SELF-SELECTION OPTION OF INFORMATION PRESENTATION:
THE FUTURE OF ONLINE ADVERTISING?
Master’s Thesis
Isabella Ingelman-Sundberg (11139099) Graduate School of Communication Master’s Program Communication Science Supervised by dr. A. J. Linn and M. H. Nguyen
Abstract
In order to increase message persuasiveness, tailored health communication strategies have focused on adjusting message content to suit individual characteristics, as well as on tailoring the mode of information presentation. This study investigated the effect of a self-selection option of information presentation (compared to an absent self-selection option) in an online advertising context, on recall of information and attitude towards the ad. The study also explored whether this effect differed between older [>65] and younger [<65] participants. Results from a 2 (selection: present vs. absent) x 2 (age: younger vs. older) experimental design showed that self-selection did not improve recall. In contrast to what expected, younger participants in the control condition recalled more information compared to younger participants in the self-selection condition. Further analysis showed that within the control condition, the audio format worked best on recall (compared to the text and image formats). Moreover, self-selection showed to improve motivation of information processing which, in turn, improved recall. These findings suggest that audio formats could be an appropriate way of conveying information in online ads. Additionally, self-selection can improve recall via motivation to process information. Although more research is needed, this study provides useful suggestions to advertisers on how to improve online advertising.
Keywords: self-selection, tailoring, motivation, self-efficacy, recall, attitude, online ads, health communication, physiological ability, cognitive ability, information processing.
Introduction
In 2014, an American research indicated that 59 percent of older individuals [>65] were using the Internet, which was a six percent increase from the previous year (Pew Research
Center, 2014). This figure indicates that Internet use among older individuals is increasing. When considering the continuous shift from traditional services and print sources towards a greater and––sometimes––almost exclusive online presence, it is not surprising that people are drawn to use the Internet more often for information search, shopping or healthcare management
(Lippincott, 2004; Wagner, Hassanein, & Head, 2010).
Nevertheless, despite older individuals’ increase in Internet use over the years, they are still behind the younger generations when it comes to technological adoption (Pew Research Center, 2014). This group is very heterogeneous and there are important distinctions in their technological adoption patterns, Internet use, life-experiences as well as physiological differences (Woodhouse, Wynne, Baillie, James, & Rawlins, 1988). Vision and hearing impairments, for example, are among the most common physiological conditions affecting older individuals today (Reuben, Mui, Damesyn, Moore, & Greendale, 1999). Apart from physiological limitations, older individuals are often struggling with cognitive decline as they age (Salthouse, 2011). Hence, physiological and cognitive limitations could make it harder for older individuals to attend to, and process, information online.
Tailored communication strategies, in which information is adjusted to suit individual characteristics, have proven many times to be an effective way to enhance message
persuasiveness (Webb, Hendricks, & Brandon, 2007; Lustria et al., 2016), although, with
discrepancies in findings (Maslowska, Smit, & Van den Putte, 2011). Because individuals differ on how they process information (Ausburn & Ausburn, 1978), one way to improve tailored
communication strategies can be to base them on individuals’ self-selection of information presentation (i.e., letting the individuals choose whether they want to receive the information verbally, written or orally). This way, individuals can take into account their physiological and cognitive abilities to process a message (Smit, Linn, & Van Weert, 2015), which has shown to influence individuals’ choices of information modality (Wright et al., 2008). Individuals with limited vision, for example, might prefer to receive information in an audio-format, rather than reading a written message. Self-selection of information presentation based on personal cognitive and physiological limitations could, therefore, improve persuasive communication outcomes.
However, whether a self-selection option is an effective method to improve message persuasiveness could be explained by an individual’s motivation to process information (Rimer & Kreuter, 2006) and perceived self-efficacy. Persuasive outcomes such as the intention to perform a specific behaviour, attitude and recall have previously been explained by individuals’
motivation to process information and self-efficacy beliefs (Zhao, Seibert, & Hills, 2005;
MacInnis, Moorman, & Jaworski, 1991). If individuals are able to self-select a message based on their abilities to process the message, this could lead them to be more motivated and boost their confidence in processing the information better. This could improve persuasive communication outcomes.
Hence, to examine the effects of self-selection on recall of information and attitude towards an online ad, the following research question will be investigated: To what extent does self-selection of information presentation affect the persuasiveness of a message; is it explained by motivation and self-efficacy to process information and, does it depend on age?
This study will focus on self-selection of information presentation based on individuals’ preferences in receiving the message in an image (supported by text), text (only) or audio (only)
format. It will contribute to the existing literature on tailored communication and, in particular, on the effects of a self-selection option of online information presentation applied to online advertising.
Theoretical Framework Information Processing
How people process information highly depends on how that information is presented. According to the dual coding theory (DCT) information can be presented in a verbal or visual way and individuals process information through the verbal and the non-verbal channels (Paivio, 1971). The DCT suggests that the ability to recall information through multiple channels (the visual and verbal channels) rather than the verbal channel alone increases the possibilities of recalling information (Clark & Paivio, 1991). However, according to the redundancy effect (Kalyuga, Chandler, & Sweller, 1999) multimedia (printed text and spoken text) should never be presented together when they cover the same type of information. According to this principle, the visual and verbal processing channels have limited processing capacity and, thus, presenting words in one way (printed or spoken) is better than presenting it in two ways simultaneously (printed and spoken). This principle contradicts the DCT (Mayer, Heiser, & Lonn, 2001).
The fact that human beings can only process a limited amount of information in each channel at the same time (Mayer, 2014) can be referred to as the cognitive load theory (CLT). This theory explains that to best manage the working memory––to be able to preserve and process the same or new information (Salthouse & Babcock, 1991)––any unnecessary cognitive load should be minimized in order for the available resources to be used efficiently (Van Gerven, Paas, Van Merriënboer, & Schmidt, 2000). Also Mayer (2002), states that addressing the verbal and non-verbal channels simultaneously can cause overload in the visual channel if not combined
adequately. Thus, there are different theories on how to best manage processing and recall which are, however, contradicting each other. It is, thus, unclear which is the most appropriate
approach.
Several studies on the topic have shown mixed results. Information recall has shown to be higher for a word-only condition compared to a multimodal (word and picture) condition (Chen & Fu, 2003). A written format has also shown to be more effective in communicating complex information, compared to audio-visual or audio-only formats, because of its less distracting feature (Byrne & Curtis, 2000). On the other hand, other studies have shown that text with icons as well as audio-visual modalities increase recall compared to purely written formats (Morrow, Hier, Menard, & Leirer, 1998) for both younger and older individuals (Bol, Van Weert, De Haes, Loos, & Smets, 2015). Another research also indicated that working memory could be improved by combining audio and visual presentations (Mousavi, Low, & Sweller, 1995).
The differences in results might be due to the fact that individuals have different needs and preferences on how information is presented (written, spoken or visually), which suggests the need to introduce of a self-selection option in order to improve information processing.
Self-Selection of Information Presentation on Recall and Attitude
Rather than presenting a universal “one-fits-all” type of message, a self-selection option of information presentation can accommodate user diversity. This means that the difference in user preferences might be grounded on individuals’ available cognitive resources and
physiological abilities. For example, it has been shown that when poor readers and older individuals with sensory impairments are given the option to select their preferred mode for onscreen tasks, they have a tendency to choose audio formats, compared to pictures and text formats (Soroka et al., 2006; Wright et al., 2009).
