study also provides a
quantitative validation of the persuasiveness of the strategies overall. Some of the
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
43
strategies are perceived as highly persuasive by the participants overall, while
others were scored low in persuasiveness with respect to their efficacy to
motivate healthy behavior change. Yet, some of the strategies are intermediately
persuasive.
2 Background
Over the years, a number of strategies for designing behavior change support
systems have been developed. For example, Fogg [9] developed seven
persuasive tools, and Oinas-Kukkonen [20] built on Fogg’s strategies to develop
28 persuasive system design principles. These strategies are often applied in
combinations when incorporated in actual software [13]. Therefore, it is common
practice for researchers in persuasion to select a combination of strategies from
various authors to inform their design. The choice of the strategies based on their
persuasiveness and their suitability for particular users or user group are often
based on a designer’s own intuition, making it difficult to tailor strategies to users
or user groups.
Considering that the large number of PT strategies in existence today cannot be
exhausted in a studio, in this paper, we adopt 10 strategies (from Fogg and Oinas-
Kukkonen). Personalization offers system-tailored contents and services to its
users, tailoring content and functionality to a particular user’s need based on a
user’s characteristics. For a detailed discussion of the strategies see [20].
Simulation provides the means for a user to rehearse the behavior and to observe
the cause-and-effect linkage of their behavior. It is one of the rarely employed
strategies in health game design. Self-monitoring allows people to track their own
behaviors, providing information on both past and current states. It is one of the
most common strategies for healthy eating and physical activity motivating
applications [3,32]. The Suggestion strategy suggests certain tasks (for achieving
favorable behavior outcomes) to users during system use. Praise applauds the
user for performing the target behavior via words, images, symbols, or sounds as
a way to give positive feedback to the user (for example in [1,30]). Reward
offers virtual rewards to users for performing the target behavior. It is one of the
commonly employed strategies [25]. Competition allows the user to compete with
others. Comparison provides a means for the user to view and compare his/her
performance with the performance of other user(s). Competition, and Comparison
are included among the commonly used strategies. Cooperation requires users to
cooperate (work together) to achieve a shared objective and rewards them for
achieving their goals collectively. Customization is a strategy that provides the
user an opportunity to adapt a system’s contents and functionality to their needs
or choices. These strategies have been employed in the design of several health
behavior change support systems (for examples, see [3,15,25,30]).
3 Study Design and Methods
For the purpose of this study, we chose to focus on common application of
behavior change technology to ensure uniformity: behavior change technology
44
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
for encouraging healthy eating behavior. Through a review of related work in
designing behavior change support systems, we established a comprehensive list
of persuasive strategies and how they have been operationalized in behavior
change support systems. Storyboards provide a common visual language that
individuals from diverse backgrounds can read and understand [18]. Considering
that we cannot exhaustively study the large number of behavior change strategy
from the literature, we selected 10 commonly employed – competition,
comparison, cooperation, customization, personalization, praise, simulation,
Self-monitoring and Feedback, suggestion, and reward (from Fogg [9] and Oinas-
Kukkonen [20]). Recent reviews also identified these strategies among the
commonly used PT strategies in persuasive systems design [17,35]. However, it
is important to note that these ten strategies are not more important than the rest
and may not be representative of all strategies.
To collect data for our model, we follow the approach described by Halko and
Kientz [12]. Specifically, we represented each behavior change strategy in a
storyboard. Although we could implement the individual strategies and then
evaluate their persuasiveness in actual BCSS, we chose to use storyboards
because actual implementation may create additional noise as it involves many
other design decisions and the results can easily be biased by specific
implementation decisions. The storyboards show a character and his/her
interactions with a persuasive application for promoting healthy eating. The ten
storyboards were drawn by an artist and were based on storyboard design
guidelines by Truong et al. [31]. Figure 1 shows examples of two of the ten used
persuasive strategies, reward and self-monitoring. Prior to assessing the
persuasiveness of the various strategies, we ensured that the participants
understood the strategy depicted in each storyboard by asking them two
comprehension questions – first, to identify the illustrated strategy from a list of
ten different strategies; and second, to describe what is happening in the
storyboard in their own words. To elicit feedback on the persuasiveness of the
strategies, each storyboard was followed by a validated scale consisting of four
questions for measuring perceived persuasiveness, adapted from Drozd et al. [8].
