The Effectiveness of Persuasive Appeals Tailored to Recipient’s Personality Traits:

The Effectiveness of Persuasive Appeals

Tailored to Recipient’s Personality Traits:

The Mediating Role of Self-referencing and Resistance to Persuasion

Anton Jatula (1895230) University of Groningen

The Effectiveness of Persuasive Appeals

Tailored to Recipient’s Personality Traits:

The Mediating Role of Self-referencing and Resistance to Persuasion

Anton Jatula University of Groningen Faculty of Economics & Business Master thesis MSc. Marketing Management

20-06-2016 Vrijheidslaan 22 9301 ES Roden 0637385438 a.jatula@student.rug.nl S1895230

MANAGEMENT SUMMARY

Marketers today are able to capture and store vast amounts of data and use it to tailor messages to an individual’s characteristics. Message-person congruence effects have been studied in relation to a variety of psychological characteristics (Dijkstra, 2008), however, they have not yet been systematically related to a comprehensive model of personality traits. This study set out to replicate the work of Hirsh et al. (2012) by examining whether message-person congruence effects can be obtained by framing persuasive messages in terms of the Big Five personality dimensions (Goldberg, 1990). Furthermore, it examined the mediating effects of self-referencing and resistance to persuasion on the relationship between messages tailored to recipients’ personality traits and attitude towards the message.

The results of this study indicate that a persuasive message would not be more effective when framed to be congruent with the recipient’s personality profile. Mediating effects of self-referencing were detected in three of the five advertisements (Conscientiousness, Agreeableness, and Neuroticism). The mediating effect of resistance to persuasion was only found in one advertisement (Openness/Intellect).

TABLE OF CONTENTS

TABLE 1:MEASUREMENTS OF CONSTRUCTS ... 13

3.4METHOD OF ANALYSIS ... 14 4. RESULTS ... 15 4.1 MEASUREMENT MODEL ... 16 4.2 STRUCTURAL MODEL ... 18 4.2.1 Direct effects ... 18 4.2.2 Indirect/mediation effects ... 21 4.2.3 Robustness check ... 27 5. DISCUSSION, LIMITATIONS, AND CONCLUSION ... 28 5.1 DISCUSSION ... 28

5.2 LIMITATIONS AND FUTURE RESEARCH DIRECTIONS ... 29

5.3 CONCLUSION ... 29

REFERENCES ... 30

APPENDICES ... 38

APPENDIX 1: TEXT FROM THE ADVERTISEMENTS (HIRSH ET AL., 2012) ... 38

APPENDIX 2: MEASUREMENT OF CONSTRUCTS – THE BIG FIVE ASPECT SCALES ... 39

APPENDIX 3: PLS-SEM MODELS FOR EACH TARGETED PERSONALITY DIMENSION ... 41

APPENDIX 4: CONSTRUCT VALIDITY AND CONVERGENT VALIDITY ... 43

APPENDIX 5: SQUARED ROOTS OF THE AVERAGE VARIANCE EXTRACTED (AVE) AND CORRELATION MATRIX ... 44

1. INTRODUCTION

The internet has profoundly influenced the way companies do their marketing. Marketers today are able to capture real-time behavioural data, such as search behaviour and purchasing patterns, and use it to target individual consumers with specific advertisements. One method commonly used to customize advertising messages is message tailoring. This method involves adapting persuasive messages to the individual, and the focus is on fitting a message to meet personal needs and characteristics (instead of targeting group criteria, which is utilized in message targeting). The most basic approach to message tailoring is personalization, which involves identifying an individual’s characteristics, such as a person’s name or age, and incorporating this information into marketing communication messages. A number of studies have shown that personalization may result in positive effects such as higher customer loyalty (Komiak & Benbasat, 2006), improved communication, preference match, and overall customer experience (Wind & Rangaswamy, 2001). Another approach to tailoring involves adapting messages to an individual’s motivational orientation by focusing either on promoting gains or preventing losses. Several studies have demonstrated that messages which are congruent with an individual’s motivational orientation are processed more fluently and evaluated more positively than incongruent messages (Cesario, Grant, & Higgins, 2004; Higgins, Idson, Freitas, Spiegel, & Molden, 2003; Lee & Aaker, 2004). A more recent sophisticated approach to message tailoring messages involves tailoring messages to a recipient’s personality traits. This promising technique was tested in the context of product advertisements by Hirsh et al. (2012). They observed message-person congruence effects by manipulating the framing of an appeal to the Big Five personality traits. Respondents in their study evaluated a mobile phone advertisement more positively when it emphasized a motivational concern that was congruent with their own personality characteristics. This kind of message tailoring is becoming increasingly practical. Several authors have shown that reliable inferences about personality can be obtained from an individuals’ Facebook profile (Back et al., 2010), e-mail address (Back,Schmukle, & Egloff, 2008), and language use (Hirsh & Peterson, 2009; Yarkoni, 2010).

processing (Baumeister, 1998). Thus, self-referencing is viewed as a mediator between an individual’s exposure to ad messages tailored to the recipient’s personality traits and their subsequent attitudes. Furthermore, we test for the mediation effect of resistance to persuasion. Ringold (2002) has demonstrated that people do not want to be influenced and that they are motivated to resist persuasion. We therefore expect that message recipients are likely to understand that the message tailoring to personality traits is utilized to influence them and as a result try to resist persuasion.

The theoretical framework proposed to explore the mediation effects attempts to achieve two research objectives. First, it uses the Big Five framework to explore the effectiveness of tailoring persuasive appeals to a recipient’s personality profile by highlighting the motivational concern associated with one of the Big Five major personality dimensions. Second, it tests a self-referencing-mediated psychological process by which these tailored messages may enhance perceived advertisement message effectiveness or prohibit them through the mediating role of resistance to persuasion. The remainder of this paper is organized as follows. Chapter two will provide a literature review. The methodology and data analysis are presented next in chapter three and four respectively. The paper continues with a discussion of the study’s finding, limitations, areas for future research, and it ends with a conclusion.

2. THEORETICAL FRAMEWORK

This chapter provides on overview of relevant literature. Section 2.1 will introduce relevant literature pertaining to message tailoring. Section 2.2.1 will introduce the Big Five Framework. Next, in section 2.2.2 and 2.2.3 literature relevant to the effect of message tailoring on self-referencing and resistance to persuasion will be discussed. Lastly, in section 2.2.4 and section 2.2.5 literature on the potential mediating influence of self-referencing and resistance to persuasion on attitude towards the message will be reviewed. Section 2.3 will provide a graphical summary of the expected relationships.

2.1 Message Tailoring

orientation are processed more fluently and evaluated more positively than incongruent messages (Cesario, Grant & Higgins, 2004; Higgins, Idson, Freitas, Spiegel, & Molden, 2003; Lee & Aaker, 2004).

One important first step is to engage in audience segmentation and carefully define the audience (Grunig, 1989; Rogers & Storey, 1987). This practice of crafting persuasive campaign messages for specific audience segments is known as message targeting (Kreuter, Strecher, & Gassman, 1999; Palmgreen & Donohew, 2003). Audiences can be segmented on a wide number of variables, and there is a significant amount of literature that describes different approaches to segmentation and targeting. For example, common variables used to segment audiences are demographic, geographic, psychographic, attitudinal, cultural, and behavioural variables (Albrecht & Bryant,1996; Goldberg, Fishbein, & Middlestadt, 1997; Slater, 1996). The focus of this study will be on message tailoring that involves adapting persuasive messages to the recipients’ personality characteristics.

2.2 Hypotheses

2.2.1 The Big Five Traits

creative, and like to come up with new ideas (McCrae & Costa, 1997). Each trait can be considered continuous, and by knowing where an individual is on the Big Five trait continuum, one can compare and differentiate among people’s behaviours across various times and places (John & Srivastava, 1999).

