Building a University City Brand: Colombian University Students’ Perceptions

Abstract

Modern city brand trends aim to position cities according to the interests of their stakeholders; in the case of university students, it is necessary to identify the factors that determine their choice of a university city. Consequently, this study applied an exploratory model that allowed researchers to validate the factors and characteristics that influence this process in the major Colombian university cities. The partial least square statistical analysis methodology was used to collect 585 surveys from students from different cities in Colombia. The study found that the factors determining cities’ cognitive or affective image directly or indirectly influenced the university city brand. This paper’s contributions lie in exploring university city branding from students’ perceptions.

1. Introduction

The concept of a city brand was born in the 19th century to market urban places and evolved into a more complex modern concept involving disciplines such as marketing, architecture, urban planning, and tourism, among others [1,2,3,4]. A city brand aims to attract and retain visitors by developing a local identity that translates into offering distinctive benefits [5]. This process is achieved in the long term and is linked to the city’s public policies [6], with the city brand being a key factor for the attraction and retention of interest groups (companies–citizens), since it allows them to promote and position the city as ideal for these groups to establish themselves in [7].

From cities, branding has been developed and applied to the city brand with the aim of achieving an image positioning, more precisely of a corporate brand, seeking to make the city attractive to its target audiences (citizens, visitors, and companies) [8,9]. According to Oguztimur et al. [7], four lines of research have been developed in city branding: the first is the concept, processes, and measurement of the city brand; the second is the city brand strategy; the third is related to brand culture and tourism; and the fourth to urban planning and the city.

1.1. Cities and Knowledge

In the knowledge society, the university is the centre of creativity, strengthening its role as a leader in social, economic, and cultural development [10]. Universities could be part of the city brand since their functions and benefits are positive factors for the development and sustainability of cities. Indeed, they are one of the most decisive elements of the city’s university brand image. However, the relations between universities and cities vary from one case to another. In some cases, there are strong links between the citizen governments and the university centres, especially when there is a direct dependency given the public nature of many universities. Likewise, the university generates high social impact in education, employment, economic impact, and support to industry, among others [11,12].

The concept of the city of knowledge brand then arises, which focuses on offering education, research, and innovation services, with the university being the central actor in achieving this position [12].

Some authors categorise cities of knowledge depending on the type: some cities focus on technological-innovative areas (cities that generate centres of industrial innovation, such as Detroit, USA, with Fordism), whilst others are focused on the cultural-intellectual (cities focused on the so-called soft sciences, art and culture, classical Athens), the cultural-technological (cities that combine both cultural and technological developments, such as Silicon Valley in the USA), and technological-organisational cities (cities that generate innovations that solve problems derived from the large-scale urban life such as infrastructure, transport and housing, such as the city of Barcelona in Spain) [13].

The type of university (private or public) and its impact on the city becomes important. Likewise, the geolocation of these and their proximity to each other can be strategic factors to enhance the effect and importance of cities’ interests. For example, cities in which several universities are concentrated, and cities that integrate joint research and innovation programmes with other actors in the city take a step towards positioning their identity and image as a city of knowledge [14,15]. This type of impact can even be generated in small cities or cities with few universities [16].

The characterisation of a city of knowledge has always been constituted by the central participation of the universities, being a relationship of mutual dependence. In recent centuries, the cities have evolved with the universities in the urban, social, cultural development, scientific, and industrial areas. This is a challenge for cities of the future as this synergy is in danger of remaining only in symbolism or within a “status”. Universities are fundamental in developing knowledge cities [17], so the city of knowledge will be one in which knowledge, creativity, and innovation reach their full maturity, which requires concentrating enough diversity, stability and image projection [18].

1.2. University City Brand Research

There is little research on the university city brand, and the available studies are mostly case studies of cities that have positioned themselves as cities of knowledge focused especially on their universities and the measurement of their impact on economic, demographic, social, cultural and cultural and scientific indicators.

In a study of three cities by Popescu [12] that aligned their brand image with knowledge, the results revealed that investments in education and research provide an image of how a city values its people and the degree to which public and private actors invest in the development of human resources. Likewise, it demonstrated that universities are central actors in the knowledge regions and, in most cases, coordinate their efforts with the city’s brand strategy. The strategies of the leading universities studied proved to be consistent with the brand strategies of the cities in which they are located.

Moreover, the branding strategies of the cities emphasise to a great extent the role of higher education, with the prestige of the universities being the guarantor of this positioning of the city brand. This factor is aligned with the theoretical developments of branding in which awareness and brand loyalty results from a long-term relationship between universities and influence groups [19,20].

The city of knowledge brand was also analysed regarding the factors that attract and maintain interest groups [21], firstly highlighting that the cities and universities that offered the tools and infrastructure of services, habitat, and urban planning focused on the needs of scientists, students, businessmen, and others were successful in their endeavour to attract potential residents. For example, New Mexico has the highest number of physicists per capita in the United States, thanks to the laboratories it offers. On the other hand, Boston has a strong base of basic science that has attracted companies and institutes that have created wealth and attracted more top scientists. The case of San Francisco is similar, offering an attractive climate (culture of adventurers, investments and laws that favour creative workers). This has allowed San Francisco to develop into one of the cities of knowledge and information technologies. Although these cases of university or knowledge brand cities are evident, no studies were found that empirically defined the perception by the various interest groups of the university city brand. Instead, previous studies only developed the identification of the important factors for the choice of universities and destinations, including public services, specific government policies in science and technology, image, prestige, urban development, and characterisation of the city’s groups of influence [22,23].

There is a wide field of research around the cities of knowledge-universities. Thus, this study posed the following question: How can the perception of the city university brand be measured? For the development of this question, this paper initially starts with the construction of a measurement tool based on previous empirical studies related to student perceptions regarding what motivates them to choose a university or a university city, as well as the index of university cities for Colombia, which will be contrasted with the empirical results applied to the main cities that this index classifies as university cities. Next, the development of the field work, the analysis of the data, the discussion of the results and the final conclusions are presented.

1.3. Aspects That Influence the Choice of a University

It can be said that the target audiences of the highest interest in the city of knowledge brand are students, especially at secondary and higher education levels. They are the ones to whom the offers related to education, their first job, research, innovation, and entrepreneurship are directed.

For this reason, some studies have analysed and attempted to ascertain the main aspects that influence the choice of an educational centre for young applicants in order for the public-private educational sector to properly manage the offer and achievement in education [24,25,26,27,28,29]. This decision-making process is complex [24]. Indeed, it is remarkably similar to the classic purchase decision behaviour, in which there are three stages: the first is the “awareness stage”, where the student generates the desire to carry out their studies; the second where the “information search” begins by analysing the different alternatives; and thirdly, the final phase of the “decision” where the student finally chooses where to carry out their studies.

Therefore, several characteristics of the place where the studies are carried out are important for the student when deciding where to study; places that have cultural and geographical proximity, for example, will be more attractive [30].

There has been limited academic research regarding the influence of the specific positioning of the university or knowledge city brand. However, it has been proven that the positioning of the destination has influenced the choice of a country or city as a place for studying and pursuing the life project of young students. For instance, factors such as the reputation of the place, quality of life, and infrastructure, among others, play a significant role (see Table 1).

1.4. Measurement Index of the Concept of Colombian University Cities-ICU

The network of higher education institutions called: “How We Are Going” has built the measurement index of university cities (ICU), which aims to measure and categorise the cities of Colombia as university cities. This index has 4 dimensions and a total of 19 variables built with data from the National Ministry of Education and other institutions specialised in education in this country. Information is taken from the QS international ranking, which collects information on prices and other factors in each city periodically (See Table 2).

