The effects of negative news about
celebrity endorsers in sports advertising
Author: P.T. Westra
The effects of negative news about
celebrity endorsers in sports advertising
Master Thesis MSc Business Administration, Marketing Management
Author P.T. Westra
Address: Legebuorren 61, 8525 GX, Langweer Phone number +31 (0)6 46580239
Mail ptwestra@hotmail.com Student number s1903276
Organization University of Groningen Faculty Economics and Business
1st Supervisor University of Groningen Prof. dr. T.H.A. Bijmolt 2nd Supervisor University of Groningen: Dr. J. Liu
MANAGEMENT SUMMARY
This thesis studied the effects of negative news about a celebrity endorser on brand attitude and purchase intention, and which factors increase or decrease this influence. This was examined by exposing consumers to negative and positive sport and non-sport web articles with different advertisements on the same page.
There are numerous examples of sports celebrities who created negative news around them, for example Tiger Woods. Knittel and Stango (2009) estimated that EA sports lost a 4% market value and Gillette lost 2 % market value due to the Tiger Woods scandal. So these effects were big. But it is not clear which factors increase or decrease this
influence of negative news about the celebrity endorser towards brand attitude and purchase intention. Nike has spent in 2006 476 million dollars in exchange for
endorsements by celebrity athletes, this was part of the company's 1.7 billion dollars advertising budget for that year. It is a multi million dollar business and it deserves more research in the effects of this negative news around the celebrity endorser towards brand attitude and purchase intention.
To test the hypotheses two general linear models combining ANOVA and regression were performed, one with the dependent variable brand attitude and one with
trough negative news about the celebrity in advertisements. They will or won’t buy the shoes and clothes of Nike even if the Nike celebrity endorser comes badly into the news. This effect were not strengthened when it was a sports celebrity or when it was
negative sport news. People made not a distinction between a sports celebrity and non-sports celebrity when these came negatively into the news when they intend to buy Nike shoes or clothes.
As with any other study, there are several limitations and advices for future research. Most important limitation of this research is that in this experiment only different WebPages were used and there was only one moment of contact with the negative news. Future researchers should also use a couple of different negative experiences, for example show a ‘negative’ WebPage, show a ‘negative’ newspaper article, let them hear a ‘negative’ radio fragment about the endorser etc.
This research will not be the last one about the effects of negative news about
celebrities in celebrity endorsements. Nowadays it is much easier for celebrities to come negatively into the news. Most people have a cell phone with a camera, so celebrities can be filmed or photographed at any time of the day, for example in night clubs or restaurants. In addition to this social media plays an important role, a lot of people have Twitter or Facebook where they can post tweets and messages of the activities
PREFACE
With this master thesis I finish my study MSc Business Administration Marketing
Management. With this thesis I finish my school period and it is time for the next step in my life, namely work full-time. The main reason for conducting a research into celebrity endorsement was that ‘the negative news’ celebrities in the sports advertising interests me very much. Next to this I have affinity with sport, especially soccer. I play this sport for a long time and not only the sport itself interest me but also everything around soccer, for example: interviews (on TV or radio), soccer news and sport advertising.
Besides thanking my parents and my girlfriend, I would like to thank my supervisor Prof. dr. T.H.A. Bijmolt for his feedback during my thesis and the meaningful and fine
meetings we had. Also I would like to thank my second supervisor dr. J. Liu for her feedback.
I hope you will enjoy reading my thesis.
Pieter Westra
TABLE OF CONTENTS
1 INTRODUCTION 8 1.1 Problem statement 10 1.2 Conceptual model 11 1.3 Academic relevance 11 1.4 Managerial relevance 121.5 Structure of the thesis 12
2 LITERATURE FRAMEWORK 13
2.1 Why use a celebrity endorser? 13
2.2 Negative news about celebrity endorsers 14
2.3 Brand attitude 14
2.4 Purchase intention 17
2.5 Celebrity product fit 17
2.6 Overview hypotheses 20 3 RESEARCH DESIGN 21 3.1 Articles 22 3.2 Methodology 24 3.3 Procedure 25 3.4 Brand attitude 26 3.5 Purchase intention 27 4 RESULTS 29 4.1 Brand attitude 29 4.2 Purchase intention 30
4.3 Testing hypotheses 3 and 4 31
4.4 Control variables 32
4.5 Testing hypothesis 5 33
4.6 Control variables 35
4.7 Overview of conclusions per hypothesis 36
5 CONCLUSION AND DISCUSSION 38
5.1 General conclusion 38
5.2 Managerial implications 40
5.3 Academic implications 40
5.4 Limitations and future research 41
1
INTRODUCTION
Celebrities are widely used in advertisements and are world wide a popular way to increase sales (Agrawal & Kamakura (1995). Celebrities have been endorsing products since the late nineteenth century, one of the earliest examples of the past is Queen Victoria in association with Cadbury’s Cocoa. In 1979, celebrity endorsers in commercials were estimated as one in every six advertisements, around the nineties this was around one in every four advertisements. (Erdogan, 1999). So advertisers make extensive use of celebrities endorsers over the years.
A celebrity endorser is defined as any individual who enjoys public recognition and who uses this recognition on behalf of a consumer product by appearing with it in an advertisement (Carroll, 2008). But specifically sports celebrities have become major spectacles of today’s media culture. Sports celebrities have been looked upon as role models for decades, and with the technological advances in broadcast and interactive media, it appears that famous and infamous athletes are everywhere (Bush et al, 2004). A sports celebrity endorser may be defined as “a famous athlete or coach who uses public recognition to recommend or co-present with a product in an ad” (Stafford et al, 2003).
The benefits of using celebrity endorsement to support a brand strategy can include according to Erdogan (1999): increased attention, image polishing, brand introduction, brand repositioning and the potential to underpin global campaigns. Overall research points to accumulated positive influence over the audience’s recall and purchase intentions (Atkin & Block, 1983) with
celebrities deemed more effective than using a typical consumer or expert (Kamins, 1989).
Till and Shimp (1998) stated that the impact of celebrity endorsement depends on the fit between the celebrity and the product next to the type of information about the celebrity. It is plausible that Real Madrid star Christiano Ronaldo has more effect in a soccer shoe
advertisement than George Clooney for example. So the fit between the celebrity and the advertised product category could be very important.
change of the celebrity and public controversy of the celebrity. Especially the last one could harm the brand and the sales of products. In the past there have been numerous examples of celebrity endorsers who created negative news around them but the 5 most costly failed sports celebrity endorsements, according to Businessinsider.com and Financialtimes.com, can be found under here:
1 Tiger Woods
Following his infidelity scandal, Woods lost his Gatorade, AT&T and Accenture endorsements, costing him 22 million dollars. Nike kept sponsoring him.
2 Kobe Bryant
In 2001, Kobe Bryant lost endorsement deals with McDonald's, Sprite and Nutella when he was charged with sexual assault but was not convicted because the girl dropped the charge cause of a cash settlement. The Basketball star lost an estimated $10 million when McDonald's declined to renew his contract at the end of the 2003.