Specifically, we asked participants the following questions after they have
successfully answered the comprehension questions that show that they
understood the strategy depicted in the storyboard:
Imagine that you are using the system presented in storyboard above to track
your daily eating, on a scale of 1 to 7 (1-Strongly disagree and 7-Strongly agree),
to what extend do you agree with the following statements:
a. The system would influence me.
b. The system would be convincing.
c. The system would be personally relevant for me.
d. The system would make me reconsider my eating habits.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
45
The questions were measured using participant agreement with a 7-point Likert
scale ranging from “1 = Strongly disagree” to “7 = Strongly agree”.
To eliminate possible bias due to the ordering of the storyboards in the survey,
we used a Latin Square to balance the order of presentation of the persuasive
strategies. We created ten surveys that varied the position of each strategy and
randomly assigned participants to one of the ten surveys.
We recruited participants for this study using Amazon’s Mechanical Turk
(AMT). AMT has become an accepted method of gathering users’ responses [19].
It allows access to a global audience at a relatively low cost, and ensures efficient
survey distribution, and high quality results [4,19]. We followed the
recommendations for performing effective studies on the AMT by Mason and
Suri [19], and used a similar approach to the one described by Halko and Kientz
[12]. The study took an average of an hour to complete. Before the main study,
we conducted pilot studies to test the validity of our study instruments.
A total of 1384 participants responded to our study. A total of 1108 valid
responses were retained and included in the analysis. The participants
demographic information is summarized in Table 1. In general, our participants
are fairly distributed across the gender groups. With respect to age and education
level attained, we have a diverse population.
Figure 1: Storyboard illustrating reward and self-monitoring strategy
46
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
Table 1: Participants’ demographic information
Total Participants = 1108
Gender Females (533, 48%), Males (575, 52%)
Age 18-25 (418, 38%), 26-35 (406, 37%), 36-45 (168, 15%), Over 45
(116, 10%).
Education Less than High School (12, 1%), High School Graduate (387, 35%),
College Diploma (147, 13%), Bachelor’s Degree (393, 35%),
Master’s Degree (141, 13%).
4 Data Analysis and Results
We begin our analysis by validating our study instrument. First, to ensure that
participants understood the intended persuasive strategy in each of the
storyboards, we ran chi-squared tests on the participants’ responses to the
multiple-choice questions that required them to identify the represented
persuasive strategy for each of the storyboards. The results for all the strategies
were significant at p<. 001. Second, we determined the consistency of the scale
using Cronbach’s alpha (α). The α for the strategies were all greater than 0.70
showing that the scales have good internal consistency. Third, to determine
whether responses to each strategy were unique in our data, we performed
Exploratory Factor Analysis (EFA), which showed that self-monitoring and
suggestion loaded into one factor and competition and comparison loaded into
one factor as well. Hence, the total number of factors examined in this study was
reduced from ten to eight. Next, we examine the persuasiveness of the strategies.
Alongside examining the differences in perceived persuasiveness between males
and females, validating the overall persuasiveness of the individual strategies for
promoting healthy behavior is of interest. To achieve this, we performed one-
sample t-test separately on the data for males and females and on the combined
data – to obtain an overall persuasiveness of the strategies. We compared this
data against a neutral rating for the perceived persuasiveness scale of 4. Figure 2
and Table 2 present the details of the overall persuasiveness of the individual
strategies.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
47
In general, participants perceived most of the strategies as persuasive.
Specifically, all the strategies apart from reward and customization were
perceived as persuasive. Customization is significantly below the neutral rating of
4 making it the least persuasive among all the strategies studied, Table 2. On the
other hand, personalization and simulation emerged as strategies that are
perceived as most persuasive (capable of motivating health behavior change) with
mean ratings quite high and well above the neutral rating of 4, with mean
differences close to 1 – see Figiure 2 and Table 2.
Figure 2: A bar graph of the mean of individual strategies showing their overall
persuasiveness. Error bars represent a 95% confidence interval.
48
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
Table 2: Mean and Standard Deviations (SD), Mean Difference (MD), t-values
(t2), and significant levels for the individual strategies on a scale from 1(low) to
7(high) for overall persuasiveness.
