Running head: CHOICE AS A PERSUASIVE STRATEGY
MASTER’S THESIS
Choice as a persuasive strategy: The effect of different choice options on the persuasiveness of health messages about physical activity and the moderating role of need for autonomy
Adriana Solovei 11107464
Graduate School of Communication
Master’s Programme of Communication Science (Research Master) Supervisor: Dr. Eline Smit
CHOICE AS A PERSUASIVE STRATEGY 2 Abstract
Choice is a message feature commonly used in online health communication (e.g., through hyperlinks that can be clicked on or by emphasizing it verbally). However, results regarding the persuasiveness of choice are mixed. The goal of this study is to test the separate effects of two different types of choice (i.e., verbal and physical). Also, the moderating role of need for autonomy and the mediating effects of autonomous motivation and reactance (affective and cognitive) in the persuasiveness of choice are assessed. To our knowledge, this study is the first to differentiate between two types of choice in an online message: physical choice and verbal choice. In a two (verbal choice: present or absent) by two (physical choice: present or absent) experimental design, 218 participants were randomly assigned to an online health message (on an existing website and based upon the original website text) about the role of physical activity in cancer prevention. Structural equation modeling was employed for data analysis. Results showed that verbal choice had a negative effect on the intention to increase physical activity, but only when physical choice was not present in the message. Verbal choice also had a negative effect on cognitive reactance, which negatively influenced behavioral intention through attitude. This indirect path diminished the negative effect found of verbal choice on intention, which however remained negative (b*total effect = -.18). Moreover, the strength of the effect of verbal choice on cognitive reactance was positively influenced by need for autonomy. No main effects were found of physical choice on attitude and intention towards increasing physical activity, nor of verbal choice on attitude. Also, no effects were found via autonomous motivation or affective reactance. These findings contribute to the understanding of circumstances under which
providing choice leads to mixed persuasive outcomes. Verbal choice should be accompanied by physical choice in online health messages, in order to avoid backfiring persuasive effects.
Furthermore, this study shows that, especially for target groups with members with a higher need for autonomy (vs. lower need for autonomy), verbal choice can be used in order to decrease cognitive reactance against the message.
Keywords: verbal choice, physical choice, need for autonomy, reactance, autonomous motivation, online health communication, physical activity, cancer prevention.
CHOICE AS A PERSUASIVE STRATEGY 3 Choice as a persuasive strategy: The effect of different choice options on the persuasiveness of
health messages about physical activity and the moderating role of need for autonomy Cancer is the second leading cause of death, accounting for 1 in 6 deaths worldwide (World Health Organization, 2017). Evidence suggests that between 30% and 50% of cancer deaths can be prevented through healthier lifestyle choices, such as a more balanced nutrition, more physical exercise, avoiding tobacco products and reducing alcohol drinking (World Cancer Research Fund International, 2017). Researchers and practitioners in health communication therefore continuously explore and develop strategies aimed at convincing people to adopt healthier lifestyles. One of the most popular channels used for health interventions nowadays is the internet (Chou, Prestin, Lyons, & Wen, 2013). However, recent studies suggest that the effect sizes of online computerized health interventions are rather low (Krebs, Prochaska, & Rossi, 2010; Lustria, Noar, Cortese, Van Stee, Glueckauf, & Lee, 2013). Therefore, it is of importance to explore the nature of persuasive effects of online health communication, as well as the circumstances under which these effect occur (Maher, Lewis, Ferrar, Marshall, De Bourdeaudhuij, & Vandelanotte, 2014).
One of the strategies often used in online (health) communication is providing choice, so that the message recipients experience an increased feeling of freedom and autonomy, which is expected to raise the persuasiveness of the message (Deci & Ryan, 2000; Fransen, Smit, & Verlegh, 2015). For example, individuals are offered the opportunity to choose the behavior(s) they want to receive information about (Patall, Cooper, & Robinson, 2008) or the mode in which the content is delivered (Nguyen et al., 2017). However, results regarding the persuasive effects of choice are rather mixed. Although a majority of studies indicate that choice has positive effects on attitude and intention (for a review see Patall et al., 2008), there are also studies that
CHOICE AS A PERSUASIVE STRATEGY 4 found null (Reeve, Nix, & Hamm, 2003) or negative effects of choice (Schwartz, 2000).
Therefore, further investigations of the mechanisms and circumstances under which persuasive effects of choice take place are necessary.
Patall et al. (2008) suggested that a possible explanation for the mixed results regarding the persuasive effects of choice may be that different types of choice lead to different effects. In light of this, the current study proposes a differentiation between two types of choice, that are typical in online (health) messages: physical choice (i.e., physically giving the individuals the opportunity to make a choice regarding the message they are exposed to, for example by clicking on a hyperlink) and verbal choice (i.e., emphasizing to the message recipients that they can choose what they want to read in the message). Therefore, the first aim of this study is to explore whether different types of choice have different persuasive effects on message recipients.
A suggested underlying process which explains the effects of choice on persuasive outcomes is through influencing autonomous motivation. Specifically, choice increases autonomous motivation to engage in a behavior, because choice supports the feeling of autonomy in the message recipient (Weinstein & Ryan, 2010). Subsequently, increased autonomous motivation leads to a more positive attitude and intention towards the behavior (Hagger, Chatzisarantis, & Harris, 2006). Next to autonomous motivation, decreased reactance has also been suggested to explain the persuasive effects of providing choice. By receiving choice, individuals experience an increased freedom of decision, which results into less reactance against the message (Burgoon, Alvara, Grandpre, & Voulodakis, 2002; Shen, 2015).
Subsequently, decreased reactance allows for more positive attitudes and intentions to engage in the behavior (Ringold, 2002). As it is not yet clear which of these two mechanisms is stronger in the case of providing choice (Patall et al., 2008), the second aim of this study is to shed light on
CHOICE AS A PERSUASIVE STRATEGY 5 the underlying processes through which choice has an effect on persuasive outcomes of health messages.
A factor that appears to influence the effects of choice as an autonomy supportive message feature (i.e., supporting the freedom of decision of the message recipient) is need for autonomy (Smit, Linn & van Weert, 2015). Need for autonomy stems from the
self-determination theory (SDT; Deci & Ryan, 2000) and refers to the psychological need of people to behave based on their own choices and decisions. Resnicow et al. (2014) suggested that need for autonomy influences the persuasive effects of autonomy supportive language style in a health message. However, there is no empirical evidence yet regarding the influence of need for
autonomy on the effects of providing choice, as an autonomy supportive message feature. To fill this research gap, the third aim of this study is to look into the moderating effects of need for autonomy on the effects of choice on attitude and behavioral intention.
Theoretical framework Choice
Until about two decades ago, the concept of choice has seldom been researched in persuasion studies. This is partly because traditional media, such as TV and newspapers, which were dominant at that time, provided a rather low level of control of the information flow (e.g., choice) for their users (Liu & Schrum, 2002). However, with the increasing dominance of electronic and online communication, the potential for providing choice in a message increased exponentially (Ariely, 2000; Hoffman & Novak, 1996; Rodgers & Thorson, 2000), and more studies started to investigate the effects of choice on user experience and persuasion (Patall et al., 2008).