2.2.2 Message Tailoring to Personality Traits and Self-referencing

A self-schema can be described as a cognitive generalization about the self that is comprised of a more or less comprehensive set of traits, values and beliefs (Markus, 1977). The salience of a self-schema is a key source for inference making. The knowledge about one’s traits and personality characteristics has been shown to be abundant and has the potential to considerably influence information processing (Baumeister, 1998). The definition of self-referencing informs us how this influence on information processing might be understood. Self-referencing describes the notion that people interpret incoming information in relation to themselves, using their self-concept as a background for new information (Rogers et al., 1977). Following this definition, it is expected that messages tailored to recipients’ personality traits will stimulate them to relate the message content to themselves, using their self-concept as a background for new information. Therefore, the following hypothesis is proposed.

Hypothesis 1. Message appeals tailored to recipients’ personality traits will lead to higher levels of self-referencing.

2.2.3 Message Tailoring and Resistance to Persuasion

or behaviour. Empowerment strategies can be described as individuals asserting their own, existing views instead of challenging the persuasive communication. Counter arguing messages involves recipients actively challenging messages content, source, or persuasion tactics and is one frequently used resistance strategy by consumers (Wright, 1975; Zuwerink Jacks and Cameron, 2003). It is also the most extensively studied means of resistance, especially in a context of high processing motivation and ability (Petty & Cacioppo, 1979; Wood, Rhodes & Biek, 1995). Counter arguing strategies may involve: the content of the message, the source, or persuasive strategies used in the message. Friestad and Wright (1994) put forward in their Persuasion Knowledge Model that people hold theories and beliefs about how marketers attempt to influence them. They argue that the detection of persuasion tactics results in a change of message meaning, and this may subsequently result in resisting the persuasion attempt. As the self-concept can direct attention to information and guide behaviour (Markus and Wurf, 1987), it is expected that tailoring messages to recipient’s personality traits results in higher levels of resistance to persuasion. This is because recipients will probably understand that the message tailoring is used to influence them and start to counter argue the message. The following hypothesis is proposed.

Hypothesis 2. Message appeals tailored to recipients’ personality characteristics will result in higher levels of resistance to persuasion.

2.2.4 Self-Referencing and Message Effectiveness

persuasion when message arguments or product features are strong, but not when they are weak. In their study they instructed participants to relate ad messages to their own experiences, which resulted in strong arguments generating more favourable product attitudes than weak arguments. This is in line with the work of several other authors who have contended that advertising messages that are congruent with ad recipients’ self-concepts are more effective then incongruent messages because they trigger self-referencing (Chang, 2001a,2005b; Debevec and Iyer, 1988). As such, it is suggested that self-referencing can enhance the effectiveness of messages by making the information more meaningful.

Hypothesis 3. Self-referencing positively influences recipients’ attitude towards the message.

Furthermore, there is reason to suspect that self-referencing mediates the relationship between messages tailored to personality traits and message effectiveness. A recent study by Hoekstra et al. (2016) showed that personalization of direct mailings has a positive effect on response through referencing. Another study by Chang (2011) tested the hypothesis that self-referencing mediates the effects of the interaction between heading type and issue concern on perceived ad effectiveness. Chang found that when the message contained second-person pronouns, participants concerned with the specific issue engaged in greater self-referencing and this had a positive effect on ad effectiveness. As such, in line with prior literature on message tailoring to personality traits and self-referencing, this study proposes that self-referencing mediates the relationship between tailored appeals to recipient’s personality traits and perceived ad effectiveness. Thus, to the degree that tailoring fundraising message appeals to the recipient’s personality profile enhances self-referencing, it should improve perceived ad effectiveness.

Hypothesis 4. Tailoring message appeals to recipients’ personality traits has a positive effect on a recipient’s attitude towards the message through self-referencing. Hence, self-referencing is expected to have a mediating effect.

2.2.5 Resistance to Persuasion and Message Effectiveness

and Petty (2002) showed that when people resist persuasion, they can notice their own resistance and subsequently infer that their decision to resist persuasion must be correct. As such, the following hypothesis is proposed.

Hypothesis 5. Higher resistance to persuasion negatively influences the recipient’s attitude towards the message.

Festinger and Maccoby (1964) and Silvia (2006) have shown that a resistance strategy plays a mediating role between a persuasive message and outcomes such such as attitude and behaviour. Moreover, Hoekstra et al. (2016) also found a mediating effect of message resistance. In their research on the effect of email personalization on response they found a significant indirect effect of email personalization on website visit (response) through message resistance. The following hypothesis is proposed.

Hypothesis 6. Tailoring message appeals to recipients’ personality traits has a negative effect on a recipient’s attitude towards the message through resistance to persuasion. Hence, resistance to persuasion is expected to have a mediating effect.

2.3 Research Framework

3. METHOD

This chapter starts with a discussion of the data collection process in section 3.1. Next, the design of this study is outlined in section 3.2. Section 3.3 discusses the measurement of constructs. Finally, the method of analysis will be outlined in section 3.4.

3.1 Data Collection

We recruited 119 random passers-by at the University of Groningen in the Duisenberg Building of the Faculty of Economics and Business at the Zernike Campus. Participants were asked to participate in a Master thesis survey. After agreement they were asked for their email to which a survey link was sent. This email informed respondents that they had three days to complete the survey. To persuade participants to fill out surveys, the respondents were rewarded with a €1,30 Starbucks coupon. In total, 99 respondents completed the online survey. Just over half of the sample (50,5%) was female, and the majority of the respondents (73,7%) were students.

3.2 Design

Figure 2 The Conscientiousness advertisement

The questionnaire was administered using Qualtrics software, a web-based survey creation and distribution tool to conduct online surveys. The five advertisements were presented in random order to each participant. After reading one advertisement, respondents had to respond to several questions regarding their attitude towards the advertisements, extent of self-referencing, and extent of resistance to persuasion before being randomly assigned to another advertisement. After having completed this for all five advertisements participants’ personality was assessed with the Big Five Aspect Scales (DeYoung, Quilty, & Peterson, 2007).

3.3 Measurement of Constructs

For all questions 5-point Likert scales were used, ranging from strongly disagree to strongly

agree. The constructs were measured with existing scales, see Table 1.

Table 1: Measurements of constructs

Construct Items Source Loading

Attitude towards the advertisement

1. I find this advertisement to be persuasive; 2. This is an effective advertisement; 3. I would purchase this product after seeing

this advertisement;

4. Overall, I like this advertisement; 5. This advertisement has made me more

interested in the product; and

6. I am interested in learning more about this product after seeing this advertisement.

3.4 Method of Analysis

Since this study uses an analytical method that is comprised of latent constructs and two parallel mediators it was decided to use Partial Least Squares Structural Equation Modelling (PLS-SEM). This method was justified because it is a variance-based method that combines factor analysis and multiple regression (Hair, Ringle, and Sarstedt, 2011) and is recommended to be used to avoid reliability problems that exist when estimating mediation using regression analysis (Baron and Kenny, 1986). Moreover, regression analysis assumes there is no measurement error in the mediator. As such, it is not possible to fully control the effect of the mediator on the dependent variable when the effect of the independent variable on the dependent variable is estimated. Structural modelling methods have the advantage that they do incorporate the measurement error(s) of the mediator(s) in the model. The PLS-SEM analysis was performed using WarpPLS 3.0 software.

