xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2010 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003346
The effect of color congruency and involvement on non-profit organizational messaging
h [electronic resource] /
by Claire Selius.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Thesis (M.A.)--University of South Florida, 2010.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: Color is a major factor in persuasive communication and messaging. This study examines the role of color and other variables in the communication of environmental causes. Many environmental organizations are starting to use color to generate awareness and create symbolic color associations. Based on the congruency theory and the elaboration likelihood model, this study tested specific research hypotheses pertaining to the effects of color-cause congruency and involvement with the cause on consumer reactions through a controlled experiment. Results reveal that involvement and the congruency between color and cause had an effect on attitude toward the organization and attitude toward the message. However, the hypothesized interaction effects of congruency and involvement failed to reach statistical significance. The study thus provides support for the congruency theory and elaboration likelihood model by showing the independent influence of color-cause congruency and cause involvement.
Advisor: Claire Selius
x Mass Communications
t USF Electronic Theses and Dissertations.
The Effect of Color Con gruency a nd Involvement on Non P rofit Organizational Messaging b y Claire Selius A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts School of Mass Communications College of Arts and Sciences University of South Florida Major Professor: Scott Liu, Ph.D. Kelli Burns, Ph.D. Kelly Page Werder, Ph.D. Date of Approval: Febr u ary 16 2010 Keywords: involvement, congruency, intention, non profit, attitudes, reactions Copyrigh t 2010 Claire Seliu s
Acknowledgements This thesis could not have been achieved had it not been for the love and support of so many individuals. To my parents, your inspiration has moved me to strive to fill your footsteps by continuing my education and working hard to achieve success. I cannot thank you enough Dr. Liu, for being my thesis advisor. I appreciate so much your on and off the clock time to work with me, and your push and encouragement has helped to turn this into a work which will mak e us both proud. To Dr. Burns and Dr. Werder, as you have endless obligations and your time is so valuable, I appreciate all of the long hours you have invested in my success. To my family and friends who have been there to support me in the long days a nd even longer nights, there are no words to express my gratitude and appreciation for your support. To my fellow graduate students from whom I have learned so much, this journey is one I will never forget. Our drastically different life experiences have created an opportunity to learn and grow as individuals and friends.
i Table of Contents List of Tables ii List of Figures iv Abstract v Chapter 1: Introduction 1 Chapter 2: Literature Review Elements of Color and Perceptions 4 Cognitive Reacti ons 6 Emotional Reactions 7 Congruency Theory 9 Elaboration Likelihood Model 11 Research Hypotheses 21 Chapter 3: Methodology 25 Participants 25 Experimental Design 26 Procedure 28 Stimulus Materials 28 Dependent Measures 30 Manipulatio n Check Measures 31 Chapter 4: Results 35 Test of Hypotheses 35 Chapter 5: Discussion 49 Discussion of the Findings 49 Study Limitations 54 Pragmatic Implications 56 Theoretical Implications 57 Conclusion 58 References 61 Appendices 66 Ap pendix A: Stimulus Materials 66 Appendix B : Questionnaire 75
ii List of Tables Table 1: Participation Profile 25 Table 2: Experimental Design 26 Table 3 : Distributi on of Participants in Experimental Conditions 29 Table 4 : Reliability Tests for Dependent Measures 31 Table 5 : Reliability Tests for Involvement 32 Table 6 : Reliability Tests for Congruency 32 Table 7: Manipulation Check for Involvement 33 Table 8: T test for Involvement 33 Table 9: Man ipulation Check for Congruency 34 Table 10: T test for Congruency 34 Table 11: Attitude Toward the Cause Mean Scores: Manatee vs. Florida Plants 36 Table 12: Interaction Effect of Involvement and Congruency on Attitude Toward the Cause: Manatee vs. Florida Plants 36 Table 13: Attitude Toward the Cause Mean Scores: Tampa Wildlife vs. Denver Wildlife 37 Table 14: Interaction Effect of Involvement and Congruency on Attitude Toward the Cause: Tampa Wildlife vs. Denver Wildlife 38 Ta ble 15: ATTO: Attitude Toward the Organization Mean Scores: Manatee vs. Florida Plants 39 Table 16: Interaction Effect of Involvement and Congruency on Attitude
iii Toward the Organization: Manatee vs. Florida Plants 40 Table 17: ATTO: Attitude Towa rd the Organization Mean Scores: Tampa Wildlife vs. Denver Wildlife 41 Table 18: Interaction Effect of Involvement and Congruency on Attitude Toward the Organization: Tampa Wildlife vs. Denver Wildlife 42 Table 19: ATTF: Attitude Toward the Flier Mean Scores: Manatee vs. Florida Plants 43 Table 20: Interaction Effect of Involvement and Congruency on Attitude Toward the Flier: Manatee vs. Florida Plants 43 Table 21: ATTF: Attitude Toward the Flier Mean Scores: Tampa Wildlife vs. Denve r Wildlife 44 Table 22: Interaction Effect of Involvement and Congruency on Attitude Toward the Flier: Tampa Wildlife vs. Denver Wildlife 45 Table 23: BI: Behavioral Intention Mean Scores: Manatee vs. Florida Plants 46 Table 24: Interaction Effect of Involvement and Congruency on Behavioral Intention: Manatee vs. Florida Plants 46 Table 25: BI: Behavioral Intention Mean Scores: Tampa Wildlife vs. Denver Wildlife 47 Table 26: Interaction Effect of Involvement and Congruency on Behavior al Intention: Tampa Wildlife vs. Denver Wildlife 48
iv List of Figures Figure 1: The Elaboration Likelihood Model of Pers uasion 14
v The Effect of Color Con gruency and Involvement on Non P rofit Organizational Messaging b y Claire Selius ABS TRACT Color is a major factor in persuasive communication and messaging. This study examines the role of color and other variables in the communication of environmental causes. Many environmental organizations are starting to use color to generate awaren ess and create symbolic color associations. Based on the congruency theory and the elaboration likelihood model, this study tested specific research hypotheses pertaining to the effects of color cause congruency and involvement with the cause on consumer r eactions through a controlled experiment. Results reveal that involvement and the congruency between color and cause had an effect on attitude toward the organization and attitude toward the message. However, the hypothesized interaction effects of congrue ncy and involvement failed to reach statistical significance. The study thus provides support for the congruency theory and elaboration likelihood model by showing the independent influence of color cause congruency and cause involvement.
1 Chapter One I ntroduction Colors are everywhere, and for profit as well as non profit organizations commonly use them to their advantage. Positive and negative color associations are created through our life experiences (Cerbus & Nichols, 1963). We respond cognitively, emotionally and behaviorally to colors, which explains why color influences decision making in today's consumer world. For example, diamond jewelers are beginning to allow consumers to choose the color of the stones in their jewelry using the Pantone Col or Guide (Gambhir, 2009). With the color guide, jewelers are able to track the most popular colors and styles and consumers can then pick the colors based on their popularity. These jewelry trends help make up Pantone's annual color report on the latest co lor crazes in the fashion world. Just to name a few, Fushia Red (Pantone 18 2328), Dark Citron (Pantone 16 0435), Super Lemon (Pantone 14 0704), and Salmon Rose (Pantone 15 1626) earn the highest honors for this year's list of top fashion colors (Gambhir, 2009). Colors are not only symbolic in nature; they also define some organizations. Coca Cola is globally known as having schema of red ; the McDonald's arch is easily recalled as yellow ; and Starbucks quickly comes to mind as green. These corporatio ns have created brand awareness by linking their brands to specific colors. Once a strong brand color link is created, it is nearly impossible to break ( Skorinko, Kemmer, Hebl, & Lane, 2006). Most corporations are not as lucky, and such a clear associatio n of colors rarely occurs.
2 Non profit and cause related organizations utilize colors for association purposes. For example, Susan G. Komen for the Cure promotes breast cancer awareness by featuring the color pink. Its success has attracted support from commercial organizations such as Yoplait yogurt For each Yoplait yogurt container lid mailed back, Yoplait donates 10 cents toward s breast cancer research. Also, Mothers Against Drunk Driving (MADD) is known for its emblematic red ribbons and Autism Soci ety of America is known for dispersing pins with multi colored puzzle piece s to raise awareness. However, as Grossman & Wisenblit (1998) pointed out, non profit organizations need to choose their colors wisely by examining stakeholders' reactions to specif ic color organization associations. All organizations rely on the acceptance of the public for survival and success. Non profit organizations goals differ from those of for profit organizations. Also, they often must create awareness and a public connecti on with limited resources. Unlike for profit organizations whose communication strategies aim at selling products and services, non profit organizations communicate with the public to promote awareness of specific causes, solicit support for a target grou p or area and provide a solution or "cure" to pervasive issues or problems. It is in this unique context that color cause associations render an opportunity for achieving great success in non profit communications. Despite the growing popularity of the use of colors in non profit communications, little empirical research has been done to examine the effectiveness of colors and their association with different causes. Conclusions which link color to cause are often made on the basis of limited data, with little or no consideration of the conditions in which certain these color associations may be more or less effective Preoccupied with short
3 term pragmatic implications, most industry sponsored studies of colors are poor in theoretical reasoning, if not outright atheoretical (e.g., Gambhir, 2009) As such, they lack the power in describing, explaining, and predicting the effects of color in non profit communications. This thesis attempts to provide a theoretical framework for approaching the role of c olor in persuasive communications. A major argument derived from the theoretical framework is that color effects are contingent upon (1) the perceived congruency or fit between the color and the cause, and (2) the levels of involvement or personal relevan ce with non profit causes. This thesis then reports a controlled experiment designed to test research hypotheses derived from the theoretical framework. In presenting the theoretical discussions and empirical data this thesis has been divided into five chapters. Following this Introduction, Chapter Two reviews relevant literature from which research hypotheses were derived. Chapter Three describes the research methodology that was used in the collection of empirical data. Chapter Four presents results of the experiment in detail. Chapter Five summarizes the results and discusses their theoretical and practical implications.
