From the brand logo to brand associations and the corporate identity: visual and identity-based logo associations in a university merger

Brand logos are a fundamental part of the corporate visual identity, and their reception has been vigorously researched. The focus has been on the visual traits of the logo and their effect on the reception process, whereas little attention has been paid to how the logo becomes part of the brand. This article narrows this research gap in investigating how a new logo is evaluated, how the perception evolves, and what underlying dimensions emerge from the reception process. We adopted a longitudinal free-association approach and followed the qualitative and quantitative changes in logo associations among first-year students at Aalto University as it was going through a merger accompanied with a radical visual-identity redesign. We show how the new logo faced initial resistance before it became a source of positive brand associations, and how it became anchored in the university´s corporate identity. We argue that logo evaluations span three dimensions: they may be congruent or incongruent with the disposition of the individual toward the change: they may be congruent or incongruent with the visual preferences of the individual; and they may be based on the visuals of the logo or on its identity-expressing capabilities.


Introduction
What defines the image of an educational institution? Discussion, evaluation, change. A school is constantly re-evaluating, discussing and adjusting the way it organizes itself and its activities. The image of a school is thus undefinable-it is in constant change. This is why the logo should not provide a static predefined image, but a blank canvas, which will in time be attached with meaning-an identity created and experienced by the students, researchers and employees.
The above quotation is from graphic designer Rasmus Snabb, when he was in the process of designing a logo for Aalto University in 2009. The starting point was unconventional. Literature on logos emphasizes the role of organizations as their creators, whereas customers are mere recipients (Kim and Lim 2019). In the case of Aalto, the purpose was to create a symbol to communicate change and a fresh start. At the same time, the logo was meant to be a tool for identity building-a blank canvas on which the community could reflect the identity of Aalto University. More and more universities are facing such challenges. Competition is intensifying as the field becomes increasingly marketized (Wedlin 2008), but as organizations valuing diversity and academic freedom they constitute a potentially challenging environment for brand management (Melewar et al. 2018). Fragmented identities may make it difficult to create a uniform brand (Waeraas and Solbakk 2009). In the following, we focus on how resistance to a new logo turns into acceptance, and how the logo becomes a source of brand association and symbol of a shared identity.
A logo is a graphic design, which companies use to identify themselves and their products (Henderson and Cote 1998). They are known to contribute to recognizability (Balmer and Gray 2000), brand equity (Abratt and Kleyn 2012) and a sense of familiarity (Foroudi et al. 2014). There has also been research on the logo-creation process and consumer evaluation (Henderson and Cote 1998;Buttle and Westoby 2006;van Der Lans et al. 2009), as well as on the psychological mechanisms behind their reception (Park et al. 2013; van Grinsven and Das 2015;Miceli et al. 2014). A logo could thus be described as a brand-management tool used in a carefully planned process to create brand equity, customer commitment and competitive distinctiveness. As a visual presentation of a corporation, the logo has also been viewed as the root of corporate identity (Foroudi et al. 2017;van Riel and Balmer 1997), and in addition to their functional benefits and aesthetic appeal, they have been found to facilitate identity expression (Park et al. 2013). A logo may display desired identities, hence congruence of the self and the brand is crucial (Japutra et al. 2016). According to this line of thinking, logos convey associations between the brand and the self-from the organization they represent to the identity of the consumer.
Although there has been vigorous and ambitious research on logos, the majority of studies use quantitative metrics with relatively short time spans, and target groups with no meaningful relationship to the brand in question. We know much of what happens during the reception phase and about the effect of different design elements (see e.g., Henderson and Cote 1998;Janiszewski and Meyvis 2001; van Grinsven and Das 2016;Miceli et al. 2014;Bresciani and Del Ponte 2017), as well as about the consequences of using a logo (see e.g., Park et al. 2013;Abratt and Kleyn 2012), but little is known about how the process unfolds. We therefore take a complementary approach in this article, considering logo associations from a longitudinal perspective. Our investigation covers: (1) how the new logo of the university is being evaluated and what is associated with it, (2) how perceptions and associations evolved following the founding of the university and (3) what underlying dimensions emerge from the logo associations.