For the statistical procedure we selected the default option of Warp3 PLS Regression, which tries to find an S-curve relationship between latent variables was selected. If it succeeds in identifying an S-curve relationship, the scores of the predictor latent variables are transformed to reflect the curve relationships in the estimated path coefficients in the model. In case a

S-Self-referencing 1. The ad message relates to me personally; 2. The ad message made me think about my

personal experience with a mobile phone; and

3. I read the ad as if I own the xPHONE.

Burnkrant and

Ungava, 1989 Conscientiousness model: (!= 0.854) Extraversion model: (!= 0.803) Agreeableness model: (!= 0.854) Neuroticism model: (!= 0.882) Openness/Intellect model: (!= 0.850) Resistance to Persuasion

1. The ad message triggers a sense of resistance in me; and

2. I resist to the attempt of the ad message to influence me.

Hong & Faedda (1996) Conscientiousness model: (!= 0.847) Extraversion model: (!= 0.0853) Agreeableness model: (!= 0.867) Neuroticism model: (!= 0.847) Openness/Intellect model: (!= 0.868)

Personality Traits See Appendix 2. De Young, C.G., Quilty, L. C., & Peterson, J. B. (2007)

curve is not identified, it tries to fit a U-curve, and ultimately a straight line (Knock, 2012). Jackknifing was selected as a resampling method, since it is known to generate the most stable resampling coefficients with smaller samples (lower than 100), and with samples containing outliers (Knock, 2012). For each advertisement manipulation highlighting the motivational concern of one of the Big Five personality dimensions, a separate PLS-SEM model was developed, Figure 3 shows this visually for the Conscientiousness advertisement framing. Please refer to Appendix 3 for the figures of the other four PLS-SEM models.

Figure 3 –PLS-SEM model for Conscientiousness

The sample size of 99 respondents exceeds the minimum required sample size. As suggested by Chin (1998), the minimum sample size for a PLS study should be equal to the larger of the next two conditions: (1) ten times the amount of the indicators of the scale with the largest number of formative indicators, or ten times the largest number of structural paths directed to a dependent variable (in our model: seven structural paths per model).

4. RESULTS

4.1 Measurement Model

To test the adequacy of the measurement model both its convergent and discriminant validity were assessed. Convergent validity is the extent to which indicators of the same latent construct converge or share a high proportion of variance in common, while discriminant validity is the degree to which a latent construct is truly different from other latent constructs (Hair, Black, Babin, and Anderson, 2010). First, the indicator reliability was assessed, all indicator loadings that have a p-value lower than 0.001, and were equal to or greater than 0.5 met the requirements for acceptable convergent validity (Hair et al., 2010). All items that did not meet these requirements were removed from the measurement model. Only the personality construct had items that did not meet the requirements. For example, AGREE14 had a standardized indicator loading of 0.492 (p=0.001), and was removed from the measurement model. In total, 11 items were removed from the agreeableness personality dimension, 4 items from Extraversion, 9 items from Conscientiousness, 6 items from Neuroticism, and 8 items from Openness/Intellect. As the personality scales use a reflective approach to measurement, this was deemed acceptable (Coltman et al., 2008). Henseler et al. (2009) recommend the use of composite reliability in PLS path models. Accordingly, in addition to Cronbach’s alpha, composite reliability values were assessed and they show acceptable convergent validity (All values were > 0.7). Table 2 shows the assessment of the convergent validity of the final measurement model, only reflective scales were used. The construct validity and convergent validity of the personality construct was the same for all five models, whereas the results for ‘Attitude towards advertisement’, ‘Self-referencing’ and ‘resistance to persuasion’ differed for each model. See Appendix 4 for further details on construct validity and convergent validity results.

Table 2: Construct validity and convergent validity of reflective scales for the Conscientiousness model

Discriminant validity was assessed by comparing the square roots of the average variances extracted (AVE) of a specific latent construct with the correlation coefficients of that specific construct and the other latent constructs. Table 3 show the discriminant validity results for the used constructs. As can be seen in Table 3 discriminant validity holds since the square root of the AVE is higher than any of the correlations involving that construct (Fornell and Larcker, 1981). Moreover, all correlation coefficients were significant at the 0.05-level. As such, the convergent and discriminant validity statistics indicate that the measurement model is sufficiently strong to enable interpretations of structural estimates. See Appendix 5 for the Squared roots of the average variance extracted (AVE) and correlation matrix of the other four PLS-SEM models.

Table 3: Conscientiousness advertisement-- Squared roots of the average variance extracted (AVE) and correlation matrix. 1. 2. 3. 4. 5. 6. 7. 8. 1. Agreeableness 0.640 2. Extraversion 0.345 0.613 3. Conscientiousness 0.168 0.256 0.610 4. Neuroticism -0.366 -0.423 -0.493 0.635 5. Openness/Intellect 0.557 0.403 0.179 -0.440 0.677 6. Attitude towards Advertisement 0.287 0.034 0.257 -0.160 0.261 0.832 7. Self-referencing 0.341 0.170 0.341 -0.353 0.420 0.703 0.880 8. Resistance to Persuasion -0.015 -0.167 -0.223 0.130 -0.043 -0.417 -0.273 0.931 Note: The squared roots of the average variances extracted (AVE) are shown on the diagonal, while the off-diagonal coefficients represent the correlation coefficients.

4.2 Structural Model

4.2.1 Direct effects

The five model fit indices indicate that for all advertisement manipulations our structural model describes reality quite well. Table 4a shows the average path coefficients (APC) and average R-squares (ARS), average variance inflation factors (AVIFs) of all five models, with significance level. Table 4b depicts the explained variance and predictive validity measures of the endogenous latent variables, and the full collinearity measures (VIFs) of all latent variables.

Table 4a: Direct Effects – APC, ARS and AVIF coefficients.

Model/Coefficients APC ARS AVIF

Table 4b: Full Collinearity VIFs, explained variance, and predictive validity measures (PLS-SEM). Advertisement Full collinearity VIFs Explained Variance (R-square): Predictive validity (Stone-Geiser Q-square): Agreeableness advertisement 1. Agreeableness 2. Extraversion 3. Conscientiousness 4. Neuroticism 5. Openness/Intellect

6. Attitude towards advertisement 7. Self-referencing 8. Resistance to persuasion 1.530 1.370 1.345 1.738 1.728 2.091 2.480 1.673 0.410 0.376 0.372 0.403 0.459 0.751 0.774 0.883 0.630 0.028 0.013 Extraversion advertisement 1. Agreeableness 2. Extraversion 3. Conscientiousness 4. Neuroticism 5. Openness/Intellect

6. Attitude towards advertisement 7. Self-referencing 8. Resistance to persuasion 1.566 1.440 1.416 1.956 1.818 3.174 1.375 2.639 0.410 0.376 0.372 0.403 0.459 0.681 0.718 0.872 0.629 0.034 0.031 Conscientiousness advertisement 1. Agreeableness 2. Extraversion 3. Conscientiousness 4. Neuroticism 5. Openness/Intellect

6. Attitude towards advertisement 7. Self-referencing 8. Resistance to persuasion 1.587 1.426 1.439 1.750 1.799 2.411 2.409 1.320 0.410 0.376 0.372 0.403 0.459 0.691 0.774 0.867 0.577 0.131 0.071 Neuroticism

advertisement 1. 2. Agreeableness Extraversion 3. Conscientiousness 4. Neuroticism 5. Openness/Intellect

6. Attitude towards advertisement 7. Self-referencing 8. Resistance to persuasion 1.596 1.462 1.385 1.864 1.759 3.172 3.213 1.781 0.410 0.376 0.372 0.403 0.459 0.691 0.774 0.867 0.723 0.126 0.144 Openness/Intellect advertisement 1. Agreeableness 2. Extraversion 3. Conscientiousness 4. Neuroticism 5. Openness/Intellect

6. Attitude towards advertisement 7. Self-referencing 8. Resistance to persuasion 1.547 1.456 1.335 1.809 1.726 2.091 2.480 1.673 0.410 0.376 0.372 0.403 0.459 0.887 0.769 0.884 0.800 0.015 0.086

dimensions did predict effectiveness ratings. For the Extraversion advertisement, scores on Neuroticism (Beta=-0.227, p<0.05) and scores on Openness/Intellect (Beta=0.253, p<0.05) were found to predict effectiveness ratings. In the case of the advertisement targeting the personality dimension of Agreeableness, we found predictive effectiveness ratings for scores on Extraversion 0.203, p<0.001). For neuroticism, scores on Agreeableness (Beta=-0.0178, p<0.05) and Openness/Intellect (Beta=-0.258, p<0.05) were found to be predictors of effectiveness ratings. And for the advertisement targeting the Openness/Intellect personality dimension, we found a negative effect of the targeted personality dimension (Beta=-0.165, p<0.1). Table 4d shows the standardized path coefficients when the two mediators ‘self-referencing’ and ‘resistance to persuasion’ are added in the PLS-SEM model.