N = 1108
Mean SD MD t2 p
COOP 4.40 1.76 0.41 7.69 <.0001
CUST 3.35 1.75 -0.65 12.38 <.0001
PERS 4.84 1.64 0.83 17.04 <.0001
PRAS 4.22 1.75 0.21 4.01 <.0001
REWD 3.91 1.82 -0.09 1.67 <.0960
SIML 4.62 1.72 0.62 11.88 <.0001
CMPTCMPR 4.40 1.72 0.40 7.81 <.0001
SEMSUG 4.31 1.59 0.31 6.57 <.0001
With respect to gender differences, males and females perceived most of the
strategies as persuasive, see Figure 3 and Table 3. Similar to the general group,
personalization and simulation emerged as the most persuasive strategies that is
capable of motivating health behavior change for both males and females.
Customization is significantly below the neutral rating of 4 – making it the
strategy that is perceived as least persuasive for both males and females. Reward
on the other hand is a borderline strategy – that is exactly equal to the neutral
rating of 4 – for females while it is below the neutral rating of 4 for males and
therefore listed among the least persuasive together with customization for males
– Figure 2 and Table 2.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
49
Figure 3: A bar graph of the mean of individual strategies showing their
persuasiveness for males and females. Error bars represent a 95% confidence
interval.
Table 3: Means and Standard Deviations (SD), Mean Difference (MD), t-values
(t2), and Significant levels (p) of the persuasiveness rating of the ten strategies on
a scale from 1 (low) to 7 (high) for females and males separately.
N = 533 N = 575
Females Males
Mean SD MD t2 p Mean SD MD t2 p
COOP 4.52 1.79 0.52 6.73 <.000 4.30 1.73 0.30 4.16 <.0001
CUST 3.46 1.71 -0.54 7.25 <.000 3.24 1.78 -0.76 10.19 <.0001
PERS 5.02 1.57 1.03 15.04 <.000 4.66 1.68 0.67 9.52 <.0001
PRAS 4.33 1.74 0.33 4.31 <.153 4.10 1.75 0.10 1.43 <.0001
REWD 4.00 1.84 0.00 0.041 <.017 3.82 1.79 -0.18 2.39 <.967
SIML 4.78 1.70 0.78 10.62 <.000 4.46 1.73 0.46 6.39 <.0001
CMPTCMPR 4.35 1.76 0.36 4.64 <.000 4.45 1.68 0.45 6.42 <.0001
SEMSUG 4.36 1.59 0.35 5.17 <.000 4.27 1.59 0.27 4.15 <.0001
COOP = cooperation, CUST = customization, PERS = personalization, PRAS = praise, SIML = simulation,
REWD = reward, CMPTCMPR = competition and &comparison, SEMSUG = self-monitoring and suggestion.
50
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
4.1 Interaction Between Gender and Behavior Change Strategies
From the t-test, we established that both males and females perceive some
strategies as highly persuasive (e.g., personalization and simulation) while other
strategies (e.g., customization) scored low in the persuasiveness scale. However,
the magnitudes of persuasiveness rating for the individual strategies were
different, suggesting possible differences in the persuasiveness of the strategies
for males and females – Table 3. To explore for significant differences between
males and females with respect to the persuasiveness of various strategies, we
performed the Repeated-Measure ANOVA (RM-ANOVA) on our data.
Specifically, we examine the effect of gender on the persuasiveness of the various
PT strategies using RM-ANOVA in SPSS 21. The analysis was performed after
validating our data for ANOVA assumptions, with no violations. When the
sphericity assumption was violated, we used the Greenhouse-Geisser method of
correcting the degrees of freedom. Pairwise comparison used the Bonferonni
method of adjusting for multiple comparisons.
The results of the RM-ANOVA show significant main effects of strategy
(F6.05,6687.58=184.718, p≈.000, η2=.143) and gender (F1,1106= 5.331, p≈.021,
η2=.005) on persuasiveness (see Table 4 and Figure 4). Overall, females rated the
strategies as more persuasive than males, however; there was also a significant
strategy by gender interaction on persuasiveness (F6.05,6687.58=4.463, p≈.000,
η2=.004). Pairwise comparisons show that females found five out of the eight
strategies significantly more persuasive than males: personalization
(F1,1106=13.153, p≈.000, η2=.012); simulation (F1,1106=9.831, p≈.002, η2=.009);
cooperation (F1,1106=4.418, p≈.036, η2=.004); customization (F1,1106=4.386,
p≈.036, η2=.040); and praise (F1,1106=4.428, p≈.036, η2=.004).