4 Chapter Two Review of Literature Elements of Color and Perceptions Colors are seen constantly and often go unnoticed. The color wheel as Poynter Institute (2008) illustrates, is composed of the two basic color sets of primary (yellow, blue, red) and complementary (orange, green, purple). I n order to recognize exactly what the eye sees it is important to look deeper into t he elements that make up a color The main components of color are hue and brightness. Both of these involve visual sensation s of a color and how we respond to it. Hue is the pigment of a color. In other words, it is the degree to which a color is co mparable to the colors blue, green, red, yellow, or a combination. Hue is determined by wavelength. Cooler colors such as violet, blue and their deviations have short wavelengths, and warm colors like red and orange have longer wavelengths (Rose, 2007). Our eyes prefer shorter wavelengths that can lead to positive and/or improved emotional states (Babin, Hardesty & Suter, 2003). Value is also an important element as it refers to the lightness or darkness of the color (Birren, 1978; Rose, 2007). Adding various amounts of black or white to a color creates different values. Elements of the gray scale can be used if color is not a possibility. An example of this is black and white photography. Texture can still be created because of the variance of color Different colors have various perceived meanings and it is common that individuals respond in specific ways. Red is often perceived as a harsh color that arouses
5 energy level and appetite. It also raises blood pressure and body temperature. It is a very poignant color and tends to stir up such emotions as love and anger. Red is often attached to sexual emotions and connotations including passion, excitement, and infatuation. Pink, a lighter shade of red tends to have calming effects. Often associ ated with young females, pink also connotes sweetness and innocence. Orange most commonly stimulates thirst and can cause the feelings of sociability, cheerfulness, and vigorousness (Babin, Hardesty & Suter, 2003). Yellow, which stimulates memory, is the hardest color for us to see and process, due to its exceptionally long wavelength. It is associated with enlightenment and sunshine. Green behaviorally induces quietness, comfort, and relaxation. It has an extremely short wavelength, so it is th e easiest color for the eye to see. As such, green creates a sense of visual relaxation, which progresses throughout the whole body. Green is also associated with nature, growth, and new beginnings. Blue and its deviations hold the distinction of being t he most popular colors. Blue is proven to lower blood pressure and is known as the color of reliability, longevity, and trust (Cerbus & Nichols, 1963). Also known to help lower blood pressure, purple calms overactive glands and suppresses appetite. It is most commonly associated with religion and loyalty (Babin, Hardesty & Suter, 2003). Finally, black is a very mysterious color and has various meanings. Philosophically, it is the absence of color. Black is frequently coupled with death and evil sugges ting power and dignity Its opposite, white, expresses cleanliness, purity, and innocence (Cerbus & Nichols, 1963). The importance of color in communication is not limited to its influence on human perceptions. Over the years, researchers have identified important cognitive and
6 emotional reactions to color in a variety of contexts. Cognitive Reactions A seminal study by Siple and Springer (1983) examined color memory and matching of hue, chroma and saturation. Participants were exposed to sets of photo graphs containing different fruits and vegetables. They were then asked to match the fruit and vegetable colors as accurately as possible with colors presented to them from the Munsell color chart. They were also instructed to select colors that they beli eved the fruits and vegetables should be. It was discovered that memory of the hue and brightness of colors was precise in matching the colors to the objects. T his was not the case when saturation was recalled. Participants of the study recollected obj ects to be more saturated and when asked, expressed preference for an increased amount of saturation of the color. Siple and Springer (1983) concluded, Such results are more consistent with a theory of memory representation in which the color of an object is stored as an independent attribute rather than one in which color is an integral part of a prototypic representation" (p. 369). Color is often used in advertising to catch the eye of the reader. In magazine advertising, it is difficult to make an ad stand out. People need a reason not to turn the page, and color plays an important role. Schindler (1985) studied the effects of color in magazine advertising in terms of readability and legibility. His results showed that increased contrast achieved h igher readability. However, only 14 percent of the ads studied used high contrast (out of the options of high, moderate, and low). According to Schindler (1985), "Many advertisers fail to optimize the legibility and readability of their
7 ads by the colors they select. Color should enhance, not restrict, legibility" (p. 75). He further cautioned that organizations might have excessively tried to catch the reader's eye. Although some colors may be more fashionable, the eye does not respond to fashion trend s. Some of the basic findings of the effects of color on memory remain relatively unchanged, however. After reviewing extant literature, Clark and Clark (1978) summarily stated: " three main color preferences exist: 1) light colors are remembered mo re and chosen as the best example of a particular color over dark colors, 2) bright colors are remembered more and chosen as the best example of a particular color over dull colors, and 3) focal (or primary) colors (e.g., red, yellow, and blue) are remembe red more and chosen as the best example of a color over secondary colors" (p. 978). Emotional Reactions Emotions are complicated and vary with each individual. Starting at a young age, color can cause an individual to feel specific emotions (Boytzis & Var ghese, 1994). At the age of 5, children have already developed color emotion associations. Brighter colors are preferred (pink, blue, red), while darker colors (brown, black, grey) evoke negative emotions. The partiality for bright colors increases as w e grow. It is hard to ignore our response to colors, especially emotionally. This section looks at research which reveals a connection between color and emotions. Gorn, Chattopadhyay, Yi, and Dahl (1997) studied color in advertising by linking hue, chro ma, and the value of color in an ad to affective and emotional responses. Color was treated as an executional element in a print ad vertisement and the Munsell system
8 was used as the color collaboration system. Gorn et al. (1997) purposely used a product w ith no color associations in order to ensure no bias was formed. Results of the study indicated that increased levels of chroma and value could influence excitement and relaxation feelings. Specifically, high chroma and hues generated positive ad vertiseme nt attitudes by inducing feelings of excitement. This study explains the components of color and illustrates their influence on affective responses of the consumer. The research clearly support s the influence of color on human cognitions and emotions. W hat remains relatively unclear is the power of color in shaping or changing attitudinal and behavioral responses to persuasive communications. Another noticeable limitation of existing research is the assumption of a direct relationship between color and response variables, with little or no consideration of other message elements in communication or the context in which communication takes place. In addition to asking whether color has an effect on communication, an equally important question is under wh at conditions color may exert stronger or weaker influence on the acceptance of a message. Cerbus and Nichols (1963) indirectly explored the color emotion link by relating color response to a number of personality variables. Specifically, they derived r esearch hypotheses from Rorschach (1940) which stated that responses to colors were related to such personality variables as impulsivity or difficulties in control, suggestibility and passivity, and abundance of affect or emotion. Their results showed an inverse relationship between impulse control and responsiveness to color. That is, participants who exhibited the least amount of impulse control were most responsive to color.
9 Similarly, emotionally abundant and disobedient participants were more respons ive to colors than those who were less emotional and obedient. Valdez and Mehribian (1994) more directly examined the relationship between colors and emotions. Unlike previous studies that adhered theoretically to Rorschach (1942), Valdez and Mehribian (1994) derived their hypotheses from the Pleasure Arousal Dominance Emotion Model (PAD). First introduced by Mehrabian and Russel (1974), this PAD model claims that human emotions vary along the dimensions of pleasure displeasure, arousal nonarousal, and dominance submissiveness. The model can thus be used to describe and determine varying emotional states with sufficient detail and flexibility. After carrying out multiple experiments, Valdez and Mehribian (1994) concluded that color brightness and satur ation held a strong relationship with emotional response. In contrast, the relationship between hue and emotion was significantly weaker. In the next sections, we will examine two of the conditioning or qualifying variables in the context of persuasive communication. The first variable is the degree of congruency between color and the theme or cause promoted in the message. The second variable is the level of involvement or personal relevance perceived by the message recipient upon exposure to persuasi ve communication. Congruency Theory and Color Congruency theory was chosen for this study because it explicitly addresses the interactive effects of elements in persuasive communication. In the present study, the theory prescribes that a specific color may have greater (weaker) influence when its perceived meaning is congruent (incongruent) with other elements of the message.
10 Osgood and Tannenbaum (1955) were among the first to explore congruency theory. Their original conception emphasized the im portance of congruency among three key variables: existing attitude toward the source of message, existing attitude toward the concept evaluated by the source, and the nature of the evaluating assertion which relates source and concept in the message. Acc ording to their theory, congruent attitudes would likely result in yielding to persuasive communication. However, incongruent attitudes may give rise to psychological discomfort which creates pressures to seeing resolution, thus prolonging or impeding per suasion (Gassenheimer, Davis, & Dahlstrom, 1998). Campbell and Goodstein (2001) extended the reasoning by showing that congruency is related to conforming to the norm. The level of risk or threat involved in complying with, or deviating from, the percei ved norm may thus determine congruency and incongruency. In the context of non profit communication, for example, the perceived risk in ignoring or rejecting the promoted cause could affect one's reaction to the communication. More pertinent to the prese nt study is Osgood and Tannenbaum's (1955) assertion that "The issue of congruity arises whenever a message is received which relates two or more objects of judgment via an assertion" (p. 43). The "objects" of judgment (or message elements) examined in t his study are the color used and the promoted cause in non profit communication. To illustrate, let's consider pairing money (object 1) and the color black (object 2). We can further assume that money has a positive attitude rating of +5 and black has a negative rating of 5. Pairing the two would result in incongruity and consequently, according to Osgood and Tannenbaum (1955), both money and black would have an
11 assessment of somewhere between +5 and 5. In other words, negative attitudes become less negative, while positive attitudes also become less positive. The reasoning may be extended to congruent pairings between, for example, money (+5) and the color of orange (+3). The theory predicts that money and orange color will be perceived more positi vely when they are presented together than when are presented separately. Intentionally or otherwise, the principle of congruency is often applied in designing advertising messages. For instance, Rolex consistently uses black or dark grey background in it s ads to convey the image of sophistication, classiness, and power (Cerbus & Nichols, 1963). In the case of non profit organizations, color is often used to create congruency. It would be inconceivable for Green Peace not to maximize the congruency of i ts cause with the color of green. Equally inconceivable is the notion of presenting Green Peace in a sea of red or orange. Unlike Green Peace, however, most non profit organizations are not strongly associated with specific colors, and the congruency bet ween the colors used and the causes promoted is often based on intuitive judgment or guesswork. One of the main purposes of this study is to present some initial empirical evidence of the importance of color cause congruency. Elaboration Likelihood Mode l The elaboration likelihood model (ELM) is a major theory in persuasion research. Developed by psychologists Petty and Cacioppo (1986), the model takes into account the different depth or amount of information involved in processing persuasive informat ion. The model may be best be illustrated through consumer processing of advertising information. At one extreme, the consumer can consciously and diligently consider the information provided in the ad in forming attitudes towards the advertised brand.