Table 4c: Standardized path coefficients to check for congruence effects (PLS-SEM).

Advertisement Framing

Predictor Extraversion Agreeableness Conscientiousness Neuroticism Openness/Intellect Extraversion 0.085 -0.203*** -0.058 -0.175 0.111

Agreeableness 0.056 0.097 0.217 -0.178** -0.026 Conscientiousness 0.061 -0.085 0.270** 0.083 -0.111 Neuroticism -0.227** 0.114 -0.081 0.092 0.075 Openness/Intellect 0.253** -0.025 0.117 -0.258** -0.165* Note: p-values in parentheses (***p ≤ .01;**p ≤ .05; * p ≤ 0.10).

Table 4d: Standardized path coefficients with mediators included

Advertisement type Independent dependent

Beta

Conscientiousness Conscientiousness

Attitude towards the message Self-referencing

Resistance to persuasion Agreeableness

Attitude towards the message Extraversion

Attitude towards the message Neuroticism

Attitude towards the message Openness/Intellect

Attitude towards the message Self-referencing

Attitude towards the message Resistance to persuasion

Attitude towards the message

0.091 0.352*** -0.246** 0.121 -0.021 -.167 -0.034 0.634*** -0.233** Agreeableness Agreeableness

Attitude towards the message Self-referencing

Resistance to persuasion Conscientiousness

Attitude towards the message Extraversion

Attitude towards the message Neuroticism

Attitude towards the message

4.2.2 Indirect/mediation effects

In order to test the hypothesized mediation effects of self-referencing and resistance to persuasion of advertisement effectiveness a two-step PLS-SEM analysis was performed.

Openness/Intellect

Attitude towards the message Self-referencing

Attitude towards the message Resistance to persuasion

Attitude towards the message

0.019

0.598***

-0.305***

Extraversion Extraversion

Attitude towards the message Self-referencing

Resistance to persuasion Conscientiousness

Attitude towards the message Agreeableness

Attitude towards the message Neuroticism

Attitude towards the message Openness/Intellect

Attitude towards the message Self-referencing

Attitude towards the message Resistance to persuasion

Attitude towards the message

0.062 -0.165 -0.182** 0.073 0.088 -0.015 0.126 0.687*** -0.107 Neuroticism Neuroticism

Attitude towards the message Self-referencing

Resistance to persuasion Conscientiousness

Attitude towards the message Extraversion

Attitude towards the message Agreeableness

Attitude towards the message Openness/Intellect

Attitude towards the message Self-referencing

Attitude towards the message Resistance to persuasion

Attitude towards the message

-0.004 0.354*** -0.378*** 0.029 -0.013 -.002 -0.153 0.747*** -0.049 Openness/Intellect Openness/Intellect

Attitude towards the message Self-referencing

Resistance to persuasion Conscientiousness

Attitude towards the message Extraversion

Attitude towards the message Neuroticism

Attitude towards the message Agreeableness

Attitude towards the message Self-referencing

Attitude towards the message Resistance to persuasion

Attitude towards the message

WarpPLS3.0 only estimates the total indirect effect and its associated p-value, a second step was necessary to estimate the p-value of the mediation effect of each mediator separately. The mediation effect of each mediator was isolated one at a time, by deleting the links between the independent variable and the other mediator variable, without altering the coefficients of the structural model. Figure 4a shows depicts this process conceptually for the ‘conscientiousness’-model, where we estimate the mediation effect of self-referencing. We estimate the hypothesized indirect effect between a tailored ‘conscientiousness’-message and respondents’ effectiveness rating behaviour by isolating the mediation effect. This is done by deleting the link between a respondents’ conscientiousness personality score variable (shown as ‘Consct (R)20i’ in Figure 4a) and resistance to persuasion variable (shown as ‘Con_res (R)2i’ in Figure 4a. In Figure 4b, one can see how this would apply for estimating the hypothesized mediation effect of resistance to persuasion. The results of the mediation analysis are shown in Table 5.

Figure 4b: Estimating the mediating effects of self-referencing and resistance to persuasion for the Conscientiousness advertisement using a two-step PLS-SEM analysis.

Table 5: Estimated mediation effects (PLS-SEM)

Advertisement effectiveness rating

Mediation effect Standardized Coefficients (p-values): Effect size

M+),-./)-)/)(#'(0 0.264*** 0.950 M/)+'+12(#) 1$ 4)/+52+'$( 0.018 0.065 Openness/Intellect advertisement effectiveness rating Total effect -0.266 M#$%&'()* -0.149** -0.117 0.560 M+),-./)-)/)(#'(0 -0.067 0.252 M/)+'+12(#) 1$ 4)/+52+'$( -0.082** 0.308

Note: p-values in parentheses (***p ≤ .01;**p ≤ .05; * p ≤ 0.10).

1 _{This is a case of inconsistent mediation (MacKinnon, Fairchild, and Fritz, 2007). Therefore, the estimated strength of mediation is not}

informative in these cases.

In order to test the hypotheses, the estimated direct effects and the estimated indirect effects were taken into consideration. First, for each type of advertisement framing we discuss the direct effects of hypotheses 1-3 and hypothesis 5 using the results depicted by Table 4d. Second, the mediation hypotheses will be discussed using the results presented in Table 5. Third, we check the robustness of the outcomes of the mediation analysis by performing one additional method developed by Hayes and Preacher (2010), using the WarpPLS generated path coefficients and standard errors.

Hypothesis testing for the Conscientiousness advertisement framing

A recipient’s score on the targeted conscientiousness personality dimension was found to significantly affect the level of self-referencing positively (β=0.352, p<0.001). Contrary to expectation, a significant negative effect on resistance to persuasion was found (β=-0.246, p<0.05). Hence, the data indicate support for H1 and a lack of support for H2. Self-referencing was found to positively affect a recipient’s attitude towards the message (β=0.634, p<0.01), thereby, indicating support for H3. We found a negative effect of resistance to persuasion on attitude towards the message (β=-0.233, p<0.01). Hence, H5 was supported. Furthermore, the data indicates that self-referencing significantly partially mediates the relationship between the tailored message appeal and advertisement effectiveness positively (ab=0.223, p<0.01). The estimated strength of mediation is 0.601. Hence, H4 is fully supported. The mediation effect of resistance to persuasion was not found to be significant at the p=0.5 level (ab=0.057, p<0.1), we conclude that H6 was not supported.

Hypothesis testing for the Agreeableness advertisement framing

score on the targeted personality dimension was found to be insignificant on resistance to persuasion (β=0.101, p>0.05). Self-referencing was found to positively influence a recipient’s attitude towards the message highlighting the motivational concern of Agreeableness (β=0.598, p<0.01). Moreover, we found a significant negative effect of resistance to persuasion on attitude towards the message (β=-0.305, p<0.01). Hence, the data indicate support for H3 and H5. The mediation effect of self-referencing was found to be significant (ab=0.096, p<0.01), hence, H4 was supported. The estimated strength of mediation could not be calculated, because of inconsistent mediation (MacKinnon, Fairchild, and Fritz, 2007). Contrary to expectation, no significant effect was found for the mediating effect of resistance to persuasion (ab=-0.031, p>0.05). Thus, H6 was not supported by the data from the Agreeableness advertisement framing.