CHOICE AS A PERSUASIVE STRATEGY 6 A relatively large amount of studies show that choice has a positive impact on outcomes such as increased sense of personal control (Taylor, 1989), increased liking for a task
(Zuckerman, Porac, Lathin, & Smithz, 1978), or higher relevance of the content (Payne, Bettman, & Johnson, 1993). In the field of marketing communication, Schlosser and Shavitt (2009) showed that consumers give more positive brand and product evaluations, and have an increased purchase intention, when they are allowed to choose the product information that they receive on an online page. Moreover, Fransen et al., (2015) argue that offering choice is a way to neutralize consumers' motivation to adopt resistance strategies against the message, such as attitude bolstering, social validation and self-assertion. Yet, some scholars argue that choice has negative effects on persuasive outcomes (Flowerday & Schraw, 2003; Muraven & Baumeister, 2000; Schwartz, 2000). For example, Schlosser and Shavitt (2009) also found that when choice is emphasized by the company in a marketing message, this can be perceived as a persuasive strategy by the reader, thus leading to backfiring persuasive effects. Other scholars found null effects of choice on the persuasiveness of messages (Parker & Lepper, 1992; Reeve et al., 2003).
In order to explain these mixed results, it has been suggested that different types of choice should be taken into account in research (Moller, Deci & Ryan, 2006; Williams, 1998). This idea is particularly relevant in the context of online (health) communication, which provides a large amount of multimedia content formats, out of which the user can choose what he/she wants to be exposed to (Ariely, 2000). Schlosser and Shavitt (2009) used in their study on the persuasiveness of online choice in a marketing message two intertwined types of choice: 1) making a verbal suggestion to the reader that he/she can choose what to read in the message, and 2) including hyperlinks in the message on which the reader can choose to click, depending on his/her interests. The authors found some interesting results regarding different combinations of
CHOICE AS A PERSUASIVE STRATEGY 7 these types of choice. For example, when choice through hyperlinks was strongly emphasized by the marketeer in a verbal way, it led to less favorable product evaluations, as compared to when the presence of hyperlinks was only briefly mentioned in the message. However, to our
knowledge, the separate effects of these two abovementioned types of choice have not yet been assessed in research. This, in spite of the fact that they can be commonly met in online
environments (Baggio & Corigliano, 2009) and, as shown, may lead to different persuasive effects. To fill this research gap, the current study focuses on exploring the difference between these two types of choice, which are labeled verbal choice and physical choice, respectively. To illustrate, a verbal choice is present on a webpage when it is verbally suggested to the users that they may choose what they want to read on the page. A physical choice, on the other hand, is present when the layout of the webpage is built in such a way that it allows the users to decide (i.e., choose) the content to which they are exposed. This can be done, for example, by inserting the texts under drop-down menus or hyperlinks, so that the users read them only they are
interested – and consequently click on the hyperlinks.
Based on the fact that generally more empirical studies found positive (vs. negative) persuasive effects of choice (Patall et al., 2008), in this study it is expected that when either physical choice or verbal choice is included, the online health message will lead to a more positive attitude and behavioral intention, as compared to when no choice is included. Moreover, it is expected that when both types of choice are present in a message, the general persuasive effect will increase, as compared to when a single type of choice is offered, as well as when no choice is offered. This latter effect is proposed because it is expected that when the choice is conveyed using more than one mode, it will become more salient (Mayer & Moreno, 2003). Additionally, as the theory of planned behavior (Ajzen, 1991) posits that attitude is a predictor of
CHOICE AS A PERSUASIVE STRATEGY 8 behavioral intention and this was empirically proven in a multitude of studies (for reviews see Godin & Kok, 1996; Montano & Kasprzyk, 2015), it is expected that attitude will mediate the effects of choice (physical and verbal) on intention. Therefore, we pose the following hypothesis:
H1: Being exposed to a) physical choice (compared to no choice) and b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only verbal choice, only physical choice, or no choice) in a health message leads to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior.
Autonomous motivation
The SDT posits that there is a difference between autonomous motivation and controlled motivation to engage in behaviors (Deci & Ryan, 2000). Autonomous motivation is experienced when individuals want to engage in a behavior based on their sense of volition and because the behavior is of personal importance to them (Levesque, Williams, Elliot, Pickering, Bodenhamer, & Finley, 2006; Williams, Gagne, Ryan, & Deci, 2002). With an autonomous motivation, individuals experience an “internal perceived locus of causality” (Weinstein & Ryan, 2000, p. 223) to engage in a behavior. For example, an autonomous motivation to increase physical activity is to personally find the outcomes of this behavior important, or to simply enjoy engaging in it (Levesque et al., 2006). Controlled motivation, on the contrary, is a result of coercive external forces, that are not internalized by the individual. For example, being
motivated to increase physical activity in order to avoid feelings of guilt or in order to follow the physician‘s recommendations (Ng et al., 2012).
The SDT proposes that autonomous motivation is a stronger predictor of behavior, compared to controlled motivation, because when someone genuinely appreciates the outcomes of the behavior, he/she is more likely to engage in and maintain the behavior (Hagger et al.,
CHOICE AS A PERSUASIVE STRATEGY 9
2006).This is also supported by empirical evidence, for example in the context of physical activity (Chatzisarantis, Hagger, Biddle, Smith, & Wang, 2003), weight loss (Silva et al., 2011; Williams, Grow, Freedman, Ryan, & Deci, 1996) or smoking cessation (Williams et al., 2002). According to the SDT, autonomous motivation is increased when messages are autonomy supportive (Deci & Ryan, 2000) because individuals perceive an increased sense of autonomy and freedom in their decisions (Teixeira, Patrick, & Mata, 2011). As choice is an autonomy supportive message feature (Williams, Cox, Kouides, & Deci, 1999), providing choice in a message should therefore lead to an increased feeling of autonomous motivation. Subsequently, as argued above, this feeling of autonomous motivation should lead to a more positive attitude, which then results in a stronger behavioral intention. Thus, it is expected that the effect of choice on attitude, and subsequently intention, is mediated through increased autonomous motivation. In line with this it is hypothesized that:
H2: Being exposed to a) physical choice (compared to no choice), b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only verbal choice, only
physical choice, or no choice) in a health message leads to more autonomous motivation, and subsequently to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior.
Reactance
Reactance is the motivational state aimed at “reestablishing free behaviors that have been eliminated or are threatened with elimination” (Burgoon et al., 2002, p. 216). The theory of reactance (Brehm, 1966) posits that when a person is exposed to a persuasion attempt and feels a potential threat to his/her autonomy of making decisions, a state of reactance will occur. This is reflected in a desire to reestablish the threatened freedom, by holding to the threatened attitude or
CHOICE AS A PERSUASIVE STRATEGY 10 engaging in the threatened behavior. Dillard and Shen (2005) proposed that reactance consists of two measurable dimensions: affective reactance and cognitive reactance. Affective reactance refers to the anger experienced in response to a freedom-limiting message, while cognitive reactance refers to negative thoughts (i.e., counterarguing) of the message recipient against a message (Rains & Turner, 2007).