12 Attitudes are thus changed or formed by careful consideration, thinking, and integration of information relevant to the advertised product. This type of persuasion is termed the central route to attitude change. According to Petty, Cacioppo, and Schumann (1983), "the central route, views attitude change as resulting from a person's diligent consideration of information that s/he feels is central to the true merits of particular attitudinal position" (p.132). Through the central route, advertising could persuade the consumer to take action by presenting convincing product information, rational arguments and concrete evidence. In response, the consumer is likely to process the advertising message deeply and result in long term attitude change and product adoption. A good example of central route persuasion is to tell consumers how fuel efficient and reliable an automobile is. In contrast to such central processing, there also exists the peripheral route to attitude change. "Attitude changes that occur via the peripheral route do not occur because an individual has personally considered the pros and cons of the issue, but because the attitude issue or object is associated with positive and negative cues" (Petty, Cacioppo, and Schumann, 1983, p. 132). I n advertising communication, brand attitudes are formed and changed without active thinking about brand attributes and the strengths and weaknesses. Rather, the persuasive impact occurs by associating the product with positive or negative aspects or execu tional cues in the ad that really are not central to the product. Through the peripheral route, advertising may attempt to persuade the consumer through superficial means. A good example of peripheral persuasion is the use of a well known celebrity as pr oduct spaceperson. The goal is to create positive association between the celebrity and the product without explaining the real reasons why the
13 product should be considered. Compared to central route, peripheral route tends to produce short term attitude change, if any change at all. According to Petty and Cacioppo (1986), a key determining factor in the choice between central and peripheral routes is the level of involvement (or personal relevance) the consumer experiences in the decision process. Wh en involvement is high, as in the process of buying a new car, the consumer is more likely persuaded through the central route. When involvement is low, as in the case of buying a can of soda, the consumer is more likely persuaded through the peripheral r oute. Involvement is a direct function of motivation. Central processing requires first the motivation to process information, because information processing requires effort. Unless there is some reason to expend the energy or effort, the information in the ad will not be actively processed. Such effort will not be made unless the consumer is involved with the product and only unless the information in the ad is relevant and important (Petty & Wegener, 1999). Figure 1 shows how the different routes of persuasion operate
14 Figure 1. The Elaboration Likelihood Model of Persuasion (Petty & Wegener, 1999)
15 From the consumer's point of view, the key implication of the ELM is that motivational factors play a big role in the processin g of information. If motivation is hi gh and the information is processed centrally the consumer is more likely to focus on the reasons why a product or brand is better, resulting in more informed purchase decisions. When the motivation is low and the cha nce for peripheral processing rises the consumer is increasingly affected, willingly or not, by unimportant or perhaps irrelevant product claims in making purchasing decisions. As Batra, R., Myers, J., & Aaker, D. (1996) point out, the fact that advertis ing may be persuasive without providing strong convincing arguments of th e merits of the product may be good news to some advertisers. It also helps explain why consumers sometimes make foolish or erroneous product decisions at a result of advertising exp osure. Involvement Involvement is a facet of ELM that will be used and manipulated as a variable in this experiment. It has also been named personal relevance' by the original theory creators Petty and Cacioppo (1986). Personal relevance is one of the t hree main situational variables introduced which affect our internal motivation to process the information or message. The other two are personal responsibility for message evaluation and the number of message sources (p. 81). The factor of personal relevance is crucial to the model because i t allows us to seek out individual significance in a message. Creating personal relevance is a difficult task for organizations, because it varies so much with each individual. For example, it may be difficult to create personal relevance among college students when speaking of social security, or saving the beaches if reaching out to those who live in mountainous areas. Involvement is a key element
16 when creating non profit organizational communication, and this s tudy recognizes the importance of ELM. Non profit causes reach out to their audience with the hopes that their goals and what the organization is trying to accomplish will be cared about. The use of ELM in this study will help to explain if involvement i ncreases audience awareness and attention, or creates an unexciting, tedious environment for the reader, giving them information or ideas that they have heard many times prior. Sherif and Hovland (1961) originally identified the greater difficulty of per suading a consumer who was already personally invested in an idea, simply because he or she already had set opinions about the issue (p. 82). According to Gardner, Mitchell, and Russo (1985), "Involvement is conceptualized as a state variable with the comp onents, intensity and direction" (p. 4). Intensity is denoted by the amount of consideration devoted to a specific ad, while direction is the processing strategy. It is also stated that low involvement through low attention actually develops a more posit ive attitude toward the ad (p. 5). As involvement increases, the importance of message elaboration in producing persuasion increases (p. 87). This relates directly back to the central and peripheral routes of persuasion. It can be more difficult to adju st or change ideas, views and opinions, specifically if an individual has taken the central route of persuasion, and has high involvement in that issue or message. Message involvement in advertising has an increasingly deciding impact on brand attitude, brand loyalty, and the forming and changing of brand attitudes ( Laczniak, Muehling, & Grossbart, 1989). When devising communication for non profit organizations, this information should be taken into account. It may be more or less difficult to reach an audience depending on their involvement with a specific issue. Through this research, involvement is elucidated and
17 can be manipulated in the experiment in order to further understand the conscious and unconscious thoughts of consumers. ELM and Color The results from prior research studies using ELM do not show a clear consensus on whether the cognition of color and other forms of aesthetic message elements such as music take place in the peripheral or central route. It is necessary to identify and evalua te the elements of both routes in order to choose a starting point and a cognitive basis for this study. Levy and Peracchio (1995) used ELM to identify which type of processing occurs when colors are viewed. Advertisements were used in which color or bl ack and white photos were displayed. The results explain that color's effect on consumer attitude depends on three factors: (1) consumers are motivated to process and substantiate the ad effortfully, (2) ad processing and/or substantiation is highly re source consuming, and (3) the resources available for processing and/or substantiating the ad are sufficient to accommodate both the resources consumed by the presence of potentially distracting color and those required to draw sensory or subjective infere nces from the color" (p. 134). In other words, if the consumer already has high involvement in the issue, this research moves forward the idea that the color in the advertisements has little to no effect. However, if the consumer has little to no inter est or involvement in the issue and the peripheral route is being taken, the color has a large impact in determining attitude toward the product. When color is used as a peripheral cue, it can affect attitude change. This is explained initially by Petty and Cacioppo (1986), as stating, "Variables can affect the amount and direction of attitude change by (a) serving as persuasive arguments, (b) serving as peripheral cues and/or (c) affecting the extent or direction of issue and
18 argument elaboration" (p. 16 ). People hold opinions and views for different reasons, and some are more easily influenced or persuaded than others. Because an individual may or may not hold as headstrong of an opinion, some persuasion techniques are more effective in different situa tions. Color used as a peripheral cue can be more effectual depending on the strength of a position held. Individuals respond differently to not only social and personal situations, but also to messages. In other words, the personal elaboration encounte red with each message depends on the color, depends on involvement as well as other factors with each message It is crucial to understand not only audience respon se to specific colors, but also congruency responses of the consumer. ELM explains involvemen t as an important attribute of message elaboration. However, there is a lack of research combining involvement with other variables. This study will combine involvement with congruency to further understand this relationship. I nvolvement and color/cause c ongruency will be the main independent variables in this experiment ELM and Non Profit Organizations Non profit organizations are only successful with an excessive amount of donations and volunteering. They stay afloat on the public's foundation of hel ping the community, or donating to a cause they find relevant and important. It can be stated that people generally want to do the right thing. The right thing may be defined by society, by family, or by peers. The ELM offers an explanation involving de velopment of elaboration. For example very young children base their decisions on personal preference and what feels good or bad to only themselves. They pay no attention, and doubtfully understand the choices that society wants or prefers them to make. However,
19 once they become older, they begin to notice society's norms. When old enough to understand society, one commences to become influenced by it. Because their ideas and opinions of correct attitude are based on society, the ability to sway opini ons may be increased. Petty and Cacioppo (1986) specify that even though individuals may posses the obligation and motivation to analyze probe issues, "they will lack motivation and ability on others. Thus, simple inferences and affective cues may still product attitude change in adults" (p. 10). Petty and Cacioppo (1986) also explain that individuals may be more motivated and/or able to process a message in an objective manner when argument scrutiny is increased or decreased. Knowing this, non profit o rganizations can take risks by overloading individuals with information through public service announcements, advertisements or collateral materials. According to ELM, it may be more beneficial, and increase motivation if they leave the message recipient some curiosity, or room to find things out for themselves. Allowing the recipient to be active may create a more positive response. If the recipient is receiving the information with high involvement and is motivated, it can be assumed according to ELM, t hat s/he is attempting to do the right thing and is seeking out the truth (p. 19). This can be positive or negative for the organization, depending on the ethical standards and other factors of the organization. ELM furthers the statement that individuals are motivated to hold correct attitudes. But it is also explained that involvement is a factor when correct attitudes are identified. Specifically, Petty and Cacioppo (1986), "Although people want to hold correct attitudes, the amount and nature of is sue relevant elaboration in which they are
20 willing or able to engage to evaluate a message va ry with the individual and situational factors" (p. 6). In other words, individuals will pay attention to correct attitudes only if they are able to stay engag ed in the message. Therefore when related to non profit organizations, the theory explains that non profit organizations have a slim window of opportunity to attract and retain the attention of their target audience. The message must catch the attention of the audience, and must be relevant to even begin to make an impact. When the conditions engage individual's motivations and create the ability to involve issue relevant thinking, the elaboration likelihood' is increased (p. 7). This experiment will a ttempt to gain and keep the attention of the audience by manipulating involvement and color/cause congruency. Park (2004) also used the Elaboration Likelihood Model to study public service announcements for the Special Olympics. Empathetic tendency was i ntroduced as a moderator. Furthermore, involvement was applied as a covariate. As predicted, subjects with lower empathy had an increased motivation to process the strong arguments because the messages were focused on empathy appeal. Also, involvement was successfully manipulated and was established to have an influence on the argument processing of low empathy subject s. Involvement and empathy were successfully prove n to sway subjects one way or the other. Park (2004) adds to the knowledge of ELM by int roducing the variable of empathy. ELM was used in a study by Carson (1991), in order to investigate profit and non profit sponsorship in comparative messaging. Specifically, a 2x3x2 experimental design was used: (profit / non profit sponsor) x (direct comparative ad, indirect comparative ad,
21 and non comparative ad) x (high/low involvement). Through this research, it was unveiled that non profit sponsors were deemed more credible than profit sponsors. Furthermore, attitude toward the ad, attitude to ward the service, and measures of intention to seek information and purchase rated more positively when the ad was sponsored by the non profit organization. As predicted by the ELM, the differences were increased in the low involvement condition. Becau se a new view of involvement is introduced, as well as an insight into what may connect viewer to non profit messaging; these studies directly relate and add insight into this experiment where involvement and cause/color congruency will be examined. Res earch Hypotheses The main objective of this study is to explore the influence of color congruency with varied levels of involvement using the variables of attitude toward the flier and behavioral intention. Non profits use a variety of communication devic es, such as fliers, in an attempt to get noticed by the public and to build the public's knowledge of the organization. In addition, participation is vital in all non profit organizations. Non profits typically focus on building their volunteer base and on fundraising. Because of the unique attributes of non profit organizations, behavioral intention is introduced as a variable in this experiment. In this study, e xperimentation will identify the conditions in which color influences consumer reactions to cause related communications. Specifically, this study attempts to examine cognitive and emotional responses using color in combination with other variables. As attitude is a common term in this experiment, and can be defined by Petty and Cacioppo (1986) as, a general favorable, unfavorable, or neutral evaluation of a
22 person, object, or issue" (p. 26). Effect of consumer response will be moderated by level of involvement and color/cause congruency. Through literature and theoretical framework, the followi ng hypotheses are presented: H ypothesis 1 H1 a. The higher the involvement, the more favorable the attitude toward the cause. H1 b. The higher the involvement, the more favorable the attitude toward the organization. H1 c. The higher the involvement, th e more favorable the attitude toward the flier. H1 d. The higher the involvement, the more higher the behavioral intention. Hypothesis 1 explains that participants will respond more positively to causes in which high involvement is expressed. The Elabora tion Likelihood Model states that a message can change attitudes or create resistance to change. As involvement increases, the importance of message elaboration in producing persuasion increases. If participants have know ledge and aware ness of the cause, a nd they process the cause with high involvement, a positive response will occur. If the participants ideals, interests, and lifestyle characteristics are congruent with the cause, they will be more likely to have a positive response. Hypothesis 2 H2 a. T he higher the color/cause congruency the more favorable the attitude toward the cause.