The general results of the ICU positioned Manizales, Medellín, and Pereira as the leading Colombian cities in 2019, followed by a group of four cities that achieved around 60 points in the index: Barranquilla, Bucaramanga, Cali, and Bogotá. However, the results by dimension vary in all cases; therefore, we cannot conclude that any of the best-valued cities are the best across all the measured aspects. For example, in the cost of living dimension, the city of Quibdó has the highest score but is the last city across the total scores (See Table 3).

Considering the results of the ICU index, similar perceptions should be had of students who are choosing where to carry out their university studies or who have just chosen a city to carry out their university studies with respect to the ICU index. If applicants were asked directly, Colombian university students would be expected to evaluate according to their perceptions in the same way as the ICU index.

Empirical studies on students’ perceptions of destinations for university studies have concluded that their perceptions based on the information and previous experiences with destinations generate an image and give each place a personality [29]. Therefore, it could also be assumed that the students, based on the information they seek and receive from the cities, will generate both a cognitive and an affective image and may or may not perceive a destination as a university city brand.

Completing the last and most important assumption, the cognitive and affective image and the personality of the city influence students’ choice of where to conduct studies [29].

Therefore, the following hypotheses are proposed (Figure 1):

H1a. 

The perception of rankings value influences the cognitive image of the university city brand.

H1b. 

The perception of rankings value influences the perception of the university city brand.

H2a. 

The perception of the cost of living influences the cognitive image of a city with a university brand.

H2b. 

The perception of the cost of living influences the perception of a university city brand.

H3a. 

The perception of job opportunities influences the cognitive image of the university city brand.

H3b. 

The perception of job opportunity influences the perception of a university city brand.

H4a. 

The perception of the physical environment influences the cognitive image of the university city brand.

H4b. 

The perception of the physical environment and facilities influences the perception of a university city brand.

H5. 

The cognitive image of a university city influences the perception of a university city brand.

H6a. 

Social influence influences the affective image of the university city.

H6b. 

Social influence influences the perception of the university city brand.

H7a. 

E-wom influences the affective image of the university city brand.

H7b. 

E-wom influences the perception of the university city brand.

H8. 

The affective image of a university city brand influences the perception of the university city brand.

2. Materials and Methods

Given its objective, this study was structured according to an exploratory approach to conform and test a model that would allow researchers to empirically explain a phenomenon that has not yet been widely verified. To achieve this, the methodology used was modelling using the structural equations method (SEM), given its benefits regarding the verification of cause-effect relationships between constructs for the explanation of behavioural models, which is widely accepted in marketing studies. The method that best allows working with exploratory or rarely studied models is partial least square (PLS) [30]. Therefore, the data collected was analysed following the main analysis tests for the measurement tool and the proposed hypothesis model.

Measurement Tool

The measurement tool was built through the following process. First, a search for empirical studies was carried out through the in-depth content analysis of scientific publications (Scopus and Web of Science databases) investigating research on decision-making in young people regarding where to carry out their university studies, without taking into account bibliometric document filter processes given the scarce development of this type of research. From this first phase, we discovered that the authors found different constructs to explain the reasoning behind deciding on a place to conduct university studies. These were tested in different contexts, such as choice of university, choice of course abroad, and choice of city to study. The constructs were grouped into a total of 10 and presented with their respective measurement indicators in Figure 1.

Next, the measurement tool was constructed through a psychometric test process that required the translation and retranslation of the items from English to Spanish and then a pre-test with a group of five marketing researchers who verified the semantic and comprehension aspects of the questions in Spanish. They also removed or added words that would facilitate clarity and precision in the measurement (all items were measured using Likert scales with values between 1–7 to have a better level of contrast).

Finally, a pilot test was carried out with 50 university students in order to verify whether there were any comprehension problems with the items. Statistical tests were not carried out as the constructs, and their elements have already been tested in previous empirical studies with high rigour. Additionally, in the analysis phase of this study, tests corresponding to the validity of the measurement tool used were carried out. The final questionnaire was distributed digitally among various universities in the seven selected cities.

3. Results

3.1. Sample

The sample was defined by a quota method based on the first seven cities in the ranking of Colombian university cities. A minimum value of 50 samples per city was set to have a minimum sample of 350 responses, an acceptable value for applying the PLS methodology. The final result was a total of 585 valid answers (excluding missing values and incomplete answers). The characterisation of the sample revealed that 45% were men and 55% were women, 83% were young people between 16 and 25 years old, 15% were between 26 and 35 years old, and 2% were over 35 years old. Meanwhile, 87% were students studying professional university programmes, and 13% were in postgraduate courses. Subsamples by the city are detailed in Table 4.

3.2. Reliability and Validity of the Measurement Tool

The results of the reliability tests (see Table 5) show that the individual reliability of each item is supported and all the tests are met: (loads > 0.7—significant difference: p-value < 0.05), simple reliability (Cronbach’s Alpha > 0.7) and composite reliability (IFC > 0.7). The values of the multicollinearity test (VIF < 3.3) confirm that the data comply with the measurement behaviour of a high level of reliability [32].

Regarding the validity tests, the convergent validity test (see Table 2; AVE > 0.7) and the discriminant validity test (see Table 6 and Table 7), the Fornell–Larcker criterium test [33,34], ascertained that the indicators measure the constructs associated with them and do not measure others, since the values below the diagonal of the matrix are lower than the values of the construct itself. The Heterotrait–Monotrait Ratio (HTMT) test [33] did not present values higher than 8.0, reaffirming that the convergent validity is adequate.

3.3. Prediction and Validity of the Empirical Model

The ${\mathrm{R}}^2$ value (Table 8) for the city brand university explained variable was greater than 0.6, so the predictive relevance of the model was validated [35].

Similarly, when performing the validation test of the model hypotheses, the bootstrapping process was carried out with an N > 10,000 subsamples to have greater verification rigour [34] (Table 6). The results supported the hypothesis: H1a (rankings value -> cognitive image, β = 0.214*); H1b (rankings value -> city brand university, β = 0.200*); H3a (job opportunity -> cognitive image, β = 0.278*); H4a (physical environment -> cognitive image, β = 0.166*); H5 (cognitive image -> city brand university, β = 0.269*); H6a (social influence -> affective image, β = 0.346*); H7a (E-wom -> affective image, β= 0.214*); H7b (E-wom -> city brand university, β = 0.137*); and H8: affective image -> city brand university, β = 0.424*). Hypotheses that were not supported were H2a (cost life -> cognitive image, β = 0.058); H2b (cost life -> city brand university, β = −0.015); H3b (job opportunity -> city brand university, β = 0.040); H4b (physical environment -> city brand university, β = −0.010); and H6b (social influence -> city brand university, β = −0.056) (Table 9) (Figure 2). The VIF values < 0.3 reject multicollinearity problems between the constructs (Table 10).

However, the results of the total effects show that the job opportunity construct exerts a positive effect on the perception of the university city brand of β = 0.114, which indicates there is a partial mediating effect (see Table 11 and Table 12) exerted by the cognitive image regarding the perception of the city brand. The cities that offer high job opportunities are also important for students in their perception of the university city brand. The same is true of social influence, which was not supported. After analysing its partial and total effect, it generates a positive influence β = 0.090, which, although it is less than 1, can be translated into the fact that the mediating effect of the affective image generates a positive effect of social influence on the perception of the university city brand.

Therefore, the results find a high causal relationship between the perception of the cognitive image and the affective image that a student has of a city on the perception of the university city brand, as well as all the independent constructs: rankings value, job opportunity, and physical environment, excluding the cost of living. These are influential in a mediated or direct way on the perception of the university city brand.

3.4. Unobserved Heterogeneity Test—Measure Invariance of Composite Models

Given the nature of the sample in which data was taken from different cities in Colombia, which shows heterogeneity by the geographical area of origin of the samples, the sample should be considered to perform the unobserved heterogeneity test [34]. The test is to verify if there are differences between the samples of each city. Subsequently, there may be non-homogeneous behaviour from the data, which may lead to incorrect generalised results and, therefore, the grouped invalidity of the data [34].