Table 4: Mean and Standard Deviations (SD) for the strategies by gender. Bolded
means are significantly different across males and females.; p<. 05.
Strategies CMPT/ COOP CUST PERS PRAS SEM SIML REWD
CMPR SUGG
mean(SD) mean(SD) mean(SD) mean(SD) mean(SD) mean(SD) mean(SD) mean(SD)
Males 4.45(1.68) 4.30(1.73) 3.24(1.79) 4.67(1.68) 4.10(1.75) 4.27(1.58) 4.46(1.73) 3.82(1.79)
Females 4.35(1.76) 4.52(1.79) 3.46(1.71) 5.02(1.57) 4.33(1.74) 4.36(1.59) 4.78(1.70) 4.00(1.84)
COOP = cooperation, CUST = customization, PERS = personalization, PRAS = praise, SIML =
simulation, REWD = reward, CMPTCMPR = competition and &comparison, SEMSUG = self-
monitoring and suggestion.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
51
Figure 4: Paired mean of individual strategies by gender group. Error bars
represent a 95% confidence interval.
5 Discussion
This study presents the results of a large-scale evaluation of ten persuasive
strategies that are commonly employed in developing behavior change support
systems. Many of these strategies are conjecture and their effectiveness have not
been validated in a large scale study while few of them have only been
qualitatively evaluated – with systematic validation. The study presented in this
paper provides a quantitative validation of the persuasive strategies and the
influence of gender on the persuasiveness of the strategies. To achieve this we
represented the individual strategies in a storyboard showing persuasive
application for promoting healthy eating and collected quantitative measures
from 1108 participants – 533 females and 575 males – using the storyboard. The
results of analysis of the data show that as expected, most of the strategies are
perceived as highly persuasive by the participants overall, while others were
scored low in persuasiveness with respect to their efficacy to motivate healthy
behavior change.
5.1 Comparing the Persuasiveness of the Strategies by Males and Females
The results show that males and females differ with regard to their perceived
persuasiveness of five out of the eight strategies examined in this paper.
Surprisingly, females perceive five strategies: cooperation, customization,
personalization, praise, and simulation as being more persuasive than males –
52
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
Table 4 and Figure 4. Below, we discuss these results with respect to the
persuasiveness of the strategies.
Personalization and Customization: Personalization and customization
represent two different ways of tailoring from literature. Both personalization and
customization emphasize tailoring system contents to the user group. However, in
personalization the system initiates and control the tailoring to users based on
user characteristics – system-controlled tailoring – while in the customization, the
user initiates and controls the tailoring – user-controlled tailoring. Although
customization strategy is not explicitly included in the PSD model as a persuasive
strategy, research has identified it as strategy different from the popular
personalization strategy that is listed in the PSD model [27,28]. The result from
our study also confirmed that they are in fact different.
While personalization emerged as the most persuasive strategy for both males
and females from our study, customization emerged as the least persuasive
strategy that may not motivate meaningful behavior change. A possible
explanation while customization is perceived as less persuasive than
personalization is that most users tend to use only the default system features [29]
and tend to dislike systems that require a lot of input from them [24] –
customization. Therefore, although, most people would prefer systems that tailor
their contents to them, they would prefer a system that does that automatically
(personalization) to a system that requires their input – customization. Therefore,
behavior change support systems should be designed to require minimal user
input for tailoring purposes. This suggests a need for various ways of tracking
and sensing users’ behaviors automatically to aid system adaptation –
personalization.
Fortunately, personalization is among the strategies that are moderately employed
in health behavior change systems design [17]. Interestingly, although males and
females perceive personalization as highly persuasive, personalization is also a
differentiator of males and females. Females perceive both personalization and
customization as more persuasive than males.