23 H2 b. The higher the color/cause congruency the more favorable the attitude toward the organization. H2 c. The higher the color/cause congruency the more favorable the attitude toward the flier. H2 d. The higher the color/cause congruency the more higher the behavioral intention. As previously stated, congruency through colors creates a connection between brand and audience. A specific color may have greater (weake r) influence when its perceived meaning is congruent (incongruent) with other elements of the message. It is hypothesized that color cause congruence can root shifts in opinion, thought, and feeling towards a particular cause or message. Hypothesis 3 H3 a. Involvement effects on attitude toward the cause are dependent upon color/cause congruency. H3 b. Involvement effects on attitude toward the organization are dependent upon color/cause congruency. H3 c. Involvement effects on attitude toward the flier are dependent upon color/cause congruency. H3 d. Involvement effects on behavioral intention are dependent upon color/cause congruency. H3: Involvements effects on attitudes and behavioral intention are dependent upon the congruency between color and cause Therefore, involvement levels will have a
24 stronger(weaker) influence on attitudes and behavioral intention when color/cause congruency is high(low). Because of expected interaction effects between variables, if positive attitude toward a high involvemen t cause are present, there will be an increased amount of cause/color congruency and vice versa. H3 introduces the variable of interaction effects between involvement and color cause congruency. Much research has been done on color and color effects. Ho wever, H3 fills a gap in social science research in which it will help to create not only a greater understanding of involvement, but demonstrate a significant connection between the congruency of the color and the cause; w hile also linking these variables to various attitude scales. Previous research listed in the review of literature has made elementary connections between these variables, but never tested for specific interaction effects. Through this hypothesis, it is expected that interaction between attitude toward the cause attitude toward the organization, attitude toward the flier, and behavioral intention will create significant results revealing that these variables are directly related and dependant upon color/cause congruency and involvement. Ideally, this hypothesis will add to the ELM as well as Congruency Theory.
25 Chapter Three Methodology Participants A total of 103 undergraduate students at the University of South Florida participated in the experiment. The students majors wer e mass communications and biology. Some received extra credit for participation, but most completed the experiment voluntarily. A participant profile is presented in Table 1. Table 1 Distribution of Participants in Experimental Conditions Frequency Pe rcent Gender Male 35 34.0 Female 68 66.0 Total 103 100.0 Student Status S o phomore 25 24.3 Junior 35 34.0 Senior 43 41.7 Total 103 100.0 Ethnicity Caucasian 69 67.0 African American 10 9.7 Hispanic 17 16.5 Asian 5 4.9 Other 2 1.9 Tota l 103 100.0 Home town Tampa, FL 19 18.4 Florida 90 87 Other States 14 14 Total 103 100.0
26 Experimental Design The experiment al design was a 2x2 factorial: Involvement with cause (high vs. low) and color/cause congruency (high vs. low). The design created four between subjects conditions: High color/cause congruency + high cause involvement, high color/cause congruency + low cause involvement, low color/cause congruency + high cause involvement, and low color/cause congruency + low cause involvemen t. Each of the four conditions were replicated twice using different color/cause combinations, resulting in eight experimental stimuli as shown in Table 2. Table 2. Experimental Design High Involvement Cause Low Involvement Cause High Color/Cause C ongruency Manatee Protection Blue Tampa Wildlife Protection Green Florida Plants and Palms Blue Denver Wildlife Protection Green Low Color/Cause Congruency Manatee Protection Orange Tampa Wildlife Protection Red Florida Plants and Palms Or ange Denver Wildlife Protection Red Involvement with the cause wa s the first independent variable manipulated in the design. Consistent with ELM, involvement with the cause determine s how much personal atta chment or relevance one may have in reaction to a specific cause (Petty & Cacioppo, 1986) A set of four fictional organizat ions were created for this study. Fictional organizations were used to minimize the influence of prior familiarity and other extraneous factors. Two of the organizations, Man atee Protection Association and Florida Plants and Palms Association, represented high and low involvement causes as determined by perceived relevance and importance of the organizations' purposes. Two other organizations, Tampa Wildlife Protection and De nver Wildlife Protection, represented high and low involvement as determined by geographic proximity.
27 Geographic proximity has frequently been used in previous studies to create variations in personal involvement. Petty and Cacioppo (1986), for example, manipulated involvement by comparing issues occurring at near and far locations. The second independent variable, color/cause congruency, was manipulated by paring organizations with different background and font colors in communication flyers. Specific ally, four different colors were chosen for the present study: green, blue, orange and red. Green symbolizes verdancy of growing things (Sloane (1989), p. 123) and is commonly associated with nature and the environment schema (Dee & Taylor, 2002). Blue a calming and pleasing color is rated the most popular color in America It is related to perceptions of longevity, trust, reliability and often associated with water, specifically the ocean ( Cerbus & Nichols, 1963) Opposite to blue, orange is an ene rgetic color that symbolizes exuberance, playfulness, and liveliness However, it also connotes such negative meanings as laziness and dependency ( Nor ris, 2001; Dee and Taylor 2002; Jasper 2005) Red has a similar wavelength as orange and is also a bri ght and ha rsh color. Its vibrancy is known to stir up emotions related to sex, passion, lust and love (Babin, Hardesty & Suter, 2003) A series of pretests were conducted to determine the congruency between the four fictional organizations and colors. S pecifically, each organization was presented in four different flyers, and each flyer used a different color background and font. A total of sixteen flyers (4 organizations x 4 colors) were thus created. They were presented to groups of participants simi lar to those in the main experiment for congruency and "fit" rankings. Based on the ranking data, the following organization/color combinations were selected for the main experiment.
28 High Congruency Manatee Protection Association Blue Tampa Wildlife As sociation Green Florida Plants and Palms Association Blue Denver Wildlife Association Green Low Congruency Manatee Protection Association Orange Tampa Wildlife Association Red Florida Plants and Palms Association Orange Denver Wildlife Associat ion Red Stimulus Materials The eight flyers created for the experiment are presented in Appendix A. Each flyer (letter sized, single sided) contains pictures of the causes promoted by the organization. The headline and body copy describe the causes and calls for participation, donation and/or request for further information. To maximize internal validity, the verbal content of the flyers were kept nearly identical. Procedure Eight experimental sessions took place in regularly scheduled undergraduate classes. Each session consisted of eight to twenty participants and they were randomly assigned to one of the four experimental conditions. At the beginning of each session, participants were informed that the purpose of the study was about organizationa l messaging. They were then given the response questionnaire and instructed to pay attention to the PowerPoint presentation on the screen. The PowerPoint presentation
29 showed two fliers, one at a time, in each session. Each flier was presented for three minutes, followed by instructions to answer questions pertaining to the flyer on the questionnaire. On average, the experimental sessions took about 15 20 minutes to complete. Due to time and resource limitations, the distribution of participants in the four experimental conditions was uneven. The actual number of participants in each of the experimental conditions is presented in Table 3. Table 3 Distributi on of Participants in Experimental Conditions Condition n % High Involvement/ High Congruenc y 27 26 High Involvement/ Low Congruency 33 32 Low Involvement/ High Congruency 20 19 Low Involvement/ Low Congruency 23 22 Total 103 100 Specifically, participants in the four conditions were exposed to the experimental stimuli as follows : The fir st group ( n =27) viewed two high involvement and high color/congruency fliers (Manatee Protection Association Blue, Tampa Wildlife Protection Association Green); the second group ( n =33) viewed two high involvement and low color/congruency fliers (Manatee Protection Association Orange, Tampa Wildlife Protection Association Red); the third group ( n =20) viewed two low involvement and high color/congruency fliers (Florida Plants and Palms Society Blue and Denver Wildlife Protection Association Green); and the fourth group ( n =23) viewed two
30 low involvement and low color/congruency fliers ( Florida Plants and Palms Society Orange and Denver Wildlife Protection Association Red). Dependent Measures Four sets of dependent measures were taken to determine participant s' reaction to the experimental stimuli (see Appendix B). They were attitude toward the cause (ATTC), attitude toward the organization (ATTO), attitude toward the flier (ATTF), and behavioral intention (BI). ATTC and ATTF were measured by four 7 point se mantic differential items (Good/Bad, Favorable/Unfavorable, Likeable/Unlikeable, Positive/Negative). ATTO was measured by five 7 point semantic differentials (Good/Bad, Favorable/Unfavorable, Likeable/Unlikeable, Positive/Negative, Important/Unimportant). Behavioral intention was measured by the likelihood of (1) obtaining more information from the organization, (2) donating to the organization, (3) telling others about the organization, and (4) volunteer for the organization on 7 point (Likely/Unlikely) scales. All attitude and behavioral intention measures were adopted from previous studies of persuasive communications (e.g., MacKenzie, Lutz, and Belch). Reliability tests were performed to ensure the internal consistency of the dependent measures. Resu lts (Table 3) indicated all measures achieved "acceptable" reliability, as determined by Cronbach's alphas (Morgan, Reichert, & Harrison, 2002) Specifically, alphas ranged from .789 to .901 for ATTC, .896 to .942 for ATTO, and .863 to .908 for ATTF acros s high and low involvement organizations. Similarly, alphas for BI were .892 and .918 for high and low involvement organizations respectively.