3 Earvin Johnson
Earvin Johnson is one of the most successful basketball players in history. When Johnson announced in 1991 that he had contracted HIV, his endorsement deals, worth nearly 12 million had disappeared. Pepsi and Converse cancelled the endorsement deals.
4 Floyd Landis
Floyd Landis, Tour de France winner of 2006, used enhancing steroids during the competition. He lost his title and endorsement worth 9 million dollar.
5 Justin Gatlin
Sprinter Justin Gatlin has won the 100 meter gold medal during the summer Olympics of 2004. He earned 4 million in annual endorsements and sponsorships. In 2006 he tested positive for testosterone. Nike and several other sponsors dropped him from its sponsorships.
Nike has spent in 2006 476 million dollars in exchange for endorsements by celebrity athletes, this was part of the company's 1.7 billion dollars advertising budget for that year.
The use of celebrity endorsers is a popular tool but not without risks as stated above. Till and Shimp (1999) stated that brands perceived with a strong connection with the endorser are more vulnerable when negative information about the endorser is exposed.
Chung et al (2011) found that Nike Golf even in the midst of the Tiger Woods scandal was actually better off with Tiger Woods than without, so that negative news had less effect. Knittel and Stango (2009) estimated that EA sports lost a 4% market value and Gillette lost 2 % market value due to this scandal. So these effects were big.
1.1 Problem statement
But it is not clear which factors are responsible for the increase or decrease of this influence of negative news towards brand attitude and purchase intention. This thesis will focus on the effects of negative news about the celebrity on the brand attitude and the purchase intention and will investigate which factors have effect on this influence.
Therefore the following problem statement:
What is the influence of negative news about the celebrity endorser in sports advertising on the brand attitude and purchase intention of consumers, and which factors increase or decrease this influence?
The following sub questions will help to answer the problem statement: 1. Why use a celebrity endorser?
2. Should a celebrity endorser fit to the advertised product?
3. What are the potential negative effects of celebrity endorsement? 4. Does negative news about the celebrity affects the brand attitude? 5. Does negative news about the celebrity affects the purchase intention?
1.2 Conceptual model
To answer the problem statement above the following conceptual model was set up to visualize the research:
Figure 1: Conceptual model
The celebrity could be a sports celebrity or a non-sports celebrity. A sport celebrity is for example football player Ruud van Nistelrooij and a non sports celebrity is for example actor Bas Muijs. The negative news could also be sports related or non sports related; sport news for example is a doping scandal and non-sport news is for example a sexual assault.
1.3 Academic relevance
Many researchers have investigated the positive effects of celebrity endorsements towards brand attitude and sales of products. But about specifically (sports) celebrities in sports
advertising is less information available and particularly the effects of negative aspects towards it: when a sports celebrity comes badly in the news and creates negative news to the brand. Athletes receive millions of dollars / euros for their endorsements for sport brands, earlier research found out that most of the times this investment is profitable. But about the effects
when negative news about the celebrity will arise is less information available. This thesis is an addition to the existing literature about the negative aspects of celebrity endorsement.
1.4 Managerial relevance
The effects of this negative news about (sports) celebrities towards brand attitude and purchase intention is very useful information for advertisers and marketing managers of sports
companies. It is useful to know what risks there are of using sports celebrities in their advertising campaigns and they can assess whether a certain celebrity will fit to the
advertisement and brand and if this could harm their brand and sales. Also large sponsors of celebrities could use this research to see what the effects on sales and brand attitude can be when negative news arises about the celebrity. Famous athletes can also benefit from this research, this will inform them which risks there are when endorsing famous brands and what effects they can cause when coming badly into the news.
1.5 Structure of the thesis
2
LITERATURE FRAMEWORK
2.1 Why use a celebrity endorser?
According to Clark and Horstman (2005) a consumer that observes messages for two different firm’s products, one product’s message containing a celebrity endorsed and the other not, believes the celebrity endorsed product will have more purchases and so be of higher value.
In return for the high costs of celebrity endorsers, these athletes endorsers are expected to accomplish a number of objectives, including: capture the attention of consumers, strengthen recall of the brand name, reinforce the image of the product, give the message credibility, increase product attractiveness, increase liking and recall of the ad, and increase the likelihood of purchase (Burnett et al, 1993).
There have been lots of researches done about celebrity endorsement but what determines the success of the endorser, under here a small review of different options:
Source creditability and attractiveness
The effectiveness of the message depends on the perceived expertise and trustworthiness of an endorser. Pursuing a celebrity-endorsed strategy enables advertisers to project a credible image in terms of expertise, persuasiveness, trustworthiness and objectiveness (Till and Shimp, 1998). Solomen (2002) stated that source attractiveness refers to the endorsers physical approach personality, likeability, and similarity to the receiver, thus to the perceived social values of the source.
Meaning transfer
McCracken (1989) explain that the celebrity spokesperson would be effective only if clarity is achieved in assessing the meanings consumers associated with the endorser which are eventually transferred to the brand.
Selecting the best endorser
2.2 Negative news about celebrity endorsers
There is also a less positive side of celebrity endorsement, the case when a celebrity comes negative into the news. Negative news about celebrity endorsers can put a firm’s products and image at risk. And negative information gets more attention, processing, and impact than positive information depending on which type of information is presented first and which is less expected (Smith and Petty, 1996).
Till and Shimp (1998) stated that negative celebrity information lowers brand evaluations conditioned that there must be a strong associative link between celebrity and product present. Since repeated pairings of a brand and celebrity strengthen the associative link consumers establish between brand and celebrity, negative information about the celebrity may negatively impact the endorsed brand (Erdogan and Baker, 2001)
Negative behavior of celebrities attracts more attention, is better encoded and is more easily recalled than positive information (Ybarra and Stephan, 1996). There is some evidence that positive information about a brand can reverse some of the damage caused by negative information, lots of research shows it is very difficult to refurbish a celebrity endorser’s image once it is harmed by negative publicity (Butler and Berry, 2002).
2.3 Brand attitude
An attitude is an overall evaluation that expresses how much we like or dislike an object, issue, person or action. Attitudes are important because they guide our thoughts, influence our feelings and affect our behavior (Hoyer and MacInnes, 2008). Attitude consists out of three interrelated components:
1. Cognitive function – thoughts / beliefs 2. Affective function – feelings / emotions 3. Connative function – behavior
Marketers need to consider multiple factors when trying to influence cognitive attitudes: strength and importance of consumers ‘beliefs, the likelihood that consumers will form
have about information received from an external source (such as advertising, sales people, internet or friend) or on information we recall from memory (Hoyer & Macinnes, 2008). So for instance marketers could use a celebrity, who does not advertise many other products for an endorsement, in an advertisement because of the believability between endorser and product. And when negative news occurs about the celebrity endorser this could also damage the believability of the brand. But due to this negative news people could remember the brand and not the negative news, so they recall only the brand from their memory.