The unobserved heterogeneity test means that it is possible to find the different behaviours between these groups, if any. In the case of this study, it makes it possible to analyse whether there is a different behaviour for each city (city number 1 in the ranking was chosen and contrasted with three other cities).

There are several methodologies to evaluate invariance for covariance-based SEM (CB-SEM), the most frequently used being multigroup confirmatory factor analysis. However, for PLS-SEM models, the measurement invariance of composite models (MICOM) procedure has been developed [34,36]. This test consists of the following steps: (1) the configuration invariance is determined, (2) the compound invariance is determined, and (3) the equality of means and variances of the compounds are identified. It is mandatory to confirm each invariance to determine the partial invariance (of configuration and composite) or the total (of configuration, composite, and equality of means and variances between groups). Finally, the verification of the total or partial invariance allows the performance of the multigroup analysis to determine the possible differences between the path coefficients. If the invariance is not verified, each model must be analysed separately, and the possible differences in the path coefficient results must be compared.

The measurement tool and the treatment of the data guaranteed that the same indicators and the same scales were used in all the groups. Similarly, the coding and screening of the final data were treated similarly. Finally, the validity tests of the indicators complied with the recommended quality standards and tests for the reliability of the tool (Table 5). Therefore, the configuration invariance is satisfied.

Next, the invariance of the compound was analysed, establishing if the scores of the compounds in all the groups were the same. This is achieved by examining the original correlations between the first group (city 1) and each of the other groups (the following six cities) using the following process: (1) PLS nomograms were estimated for each group to obtain the scores of the compounds of the groups (number of observations per group and total). (2) The correlations between the scores of the compounds of the groups were calculated. (3) A random permutation was carried out (with a minimum of 1000 permutations). (4) For each permutation, the calculations of the corrections between the scores of the compounds were made, comparing the first group with respect to each of the other groups. (5) A hypothesis test was carried out, classifying the results of the correlations of the permutations in descending order with a cut-off value of 95%. The hypothesis or invariance of the model is accepted when the value is significantly different from 1 with a level p < 0.05.

Finally, the total invariance was verified through the test of equality of means and variances of the compounds by determining the logarithm of the ratio of the variance of the scores of the compounds for the groups. If the logarithm of this variance is not statistically different from 0, it is concluded that the variances are equal in all groups; therefore, the total invariance can be verified [34] (Table A1, Table A2 and Table A3).

3.5. Differences between Groups–Analysis Based on Permutations–Parametric Test–t-Test Welch–Satterhwaite–PLS-MGA

We chose to only make comparisons between pairs of cities to explore the possible differences in the model results. As a result, the tests were applied to analyse two groups, i.e., applying them to the first city in the ranking of university cities, Manizales, and comparing it with Medellín. It was then compared with Pereira and finally with Bogota. The multigroup tests resulting from the permutations, including the parametric test and the t-test Welch–Satterhwaite, are shown. In addition, the non-parametric PLS-MGA approach was applied, which compares each estimate made in a bootstrapping procedure [34]. The Welch–Satterthwaite parametric test confirms no significant differences between the groups (Table A4, Table A5, Table A6, Table A7, Table A8, Table A9, Table A10, Table A11 and Table A12).

These results show that the model fits very similarly in all cities and, therefore, it can be inferred that students behave in the same way with respect to the same important constructs when choosing a city in which to carry out their studies, as well as their perception of the university town.

4. Discussion

The objective of this paper was to explore an empirical model that would determine what factors generate university students’ perception of the university city brand. A model was formed that included the main constructs developed to date of student behaviour around their choice of place for their university studies and the positioning of their perception of the city.

A model was built that started from six independent constructs that had been tested in previous empirical studies as incidents in the choice of a centre, place, or country to carry out university studies. However, these had not yet been explored as determining elements of students’ image perception. Consequently, the cognitive image, the affective image, and the image of the university city brand were considered dependent constructs.

The results show that the cognitive image that a student generates of a city is influenced by their perception of the quality rankings on the offers of university education that the city offers, followed by the international rankings of university quality (e.g., ARWU, CWUR, QS and others). This is in addition to students’ evaluation of universities and the academic programmes and degree prestige that the universities have. The rankings’ quality significantly affects a student’s perception of a city in their election process, which could mean that those cities that appear in the first positions in educational rankings will have more positioning as a university city.

Regarding the cost of living, the results were insignificant in their effect on the cognitive image or the perception of the university city brand. This element is complex and cannot be excluded from future analyses, given that the cost generated by undertaking university study includes maintenance, housing, transportation, school supplies, complementary courses, and enrollment. In the Colombian case, students may perceive cities to be very similar, meaning they are not seen as a determining factor.

Job opportunities were validated as a factor that influences the cognitive image of a city, although it was not directly for the perception of a branded university city. This result shows that students positively perceive cities with high employment opportunities as places to live. As a result, this would, to a certain extent, motivate their choice of a city as a place in which to carry out their university studies. Furthermore, the non-validity of the direct influence could be because students, when they potentially choose a city in which to study, think in the short term since many will only dedicate themselves to studying full time for several years and do not even think about the future working market.

For its part, the physical environment factor was significant in its impact on the cognitive image of the city, although not directly on the perception of the university city brand and not mediated by its effect on the cognitive image. The result may be because it is a broad factor that the student perceives in a complex way since it refers to the city’s infrastructure and means of transportation, among many other elements that students may not directly associate with university campuses. However, as a great result, this study shows that the perception of a positive cognitive image regarding a city positively affects the perception of the university city brand, even being decisive.

Social influence is another significant factor in its incidence for forming the positive affective image of a city. However, although it was not directly significant with respect to the university city brand, the mediation analyses establish that it has a positive influence through the cognitive image on the perception of the university city. This result is interesting because students form feelings and emotions towards cities from their families and close nuclei. This factor may therefore be decisive for students, potentially even the vast majority, to perceive their own hometowns as university cities.

More influentially, this study found that the e-wom is decisive in the perception of both the affective image of a city and the perception of the city university brand. This demonstrates the power of digital information related to comments, experiences, evaluations, suggestions and other sources of e-wom that the internet openly provides students [35]. That could be the greatest consultation they use when deciding on a university.

Finally, this study also showed that affective image positively influences the perception of a university city. This result shows that students who have an emotional image towards a city will consider it an optimal city to carry out their university studies.

5. Conclusions

The theoretical contributions of this study are valuable, given that this study is one of the few that has analysed students’ perceptions of a city and whether they consider it a university city. Indeed, most studies in this line of research have presented case analyses or qualitative studies that did not test constructs only analysed as a university choice. Consequently, this study provides a new vision around the decision-making of the university centre as something more aligned with city marketing, demonstrating that students perhaps value branded cities more as the initial factor for choosing a university study centre and not the other way around. However, this is not directly analysed in this study.

Considering the aforementioned findings, this study explored the possible constructs of greatest importance in choosing a university city. For example, the ranking of the university cities determines whether a city is categorised as a university city brand. Therefore, another great contribution of this article has been to verify that this position is due to the cognitive image and the affective image that a student has of a city.

Moreover, the results of this study, applying the unobserved heterogeneity and measurement invariance tests, revealed that the data behaved similarly in all the cities. Therefore, in an exploratory way, this behaviour could be extrapolated to decision making in general for the election of a city university by students.

The practical and managerial contributions of this study offer valuable information so that city governments take the key elements for the construction of policies and actions focused on strengthening the factors that interest students. This will generate the positioning of both cognitive and affective images, the two central elements for the good positioning of a university city brand, which could differentiate cities to retain and attract university students in the city educational ecosystem.

On the other hand, the measurement tool that was built and adequately tested for its validity and measurement can be used by mayors of cities interested in measuring the positioning of the city brand so that they can analyse these perceptions within their interest group (students).