31 Table 4. Reliability Tests for Dependent Measures Measure Cronbach's Alpha Cronbach's Alpha Based on Sta ndardized Items N of Items ATTC (High Inv.) .789 .794 4 ATTC (Low Inv.) .901 .905 4 ATTO (High Inv.) .896 .900 5 ATTO (Low Inv.) .942 .946 5 ATTC (High Inv.) .863 .858 4 ATTC (Low Inv.) .908 .908 4 BI (High Inv.) .892 .894 4 BI (Low Inv.) .918 .91 9 4 Manipulation Check Measures Manipulation checks offer evidence that the variables manipulated were accurately understood by the receivers the way they were meant to be" (Morgan, Reichert & Harrison, 2002). For each flier, two sets of measures were included in the questionnaire to determine if the experiment successfully manipulated the independent variables of interest. The first independent variable, involvement with the organization/cause, was measured by the Personal Involvement Inventory (PII) developed by Zaichkowsky (1985). The PII is composed of 20 semantic differential items scored on 7 point scales. The second independent variable, color/cause congruency, was measured by seven 7 point semantic differential items (good/bad, fitting/unfitti ng, congruent/incongruent, match/mismatch, creative/uncreative, correct/incorrect,
32 suitable/unsuitable). Both sets of manipulation check measures showed high reliability, as measured by Cronbach's alpha ( Table 5 and 6 ). Table 5. Reliability Tests for In volvement Condition Cronbach's Alpha Cronbach's Alpha Based on Standardized Items N of Items High Involvement .951 .952 20 Low Involvement .951 .951 20 Table 6. Reliability Tests for Congruency Condition Cronbach's Alpha Cronbach's Alpha Based o n Standardized Items N of Items High Congruency .900 .899 7 Low Congruency .939 .940 7 A series of t tests were performed to determine if the experiment manipulated the independent variables successfully. Since the experiment contained two replication s (i.e., both involvement with the cause and color/cause congruency were tested twice in each of the four experimental conditions), the t tests were performed separately for each replication. Test results indicate that participants were more involved with the Manatee Protection (Mean = 4.60) than Florida Plants Protection (Mean = 4.09) ( t = 2.67, df = 86, p= .009 ) They were also more involved with Tampa Wildlife (Mean = 4.73) than Denver Wildlife (Mean = 4.37), although the difference was only marginally significant ( t = 1.82, df = 86, p = .072 ). The manipulation of color/cause congruency was also successful. Blue color (Mean = 3.59) was rated more congruent than orange (Mean = 2.85) for Manatee
33 Protection and Florida Plants Protection ( t = 3.195, df = 84, p = .002 ); green color (Mean = 4.15) was rated more congruent than red (Mean = 3.31 for Tampa Wildlife Protection and Denver Wildlife Protection (t = 2.901, df=83, p = .005 ). Table 7 Manipulation Check for Involvement Involvement /Replication Fliers N Mean Std. Deviation High /A Manatee/Blue & Manatee/Orange 50 4.6000 .94031 Low /A Florida Plants/Blue & Florida Plants/Orange 38 4.0900 .78466 High /B Tampa Wildlife/Green & Tampa Wildlife/Red 51 4.7300 .95302 Low /B Denver Wildlife/Green & Denver Wild life/Red 37 4.3700 .90094 Table 8 T test for Involvement t df Sig. Mean Diff. Std. Error Difference Involvement A 2.67 86 .009 .507 .189 Involvement B 1.82 86 .072 .367 .201
34 Table 9 Manipulation Check for Congruency Congruency/ Replication F liers N Mean Std. Deviation High /A Manatee/Blue & Florida plants/Blue 37 3.5900 1.04446 Low /A Manatee/Orange & Florida Plants/Orange 49 2.8500 1.08304 High /B Tampa Wildlife/Green & Denver Wildlife/Green 36 4.1500 1.19618 Low /B Tampa Wildlife/Red & Den ver Wildlife/Red 49 3.3100 1.39117 Table 10 T test for Congruency t df Sig. Mean Diff. Congruency A 3.195 84 .002 .742 Congruency B 2.901 83 .005 .836
35 Chapter Four Results In this chapter, hypothesis testing results are presented. Each hypothesi s was tested twice, one in each experimental replications. To facilitate presentation, test results are organized by the dependent variables of interest. The principal statistical procedure used was Analysis of Variance (ANOVA) which allowed comparisons between sample means. H1 a H2 a, H3 a: Effects on Attitude Toward the Cause Hypotheses 1 a, 2 a, and 3 a predict the main and interaction effects of cause involvement and cause/color congruency on attitude toward the cause (ATTC). For Florida Manatee Prot ection vs. Florida Plants Protection, ANOVA results showed that both main effects of cause involvement and cause/color congruency reached statistical significance. Supporting H1 a, attitude toward Manatee Protection (Mean = 6.15), the high involvement cau se, was more favorable than attitude toward Florida Plants (Mean = 5.57), the low involvement cause (F (1, 82) = 8.24 p = !""#$%& 2 = .09) Consistent with H2 a, high cause/color congruency (Manatee Protection/Blue and Florida Plants Protection/Blue, Mean = 6.14) resulted in more positive attitude toward the cause than low cause/color congruency (Manatee Protection/Orange and Florida Plants Protection/Orange, Mean = 5.71) ( F (1, 82) = 5.20 p = !"'#$%& 2 = .06 ). The interaction effect between cause involveme nt and cause/color congruency, however, failed to reach
36 statistical significance (F(1,82) = .564, p = .455 $%& 2 = .0 07). H3 a was thus not supported in this group. Table 11 Attitude Toward the Cause Mean Scores: Manatee vs. Florida Plants Involvement Congruency Mean Std. Deviation N High : Manatee High : Blue 6.3100 .83256 21 Low : Orange 6.0200 .93265 27 Total 6.1500 .89299 48 Low : Florida Plants High : Blue 5.1900 .58363 16 Low : Orange 5.3300 .97708 22 Total 5.5700 .87361 38 Total High : Blue 6 .1400 .75368 37 Low : Orange 5.7100 1.00435 49 Total 5.8900 .92477 86 Table 12 Interaction Effect of Involvement and Congruency on Attitude Toward the Cause: Manatee vs. Florida Plants ATTC Type III Sum of Squares df Mean Square F Sig. Partial Eta Sq uare d Corrected Model 11.056 3 3.685 4.903 .003 .152 Intercept 369.496 1 369.496 491.570 .000 .857 Involvement 6.194 1 6.194 8.240 .005 .091 Congruency 3.909 1 3.909 5.200 .025 .060 Involvement Congruency .424 1 .424 .564 .455 .007 Error 61.637 82 .752 Total 454.688 86 Corrected Total 72.693 85
37 For Tampa Wildlife vs. Denver Wildlife, however, ANOVA results showed none of the main and interaction effects of cause involvement and cause/color congruency. For lack of statistical signific ance in this replication, H1 a, H2 a and H3 a were not supported. (Involvement, (F (1, 82) = 1.64 p = !'"($%& 2 = .02) Cause/color Congruency, (F (1, 82) = .693 p = !)"*$%& 2 = .008) Interaction, (F (1, 82) = .003 p = !+#,$%& 2 = .00) Table 13 ATTC: Attit ude Toward the Cause Mean Scores: Tampa Wildlife vs. Denver Wildlife Involvement Congruency Mean Std. Deviation N High : Tampa Wildlife High : Green 6.1200 .85739 21 Low : Red 5.9300 1.03963 27 Total 6.0100 .95922 48 Low : Denver Wildlife High : Green 5. 8300 .59665 16 Low : Red 5.6600 1.24295 22 Total 5.7300 1.01406 38 Total High : Green 5.9900 .76031 37 Low : Red 5.8100 1.13108 49 Total 5.8900 .98791 86
38 Table 14. Interaction Effect of Involvement and Congruency on Attitude Toward the C ause: Tampa Wildlife vs. Denver Wildlife ATTC Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 2.370 a 3 .790 .804 .495 .029 Intercept 372.273 1 372.273 378.80 0 .000 .822 Involvement 1.615 1 1.615 1.643 .203 .020 Congrue ncy .681 1 .681 .693 .408 .008 Involvement Congruency .003 1 .003 .003 .956 .000 Error 80.587 82 .983 Total 467.063 86 Corrected Total 82.957 85 H1 b, H2 b, H3 b: Effects on Attitude Toward the Organization Hypotheses H1 b, H2 b, and H3 b predict the main and interaction effects of cause involvement and cause/color congruency on attitude toward the organization (ATTO). Results pertaining to attitude toward the organization (ATTO) were nearly identical to that of attitude toward the caus e (ATTC). For Florida Manatee Protection vs. Florida Plants Protection, ANOVA results revealed a significant main effect of cause involvement. For Florida Manatee Protection vs. Florida Plants Protection, ANOVA results showed that both main effects of c ause involvement and cause/color congruency were statistically significant. Supporting H1 b, attitude toward Manatee Protection (Mean = 6.08), the high involvement cause, was more favorable than attitude toward Florida
39 Plants (Mean = 5.32), the low involv ement cause (F (1, 82) = 10.75 p = ""'$%& 2 = .116) Consistent with H2 b, high cause/color congruency (Manatee Protection/Blue and Florida Plants Protection/Blue, Mean = 6.10) resulted in more positive attitude toward the cause than low cause/color congruency (Manatee Protection/Orange and Florida Plants Protection/Orange, Mean = 5.433) ( F (1, 82) = 8.983 p = !"")$%& 2 = .099 ). The interaction effect between cause involvement and cause/color congruency, however, failed to reach statistical significance and H3 a was not supported (F(1,82) = .150, p = .700 $%& 2 = .0 02). Table 15 ATTO: Attitude Toward the Organization Mean Scores: Manatee vs. Florida Plants Involvement Congruency Mean Std. Deviation N High : Manatee High : Blue 6.4500 .76395 21 Low : Orange 5.7100 1.21444 27 Total 6.0800 1.09570 48 Low : Florida Plants High : Blue 5.6500 .72111 16 Low : Orange 5.0800 1.04318 22 Total 5.3200 .95356 38 Total High : Blue 6.1000 .83749 37 Low : Orange 5.4300 1.17260 49 Total 5.7200 1.08915 86
40 Table 16. Interaction Effect of Involv ement and Congruency on Attitude Toward the Organization: Manatee vs. Florida Plants ATTO Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 20.158 a 3 6.719 6.830 .000 .200 Intercept 430.828 1 430.82 8 437.9 22 .000 .842 Inv olvement 10.571 1 10.571 10.74 5 .002 .116 Congruency 8.838 1 8.838 8.983 .004 .099 Involvement Congruency .147 1 .147 .150 .700 .002 Error 80.672 82 .984 Total 548.440 86 Corrected Total 100.830 85 Just as in attitude toward the cause ( ATTC), the ANOVA results for Tampa Wildlife vs. Denver Wildlife showed none of the main and interaction effects of cause involvement and cause/color congruency. Therefore in this replication, H1 b, H2 b and H3 b were not supported. (Involvement, (F (1, 82) = 2.36 p = !-'*$%& 2 = .028) Cause/color Congruency, (F (1, 82) = 2.14 p = !-).$%& 2 = .025) Interaction, (F (1, 82) = .255 p = !$%& 2 = .003)
41 Table 17. ATTO: Attitude Toward the Organization Mean Scores: Tampa Wildlife vs. Denver Wildlife Invol vement Congruency Mean Std. Deviation N High : Tampa Wildlife High : Green 6.0900 .76830 21 Low : Red 5.8700 1.11907 27 Total 5.9600 .97776 48 Low : Denver Wildlife High : Green 5.8500 .81813 16 Low : Red 5.4000 1.28285 22 Total 5.5900 1.12076 38 Tota l High : Green 5.9800 .78793 37 Low : Red 5.6600 1.20554 49 Total 5.8000 1.05362 86 Table 18. Interaction Effect of Involvement and Congruency on Attitude Toward the Organization : Tampa Wildlife vs. Denver Wildlife ATTO Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 5.394 a 3 1.798 1.657 .183 .057 Intercept 401.837 1 401.837 370.37 5 .000 .819 Involvement 2.561 1 2.561 2.361 .128 .028 Congruency 2.324 1 2.324 2.142 .147 .025 Involvement Congruen cy .277 1 .277 .255 .615 .003 Error 88.966 82 1.085 Total 511.480 86 Corrected Total 94.360 85
42 H1 c, H2 c, H3 c: Effects on Attitude Toward the Flier Results pertaining to attitude toward the flier (ATTF) varied from attitude toward the c ause (ATTC) and attitude toward the organization (ATTO). Hypotheses H1 c, H2 c, and H3 c predict the main and interaction effects of cause involvement and cause/color congruency on attitude toward the flier (ATTF). For Tampa Wildlife vs. Denver Wildlife, A NOVA results revealed a significant main effect of cause involvement. For Florida Manatee Protection vs. Florida Plants Protection, ANOVA results showed no statistical significance for the main or interaction effects. Therefore, this replication revealed no support for H1 c, H 2c, and H3 c. Table 19 ATTF: Attitude Toward the Flier Mean Scores: Manatee vs. Florida Plants Involvement Congruency Mean Std. Deviation N High: Tampa Wildlife High: Green 3.63 1.23132 26 Low: Red 4.33 1.43324 32 Total 4.02 1.37957 58 Low: Denver Wildlife High: Green 5.12 1.03714 20 Low: Red 4.27 1.25886 23 Total 4.67 1.22562 43 Total High: Green 4.28 1.36179 46 Low: Red 4.31 1.35117 55 Total 4.29 1.34926 101
43 Table 20. Interaction Effect of Involvement and Congru ency on Attitude Toward the Flier: Manatee vs. Florida Plants ATTF Type III Sum of Squares df Mean Square F Sig. Partial Eta Squared Corrected Model 7.794 a 3 2.598 1.608 .194 .056 Intercept 1105.028 1 1105.028 684.082 .000 .894 Involvement 3.850 1 3.850 2.383 .127 .029 Congruency .188 1 .188 .116 .734 .001 Involvement Congruency 5.088 1 5.088 3.150 .080 .037 Error 130.843 81 1.615 Total 1300.438 85 Corrected Total 138.637 84 For Tampa Wildlife vs. Denver Wildlife, however, ANOVA resu lts did show significance in the main effect of involvement, consistent with H1 c. In this replication, attitude toward Tampa Wildlife (Mean = 4.18), the high involvement cause, was more favorable than attitude toward Denver Wildlife (Mean = 5.06), the low involvement cause ( F (1, 82) = 10.291 p = !""'$%& 2 = .112 ). However, H2 b, cause/color congruency showed no statistical significance ( F (1, 82) = .836 p = !(,($%& 2 = .112) And just as in the previous results for interaction effects, ATTF showed no significance and was inconsistent with H3 c. The refore, in this replication, H1 a, was supported, while color/cause congruency and interaction effects showed no statistical significance, revealing no support for H2 c and H3 c.