Hypothesis testing for the Extraversion advertisement framing

In case of the Extraversion advertisement framing we found no support for H1 (β=-0.165, p>0.05). Contrary to expectation, we found a significant negative effect of a recipient’s Extraversion personality score on resistance to persuasion (β=-0.182, p<0.05), hence, H2 was also not supported. Furthermore, we found that self-referencing positively influences attitude towards the message (β=0.687, p<0.01), indicating support for H3. Resistance to persuasion was not found to have a significant effect on attitude towards the tailored message (β=-0.107, p>0.05), hence, H5 was not supported. The mediating effect of self-referencing (ab=0.113, p>0.05) and resistance to persuasion (ab=0.019, p>0.05) was not significant. Hence, the data indicate a lack of support for H4 and H6 in the case of the Extraversion advertisement framing.

Hypothesis testing for the Neuroticism advertisement framing

the mediating effect of resistance to persuasion was positive and not found to be significant at the p=0.05 level (ab=0.018, p>0.05), so we conclude that H6 was not supported.

Hypothesis testing for the Openness/Intellect advertisement framing

For the Openness/Intellect advertisement we did not find a significant effect of a recipient’s score on the Agreeableness personality dimension on self-referencing (β=-0.094, p>0.05), indicating a lack of support for H1. However, we did find support for H2, the positive effect of a participant’s score on the targeted personality dimension on resistance to persuasion was found to be significant (β=0.289, p<0.01). Furthermore, self-referencing was found to positively influence a recipient’s attitude towards the message highlighting the motivational concern of Agreeableness (β=0.709, p<0.01). We also found a significant negative effect of resistance to persuasion on attitude towards the message (β=-0.283, p<0.01). Hence, the data indicate support for H3 and H5. The mediation effect of self-referencing was found not to be significant (ab=-0.067, p>0.05), hence, H4 was not supported. As expected, a significant effect was found for the mediating effect of resistance to persuasion (ab=-0.082, p<0.01), hence, H6 was supported by the data from the Openness/Intellect advertisement framing. The estimated strength of mediation is 0.308. The results are summarized in Table 6.

Table 6: Overview of supported and rejected hypotheses (PLS-SEM)

Advertisement framing Hypothesis H1-H6 Summary Supported/rejected Summary Conscientiousness H1 H2 H3 H4 H5 H6 Summary Supported Rejected Supported Supported Supported Rejected

Mediation effects found for self-referencing. Agreeableness H1 H2 H3 H4 H5 H6 Summary Supported Rejected Supported Supported Supported Rejected

4.2.3 Robustness check

In order to test the robustness of the outcomes of the mediation analysis we performed one additional test to assess mediation effects. This alternative test, developed by Hayes and Preacher (2010) was based on the estimation of indirect effects and related p-values calculated by the WarpPLS software. From Table 7 we can conclude that the results of the robustness check are almost identical to the indirect effect estimates of Table 5, which indicates that these estimates are stable outcomes. There is only one difference, for the Agreeableness advertisement framing the mediation effect of self-referencing is not found to be significant.

Table 7: Estimated mediation effects (PLS-SEM, Hayes and Preacher (2010)).

Advertisement Mediation effect: Indirect effect: t-value: p-value:

Conscientiousness M+),-./)-)/)(#'(0 0.2232 3.3348 0.0006*** M/)+'+12(#) 1$ 4)/+52+'$( 0.0573 1.4739 0.0719* Extraversion M+),-./)-)/)(#'(0 0.1134 1.5765 0.0581* M/)+'+12(#) 1$ 4)/+52+'$( 0.0195 0.8171 0.2079 Agreeableness M+),-./)-)/)(#'(0 0.0963 0.5306 0.2985 M/)+'+12(#) 1$ 4)/+52+'$( -0.0308 -0.8636 0.1950 Neuroticism M+),-./)-)/)(#'(0 0.2627 4.0779 0.0000*** M/)+'+12(#) 1$ 4)/+52+'$( 0.0189 0.5754 0.2832 Openness/Intellect M+),-./)-)/)(#'(0 -0.0666 -0.9415 0.1744 M/)+'+12(#) 1$ 4)/+52+'$( -0.0818 -2.2274 0.0141**

Note: p-values in parentheses (***p ≤ .01;**p ≤ .05; * p ≤ 0.10). Neuroticism H1 H2 H3 H4 H5 H6 Summary Supported Rejected Supported Supported Supported Rejected

Mediation effects only found for self-referencing Openness/Intellect H1 H2 H3 H4 H5 H6 Summary Rejected Supported Supported Rejected Supported Supported

5. DISCUSSION, LIMITATIONS, AND CONCLUSION

This chapter starts with a discussion of the results. Next, it provides the limitations of the study and directions for future research. It ends with a conclusion.

5.1 Discussion

The current study replicated the study of Hirsh et al. (2012) in the sense that it also included a comprehensive model of recipient’s personality traits. Moreover, we added two mediators to our theoretical framework: self-referencing and resistance to persuasion, to study the mechanics by which tailored message appeals influence message effectiveness. However, contrary to the findings of Hirsh et al. (2012) we only found message-person congruence effects for the Conscientiousness advertisement. This finding provides a lack of support for the conceptual premise that tailoring a message to an individual’s personality profile positively influences the recipient’s attitude towards the message. A possible explanation for why we did not find message-person congruence effects in the other four advertisements might be that the messages did not include information on attributes such as price, screen size, screen resolution and quality, the type of camera, and battery life. Kreuter et al. (1999) showed that congruently framed messages result in increased attention, which often appears to be associated with a more positive evaluation. However, it may also result in more negative evaluations when the message itself is of low quality (Updegraff et al. 2007). One question that needs to be asked, however, is why we did find congruence effects for the Conscientiousness advertisement. It is difficult to explain this result, but it might be related to the fact that the message highlighted how the xPHONE could boost their productivity and help them stay focused on their long-term plans. Since the majority of the sample was student, it is possible that this was a benefit relevant to their lifestyle as a student.

the message source. This is in line with Sagarin et al. (2002), who demonstrate that resistance to persuasion can be observed as the lack of attitude change in response to a persuasion attempt.

5.2 Limitations and Future Research Directions

A first limitation of this study is that the current study focuses solely on the evaluations of advertisements for a smartphone. This limits the generalizability of the findings, hence, future research could focus on other persuasive appeals, such as health-promotion and political campaigns. Secondly, the advertisements did not include information on attributes such as price, battery life, or screen size. As a result, respondents might have found it confusing to evaluate the advertisement of a smartphone brand that did not indicate this type of information. In preference measurement, products (such as smartphones) represent attribute bundles, i.e., combinations of attribute levels. Future research could use conjoint measurement techniques (such as conjoint analysis) to make sure that the advertisements feature relevant attributes. Subsequently, psychographic segmentation could be utilized to measure psychological variables of the targeted audience with the goal of identifying individuals who will be most receptive to these attributes. This will result in a group of individuals who have common needs and wants, but have different personality traits. Ahmad et al. (2010) demonstrated that psychographic segmentation is one of the best approaches to better understand customers. Hence, researchers could subsequently test whether tailoring smartphone advertisements (that also include relevant attributes) positively influences recipients’ attitudes towards the messages.

5.3 Conclusion

This study set out to replicate the work of Hirsh et al. (2012) by examining whether message-person congruence effects can be obtained by framing persuasive messages in terms of the Big Five personality dimensions (Goldberg, 1990). Furthermore, it examined the mediating effects of self-referencing and resistance to persuasion on the relationship between messages tailored to recipients’ personality traits and attitude towards the message.