Messages with a persuasive intent, such as health messages, often lead to reactance (Knowles & Linn, 2004). This is because messages that advocate that people hold/change specific beliefs and/or perform/refrain from certain behaviors, are often perceived as threats to autonomy (Jenkins & Dragojevic, 2013). A way to reduce this reactance is by including choice options in a message (Fransen et al., 2015). Knowles and Linn (2004) argue that by being offered a choice, a person may focus his/her reactance on one or more of the alternatives, and, thus, is more likely to actually become engaged in the remaining alternative(s). By rejecting some of the choice alternatives in the message, the person feels an increased sense of autonomy and of being "critical and discerning” (p. 141), which lowers the overall reactance towards the whole
message. Several studies have shown that a decreased reactance has a positive effect on the attitude and intention towards the health behavior (Miller, Lane, Deatrick, Young, & Potts, 2007; Moyer-Guse & Nabi, 2010; Rains & Turner, 2007). This is because a lower level of reactance allows for more positive beliefs regarding the message and less counterarguments that can elicit negative beliefs (Burgoon et al, 2002). In light of this, it can be expected that decreased
reactance (both affective and cognitive) mediates the effect of choice on attitude and, subsequently, intention. Therefore, the following hypothesis is posited:
H3: Being exposed to a) physical choice (compared to no choice), b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only verbal choice, only
CHOICE AS A PERSUASIVE STRATEGY 11
physical choice, or no choice) in a health message leads to a decreased ((i) affective and (ii) cognitive) reactance, and subsequently to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior.
Need for autonomy
The SDT posits that every person has a basic need for autonomy, that is, to feel
autonomous and to like holding control over his/her actions (Deci & Ryan, 2000). The level of need for autonomy may differ among individuals (Deci & Ryan, 1985). People with a higher need for autonomy tend to engage in behaviors based on their own personal interests and values. People with a lower need for autonomy have a stronger controlled orientation, and when
engaging in behaviors, focus on directives from other people (Ng et al, 2012). Smit et al. (2015) argue that need for autonomy may moderate the persuasiveness of autonomy supportive
messages. Two studies have previously tested this moderation effect in an offline tailored
communication setting, in the context of fruit consumption (Resnicow et al., 2008) and colorectal cancer screening (Resnicow et al., 2014). Both studies showed that autonomy supportive
messages (i.e., using non-controlling language styles) were more effective for people with a higher (vs. lower) need for autonomy. However, this was not yet investigated with choice as an autonomy supportive message feature. In line with these results, it is expected that need for autonomy moderates the effects of choice on attitude and intention, in such a way that a message with choice will be more persuasive for people with a higher need for autonomy, compared to a lower need for autonomy.
Moreover, as already mentioned, the SDT posits that autonomous motivation appears in a person when his/her need for autonomy is satisfied. Thus, because choice satisfies one’s need to take his/her own decisions (i.e., need for autonomy; Deci & Ryan, 2000), a stronger autonomous
CHOICE AS A PERSUASIVE STRATEGY 12 motivation should be experienced by those people with a higher need for autonomy, compared to a lower need for autonomy. Also, Burgoon et al. (2002) argue that reactance, experienced as a threat to the freedom of decision, is positively influenced by one’s need for autonomy.
Therefore, the effects of providing choice on reducing reactance should be more visible among individuals with a higher (vs. lower) need for autonomy. In light of these arguments, as shown in Figure 1, it is expected that all of the effects described in the previous hypotheses are more likely to happen in those people with a higher need for autonomy, compared to a lower need for
autonomy. Thus, the following hypothesis proposes:
H4: Need for autonomy moderates the effects mentioned in H1-3, so that those effects are stronger for individuals with a higher need for autonomy, compared to a lower need for autonomy. physical choice (absent vs. present) verbal choice (absent vs. present) attitude affective reactance cognitive reactance intention autonomous motivation
need for autonomy
CHOICE AS A PERSUASIVE STRATEGY 13 Method
Participants
Participants (Dutch speakers) were recruited through a message in the bimonthly newsletter sent by the Dutch branch of the World Cancer Research Fund (WCRF; sent to approximately 2000 Dutch people), as well as through the researcher’s personal network. The sample consisted of 218 participants (69% female) between the age of 18 and 84 (M = 37.77, SD = 16.27). As a compensation for their participation, respondents were offered a WCRF
wristband, and the opportunity to enter a prize draw in which they could win one of five €50 gift vouchers for an online Dutch retailer.
Experimental design, procedure and materials
Participants were randomly assigned to one of the four conditions in an online 2 (physical choice: present vs. absent) X 2 (verbal choice: present vs. absent) between-subjects factorial experiment. The aim was to assess the main and interaction effects of the two types of choice (independent variables) and of need for autonomy (moderator) on reactance and autonomous motivation (mediators), as well as their indirect effects on attitude and behavioral intention (dependent variables).
When participants opened the online questionnaire, they were first presented with general information about the study and the informed consent form, with which they had to agree in order to proceed with the study. Next, they had to answer general questions regarding their gender, age and current level of physical activity level. Afterwards, participants’ need for
autonomy was assessed. Thereafter, they were randomly directed to one of the four experimental stimuli. These were four website pages created for the purpose of the study, in collaboration with and based on the official website of the Dutch branch of the WCRF (see Appendix A). Each page
CHOICE AS A PERSUASIVE STRATEGY 14 contained the same information (approximately 600 words) about the importance of physical activity for preventing cancer. The information included an introductory text about the
importance of physical activity, followed by four content sections: 1) the role of physical activity in preventing cancer, 2) what types of physical activity count towards preventing cancer, 3) tips to increase physical activity and 4) sedentary lifestyle and cancer. The information depicted in the stimuli is used on the existing WCRF website, and contained scientifically proven facts about the importance of physical activity for cancer prevention, written in an accessible language. After reading the webpages, participants were redirected to the questionnaire where manipulation checks were conducted. Next, participants answered questions about their reactance against the message (affective and cognitive), autonomous motivation to increase physical activity, as well as attitude towards and intention to increase physical activity. Lastly, participants could
voluntarily provide their contact details for receiving a WCRF wristband and entering the prize draw.
A pretest of the entire survey and manipulations, with 12 participants, was performed, in order to make sure the manipulations and randomization worked out well, and the survey items were clear. Based on the pretest, the physical choice manipulation was slightly adapted, by excluding a verbal element from it, so that it could be better distinguished from the verbal choice manipulation. Overall, the pretest indicated that the manipulations and randomization worked out as intended, and the survey questions were clear to the participants.
Manipulations
The presence of verbal choice was manipulated by including a sentence in the
introductory text on the webpage: You can choose for yourself about which topics you want to
CHOICE AS A PERSUASIVE STRATEGY 15 give respondents a feeling of choice and freedom to read the content that is line with their
personal preferences. The presence of physical choice was manipulated by including hyperlinks that had to be clicked on, in order to read each of the four abovementioned content sections (e.g., the role of physical activity in preventing cancer). If the participants wanted to read about a certain content section, they could click on its respective hyperlink, and a new page would open with that specific information. Afterwards, participants could return to the page listing the four content sections, and either click on another hyperlink with information, or go back to the main questionnaire. In the conditions without physical choice, all four content sections were included on the same page in a static manner (i.e., not under hyperlinks). All content (i.e., text and visual elements) was kept identical across conditions.
Manipulation checks
The verbal choice manipulation was checked for by asking participants to answer yes or
no to the statement: It was emphasized in the text that I could choose about which topics I wanted to read, depending on my interests and lifestyle. The physical choice manipulation was
checked for by asking participants to answer yes or no to the statement: I could choose myself
about which topics I wanted to read by clicking on those topics.