Petty and Cacioppo’s (1986) elaboration likelihood model (ELM) of persuasion is a theory that explains that an adult can process information in two different ways; through the central route or the peripheral route. The central route represents high information processing, which is characterized by issue-relevant thinking, whilst the peripheral route is characterized by low elaboration processing where peripheral cues are guiding attitudes. Because tailored
messages are adapted to individuals in order to increase the perception of personal relevance in the message, central processing will take place, which will lead to persuasion (Noar, Benac, & Harris, 2007), such as better information recall (Muehling, Stoltman, & Grossbart, 1990) and positive attitude towards the ad. With regards to the latter, when elaboration is high (in central processing) the persuasive outcome will depend on the individual’s thought about the message and its quality of arguments (O’Keefe, 2008): if the individual has positive thoughts in high elaboration, this will influence the attitude towards it positively, and vice versa. Similarly to tailored message contents, by tailoring the way information is presented could increase the perception of message relevance––because individuals are able to consider and choose their preference based on their individual abilities. Thus, it can be expected that when a self-selection option is proposed to individuals, this will lead to central processing because they evaluated and thought about an option that was personally relevant to them, which will have a positive
influence on message attitude and recall.
Studies on tailoring the mode of information presentation have shown to have positive intentions for breast cancer screening (Jensen, King, Carcioppolo, & Davis, 2012) and positive effects on the satisfaction with the attractiveness and comprehensibility of a website (Nguyen, Smets, Bol, Loos, & Van Weert, submitted). Furthermore, the same way research on tailored message content has proven to have a positive effect on attitude towards online newsletters,
brands and ads (Maslowska, Smit, & Van den Putte, 2013, 2016), tailoring the mode of
information presentation (by adding a self-selection option) can be expected to have a positive effect on recall of information and attitudes towards an online ad.
H1: Self-selection of information presentation online (vs. no self-selection option) will lead to (a) a higher recall of information and (b) a more positive attitude towards the ad.
Self-Selection of Information Presentation and Age Difference on Recall and Attitude Two in five older individuals have physiological issues, which make it challenging for them to read online (Pew Research Center, 2014). In addition, cognitive decline is associated with aging and with previous health and lifestyle factors. For example, cognitive memory decline can be related to problems such as a lack of concentration and limited executive functions. This means that individuals with cognitive memory decline are unable to concentrate on several aspects of a task at the same time (Lindeboom & Weinstein, 2004). Online, older individuals are often required to focus on many elements at the same time as they are faced with large quantities of inadequately and complex information, which affects their attitude towards the Internet negatively––especially for less experienced computer users (Van Gerven et al., 2000; Morris, 1994). Therefore, it is suggested that older individuals will benefit the most by being able to choose their preferred format.
H2: The positive effect of self-selection of information presentation online (vs. no self-selection option) on (a) recall of information and (b) attitude towards the ad is stronger for older individuals than for younger individuals.
Motivation is a theoretical construct that has been frequently used to define differences in the strength and direction of behaviors. It results from a mixture of individual needs and desires (Humphreys & Revelle, 1984) and the more individuals are motivated, the more likely it is that they will perform a certain task or behaviour (Deci & Ryan, 2008). The role of motivation has not only shown to be an influential part in information processing within academic learning (Manolopoulou-Sergi, 2004): Previous research within health communication has also shown that the more individuals are motivated the more they acquire health related information (Moorman & Matulich, 1993) and recall online cancer information (Bol et al., submitted.). Also within
advertising, motivation, ability and opportunity to process brand information have shown to mediate the relationship between brand information processing and brand attitude and recall (MacInnis et al., 1991). These examples highlight the importance of motivation in human actions.
Furthermore, a recent research on tailored communication showed that when older individuals were able to choose their preferred mode of information presentation online, their attention increased, which resulted in higher information recall (Nguyen et al., 2017). Thus, it can be expected that when individuals are able to self-select the message format (text, audio or
image–supported by text), they will become more motivated to process the message, which in turn will lead to a more positive effect on recall and attitude towards the ad.
H3: Self-selection of information presentation online (vs. no self-selection option) on (a) recall of information and (b) attitude towards the ad is mediated by motivation to process online information.
Perceived Self-Efficacy of Information Processing
Self-efficacy is a main construct of the social cognitive theory (SCT), and it is largely defined as an individual’s belief in the ability to successfully perform a certain behavior.
Self-efficacy is an important contributor to cognitive and task improvements (Bandura, 1977). Before individuals select their choices and begin their work efforts, they evaluate and integrate
information about their perceived abilities. Self-efficacy determines how much effort an
individual invest into a task and for how long the effort will be sustained (Stajkovic & Luthans, 1979). Strong perceived efficacy beliefs can lead individuals to successfully challenge difficult projects as well as to work towards effective task performances and learning. According to a study by Berry (as cited in Bandura, 1993, p. 121) “perceived self-efficacy affects memory performance both directly and indirectly by raising cognitive efforts”. The more individuals believe in their capabilities of remembering information, the more cognitive processing effort they apply to a task. This higher cognitive effort, in turn, also leads to improved memory.
Self-efficacy techniques have been widely used in different types of studies and have shown to increase persuasive outcomes in several health areas such as, to improve breast cancer screening intentions, sun protection use, improve work performance and linked with fear appeals (Meyerowitz & Chaiken, 1987; Good & Abraham, 2011; Stajkovic & Luthans, 1998; Witte, 1992). Based on previous positive persuasive outcomes in using self-efficacy techniques, it is expected that by permitting individuals to self-select their preferred message format it will strengthen their perceived self-efficacy beliefs in processing the online information better. This will, in turn, lead to improved information recall and attitude towards the ad.
H4: The relationship between a self-selection option (vs. an absent self-selection option) on (a) recall of information and (b) attitude towards the ad is mediated by perceived self-efficacy to process online information.
Figure 1 (below and in Appendix D) shows the conceptual model of the current study. It illustrates the paths of the proposed hypotheses (H1, H2, H3 and H4).
H2
H3, H4
Age
Attitude and recall Self-selection (absent vs. present) Motivation Self-efficacy H1
Figure 1. The conceptual model of the study. This figure illustrates the paths of the hypotheses
(H1, H2, H3 and H4).
Method Design and Material
The experiment was carried as an online survey on the software tool Qualtrics. To test the hypotheses in the study, data were collected from female and male respondents who were
exposed to one out of three types of mock ads in a 2 (self-selection: present vs. absent) x 2 (age: younger vs. older) between-subjects factorial design. Ethical approval of this study was provided in advance by the Amsterdam School of Communication Research.
Three different versions of a health insurance ad were created for the experiment: a text (only), an image (supported by text) and an audio (only) format. The ads contained information about a health insurance policy, such as the yearly insurance fee, where and what it is covering and where to purchase it (see Appendix A for an overview). The ads were identical regarding their content and only differed in the way they were presented. The audio option was narrated by a female voice in clear Swedish language.