Simulation: Simulation strategy which deals with providing users opportunity to
rehearse their behavior and to observe the cause-and-effect linkage of their
behavior emerged as the second highly persuasive strategy that is capable of
motivating health behavior change for both males and females. Although
simulation is not among the commonly employed persuasive strategies in health
behavior promoting applications, the persuasiveness score stresses a need for
behavior changing application to include some features that allows people to
rehearse and observe the simulated impact of their behaviors both in short and
long-term. This is important because the intangible and the gradual nature of
achieving the benefit of adopting healthy behaviors are often barriers to adopting
healthy behavior. Adopting healthy behavior is a lifestyle than spans over a
lifetime with no quantifiable benefit [22], therefore, simulation strategy that
allows users to view both immediate and projected impacts of their health
behavior may bridge this gap and make the benefit of adopting health behavior
more visible and tangible. Similar to personalization and customization, females
perceive simulation as more persuasive than males.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
53
Cooperation: According to the PSD framework, “a system can motivate users to
adopt a target attitude or behavior by leveraging human beings’ natural drive to
cooperate.”[20]. From the results of our study, females found cooperation more
persuasive than males and therefore will be motivated to change their behavior by
any behavior change system that employs the cooperation strategy. This is
probably because females are more susceptible to social influence, social
facilitation [9], and social support [20] and therefore, more inclined to performing
the target behavior when they are working together with others than males. This
result is in line with research in other domain that found that females generally
cooperated and their cooperation unlike males are largely unconditional [33].
Our study is also in line with previous studies that found that social influence is a
contributing factor that influence how females perceive their weight and how it
affect their behavior [23,26], while it is not significant for males. Unfortunately,
cooperation strategy is rarely used in health intervention design [17]. However,
cooperation is the third strategy that is perceived as persuasive (after
personalization and simulation) for females and therefore should be employed in
designing behavior change systems especially those targeting healthy behaviors.
Competition and Comparison: Competition and Comparison are listed as two
separate strategies by the PSD model [21]. However, according to our analysis,
they belong together. This is understandable considering that in most situations;
competition is often a by-product of comparison. Competition/comparison is
among the strategy that is frequently used in health behavior change intervention.
The results from our study show that competition/comparison is moderately
persuasive. According to previous research, males are more inclined to
competition and can even be motivated to cooperate to win a competition than
females [33]. The results from our study support this finding by showing that
competition/comparison is the only strategy that is perceived as more persuasive
by males than females (although the difference is not significant). This suggests
that employing competition strategy in the design of a behavior change system
will motivate behavior change in males than females.
Self-monitoring and Suggestion: Self-monitoring and suggestions are listed as
two separate strategies by the PSD model [21]. However, according to our
analysis, they belong together. This is understandable considering that effective
suggestion would require context awareness (that is often achieved through
monitoring) to determine the opportune moments.
Self-monitoring is among the most frequently employed persuasive strategies,
especially those aimed at promoting healthy eating behaviors [17]. However,
from the results of our study, self-monitoring is intermediately persuasive. This is
probably because of the labour intensive nature of current (diet) self-monitoring
systems that often requires some level of input from the user to be effective. Self-
monitoring is equally persuasive for both males and females.
Praise: According to the PSD model, systems that applaud users for performing
the target behavior are more likely to motivate them to adopt healthy behavior
[20]. Praise is intermediately persuasive and therefore, can moderately motivate
behavior. It is infrequently used in behavior change motivating systems [17].
Males and females differ with respect to the persuasiveness of praise. Females
54
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
perceive praise as more persuasive than males. This is probably because females
are more inclined to respond to strategies that appeal to emotions than males.
Reward: Reward is the least persuasive strategy after customization. This is
probably contrary to popular expectations. Many behavior change systems offer
one type of reward or the other to the users to encourage them to perform the
behavior. The use of any form of reward to motivate behavior change has been a
subject of debate because of the tendency of reward to trivialize the benefit of
adopting healthy behavior and make it extrinsically motivated [5,10]. The results
from our study show that reward is not all that important a strategy for motivating
behavior change and therefore, can be excluded. There is no difference between
males and females with respect to the persuasiveness of the reward strategy.
6 Limitation
This study examined the perceived persuasiveness using the storyboards
implementation of the strategies, however, actual persuasiveness may be different
when implemented and used in actual behavior change support system. Although
we use the application for motivating healthy eating as a sample in our
storyboards, the storyboards were drawn at a high enough level that it does not
encapsulate much of a specific application domain. However, further work is
needed to establish the applicability of our result in other domains.