Our research method originates from the theory of social representations (Moscovici 1973;Abric 2001). Using both qualitative and quantitative analysis, we show how the university's new logo faced initial resistance before it became a vessel for positive brand associations, and how it came to be anchored in Aalto's corporate identity. The approach resembles that of Van Riel and van den Ban (2001), who categorized logo associations as either intrinsic (focusing on visual elements directly resulting from confrontation with the logo) or extrinsic (originating from associations with the organization behind it). They refer to the former as graphical and the latter as referential associations. Building on this work, we argue that logo evaluations span three dimensions: they may be congruent or incongruent with the disposition of the individual toward the change: they may be congruent or incongruent with visual preferences of the individual; and they may be based on the visuals of the logo or on its identity-expressing capabilities.
Our approach has several benefits. First, it allows us to tap into a real-life setting in which the participants are deeply invested in the brand that is being revamped, unlike most modern research on logos that relies on controlled laboratory experiments or traditional surveys (Kim and Lim 2019). Second, our longitudinal setting covers an exceptionally long period of time-seven years-which enables us to follow changes as they transpired after the founding of the university. Third, our focus on the qualitative associations evoked by the logo enhances understanding of the reception process and the underlying dimensions from the recipients' perspective. In essence, the method enables us to monitor how the link between the symbol and the organization evolved, and how the logo became a synonym for the organization invoking associations related to the brand. Fourth, applying the theory of social representations in our analysis gave us the opportunity to interpret many of the phenomena we monitored in the logo-reception process.

Brand logos and corporate visual identity
Logos are a fundamental part of the visual identity of brands and corporations. How they are received has attracted attention in the literature during recent decades, which is not surprising given the considerable resources corporations devote to their design. The focus in the research has been on the reception process, and more lately also on the benefits of logos as well as their meaning and identity-based elements.
According to Henderson and Cote (1998), a logo should evoke recognition, affect and meaning, and reflect the organization it represents. It should also arouse positive reactions and carry the same meaning across people and contexts. Building on this, van Grinsven and Das (2016) found that increased exposure to the logo can lead to stronger brand recognition and positive attitudes, especially if the logo is complex. According to Miceli et al. (2014), however, there is a difference between visual and conceptual complexity. Whereas the former is evaluated more positively on the first exposure, but the evaluation turns more negative thereafter, the opposite is the case with conceptually complex logos, which may evoke multiple meanings.
Logos have purposes other than being just symbolic reminders of a particular corporation or product: they may have symbolic and functional benefits (Park et al. 2013) or convey a corporate identity (Foroudi et al. 2014). Above all, the logo should be suited to the organization (Foroudi et al. 2014), whose image may end up "colouring" the logo in the long run (van Riel and van den Ban 2001). In other words, there should be congruence between the organization and its symbol, because the two may have a reciprocal relationship.
The link between the logo and the identity of the organization behind it is especially evident in strongly value-driven organizations such as universities. A new visual image may threaten the sense of tradition in a university, which may provoke resistance (Idris and Whitfield 2014). It is nevertheless possible to mitigate the resistance by means of empathetic communication (Walsh et al. 2019) and focusing on the congruence between the logo and the organization (Japutra et al. 2016). It has been shown that incongruence may lead to resistance, especially if the new logo comes as a surprise (Grobert et al. 2016).
There has been a vast amount of research on identity, which is a widely used term. In the context of logos, it often refers to corporate visual identity, meaning the visible part of the corporate identity (see e.g., Melewar and Saunders 1998;Foroudi et al. 2014). Corporate identity is the distinctive public image a corporation communicates and the shared meanings that the corporate entity is understood to have (Cornelissen et al. 2007). Balmer (2008) makes a distinction between the identity of the corporation and identification from the corporation: in this sense corporate visual identity is the latter, the focus being on the symbolism and how the corporation wishes to be seen. The identity of a corporation, on the other hand, concerns its distinctive traits. This article follows the same line of reasoning, according to which a logo is a fundamental component of corporate visual identity. By analyzing responses to the visual manifestation of the corporate identity, we shed light on its distinctive traits. We see the logo as a part of a brand's associative network, which evokes associations related to the brand, the corporation and the visual image of the logo.

Brand associations and social representations
Brand associations are the informational nodes in the memory of consumers that contain the meaning of the brand to them. They vary in favourability, strength and uniqueness and may refer to attributes or benefits of the product or service, or to overall evaluations of the brand such as attitudes toward it (Keller 1993). They could also be viewed as either owned or shared: owned associations are actively communicated by the organization whereas shared associations are perceived and produced by consumers (Mirzaei et al. 2016). In this sense, there is active communication between the owner-the organization-and the consumer, during which the meaning of the brand is negotiated. Another major factor is the congruity between owned and shared associations (Crawford Camiciottoli et al. 2014;Mirzaei et al. 2016), in other words the match or mismatch between associations that are actively communicated by the organization and shared by consumers.