The study focused on Swedish nationals living in Sweden or abroad and on non-Swedish nationals living in Sweden. Participants that were not Swedish nationals and did not live in Sweden (although Swedish speakers), as well as those who did not complete the survey or provided invalid data, were excluded from the study. From an initial 333 participants, the final valid sample consisted of 296 participants. The age of the participants (N = 296) ranged from 22 to 82 (M = 50.17, SD = 15.66). Most of them (74.3%) were between 22 and 65 years old (the younger age group) and the rest (25.7%) were older than 65 (the older age group). Most
participants were female (79.7%) and the rest were male (20.3%). Their level of education was computed into a low and a highly educated group; the low education group included participants who had a mandatory education up to high school (52%) and the high education group included participants with a bachelor’s degree or higher (48%).
The participants were recruited in December 2016 via social media (Facebook) and e-mail. In order to motivate participants to undertake the survey they were given the option to join in a prize draw for the chance of winning a gift card valid SEK 450 (EUR 45) from a large Swedish department store (Åhléns). The first part of the online survey consisted of a recruitment text. In this part, they were briefed on who was behind the experiment, what they were asked to do and what to expect. They also had to read and accept a consent form in order to participate.
Subsequently, and before they were randomly assigned into one of the two conditions (self-selection or absent self-selection), they answered questions about their current physiological and cognitive abilities. Participants in the self-selection condition (N = 146) were told that they could choose their preferred ad format, whilst in the absent self-selection condition participants (N = 150) were
randomly exposed to one of the three ads (text, image–supported by text, or audio). Although, just before the participants viewed, read or listened to the ad, they were asked questions aimed at
assessing their motivational state and their perceived self-efficacy. The outcome variables (recall and attitude) were measured immediately after their exposure to the ad. In addition, questions for the control variables, the manipulation check and demographics were also included such as, age, gender, nationality, education and residency.
Measures
Recall of product characteristics. Six open questions, adapted from studies by Marks (1973) and Dahlén, Rosengren, Törn, and Öhman (2008), measured the participants’ recall of information in the ads. The questions were formulated such as; ‘Name the health insurance
company featured in the ad?’ and ‘How much does the health insurance package cost per month?’ Participants were given three answer options, which were adapted from the Netherlands Patient Information Recall Questionnaire (Jansen et al., 2008), such as; (i) ‘this was not mentioned in the ad’; (ii) ‘this was mentioned, although, I am unable to recall’ and; (iii) ‘I recall, namely…’ When the participants chose the last option, they were able write down their answer.
A codebook (see Appendix C) was developed beforehand to calculate the recall scores for each answer. Scores on each question ranged from 0 (wrong answer or not recalled), 1 (partially recalled) or 2 (recalled correctly). The scores of all the six questions were computed into a summed recall score with possible scores ranging from 0 to 12, (M = 8.17, SD = 3.00).
Attitude towards the ad. Seven items assessed participants’ attitude towards the ad. These were adapted from studies by Barksdale and Darden (1972), Sandage and Leckenby (1980) and Dahlén et al. (2008). Before answering, the participants were told to think about the ad they had just seen, heard or read. The variable was measured on a seven-point semantic differential scale consisting of items such as, ‘bad/ good’, ‘unclear/clear’ or ‘unpleasant/pleasant’, (α = .89, M = 3.61, SD = 0.97).
Motivation. The participants’ motivation was measured with five questions with
answer possibilities ranging from ‘1 = not at all’ to ‘7 = extremely’, such as ‘I am curious about the upcoming ad’ and ‘Right now, I feel interested in the upcoming ad’, (α = .70, M = 4.49, SD = 1.07). The scales were adapted from studies carried out by Guay, Vallerand, and Blanchard (2000), and Nijboer, Birbaumer, and Kubler (2010).
Self-efficacy. The participants’ perceived self-efficacy was measured with five questions; ‘Right now, I feel that I can overcome any task’ and ‘I feel confident that I will understand the upcoming ad’, (α = .86, M = 4.99, SD = 1.10). The questions were adapted from a study carried out by Jerusalem and Schwarzer (1979) and Bandura (2006). The answer possibilities ranged from ‘1 = not at all’ to ‘7 = extremely’.
Physiological (vision and hearing) and cognitive ability. Several questions, adapted from Peters, Boter, Buskens, and Slaets (2012) assessed participants’ physiological abilities (vision and hearing) as well as their cognitive abilities. The variables were measured using a 7-point Likert scale with answer possibilities ranging from ‘1 = not at all’ to ‘7 = extremely’. Participants’ vision was measured with two questions, such as ‘Do you have poor vision?’ and ‘Do you experience problems in daily life because of poor vision?’ (M = 1.98, SD = 1.06). Similarly, participants’ hearing was measured with two questions; ‘Do you suffer from poor hearing?’ and ‘Do you experience problems in daily life because of being hard of hearing?’ (M = 1.55, SD = 1.13). Spearman's correlation coefficient indicated that there was a positive
correlation between the items for vision and hearing (rs = .58, p = <.001 and rs = .80, p = <.001)
respectively.1
To measure the participants’ cognitive ability five questions were formulated; ‘Do
1
Eisinga, Grotenhuis, and Pelzer (2013) regard Spearman-Brown as the most appropriate reliability coefficient test for a two-item scale.
you have any complaints about your memory?’ and ‘Do you have any complaints about your ability to concentrate when you are browsing on the Internet?’ (α = .67, M = 2.73, SD = 0.95).
Control variables. Two control variables were included in the study; familiarity with health insurance and previous purchase of health insurance policies. The variables were measured with questions such as ‘Have you seen ads on health insurance policies before this one?’ and ‘Have you personally purchased health insurance policies before?’ (Yes/no/don’t know). Most participants (62.8%) had never seen ads on health insurance policies before and some answered ‘don’t know’ (17.2%). Most participants also reported not to have ever purchased health
insurance before (63.2%). Some participants (2.4%) answered ‘don’t know’.
Manipulation check. To check if the manipulation of the self-selection option had been successful, the following question was created: ‘The way in which I was exposed to the ad (via image–supported by text, text only or audio only) corresponds to my preferred way of receiving information’. The answer options were; ‘yes’, ‘no’ or ‘I don’t know’.
Appendix B contains a full list of all the measured used in the study.
Statistical Analysis
A chi-square test was carried out for the manipulation check in order to establish if the manipulation of the self-selection option had been successful. Randomization checks, in order to check for unequal distributions of variables between the two conditions, were also performed using independent t-tests and chi-square tests. The ‘don’t know’ answers for the manipulation and randomization checks were recoded into ‘no’ answers before analyses of the data.
To examine the main effect of self-selection and the interaction effect of self-selection and age group on recall and attitude, two separate two-way analyses of variance (ANOVAs) were carried out with recall and attitude as dependent variables and self-selection and age as (fixed)
factors. A simple effects analyze was included to look at the difference between the conditions within the older and younger adult groups on recall of information. Levene’s tests were also carried out in order to check for equality of variances in the groups. The analyses were not significant, which means that the variances in the groups were equal and that the assumptions are justifiable (Field, 2013).
To examine the extent to which motivation and self-efficacy mediates the relationship between self-selection (0 = absent, 1 = present) on recall and attitude, Hayes PROCESS statistical software (model 4) was used for these analyses. 5,000 bootstrap samples were used to estimate the bias-corrected bootstrap confidence interval (BCB CI) with a 95% confidence interval. Furthermore, to test if physiological and cognitive limitations influenced the choice of format chi-square tests of independence were performed.