Marketers like to feature celebrities in ads because of two factors who evokes favorable and non-favorable affective reactions: physical attractiveness and its likability (Hoyer & Macinnes, 2008). Nike found that the football star Cristiano Ronaldo is an attractive and likable endorser in their ads for football equipment like shoes and cloths. These likable sources may serve as unconditioned stimuli, create a positive mood that affects consumers’ evaluations of the ad or brand, and make consumers feel more positive about the endorsed product (Hoyer & Macinnes, 2008). But this has also a downside, when negative news about this endorsers arises. The prediction is that the negative news about the celebrity will have impact on the affective side of the attitude because of the feelings and emotions people have when they face the negative news about the celebrity in celebrity endorsements.
The third component is concerned with the likelihood or tendency that a specific action will be undertaken by an individual regarding attitude object. It is treated as an expression of
consumers’ intention to buy. Consumers usually make purchases for positively evaluated brands, when a negative evaluation occurs the number of purchases could be harmed.
There are multiple scales to measure these three components of brand attitude. Muehlink and Laczniak (1988) had found some reliable scales to measure like positive – negative, good – bad and favorable – unfavorable rated on 7-points scales. Mitchel (1986) used the scales dislike very much – like very much and pleasant – unpleasant.
image and product image (Kamins, 1989). This author suggests that endorsers are more effective when there is a fit between the endorser and the endorsed product. So for example Kobe Bryant was used in Nike advertisements to promote the basketball department of Nike, he is seen as one of the best players in the NBA.
Picture 1 : Kobe Bryant in Nike advertisement (source Nike)
Above there is a Nike advertisement with Kobe Bryant, slogan is ‘supernatural’. He is presented in this ad as an angel who can perform optimally with this Nike outfit. As stated before, Kobe Bryant was arrested in 2003 due to a sexual assault complaint. This collides with the message that Nike spread in this ad of ‘supernatural’. This could particularly affect the affective part of the brand attitude, this part is based on emotions and feelings. Negative feelings could arise about the brand Nike, this leads to hypothesis 1:
2.4 Purchase intention
Lee and Thorson (2008) showed that celebrity endorsements are evaluated more favorably in terms of purchase intention when there is a moderate mismatch than when there is a complete mismatch between celebrity and product. The results of Agrawal & Kamakura (1995) indicated that celebrity endorsement has a positive impact on expected future profits.
The study of Lafferty and Goldsmith (1999) indicates that the credibility of the corporation is significantly more important than the endorser in determining purchase intentions. Both play a role in the attitude formation, but the endorser has considerably less impact when it comes to making actual purchases. This could suggest that the respondents may be more likely to
purchase a product based on the product’s attributes rather than relying on the endorsement of the celebrity. Petty et al (1983) found some evidence that for most people the impact of the celebrity endorser is irrelevant to the evaluation of the true merits of a product and serves only as a peripheral cue.
Therefore I predict that negative news about the celebrity endorser has no impact on the purchase intention of the advertised products. I argue that negative news will not harm the intention to buy the advertised product, therefore we can derive hypothesis 2:
H2: Negative news about the celebrity in celebrity endorsements will affect the purchase intention of the advertised product compared to positive news about the celebrity in celebrity endorsements.
2.5 Celebrity product fit
(football shoes), non-sports and sports related. An advertisement of Nike football shoes and Ruud van Nistelrooij should normally fit better than Van Nistelrooij with L’Oreal.
Picture 2 : Ruud van Nistelrooij in Nike ad (left) and L’Oreal ad (right), source: Nike and L’Oreal
McCracken (1989) stated that the foundation of the success of a celebrity endorsement is based on whatever gives the spokesperson celebrity status. In the case of an athlete spokesperson, the base for success rests in the sport from which the athlete comes. The more similar the image of the product is to the image of the sport, the more positive the consumer response is to the endorsement (Martin, 1996).
Negative news Sport Non-sport Sport ++ + C e le b ri ty e n d o rs e r Non-sport 0 0
Table 1: Forecast of effects on purchase intention and brand attitude (0 = no effect and ++ = much effect)
From this table we can derive the following hypotheses:
H3 For sports advertising, negative news about sports celebrity endorsers has a stronger negative effect on a) brand attitude and b) purchase intention, compared to non-sports celebrity endorsers.
H4 For sports advertising, negative sport news has a stronger negative effect on a) brand attitude and b) purchase intention, compared to negative non-sport news.
2.6 Overview hypotheses
Under here there is an overview of the earlier stated conceptual model containing the hypotheses and the different effects of these hypotheses.
3
RESEARCH DESIGN
The goal of this research is to test the influence of negative news about the celebrity in celebrity endorsements in sports advertising on the brand attitude and purchase intention of consumers. And which factors increase or decrease this influence? An experiment is conducted to study these questions.
For the experiment sports celebrity Ruud van Nistelrooij (famous soccer player) was chosen and for the non-sports celebrity Bas Muijs (famous Dutch actor) was chosen. These two celebrities have similar looks and image, both are very famous celebrities in the Netherlands each in their own specialty.
The brand Nike will be used in the different advertisement, reason for this choice is that Nike is a global sports brand and Nike is specialized in soccer equipment. So the combination with Ruud van Nistelrooij in Nike advertisements is also more credible. Bas Muijs is especially famous because he is an actor but he plays also soccer and does that a couple of times a year with other famous non sport celebrities in fundraising games.
3.1 Articles
To test the different hypotheses as stated in the previous chapter there are six different articles designed, two of them are positive and four of them are negative. The two positive news articles are used to test the first two hypotheses, the four negative news articles were used to test the remaining hypotheses and were also used to compare with the first two articles.
There are two ‘positive sport news’ articles, one about a sports celebrity (Ruud van Nistelrooij) and one about a non-sports celebrity (Bas Muijs)
Positive sport news – Ruud van Nistelrooij
Van Nistelrooij trefzeker
Van onze redactie
AMSTERDAM – Voetballer Ruud van Nistelrooij heeft zich direct laten gelden bij zijn debuut voor Malaga. De Nederlandse spits was zaterdag drie keer trefzeker in de oefenwedstrijd tegen UD Horadada (4-1 winst).
Na een kwartier maakte Van Nistelrooij zijn eerste, van 11 meter. Nummer twee kwam nog voor rust, in de 25e minuut om precies te zijn. Kort na de hervatting schoot de voormalige PSV’er zijn derde binnen, opnieuw een penalty.
Net als Van Nistelrooij maakte ook Joris Mathijsen zijn officieuze debuut voor de Spaanse club. De verdediger viel twintig minuten voor tijd in.
Positive sport news – Bas Muijs
Bas Muijs trefzeker
Van onze redactie
AMSTERDAM – Acteur Bas Muijs heeft zich direct laten gelden bij zijn debuut voor de KIKA All-stars. De 34-jarige acteur was zaterdag drie keer trefzeker in het benefiet duel voor KIKA tegen de profs van Cambuur (5-3 verlies).
Na een kwartier maakte Muijs zijn eerste, van 11 meter. Nummer twee kwam nog voor rust, in de 25e minuut om precies te zijn. Kort na de hervatting schoot de acteur zijn derde binnen, opnieuw een penalty.