Social representations, on the other hand, are "fields of knowledge" or forms of common sense (Moscovici 1973;Moscovici and Marková 1998). They could be considered a set of organized judgements, attitudes and information concerning a social object with a hierarchical structure, which are shared by a social group (Abric 2001). From a psychological perspective, a brand is a social representation (Schmitt 2011) that enables social groups to communicate, interact and organize around social objects such as products and services. In the context of brand research, the theory of social representations has been applied to consumer perceptions (Roininen et al. 2006), brand positioning (Lebrun et al. 2013) and brand associations (Michel and Donthu 2014). Social representations and brand associations are complementary: both purport to organize the contents of social objects to enable communication and the formation of a shared vision.
This study applies a free-association approach adapted from the social-representation tradition, the aim being to analyze associations evoked by a symbol of the brandthe logo. In our view, once the logo is well established, it provides a lens through which one can view the content of the brand and the associations related to the organization in question. Establishing this link between the logo and the brand may take time, however.

Context and research questions
This article focuses on how the Aalto University logo (see Fig. 1) was received. The university was founded in 2010 following a merger of Helsinki University of Technology (HUT), Helsinki School of Economics (HSE) and the University of Art and Design Helsinki (UIAH). HUT was the largest of the merging universities, and all three were top The image was used as a stimulus in the survey institutions in their fields in Finland. The merger was one of the flagship projects of the Finnish higher-education reform (Aula and Tienari 2011) and it marked a significant turn toward marketization. The merging universities were all state-owned, whereas Aalto University is controlled by a privately owned fund and a board including representatives from outside academia. Additional consequences of marketization included the strong financial support from the Finnish government and industry (Aspara et al. 2014) and the allocation of attention and resources to branding.
Aalto University is Finland's most renowned highereducation institution in the fields of business, technology and design and has attracted research interest in various areas such as reputation building (Aula and Tienari 2011), rebranding (Aspara et al. 2014), strategy (Tienari et al. 2016) and organizational identification (Edwards et al. 2017). During its early years, the university struggled to establish its reputation (Aula and Tienari 2011) and to adopt a service-dominant branding logic (Aspara et al. 2014). Edwards et al. (2017) studied different integration trajectories and found that employee adaptation to the new organization was also slow, especially among employees from the former Helsinki School of Economics and the University of Art and Design. According to Tienari et al. (2016), the fact that Aalto University positioned itself as a "world class" university caused controversy among employees. According to Ripoll-Soler and de-Miguel-Molina (2014,2019), however, the merger could be considered a relative success in terms of what are considered crucial factors in university mergers, and has brought about improvements in terms of global ranking. The Aalto merger was horizontal and complementary in academic profile, the size of the merged institution was optimal in terms of student numbers, and the original aim was to create something more than the mere sum of its parts in response to contextual demands in its environment.
To better understand the context of the logo-design process at Aalto University, we interviewed selected key informants involved in the creation of its visual identity. The thinking was that the new university would need a logo that would communicate something totally new. The name of the university was decided through a competition, and the same process was used for the logo. Hence, a logo-design competition was launched in 2009, which attracted 177 proposals from students and alumni of the former universities. The winning entry-designed by a student at UIAH-was called "Invitation" containing the text "Aalto-yliopisto" (Aalto University, in English) accompanied by a question mark, an exclamation mark and quotation marks. The final logo was modified from this design. There are nine variations, each containing only the letter A accompanied with either a question mark, an exclamation mark or quotation marks in different colours (see Fig. 1).
From the university's perspective, the logo was intended to communicate that the multidisciplinary Aalto University was something different and completely new. The university community was involved in the creation of the visual identity, and the logo was meant to be neutral so that the organizational entity could give it meaning-or "colour" in the words of van Riel and van den Ban (2001)-in the years to come. In terms of brand management and design, the case has many interesting angles. First, the logo is unorthodox in the traditional world of brand management. Instead of a single symbol evoking the same meaning, there are nine variations that are to be used randomly. Second, referring to Miceli et al. (2014), the logo is visually simple but conceptually complex, thereby allowing room for in-depth analysis of how this type of logo can be adopted as a symbol of an existing social group. Third, much of the literature on logos emphasizes the role of the organization as the creator and of the marketing department as the identity builder (e.g., Kim and Lim 2019). A complementary view would be to consider the organization a facilitator of the branding process (Brodie et al. 2016), which is exactly what the Aalto case exemplifies.