Results Manipulation and Randomization Checks
Manipulation check. A manipulation check was carried out in order to establish whether the two conditions (self-selection vs. absent self-selection) differed to the extent that the way the participants were exposed to the ad corresponded to their preferred way of receiving the ad. As 70.5 percent of the participants (N = 103) in the self-selection condition answered ‘yes’ to the question assessing the success of the manipulation, it can be established that the manipulation was, indeed, successful, χ2
(1, N = 296) = 32.83, p = < .001.
Randomization check. To check for successful randomization, independent t-tests and chi-square tests were performed. Participants’ level of education was equally distributed between the two conditions, χ2
(1, N = 296) = 0.00, p =.992. So was the participants’ age, t(294) = -.06, p = .284, 95% CI [-3.70, 3.48], gender, χ2
insurance, χ2
(1, N = 296) = 0.31, p =.581 and previous purchase of health insurance policies χ2
(1, N = 296) = 0.03, p =.866 . Because of the non-significant results, no covariates were taken into account for further analysis.
Participants’ Choice of Format
Table 1 shows an overview of the participants’ choice of format in the self-selection condition. Most participants (69.2%) chose the image (supported by text) format (N = 101), whilst the rest (28.8%) chose the text format (N = 42). Only three participants (2.1%) chose the audio format. There were no differences in the choice of format between the young and older age groups; the younger group chose the image option the most compared to the older group (69.7% and 67.6% respectively), followed by the text format (28.4% and 29.7% respectively).
Cognitive and Physiological Abilities and Choice of Format
In general, most participants (N = 167) reported to have a high cognitive ability (56.4%) compared to participants (N = 129) with a low ability. They also reported to have only slightly impaired vision and hearing abilities (N = 214), 72.3 percent. The remaining reported to have perfect vision and hearing abilities (82), which was 27.7 percent of the participants.
The relation between participants’ cognitive ability and choice of format was significant, X2
(2, N = 146) = 6.16, p = .046. Even though participants in general choose the image format,
Table 1
Overview of the participants’ choice of format in the self-selection condition
Formats
Age Groups Text (%) Audio (%) Image (%) Total (%)
Younger 31 (28.4) 2 (1.8) 76 (69.7) 109 (100)
Older 11 (29.7) 1 (2.7) 25 (67.6) 37 (100)
participants with a perceived low cognitive ability tended to choose the text image (38.3%) compared to participants with a reported high cognitive ability (22.1%). The participants that chose the audio format (N = 3) had a high-perceived cognitive ability. Table 2 shows an overview of the participants’ reported cognitive ability and their choice of format.
There were no significant differences between participants with perfect physiological ability (vision and hearing) and those with slightly impaired physiological ability regarding the choice of format, X2
(2, N = 146) = 1.89, p = .388. In general, however, the image format was the most frequently chosen for individuals that reported to have a perfect vision and hearing (N = 31) as well as slightly impaired vision and hearing condition (N = 70), compared to the other formats. Table 3 shows an overview of the participants’ reported physiological ability and their choice of format.
Table 2
Overview of the participants’ reported cognitive ability and their choice of format
Formats Cognitive
Ability Text (%) Audio (%) Image (%) Total (%)
Low 23 (38.3) 0 (0) 37 (61.7) 60 (100)
High 19 (22.1) 3 (3.5) 64 (74.4) 86 (100)
Total 42 3 101
Table 3
Overview of the participants’ reported physiological ability and their choice of format
Formats Physiological
Ability Text (%) Audio (%) Image (%) Total (%)
Slightly
Impaired 32 (30.5) 3 (2.9) 70 (66.7) 105 (100)
Perfect 10 (24.4) 0 (0) 31 (75.6) 41 (100)
The Effect of Self-Selection and Age Group on Recall
There was no significant main effect of self-selection on recall (H1a), F(1, 292) = 2.68, p = .102, η2
= .01, which suggests that self-selection did not directly affect recall of information. H1a is, therefore, not supported. However, a simple effect analysis shows that there is a
significant difference between conditions within the younger age group, F(1, 292) = 6.32, p = .012, η2
= .02. Younger participants in the control condition performed better (M = 8.87, SD = 0.28) than the younger participants in the self-selection condition (M = 7.87, SD = 0.28). This means that the younger participants in the control condition recalled significantly more
information in the ads compared to younger participants in the self-selection condition (Mdifference =
1.00, SE = 0.40, p = .012). There was, however, no significant difference between conditions on recall within the older age group, F(1, 292) = 0.18, p = .673, η2
= .00. The interaction effect of self-selection and age group on recall was not significant either, F(1, 292) = 0.83, p = .364, η2
= .00. This means that self-selection does not differ on recall between young and older individuals. Hence, H2a is not supported.
A post-hoc analysis within the control condition on recall showed a significant difference between the ad formats, F(2, 147) = 4.42, p = .014. Those participants who were exposed to the audio format recalled more information (M = 9.56, SD = 2.05) compared to the participants that were exposed to the text (M = 8.56, SD = 3.14) and image (M = 7.77, SD = 3.28) formats. The participants in the control condition who listened to the ad differed significantly from participants who saw the image (Mdifference = 1.78, SE = 0.60, p = .010).
Within the self-selection condition there was a significant difference between the formats, F(2, 143) = 3.56, p = .031. However, only three participants chose the audio format (M = 9.00, SD = 1.00). Participants who chose the text format recalled more (M = 8.69, SD = 2.30)
compared to participants who chose the image format (M = 7.33, SD = 3.14), Mdifference = 1.36, SE
= 0.53, p = .035.
The Effect of Self-Selection and Age Group on Attitude
There is no significant main effect of self-selection on attitude (H1b), F(1, 292) = 0.04, p = .838, η2
= .00, which suggests that self-selection did not directly affect attitude towards the ad. H1b is, therefore, not supported. The interaction effect of self-selection and age group on attitude was not significant either, F(1, 292) = 1.17, p = .281, η2
= .00, which suggests that self-selection does not differ on attitude between young and older individuals. Hence, H2b is not supported. Participants in the self-selection condition had a higher attitude towards the ad (M = 3.70, SD = 0.09) compared to participants in the present control condition (M = 3.68, SD = 0.09), although the difference was not significant. Table 4 shows an overview of the effect of self-selection and age group on recall of information and attitude towards the ad. An overview of all the tables and figures can be found in Appendix D.
Table 4
Overview of the effect of self-selection (versus absent self-selection) and age group on recall and attitude
Recall Attitude
Condition N M SD M SD
Absent Self-selection
(control condition) Total 150 8.57 2.99 3.63 0.96
Young 111 8.87* 2.84 3.58 0.93
Old 39 7.69 3.27 3.77 1.04
Present Self-selection Total 146 7.75 2.95 3.59 0.99
Young 109 7.87 3.02 3.47 0.93
Old 37 7.41 2.76 3.93 1.11
Age group Total 296 8.17 3.00 3.61 0.97
Young 220 8.38 2.97 3.52 0.93
Old 76 7.55 3.02 3.85 1.07
The Mediating Effect of Motivation and Self-Efficacy on Recall
Figure 2 represents the statistical mediation model that was tested, including the results for H3a and H4a. The results showed that there is an indirect effect of self-selection on recall through motivation, b = 0.201, SE = 0.10, 95% BCa CI [0.057, 0.428]. This indicates that self-selection of information presentation lead to higher motivation to process information, which subsequently improved recall of information in the ad. Hence, H3a is supported. Regarding self-efficacy, there was no indirect effect of self-selection on recall through self-self-efficacy, b = -0.049, SE = 0.05, 95% BCa CI [-0.210, 0.020]. Thus, H4a is not supported.