REFERENCES

Albrecht, Terrance L. and Bryant Carol (1996), “Advances in Segmentation Modelling for Health Communication and Social Marketing Campaigns,” Journal of Health Communication, 1 (January-March), 65-83.

Back, Mitja D., Schmukle, Stefan C., and Boris Egloff (2008), “How Extraverted is honey.bunny77@hotmail.de? Inferring Personality from E-mail Addresses,” Journal of

Research in Personality, 42 (August), 1116–1122.

Back, Mitja D., Stopfer, Juliane M., Vazire, Simine, Gaddis, Sam, Schmukle, Stefan C., Egloff, Boris, and Samuel D. Gosling (2010), “Facebook Profiles Reflect Actual Personality, Not Self-Idealization,” Psychological Science, 21 (3), 372–374.

Baron, Reuben M., and David A. Kenny (1986), “The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic and Statistical Considerations,”

Journal of Personality and Social Psychology, 51 (6), 1173-1182.

Baumeister, Roy F (1998), “The Self,” in The Handbook of Social Psychology, Vol. 1, Daniel T. Gilbert, Susan T. Fiske, and Gardner Lindzey, eds. New York, NY: McGraw-Hill, 680–740. Burnkrant, Robert E. and H. Rao Unnava (1989), "Self-referencing: A strategy for Increasing Processing of Message Content," Personality and Social Psychology Bulletin, 15 (December), 628-638.

Carver, Charles S., Steven K. Sutton, and Michael F. Scheier (2000). “Action, Emotion, and Personality: Emerging Conceptual Integration,” Personality and Social Psychology Bulletin, 26 (6), 741–751.

Cesario, Joseph, Heidi Grant, and E. Torry Higgins (2004), “Regulatory Fit and Persuasion: Transfer from “Feeling Right.”,” Journal of Personality and Social Psychology, 86 (3), 388– 404.

--- (2005b), “Ad-Self-Congruency Effects: Self-Enhancing Cognitive and Affective Mechanisms,” Psychology and Marketing, 22 (November), 887–910.

--- (2011), “Enhancing Self-referencing to Health Messages,” The Journal of Consumer

Affairs, 45 (1), 147-164.

Coltman, Tim, Timothy M. Devinney, David F. Midgley, and Sunil Venaik (2008), “Formative versus Reflective Measurement Models: Two Applications of Formative Measurement,”

Journal of Business Research, 61 (12), 1250-1262.

Chin, Wynne W. (1998), “The Partial Least Square Approach to Structural Equation Modelling,” in Modern methods for business research, Vol. 2, George A. Marcoulides, ed. Mahwah, NJ: Lawrence Erlbaum Associates, 295-336.

Debevec, Kathleen, and Easwar Iye (1988), “Self-Referencing as a Mediator of the Effectiveness of Sex-Role Portrayals in Advertising,” Psychology and Marketing, 5 (Winter), 71–84.

DeYoung, Colin G., Lena C. Quilty, and Jordan B. Peterson (2007), “Between Facets and Domains: 10 Aspects of the Big Five,” Journal of Personality and Social Psychology, 93 (5), 880–896.

Digman, John M (1997), “Higher-Order Factors of the Big Five,” Journal of Personality and

Social Psychology, 73 (6), 1246–1256.

Dijkstra, Arie (2008), “The Psychology of Tailoring-ingredients in Computer-tailored Persuasion,” Social & Personality Psychology Compass, 2 (2), 765-784.

Festinger, Leon, and Nathan Maccoby (1964), “On Resistance to Persuasive Communications,”

The Journal of Abnormal and Social Psychology, 68 (4), 359.

Fornell, Claes, and David F. Larcker (1981), “Evaluating Structural Equation Models with Unobservable Variables and Measurement error,” Journal of Marketing Research, 18(1): 39-50.

Goldberg, Lewis. R (1990), “An Alternative “Description of Personality”: The Big-Five Factor Structure,” Journal of Personality and Social Psychology, 59 (6), 1216–1229.

Goldberg, Marvin E., Martin Fishbein, and Susan E. Middlestadt (1997), Social Marketing:

Theoretical and practical perspectives. Manwah, NJ: Lawrence Erlbaum Associates.

Graziano, William G., and Renée M. Tobin (2002), “Agreeableness: Dimension of personality or social desirability artifact?,” Journal of Personality, 70 (5), 695–727.

---, and Eisenberg, Nancy (1997), “Agreeableness: A dimension of personality,” in Handbook

of personality psychology, R. Hogan, S. Briggs, and J. Johnson, eds. San Diego, Academic

Press, 795– 824.

---, Meara M. Habashi, Brad E. Sheese, and Renée M. Tobin (2007), “Agreeableness, Empathy, and Helping: A Person Situation Perspective,” Journal of Personality and Social Psychology,

93 (4), 583–599.

Grunig, James (1989), “Publics, Audiences, and Market Segments: Segmentation Principles for Campaigns,” in Information campaigns: Balancing social values and social change, C. Salmon, ed. Newbury Park, CA: Sage Publications, 199-228.

Hair, Joseph F., Christian M. Ringle, and Marko Sarstedt (2011), “PLS-SEM: Indeed a Silver Bullet,” Journal of Marketing Theory and Practice, 19(2), 139-152.

---, William C. Black, Babin, Barry J., and Anderson, Rolph E (2010), Multivariate data

analysis: A global perspective. Upper Saddle River, NJ: Pearson.

Hayes, Andrew F., and Kristopher J. Preacher (2010), “Quantifying and Testing Indirect Effects in Simple Mediation Models when the Constituent Paths Are Nonlinear,” Multivariate

Heatherton, Todd F., Neil C. Macrae, and William M. Kelley (2004), "What the Social Brain Sciences Can Tell Us About the Self." Current Directions in Psychological Science, 13 (5), 190-193.

Henseler, Jörg, Christian M. Ringle, and, Rudolf R. Sinkovics (2009), “The Use of Partial Least Squares Path Modelling” in International Marketing,” Advances in International Marketing, 20 (1), 277-319.

Hirsh, Jacob B., Sonia K. Kang, and Galen V. Bodenhausen (2012), “Personalized Persuasion: Tailoring Persuasive Appeals to Recipients’ Personality Traits,” Psychological Science, 23 (6), 578-58.

---, and Jordan B. Peterson (2009), “Personality and Language Use in Self-Narratives,” Journal

of Research in Personality, 43 (3), 524–527.

---, and Michael Inzlicht (2008), “The Devil You Know: Neuroticism Predicts Neural Response to Uncertainty,” Psychological Science, 19 (10), 962–967.

Higgins, E. Tory, Lorraine C. Idson, Antonio L. Freitas, Scott Spiegel, and Daniel C. Molden (2003), “Transfer of Value from Fit,” Journal of Personality and Social Psychology, 84 (6), 1140–1153.

Hoekstra, Janny C., Daniela A. Naydenova, and Jaap E. Wieringa (2016), “The Effect of Personalization and Message Framing on Marketing Communication Processing and Response,” Working Paper.

Hong, Sung-Mook, and Salvatora Faedda (1996), “Refinement of the Hong Psychological Reactance Scale,” Educational and Psychological Measurement, 56 (1), 173-182.

Jensen-Campbell, Lauri A. Jensen-Campbell, Ryan Adams, David G. Perry, Katie A. Workman, Janine Q. Furdella, and Susan K. Egan (2002), “Agreeableness, Extraversion, and Peer Relations in Early Adolescence: Winning Friends and Deflecting Aggression,” Journal of

Research in Personality, 36 (3), 224–251.