The focus of our study is on the reception of the logo among students at Aalto University School of Business and its predecessor Helsinki School of Economics. The school is an interesting target because it is one of the smaller universities comprising Aalto, and in this sense it was not in a dominant position in the merger. At the same time, the new logo was designed by a UIAH student, so in this sense it could be seen as an "art and design" project coming from outside the School of Business.
From these premises, we developed our research questions for this article: RQ1: How is the logo evaluated and what associations are linked to it? RQ2: How did the perception and the associations evolve following the founding of the university? RQ3: What underlying dimensions do the associations reveal?

Methods
We used a free association approach to tap into the respondents' impressions of the new logo in this real rebranding situation. The method has been used to study the structure of social representations in various contexts such as brand positioning (Lebrun et al. 2013), consumer perceptions (Roininen et al. 2006;Mäkiniemi et al. 2011) and marketing (Penz 2006), and has proved efficient and practical as a way of collecting data without imposing a predefined structure or discourse on respondents.
The data were collected via an online survey, conducted in Finnish, in which the participants were asked to list up to five associations that the Aalto University logo brought to mind. An image with all the logo variations was provided as a stimulus (see Fig. 1). The respondents were then asked to rate the tone of the associations as positive, negative or neutral. Following the free-association task there were further questions about the merger, such as how the respondents felt about the name and the brand of the new university. Background information on aspects such as age and gender was collected at the end of the survey. To ensure data-anonymization, no identifying information was collected from the participants in the surveys.
The survey was distributed at three different time points. First-year students at Helsinki School of Economics were approached in November 2009, just a few months before the merger officially took place (January 2010) and 6 months after the new logo had been released (May 2009). A second survey was sent out to first-year students at Aalto University School of Business in November 2011, and a third survey to first-and second-year students in April 2016. Secondyear students were included in this third target group to equalize the sample sizes, given that between 2011 and 2016 it became university policy to allow students to opt out of receiving invitations such as this. The total number of respondents was 162 (86 female), yielding a total of 792 associations. Of these, 59 responded in 2009, 52 in 2011 and 51 in 2016. The response rates were 16.8, 16.9 and 17.5 per cent, respectively. The average age of the respondents was 23 (median age 22).
It should be pointed out that the three separate groups of respondents (2009, 2011 and 2016) differed in terms of how they had been exposed to the old and the new logos. Those who started their studies in 2009 had applied to the old university and had studied there for 3 months; thus, the logo redesign represented a clear change from the old to the new. Those who started in 2011 began their studies at Aalto University with its new logo, but were also somewhat exposed to the old logo because it was actively used in marketing until 2010. Students starting their studies in 2016 were only 16-17 years old when Aalto University was founded and were thus only exposed to the new logo, which at this point had been used exclusively in all marketing and communication for six years (2010)(2011)(2012)(2013)(2014)(2015)(2016). Given that Aalto is the most well-known Finnish university in the field of business, it is safe to say that all of the respondents to the 2016 survey had seen the logo beforehand, and that most-if not all-of them had been exposed to the marketing efforts of the university.

Empirical analysis
The diversity of the collected qualitative data reflected the controversial nature of the subject. The first step in the analysis was to lemmatize the associations, in other words to identify the root words so they could be analyzed as single items. We then combined synonyms in classes reflecting the most representative word in the class and created a further qualitative categorization reflecting the semantics and tone of the associations. In unclear cases, we considered all the associations provided by the respondent concerned in an attempt to establish the meaning. This was especially helpful when the tone was sarcastic, such as "huippuyliopisto" ("top university" in English), which can be used in its literal meaning or sarcastically to ridicule the university.
The final classification scheme comprised 21 categories covering 89 per cent of all the associations. The remaining 11 per cent did not fit any of the classes and were too rare to justify a separate class. A threshold of 10 total mentions was used to focus on the associations that were shared by the group. The first author of this article devised the categorization scheme, which was tested by a second researcher. Contradictory cases were discussed and agreed in collaboration. Table 1 shows the classification scheme including the number of associations per year.