And there are four categories of ‘negative news’ articles: 1. Sport endorser – sport news
2. Sport endorser – non-sport news 3. Non-sport endorser – sport news 4. Non-sport endorser – non-sport news.
The four different ‘negative news’ articles are stated under here: 1. Sport endorser – sport news
Dubbele beenbreuk na charge Van Nistelrooij
Van onze redactie
AMSTERDAM - Voetballer Ruud van Nistelrooij heeft Sebastian Schweinsteiger een dubbele beenbreuk bezorgd, zo blijkt uit de eerste onderzoeken in het ziekenhuis. De middenvelder van Bayern Munchen moest zondag al na vier minuten het veld verlaten tijdens het Champions League duel.
In het ziekenhuis bleek de 26-jarige Duitser het linker scheen- en kuitbeen te hebben gebroken. Hij zal maandag worden geopereerd en daarna maandenlang moeten revalideren Na de tackle van Van Nistelrooij kreeg Schweinsteiger een zuurstofmasker op. Hij verliet per brancard het veld.
De scheidsrechter zag in de forse tackle geen overtreding, maar floot iets later wel af. Van Nistelrooij speelde de bal volgens de leidsman.
2. Sport endorser – non-sport news
Van Nistelrooij aangeklaagd door vriendin wegens zware mishandeling
Van onze redactie
AMSTERDAM - Voetballer Ruud van Nistelrooij heeft zichzelf in een lastig parket
gemanoeuvreerd. Justitie in Amsterdam start namelijk een strafrechtelijk onderzoek naar de Malaga-speler wegens bedreiging en zware mishandeling van zijn ex-vriendin Sarah H.
3. Non-sport endorser – sport news
Dubbele beenbreuk na charge Bas Muijs
Van onze redactie
AMSTERDAM - Acteur Bas Muys heeft Koen Brack een dubbele beenbreuk bezorgd, zo blijkt uit de eerste onderzoeken in het ziekenhuis. De verdediger van Cambuur moest zondag al na vier minuten het veld verlaten tijdens het benefietduel voor KIKA.
In het ziekenhuis bleek de 26-jarige Brack het linker scheen- en kuitbeen te hebben gebroken. Hij zal maandag worden geopereerd en daarna maandenlang moeten revalideren Na de tackle van Muijs kreeg Brack een zuurstofmasker op. Hij verliet per brancard het veld.
De scheidsrechter zag in de forse tackle geen overtreding, maar floot iets later wel af. Bas Muijs speelde de bal volgens de leidsman.
4. Non-sport endorser – non-sport news
Bas Muijs aangeklaagd door vriendin wegens zware mishandeling
Van onze redactie
AMSTERDAM - Acteur Bas Muijs heeft zichzelf in een lastig parket gemanoeuvreerd. Justitie in Amsterdam start namelijk een strafrechtelijk onderzoek naar de oud GTST ster wegens bedreiging en zware mishandeling van zijn ex-vriendin Madelon L.
Madelon L., een fotomodel uit Amsterdam, deed aangifte tegen Muijs, met wie ze jarenlang samenwoonde en met wie ze een tweejarig kind heeft. Ze zegt dat ze de afgelopen jaren veelvuldig ernstig is mishandeld en bedreigd door de acteur. De recherche van bureau Amstelland begint vandaag een uitgebreid onderzoek, waarbij ook Bas Muijs zal worden verhoord.
3.2 Methodology
Picture 4 : Webpage negative sport news and sports celeb (RvN)
3.3 Procedure
Respondents are asked to read the text carefully and take a good look of the whole web page. After that several questions were asked (on 7 points Likert scales) to measure the brand attitude and purchase intention.
For each category a minimum of 40 respondents are needed to get valid results, the questionnaire is set out online using thesistools.com. This is an online tool to spread questionnaires, respondents are reached using the mail tool of RUG, every RUG student will receive the questionnaire. The respondents were also asked to forward the questionnaire to people they know.
Number of respondents Ruud van Nistelrooij and positive sport news 42
Bas Muijs and positive sport news 42
Ruud van Nistelrooij and negative sport news 70 Ruud van Nistelrooij and negative non-sport
news
68
Bas Muijs and negative sport news 50
Bas Muijs and negative non-sport news 66
Total 341
Table 2 : Number of respondents per category
A total number of 341 man and women filled in the questionnaire, there were no missing values due to the fact that every question has to be filled in to finish the questionnaire.
The sample is randomly chosen and approached by email to fill one of the six questionnaires, each questionnaire has the same questions but the six Web Pages are different, as stated in the experimental design. People were invited by email so these respondents have affinity with surfing on the internet, this will represent somewhat the whole Dutch population.
3.4 Brand attitude
As mentioned in the previous chapter, the affective component of the brand attitude will be expected to be influenced by negative news. Brand attitude is divided in three questions about the brand Nike:
- Dislike – Like - Negative – Positive
- Not attractive – Very attractive
3.5 Purchase intention
After asking the respondents questions about the brand attitude questions will be asked about purchase intention, this is divided in two questions:
- The intention to buy Nike shoes - The intention to buy Nike clothes
For these questions also Likert scales (1 to 7) was used. The internal consistency between these two questions was 0.676, which is reliable. The average scores of these two questions were used to form a new variable, namely purchase intention.
Finally there are some general questions:
- Questions about the advertisement (do the respondent like the ad and how convincing is the ad)
- Demographic questions (age, gender and education)
- Questions about how sporty the respondent is (favorite sport and how many hours the respondent sports).
See appendix B for the complete questionnaire.
3.6 Demographics
In total 204 men (59.8 %) and 137 women (40.2%) filled in the questionnaire. The age category 18 to 25 years covers 69.5 % of the respondents, category 26 to 35 years covers 19.6 %, category 36 to 50 years covers 7 % and category 50 plus covers 3.9 % .The following graph shows the gender / age spread across the 4 scenarios
Among the respondents, the education level was mainly WO and HBO, followed by MBO, secondary school and primary school.
Graph 2 : education
The respondents were fairly sporty, 58.1 % of the respondents sports less than 4 hours a week, 38.1 % sports 4 to 10 hours a week and the remaining 3.8 % sports more than 10 hours a week.
4
RESULTS
4.1 Brand attitude
To test the hypotheses a general linear model combining ANOVA and regression is performed. First brand attitude is used as dependent variable to test H1.
The following factors / variables were conducted and used to see if there are significant differences between the following groups:
- Positive news vs. negative news - Sport celebrity vs. non-sport celebrity - Sport news vs. non-sport news
Also gender, age and education were added to the test, to see if one of these control variables has an effect on the dependent variables.
The followings tables 3 and 4 provides an overview of the results.
Factors df F-value P-value
Positive vs negative news 1 2,605 0,107 Sport celebrity vs non-sport celebrity 1 2,226 0,137
Sport news vs non-sport news
1 2,140 0,144
Gender 1 8,171 0,005
Age 3 0,489 0,690
Education 3 2,311 0,076
Table 3 : Results ANOVA – Brand attitude
Parameter B T-value P-value
Age 18 to 25 years 26 to 35 years 36 to 50 years 50 plus 0,097 0,073 - 0,198 0,295 0,210 - 0,508 0,768 0,834 0,612 Education Secondary school MBO HBO WO - 0,260 0,230 0,279 - - 0,959 1,004 2,020 0,338 0,316 0,044
Table 4 : Parameter estimates ANOVA – Brand attitude
There is no significant effect between positive news and negative news about a celebrity in celebrity endorsements towards brand attitude (F = 2.605, p=0.107), see table 3. Therefore H1 is rejected. Contrary to expectations, the respondents in the negative news condition indicated to have an higher brand attitude (M = 4.8) than the respondents in the positive news condition (M= 4.5), but this difference is not significant.