Measures
Need for autonomy. To measure need for autonomy, the Health Causality Orientations Scale (HCOS) was used, adapted based on the General Causality Orientations Scale (Deci & Ryan, 1985). HCOS measures autonomy and controlled motivational orientations in health contexts. In this study, participants read five health scenarios, for example that they would consider improving their health by increasing their physical activity (for complete scale see Appendix C). Then, for each scenario, they were asked to indicate how likely they would be to:
CHOICE AS A PERSUASIVE STRATEGY 16 1) seek an expert’s advice and 2) rely on their own opinions. The answer options ranged from
very unlikely (1) to very likely (7). The final score for need for autonomy (M = 1.34, SD = 1.55)
was calculated by subtracting the need for expert advice score (M = 4.15, SD = 1.14, Cronbach’s α =.73) from the need for relying on own opinion score (M = 5.50, SD = .92, Cronbach’s α =.73).
Reactance. Reactance was measured by separately assessing affective reactance (i.e., negative affect) and cognitive reactance (i.e., counterarguing), as suggested by Dillard and Shen (2005). Negative affect was measured with four Likert scale items, based on Dillard and Peck (2000). The items asked whether, while reading the webpage, the participant felt irritated, angry,
annoyed, and aggravated. The answer options ranged from very unlikely (1), to very likely (7;
Cronbach’s α = .94). The affective reactance score was computed as the average of the four items (M = 1.57, SD = 1.88). A higher score indicated a stronger affective reactance. Cognitive reactance was measured with four Likert scale items which assessed counterarguing, based on Van Reijmersdal et al. (2016). The items asked whether, while reading the webpage, participants
contested, refuted, doubted and countered the information. The answer options ranged from very unlikely (1) to very likely (7; Cronbach’s α = .86). The cognitive reactance score was computed
as the average of the four items (M = 2.17, SD = 1.19). A higher score indicated a higher level of cognitive reactance.
Autonomous motivation. Autonomous motivation was measured using six Likert scale items from the Treatment Self-Regulation Questionnaire (TSRQ; Levesque et al., 2006). The items referred to the statement If I were to increase my physical activity, I would do it because, and were, for example, I want to take responsibility for my own health or I think personally that
increasing physical activity is the best for my health (for complete scale see Appendix C). The
CHOICE AS A PERSUASIVE STRATEGY 17 autonomous motivation score was computed based on the average of the six items (M = 5.85, SD = 1.14). A higher score indicated a higher autonomous motivation to increase physical activity.
Attitude. Attitude towards increasing physical activity was measured on a seven-point scale based on Moorman and van den Putte (2008). The scale included four semantic differential items: bad (1) – good (7), unhealthy (1) – healthy (7), unwise (1) – wise (7), and unpleasant (1) –
pleasant (7; Cronbach’s α = .89). The attitude score was calculated as the average of the four
items (M = 6.08, SD = 1.23). A higher score indicated a more positive attitude towards
increasing physical activity.
Intention. Intention to increase physical activity was measured with three seven-point Likert scale items, adapted from De Bruijn and Rhodes (2011). The items asked whether the participant has the intention to increase physical activity, plans to increase physical activity and
probably will increase physical activity in the next four weeks. The answer options ranged from very unlikely (1) to very likely (7; Cronbach’s α = .95). The intention score was computed based
on the average of the three items (M = 4.47, SD = 1.78). A higher score indicated a higher intention to increase physical activity.
Control variables
Age (in years) and gender (i.e., male, female) were each assessed with one question in the survey and used as control variables in this study. Also, the level of physical activity was
measured as a control variable, using four items adapted from the International Physical Activity Questionnaire (IPAQ; Craig et al., 2003). Two items measured how many days per week
participants performed vigorous (e.g., running, playing tennis) and moderate (e.g., biking, working in the garden) physical activities, respectively (for complete scales see Appendix C). Answer options ranged from less than one day per week (0) to seven days per week (7). Two
CHOICE AS A PERSUASIVE STRATEGY 18 items measured how many hours on average per day participants performed vigorous and
moderate physical activities, respectively. The moderate physical activity score was computed by multiplying the amount of days per week the activity is practiced with the amount of hours per day the activity is practiced (M = 7.59, SD = 11.30). The vigorous physical activities was computed in the same way, and at the end, multiplied with 2 (M = 5.62, SD = 9.17), because, according to the IPAQ, vigorous physical activity weights twice as much as moderate physical activity in the general physical activity score. Then, the scores for vigorous physical activity and moderate physical activity were summed-up, to compute the general physical activity score (M = 13.22, SD = 15.74). A higher score indicated a higher rate of physical activity.
Results Randomization checks
To verify that the randomization of participants into conditions worked out as intended, a chi-square test was conducted, between the gender variable and the four experimental conditions. None of the associations were significant. Additionally, two factorial ANOVAs were performed with age and physical activity as dependent variables, respectively, and verbal choice and physical choice as independent variables. Again, no significant results were found. Thus, in terms of these three variables, the randomization of participants was successful. Therefore, none of the control variables were further included in the main analyses.
Manipulation check
To check whether the manipulation of choice in the messages worked out as intended, two chi-square tests were performed, with the physical/verbal choice variable (present/absent) and the respective manipulation check variable (yes/no). Results revealed significant effects in the expected direction, both for physical choice (χ2 (1) = 52.02, p < .001), and for verbal choice
CHOICE AS A PERSUASIVE STRATEGY 19 (χ2 (1) = 8.25, p = .005). Participants exposed to the physical choice more often answered that they could choose what they wanted to read by clicking on the topics (n = 105), than that they could not choose by clicking (n = 5). Similarly, participants who were exposed to the verbal choice message more often answered that it was emphasized in the text that they could choose what they wanted to read (n = 78), compared to answering that it was not emphasized that they could choose (n = 31). Thus, the choice manipulations were noticed as intended.
Model specification
The conceptual model specified in Figure 1 was tested using path analysis in structural equation modeling. This technique was used because it is a powerful statistical test, which allows for testing the hypothesized moderated serial mediation at once, thus allowing for increased validity of the results (Kline, 2011). The model was specified in the program AMOS 21 and indirect effects were tested with bootstrapping. Because of the sample size, which was acceptable for a path analysis, but less recommendable for structural regression (Wolf, Harrington,Clark,& Miller, 2013), all variables in the model were included as observed (i.e., computed in SPSS, as described in the Methods section), rather than latent variables with indicators. All the hypothesized relationships were added to the model. Independent variables (i.e., verbal choice and physical choice), the moderator (i.e., need for autonomy) and the
interaction variables between need for autonomy and each of the choice types, as well the three-way interaction between need for autonomy and the two types of choice, were allowed to correlate freely among themselves. Also, correlations were added between the error terms of the mediators (i.e., autonomous motivation, affective reactance and cognitive reactance). Zero-order correlations between all variables entered in the model can be found in Appendix D. The model fit was evaluated with three indicators: chi-square (should be not significant), RMSEA (should
CHOICE AS A PERSUASIVE STRATEGY 20 be smaller than .05) and CFI (should be higher than .95). The model was identified and had a good model fit (χ2 (3) = 1.53, p = .675, RMSEA = .00 and CFI = 1), allowing to proceed to hypothesis testing.