7 Conclusion and Future Work
The study validated that the persuasiveness of various persuasive strategies in use
today (which have not been validated in large-scale studies). The results suggest
that these strategies could be employed to design behavior change support
systems to motivate healthy behavior change overall. We also establish that
gender influences the persuasiveness of the strategies. Specifically, males and
females differ with regard to the perceived persuasiveness of five out of the eight
strategies examined in this paper. Surprisingly, females perceive five strategies:
cooperation, customization, personalization, praise, and simulation as being more
persuasive than males. This implies that females can be more easily persuaded
using these strategies. It also suggests that females are more persuadable than
males with respect to the influence of the strategies on their behavior. The
gender-related differences across a number of strategies also suggest that gender-
dependent approaches would generally be more appropriate for designing
behavior change support systems that will effectively promote health behavior
change than the one-size-fits all approach.
In general, regardless of gender, personalization and simulation emerged as the
most persuasive (significantly different from all other strategies), whereas reward
and customization were the least persuasive (also significantly different from all
others). The rest of the strategies – competition/ comparison, cooperation, self-
monitoring, and praise – were in the middle with competition/comparison and
cooperation leading the group.
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
55
Future work should examine the applicability of our result in other domains by
examining the persuasiveness of the strategies using application from other
domains. Research should also design and compare the effectiveness of behavior
change support systems designed using strategies that are listed as highly
persuasive (personalization and simulation) with those that are scored low in the
persuasiveness scale (e.g., reward and customization).
Acknowledgements: The author of this paper is being sponsored by the Natural
Sciences and Engineering Research Council of Canada (NSERC) Vanier
Graduate Scholarship. Many thanks to the reviewers for their insightful
comments.
References
1. Bang, M., Torstensson, C., and Katzeff, C. The PowerHhouse: A
Persuasive Computer Game Designed to Raise Awareness of Domestic
Energy Consumption. Persuasive Technology, (2006), 123–132.
2. Berkovsky, S., Freyne, J., and Oinas-kukkonen, H. Influencing
Individually. ACM Transactions on Interactive Intelligent Systems 2, 2
(2012), 1–8.
3. Brown, B., Chetty, M., Grimes, A., and Harmon, E. Reflecting on
health: a system for students to monitor diet and exercise. CHI’06
extended abstracts on Human factors in computing systems, ACM
(2006), 1807–1812.
4. Buhrmester, M., Kwang, T., and D, G.S. Amazon’s Mechanical Turk A
New Source of Inexpensive, Yet High-Quality, Data? Perspectives on
Psychological Science 6, 1 (2011), 3–5.
5. Colineau, N. and Paris, C. Can Beneficial Habits Be Induced through
Reflection ? Workshop on User Models and Motivational Systems: the
effective and the rational routes to persuasion, with UMAP’2010., .
6. Consolvo, S., McDonald, D.W., and Landay, J.A. Theory-driven design
strategies for technologies that support behavior change in everyday life.
Proceedings of the 27th International Conference on Human factors in
Computing Systems, ACM (2009), 405–414.
7. Dawson, K. a, Schneider, M. a, Fletcher, P.C., and Bryden, P.J.
Examining gender differences in the health behaviors of Canadian
university students. The Journal of the Royal Society for the Promotion
of Health 127, 1 (2007), 38–44.
8. Drozd, F., Lehto, T., and Oinas-Kukkonen, H. Exploring perceived
persuasiveness of a behavior change support system: a structural model.
Persuasive Technology., (2012), 157–168.
56
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
9. Fogg, B.J. Persuasive Technology: Using Computers to Change What
We Think and Do. Morgan Kaufmann, 2003.
10. Gneezy, U. and Rustichini, A. Pay Enough or Don’t Pay at All*.
Quarterly Journal of Economics 115, 3 (2000), 791–810.
11. Grimes, A., Kantroo, V., and Grinter, R.E. Let’s play!: mobile health
games for adults. Proceedings of the 12th {ACM} International
Conference on Ubiquitous computing, ACM (2010), 241–250.
12. Halko, S. and Kientz, J. Personality and Persuasive Technology: An
Exploratory Study on Health-Promoting Mobile Applications.
Persuasive Technology, (2010), 150–161.
13. Harjumaa, M. Understanding persuasive software functionality in
practice: a field trial of polar FT60. Persuasive Technology, (2009).
14. Kaptein, M., Duplinsky, S., and Markopoulos, P. Means based adaptive
persuasive systems. Proceedings of the 2011 annual conference on
Human Factors in Computing Systems, (2011), 335.