Most of the classes were clear and easy to define, most of the associations being identical or close synonyms. A good example of such a class is "simple", in which almost twothirds of the associations are versions of the root word. Other homogenous association classes included "ugly", "colourful", "unclear" and "diverse". At the same time, many classes required more consideration, "inappropriate" and "academic" being good examples. The "inappropriate" class contains a diverse range of associations questioning the suitability of the logo for a university, varying from "unacademic" to "incompetent" and referring to various non-academic organizations (e.g., circus, pharmacy). The "academic" class, on other hand, contains a wide variety of associations related to academic life, such as "learning", "science", "research", "open" and "critique" in a positive sense. Among the other heterogenous classes were "thought", "questionable", "innovation" and "influence".
We noted during the classification process that many of the classes had counterparts, including contradictory and opposing pairs such as clear-unclear, boring-interesting and simple-diverse. Some pairs were quite simple and easy to recognize-such as ugly (aesthetically displeasing) and stylish (aesthetically pleasing), whereas others were conceptually fairly complex, forming semantic bundles rather than simple pairs. The classes "inappropriate" and "childish" serve as a good example of a conceptually complex bundle, the associations questioning the logo as a symbol of an academic institution and the counterparts deriving from the classes "academic", "thought" and even "innovation".
We conducted Multiple Correspondence Analysis (MCA) and K-Means cluster analysis in combination to identify the underlying response dimensions and thereby better understand the structure of the data. MCA is a generalization of principal component analysis (PCA) for nominal variables; hence, it can be used to analyze patterns of relationships among several categorical dependent variables (Abdi and Valentin 2007;Vehkalahti and Everitt 2019). In this sense, the method is related to factor analysis, which is often used in experimental aesthetics to discover underlying dimensions (see e.g., Henderson and Cote 1998). K-Means, on the other hand, is a traditional clustering method, which is used to discover natural groupings in a given data set (Jain 2010). The combination is widely used with free association data to analyze social representations.
The post-classification analysis covered the whole data set, and the cluster analysis revealed four separate respondent groups. We named the clusters Visual incongruence (n = 30), Identity incongruence (n = 49), Visual congruence (n = 36) and Identity congruence (n = 47), reflecting the responses in each one and the evaluation dimensions in the final MCA model (Table 2). For each cluster, we calculated an index of polarity, which indicates the relative sentiment of the cluster varying between 1 (extremely positive) and −1 (extremely negative). The clusters are discussed in more detail in the next section. Finally, Multiple Correspondence Analysis (MCA) allowed us to identify and visualize the relations among the associations, clusters and time points.

Results
The analysis revealed four respondent clusters and our final MCA model identified three dimensions explaining a total of 35.1 per cent of the variance. The three dimensions differentiated responses along continuums ranging from i) incongruence to congruence, ii) visual congruence to visual incongruence and iii) visual to identity-based associations, each dimension explaining 16.5, 10.3 and 8.3 per cent of the variance, respectively. In other words, the respondents were either resistant to or supportive of the logo, they reacted to the visual design either positively or negatively, and their associations related either to the visual aspects of the logo or to the organization and the brand behind it.
Visual incongruence This cluster had 30 members whose associations focused on the visual aspects of the logo, expressing dissatisfaction with the design. The most typical were "boring", "ugly", "unclear" and "stupid", which alone covered over two-thirds of the associations. To a smaller extent the associations of members of the visual-incongruence cluster concerned the simplicity of the logo, as well as its "questionable" nature, in particular the underlying logic. There were also associations linking Aalto University with technology and the arts-in other words the non-business "out-group" aspects: examples included "engineer" and "modern art". The cluster as a whole was highly negative with a polarity index of -0.60.
The cluster was at its peak in 2009 and declined through- Identity incongruence This cluster had 49 members whose associations focused mainly on identity aspects expressing resistance, the most typical being "inappropriate", "childish" and "questionable". The logo was considered not to be 'representative of us', but childish, unacademic and corporate-like. The words "boring" and "simple" (with a negative connotation) were also fairly typical. The cluster was highly negative with a polarity index of −0.69.
Similar to the visual-incongruence cluster, this one was at its peak in 2009, also declining and becoming slightly less negative over the time period. It covered 55.9 per cent of respondents in 2009, 21.2 per cent in 2011 and 9.8 per cent in 2016, the respective indexes of polarity being −0.73, −0.69 and −0.56. The cluster was more representative than the visual-incongruence cluster in 2009, but also declined more rapidly: by 2016 most of the incongruence related to the visual aspects of the logo.