4.2 Purchase intention
Again a general linear model combining ANOVA and regression is performed, this time with purchase intention as dependent variable to test H2, also the same variables / factors as in the previous test are used in this test.
The following tables 5 and 6 provides an overview of the results:
Factors df F-value P-value
Positive vs negative news 1 2,495 0,115 Sport celebrity vs non-sport celebrity 1 1,772 0,184
Sport news vs non-sport news
1 0,392 0,532
Gender 1 0,069 0,793
Age 3 0,210 0,889
Education 3 0,798 0,496
Parameter B T-value P-value Negative news Positive news 0,338 - 1,580 0,115 Sports celebrity Non-sports celebrity - 0,210 - - 1,331 0,184 Sport news Non-sport news 0,112 - 0,626 0,532 Gender Men Women 0,042 - 0,262 0,793 Age 18 to 25 years 26 to 35 years 36 to 50 years 50 plus 0,259 0,218 0,087 0,620 0,493 0,175 0,536 0,622 0,861 Education Secondary school MBO HBO WO 0,104 - 0,139 0,219 - 0,302 - 0,477 1,253 0,762 0,663 0,211
Table 6 : Parameter estimates ANOVA –Purchase intention
The respondents in the negative news condition indicated to have a higher purchase intention (M = 3.60) than the respondents in the positive news condition (M= 3.26), but this difference is not significant. There is no significant effect (F = 2.495, p=0.115) in purchase intention between positive and negative news about the celebrity in celebrity endorsements.
There is no significant difference between these two groups, so people will not be influenced by negative news when they have the intention to buy Nike products. So H2 is rejected. People’s purchase intention will not be affected due to negative news about the endorser.
4.3 Testing hypotheses 3 and 4
Previous in hypothesis 1 and 2 the results of the effects of negative versus positive news regarding brand attitude and purchase intention were presented. The hypotheses, 3 till 5, will present the results of the effects of the following moderators on this relationship:
First the effects of a sport celebrity and a non-sports celebrity in celebrity endorsements towards brand attitude were tested. To test this Hypothesis H3a the same test is used as for H1, see table 3 ANOVA brand attitude.
The estimated parameter of -0.185 indicates a difference between a sports celebrity and a non- sports celebrity but this effect is not significant (T = - 1.492, P = 0.137). Sports celebrity mean is 0,185 smaller than the non-sports celebrity mean, but this is not significant and therefore there is no significant difference between a sports celebrity and non-sports celebrity regarding to brand attitude. Hence, no support is found for H3a.
The same is done but now for purchase intention. To test this Hypothesis H3b, the same test is used as for H2, see table 5 ANOVA purchase intention. The sports celebrity mean is 0.21 smaller than the non-sports celebrity mean. But there is also no significant effect (T=- 1.331, P=0.184) between negative news about a sports celebrity and non-sports celebrity regarding purchase intention. So we reject H3b.
Secondly the effects of negative sport news and non sport news towards brand attitude were tested. For the results of H4a, see table 4 brand attitude ANOVA. The sport news mean is 0.207 larger than the non-sport news mean but there is no significant difference (T=1.463, p=0.144) between negative sport news and negative non-sport news regarding brand attitude. There is not enough evidence for accepting H4a.
This was also tested on purchase intention, for the results of H4b see table 5 ANOVA purchase intention. Sport news mean is 0,112 larger than the non-sports mean but forthis difference is not enough evidence (T=0.626, P=0.532). There is no significant result for accepting H4b, that negative sport news has a stronger effect on purchase intention, compared to negative non-sports news.
4.4 Control variables
Men have a significant higher brand attitude compared to women (T = 2.859, P = 0.005). Men’s mean is 0.358 larger than the women’s mean. Also people whose education was HBO had a significant lower brand attitude (T = 2.020, P=0.044) compared to people whose education was WO. Variable age had not significant differences.
All control variables (gender, age and education) checked for purchase intention were not significant. These variables have no significant effect on purchase intention.
4.5 Testing hypothesis 5
To test hypothesis 5 two general linear models combining ANOVA and regression are
performed, one for brand attitude and one for purchase intention. All SPSS output can be found in appendix C.
The following factors were used in this test to see if there are significant differences: - Sport celebrity vs non-sport celebrity
- Sport news vs non-sport news
And most important the interaction between the two above factors will be tested.
Next to checking the significance levels to accept or reject the hypothesis two tables (and graphs) were conducted. In these tables the predicted values of the means, under reserve of the parameters estimates, were presented. Table 7 and graph 3 for brand attitude and table 8 and graph 4 for purchase intention.
Brand attitude Celebrity
News Sports celebrity Non-sports celebrity
Sport news 4.481
-.10 ( = 4.381)
4.481
-0.113 (= 4.368) Non sport news 4.481
0.154 (= 4.635)
4.481
(=4.481)
Graph 3 : brand attitude
Negative sport news about sports celebrity endorsers has a stronger negative effect (M = 4,381) on brand attitude compared to negative non-sport news about non-sports celebrity endorsers (M = 4,481) but this difference is not significant (T = -0.399, P = 0,69).
There is also not enough evidence (T = -0.399, P = 0,69) to assume that negative sport news about non-sports celebrity endorsers has a stronger effect (M = 4,368) on brand attitude compared to negative non-sport news about non-sports celebrity endorsers (M = 4,481). Hence, no support is found for H5a. For these results of H5a see table 7 and graph 3.
Purchase Intention
Celebrity
News Sports celebrity Non-sports celebrity
Sport news 3.293
-0.332 (= 2.961)
3.293
.004 (= 3.297) Non sport news 3.293
0.157 (=3.45)
3.293
(= 3.293)
Graph 4 : purchase intention
The effect of negative sport news on purchase intention for sports celebrity endorsers (M = 2.961) is more negative compared to negative non-sport news about non-sports celebrity endorsers (M = 3.293). But this negative effect is not significant (T = -1.049, P = 0.295).
The effect of negative sport news about non-sports celebrity endorsers (M = 3.297) isn’t more negative compared to negative non-sport news about non-sports celebrity endorsers (M = 3.293) on purchase intention, but again there is no support found for this difference (T = -1.049, P = 0.295). We have not enough evidence for accepting H5b see graph 4 for graphical
presentation.
4.6 Control variables
4.7 Overview of conclusions per hypothesis
The following table gives a complete overview of the conclusions of the results for each hypothesis.
Hypotheses Conclusions
H1 Negative news about the celebrity in celebrity endorsements leads not to a negative affective attitude towards the brand compared to positive news about the celebrity in celebrity endorsements. There is no significant effect between positive news and negative news about a celebrity in celebrity endorsements towards brand attitude (F = 2.605, p=0.107).