This study raises the need to generate strategies between government entities of cities and universities to jointly plan and manage the positioning of the university city brand, since a correlation is evidenced between the choice of the university based on the image of the city that prospective students have.

This study’s limitations lie in exploring a new empirical model, which requires it to be tested in other cities to evaluate its behaviour. However, as demonstrated in the unobserved heterogeneity and measurement invariance tests, it possibly behaves similarly. Therefore, confirmatory analyses of the model in other cities are needed. On the other hand, other variables such as environment and educational facilities that may indicate the perception of a city with a university brand may have been left out of the model. Consequently, we suggest that future studies include them if their importance is evident.

Finally, this study contributes to developing the research line in city branding, which is becoming increasingly important for modern and sustainable cities.

Author Contributions

Conceptualisation, E.J.D.O. and J.A.S.-T.; methodology, E.J.D.O. and J.A.S.-T.; software, E.J.D.O. and J.A.S.-T.; validation, E.J.D.O. and J.A.S.-T.; formal analysis E.J.D.O. and J.A.S.-T.; investigation E.J.D.O. and J.A.S.-T.; resources, E.J.D.O. and J.A.S.-T.; data curation, E.J.D.O. and J.A.S.-T.; writing—original draft preparation, E.J.D.O. and J.A.S.-T.; writing—review and editing, E.J.D.O. and J.A.S.-T.; visualisation, E.J.D.O. and J.A.S.-T.; supervision E.J.D.O. and J.A.S.-T.; project administration, E.J.D.O. and J.A.S.-T.; funding acquisition, E.J.D.O. and J.A.S.-T. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board.

Informed Consent Statement

Informed consent was obtained from all subjects involved in the study.

Data Availability Statement

Data are unavailable due to privacy or ethical restrictions.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Proposed model.

Figure 2. Structural model.* Significant difference: p-value < 0.05.

Table 1

Reasons for choosing a place to carry out university studies.

Construct	Description	Items	Author
Rankings Value	It assesses the national and international educational quality measurement indexes in choosing a city to conduct university studies (including evaluation of universities, academic programs, and degree prestige).	To choose the destination where to carry out my studies:R1. I have considered the reputation of the universities in that place.R2. I had/I have in mind the place that the universities of that place occupy in the international rankings.R3. I have considered the quality standards in the education of the universities of that place.R4. I have considered the prestige that the graduates of the universities of that place have.	[24,25,27,29]
Cost of living	It refers to considering all the costs of living and expenses in general for choosing a place of study (including tuition costs, living expenses, food, transportation, leisure, and health, among others).	To choose the destination where to carry out my studies:CL1. I have considered the costs of accommodation.CL2. I have considered the costs of living.CL3. I looked/I look for it to have reasonable costs of living.	[24,25,26,27,28,29]
Job opportunity	It refers to considering the favourable opportunities to build a life project in that place after finishing. The career opportunities when choosing a destination for university studies focus on employment and habitat possibilities.	To choose the destination where to carry out my studies:EO1. I had/I am considering the opportunity to stay living there after finishing my studies.EO2. I had/I have in mind to have the opportunity to remain residing in that place permanently.EO3. I have considered the potential for employment there after finishing my studies.	[24,25]
Environment and educational facilities	It refers to considering the offer of support and complementary services and activities offered by the said place when choosing a place to conduct university studies.	To choose the destination where to carry out my studies:EE1. I valued/I value that I can study other complementary courses (languages, complementary, postgraduate, among others).EE2. I valued/I value the facilities in educational services.EE3. I valued/I value safe and adequate spaces to carry out my studies (university campuses, libraries, etc.).	[24]
Physical environment and leisure facilities	It refers to considering the choice of a place to carry out university studies and the urban and habitat equipment of the place of educational offer (includes urban housing, accessibility, roads, supply of public services, leisure sites, spaces and sports facilities, among others).	To choose the destination where to carry out my studies:PE1. I valued/I value the wide offer and good housing prices to carry out my studies.PE2. I value/I value facilities in habitat services.PE3. I valued safe and adequate urban spaces to live in.	[24,25,26]
E-wom	It refers to the information (comments or opinions) on the internet from different digital sources that the student consults about the places of study to decide their destination.	To choose the destination where to carry out my studies:EW1. I valued/I value the comments on social networks about that destination.EW2. I searched/I am looking for opinions and information on the internet about that placeEW3. I/I take into account all kinds of information from the internet about a place.	[25,27,29]
Social influence	The people close to the student whose opinions are valued in the decision-making process of the place of study. Mainly it has been found that parents and close family influence the decision of choosing where to study, but also groups of friends, people or leaders of influence and students who have already had previous experiences.	To choose the destination where to carry out my studies:YES1. I value/I value the opinions of someone who studies in a place.YES 2. I/I have taken into account the opinion of my parents to study in that place.YES 3. I valued/I value the opinions of close relatives about studying in that place.SI4. I valued/I value the opinions of my teachers about studying in that place.	[24,25]
Cognitive image of the country/city	It refers to the perception of the quality of life, the level of technological advances, and the development of a city or country.	Regarding the city of choice as a destination to carry out my studies:CI1. I consider that this city has high levels of quality of life.CI2. I consider this city to have high levels of economic development.CI3. I consider that this city has high levels of technological and industrial development.	[31]
Affective image of the country/city	It refers to the feeling measured by love or hate, admiration or rejection, or a positive/negative image of a city or country.	Regarding the city of choice as a destination to carry out my studies:AI1. This city generates a good image for me, and I have good impressions.AI2. This city has a good image, and I have good comments about it.AI3. This city generates an image of a welcoming city for me to carry out my studies.	[31]
University City Brand Personality	It refers to the characteristics associated with a university city that represents the name of a city-region-country based on the perceptions and experiences of the institutions and inhabitants of that place.	Regarding the city of choice as a destination to carry out my studies:PC1. I identify this city as ideal for studying because of the name it represents worldwide.PC2. I identify this city as ideal for studying because of the reputation of its most representative institutions.PC3. I identify this city as ideal for studying because of the characteristics of the people who originate from there.PC4. I identify this city as ideal to study because its characteristics are aligned with knowledge.	[31]

Table 2

Index of Colombian university cities—ICU.

Construct	Item (Measured with Aggregate Data and Student Interviews)
Quality university education	Availability of highly trained teachers
	Academic achievement of city students
	High-quality accreditation coverage
	Presence of world-class universities in the city
University environment	Presence of university students in the city
	Variety of local university offer
	tolerance and inclusion
	Cultural and recreational offers of the city
Living cost	feed cost
	housing cost
	Transport cost
	Price level change
Quality of life	Satisfaction with the city
	Environment in the city
	Mobility in the city
	Perception of city safety
Graduate employability	Probability of job attachment
	Down payment salary
	Youth unemployment rate

Table 3

ICU results by the main cities in Colombia as of 2019.

City	Total Value	Quality University Education	University Environment	Living Cost	Quality of Life	Graduate Employability
Manizales	70.88	81.54	88.68	54.70	96.00	48.90
Medellín	66.58	85.07	78.40	23.70	66.70	72.90
Pereira	64.54	54.63	73.30	60.40	73.10	67.30
Barranquilla	61.55	58.48	76.48	45.5	69.30	32.40
Bucaramanga	61.47	73.17	65.36	47.60	56.30	59.80
Cali	61.15	76.53	54.18	54.50	48.30	61.80
Bogotá	59.18	85.66	58.39	25.20	34.20	70.50
Ibagué	52.22	51.92	56.86	70.70	71.00	26.50
Cartagena	46.30	36.49	55.58	23.20	48.40	66.30
Santa Marta	45.47	20.99	56.98	70.70	52.30	37.80
Montería	45.39	27.83	67.83	42.70	84.00	31.70
Cúcuta	42.06	14.43	59.84	79.50	51.00	22.00
Armenia	37.67	29.41	47.12	41.00	56.00	28.4
Quibdó	35.51	18.34	33.19	94.10	38.20	6.60

Table 4

Sample.