The three most typical associations in the cluster were "inappropriate", "childish" and "questionable". Of these, "inappropriate" and "questionable" are highly heterogenous while "childish" is fairly homogenous, the associations including "kindergarten" and "elementary school" in addition to the word "childish".
Visual congruence This cluster had 36 members whose associations focused on the positive visual aspects of the logo. The most typical associations included "stylish", "Aalto University", "colourful" and "simple" with a positive connotation, but also "questionable" and "strange". The latter two represent the more positive end of their respective classes compared to the same class associations in the visual-and identity-incongruence clusters. The cluster is fairly positive with a polarity index of 0.29.
Whereas Identity congruence This final cluster had 47 members with associations focusing mainly on positive brand associations and identity-related themes. The most representative associations were "academic", "thought", "modern", "influence" and "innovation", but "simple" (with a positive connotation), "interesting" and "diverse" also featured. Most of the classes were heterogenous and ambiguous, such as "academic", "thought" and "influence". The cluster is the most positive in the data set with a polarity index of 0.60.
Unlike the other clusters, this one was at its peak in 2016. It covered only 8.5 per cent of respondents in 2009, rising to 36.2 per cent in 2011 and 55.3 per cent in 2016. The respective indexes of polarity were 0.80, 0.54 and 0.60. It is clear from the rate and index for 2009 that at the beginning there was already a small but extremely positive group of people who took the logo as their own and linked it to positive and multidisciplinary themes such as "technology", "invention", "intelligent" and "evolved".
The reception of the logo The initial response was one of shock. Almost two-thirds of the associations in 2009 were negative (see Table 3), the five most typical being inappropriate, childish, questionable, ugly and boring, reflecting both visual and identity incongruence. These classes alone covered almost half of the associations in the first year. The tone quickly became more positive; however, the rate of negative associations had already halved in 2011, whereas the rate of positive associations had more than doubled. This trend continued in 2016, at which point half of the logo associations were positive and less than a quarter were negative.
The same trend is visible in the development of the clusters throughout the time period. The most typical clusters in 2009 were identity incongruence (55.9%) and visual  The different trend of the visual-congruence cluster is also visible in Fig. 2. Incongruence with the visual and identity aspects of the logo dominate the cohort of 2009, which then changes to congruence on the visual and identity levels. Visual congruence is especially prominent in the first phases of accepting the new logo, as the cluster peaks in 2011. A possible explanation for this is that visual congruence precedes identity congruence-if the positive brand associations are to become dominant, the visual aspects of the logo first have to be accepted. Figure 2 further shows how time relates to the three dimensions: reception moves from incongruence to congruence in a fairly linear fashion (2009 via 2011-2016) along the first dimension. There is no linear trend on the dimensions from visual incongruence to visual congruence or from the visual to the identity aspects, but it is clear that 2011 was slightly more visually oriented than the other years. This also supports the interpretation that visual congruence precedes identity congruence, and that visual acceptance of the logo could be a prerequisite for identitybased associations to emerge.
As a final interesting observation, a comparison of the cluster rates of each year reveals that 2009 and 2016 are almost exact mirror images. Identity incongruence dropped from 55.9 to 9.8 per cent, whereas identity congruence rose from 6.8 to 51.0 per cent. Visual incongruence, on the other hand, dropped from 23.7 to 13.7 per cent whereas visual congruence rose from 13.6 to 25.5 per cent. The year 2011 is the exception, with a more evenly distributed cluster representation.

Discussion
This study enhances understanding of how logos are received, building on a novel and practical research method that is well-suited to the domain. The focus was on associations related to the logo that was introduced in a newly established university following the merging of three higher-education institutions, as part of a major redesign of its visual identity. We used both qualitative and quantitative analysis in combination. According to the findings, logos are evaluated on the basis of resistance to or acceptance of the change, and in terms of whether or not they are considered aesthetically pleasing. Moreover, evaluations are either identity-based, focusing on the organization and its brand, or visuality-based in which case the focus is on the visual aspects. Finally, how the new logo is received moves from the negative to the positive-from resistance to acceptance-on both the visual and the identity dimension. It is typically heavily resisted at first because it is not considered aesthetically pleasing, or representative of 'us', but in time it becomes acclaimed and it is accepted as a common symbol. Eventually, the logo becomes synonymous with the organization, yielding associations related to the brand and the corporate identity.