44 Table 21 ATTF: Attitude Toward the Flier Mean Scores: Tampa Wildlife vs. Den ver Wildlife Involvement Congruency Mean Std. Deviation N High : Tampa Wildlife High : Green 4.0800 1.42156 21 Low : Red 4.2500 1.66266 27 Total 4.1800 1.54796 48 Low : Denver Wildlife High : Green 5.4700 .56917 16 Low : Red 4.7600 1.22833 22 Total 5.0 600 1.05497 38 Total High : Green 4.6800 1.31979 37 Low : Red 4.4800 1.49115 49 Total 4.5700 1.41547 86
45 Table 22 Interaction Effect of Involvement and Congruency on Attitude Toward the Flier: Tampa Wildlife vs. Denver Wildlife ATTF Type III Sum of Sq uares df Mean Square F Sig. Partial Eta Squared Corrected Model 21.467 a 3 7.156 3.942 .011 .126 Intercept 937.336 1 937.336 516.419 .000 .863 Involvement 18.679 1 18.679 10.291 .002 .112 Congruency 1.518 1 1.518 .836 .363 .010 Involvement Congruency 3.966 1 3.966 2.185 .143 .026 Error 148.836 82 1.815 Total 1183.938 86 Corrected Total 170.303 85 H1 d, H2 d, H3 d: Effects on Behavioral Intention Hypothesis H1 d, H2 d, and H3 d predict the main and interaction effects of cause involvem ent and cause/color congruency on behavioral intention (BI). Results relating to behavioral intention (BI) revealed no statistical significance in the main or interaction effects. (Tampa Wildlife vs. Denver Wildlife, ( Involvement) F (1, 83), = .281, p = #+.$%& 2 = .003). (Congruency) F (1, 83) =2.445, p = !-''$%& 2 = .029). (Interaction) F (1, 81) = 2.566, p = !--($%& 2 = .030). Therefore, H1 d, H2 d, and H3 d were not supported in this replication.
46 Table 23 BI: Behavioral Intention Mean Scores: Manatee v s. Florida Plants Involvement Congruency Mean Std. Deviation N High : Tampa Wildlife High : Green 2.001 .72457 21 Low : Red 2.000 1.20662 28 Total 2.001 1.01869 49 Low : Denver Wildlife High : Green 2.250 1.29422 16 Low : Red 1.511 .96510 22 Total 1.82 3 1.15943 38 Total High : Green 2.108 1.00267 37 Low : Red 1.790 1.12413 50 Total 1.925 1.07975 87 Table 24 Interaction Effect of Involvement and Congruency on Behavioral Intention: Manatee vs. Florida Plants BI A Type III Sum of Squares df Mean Squa re F Sig. Partial Eta Squared Corrected Model 5.769 a 3 1.923 1.689 .176 .058 Intercept 3069.085 1 3069.085 2695.7 43 .000 .970 Involvement .320 1 .320 .281 .597 .003 Congruency 2.784 1 2.784 2.445 .122 .029 Involvement Congruency 2.922 1 2.922 2.566 .113 .030 Error 94.495 83 1.138 Total 3310.750 87 Corrected Total 100.264 86
47 The ANOVA results for Tampa Wildlife vs. Denver Wildlife showed none of the main and interaction effects of cause involvement and cause/color congruency. Therefor e, behavioral intention (BI) showed no effects and resulted in a lack of support for H1 d, H2 d and H3 d in this replication. Table 25 BI: Behavioral Intention Mean Scores: Tampa Wildlife vs. Denver Wildlife Involvement Congruency Mean Std. Deviation N High : Tampa Wildlife High : Green 2.2700 1.18560 21 Low : Red 2.2500 1.39940 28 Total 2.2600 1.29900 49 Low : Denver Wildlife High : Green 3.0000 1.46059 16 Low : Red 2.4200 1.22082 22 Total 2.6400 1.33972 38 Total High : Green 2.5900 1.34270 37 Low : Red 2.3300 1.31344 50 Total 2.4400 1.32464 87
48 Table 26 Interaction Effect of Involvement and Congruency on Behavioral Intention: Tampa Wildlife vs. Denver Wildlife BI B Type III Sum of Squares df Mean Square F Sig. Partial Eta Square d Correc ted Model 6.616 a 3 2.205 1.269 .291 .044 Intercept 2543.054 1 2543.054 1462.8 79 .000 .946 Involvement 4.203 1 4.203 2.418 .124 .028 Congruency 1.903 1 1.903 1.095 .298 .013 Involvement Congruency 1.615 1 1.615 .929 .338 .011 Error 144.286 83 1.738 Total 2843.500 87 Corrected Total 150.902 86
49 Chapter Five Discussion and Implications Discussion The analys es of the data collected revealed several patterns as well as other interesting findings which are highlighted in this section. In re view, the hypotheses of this study attempted to explain the main affect of cause involvement, color/cause congruency and the interaction of these variables. The interaction hypothesis is explained by stating that involvement levels will have a stronger(wea ker) influence on attitudes and behavioral intention when color/cause congruency is high(low) and vice versa. The research did express that attitude toward the cause and attitude toward the organization can be affect ed by variables such as involvement and cause/color congruency. Through the results, attitude toward the cause and attitude toward the organization showed the most significance when interacting with involvement and congruency specifically in Manatee Protection vs. Florida Plants. Through these outcomes, it is clear the higher the involvement and the color/cause congruency, the more favorable the attitude toward the cause and the attitude toward the organization. These findings are a step forward for these variables and succeed in proving that c onsumers are affected by the color of advertisements, and depending on how involved they are in the cause, attitudes can be shifted. Attitudes are important and can lead to loyalty and future business for that cause or organization. It is expensive for n on profits to use most communication tools; therefore they must reach their audiences in the most direct way
50 possible. Combining and focusing on these variables can give non profit and for profit organizations an edge that others may not hold. Significa nce was also shown in ANOVA testing while measuring attitude toward the flier in the variable of involvement, specifically to the Tampa Wildlife and Denver Wildlife replication. Attitude toward the flier is the most direct route, and an instinctual reacti on by the participants. In the case of this research, it can be concluded that it is possible to shift attitudes when a visual communication device connects to the audience through involvement. Also, a connection can be created with the audience if the co lor in the flier is consistent with the color expectations of the audience. For example, when water causes are blue, forest causes are green, or desert causes are orange and yellow. Attitude toward the flier showed some statistical significance, but onl y in certain replications. It can be argued that any advertisement or visual stimuli such as a flier is just a message tool, which leads to the creation of attitudinal decisions about the organization or cause. Therefore, the results for attitude toward th e flier are of less importance than that of the other variables. Although a limited number of conditions were found to be statistically significant through ANOVA testing, when looking at the means, it is illustrated that with the exception of the Manatee vs. Florida Plants replication of attitude toward the flier and attitude toward the cause in both replications, high involvement/ high congruency (Manatee/Blue and Tampa/Green) showed higher means than the high involvement/low condition (Manatee/Orange an d Tampa/Red). Furthermore, the low involvement/ high congruency condition (Florida Plants/Blue and Denver/Green) illustrated higher means
51 than the low involvement/ low congruency condition (Florida Plants/Orange and Denver/Red). With the exception of both sets of fliers measuring attitude toward the flier and the Tampa vs. Denver replication of behavioral, the affect of involvement was shown to be most significant The variable involvement illustrated the lowest p value regardless of statistical significa nce, meaning it consistently had the most affect on the dependent variables. The means offer support of the hypothesis, however the ANOVA results failed to do so. This just adds to the complication of involvement. The Elaboration Likelihood Model does s uggest that involvement is a multi dimensional variable that depends on many factors including environment, situation and social pressures ( Laczniak, Muehling, & Grossbart, 1989). Furthermore, involvement of fictitious organizations may have been an obstac le in this experiment. Because participants have no prior knowledge of the organization, involvement can only be manipulated to a certain point before extraneous factors begin to play a role. When looking at the means, it can also be argued that attitud e toward the cause and attitude toward the organization are more likely to be affect ed by the variables of involvement and congruency than that of the other variables, although there is no statistical significance proving this. All organizations seek the g oals of acceptance, loyalty, and support from their audience using various forms of messaging. The results from this experiment suggest that using the manipulation factors of involvement and color/cause congruency, attitude toward the cause and attitude t oward the organization can be altered
52 Also when looking at the means of low involvement and high congruency, support is displayed that color/cause congruency can impact attitudes and behavioral intention as displayed through mean measures. These results also further ELM by agreeing that color is more affect ive as a peripheral cue than a central cue. As previous studies have shown, and Petty and Cacioppo (1986) explain, color has more of an affect when used as a peripheral cue in messages. As described in the results section, this study yielded a variation of results not identified in the hypotheses. The complete ANOVA results revealed many questions, specifically, the unexpected difference in the replications of Manatee vs. Florida Plants and Tampa vs. Denver. There are many possibilities which could prove as insight. It is possible that because manatees are actually living creatures, rather than plants or general wildlife, they may appeal more to participants, and create more of a realistic threat. Al so, color may be contributing to the lack of statistical significance. Perhaps the participants were not able to connect to the causes enough because they didn't feel the color related enough to the cause. It is another possibility that the colors in the Manatee vs. Florida Plants fliers make more of an impact than those in Tampa vs. Denver fliers. If this is the case, blue and orange reach and are able to sway audiences in a way that green and red cannot. Furthermore, these factors could add to the ELM by explaining that elaboration of messages can be swayed more by specific peripheral cues (such as individual colors) more than others. Behavioral intention did not show any statistical significance, and showed very low means. Therefore, the variables had little to no affect on manipulating intentions to become pro active with the cause or the organization. This furthers the belief that
53 behaviors are very complicated, especially when linking to attitudes. College students often have good intentions, and as the research shows from the composite means, have positive attitudes about the organizations and causes presented. This demographic however, may fall short when it comes to actions or intentions to act, because of their distinctive set of prioritie s. It is a common concept that attitudes may not always lead to behaviors. According to Ajzen and Fishbien (1977), "Attitudinal and behavioral entities may be viewed as consisting of four different elements: the action, the target at which the action is d irected, the context in which the action is performed, and the time at which it is performed" (p. 889). It is further explained that in order for attitudes to predict behaviors, the attitudinal and behavioral entities must match, and the attitude must be strong enough to create action. In this case, it is possible that attitudes were not strong enough in this case to even consider action. For example, although ninety percent of the population feels that the declining environment is a problem, and eighty percent have access to recycling programs such as plastic bottle recycling, only 24 percent of those programs are used (American Chemistry Council, 2008). Also, Americans have created so many different diets, and are supposedly becoming more health consc ious. Restaurants have lower calories and low fat sections, and healthy cookbooks are on the rise. However, in most states, adult obesity rate s have either stayed the same or increased (Center for Disease Control and Prevention, 2008). These statistics sh ow good intentions and attitudes, but resistance to follow through and be active. The ELM is a very strong and helpful theory, which explains much message and issue elaboration. As previously stated, ELM explains that many variables combine to