Jenkins, Mikayla, and Marko Dragojevic (2011), “Explaining the Process of Resistance to Persuasion: A Politeness Theory-Based Approach,” Communications Research, 38 (5), 1-32. John, Oliver P., and Sanjay Srivastava (1999), “The Big Five Trait Taxonomy: History, Measurement, and Theoretical Perspectives,” Handbook of personality: Theory and research, 1999 (2), 102-138.

Kock, Ned, and Gary Lynn (2012), “Lateral Collinearity and Misleading Results in Variance-Based SEM: An illustration and recommendations,” Journal of the association for information

systems, 13 (7): 546-580.

Komiak, Sherrie Y, and Izak Benbasat (2006), “The Effects of Personalization and Familiarity on Trust and Adoption of Recommendation Agents,” MIS Quarterly, 30 (4), 941-960.

Kreuter, Mathew W., David Farell, Laura Olevitch, and Laura Brennan (2000), Tailoring

Health Messages: Customizing Communication with Computer Technology. Mahweh, NJ:

Lawrence Erlbaum Associates.

---, Fiona C. Bull, Eddie M. Clark, and Debra L. Oswald (1999), “Understanding How People Process Health Information: A Comparison of Tailored and Nontailored Weight-Loss Materials,” Health Psychology, 18 (5), 487-494.

---, and Ricardo J. Wray (2003), “Tailored and Targeted Health Communication: Strategies for Enhancing Information Relevance,” American Journal of Health Behavior, 27 (Supplement 3), S227–S232.

---, Strecher, Victor J. and Bernard Glassman (1999), “One Size Does not Fit All: The Case for Tailoring Print Materials,” Annals of Behavioral Medicine, 21 (4), 276-283.

Lee, Angela Y., and Jennifer L. Aaker. (2004), “Bringing the Frame into Focus: The Influence of Regulatory Fit on Processing Fluency and Persuasion,” Journal of Personality and Social

Psychology, 86 (2), 205–218.

Lucas, Richard E., Ed Diener, Alexander Grob, Eunkook M. Suh, and Liang Shao (2000), “Cross-Cultural Evidence for the Fundamental Features of Extraversion,” Journal of

Markus, Hazel (1977), “Self-Schemata and Processing Information About the Self,” Journal of

Personality and Social Psychology, 35 (February), 63–78.

---, and Elissa Wurf (1987), “The Dynamic Self-Concept: A Social Psychological Perspective,”

Annual Review of Psychology, 38 (1), 299-337.

McCrae, Robert R., and Paul T. Costa Jr. (1997). “Conceptions and Correlates of Openness to Experience,” in Handbook of personality psychology, Robert Hogan, John A. Johnson, and Stephan R. Briggs, eds. San Diego, CA: Academic Press, 825– 847.

Palmgreen, Philip, and Lewis Donohew (2003), “Effective Mass Media Strategies for Drug Abuse Prevention Campaigns,” in Handbook of drug abuse prevention: Theory, science, and

practice, Zili Slobada & William J. Bukoski, eds. New York: Kluwer Academic/Plenum

Publishers, 27-43.

Penner, Louis A., John F. Dovidio, Jane A. Piliavin, and David A. Schroeder (2005), Prosocial Behaviour: Multilevel Perspectives. Annual Review of Psychology, 56 (September) 365–392. ---, Barbara A. Fritzsche, J. Philip Craiger, and Tamara S. Freifeld (1995), “Measuring the Prosocial Personality,” Advances in Personality Assessment, 10, 147–163.

---, Barbara A. Fritzsche, J. Philip Craiger, and Tamara S. Freifeld (1995), “Measuring the Prosocial Personality,” in Advances in Personality Assessment, Vol. 10, John W. Butcher and Charles D. Spielberger, eds. Hillsdale, NJ: Erlbaum, 147-163.

Petty, Richard E., and John T. Cacioppo. (1979), “Issue Involvement can Increase or Decrease Persuasion by Enhancing Message-Relevant Cognitive Responses,” Journal of personality and social psychology, 37 (10), 1915–1926.

Ringold, D. Jones (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.

Rogers, Everett M., and J. Douglas Storey (1987), “Communication Campaigns,” in Handbook

of communication science, Charles R. Berger and Steven. H. Chaffee, eds. Bevery Hills, CA:

Sage publications, 817–846.

Rogers, Timothy B., Nicholas A. Kuiper, and William S. Kirker (1977), "Self Referencing and the Encoding of Personal Information,” Journal of Personality and Social Psychology, 35 (9), 677-688.

Sagarin, Brad J., Robert B. Cialdini, William. E. Rice, and Sherman B. Serna (2002), “Dispelling the Illusion of Invulnerability: The Motivations and Mechanisms of Resistance to Persuasion,” Journal of Personality and Social Psychology, 83 (3), 526–541.

Silvia, Paul J. (2006), “Reactance and the Dynamics of Disagreement: Multiple Paths from Threatened Freedom to Resistance to Persuasion,” European Journal of Social Psychology, 36 (5), 673–685.

Slater, Michael D. (1996), “Theory and Method in Health Audience Segmentation,” Journal of

Health Communication, 1 (3), 267-283.

Tormala, Zakary L., and Richard E. Petty. (2002), “What Doesn’t Kill Me Makes Me Stronger: The Effects of Resisting Persuasion on Attitude Certainty,” Journal of Personality and Social

Psychology, 83 (6), 1298–1313.

Updegraff, John A., David K. Sherman, Faith S. Luyster, and Traci L. Mann (2007), “The Effects of Message Quality and Congruency on Perceptions of Tailored Health Communications,” Journal of Experimental Social Psychology, 43 (2), 249-257.

Wind, Jerry, and Arvind Rangaswamy (2001), “Customization: The Next Revolution in Mass Customization,” Journal of Interactive Marketing, 15(1), 13-32.

Wood, Wendy, Nancy Rhodes, and Micheal Biek (1995), “Working Knowledge and Attitude Strength: An Information-Processing Analysis,” in Attitude strength: Antecedents and

consequences Richard E. Petty and Jon A. Krosnick, eds. Mahwah, NJ: Lawrence Erlbaum

Associates, 283-313.

Wright, Peter (1975), “Factors Affecting Cognitive Resistance to Advertising,” Journal of

Yarkoni, Tal (2010), “Personality in 100,000 Words: A Large-scale Analysis of Personality and Word Use Among Bloggers,” Journal of Research in Personality, 44 (3), 363–373.

Jacks, J. Zuwerink, and Kimberly A. Cameron (2003), “Strategies for Resisting Persuasion,”

APPENDICES

APPENDIX 1: TEXT FROM THE ADVERTISEMENTS (HIRSH ET

AL., 2012)

The first two sentences of each advertisement were always the same: “When creating xPhone, designers and engineers didn’t start with a clean sheet of paper. They started with years of experience designing and building the phones that redefined what a phone can do”. After these initial sentences, the description was tailored to the motivational concerns associated with each personality dimension as described below.

For extraversion: “XPhone is the result of everything they’ve learned so far, a phone designed for strong, active, outgoing people like you. With the new XPhone, you’ll always be where the excitement is. You’re the life of the party, and the XPhone will keep you in the spotlight. Experience the fun and rewards of the latest technology, and express yourself to the world. Get in touch with your social side and harness the power of the XPhone to get you more of the attention you deserve.”

For agreeableness: “XPhone is the result of everything they’ve learned so far, built from the ground up to bring people together. With the new XPhone, you’ll have access to your loved ones like never before, so you can always stay in touch with your community. Designed with empathy and consideration, the XPhone helps you take the time for the people you care most about. Join the XPhone family and find out where you belong. Get in touch with your caring side and the XPhone.”