Figure 2. Mediation model. This figure illustrates the effect of self-selection on recall via
motivation and self-efficacy (H3a and H4a). The model shows the unstandardized coefficients. *** p < .001, ** p < .01
The Mediating Effect of Motivation and Self-Efficacy on Attitude
Figure 3 represents the statistical mediation model that was tested, including the results for H3b and H4b. The results showed that there was no indirect effect of self-selection on attitude through motivation, b = 0.044, SE = 0.03, 95% BCa CI [-0.002, 0.125]. Furthermore, there was no indirect effect of self-selection option on attitude through self-efficacy, b = -0.016, SE = 0.02, 95% BCa CI [-0.067, 0.004]. As motivation and efficacy did not mediate the effect of self-selection on attitude, H3b and H4b are not supported.
Self-selection
(absent vs. present) Recall
.47** .43*** c’ = - .97 ** -.23 Motivation Self-efficacy .21
Figure 3. Mediation model. This figure illustrates the effect of self-selection on attitude via
motivation and self-efficacy (H3b and H4b). The model shows the unstandardized coefficients. *** p < .001.
Discussion Summary of Findings
This study investigated the effect of a self-selection option of information presentation (vs. an absent self-selection option) on recall of information and attitude towards an online health insurance ad. It also examined whether this effect differed between older and younger
individuals. The results showed that, as expected, self-selection on recall was mediated by motivation to process online information (H3a). Motivation did not, however, mediate the relationship between self-selection and attitude towards the ad (H3b).
In contrast to hypothesis H1ab, self-selection did not lead to higher information recall and more positive attitudes towards the ad compared to an absent self-selection option. There was, however, a significant difference between self-selection (vs. an absent self-selection) on recall within the younger age group. Furthermore, the relation between self-selection on recall and attitude towards the ad was not mediated by self-efficacy to process online information (H4ab).
Self-selection
(absent vs. present) Attitude
Self-efficacy .10 -.23 .07 Motivation .43*** c’ = - .07
Nor was the effect of self-selection on attitude and recall stronger for older individuals compared to younger individuals (H2ab).
General Discussion and Conclusion
As expected, self-selection had a positive effect on motivation to process information, which, in turn, improved recall of information (H3a). This indicates that when individuals have the opportunity to select their preferred way of receiving information online, their motivation to process this information increases, which subsequently improves their recall of information. This finding is in line with previous research in health communication (Bol et al., submitted) and advertising (MacInnis et al., 1991) and suggests that adding a self-selection option to online ads could be beneficial towards information recall via motivation.
However, self-selection did not directly improve information recall and lead to a more positive attitude towards the ad compared to an absent self-selection (H1ab). This result is similar to the outcome in the research by Maslowska et al. (2013). Their study showed that tailored advertising improved message attitude for Polish participants but not for Dutch participants, which suggested different tailored advertising effects in different markets. One explanation to the current findings could be that Swedes, similarly to the Dutch, might be used to advertising due to high advertising exposure rates. Sweden saw an enormous growth in advertising expenditures from 1990 onwards (“90-talet,” 2014) and, therefore, they might have become indifferent to advertising by now. They might also have negative attitudes towards ads in general: Currently, as much as 30 percent of the Swedish Internet population block online ads (Southern, 2016). This could also explain why self-selection did not improve attitudes towards the ad through motivation to process information (H3b).
Further analysis, however, showed that there was a significant difference on recall between the conditions, but only within the younger age group: Younger participants in the control group recalled more information compared to younger participants in the self-selection condition. This finding was unexpected because it is rather the opposite of what was predicted. As the post-hoc analysis showed, participants in the control group of the study who were exposed to the audio format recalled the most information, compared to those exposed to the text and image formats. Because only three participants in the self-selection condition of the study chose the audio format, this could be a reason why participants in the control group recalled more compared to participants in the self-selection, within the younger age group.
Hence, audio worked best on recall (in the control condition) compared to the multimodal (image–supported by text) and text (only) formats, which might suggest that this is a better ad format to employ. This result is in line with findings in the study by Kalyuga et al. (1999) within learning, and in the research by Meppelink, Van Weert, Haven, and Smit (2015) on health information. Although, the latter research found that only participants with low health literacy who were exposed to an audio format recalled more, compared to participants with high literacy. Thus, it would be hasty to indicate that an audio message format is a better option on recall overall: Perhaps, giving individuals the option to chose format based on their abilities could be more beneficial.
However, a reason why participants in control condition––that were exposed to the audio format––recalled more information (compared to the other formats) could simply be because they paid more attention and time to the audio format, or because the audio format was clearer
(without a fast-forward or pause option) before being able to continue with the survey. Future research could include measures of attention or involvement to the ad in order to control for this.
This finding contradicts the DCT and previous finding on e-health information, which has found that a multimodal information presentation improves recall compared to a text format (Bol et al., 2015). In the current study participants exposed to the multimodal format (image–
supported by text) recalled the least. This is in accordance with other findings on multimedia learning (Craig, Gholson, & Driscoll, 2002) and the redundancy theory that states that presenting words in one way is better than presenting them in two ways simultaneously (Mayer et al., 2001).
Unexpectedly, the results showed that there was no difference in self-selection on attitude and recall between the age groups (H2ab). The participants in the study were quite healthy; they did not report to have excessive cognitive or physiological limitations. Thus, because the older participants in the current study were not as frail as expected, this could be a reason why there was no difference on recall and attitude between the age groups. The older participants might also not have differed in their technological adoption and Internet use compared to the younger
participants––contrary to what previous American research shows (Pew Research Center, 2014). Resent research from Sweden on older Swedes’ Internet use shows that Internet accessibility and use among this group has risen intensely in the past years: 70 percent of Internet users within this group are surfing online on a daily basis (Ilbring, 2015). Furthermore, the oldest Internet blogger in Sweden is 103 years old and has almost two million visitors to date
(http://123minsida.se/bojan). This indicates that it is not uncommon that older Swedes are using the Internet and are, perhaps, just as tech savvy as the younger generations. Thus, older Swedish nationals might be more homogeneous in their technological adoption, and not feel the benefit of
an online self-selection option: They might be, overall, as able to attend to, and process, online information as well as younger individuals.
Another noteworthy result is that, participants with a low cognitive ability in the self-selection condition chose the text option more compared to individuals with a perceived high cognitive ability. This may indicate that individuals with a low cognitive ability prefer online ad formats that are simpler (i.e. text only) compared to more complex (multimodal) image formats. Mayer and Sims (1994) highlight adult differences and state that individuals with low-spatial abilities do not benefit of complex multimedia designs as they have to devote more cognitive processing to hold mental images, compared to individuals with high-spatial abilities. Similarly, the participants with low cognitive abilities might have taken their cognitive abilities into account before selecting their preferred format and, thus, preferred a single modal compared to a
multimodal one.