City	Sub-Sample	%
Manizales	86	15%
Medellín	74	12%
Pereira	79	13%
Barranquilla	70	12%
Bucaramanga	72	13%
Cali	79	14%
Bogotá	125	21%
Total sample	585	100%

Table 5

Resume test of reliability and validity of the measurement tool.

Item	Loads *	t-Statistic	p Values	VIF Values	Construct	Cronbach’s Alpha	rho_A	Composite Reliability (IFC)	Average Variance Extracted (AVE)
AI1	0.937	106.795	0.000	3.001	Affective image	0.928	0.928	0.955	0.875
AI2	0.947	137.003	0.000	2.512
AI3	0.921	97.036	0.000	2.904
CI1	0.859	56.341	0.000	1.963	Cognitive image	0.880	0.880	0.926	0.807
CI2	0.932	108.396	0.000	3.001
CI3	0.902	76.364	0.000	3.030
CL1	0.904	57.861	0.000	2.805	Cost life	0.903	0.906	0.939	0.838
CL2	0.936	84.423	0.000	2.515
CL3	0.904	52.538	0.000	2.672
CP1	0.895	81.457	0.000	2.373	City brand university	0.897	0.897	0.936	0.829
CP2	0.917	90.514	0.000	3.005
CP3	0.918	98.705	0.000	3.002
JO1	0.896	60.673	0.000	2.991	Job opportunity	0.859	0.860	0.914	0.781
JO2	0.907	69.509	0.000	2.808
JO3	0.845	45.607	0.000	1.676
EW1	0.869	55.754	0.000	1.876	E-wom	0.862	0.865	0.916	0.784
EW2	0.891	54.234	0.000	2.562
EW3	0.895	63.892	0.000	2.508
PE1	0.880	49.032	0.000	2.541	Physical environment	0.900	0.913	0.937	0.832
PE2	0.936	107.279	0.000	3.003
PE3	0.919	103.068	0.000	2.816
RV1	0.841	40.837	0.000	1.980	Rankings Value	0.846	0.866	0.896	0.783
RV2	0.738	22.579	0.000	1.598
RV3	0.889	68.361	0.000	2.397
RV4	0.828	35.368	0.000	1.891
SI1	0.822	27.781	0.000	1.287	Social influence	0.730	0.764	0.842	0.740
SI2	0.727	15.591	0.000	1.590
SI3	0.842	34.976	0.000	1.769

(Note: * Significant difference: p-value < 0.05).

Table 6

Fornell–Larcker criterium.

Affective image	0.935
City brand university	0.714	0.910
Cognitive image	0.744	0.701	0.898
Cost life	0.291	0.303	0.322	0.915
E-wom	0.346	0.453	0.394	0.326	0.885
Job opportunity	0.389	0.423	0.477	0.380	0.345	0.884
Physical environment	0.398	0.421	0.433	0.499	0.575	0.565	0.912
Rankings Value	0.314	0.478	0.406	0.352	0.412	0.389	0.380	0.827
Social influence	0.428	0.385	0.425	0.349	0.383	0.348	0.402	0.435	0.800

Table 7

Heterotrait–Monotrait Ratio (HTMT).

Affective image	-
City brand university	0.781
Cognitive image	0.723	0.788
Cost life	0.318	0.336	0.361
E-wom	0.384	0.514	0.451	0.366
Job opportunity	0.437	0.480	0.546	0.429	0.395
Physical environment	0.431	0.464	0.482	0.552	0.646	0.640
Rankings Value	0.346	0.541	0.460	0.391	0.484	0.444	0.426
Social influence	0.496	0.450	0.510	0.428	0.470	0.421	0.479	0.518

Table 8

Values R square.

Construct	${\mathbf{R}}^2$	$\mathbf{Adjusted}{\mathbf{R}}^2$
Affective image	0.222	0.219
City brand university	0.632	0.625
Cognitive image	0.310	0.304

Table 9

Test hypotheses.

Hypothesis	Validity	Original Sample (β)	t-Statistic	p Values
H1a: Rankings Value -> Cognitive image	Supported	0.214 *	4.258	0.000
H1b: Rankings Value -> City brand university	Supported	0.200 *	3.984	0.000
H2a: Cost life -> Cognitive image	Not supported	0.058	0.976	0.329
H2b: Cost life -> City brand university	Not supported	−0.015	0.347	0.729
H3a: Job opportunity -> Cognitive image	Supported	0.278 *	5.020	0.000
H3b: Job opportunity -> City brand university	Not supported	0.040	0.952	0.341
H4a: Physical environment -> Cognitive image	Supported	0.166 *	2.781	0.005
H4b: Physical environment -> City brand university	Not supported	−0.010	0.190	0.850
H5: Cognitive image -> City brand university	Supported	0.269 *	4.199	0.000
H6a: Social influence -> Affective image	Supported	0.346 *	7.437	0.000
H6b: Social influence -> City brand university	Not supported	−0.056	1.390	0.165
H7a: E-wom -> Affective image	Supported	0.214 *	4.654	0.000
H7b: E-wom -> City brand university	Supported	0.137 *	3.122	0.002
H8: Affective image -> City brand university	Supported	0.424 *	6.508	0.000

(Note: * Significant difference: p-value < 0.05).

Table 10

VIF values of the structural model.

Construct	Affective Image	City Brand University	Cognitive Image
Affective image		2.368
Cognitive image		2.635
Cost life		1.444	1.402
E-wom	1.171	1.658
Job opportunity		1.700	1.564
Physical environment		2.207	1.737
Rankings value		1.501	1.273
Social influence	1.171	1.511

Table 11

Specific indirect effects.

Specific Indirect Effects	Value
E-wom -> Affective image -> City brand university	0.091
Social influence -> Affective image -> City brand university	0.147
Cost life -> Cognitive image -> City brand university	0.016
Job opportunity -> Cognitive image -> City brand university	0.075
Physical environment -> Cognitive image -> City brand university	0.044
Rankings Value -> Cognitive image -> City brand university	0.057

Table 12

Effect’s totals.

Construct	Affective Image	City Brand University	Cognitive Image
Affective image		0.424
Cognitive image		0.269
Cost life		0.001	0.058
E-wom	0.214	0.228
Job opportunity		0.114	0.278
Physical environment		0.035	0.166
Rankings value		0.258	0.214
Social influence	0.346	0.090

Appendix A

Table A1

MICON (Manizales City—Medellin city).

Composite	Correlation	5% Quartile of the Empirical Distribution of c	p-Value	Is Invariance Proven?
Affective image	1.000	1.000	0.445	Yes
City brand university	1.000	1.000	0.619	Yes
Cognitive image	1.000	0.998	0.937	Yes
Cost life	0.999	0.997	0.189	Yes
E-wom	0.997	0.988	0.394	Yes
Job opportunity	1.000	0.996	0.714	Yes
Physical environment	1.000	0.996	0.670	Yes
Rankings value	0.996	0.989	0.324	Yes
Social influence	0.991	0.973	0.343	Yes
Composite	Composite Mean Difference (=0)	95% confidence interval	p-value	Equality of means?
Affective image	−0.033	(−0.347; 0.282)	0.667	Yes
City brand university	−0.262	(−0.327; 0.307)	0.614	Yes
Cognitive image	−0.458	(−0.329; 0.289)	0.889	Not
Cost life	−0.125	(−0.314; 0.303)	0.464	Yes
E-wom	−0.311	(−0.330; 0.310)	0.805	Yes
Job opportunity	−0.289	(−0.341; 0.294)	0.348	Yes
Physical environment	−0.147	(−0.337; 0.307)	0.635	Yes
Rankings value	−0.266	(−0.323; 0.304)	0.188	Yes
Social influence	−0.304	(−0.307; 0.295)	0.508	Yes
Composite	Difference of the logarithm of the variances of the composite (=0)	95% confidence interval	p-value	Equality of variances?
Affective image	0.161	(−0.696; 0.756)	0.667	Yes
City brand university	0.129	(−0.508; 0.519)	0.614	Yes
Cognitive image	0.036	(−0.521; 0.544)	0.889	Yes
Cost life	0.216	(−0.584; 0.600)	0.464	Yes
E-wom	−0.046	(−0.356; 0.350)	0.805	Yes
Job opportunity	0.404	(−0.434; 0.504)	0.548	Yes
Physical environment	−0.104	(−0.448;0.470)	0.635	Yes
Rankings value	0.367	(−0.533; 0.553)	0.188	Yes
Social influence	0.153	(−0.450; 0.437)	0.508	Yes

Table A2

MICON (Manizales City—Pereira city).