54 create a sp ecific context for overall persuasion. Some examples of external variables in this experiment are: responsibility, social pressures, or distractions. ELM states that these factors can be considered, "separately and independently regardless of the levels of the other variables with which it is combined" (Petty and Cacioppo, 1986, p. 198). However even if these variables can be pointed out, in the case of this experiment, there is no way to identify if they caused a difference in the results. Also, becau se significance was shown in one replication more than others, it is clear that involvement is not as clean cut or straightforward a variable as this experiment attempted to manipulate and illustrate as such. It can be multi dimensional and situational. Furth ermore, it was this experiments goal to use the theory of ELM in order to depict that affective message elements can persuade message elaboration, depending on certain situational factors. It suggests a limitation of ELM that the results of this experiment found little to no statistical significance. However, it is more likely that the reason for the relatively null results were due to the experimental limitations. Study Limitations The manipulations imposed by experimental research necessary to isolate certain causal factors are often achieved at the expense of external validity and generalizability. This study is no exception. First, the fliers used in the experiment were artificial and described a limited number of hypothetical causes promoted by non profit organizations. Second, the conditions for stimulus exposure and processing were atypical in several respects: Participants were tested in groups; exposure to fliers were forced and highly compressed into a short period of time ; fliers were project ed on screens in
55 classrooms rather than in a natural environment. All these factors may give rise to a processing mode that is different from what would be expected in real life situations. Also, in the interest of time, efficiency, and convenience, class es containing college students were used as participants in this study. It should be noted that many college students may not have the excessive funds to contribute to organizations of these types, regardless of their involvement in a particular cause or issue. Also, college students have a very different set of priorities than the rest of the population. If extra money may arise, it is more likely to be used for collegiate activities, or for basic needs and wants. According to a representative of 2,000 s urveyed adults in 2003, about half of the Generation Y population (18 26) has donated between 1 and 99 dollars to charitable and/or non profit organizations. Furthermore, fifty seven percent of Gen. Y's elucidated in the survey that they planned to volunte er time and/or donate other items such as clothes, food, etc. However, this number is significantly less than that of the over seventy percent of Generation X and Baby Boomers have donated to charitable and/or non profit organizations over the studied yea rly period (Gardyn, 2003). Therefore, since the experiment was done with college students, the results should be generalized only to people similar to the group of students participated in the study. As listed in Table 5, the participants had a large var iation of hometowns. This factor could have contributed to the study limitations. It is possible that the participants do not specifically have a home concept, and if one is present, it may not be in Tampa, or even Florida. Only 19 of the participants lis ted Tampa as their hometown, equating to 18 percent. So, although Petty and Cacioppo (1965) often used location and hometown as a manipulation of involvement when experiments contained college students; the variation
56 of cities and states may have caused a n unexpected hurdle in this specific experiment. All these limitations should be kept in mind when evaluating the results and their implications. In addition, it should be noted that it is not uncommon for individuals to have pre existing feelings, emotions, o r memories which they link to specific colors. Therefore, the colors in each flier could trigger a difference in each p articipant which further adds to the complexity of the study of color. Pragmatic Implications As strategic communication practitioners, it is vital to understand an organization's r elationship with the public. Obtain thought processing knowledge when elaborating a message is a consistent goal in the field. Non profit organizations do not have the resources that many for profit organizations possess. They may lack the employees, th e publicity, and most of all money. When non profit organizations do decide where to place their funds, knowledge of the public is vital. This study helps to comprehend feelings toward non profits when interacted with various variables. Specifically, this study shows support s that involvement and color/ cause congruency do have an affect, specifically when viewing the results from attitude toward the cause and attitude toward the organization. For non profits, this study is helpful in identifying areas that should be focused on when creating a message to be distributed. Color can have an impact on a message, and feelings towards an organization, but according to this study only in a certain circumstances, and with specific variables. As previously stated, all or ganizations eventually want advertisements and other types of messaging to lead to a positive feeling towards their organization and what is
57 being promoted. Color has often been used as the focus of multiple studies and revealed varying results. Creating a color organizational connection is difficult, especially for non profit organizations. The results of this study suggest that non profit organizations should try to create color congruency with their organization and it's initiatives, as lower color/cau se congruency displayed lower means equaling to lower on the attitudinal scale. It also should be noted that because of limited resources of non profit organizations, targeting demographics and pinpointing where to put focus is always a big decision. T hrough this study, we can see that college students may not have non profit organizations in their list of priorities. It may be more beneficial for non profits to focus on higher age demographic. Theoretical Implications This experiment married involveme nt to congruency, and in earlier sections identified focal points and assumptions based on those theories. The results further congruency theory, which was chosen for this study because of its ability to predict the conditions in which color may have stro nger or weaker influence: A specific color may have greater (weaker) influence when its perceived meaning is congruent (incongruent) with other elements of the message. Attitude toward the organization and attitude toward the cause show significance and scored means coinciding with the main ideas of congruency theory. As stated, increased color/cause congruency equated to higher attitude toward the organization and the cause. Congruency theory can be helpful in practice when for profit and non profit o rganizations are creating messaging and a brand identity for themselves. However, when discussing other variables in the experiment, such as behavioral intention and
58 attitude toward the flier, congruence may not be very significant, which leads to a possi ble limitation of congruency theory. As previously stated, attitude toward an organization cannot always coincide with intention and action. It is easier to hold an attitude than to act on it or create reality for the attitude. Also, attitudes are rarel y created totally by an individual, extraneous factors play a part in the creation of attitudes, which may be the reason the results are inconsistent. This being said, congruency may not be enough of a factor in this experiment to create significance in a ttitudinal change in the variables of behavioral intention and attitude toward the flier. As for the theory of ELM, elaboration is a very complicated process, which is consistently studied through various fields. Petty and Cacioppo (1986) explain that in dividuals learn right from wrong by their surroundings including friends, family, relationships, and media influence. This theory can help to understand why the means were extremely high when measuring attitude toward the cause and the organization. It is socially expectable and encouraged to feel that the environment is important and an issue that should take precedence. However, behavioral intention's means were lower, which puts forth the idea that individuals may think the cause is important, but not enough to use their time volunteering, or their money for assistance. This furthers the idea of ELM by explaining that attitudes may be altered by social pressures, but not enough to take action. Also, in the majority of studies that include ELM and colo r, color is used as a peripheral cue and can sway opinions or attitudes when involvement is low. As reviewed in the previous section, the means illustrated in this study explain that when involvement is low, color has more of an affect on attitudes and be havioral intention. This experiment
5 9 is very limited and exclusive in its findings, but the theory of ELM held strong throughout the majority of results. Conclusion No other study has attempted to create solid results out of such objective variables suc h as attitudes and behavior, while taking on the challenge of establishing interaction between these variables. That makes this study distinctive and unique. This study did reveal statistically significant results that can be added to the field of social science. In advertising, involvement is often taken into account, but as the results of this study show, its interaction with other variables such as color should be considered. Does the color of the advertisement match the product and the organization? Questions such as this should continue to be asked and studied. Also, non profit organizations can use this information when they are creating collateral items, and in other communications with their audience. Because it is common that non profits have f ewer resources such as money, time, and manpower; they must ensure that their communications are reaching audiences and making a difference. This has become increasingly difficult with the amount of constant stimuli that individuals are exposed to on a d aily basis from media outlets such as the internet, television, radio, and the extensive use of outdoor advertising. Although these specific results are limited to the participants of this study, it is indicated that involvement and congruency can have an affect on the audience, and influence their feelings toward their cause and towards the organization. Non profits should consider involvement and congruency in order to create a positive connection with their audience.
60 These results add to the growing cu riosity of color in advertisements, and its affects. It also opens new doors for involvement, the congruency of color and cause, as well as the underlying interaction of these variables. The attitudes and intentions of consumers are very complicated and mu lti dimensional. Once consumer patterns are uncovered, non profit communication can be improved to be more efficient. This research attempted to fill a gap in social science and non profit organizational research, and yielded interesting findings a progr ess. However, further research must be completed in order to create a coherent understanding of the depth of these variables.