In addition, the hypothesis that the relationship between a self-selection option on recall and attitude towards the ad would be mediated by self-efficacy was rejected (H4ab). A reason for this could be that the participants already had a high level of perceived self-efficacy to start with, and that self-selection did not increase their (already high) levels of self-efficacy believes.
Limitations, Implications and Directions for Future Research
The study has a few limitations that should be mentioned. One limitation is the
underrepresentation of older individuals and men, compared to women and younger individuals. The study was also carried out on Swedish nationals and foreigners living in Sweden, which restricts the generalizability of the findings. Difference in Internet usage and exposure to online advertising was not measured, which could be taken into account in future research.
Another important limitation is that the participants were quite homogenous in terms of health. Most of the participants considered themselves to be healthy and did not report to have particularly high levels of physiological or cognitive limitations. This might have been a reason to why only three participants chose the audio format in the self-selection condition, which made it impossible to further examine if individuals with limited vision would chose an audio format over a text or image format. As the older participants in the study were not as frail as expected, future research could aim at replicating the study with a more diverse and frail sample so that the effects of self-selection in frail and old populations can be investigated.
In conclusion, this study demonstrated that a single modal of information presentation (audio) might be a better option for online advertising on recall. However, although audio seemed to be the best format in this study, it should not be disregarded that the audio recording in this study was very clear. Perhaps too long or too complex audio recordings might not be as effective. Audio might also not be the best option for all types of individuals and in a more heterogeneous sample. These views could be taken into account for future research and in the creation of audio messages.
However, self-selection via motivation to process information showed to be an effective tailoring strategy in order to improve recall of online information. Thus, giving individuals a choice on how to be presented with online ads might be a more beneficial way to increase recall, through motivation. Online advertisers and website designers could take this result into account when designing websites and online ads. Parallel versions of the same website (in addition to online ads) could additionally be proposed in order to accommodate different user needs. However, it may be that this positive effect appeared because participants self-selected their
preferred format for the first time and, thus, felt more motivated by it. Future research could carry out a longitudinal study to see whether this effect lasts over time.
A company that has recently introduced an option for customers to choose between a visual or audio online account login format is Emirates (2017). Their tailored login option is an example that also indicates that there might, indeed, be a need to distinguish between individuals with different cognitive or physiological differences. Individuals should, perhaps, not be grouped into a “one size fits all” category, but be given the freedom of choice, no matter chronological age. Further research is, nevertheless, needed to investigate the effect of self-selection further.
References
90-talet: En explosion av kanaler och investeringarna skjuter i höjden. (2014, October 8). Retrieved on 19 January 2017, from http://www.annons.se/artiklar/90-talet-en-explosion-av-kanaler-och-investeringarna-skjuter-i-hojden
Ausburn, L. J., & Ausburn, F. B. (1978). Cognitive styles: Some information and implications for instructional design. Ectj, 26(4), 337-354.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
Bandura, A. (1993). Perceived self-efficacy in cognitive development and functioning. Educational Psychologist, 28(2), 117-148.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In self-efficacy beliefs of adolescents (pp. 307-337). United States of America: Information Age Publishing. Barksdale, H. C., & Darden, W. R. (1972). Consumer attitudes toward marketing and
consumerism. The Journal of Marketing, 36(4), 28-35. doi:10.2307/1250423
Bol, N., Smets, E. M. A., Burgers, J. A., Samii, S. M., De Haes, J. C. J. M., Loos, E. F., . . . Van Weert, J. C. M. (submitted). Older cancer patients’ recall of online cancer information: Do ability and motivation matter more than chronological age? Manuscript submitted for publication.
Bol, N., Van Weert, J. C. M., De Haes, H. C. J. M., Loos, E. F., & Smets, E. M. A. (2015). The effect of modality and narration style on recall of online health information: Results from a web-based experiment. Journal of Medical Internet Research, 17(4), 104.
Byrne, M., & Curtis, R. (2000). Designing health communication: Testing the explanations for the impact of communication medium on effectiveness. British Journal of Health Psychology, 5(2), 189-199. doi:10.1348/135910700168856
Chen, G., & Fu, X. (2003). Effects of multimodal information on learning performance and judgment of learning. Journal of Educational Computing Research, 29(3), 349-362. Clark, J. M., & Paivio, A. (1991). Dual coding theory and education. Educational Psychology
Review, 3(3), 149-210. doi:10.1007/BF01320076
Craig, S. D., Gholson, B., & Driscoll, D. M. (2002). Animated pedagogical agents in multimedia educational environments: Effects of agent properties, picture features and
redundancy. Journal of Educational Psychology, 94(2), 428. doi:10.1037/0022-0663.94.2.428
Dahlén, M., Rosengren, S., Törn, F., & Öhman, N. (2008). Could placing ads wrong be right?: Advertising effects of thematic incongruence. Journal of Advertising, 37(3), 57-67. Deci, E. L., & Ryan, R. M. (2008). Self-determination theory: A macro theory of human
motivation, development, and health. Canadian Psychology/Psychologie Canadienne, 49(3), 182. doi:10.1037/a0012801
Eisinga, R., Grotenhuis, M. T., & Pelzer, B. (2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health, 58(4), 1-6. doi:10.1007/s00038-012-0416-3
Emirates. (2017). Retrieved from https://www.emirates.com/account/us/english/login/login.aspx Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th
ed.). London: SAGE Publications Ltd.
Good, A., & Abraham, C. (2011). Can the effectiveness of health promotion campaigns be improved using self-efficacy and self-affirmation interventions? An analysis of sun protection messages. Psychology & Health, 26(7), 799-818.
doi:10.1080/08870446.2010.495157
Guay, F., Vallerand, R. J., & Blanchard, C. (2000). On the assessment of situational intrinsic and extrinsic motivation: The situational motivation scale (SIMS). Motivation and
Emotion, 24(3), 175-213. doi:10.1023/A:1005614228250
Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A theory of the relationship between adult differences and information processing. Psychological Review, 91(2), 153.
Ilbring, L. (2015, August 26). Majoriteten av de äldre internetanvändarna surfar dagligen. Internetstatistik. Retrieved from http://www.internetstatistik.se/artiklar/majoriteten-av-de-aldre-internetanvandarna-surfar-dagligen/
Jansen, J., Van Weert, J., Van der Meulen, N., Van Dulmen, S., Heeren, T., & Bensing, J. (2008). Recall in older cancer patients: Measuring memory for medical information. The
Gerontologist, 48(2), 149-157. doi:10.1093/geront/48.2.149
Jensen, J. D., King, A. J., Carcioppolo, N., & Davis, L. (2012). Why are tailored messages more effective? A multiple mediation analysis of a breast cancer screening intervention. Journal of Communication, 62(5), 851-868. doi:10.1111/j.1460-2466.2012.01668.x Jerusalem, M., & Schwarzer, R. (1979). The general self-efficacy scale. Retrieved from
https://cyfar.org/sites/default/files/PsychometricsFiles/General%20Self-Efficacy%20Scale%20(Adolescents,%20Adults)%20Schwarzer.pdf
Kalyuga, S., Chandler, P., & Sweller, J. (1999). Managing split-attention and redundancy in multimedia instruction. Applied Cognitive Psychology, 13(4), 351-371.