For conscientiousness: “XPhone is the result of everything they’ve learned so far, a phone that works with your schedule. With the new XPhone, you’ll never miss an important message, simplifying your work life. Designed to boost your productivity, the latest XPhone helps you stay focused on your long-term plans, no matter where you are. Achieve your goals with the latest technology, and reach new levels of personal efficiency. Organize your life with the XPhone.”

For neuroticism: “XPhone is the result of everything they’ve learned so far, and is a phone that will keep you safe. Designed to keep you safe and sound, the XPhone helps reduce the anxiety and uncertainty of modern life. With access to information about the world around you, you’ll never be in the dark again. Experience the security and protection of having instant access to the help you need, no matter where you are. The new XPhone is your defense against the unknown and a shield against emergencies. Stay safe and secure with the XPhone.”

APPENDIX 2: MEASUREMENT OF CONSTRUCTS – THE BIG

FIVE ASPECT SCALES

The Big Five Aspect Scales (DeYoung et al., 2007)

Personality Dimension Key (-/+) Item and Key (+ keyed/- keyed)

Neuroticism + + + + + + - - - - + + + + + + - - - -

Get angry easily. Get upset easily. Change my mood a lot.

Am a person whose moods go up and down easily. Get easily agitated.

Can be stirred up easily. Rarely get irritated.

Keep my emotions under control. Rarely lose my composure. Am not easily annoyed.

Am filled with doubts about things. Feel threatened easily.

Worry about things. Am easily discouraged. Become overwhelmed by events. Am afraid of many things. Seldom feel blue.

Feel comfortable with myself. Rarely feel depressed. Am not embarrassed easily. Agreeableness + + + + + - - - - - + + + + - - - - - -

Feel others’ emotions.

Inquire about others’ well-being. Sympathize with others’ feelings. Take an interest in other people’s lives. Like to do things for others.

Am not interested in other people’s problems. Can’t be bothered with other’s needs. Am indifferent to the feelings of others. Take no time for others.

Don’t have a soft side. Respect authority. Hate to seem pushy.

Avoid imposing my will on others. Rarely put people under pressure. Insult people.

Believe that I am better than others. Take advantage of others. Seek conflict.

Love a good fight.

Am out for my own personal gain. Conscientiousness + + + + - - - - - - + + + + + + - - - -

Carry out my plans. Finish what I start. Get things done quickly. Always know what I am doing. Waste my time.

Find it difficult to get down to work. Mess things up.

Don’t put my mind on the task at hand. Postpone decisions.

Am easily distracted. Like order. Keep things tidy. Follow a schedule.

Extraversion + + + + + - - - - - + + + + + + - - - -

Make friends easily Warm up quickly to others. Show my feelings when I’m happy. Have a lot of fun.

Laugh a lot.

Am hard to get to know. Keep others at a distance. Reveal little about myself.

Rarely get caught up in the excitement. Am not very enthusiastic person. Take charge.

Have a strong personality. Know how to captivate people. See myself as a good leader. Can talk others into doing things. Am the first to act.

Do not have an assertive personality. Lack the talent for influencing people. Wait for others to lead the way. Hold back my opinions. Openness/Intellect + + + + + + + - - - + + + + + + - - - -

Am quick to understand things. Can handle a lot of information. Like to solve complex problems. Have a rich vocabulary. Think quickly. Formulate ideas clearly.

Have difficulty understanding abstract ideas. Avoid philosophical discussions.

Avoid difficult reading material. Learn things slowly.

Enjoy the beauty of nature. Believe in the importance of art. Love to reflect on things. Get deeply immersed in music.

See beauty in things that others might not notice. Need a creative outlet.

Do not like poetry. Seldom get lost in thought. Seldom daydream.

Seldom notice the emotional aspects of paintings and pictures.

APPENDIX 3: PLS-SEM MODELS FOR EACH TARGETED

PERSONALITY DIMENSION

Agreeableness PLS-SEM model

Neuroticism PLS-SEM model

APPENDIX 4: CONSTRUCT VALIDITY AND CONVERGENT

VALIDITY

Table 4.1: Construct validity and convergent validity of reflective scales

Advertisement Constructs/indicators Standard

loadings: Cronbach’s alpha: Composite reliability:

Agreeableness Attitude towards advertisement ATTITUDE1 ATTITUDE2 ATTITUDE3 ATTITUDE4 ATTITUDE5 ATTITUDE6 0.856 0.843 0.758 0.915 0.912 0.906 0.933 0.948 Self-referencing SELFREF1 SELFREF2 SELFREF3 0.858 0.890 0.892 0.854 0.911 Resistance to persuasion RESIST1 RESIST2 0.940 0.940 0.933 0.948

Extraversion Attitude towards advertisement ATTITUDE1 ATTITUDE2 ATTITUDE3 ATTITUDE4 ATTITUDE5 ATTITUDE6 0.839 0.818 0.583 0.913 0.881 0.873 0.902 0.926 Self-referencing SELFREF1 SELFREF2 SELFREF3 0.865 0.832 0.844 0.803 0.884 Resistance to persuasion RESIST1 RESIST2 0.934 0.934 0.853 0.931

Openness/Intellect Attitude towards advertisement ATTITUDE1 ATTITUDE2 ATTITUDE3 ATTITUDE4 ATTITUDE5 ATTITUDE6 0.968 0.943 0.884 0.947 0.955 0.952 0.974 0.979 Self-referencing SELFREF1 SELFREF2 SELFREF3 0.899 0.875 0.858 0.850 0.909 Resistance to persuasion RESIST1 RESIST2 0.940 0.940 0.868 0.938

APPENDIX 5: SQUARED ROOTS OF THE AVERAGE

VARIANCE EXTRACTED (AVE) AND CORRELATION MATRIX

1. 2. 3. 4. 5. 6. 7. 8. 1. Agreeableness 0.640 2. Extraversion 0.345 0.613 3. Conscientiousness 0.168 0.256 0.610 4. Neuroticism -0.366 -0.423 -0.493 0.635 5. Openness/Intellect 0.557 0.403 0.179 -0.440 0.677 6. Attitude towards advertisement 0.012 -0.088 -0.099 0.077 -0.355 -0.867 7. Self-referencing 0.144 -0.016 0.027 -0.067 0.103 0.705 0.880 8. Resistance to persuasion 0.093 0.041 0.100 -0.200 0.217 -0.434 -0.277 0.940

Extraversion advertisement – Squared roots of the AVE and correlation matrix

1. 2. 3. 4. 5. 6. 7. 8. 1. Agreeableness 0.640 2. Extraversion 0.345 0.613 3. Conscientiousness 0.168 0.256 0.610 4. Neuroticism -0.366 -0.423 -0.493 0.635 5. Openness/Intellect 0.557 0.403 0.179 -0.440 0.677 6. Attitude towards advertisement 0.129 0.192 0.146 -0.269 0.295 0.825 7. Self-referencing 0.248 0.145 0.053 -0.362 0.362 0.765 0.847 8. Resistance to persuasion 0.121 -0.088 0.036 -0.064 0.103 -0.355 -0.369 0.934

Neuroticism advertisement – Squared roots of the AVE and correlation matrix

1. 2. 3. 4. 5. 6. 7. 8. 1. Agreeableness 0.640 2. Extraversion 0.374 0.596 3. Conscientiousness 0.168 0.276 0.610 4. Neuroticism -0.366 -0.445 -0.493 0.635 5. Openness/Intellect 0.556 0.446 0.174 -0.442 0.664 6. Attitude towards advertisement -0.306 -0.324 -0.050 0.341 -0.355 0.912 7. Self-referencing -0.347 -0.345 -0.066 0.350 -0.287 0.817 0.899 8. Resistance to persuasion 0.364 0.248 0.049 -0.369 0.347 -0.563 -0.613 0.931

Openness/Intellect - – Squared roots of the AVE and correlation matrix