Hypotheses testing
Hypothesis 1 predicted that being exposed to a) physical (compared to no choice) and b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only physical choice, only verbal choice, or no choice) in a health message leads to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior. The analyses revealed no significant effects of physical choice on attitude and intention, thus hypothesis 1a is rejected. Moreover, no effect was found from verbal choice on attitude. However, a significant negative direct effect was found from verbal choice on intention (b* = -.21, p = .013, see Figure 3). This means that when verbal choice was present on the webpage, the readers were less inclined to increase their physical activity. This is contrary to what was hypothesized, thus hypothesis 1b is not accepted. Also, no interaction effect between the two types of choice was found on attitude. However, a marginally significant interaction effect was found between the two types of choice on intention (b* = .21, p = .060). Further plot investigations of this effect (see Figure 2) suggested that when respondents were exposed to verbal choice in a message, their intention to increase physical activity became lower, however, this happened only when physical choice was not present in the message. When both physical and verbal choice were included in the message, the intention to increase physical activity went slightly up, compared to when only physical choice or only verbal choice was present in the message. Thus, hypothesis 1c is partly accepted. The presence of both types of choices in a message had a positive effect on intention, compared to when only verbal choice or only physical
CHOICE AS A PERSUASIVE STRATEGY 21 choice were used. However, the intention remained the highest when no type of choice was used in message. The found main and interaction effects of verbal and physical choice on intention were not mediated through attitude.
Figure 2. Interaction effects between verbal choice and physical choice, on intention
Hypothesis 2 posited that being exposed to a) physical choice (compared to no choice), b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only verbal choice, only physical choice, or no choice) in a health message leads to more autonomous motivation, and subsequently to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior. With regard to this hypothesis, no significant effects were found from physical choice, verbal choice or the interaction between them on autonomous motivation. Thus, hypotheses 2a-c have to be rejected. However, as depicted in Figure 3, a significant relationship was found from autonomous motivation on
CHOICE AS A PERSUASIVE STRATEGY 22 attitude (b* = .37, p < .001). Thus, a higher autonomous motivation led to a more positive
attitude1.
Hypothesis 3 predicted that being exposed to a) physical choice (compared to no choice), b) verbal choice (compared to no choice) and c) both verbal and physical choice (compared to only verbal choice, only physical choice, or no choice) in a health message leads to decreased ((i) affective and (ii) cognitive) reactance and subsequently to a more positive attitude towards the health behavior, which results in a higher intention to perform the health behavior2. With regard to affective reactance, no significant results were found. Neither verbal choice, nor physical choice, nor the interaction between these two, influenced the level of affective reactance towards the webpage. Thus, hypotheses 3a(i), 3b(i) and 3c(i) are rejected.
Also, no effect was found from physical choice on cognitive reactance. Thus, hypothesis 3a(ii) is rejected. However, as hypothesized, cognitive reactance was significantly predicted by verbal choice (b* = -.26, p = .044). Including verbal choice in a message decreased cognitive reactance against the message. Moreover, cognitive reactance was found to have a negative effect on attitude (b* = -.16, p = .013). Bootstrapping revealed a significant indirect effect from cognitive reactance on intention (b* = -.036, p = .034), while no direct effect was found in this regard. This means that attitude fully mediated the effect of cognitive reactance on the intention to increase physical activity. In light of these results, hypothesis 3b(ii) is confirmed. The
presence of verbal choice in a message decreased cognitive reactance against the message, which
1
An equivalent path model was tested, with causal effects from both cognitive reactance and affective reactance on autonomous motivation (instead of the error correlations, from the main model). This theoretical relationship was suggested in Pavey and Sparks (2009), where it is argued that reactance may also precede, instead of co-occur with, autonomous motivation. Results revealed a significant negative relationship from cognitive reactance on autonomous motivation (b* =.18, p = .012). This means that the mechanism from the choice on intention, through cognitive reactance and attitude, may be additionally explained through autonomous motivation.
2 In addition to the analyses presented, this hypothesis was tested with the reactance variable computed as an average of affective
reactance and cognitive reactance, as suggested by Dillard and Shen (2005). This depicted neither significant results of the independent variable nor of the need for autonomy, nor their interaction, on reactance. Also, no significant effects were found from reactance on attitude and intention.
CHOICE AS A PERSUASIVE STRATEGY 23 led to a more positive attitude towards the behavior, and subsequently resulted in a higher
intention to perform the behavior. No interaction effect between the two types of choice was found on cognitive reactance, so hypothesis 3c(ii) is rejected.
Hypothesis 4 predicted that need for autonomy moderates the effects mentioned in H1-3, so that those effects are stronger for individuals with a higher need for autonomy, compared to a lower need for autonomy. Results revealed that need for autonomy moderated the effect of verbal choice on cognitive reactance (b* = -.44, p = .006). Further exploring the results (see Figure 4), it was revealed that, as hypothesized, verbal choice had a stronger negative effect on cognitive reactance for people with a higher (vs. lower) need for autonomy. In other words, when verbal choice was present in the message, cognitive reactance decreased more in people with a higher need for autonomy, compared to people with a lower need for autonomy. This confirms hypothesis 4 with respect to the relationship between verbal choice and cognitive reactance. As no other moderation effects of need for autonomy were revealed, hypothesis 4 is rejected
regarding the effects posited in the rest of the hypotheses. All non-significant results of the tested effects can be found in Appendix B.
CHOICE AS A PERSUASIVE STRATEGY 24
physical choice
verbal choice
physical choice * verbal choice
physical choice * need for autonomy
verbal choice * need for autonomy
physical choice * verbal choice * need for autonomy need for autonomy
affective reactance
cognitive reactance attitude intention autonomous motivation -.21 .37 -.26 .21 .22 -.16 -.31 -.44
Figure 3. Significant results in the tested SEM model
Note. A continuous/dashed line represents a significant/marginally significant relationship, respectively. Numbers represent standardized effect sizes (b*).
Figure 4. Interaction effects between verbal choice and need for autonomy on cognitive
CHOICE AS A PERSUASIVE STRATEGY 25 Discussion
Discussion of the results and future research
The present study aimed to examine how providing different types of choice in an online health message influences the attitude towards and intention to increase physical activity, in the context of cancer prevention. Furthermore, it investigated the underlying processes of these effects, through autonomous motivation and reactance (affective and cognitive). Finally, the influence of need for autonomy on the persuasive effects of choice was assessed. To our knowledge, this study was the first to differentiate between two types of choice that can be encountered in online messages: verbal choice and physical choice.
Overall, a direct effect was found from verbal choice on intention. Contrary to the hypothesized expectations, providing verbal choice in the message led to a smaller intention to increase physical activity. However, marginally significant results indicated that this direct effect was only present when physical choice was not included in the message. On the opposite, when physical choice was part of the message, next to verbal choice, the intention slightly increased. Thus, providing only verbal choice by mentioning in an online message that the reader can choose what to read, without giving physical choice through the option to actually click and select different subjects for reading, may in fact have negative persuasive outcomes. This is in line with the results depicted in Schlosser and Shavitt (2009), which showed that when choice is emphasized by the message source (e.g., the marketeer) it can lead to backfiring persuasive outcomes. An explanation for this negative effect may be that because the choice is emphasized by the message source, it may be perceived as a persuasion tactic by the reader, and thus lead to resistance against the message (Friestad & Wright, 1994). Regarding the null effects found of physical choice on attitude and intention, an explanation may be that online readers are used to
CHOICE AS A PERSUASIVE STRATEGY 26 choose the content they read, for example by scanning or skimming it, even when hyperlinks are not present (Hewitt, Brett, & Peters, 2007; Rosenwald, 2014). So, even though they could not click on the topics they wanted to read about, participants could have still ignored the topics that they did not want to read, as they would do it when physical choice was included in the message. An indirect path was found from verbal choice on intention, through cognitive reactance and attitude. Interestingly, this was in the opposite direction, compared to the direct effect found of verbal choice on intention. Namely, providing verbal choice led to less cognitive reactance, which resulted in a more positive attitude and, subsequently, in a higher intention to increase physical activity. In other words, the negative direct effect from verbal choice on intention was partly diminished through less cognitive reactance. The negative effect on cognitive reactance is in line with the reactance theory (Brehm, 1966), which suggests that messages supporting
autonomy in an individual (e.g., through providing choice options) decrease reactance (including counterarguing), thus making the message more persuasive. Yet, the negative direct effect from verbal choice on intention was more pronounced, compared to the indirect path through
cognitive reactance and attitude. A reason for this may be that, in general, cognitive reactance against the benefits of increasing physical activity was rather low (M = 2.11 out of 7, SD = 1.19). This might have led to a floor effect, that statistically did not allow for a stronger positive
indirect effect from verbal choice on intention, through cognitive reactance (Field, 2013). Further research may therefore benefit from looking at other types of behavior, for example addictive behaviors such as binge drinking or smoking, which typically lead to stronger cognitive
reactance from their adopters, compared to physical activity (Bensley & Wu, 1991; Erceg-Hurn & Steed, 2011). This may allow for stronger effects of choice through cognitive reactance, than the ones found in this study.