Lindeboom, J., & Weinstein, H. (2004). Neuropsychology of cognitive ageing, minimal cognitive impairment, Alzheimer’s disease, and vascular cognitive impairment. European Journal of Pharmacology, 490(1), 83-86. doi:1016/j.ejphar.2004.02.046
Lippincott, G. (2004). Gray matters: Where are the technical communicators in research and design for aging audiences? IEEE Transactions on Professional Communication, 47(3), 157-170. doi:10.1109/TPC.2004.833687
Lustria, M. L. A., Cortese, J., Gerend, M. A., Schmitt, K., Kung, Y. M., & McLaughlin, C. (2016). A model of tailoring effects: A randomized controlled trial examining the mechanisms of tailoring in a web-based STD screening intervention. Health Psychology, 35(11), 1214. doi:10.1037/hea0000399
MacInnis, D. J., Moorman, C., & Jaworski, B. J. (1991). Enhancing and measuring consumers' motivation, opportunity, and ability to process brand information from ads. The Journal of Marketing, 32-53. doi:10.2307/1251955
Manolopoulou-Sergi, E. (2004). Motivation within the information processing model of foreign language learning. System, 32(3), 427-441. doi:10.1016/j.system.2004.04.002
Marks, D. F. (1973). Visual imagery differences in the recall of pictures. British Journal of Psychology, 64(1), 17-24. doi:10.1111/j.2044-8295.1973.tb01322.x
Maslowska, E., Smit, E., & Van den Putte, B. (2011). Examining the (in)effectiveness of
personalized communication. In M. Eisend, & T. Langner (Eds.), The 10th ICORIA 2011 Berlin: June 23rd-25th 2011: conference programme [DVD]. European Advertising Academy.
Maslowska, E., Smit, E. G., & Van den Putte, B. (2013). Assessing the cross-cultural
applicability of tailored advertising: A comparative study between the Netherlands and Poland. International Journal of Advertising, 32(4), 487-511. doi:10.2501/IJA-32-4-487-511
Maslowska, E., Smit, E. G., & Van den Putte, B. (2016). It is all in the name: A study of consumers' responses to personalized communication. Journal of Interactive Advertising, 16(1), 74-85. doi:10.1080/15252019.2016.1161568
Mayer, R. E. (2002). Multimedia learning. Psychology of Learning and Motivation, 41, 85-139. doi:10.1016/S0079-7421(02)80005-6
Mayer, R. E. (2014). Cognitive theory of multimedia learning. The Cambridge Handbook of Multimedia Learning, 43, 31-48.
Mayer, R. E., Heiser, J., & Lonn, S. (2001). Cognitive constraints on multimedia learning: When presenting more material results in less understanding. Journal of Educational
Psychology, 93(1), 187.
Mayer, R. E., & Sims, V. K. (1994). For whom is a picture worth a thousand words? Extensions of a dual-coding theory of multimedia learning. Journal of Educational
Psychology, 86(3), 389. doi:10.1037/0022-0663.86.3.389
Meppelink, C. S., Van Weert, J. C., Haven, C. J., & Smit, E. G. (2015). The effectiveness of health animations in audiences with different health literacy levels: An experimental study. Journal of Medical Internet Research, 17(1), 11.
Meyerowitz, B. E., & Chaiken, S. (1987). The effect of message framing on breast self-examination attitudes, intentions, and behavior. Journal of Personality and Social Psychology, 52(3), 500-510. doi:10.1037/0022-3514.52.3.500
Moorman, C., & Matulich, E. (1993). A model of consumers' preventive health behaviors: The role of health motivation and health ability. Journal of Consumer Research, 20(2), 208-228. doi:10.1086/209344
Morris, J. M. (1994). Computer training needs of older adults. Educational Gerontology: An International Quarterly, 20(6), 541-555. doi:10.1080/0360127940200601
Morrow, D. G., Hier, C. M., Menard, W. E., & Leirer, V. O. (1998). Icons improve older and younger adults' comprehension of medication information. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 53(4), 240-254.
doi:10.1093/geronb/53B.4.P240
Mousavi, S. Y., Low, R., & Sweller, J. (1995). Reducing cognitive load by mixing auditory and visual presentation modes. Journal of Educational Psychology, 87(2), 319.
doi:10.1037/0022-0663.87.2.319
Muehling, D. D., Stoltman, J. J., & Grossbart, S. (1990). The impact of comparative advertising on levels of message involvement. Journal of Advertising, 19(4), 41-50.
doi:10.1080/00913367.1990.10673199
Nguyen, M. H., Smets, E. M., Bol, N., Loos, E. F., & Van Weert, J. C. (submitted). How tailoring the mode of information presentation influences younger and older adults’ satisfaction with health websites. Manuscript submitted for publication.
Nguyen, M. H., Van Weert, J., Bol, N., Loos, E. F., Tytgat, K. M., Van de Ven, A. W., & Smets, E. (2017). Tailoring the mode of information presentation: Effects on younger and older adults' attention and recall of online information. Human Communication
Nijboer, F., Birbaumer, N., & Kubler, A. (2010). The influence of psychological state and motivation on brain–computer interface performance in patients with amyotrophic lateral sclerosis–a longitudinal study. Frontiers in Neuroscience, 4, 55.
doi:10.3389/fnins.2010.00055
Noar, S. M., Benac, C. N., & Harris, M. S. (2007). Does tailoring matter? Meta-analytic review of tailored print health behavior change interventions. Psychological Bulletin, 133(4), 673. doi:10.1037/0033-2909.133.4.673
O’Keefe, D. J. (2008). Elaboration likelihood model. In W. Donsbach (Ed.), International Encyclopedia of Communication, 4, 1475-1480. Oxford, UK, and Malden, MA: Wiley-Blackwell.
Paivio, A. (1971). Imagery and verbal processes. New York, NY: Holt, Rinehart, and Winston. Peters, L. L., Boter, H., Buskens, E., & Slaets, J. P. (2012). Measurement properties of the
Groningen frailty indicator in home-dwelling and institutionalized elderly people. Journal of the American Medical Directors Association, 13(6), 546-551.
doi:10.1016/j.jamda.2012.04.007
Petty, R. E., & Cacioppo, J. T. (1986). The elaboration likelihood model of persuasion. Advances in Experimental Social Psychology, 19, 1-24.
Pew Research Center (2014). Older adults and technology use. Retrieved on 7 October 2016, from, http://www.pewinternet.org/2014/04/03/older-adults-and-technology-use/
Reuben, D. B., Mui, S., Damesyn, M., Moore, A. A., & Greendale, G. A. (1999). The prognostic value of sensory impairment in older persons. Journal of the American Geriatrics
Rimer, B. K., & Kreuter, M. W. (2006). Advancing tailored health communication: A persuasion and message effects perspective. Journal of Communication, 56(1), 184-S201.
doi:10.1111/j.1460-2466.2006.00289.x
Salthouse, T. A. (2011). Neuroanatomical substrates of age-related cognitive decline. Psychological Bulletin, 137(5), 753. doi:10.1037/a0023262
Salthouse, T. A., & Babcock, R. L. (1991). Decomposing adult age differences in working memory. Developmental Psychology, 27(5), 763. doi:10.1037/0012-1649.27.5.763 Sandage, C. H., & Leckenby, J. D. (1980). Student attitudes toward advertising: Institution vs.
instrument. Journal of Advertising, 9(2), 29-44. doi:10.1080/00913367.1980.10673316 Smit, E. S., Linn, A. J., & Van Weert, J. C. (2015). Taking online computer-tailoring forward:
The potential of tailoring the message frame and delivery mode of online health behaviour change interventions. European Health Psychologist, 17(1), 25-31.