H2 The respondents in the negative news condition indicated to have a higher purchase intention (M = 3.60) than the respondents in the positive news
condition (M= 3.26), but this difference is not significant (F = 2.495, p=0.115). So negative news about the celebrity in celebrity endorsements will not affect the purchase intention of the advertised product compared to positive news about the celebrity in celebrity endorsements.
H3a The estimated parameter of -0.185 indicates a difference between a sports celebrity and a non- sports celebrity regarding brand attitude but this effect is not significant (T = - 1.492, P = 0.137).
H3b There is no significant effect (T=- 1.331, P=0.184) between negative news about a sports celebrity and non-sports celebrity regarding purchase intention. So
negative news about sports celebrity endorsers has not a stronger negative effect on brand attitude and purchase intention, compared to non-sports celebrity endorsers.
H4a The sport news mean is 0.207 larger than the non-sport news mean but there is no significant difference (T=1.463, p=0.144) between negative sport news and negative non-sport news regarding brand attitude.
H4b Sport news mean is 0,112 larger than the non-sports mean regarding purchase intention but forthis difference is not enough evidence (T=0.626, P=0.532). So negative sport news has not a stronger negative effect on brand attitude and purchase intention, compared to negative non-sport news.
H5a Negative sport news about sports celebrity endorsers has a stronger negative effect (M = 4,381) on brand attitude compared to negative non-sport news about non-sports celebrity endorsers (M = 4,481) but this difference is not significant (T = -0.399, P = 0,69).
H5b The effect of negative sport news on purchase intention for sports celebrity endorsers (M = 2.961) is more negative compared to negative non-sport news about non-sports celebrity endorsers (M = 3.293). But this negative effect is not significant.
The effect of negative sport news about non-sports celebrity endorsers (M = 3.297) isn’t more negative compared to negative sport news about non-sports celebrity endorsers (M = 3.293) on purchase intention, but again there is no support found for this difference (T = -1.049, P = 0.295).
5
CONCLUSION AND DISCUSSION
Advertisers make extensive use of celebrity endorsers over the years, and in sports advertising sports celebrity endorsements are an important tool. But there are also certain risks and potential negative effects of using celebrities in advertisements like: the celebrity could overshadow the brand, it is expensive, image change of the celebrity and public controversy of the celebrity. This last one happens when the celebrity creates negative news around him or her. This is a risk because most celebrities live no ordinary life, there are numerous examples of ‘bad news’ celebrities, the most famous one is Tiger Woods. But will this negative news harm the brand attitude and purchase intention of consumers? This thesis studied the effects of negative news about a celebrity endorser on brand attitude and purchase intention, and which factors increase or decrease this influence. This was examined by exposing consumers to negative and positive sport and non-sport web articles with different advertisements on the same page.
5.1 General conclusion
The central question in this thesis was:
What is the influence of negative news about the celebrity endorser in sports advertising on the brand attitude and purchase intention of consumers, and which factors increase or decrease this influence?
The reasons and objectives of companies for using celebrities in advertisements according to Burnett et al (1993) are the following: capture the attention of consumers, strengthen recall of the brand name, reinforce the image of the product, give the message credibility, increase product attractiveness, increase liking and recall of the ad, and increase the likelihood of purchase.
Attractive and likable endorsers in ads will create a positive mood that affects consumers’ evaluations of the ad or brand, and make consumers feel more positive about the endorsed product (Hoyer & Macinnes, 2008). But when there is negative information around these attractive and likable endorsers in ads this may negatively impact the endorsed brand (Erdogan and Baker, 2001). These findings collides with the findings of this thesis; there is no significant effect of negative news compared to positive news about celebrities on brand attitude. The respondents won’t get a negative affective attitude towards the brand Nike when they are confronted with negative news about the celebrity in a celebrity endorsement. The respondents in the negative news condition even indicated to have an higher brand attitude than the
respondents in the positive news condition: respondents liked the brand more, had a more positive feeling about the brand, and found the brand more attractive. This could indicate that negative news will be more easily recalled and remembered than positive news. Ybarra and Stephan (1996) confirms this by stating that negative behavior of celebrities will attract more attention, is better encoded and is more easily recalled than positive information. The type of celebrity (sport and non-sports) and type of news (sport and non-sport) did not strengthen the negative effect on this relationship. This could indicate that people do not care what celebrity or which news they see, they will hold their same attitude towards the brand Nike. These two factors do not increase the negative effect.
5.2 Managerial implications
This study provides valuable insights for managers responsible for the advertising and promotion of sports products and for companies that sponsors sports celebrities, in this case large sport brands. There was no evidence found that negative news about the celebrity will harm the sales of Nike products and there was no evidence found that negative news will harm the brand Nike. As earlier stated, this could indicate that negative news could be remembered more. So large sports brand like Nike, Adidas, Puma etc. could use this in their marketing campaigns, choose a celebrity who is not that brave or make up a ‘negative’ story and get free publicity. These negative stories will be picked up by all the media and this will create
‘commercials’ on TV and radio for free, afterwards this can be refute as marketing stunt, again publicity. This study also confirms what previous literature stated, when a celebrity comes negatively into the news this will not affect the intention to purchase shoes or clothes of Nike. So the risks of choosing a ‘bad’ celebrity for their endorsements are not that big.
Also for sportsmen and sportswomen this thesis gives valuable insights in how they can become interesting ‘advertising objects’, not always the brave sports celebrities are good but also the more bad sports celebrities could be interesting. This thesis gives also useful information for sportsmen who created already negative news about them, results showed that these sportsmen do not have to worry because the effects on the endorsed brand are not that big.
5.3 Academic implications
Scientific contribution of this study is that this thesis provides an overview of the existing literature about celebrity endorsement till now: it gives an valuable analysis of negative news about celebrity endorsement, advantages of using celebrities in commercial outings, and the importance of the fit between product and celebrity.
any effect on brand attitude, reason for this difference could be that this thesis has combined scales of Muehlink and Laczniak (1988) and Mitchel (1986) for measuring the brand attitude.
As third this study provides, compared to existing literature, new insights into different types of news and celebrities, in this case: sport news and sport news and, sport celebrity and non-sports celebrity. This will give a clear insight into the differences between the four groups. This kind of framework was not studied previously. Not all outcomes of the tests provided significant outcomes but this framework could be used in another setting, see chapter 5.4 for future research. The method which brand attitude and purchase intention was measured can also be used by future researchers when they study the effects on these two components.
5.4 Limitations and future research
As with any other study, there are several limitations and advices for future research. First of all, in this experiment only different WebPages were used and there was only one moment of contact with the negative news. Future researchers should also use a couple of different negative experiences, for example show a ‘negative’ WebPage, show a ‘negative’ newspaper article, let them hear a ‘negative’ radio fragment about the endorser etc. Maybe people will not get a negative feeling or negative attitude towards the brand when reading one negative article about the celebrity.
Third, the brand Nike is a very strong brand and most of the respondents know the brand Nike very well, they or their friends / family have already an experience with Nike. Another advice for future research would be choosing another (unknown) brand, so that people have no
experience with the brand and do not know the brand well. The number of respondents could also be the reason for the lack of significance, this can also be extended in future research. Most respondents are recruited via the university, when a more diverse respondents group will be chosen this could also have a positive effect on the significance levels.