CHOICE AS A PERSUASIVE STRATEGY 27 Regarding the need for autonomy as a moderator, results indicate that this variable indeed moderated the effect of providing verbal choice on cognitive reactance. As expected, including verbal choice in a message had a stronger, negative effect on cognitive reactance for people with a higher (vs. lower) need for autonomy. This is in line with previous empirical studies that have suggested that messages with autonomy supportive features are more persuasive for people who have a higher need for autonomy, compared to a lower need for autonomy (Resnicow et al., 2008; Resnicow et al., 2014). However, need for autonomy did not moderate the effects of physical choice. This is probably because, as mentioned above, both in messages with and without physical choice, participants could have selected the content they wanted to read (e.g., by scanning or skimming the page). Thus, their need for autonomy could be satisfied, irrespective of the presence or absence of physical choice. Future studies may control for this by using, for example, eye tracking technology (Duchowski, 2007) in order to precisely assess the reading style of participants and find out whether choice is made even on pages without hyperlinks.
Contrary to the expectations, providing choice in the message had no effect on autonomous motivation. A reason for this may be that the choice manipulations were not
perceived as more autonomy supportive by respondents in this study, compared to messages
without choice. A two-way ANOVA test with perceived autonomy support as the dependent variable (adapted from the Virtual Care Climate Questionnaire; Smit, Dima, Immerzeel, van den Putte, & Williams, 2017) and the two choice variables as the independent variables portrayed no significant results (data not reported). Participants exposed to verbal choice, physical choice or both types of choice on the webpage did not perceive a higher autonomy support in the message, compared to participants that were not exposed to any choice. This may explain the lack of
CHOICE AS A PERSUASIVE STRATEGY 28 effects found on the perceived autonomous motivation to increase physical activity, as it is through increased perceived autonomy support that autonomous motivation is likely to be experienced (Levesque et al., 2006). Also, contrary to the expectations, providing choice did not have an effect on the affective reactance experienced while reading the message. An explanation may be that people typically agree with the fact that increasing physical activity has positive effects on health (US Sports Academy, 2008). Because of this, reading the messages from the current study might not have made them feel irritated. This may have led to a floor effect in the measurement of affective reactance as well (M = 1.57 out of 7, SD = 1.88), which did not allow for statistically significant effects to be found. Future studies may focus on behaviors that may lead to a higher affective reactance, for example detective behaviors (e.g., cancer screening), which involve more risk and may result in more negative affect (Latimer, Salovey & Rothman, 2007). Such behaviors may reveal stronger effects from autonomy supportive message features, such as choice, on decreasing affective reactance. It should be mentioned that checking
hypothesis 3 with only one general reactance variable (created as an average of affective reactance and cognitive reactance; Dillard & Shen, 2005), revealed no significant results either. An explanation may be that the low scores of affective reactance suppressed the significant effects of cognitive reactance. Thus, for future studies, especially regarding behaviors that may lead to low affective reactance (such as physical activity), it could be fruitful to disentangle reactance it in its two dimensions and analyze their separate effects.
Strengths and limitations
One of the strengths of this study is that, although experimental, ecologically valid stimuli were included, by using the existing website of the WCRF. This allowed for more
CHOICE AS A PERSUASIVE STRATEGY 29 testing of the model as a whole, through structural equation modelling, allowed for higher
validity of the results, compared to separate testing of the hypothesized effects, through regression analyses (Kline, 2011).
However, a first limitation of this study is that no behavioral measurement was
performed, which is often the preferred outcome in persuasive communication studies. However, in the context of physical activity – as well as other lifestyle associated behaviors – intention was shown to be a relevant predictor of behavior (for a review see Hagger, Chatzisarantis, & Biddle, 2002), based on which we consider behavioral intention a good proxy for behavior. A second
limitation resides in the fact the WCRF website used for the experimental stimuli included four buttons at the top, that could be clicked on, linking to different rubrics of the website. Due to technical reasons, these buttons could not be deactivated during our experiment. Although participants were instructed to only read the information about physical activity and then come back to the questionnaire, web analytics statistics (i.e., Google Analytics) showed that, on average, 35% of the visitors clicked on at least one other button on the website. This may have caused readers of the pages without physical choice to still perceive a certain extent of physical choice. The web analytics statistics do not reflect exclusively the sample included in our study, thus it is difficult to estimate the actual impact that those active buttons had on our results. Future studies should however pay increased attention to controlling the physical choice, especially in online environments.
Conclusion and contribution
To our knowledge, this study is the first one to look into different operationalizations of choice in an online health communication setting. The main conclusion is that verbal choice has a negative direct effect on the intention to increase physical activity, in the context of cancer
CHOICE AS A PERSUASIVE STRATEGY 30 prevention. However, importantly, this effect happens only when physical choice is not present in the message. It seems thus, that including choice in an online message has to be done with caution. If this choice is not sufficiently supported by the physical design of the text, through clickable hyperlinks, it can in fact lead to negative persuasive outcomes.
Also, part of the negative direct effect of verbal choice on intention is indirectly decreased through cognitive reactance and attitude. Moreover, the effect of verbal choice on cognitive reactance is stronger for those people with a higher need for autonomy, compared to lower need for autonomy. These results extend the current SDT literature and provide insights for practitioners in the field of online health communication. Practitioners should not only pay attention to how they include choice in their messages, but also to whom they address these messages. When the target group consists of people with a higher need for autonomy, providing choice has a stronger effect on decreasing their cognitive reactance, compared to people with a lower need for autonomy.
Acknowledgements
I would like to thank my supervisor dr. Eline Smit for her constructive feedback, guidance and constant support throughout writing this thesis. Also, I would like to thank the Wereld Kanker Onderzoek Fonds team and, in particular, Germund Daal, for the help offered in developing the experimental stimuli and data collection. Moreover, I would like to thank the Graduate School of Communication for the funding offered for data collection.
References
Ajzen, I. (1991). The theory of planned behavior. Organizational behavior and human decision
processes, 50(2), 179-211.
Ariely, D. (2000). Controlling the information flow: Effects on consumers' decision making and preferences. Journal of consumer research, 27(2), 233-248.