15. Kaptein, M., De Ruyter, B., Markopoulos, P., and Aarts, E. Adaptive
Persuasive Systems. ACM Transactions on Interactive Intelligent
Systems 2, 2 (2012), 1–25.
16. Lau, D.C.W., Douketis, J.D., Morrison, K.M., Hramiak, I.M., Sharma,
A.M., and Ur, E. 2006 Canadian clinical practice guidelines on the
management and prevention of obesity in adults and children
[summary]. CMAJ : Canadian Medical Association journal = journal de
l’Association medicale canadienne 176, 8 (2007), S1–13.
17. Lehto, T. and Oinas-kukkonen, H. Persuasive Features in Six Weight
Loss Websites : A Qualitative Evaluation. (2010), 162–173.
18. Lelie, C. The value of storyboards in the product design process.
Personal and Ubiquitous Computing 10, 2-3 (2005), 159–162.
19. Mason, W. and Suri, S. Conducting behavioral research on Amazon’s
Mechanical Turk. Behavior Research Methods 44, 1 (2012), 1–23.
20. Oinas-Kukkonen, H. and Harjumaa, M. A systematic framework for
designing and evaluating persuasive systems. Persuasive Technology,
(2008), 164–176.
21. Oinas-Kukkonen, H. and Harjumaa, M. Persuasive systems design: Key
issues, process model, and system features. Communications of the
Association for Information Systems 24, 1 (2009), 28.
22. Orji, R., Mandryk, R.L., Vassileva, J., and Gerling, K.M. Tailoring
persuasive health games to gamer type. Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems - CHI ’13, ACM
Press (2013), 2467–2476.
23. Orji, R., Vassileva, J., and Mandryk, R. Towards an effective health
interventions design: an extension of the health belief model. Online
journal of public health informatics 4, 3 (2012).
24. Orji, R., Vassileva, J., and Mandryk, R.L. Modeling the Efficacy of
Persuasive Strategies for Different Gamer Types in Serious Games for
Second
International
Workshop
on
Behavior
Change
Support
Systems
(BCSS
2014)
57
Health. User Modeling and User Adapted Interaction: Special Issue on
Personalization and Behaviour Change, .
25. Orji, R., Vassileva, J., and Mandryk, R.L. LunchTime: a slow-casual
game for long-term dietary behavior change. Personal and Ubiquitous
Computing, (2012), 10.1007/s00779–012–0590–6.
26. Orji, R.O., Vassileva, J., and Mandryk, R.L. Modeling Gender
Differences in Healthy Eating Determinants for Persuasive Intervention
Design. Persuasive Technology 7822, (2013), 161–173.
27. Sundar, S.S., Bellur, S., and Jia, H. Motivational Technologies: A
Theoretical Framework for Desiging Preventive Health Application.
Persuasive technology 7284, (2012), 112–122.
28. Sundar, S.S. and Marathe, S.S. Personalization versus Customization:
The Importance of Agency, Privacy, and Power Usage. Human
Communication Research 36, 3 (2010), 298–322.
29. Sundar, S.S. and Marathe, S.S. Personalization versus Customization:
The Importance of Agency, Privacy, and Power Usage. Human
Communication Research 36, 3 (2010), 298–322.
30. Toscos, T., Faber, A., An, S., and Gandhi, M.P. Chick clique: persuasive
technology to motivate teenage girls to exercise. {CHI} ’06 extended
abstracts on Human factors in computing systems, ACM (2006), 1873–
1878.
31. Truong, K., Hayes, G., and Abowd, G. Storyboarding: an empirical
determination of best practices and effective guidelines. DIS, (2006),
12–21.
32. Tsai, C.C., Lee, G., Raab, F., et al. Usability and Feasibility of PmEB: A
Mobile Phone Application for Monitoring Real Time Caloric Balance.
Mobile Networks and Applications 12, 2-3 (2007), 173–184.
33. Van Vugt, M., De Cremer, D., and Janssen, D.P. Gender differences in
cooperation and competition: the male-warrior hypothesis.
Psychological science 18, 1 (2007), 19–23.
34. Wansink, B. Mindless Eating: Why We Eat More Than We Think.
Bantam, 2006.
35. Wiafe, I. and Nakata, K. Bibliographic Analysis of Persuasive Systems:
Techniques, Methods and Domains of Application. Persuasive
Technology, (2012), 61 – 64.