Soroka, A. J., Wright, P., Belt, S., Pham, D. T., Dimov, S. S., De Roure, D., & Petrie, H. (2006). User choices for modalities of instructional information. IEEE International Conference on Industrial Informatics, 411-416. doi:10.1109/INDIN.2006.275835
Southern, L. (2016, March 14). Sweden’s publishers are joining forces to simultaneously block ad-block users. Digiday. Retrieved on 18 January 2017, from
http://digiday.com/publishers/swedens-publishers-gearing-block-
Stajkovic, A. D., & Luthans, F. (1979). Social cognitive theory and self-efficacy: Implications for motivation theory and practice. In R. M. Steers, L. W. Porter, & G. A. Bigley
(Eds.), Motivation and Work Behavior (pp. 126-140). Boston: MA. McGraw-Hill. Stajkovic, A. D., & Luthans, F. (1998). Self-efficacy and work-related performance: A
Van Gerven, P.W.M., Paas, F. G. W. C., Van Merriënboer, J. J. G., & Schmidt, H. G. (2000). Cognitive load theory and the acquisition of complex cognitive skills in the elderly: Towards an integrative framework. Educational Gerontology, 26(6), 503-521. doi:10.1080/03601270050133874
Wagner, N., Hassanein, K., & Head, M. (2010). Computer use by older adults: A multi-disciplinary review. Computers in Human Behavior, 26(5), 870-882.
doi:10.1016/j.chb.2010.03.029
Webb, M. S., Hendricks, P. S., & Brandon, T. H. (2007). Expectancy priming of smoking cessation messages enhances the placebo effect of tailored interventions. Health Psychology, 26(5), 598. doi:10.1037/0278-6133.26.5.598
Witte, K. (1992). Putting the fear back into fear appeals: The extended parallel process
model. Communications Monographs, 59(4), 329-349. doi:10.1080/03637759209376276 Woodhouse, K. W., Wynne, H., Baillie, S., James, O. F., & Rawlins, M. D. (1988). Who are the
frail elderly? QJ Med, 68(255), 505-506.
Wright, P., Dimov, S., Soroka, A., Petrie, H., Belt, S., & De Roure, D. (2009). Effects of language fluency and graphic animation on modality choices by adults when following online explanatory demonstrations. IEEE International Professional Communication Conference, 1-6. doi:10.1109/IPCC.2009.5208675
Wright, P., Soroka, A. J., Belt, S., Pham, D. T., Dimov, S., DeRoure, D. C., & Petrie, H. (2008). Modality preference and performance when seniors consult online
Zhao, H., Seibert, S. E., & Hills, G. E. (2005). The mediating role of self-efficacy in the
development of entrepreneurial intentions. Journal of Applied Psychology, 90(6), 1265. doi:10.1037/0021-9010.90.6.1265
Appendix A Stimulus Material
Health insurance ad (original Swedish version of the text-only condition)
Health insurance ad (English translation of the Swedish text-only condition) HARMONI - privat sjukförsäkring
Visste du att vår privata sjukförsäkring nu också täcker 90% av alla sjukhusbesök i alla länder inom Europa? Försäkringen inkluderar också ett kostnadsfritt tandläkarbesök per år. Bli medlem hos oss före den 31 december 2016 och få 20% rabatt på din första räkning. Du
får också en hemlig julklapp på köpet!
Försäkringen kostar bara 500 SEK per månad med en bindningstid på två år. Bli medlem på vår hemsida; www.harmoni.se/medlemskap
Förtroende. Omsorg. Harmoni.
HARMONY – private health insurance
Did you know that our private health insurance covers 90% of all hospital visits in any European country? Our insurance also covers one dental appointment per year.
Harmony costs only 50 Euros per month if you sign up for two years.
Sign up by December 31, 2016 and receive 20% off your first bill. In addition, you will receive a complementary Christmas present!
Sign up today on our website; www.harmony.se/membership Trust. Care. Harmony.
Health insurance ad (original Swedish version of the image format supported by text condition)
Health insurance ad (audio recording – 44 seconds long)
This is the transcribed version of the audio condition, which has been translated from the Swedish audio version (this transcription contains the same information as the text only and the image–supported by text condition).
Harmony, private health insurance. Did you know that our private health insurance covers 90% of all hospital visits in any European country? Our insurance also covers one dental appointment per year. Harmony costs only 50 Euros per month if you sign up for two years. Sign up by December 31, 2016 and receive 20% off your first bill. In addition, you will receive a
complementary Christmas present. Sign up today on our website, www.harmony.se/membership. Trust. Care. Harmony.
Appendix B Measures
(Translated into Swedish language on Qualtrics) Physiological ability
1. Are you able to carry out these tasks single-handedly and without any help? (Yes/no) a. Shopping
b. Walking around the house or on the street c. Dressing and undressing
2. What mark do you give yourself for your physical fitness? (From 0-10)
3. Do you experience problems in daily life because of poor vision? (1 = not at all, 7 = extremely)
4. Do you have poor vision? (1 = not at all, 7 = extremely)
5. Do you experience problems in daily life because of being hard of hearing? (1 = not at all, 7 = extremely)
6. Do you suffer from poor hearing? (1 = not at all, 7 = extremely) Cognitive ability
The items with an asterisk (*) are reversed items. (1 = not at all, 7 = extremely)
1. Do you have any complaints about your memory?
2. Do you have any complaints about your ability to concentrate when you are browsing the Internet?
3. Do you feel that you are good at problem solving? *
4. Do you feel that you comprehend most text on the Internet? * 5. Do you feel that you have a good attention spam? *
Motivation
The items with an asterisk (*) are reversed items. (1 = not at all, 7 = extremely)
1. I look forward to being presented with an ad. 2. Right now, I feel interested in the ad.
3. Right now, I feel focused on reading these questions. 4. Right now, I feel under pressure to perform well. * 5. I am curious about the upcoming ad.
Self-efficacy
1. I feel confident that I will understand the upcoming ad. 2. Right now, I feel optimistic.
3. Right now, I feel confident that I will be able to efficiently answer questions about the ad I will be presented with.
4. Right now, I feel that I can overcome any task. 5. Right now, I feel in control.
Recall (Free recall)
1. Name the health insurance company featured in the ad?
2. How much does the health insurance package cost per month?
3. How much discount is given if you sign up before December 31, 2016 4. Which geographical region does the insurance cover?
5. Where can you purchase the insurance?
6. A part from hospital visits what other service does the insurance cover?
Attitude
(Scale from 1 to 7)
1. The ad was bad/good. 2. The ad was unclear/clear. 3. The ad was unpleasant/pleasant. 4. The ad was unbelievable/ believable. 5. The ad was negative/positive.
6. The ad was unfavorable/ favorable. 7. The ad was boring/fun.
Control variables (Yes/no/don’t know)
1. Have you seen ads on health insurance policies before this one? 2. Have you personally purchased health insurance policies before?
Manipulation check
1. The way in which I was exposed to the ad (via image–supported by text, text only or audio only) corresponds to my preferred way of receiving information (yes/no).