CHOICE AS A PERSUASIVE STRATEGY 31
Baggio, R., & Antonioli Corigliano, M. (2009). On the importance of hyperlinks: A network science approach. Information and Communication Technologies in Tourism 2009, 309-318.
Bensley, L. S., & Wu, R. (1991). The role of psychological reactance in drinking following alcohol prevention messages. Journal of Applied Social Psychology, 21(13), 1111-1124. Bettman, J. R. (1979). Memory factors in consumer choice: A review. The Journal of marketing,
37-53.
Brehm, J. W. (1966). A theory of psychological reactance. Academic Press New York
De Bruijn, G. J., & Rhodes, R. E. (2011). Exploring exercise behavior, intention and habit strength relationships. Scandinavian journal of medicine & science in sports, 21(3), 482-491.
Bryman, A. (2015). Social research methods. Oxford university press.
Burgoon, M., Alvara, E., Grandpre, J., & Voulodakis, M. (2002). Revisiting the theory of psychological reactance: Communicating threats to attitudinal freedom. In J.P. Dillard & M. Pfau (Eds.), The Persuasion Handbook: Developments in Theory and Practice. California, Sage Publications.
Chatzisarantis, N. L., Hagger, M. S., Biddle, S. J., Smith, B., & Wang, J. C. (2003). A meta- analysis of perceived locus of causality in exercise, sport, and physical education contexts. Journal of Sport and Exercise Psychology, 25(3), 284-306.
Chou, W. Y. S., Prestin, A., Lyons, C., & Wen, K. Y. (2013). Web 2.0 for health promotion: reviewing the current evidence. American journal of public health, 103(1), 9-18.
CHOICE AS A PERSUASIVE STRATEGY 32
& Oja, P. (2003). International physical activity questionnaire: 12-country reliability and validity. Medicine and science in sports and exercise, 35(8), 1381-1395.
Deci, E. L., & Ryan, R. M. (1980). Self-determination theory: When mind mediates behavior.
The Journal of Mind and Behavior, 33-43.
Deci, E. L., & Ryan, R. M. (1985). The general causality orientations scale: Self-determination in personality. Journal of research in personality, 19(2), 109-134.
Deci, E. L., & Ryan, R. M. (2000). The" what" and" why" of goal pursuits: Human needs and the self-determination of behavior. Psychological inquiry, 11(4), 227-268.
Dillard, J. P., & Peck, E. (2000). Affect and persuasion: Emotional responses to public service announcements. Communication Research, 27(4), 461-495.
Dillard, J. P., & Shen, L. (2005). On the nature of reactance and its role in persuasive health communication. Communication Monographs, 72(2), 144-168.
Duchowski, A. T. (2007). Eye tracking methodology. Theory and practice, 328.
Erceg-Hurn, D. M., & Steed, L. G. (2011). Does exposure to cigarette health warnings elicit psychological reactance in smokers?. Journal of Applied Social Psychology, 41(1), 219- 237.
Field, A. (2013). Discovering statistics using IBM SPSS statistics. Sage.
Flowerday, T., & Schraw, G. (2003). Effect of choice on cognitive and affective engagement.
The Journal of Educational Research, 96(4), 207-215.
Fransen, M. L., Verlegh, P. W., Kirmani, A., & Smit, E. G. (2015). A typology of consumer strategies for resisting advertising, and a review of mechanisms for countering them.
International Journal of Advertising, 34(1), 6-16.
CHOICE AS A PERSUASIVE STRATEGY 33
about the psychology of advertising. Journal of Consumer Research, 22(1), 62-74. Godin, G., & Kok, G. (1996). The theory of planned behavior: a review of its applications to
health-related behaviors. American journal of health promotion, 11(2), 87-98. Hagger, M. S., Chatzisarantis, N. L., & Harris, J. (2006). The process by which relative
autonomous motivation affects intentional behavior: Comparing effects across dieting and exercise behaviors. Motivation and Emotion, 30(4), 306-320.
Hewitt, J., Brett, C., & Peters, V. (2007). Scan rate: A new metric for the analysis of reading behaviors in asynchronous computer conferencing environments. The American Journal
of Distance Education, 21(4), 215-231.
Hoffman, D. L., & Novak, T. P. (1996). Marketing in hypermedia computer-mediated environments: Conceptual foundations. The Journal of Marketing, 50-68.
Jenkins, M., & Dragojevic, M. (2013). Explaining the process of resistance to persuasion: A politeness theory-based approach. Communication Research, 40(4), 559-590.
Kennedy, S., Goggin, K., & Nollen, N. (2004). Adherence to HIV medications: Utility of the theory of self-determination. Cognitive Therapy and Research, 28(5), 611-628.
Kline, R. B. (2011), Principles and Practice of Structural Equation Modeling (3rd Edition). New York: Guilford Press.
Knowles, E. & Linn, J.A. (2004), Approach-avoidance model of persuasion: Alpha and omega strategies for change. In E. Knowles & J.A. Linn, Resistance and Persuasion. Lawrence Erlbaum Associates Publishers.
Krebs, P., Prochaska, J. O., & Rossi, J. S. (2010). A meta-analysis of computer-tailored interventions for health behavior change. Preventive medicine, 51(3), 214-221. Latimer, A. E., Salovey, P., & Rothman, A. J. (2007). The effectiveness of gain-framed
CHOICE AS A PERSUASIVE STRATEGY 34
messages for encouraging disease prevention behavior: Is all hope lost?. Journal of
Health Communication, 12(7), 645-649.
Levesque, C. S., Williams, G. C., Elliot, D., Pickering, M. A., Bodenhamer, B., & Finley, P. J. (2006). Validating the theoretical structure of the Treatment Self-Regulation
Questionnaire (TSRQ) across three different health behaviors. Health education
research, 22(5), 691-702.
Liu, Y., & Shrum, L. J. (2002). What is interactivity and is it always such a good thing? Implications of definition, person, and situation for the influence of interactivity on advertising effectiveness. Journal of advertising, 31(4), 53-64.
Lustria, M. L. A., Noar, S. M., Cortese, J., Van Stee, S. K., Glueckauf, R. L., & Lee, J. (2013). A meta-analysis of web-delivered tailored health behavior change interventions. Journal of
health communication, 18(9), 1039-1069.
Maher, C. A., Lewis, L. K., Ferrar, K., Marshall, S., De Bourdeaudhuij, I., & Vandelanotte, C. (2014). Are health behavior change interventions that use online social networks effective? A systematic review. Journal of medical Internet research, 16(2):e40.
Mayer, R. E., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning.
Educational psychologist, 38(1), 43-52.
Miller, C. H., Lane, L. T., Deatrick, L. M., Young, A. M., & Potts, K. A. (2007). Psychological reactance and promotional health messages: The effects of controlling language, lexical concreteness, and the restoration of freedom. Human Communication Research, 33(2), 219-240.
Moller, A. C., Deci, E. L., & Ryan, R. M. (2006). Choice and ego-depletion: The moderating role of autonomy. Personality and Social Psychology Bulletin, 32(8), 1024-1036.
CHOICE AS A PERSUASIVE STRATEGY 35
Montano, D. E., & Kasprzyk, D. (2015). Theory of reasoned action, theory of planned behavior, and the integrated behavioral model. Health behavior: Theory, research and practice, 95-124.
Moorman, M., & van den Putte, B. (2008). The influence of message framing, intention to quit smoking, and nicotine dependence on the persuasiveness of smoking cessation messages.
Addictive behaviors, 33(10), 1267-1275.