Theoretical implications
The study offers several interesting theoretical insights into the research on logos. First, reflecting the work of Van Riel and van den Ban (2001), we posit that logo evaluations are targeted either toward the logo or toward the organization behind it. Visual evaluations focus on the logo, whereas those that are identity-based focus on the organization in the form of resistance or identity expression. The logo may also be used to express resistance to and criticism of the change, in that 'it does not represent us'. Alternatively, as a method of identity expression it could even link directly to the core elements of corporate identity. In a sense, it has become a synonym for the organization in that when people look at the logo they see not only the graphical image but also the organization and brand behind it. We conclude, based on the analysis, that a logo may serve as an analytical lens through which to scrutinize the associative network of the brand. Aalto, for example, has been depicted as interdisciplinary and as having "innovation at its heart" (Aula and Tienari 2011). It could be argued that some of the associations from 2016, such as "innovation", "diverse" and "academic", are positive brand associations with deep links to the core of the university. At the same time, much of the criticism of the logo in 2009 and 2011 reflected the public discourse regarding the university. Aalto was criticized for being too business oriented and falling for "innovation hype", for example (see e.g., Aspara et al. 2014). Hence, the meaning of the organization was continuously present in the identity clusters, which also dominated the reception process throughout the time period.
Moreover, although the identity-based clusters were dominant throughout the period, there were considerable differences between the time points. For example, the identity incongruence and congruence clusters were dominant in 2009 and 2016, but there was a much more even distribution in 2011. It seems that 2009 was marked by resistance and 2016 by acceptance, whereas 2011 was the most divided and visually oriented. It seems that the process of accepting the brand and the new identity was still underway in 2011, which is evident from the even distribution of the clusters as well as from the content of the associations. A substantial part of the positive shift in associations from 2009 to 2011 could be attributed to visual congruence, which raises the question of whether visual congruence is a prerequisite for identity congruence. Do people need to accept the visual appearance of the logo before they can embrace it as an identity symbol?
Another interesting point is that 2009 and 2016 almost mirror one another. The associations change from resistance to embrace, and from "childish" and "ugly" to "stylish" and "modern". This polarity trend is also evident on a more detailed level in the data, in that many associations have an opposite counterpart. It is almost as if the respondents were engaged in a debate, voicing competing views about the logo and its meaning. The negotiated identity seems to be heterogeneous and ambiguous, representing different things to different people and mixing visual, organizational and brand elements.
Finally, although the logo was meant to be a "blank canvas", it evoked a strong response. The obvious question is, 'Why?'. One plausible explanation relates to the incongruities between the brand, the logo and what was expected from the university. The name Aalto means "wave" in Finnish, but it also refers to the famous Finnish architect and designer Alvar Aalto. In this sense, the name could be considered fairly conservative, which may have caused the discrepancy among the respondents. Another potential explanation lies in the merger context of the study. Although the new organization was nascent, its predecessors had a long history and strong significance to the respondents, who in fact had applied to a prestigious business school without knowing what Aalto would be. Thus, the reaction could have reflected their shock at the merger itself and the incongruence of the initial Aalto brand in comparison with what they had come to expect from Helsinki School of Economics. The resistance could then have been at least partly mitigated by the fact that in 2011 and 2016 the respondents already had a relationship with Aalto University. Both cohorts had applied there. The 2011 cohort was probably also very familiar with Aalto´s predecessors, whereas those in the 2016 cohort had been subjected to Aalto´s branding efforts for years.
In sum, it is clear that there was heavy resistance before the logo was accepted, and the debate proceeded along polarized lines:-ugly -stylish, clear -unclear, inappropriate for a university -academic, and so forth. Two potential interpretations arise from the theory of social representations. According to Markova (2000), social representations may be generated by themata, which are shared preconceptions or pre-categorizations and often take the form of dyadic oppositions. They may exist implicitly in our common sense or-when problematized -emerge as sources of tension and conflict (Liu 2004) to function as "first principles" or "source ideas" (Moscovici and Vignaux 2000). In this case, this would mean that respondents facing the "meaning vacuum" posed by the new logo could utilize their pre-existing themata: the visual aspects of the logo, their disposition toward the change process and whether the logo represents their idea of their university.