Composite	Correlation	5% Quartile of the Empirical Distribution of c	p-Value	Is Invariance Proven?
Affective image	1.000	1.000	0.536	Yes
City brand university	1.000	0.999	1.000	Yes
Cognitive image	0.999	0.998	0.293	Yes
Cost life	0.999	0.976	0.900	Yes
E-wom	0.991	0.990	0.066	Yes
Job opportunity	0.999	0.986	0.668	Yes
Physical environment	1.000	0.997	0.913	Yes
Rankings value	0.995	0.988	0.371	Yes
Social influence	1.000	0.947	0.975	Yes
Composite	Composite Mean Difference (=0)	95% confidence interval	p-value	Equality of means?
Affective image	0.208	(−0.315; 0.291)	0.193	Yes
City brand university	0.237	(−0.305; 0.327)	0.140	Yes
Cognitive image	0.143	(−0.291; 0.294)	0.366	Yes
Cost life	0.072	(−0.316; 0.313)	0.605	Yes
E-wom	0.138	(−0.299; 0.298)	0.382	Yes
Job opportunity	−0.087	(−0.291; 0.331)	0.573	Yes
Physical environment	0.083	(−0.300; 0.287)	0.575	Yes
Rankings value	−0.076	(−0.283; 0.288)	0.634	Yes
Social influence	−0.017	(−0.300; 0.280)	0.907	Yes
Composite	Difference of the logarithm of the variances of the composite (=0)	95% confidence interval	p-value	Equality of variances?
Affective image	0.172	(−0.696; 0.756)	0.562	Yes
City brand university	−0.013	(−0.508; 0.519)	0.938	Yes
Cognitive image	−0.148	(−0.521; 0.544)	0.502	Yes
Cost life	0.119	(−0.584; 0.600)	0.646	Yes
E-wom	−0.100	(−0.356; 0.350)	0.526	Yes
Job opportunity	0.419	(−0.434; 0.504)	0.446	Yes
Physical environment	0.093	(−0.448; 0.470)	0.671	Yes
Rankings value	0.061	(−0.533; 0.553)	0.799	Yes
Social influence	0.201	(−0.450; 0.437)	0.336	Yes

Table A3

MICON (Manizales City—Bogota city).

Composite	Correlation	5% Quartile of the Empirical Distribution of c	p-Value	Is Invariance Proven?
Affective image	1.000	0.999	0.582	Yes
City brand university	1.000	0.999	0.870	Yes
Cognitive image	1.000	0.997	0.993	Yes
Cost life	0.996	0.995	0.085	Yes
E-wom	0.996	0.990	0.058	Yes
Job opportunity	1.000	0.996	0.907	Yes
Physical environment	0.999	0.995	0.689	Yes
Rankings value	0.998	0.995	0.378	Yes
Social influence	0.998	0.954	0.860	Yes
Composite	Composite Mean Difference (=0)	95% confidence interval	p-value	Equality of means?
Affective image	0.101	(−0.275; 0.276)	0.193	Yes
City brand university	0.146	(−0.284; 0.263)	0.140	Yes
Cognitive image	0.090	(−0.273; 0.280)	0.366	Yes
Cost life	0.148	(−0.285; 0.295)	0.605	Yes
E-wom	0.141	(−0.299; 0.292)	0.382	Yes
Job opportunity	−0.152	(−0.277; 0.265)	0.573	Yes
Physical environment	0.135	(−0.288; 0.284)	0.575	Yes
Rankings value	−0.278	(−0.280; 0.288)	0.634	Yes
Social influence	0.141	(−0.261; 0.281)	0.907	Yes
Composite	Difference of the logarithm of the variances of the composite (=0)	95% confidence interval	p-value	Equality of variances?
Affective image	−0.151	(−0.401; 0.356)	0.444	Yes
City brand university	0.079	(−0.411; 0.395)	0.707	Yes
Cognitive image	−0.084	(−0.444; 0.393)	0.695	Yes
Cost life	−0.092	(−0446; 0.420)	0.700	Yes
E-wom	0.051	(−0.319; 0.289)	0.762	Yes
Job opportunity	0.153	(−0.410; 0.324)	0.387	Yes
Physical environment	−0.059	(−0.341; 0.334)	0.751	Yes
Rankings value	0.233	(−0.498; 0.428)	0.304	Yes
Social influence	−0.029	(−0.387; 0.308)	0.869	Yes

Table A4

Differences between Manizales City—Medellin city—based on permutation.

Relation	Path Coefficients Original Manizales City	Path Coefficients Original Medellin city	Path Coefficients Original Difference	Path Coefficients Permutation mean Differences	2.5%	97.5%	Permutation p-Values
Affective image -> City brand university	0.542	0.524	0.018	0.010	−0.347	0.383	0.917
Cognitive image -> City brand university	0.215	0.250	−0.034	−0.014	−0.385	0.335	0.833
Cost life -> City brand university	0.013	0.043	−0.030	−0.001	−0.294	0.310	0.847
Cost life -> Cognitive image	−0.067	0.238	−0.305	0.009	−0.447	0.480	0.196
E-wom -> Affective image	0.012 *	0.474 *	−0.462 *	0.002	−0.327	0.305	0.001
E-wom -> City brand university	0.234	0.154	0.079	0.006	−0.383	0.369	0.694
Job opportunity -> City brand university	−0.075	0.112	−0.187	−0.002	−0.312	0.291	0.227
Job opportunity -> Cognitive image	0.195	0.284	−0.089	0.018	−0.387	0.449	0.709
Physical environment -> City brand university	−0.141	0.023	−0.164	−0.010	−0.382	0.341	0.394
Physical environment -> Cognitive image	0.325	0.105	0.220	−0.018	−0.529	0.476	0.459
Rankings Value -> City brand university	0.340*	−0.161 *	0.501 *	0.010	−0.414	0.406	0.014
Rankings Value -> Cognitive image	0.159	0.324	−0.165	−0.003	−0.365	0.371	0.400
Social influence -> Affective image	0.362	0.358	0.003	−0.004	−0.360	0.359	0.991
Social influence -> City brand university	−0.171	0.088	−0.259	−0.001	−0.275	0.298	0.079

(Note: * Significant difference: p-value < 0.05).

Table A5

Differences between Manizales City—Pereira city—based on permutation.