Moyer-Gusé, E., & Nabi, R. L. (2010). Explaining the effects of narrative in an entertainment television program: Overcoming resistance to persuasion. Human Communication
Research, 36(1), 26-52.
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of limited resources: Does self-control resemble a muscle?. Psychological bulletin, 126, 247-259.
Ng, J. Y., Ntoumanis, N., Thøgersen-Ntoumani, C., Deci, E. L., Ryan, R. M., Duda, J. L., & Williams, G. C. (2012). Self-determination theory applied to health contexts: A meta-analysis. Perspectives on Psychological Science, 7(4), 325-340.
Nguyen, M. H., van Weert, J. C., Bol, N., Loos, E. F., Tytgat, K. M., van de Ven, A. W., & Smets, E. M. (2017). Tailoring the mode of information presentation: effects on younger and older adults' attention and recall of online information. Human Communication
Research, 43(1), 102-126.
Parker, L. E., & Lepper, M. R. (1992). Effects of fantasy contexts on children’s learning and motivation: Making learning more fun. Journal of Personality and Social Psychology, 62, 625–633.
CHOICE AS A PERSUASIVE STRATEGY 36 motivation and related outcomes: A meta-analysis of research findings. Psychological
Bulletin, 134, 270–300.
Payne, J. W., Bettman, J. R., & Johnson, E. J. (1993). The adaptive decision maker. Cambridge University Press.
Pavey, L., & Sparks, P. (2009). Reactance, autonomy and paths to persuasion: Examining perceptions of threats to freedom and informational value. Motivation and Emotion,
33(3), 277-290.
Rains, S. A., & Turner, M. M. (2007). Psychological reactance and persuasive health
communication: A test and extension of the intertwined model. Human Communication
Research, 33(2), 241-269.
Van Reijmersdal, E. A., Fransen, M. L., van Noort, G., Opree, S. J., Vandeberg, L., Reusch, S., & Boerman, S. C. (2016). Effects of disclosing sponsored content in blogs: How the use of resistance strategies mediates effects on persuasion. American Behavioral Scientist,
60(12), 1458-1474.
Resnicow, K., Davis, R. E., Zhang, G., Konkel, J., Strecher, V. J., Shaikh, A. R., et al. (2008). Tailoring a Fruit and Vegetable Intervention on Novel Motivational Constructs: Results of a Randomized Study. Annals of Behavioral Medicine, 35(2), 159–169.
Resnicow, K., Zhou, Y., Hawley, S., Jimbo, M., Ruffin, M. T., Davis, R. E., et al. (2014). Communication preference moderates the effect of a tailored intervention to increase colorectal cancer screening among African Americans. Patient Education and
Counseling, 97(3), 370–375.
Rimer, B. K., & Kreuter, M. W. (2006). Advancing tailored health communication: A persuasion & message effects perspective. Journal of Communication, 56, 184–201.
CHOICE AS A PERSUASIVE STRATEGY 37
Ringold, D. J. (2002). Boomerang effects in response to public health interventions: Some unintended consequences in the alcoholic beverage market. Journal of Consumer Policy,
25(1), 27-63.
Rodgers, S., & Thorson, E. (2000). The interactive advertising model: How users perceive and process online ads. Journal of interactive advertising, 1(1), 41-60.
Rosenwald, M. S. (2014). Serious reading takes a hit from online scanning and skimming, researchers say. The Washington Post, 6.
Ryan, R. M., & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American psychologist, 55(1), 68. Schlosser, A. E., & Shavitt, S. (2009). The effect of perceived message choice on persuasion.
Journal of Consumer Psychology, 19(3), 290-301.
Schwartz, B. (2000). Self-determination: The tyranny of freedom. American Psychologist, 55, 79–88.
Shen, L. (2015). Antecedents to psychological reactance: The impact of threat, message frame, and choice. Health communication, 30(10), 975-985.
Silva, M. N., Markland, D., Carraça, E. V., Vieira, P. N., Coutinho, S. R., Minderico, C. S., ... & Teixeira, P. J. (2011). Exercise autonomous motivation predicts 3-yr weight loss in women. Medicine & Science in Sports & Exercise, 43(4), 728-737.
Simoneau, H., & Bergeron, J. (2003). Factors affecting motivation during the first six weeks of treatment. Addictive behaviors, 28(7), 1219-1241.
Smit, E. S., Linn, A. J., & van Weert, J. C. (2015). Taking online computer-tailoring forward: The potentialof tailoring the message frame and delivery mode of online health
CHOICE AS A PERSUASIVE STRATEGY 38
Smit, E., Dima, A., Immerzeel, S., van den Putte, B., & Williams, G. (2016). The virtual care climate questionnaire: development and validation. European Health Psychologist, 18(S), 731.
Taylor, S. (1989). Positive illusions: Creative self-deception and the healthy mind. New York: Basic Books.
Teixeira, P. J., Patrick, H., & Mata, J. (2011). Why we eat what we eat: the role of autonomous motivation in eating behaviour regulation. Nutrition Bulletin, 36(1), 102-107.
US Sports Academy. (2008) How teens & adults really feel about physical activity and physical education: a survey conducted for NASPE. Retrieved on January 13, 2018, from
http://thesportjournal.org/article/how-teens-adults-feel-about-physical-activity-physical-education-a-survey-conducted-for-naspe/
Weinstein, N., & Ryan, R. M. (2010). When helping helps: autonomous motivation for prosocial behavior and its influence on well-being for the helper and recipient. Journal of
personality and social psychology, 98(2), 222-244.
Williams, G. C., Grow, V. M., Freedman, Z. R., Ryan, R. M., & Deci, E. L. (1996). Motivational predictors of weight loss and weight-loss maintenance. Journal of personality and social
psychology, 70(1), 115.
Williams, S. (1998). An organizational model of choice: A theoretical analysis differentiating choice, personal control, and self-determination. Genetic, Social & General Psychology Monographs, 124, 465–492.
Williams, G. C., Cox, E. M., Kouides, R., & Deci, E. L. (1999). Presenting the facts about smoking to adolescents: effects of an autonomy-supportive style. Archives of Pediatrics & Adolescent Medicine, 153(9), 959–964.
CHOICE AS A PERSUASIVE STRATEGY 39
Williams, G. G., Gagné, M., Ryan, R. M., & Deci, E. L. (2002). Facilitating autonomous motivation for smoking cessation. Health Psychology, 21(1), 40-50.
Williams, G. C., Niemiec, C. P., Patrick, H., Ryan, R. M., & Deci, E. L. (2009). The importance of supporting autonomy and perceived competence in facilitating long-term tobacco abstinence. Annals of Behavioral Medicine, 37(3), 315-324.
Wolf, E. J., Harrington, K. M., Clark, S. L., & Miller, M. W. (2013). Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety.
Educational and psychological measurement, 73(6), 913-934.
World Cancer Research Fund International (2017). Cancer preventability estimates. Retrieved on 20 September 2017, from wcrf.org/cancer-preventability-estimates
World Health Organization (2017). Cancer fact sheet. Retrieved from on January 20, 2018, from http://www.who.int/mediacentre/factsheets/fs297/en/
Zuckerman, M., Porac, J., Lathin, D., & Deci, E. L. (1978). On the importance of self-
determination for intrinsically-motivated behavior. Personality and Social Psychology
CHOICE AS A PERSUASIVE STRATEGY 40 Appendices
Appendix A Experimental Stimuli