61 References Ajzen, I. & Fishbein, M. (1977). Attitude behavior relations: A theoretical analysis and revie w of empirical research. Psychological Bulletin 84(5), 888 918 American Chemistry Council (2008 ). Info sheet. Retrieved from http://www.americanchemistry.com/ s_plastics/sec_content.asp?CID=1102&DID=5007 Anonymous. (2002). Color: The silent language. Paint & Coatings Industry 18(9), 78 80 Babin, B. Hardesty, D., & Suter, T. (2003). Color and shopping intentions: The intervening effect of price fairness and perceived affect. Journal of Business Research 56(7), 541 551 Bal, M. (1991). Reading rembrandt: Beyond the word image oppositions Cambridge, MA: Cambridge. Batra, R., Myers, J., & Aaker, D. (1996). Advertising management New York: Prentice Hall. Birr en, F. (1961). Color psychology and color therapy Secaucus, NJ: Citadel. Bloch, P. H. (1995). Seeking the ideal form: Product design and consumer response. Journal of Marketing 59(3), 16 29 Boyatzis, C. J. & Varghese (1994). Children's emotional associ ations with colors. Journal of Genetic Psychology 155 (1), 77 85
62 Campbell, M. C. & Goodstein, R. C. (2001). The Moderating effect of perceived risk on consumers' evaluations of product incongruity: Preference for the norm. Journal of Consumer Research 28 (3), 439 449 Center for Disease Control (2009). U.S. obesity trends Retrieved October 2, 2009, from Center for Disease Control website: http://www.cdc.gov/obesity/data/trends.html#Sta te Cerbus, G. & Nichols, R. (1963). Personality variables and response to color. Psychological Bulletin 60(6), 566 575 Clark, E. V., & Clark, H. (1978). Universals, relativity, and language processing In J. H. Greenberg (Ed.), Universals of human lan guage, Vol. 1: Method and theory. Stanford, CA: Stanford University Press. Dee, J. & Taylor, L. (2002). Beginners guide to color therapy New York: Sterling Publishing. Fraser, T., Banks, A. (2004) Designer's color manual: The complete guide to color theo ry and application. San Francisco: Chronicle Books. Gambhir, R. (2009). Choose your colours wisely. Canadian Jeweler 130 (2), 38 40 Gardner, M. P., Mitchell, A. A., & Russo, J. E. (1985). Low involvement strategies for processing advertisements. Journal of Advertising 14(2), 4 12 Gardyn, R. (2003). Not giving up. American Demographics : http://findarticles.com/p/articles/mi_m4021/is_8_25/ai_108538952/. Gassenheimer J. B., Davis C. J. & Dahlstrom, R. ( 1998 ), The effects of incongruency: Is dependent what we want to be? Journal of Retailing, 74 (Summer), 247 271 Gorn, G., Chattopadhyay, A., Sengupta, J. & Tripathi, S. (2004). Waiting for the web: How screen color affects time perception. Journal of Marketing Research, 41, 215 225
63 Gorn, G., Chattopa dhyay, A., Yi, T., & Dahl, D. (1997). Effects of color as an executional cue in advertising: They're in the shade. Management Science 43(10), 1387 1400 Grossman, R. P. & Wisenblit, J. Z. (1999). What we know about consumers' color choices Journal of Ma rketing Practice 5(3), 78 88 Jasper, C. (2005). Power color: Master color concepts for all media New York: Wastson Guptill Publications. Laczniak, R. N., Muehling, D. D. & Grossbart, S. (1989). Manipulating message involvement in advertising research. Journal of Advertising 18(2), 28 38 MacKenzie, S. B., Lutz, R. J., & Belch, G. E. (1986). The role of attitude toward the ad as a mediator of advertising effectiveness: A test of competing explanations. Journal of Marketing Research 23(2), 130 143 Martin, L. L., & Clore, G. L. (Eds). (2001). Theories of mood and cognition: A user's handbook Mahwah, NJ: Lawrence Erlbaum Associates. Mehrabian, A. & Russell, J.A. (1974). An approach to environmental psychology Cambridge, MA: MIT Press. Meyers Levy J. & Peracchio, L. A. (1995). Understanding the effects of color: How the correspondence between available and required resources affects attitudes. Journal of Consumer Research 22(2), 121 138 Miller, E. & Kahn, B. (2005). Shades of Meaning. Journal o f Consumer Research 32(1), 86 92 Miniard, P. W., Sirdeshmukh, D. & Innis, D. E. (1992). Peripheral persuasion and brand choice. Journal of Consumer Research 19 (2), 226 239
64 National Center for Charitable Statistics (2008). Quick facts about non profit s Retrieved October 1, 2009 from National Center for Charitable Statistics website: http://nccs.urban.org/statistics/quickfacts.cfm Norris, S. (2001). Secrets of color healing New York: D K Publishing. Osgood, C. E., & Tannenbaum, P. H. (1955). The principle of congruity in the prediction of attitude change Psychological Review 62(1), 42 55 Petty, R. E. & Cacioppo, J. T. (1986). Communication and persuasion: Central and peripheral routes to attitude change New York, NY. Springer Verlag. Petty, R., Cacioppo, J.T., & Schumann, D. (1983). Central and peripheral routes to advertising effectiveness: The moderating role of involvement. Journal of Consumer Research, 10, 132 146 Poynter Inst itute (2004). Article level page design: What matters inside. Eyetrack 3: Online News Consumer Behavior in the Age of Multimedia: poynterextra.org/eyetrack2004/articlepages.htm. Poynter Institute (2004). Recall of information presented in text vs. multim edia format. Eyetrack 3: Online News Consumer Behavior in the Age of Multimedia: poynterextra.org/eyetrack2004/multimediarecall.htm. Poynter Institute (2004). Viewing patterns for homepages. Eyetrack 3: Online News Consumer Behavior in the Age of Multimed ia: poynterextra.org/eyetrack2004/viewing.htm. Poynter Institute (2007). Eytracking the News: A study of print and online reading. Poynter Institute online www.eyetrack.poynter.org. Rorschach, H. (1942). Psychodiagnostics New York: Grune & Stratton.
65 Ros e, G. (2007). Visual methodologies (2 nd ed.). Los Angeles, CA: Sage. Rubin, R. B., Palmgreen, P., & Sypher, H. E. (Eds.). (1994). Communication Research Measures: A sourcebook. New York: Guilford. Sherif, M. & Hovland, C.I. (1961). Social judgment: Assimi lation and contrast effects in communication and attitude change New Haven, CT: Yale University Press. Shindler, P. (1986). Color and contrast in magazine advertising. Psychology & Marketing 3(2), 69 78 Simonson, A. & Schmitt, B. H. (1997). Marketing a esthetics : The strategic management of brands, identity and image New York: The Free Press. Siple, P. & Springer, R. M. (1983). Memory and preference for the colors or objects. Perception and Psychophysics 34(4), 363 370 Skorinko, J. L., Kemmer, S., He bl, M. & Lane, D. M. (2006). A rose by any other name: Color naming influences on decision making. Psychology & Marketing 23(12), 975 993 Valdez, P., Mehrabian, A. (1994). Effects of color on emotions. Journal of Experimental Psychology 123(4), 394 409 Zaichkowsky, J. L. (1985). Measuring the involvement construct. Journal of Consumer Research 12 (December), 341 352.
66 Appendix A Stimulus Materials
67 High Involvement/Low Congruency Group A
68 High Involvement/Low Congruency Group B
69 High Involvement/ High Congruency Group A
70 High Involvement/ High Congruency Group B
71 Low Involvement/ High Congruency Group A
72 Low Involvement/ High Congruency Group B
73 Low Invo lvement/ Low Congruency Group A
74 Low Involvement/ Low Congruency Group B
75 Appendi x B Questionnaire
76 Dear Participants, This research investigates environmental organizational communication. Please read the infor med consent statement below, and indicate your willingness to participate in this research by signing in the space provided. ______________________________________ Print Name __________________________ ___________ __ Signature ___________________________ Date Informed consent statement: This research is being conducted under the superv ision of Dr. Scott Liu, USF School of Mass Communications, 4202 East Fowler Ave, CIS1040, Tampa, FL 33620; (813) 974 6797. Your responses will remain confidential to the extent provided by law. You do not have to answer any questions you do not wish to ans wer, and you have the right to withdraw consent at any time without consequence. There are no anticipated risks associated with your participation in this research and you will receive no compensation for your participation. If you decide not to participat e in this study, your course grade will not be affected in any way. If you have any questions concerning the procedures used in this study, you may contact me at the e mail address email@example.com. Questions or concerns about your rights as a particip ant can be directed to the University of South Florida Institutional Review Board, 12901 Bruce B. Downs Blvd., MDC35, Tampa, FL 33612.
77 Please take 5 minutes to ra te your level of agreement with the statements about the organization represented by each flier, on the scale below by putting "X" in the appropriate section of the scale. Example : USF Parking is a large problem at the University. Strongly Agree :__ X __:___ __:_____:_____:_____:____:_____: Strongly Disagree Please complete the following measures to rate the organization. I think this organization's initiatives are Unfavorable _____:_____:_____:_____:_____:_____:_____ Favorable Negative _____:_____:_ ____:_____:_____:_____:_____ Positive Unimportant _____:_____:_____:_____:_____:_____:_____ Important Good _____:_____:_____:_____:_____:_____:_____ Bad To me, this organization is Good _____:_____:_____:_____:_____:_____:_____ Bad Favorable _____:_____:_____:_____:_____:_____:_____Unfavorable Unlikable _____:_____:_____:_____:_____:_____:_____ Likeable Positive _____:_____:_____:_____:_____:_____:_____ Negative Unimportant _____:_____:_____:_____:_____:_____:_____ Imp ortant Now please use the following measures to rate the flier. This flier is Good _____:_____:_____:_____:_____:_____:_____ Bad Favorable _____:_____:_____:_____:_____:_____:_____Unfavorable Positive _____:_____:_____:_____:_____:_____:_ ____ Negative Unlikable _____:_____:_____:_____:_____:_____:_____ Likeable Please use the following measures to rate your intention to get involved in this organization. I intend to obtain more information on this organization in the next month. Unlikely _____:_____:_____:_____:_____:_____:_____ Likely I intend to donate to this organization. Unlikely _____:_____:_____:_____:_____:_____:_____ Likely
78 I intend to tell others about this organization Unlikely _____:_____:_____:_ ____:_____:_____:_____ Likely I intend to volunteer for this organization. Unlikely _____:_____:_____:_____:_____:_____:_____ Likely Brand familiarity: I know of this organization. Strongly Disagree _____:_____:_____:_____:_____:_____:_____ Str ongly Agree Please rate the flier: The language/wording in this flier is Good _____:_____:_____:_____:_____:_____:_____ Bad Favorable _____:_____:_____:_____:_____:_____:_____Unfavorable Positive _____:_____:_____ :_____:_____:_____:_____ Negative The color in this flier is Good _____:_____:_____:_____:_____:_____:_____ Bad Favorable _____:_____:_____:_____:_____:_____:_____Unfavorable Positive _____:_____:_____: _____:_____:_____:_____ Negative The design/ graphics in this flier are Good _____:_____:_____:_____:_____:_____:_____ Bad Favorable _____:_____:_____:_____:_____:_____:_____Unfavorable Positive _____:_____: _____:_____:_____:_____:_____ Negative To me, this organization is Important : _____:_____:_____:_____:_____:_____:_____: Unimportant Of no concern :_____:_____:_____:_____:_____:_____:_____: Of concern to me Irrelevant :_____:_____:_____:_____:_____:_____:_____: Relevant Means a lot to me :_____:____ _:_____:_____:_____:_____:_____ : Means nothing to me Useless :_____:_____:_____:_____:_____:_____:_____: Useful Valuable :_____:_____:_____:_____:_____:_____:_____: Worthless Trivial :_____:_____:_____:_____:_____:_____:_____: Fundamental Beneficial :_____:_____:_____:_____:_____:_____:_____: Not beneficial Matters to me :_____:_____:_____:_____:_____:_____:_____: Doesn't matter
79 Uninterested :_____:_____:_____:_____:_____:_____:_____: Interested Significant :_____:_____:_____:_____:_____:_____:_____: Insignificant Boring :_____:_____:_____:_____:_____:_____:_____: Interesting Unexciting :_____:_____:_____:_____:_____:_____:_____: Exciting Appealing :_____:_____:_____:_____:_____:_____:_____: Unappealing Munda ne :_____:_____:_____:_____:_____:_____:_____: Fascinating Not needed :_____:_____:_____:_____:_____:_____:_____: Needed Good :_____:_____:_____:_____:_____:_____:_____: Bad Creative :_____:_____:_____:_____:_____:_ ____:_____: Uncreative Informative :_____:_____:_____:_____:_____:_____:_____: Uninformative Likeable :_____:_____:_____:_____:_____:_____:_____: Unlikable The use of color in this flier is Good :_____:_____:_____:_____:_____ :_____:_____: Bad Unfitting :_____:_____:_____:_____:_____:_____:_____: Fitting Congruent :_____:_____:_____:_____:_____:_____:_____: Incongruent Mismatch : ____ :_____:_____:_____:_____:_____:_____: Match Creativ e :_____:_____:_____:_____:_____:_____:_____: Uncreative Correct :_____:_____:_____:_____:_____:_____:_____: Incorrect Unsuitable :_____:_____:_____:_____:_____:_____:_____: Suitable I enjoyed viewing this flier. Agree :_____:_____:____ _:_____:_____:_____:_____: Disagree Demographic variables Sex _____ Male _____ Female Age ________ Academic Level: _____ Freshmen _____ Sophomore _____ Junior
80 _____ Senior Ethnicity: _______ Caucasian _______ African American _______ Hispanic _____ __ Asian _______ Other____________ Hometown (City, State, Country) _______________________________