The polarization could also be interpreted as polemical representations (Moscovici 1988) that may occur in inter-group conflicts in which there is typically a rhetorical counterpart to the dominant representation. However, the counterpart is not usually a real alternative, it is more like a shadow or a person of straw, created to reinforce the representation of the in-group (Gillespie 2008). Following this line of thought, we suggest that there may have been a shared representation of what university means to the business-school students in our study, but Aalto was not yet part of this representation in 2009. Instead, it was an outgroup project to be resisted, which would be well in line with the fact that the business school was one of the smaller universities in the merger and the logo was designed by a student from the school of art and design. Aalto had been accepted as part of the university representation by 2016, and the alternative-the shadow-had lost its meaning.
Finally, it is evident from our study that accepting a new logo may be a lengthy process. We followed its reception for 7 years, starting before the organization was founded, and the form of its evaluation was not yet finalized in 2011. This highlights the importance of longitudinal research in tracking change processes, such as how logos are received.

Managerial implications
We suggest that brand and marketing managers planning to redesign a visual identity should carefully consider their organizational context. In the case of organizations that are strongly value driven-such as universities, public-sector institutions and companies with a strong emphasis on values-we recommend including the community in the branding process and building up the stamina to endure resistance. It took several years before the resistance in Aalto University turned into positive brand associations.
Previous research has shown that careful consideration of the various logo elements could effectively reduce resistance. One might be well advised to endure the initial backlash, however, especially if the visual identity is intended to be for the long term. Resistance may simply mean that the community cares, and that its members are in the process of negotiating the meaning of the symbol and the identity of the organization. Resistance to the logo could also arise from criticism of the change in itself: people with limited resources to affect change resist where they can. A careful analysis of the reasons behind the resistance would be a good start. In sum, organizations should be able to cope with resistance, which does not prevent the symbol from becoming iconic later on.

Limitations and future research
This research focused on how a logo was received in a university setting. The context was the merging of three institutes of higher education and our data comprised the associations of students at the school of business, which was one of the smaller universities involved. The logo was controversial in itself, and the creation process-a competition-was unusual. Although these attributes make the case interesting, it is at the same time unclear how specific the findings are to this context. Is the case exceptional, or did the exceptional attributes make certain general dimensions visible? This question opens up several potential avenues for future research.
First, it would be interesting to examine the extent to which visual congruence has to precede identity congruence. We found some evidence of this, but the hypothesis remains to be validated and tested in other contexts. Second, it is our interpretation that resistance to the new logo could be attributed to the "meaning vacuum" imposed by the new visual identity-and the organization behind it. It could also reflect the inter-group conflict brought about by the merger, which in turn led to a dialogical process in which thematic concepts in the form of dyadic oppositions were used to negotiate the logo´s meaning. What would be the case in different settings?
Third, the Aalto case is an excellent example of brand cocreation through logos (Kim and Lim 2019), and there are probably limitations in terms of when this kind of approach would work. What would happen in other contexts-such as corporations or country brands-if stakeholder groups were given as big a role as in the Aalto case? What would be the optimal conditions for co-creation? A potential field of interest would be that of luxury brands, in which consumers ascribe high value to the products (see e.g., Lee et al. 2018). Fourth, given that the founding of Aalto University could, in retrospect, be considered a success (Ripoll-Soler and de-Miguel-Molina 2019), how would the dynamics between the logo and the organization play out had this not being the case?
Our study also has implications for the research on brand associations. In accordance with the theory of social representations, we suggest that the initial resistance could have originated either from shared preconceptions-themataor from the experienced intergroup conflict in the form of polemic social representations. This would be an interesting research question in itself. As we saw from our data, it took several years before the positive brand associations-which were in line with the intended Aalto University brandstarted to surface. What is the source of brand associations when the brand is unknown-a blank canvas? How long does it take to communicate the intended associations in different situations?
Finally, this study demonstrates the importance of context, both in business and in research. Managers should strive to understand their organizations so that they can select the best approaches. Sometimes the best way forward is to give control to the community and allow them to attach meaning to the symbol, although this may be painful. In terms of future studies, this study highlights the value of research on real social groups, which is one of the premises behind the theory of social representations. Our target group-students at the school of business-is cohesive and comprises people who share an understanding about their organization and discuss its meaning in day-to-day interaction. This is what enabled us to tap into the underlying dimensions of their associations. The same thinking could be more widely used in research on brands and logos. Potential research targets could consist of several internal and external social groups whose members might react to the organization differently.

Compliance with ethical standards
Conflict of interest On behalf of all the authors, the corresponding author states that there are no conflicts of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.