Relation	Path Coefficients Original Manizales City	Path Coefficients Original Medellin city	Path Coefficients Original Difference	Path Coefficients Permutation mean Differences	2.5%	97.5%	Permutation p-Values
Affective image -> City brand university	0.542	0.263	0.280	0.003	−0.302	0.312	0.074
Cognitive image -> City brand university	0.215	0.429	−0.214	−0.004	−0.327	0.344	0.223
Cost life -> City brand university	0.013	−0.156	0.169	0.006	−0.292	0.294	0.284
Cost life -> Cognitive image	−0.067	−0.012	−0.055	−0.001	−0.435	0.447	0.836
E-wom -> Affective image	0.012	0.135	−0.122	−0.006	−0.291	0.268	0.412
E-wom -> City brand university	0.234	0.185	0.049	0.006	−0.296	0.285	0.747
Job opportunity -> City brand university	−0.075	0.023	−0.097	−0.007	−0.304	0.288	0.510
Job opportunity -> Cognitive image	0.195	0.369	−0.174	0.005	−0.401	0.424	0.400
Physical environment -> City brand university	−0.141	0.063	−0.204	−0.007	−0.362	0.345	0.263
Physical environment -> Cognitive image	0.325	0.129	0.196	−0.002	−0.487	0.461	0.462
Rankings Value -> City brand university	0.340	0.120	0.219	0.004	−0.324	0.320	0.188
Rankings Value -> Cognitive image	0.159	0.239	−0.081	−0.009	−0.357	0.354	0.664
Social influence -> Affective image	0.362	0.533	−0.171	−0.001	−0.314	0.300	0.295
Social influence -> City brand university	−0.171	0.000	−0.172	0.002	−0.282	0.285	0.243

Table A6

PLS-MGA Manizales City—Pereira city.

Relation	Path-Difference Coefficients	Original p-Value (1-Tailed)
Affective image -> City brand university	0.280	1.000
Cognitive image -> City brand university	−0.214	1.000
Cost life -> City brand university	0.169	1.000
Cost life -> Cognitive image	−0.055	1.000
E-wom -> Affective image	−0.122	1.000
E-wom -> City brand university	0.049	1.000
Job opportunity -> City brand university	−0.097	1.000
Job opportunity -> Cognitive image	0.100	1.000
Physical environment -> City brand university	−0.204	1.000
Physical environment -> Cognitive image	0.196	1.000
Rankings Value -> City brand university	0.219	1.000
Rankings Value -> Cognitive image	−0.081	1.000
Social influence -> Affective image	−0.171	1.000
Social influence -> City brand university	−0.172	1.000

Table A7

Parametric test—Manizales City—Pereira city.

	Path-Difference Coefficients	t-Value	p-Value
Affective image -> City brand university	0.280	n/a	0.500
Age -> City brand university	−0.066	n/a	0.500
Cognitive image -> City brand university	−0.214	n/a	0.500
Cost life -> City brand university	0.169	n/a	0.500
Cost life -> Cognitive image	−0.055	n/a	0.500
E-wom -> Affective image	−0.122	n/a	0.500
E-wom -> City brand university	0.049	n/a	0.500
Job opportunity -> City brand university	−0.097	n/a	0.500
Job opportunity -> Cognitive image	−0.174	n/a	0.500
Level -> City brand university	0.100	n/a	0.500
Physical environment -> City brand university	−0.204	n/a	0.500
Physical environment -> Cognitive image	0.196	n/a	0.500
Rankings Value -> City brand university	0.219	n/a	0.500
Rankings Value -> Cognitive image	−0.081	n/a	0.500
Social influence -> Affective image	−0.171	n/a	0.500
Social influence -> City brand university	−0.172	n/a	0.500

Table A8

t-Test Welch-Satterhwaite—Manizales City—Pereira city.

	Path-Difference Coefficients	t-Value	p-Value
Affective image -> City brand university	0.280	n/a	0.500
Age -> City brand university	−0.214	n/a	0.500
Cognitive image -> City brand university	0.169	n/a	0.500
Cost life -> City brand university	−0.055	n/a	0.500
Cost life -> Cognitive image	−0.122	n/a	0.500
E-wom -> Affective image	0.049	n/a	0.500
E-wom -> City brand university	−0.097	n/a	0.500
Job opportunity -> City brand university	−0.174	n/a	0.500
Job opportunity -> Cognitive image	−0.204	n/a	0.500
Level -> City brand university	0.196	n/a	0.500
Physical environment -> City brand university	0.219	n/a	0.500
Physical environment -> Cognitive image	−0.081	n/a	0.500
Rankings Value -> City brand university	−0.171	n/a	0.500
Rankings Value -> Cognitive image	−0.172	n/a	0.500
Social influence -> Affective image	0.280	n/a	0.500
Social influence -> City brand university	−0.214	n/a	0.500

Table A9

Differences between Manizales City—Bogota city—Based on permutation.

Relation	Path Coefficients Original Manizales City	Path Coefficients Original Medellin City	Path Coefficients Original Difference	Path Coefficients Permutation Mean Differences	2.5%	97.5%	Permutation p-Values
Affective image -> City brand university	0.542	0.291	0.251	−0.014	−0.346	0.319	0.137
Cognitive image -> City brand university	0.215	0.193	0.022	0.012	−0.338	0.346	0.891
Cost life -> City brand university	0.013	−0.038	0.051	0.013	−0.230	0.266	0.717
Cost life -> Cognitive image	−0.067	0.078	−0.145	−0.002	−0.420	0.366	0.479
E-wom -> Affective image	0.012	0.297	−0.285	−0.003	−0.297	0.282	0.055
E-wom -> City brand university	0.234	0.139	0.095	0.002	−0.264	0.280	0.535
Job opportunity -> City brand university	−0.075	0.119	−0.194	0.002	−0.266	0.286	0.164
Job opportunity -> Cognitive image	0.195	0.229	−0.034	0.002	−0.369	0.368	0.843
Physical environment -> City brand university	−0.141	−0.064	−0.077	−0.010	−0.318	0.287	0.625
Physical environment -> Cognitive image	0.325	0.043	0.282	−0.002	−0.362	0.351	0.142
Rankings Value -> City brand university	0.340	0.378	−0.038	−0.002	−0.300	0.316	0.801
Rankings Value -> Cognitive image	0.159	0.262	−0.103	0.007	−0.334	0.312	0.546
Social influence -> Affective image	0.362	0.311	0.050	0.006	−0.242	0.260	0.684
Social influence -> City brand university	−0.171	0.003	−0.174	0.004	−0.241	0.271	0.169

Table A10

PLS-MGA Manizales City—Bogota city.

Relation	Path-Difference Coefficients	Original p-Value (1-Tailed)
Affective image -> City brand university	0.251	0.061
Cognitive image -> City brand university	0.022	0.443
Cost life -> City brand university	0.051	0.372
Cost life -> Cognitive image	−0.145	0.741
E-wom -> Affective image	−0.285	0.973
E-wom -> City brand university	0.095	0.253
Job opportunity -> City brand university	−0.194	0.921
Job opportunity -> Cognitive image	−0.034	0.567
Physical environment -> City brand university	−0.077	0.662
Physical environment -> Cognitive image	0.282	0.126
Rankings Value -> City brand university	−0.038	0.600
Rankings Value -> Cognitive image	−0.103	0.729
Social influence -> Affective image	0.050	0.359
Social influence -> City brand university	−0.174	0.894

Table A11

Parametric test—Manizales City—Bogota city.

	Path-Difference Coefficients	t-Value	p-Value
Affective image -> City brand university	0.251	1.421	0.157
Age -> City brand university	0.022	0.127	0.899
Cognitive image -> City brand university	0.051	0.361	0.719
Cost life -> City brand university	−0.145	0.709	0.479
Cost life -> Cognitive image	−0.285 *	1.986	0.048*
E-wom -> Affective image	0.095	0.656	0.513
E-wom -> City brand university	−0.194	1.407	0.161
Job opportunity -> City brand university	−0.034	0.179	0.858
Job opportunity -> Cognitive image	−0.077	0.459	0.646
Level -> City brand university	0.282	1.282	0.201
Physical environment -> City brand university	−0.038	0.247	0.805
Physical environment -> Cognitive image	−0.103	0.618	0.537
Rankings Value -> City brand university	0.050	0.345	0.730
Rankings Value -> Cognitive image	−0.174	1.291	0.198
Social influence -> Affective image	0.251	1.421	0.157
Social influence -> City brand university	0.022	0.127	0.899

(Note: * Significant difference: p-value < 0.05).