5.5 Conclusion
Not all the effects of negative news that I had expected, came out. There were differences between the groups negative news (sport news and non-sport news) vs. sports celebrity (sports and non-sports) but most of these differences were not significant. The effects on purchase intention was what I had expected but the effects on brand attitude not.
This research will not be the last one about the effects of negative news about celebrities in celebrity endorsements. Nowadays it is much easier for celebrities to come negatively into the news. Most people have a cell phone with a camera, so celebrities can be filmed or
6
REFERENCES
Articles
Agrawal, J. & W. A. Kamakura (1995), The Economic Worth of Celebrity Endorsers: An Event Study Analysis. Journal of Marketing. Volume 59, pp. 56-62.
Atkin, C., & M. Block (1983), Effectiveness of Celebrity Endorsers. Journal of Advertising Research, 23 (1), 57-61.
Burnett, J., A. Menon & D.T. Smart (1993), Sports Marketing: A new ball game with new rules. Journal of Advertising Research, 21 – 35.
Bush, A.J., G. Martin & V.D. Bush (2004), Sports celebrity influence on the behavioral intentions of generation Y, Journal of Advertising Research, 108 – 118.
Butler, L., & D. Berry (2002), The influence of affective statements on performance on implicit and explicit memory tasks. Applied Cognitive Psychology, 16 (7), 829-843.
Carroll, C (2008), Brand communications in fashion categories using celebrity endorsement. Journal of Brand Management, 17, 146 – 158.
Clark, C.J., I.J. Horstmann (2005), Celebrity Endorsements, 1 – 40.
Erdogan, B., M. Baker (2001), Selecting Celebrity Endorsers: The Practitioner’s Perspective. Journal of Advertising Research, 39 – 48.
Erdogan, B. Zafer (1999), Celebrity endorsement: a literature review. Journal of Marketing Management, 15(4): 291-314.
Kamins, M.A., K. Gupta (1994), Congruence between Spokesperson and Product Type: A Matchup Hypothesis Perspective. Psychology & Marketing , 11(6), 569-586.
Kathri, P. (2006), Celebrity Endorsement: a Strategic Promotion Perspective. Indian Media Studies Journal, Vol 1, 25 – 37.
Knittel, C. R., & V. Stango (2009), Shareholder Value Destruction following the Tiger Woods Scandal.
Lafferty, B.A., & R.E. Goldsmith (1999), Corporate Credibility’s Role in Consumers’ Attitudes and Purchase Intentions When a High versus a Low Credibility Endorser Is Used in the Ad. Journal of Business Research, 44, 109–116.
Lee, J.G., & E. Thorson (2008), The Impact of Celebrity-Product Incongruence on the Effectiveness of Product Endorsement. Journal of Advertising Research, 433 – 449.
Martin, James H. (1996), Is the athlete’s sport important when picking an athlete to endorse a nonsport product? Journal of Consumer Marketing. Vol. 13, No. 6, pp 28-43.
McCracken, G. (1989), Culture and Consumption: a theoretical account of the structure and movement of the cultural meaning of consumer goods. Journal of Consumer Research, 13 (1), 71 – 85.
Mitchel , A.A. (1986), The Effect of Verbal and Visual Components of Advertisements on Brand Attitudes and Attitude Toward the Advertisement Journal of Consumer Research, 13, 12-24.
Muehling, D., & R. Laczniak (1988), Advertising's Immediate and Delayed Influence on Brand Attitudes: Considerations Across Message Involvement Levels. Journal of Advertising, 17 (4), 23-34.
Solomen, M.R., (2002), Consumer Behavior: Buying Having, and Being, 5th ed., New Yersey: Prentice Hall.
Stafford, M. R., T.F. Stafford & E. Day (2003), A Contingency Approach: The Effects of
Spokesperson Type and Service Type on Service Advertising Perceptions. Journal of Advertising, 31 (2), 17 – 35.
Till, B. D., & T. A. Shimp (1998), Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27(1), 67–82.
Till, B.D., & M. Buster (1998), Matching Products witch Endorser: Attractiveness Versus Expertise. Journal of Consumer Marketing, 15(6), 576-58.
Ybarra, O., & W. Stephan (1996), Misanthropic Person Memory. Journal of Personality and Social Psychology, 45 (4), 961-77.
Books
Fennis, B.M., & W. Stroebe (2010), The Psychology of Advertising, Psychology Press, (Hove and New York), first edition.
Hoyer, W.D., & D.J. MacInnis (2008), Consumer Behavior, Houghton Mifflin Company, Boston (New York), 5th edition.
http://www.rvn10.com
7
APPENDICES
Appendix A – WebPagesNegative sport news – non-sports celebrity (Bas Muijs)
Positive sport news – non-sports celebrity (Bas Muijs) N
Appendix B - Questionnaire
Advertentie
1 Hoe leuk vindt u de advertentie van Nike?
Helemaal niet leuk 1 2 3 4 5 6 7 Helemaal leuk
2 Hoe overtuigend vindt u deze reclame?
Helemaal niet 1 2 3 4 5 6 7 Heel erg overtuigend
overtuigend
Brand attitude
3 Hoe leuk vindt u het sportmerk Nike?
Helemaal niet leuk 1 2 3 4 5 6 7 Helemaal leuk
4 Wat voor gevoel heeft u bij het sportmerk Nike?
Negatief 1 2 3 4 5 6 7 Positief
5 Hoe aantrekkelijk vindt u het sportmerk Nike?
Helemaal niet 1 2 3 4 5 6 7 Heel erg aantrekkelijk
aantrekkelijk
Purchase intention
6 Heeft u het afgelopen jaar producten van het merk Nike gekocht?
Ja / Nee
7 Stel u gaat binnenkort nieuwe sportschoenen kopen. Wat is de kans dat u sportschoenen van het merk Nike koopt?
Klein 1 2 3 4 5 6 7 Groot
8 Stel u gaat binnenkort nieuwe sportkleding kopen. Wat is de kans dat u sportkleding van het merk Nike koopt?
Algemene vragen
9 Hoeveel uur per week sport u? o 0 tot 2 uren
o 2 tot 4 uren o 4 tot 6 uren o 8 tot 10 uren o Meer dan 10 uren
10 Wat is uw favoriete sport? o Voetbal o Tennis o Volleybal o Basketbal o Hardlopen o Atletiek o Golf o Hockey o Anders, namelijk …… 11 Wat is uw geslacht? o Man o Vrouw 12 Wat is uw leeftijd? o 18-25 jaar o 26-35 jaar o 36-50 jaar o 51-65 jaar o 66+
13 Wat is uw hoogst genoten opleiding? o Basisschool
o Middelbare school o MBO
Appendix C – SPSS output Hypotheses 1 till 4
Tests of Between-Subjects Effects
Dependent Variable:Brand_attitude
Source
Type III Sum of
Squares df Mean Square F Sig.
Corrected Model 29,337a 10 2,934 2,332 ,011 Intercept 1600,027 1 1600,027 1271,807 ,000 Posnegnews 3,278 1 3,278 2,605 ,107 Celebrity 2,800 1 2,800 2,226 ,137 sportnews 2,692 1 2,692 2,140 ,144 Q11_geslacht 10,280 1 10,280 8,171 ,005 Q12_leeftijd 1,844 3 ,615 ,489 ,690 Q13_opleiding 8,720 3 2,907 2,311 ,076 Error 415,164 330 1,258 Total 8606,778 341 Corrected Total 444,502 340 a. R Squared = ,066 (Adjusted R Squared = ,038)
Parameter Estimates
Dependent Variable:Brand_attitude
95% Confidence Interval
Parameter B Std. Error t Sig. Lower Bound Upper Bound
[Q12_leeftijd=1] ,097 ,330 ,295 ,768 -,551 ,746 [Q12_leeftijd=2] ,073 ,349 ,210 ,834 -,614 ,761 [Q12_leeftijd=3] -,198 ,390 -,508 ,612 -,964 ,568 [Q12_leeftijd=4] 0a . . . . . [Q13_opleiding=2] -,260 ,271 -,959 ,338 -,792 ,273 [Q13_opleiding=3] ,230 ,230 1,004 ,316 -,221 ,682 [Q13_opleiding=4] ,279 ,138 2,020 ,044 ,007 ,550 [Q13_opleiding=5] 0a . . . . .
a. This parameter is set to zero because it is redundant.
Tests of Between-Subjects Effects
Dependent Variable:Purchase_intention
Source
Type III Sum of
Squares df Mean Square F Sig.
Parameter Estimates
Dependent Variable:Purchase_intention
95% Confidence Interval
Parameter B Std. Error t Sig. Lower Bound Upper Bound
Intercept 3,098 ,500 6,201 ,000 2,115 4,080 [Posnegnews=1,00] ,338 ,214 1,580 ,115 -,083 ,760 [Posnegnews=2,00] 0a . . . . . [Celebrity=1,00] -,210 ,158 -1,331 ,184 -,520 ,100 [Celebrity=2,00] 0a . . . . . [sportnews=1,00] ,112 ,179 ,626 ,532 -,241 ,465 [sportnews=2,00] 0a . . . . . [Q11_geslacht=1] ,042 ,159 ,262 ,793 -,271 ,354 [Q11_geslacht=2] 0a . . . . . [Q12_leeftijd=1] ,259 ,418 ,620 ,536 -,563 1,082 [Q12_leeftijd=2] ,218 ,443 ,493 ,622 -,653 1,090 [Q12_leeftijd=3] ,087 ,494 ,175 ,861 -,885 1,059 [Q12_leeftijd=4] 0a . . . . . [Q13_opleiding=2] ,104 ,344 ,302 ,762 -,572 ,780 [Q13_opleiding=3] -,139 ,291 -,477 ,633 -,712 ,434 [Q13_opleiding=4] ,219 ,175 1,253 ,211 -,125 ,564 [Q13_opleiding=5] 0a . . . . .
Hypothesis 5
Tests of Between-Subjects Effects Dependent Variable:Brand_attitude
Source
Type III Sum of
Squares df Mean Square F Sig.
Corrected Model 26,261a 10 2,626 2,072 ,026 Intercept 2165,310 1 2165,310 1708,473 ,000 Celebrity 2,101 1 2,101 1,658 ,199 sportnews ,866 1 ,866 ,684 ,409 Celebrity * sportnews ,202 1 ,202 ,159 ,690 Q11_geslacht 10,752 1 10,752 8,484 ,004 Q12_leeftijd 1,290 3 ,430 ,339 ,797 Q13_opleiding 11,022 3 3,674 2,899 ,035 Error 418,240 330 1,267 Total 8606,778 341 Corrected Total 444,502 340
a. R Squared = ,059 (Adjusted R Squared = ,031)
Parameter Estimates
Dependent Variable:Brand_attitude
95% Confidence Interval
Parameter B Std. Error t Sig. Lower Bound Upper Bound
[Q11_geslacht=2] 0a . . . . . [Q12_leeftijd=1] ,074 ,331 ,223 ,824 -,577 ,724 [Q12_leeftijd=2] ,033 ,350 ,095 ,925 -,655 ,721 [Q12_leeftijd=3] -,170 ,391 -,436 ,663 -,940 ,599 [Q12_leeftijd=4] 0a . . . . . [Q13_opleiding=2] -,198 ,270 -,736 ,462 -,729 ,332 [Q13_opleiding=3] ,265 ,229 1,156 ,248 -,186 ,717 [Q13_opleiding=4] ,336 ,134 2,509 ,013 ,073 ,600 [Q13_opleiding=5] 0a . . . . .
a. This parameter is set to zero because it is redundant.
Tests of Between-Subjects Effects Dependent Variable:Purchase_intention
Source
Type III Sum of
Squares df Mean Square F Sig.
Corrected Model 14,847a 10 1,485 ,730 ,696 Intercept 1212,022 1 1212,022 596,327 ,000 Celebrity 2,067 1 2,067 1,017 ,314 sportnews ,007 1 ,007 ,003 ,954 Celebrity * sportnews 2,236 1 2,236 1,100 ,295 Q11_geslacht ,199 1 ,199 ,098 ,754 Q12_leeftijd ,990 3 ,330 ,162 ,922 Q13_opleiding 7,678 3 2,559 1,259 ,288 Error 670,718 330 2,032 Total 5216,500 341 Corrected Total 685,565 340
a. R Squared = ,022 (Adjusted R Squared = -,008)
Parameter Estimates
Dependent Variable:Purchase_intention
95% Confidence Interval
Parameter B Std. Error t Sig. Lower Bound Upper Bound
Intercept 3,293 ,473 6,962 ,000 2,362 4,223
[Celebrity=1,00] ,004 ,246 ,017 ,987 -,481 ,489
[Celebrity=2,00] 0a . . . . .
[sportnews=1,00] ,157 ,231 ,679 ,498 -,298 ,611
[Celebrity=1,00] * [sportnews=1,00] -,332 ,317 -1,049 ,295 -,955 ,291 [Celebrity=1,00] * [sportnews=2,00] 0a . . . . . [Celebrity=2,00] * [sportnews=1,00] 0a . . . . . [Celebrity=2,00] * [sportnews=2,00] 0a . . . . . [Q11_geslacht=1] ,050 ,159 ,313 ,754 -,263 ,363 [Q11_geslacht=2] 0a . . . . . [Q12_leeftijd=1] ,244 ,419 ,581 ,561 -,581 1,068 [Q12_leeftijd=2] ,174 ,443 ,393 ,694 -,697 1,045 [Q12_leeftijd=3] ,132 ,495 ,267 ,789 -,842 1,106 [Q12_leeftijd=4] 0a . . . . . [Q13_opleiding=2] ,187 ,341 ,549 ,583 -,484 ,859 [Q13_opleiding=3] -,093 ,291 -,321 ,749 -,665 ,478 [Q13_opleiding=4] ,293 ,170 1,728 ,085 -,041 ,627 [Q13_opleiding=5] 0a . . . . .