Table A12

t-Test Welch–Satterhwaite—Manizales City—Bogota city.

	Path-Difference Coefficients	t-Value	p-Value
Affective image -> City brand university	0.251	1.532	0.128
Age -> City brand university	0.022	0.135	0.893
Cognitive image -> City brand university	0.051	0.346	0.730
Cost life -> City brand university	−0.145	0.654	0.515
Cost life -> Cognitive image	−0.285	1.892	0.062
E-wom -> Affective image	0.095	0.665	0.507
E-wom -> City brand university	−0.194	1.418	0.159
Job opportunity -> City brand university	−0.034	0.176	0.860
Job opportunity -> Cognitive image	−0.077	0.440	0.661
Level -> City brand university	0.282	1.173	0.244
Physical environment -> City brand university	−0.038	0.245	0.807
Physical environment -> Cognitive image	−0.103	0.615	0.540
Rankings Value -> City brand university	0.050	0.330	0.742
Rankings Value -> Cognitive image	−0.174	1.231	0.221
Social influence -> Affective image	0.251	1.532	0.128
Social influence -> City brand university	0.022	0.135	0.893

References

1. Brandt, C.; De Mortanges, C.P. City Branding: A Brand Concept Map Analysis of a University Town. Place Brand. Public Dipl.; 2011; [DOI: https://dx.doi.org/10.1057/pb.2010.37]

2. Zenker, S.; Braun, E.; Petersen, S. Branding the Destination versus the Place: The Effects of Brand Complexity and Identification for Residents and Visitors. Tour Manag.; 2017; 58, pp. 15-27. [DOI: https://dx.doi.org/10.1016/j.tourman.2016.10.008]

3. Maitland, R. How Can We Manage the Tourist-Historic City? Tourism Strategy in Cambridge, UK, 1978–2003. Tour Manag.; 2006; 27, pp. 1262-1273. [DOI: https://dx.doi.org/10.1016/j.tourman.2005.06.006]

4. Braun, E.; Kavaratzis, M.; Zenker, S. My City—My Brand: The Role of Residents in Place Branding. Proceedings of the 50th European Regional Science Association (ERSA) Congress; Jönköping, Sweden, 19–23 August 2010.

5. Hospers, G.J. A Short Reflection on City Branding and Its Controversies. Tijdschrift voor Economische en Sociale Geografie; 2020; 111, pp. 18-23. [DOI: https://dx.doi.org/10.1111/tesg.12386]

6. Goovaerts, P.; Van Biesbroeck, H.; Van Tilt, T. Measuring the Effect and Efficiency of City Marketing. Procedia Econ. Financ.; 2014; 12, pp. 191-198. [DOI: https://dx.doi.org/10.1016/S2212-5671(14)00335-9]

7. Oguztimur, S.; Akturan, U. Synthesis of City Branding Literature (1988–2014) as a Research Domain. Int. J. Tour. Res.; 2016; 18, pp. 357-372. [DOI: https://dx.doi.org/10.1002/jtr.2054]

8. European University Association. Universities without Walls—A Vision for 2030; European University Association: Brussels, Belgium, 2021.

9. Glasson, J. The Widening Local and Regional Development Impacts of the Modern Universities—A Tale of Two Cities (and North-South Perspectives). Local Econ.; 2003; 18, pp. 21-37. [DOI: https://dx.doi.org/10.1080/0269094032000073799]

10. Popescu, A. Branding Cities as Educational Centres. The Role of Higher Education Institutions. Manag. Şi Mark. (Bucureşti); 2012; 7, 493.

11. Hospers, G.-J. Creative cities: Breeding places in the knowledge economy. Knowl. Technal. Policy; 2003; 16, pp. 143-162. [DOI: https://dx.doi.org/10.1007/s12130-003-1037-1]

12. Mogaji, E. Types and Location of Nigerian Universities. Res. Agenda Work. Pap.; 2019; 1, pp. 92-103. [DOI: https://dx.doi.org/10.31124/advance.9722618]

13. Lazzeroni, M.; Piccaluga, A. Beyond ‘Town and Gown’: The Role of the University in Small and Medium-Sized Cities. Ind. High. Educ.; 2015; 29, pp. 11-23. [DOI: https://dx.doi.org/10.5367/ihe.2015.0241]

14. Lundholt, M.W.; Jørgensen, O.H.; Blichfeldt, B.S. Intra-Organizational Brand Resistance and Counter-Narratives in City Branding—A Comparative Study of Three Danish Cities. Qual. Mark. Res.; 2020; 23, pp. 1001-1018. [DOI: https://dx.doi.org/10.1108/QMR-01-2018-0012]

15. Charles, D. The Role of Universities in Building Knowledge Cities in Australia. Built Environ.; 2011; 37, pp. 281-298. [DOI: https://dx.doi.org/10.2148/benv.37.3.281]

16. Hospers, G.-J. Governance in innovative cities and the importance of branding. Innov. Manag. Policy Pract.; 2008; 10, pp. 224-234. [DOI: https://dx.doi.org/10.5172/impp.453.10.2-3.224]

17. Abbas, S.A. Brand Loyalty of Higher Education Institutions. Mark. Manag. Innov.; 2019; 1, pp. 46-56. [DOI: https://dx.doi.org/10.21272/mmi.2019.1-04]

18. Hemsley-Brown, J.; Melewar, T.; Nguyen, B.; Wilson, E.J. Exploring brand identity, meaning, image, and reputation (BIMIR) in higher education: A special section. J. Bus. Res.; 2016; 69, pp. 3019-3022. [DOI: https://dx.doi.org/10.1016/j.jbusres.2016.01.016]

19. Van Noorden, R. Cities: Building the best cities for science. Nature; 2010; 467, pp. 906-908. [DOI: https://dx.doi.org/10.1038/467906a]

20. Oliva, E.J.D.; Sánchez-Torres, J.A.; Niño, J.S.S. City brand evolution research and future agenda: Bibliometric and science mapping analysis. Heliyon; 2022; 8, e10031. [DOI: https://dx.doi.org/10.1016/j.heliyon.2022.e10031] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35991998]

21. James-MacEachern, M.; Yun, D. Exploring factors influencing international students’ decision to choose a higher education institution: A comparison between Chinese and other students. Int. J. Educ. Manag.; 2017; 31, pp. 343-363. [DOI: https://dx.doi.org/10.1108/IJEM-11-2015-0158]

22. Shamsudin, M.F.; Ali, A.M.; Ali, A.M.; Shabi, K.S. Exploratory study of students’ decision for enrolment at universiti kuala lumpur business school campus. Humanit. Soc. Sci. Rev.; 2019; 7, pp. 526-530. [DOI: https://dx.doi.org/10.18510/hssr.2019.7262]

23. Ananda, T.N.; Rini, E.S.; Karina, B.; Sembiring, F. Determinants of Students ’ Decision to Choose the Faculty of Economics and Business Harapan University, Medan. Int. J. Res. Rev.; 2019; 6, pp. 33-45.

24. Gai, L.; Xu, C.; Pelton, L.E. A netnographic analysis of prospective international students’ decision-making process: Implications for institutional branding of American universities in the emerging markets. J. Mark. High. Educ.; 2016; 26, pp. 181-198. [DOI: https://dx.doi.org/10.1080/08841241.2016.1245233]

25. Kingeski, L.; Nadal, J.O. Brazilian university students in Spain: Motivations and decision factors. Intang. Cap.; 2020; 16, pp. 91-106. [DOI: https://dx.doi.org/10.3926/ic.1651]

26. Popov, K. Factors, Affecting Students’ Decision to Enroll in a University. Pedagog. J. Pendidik.; 2019; 8, pp. 201-210. [DOI: https://dx.doi.org/10.21070/pedagogia.v8i2.2231]

27. Williams-Brown, Z.; Rhoades, G.; Smith, M.; Thompson, D. Aspiring to Higher Education? Choice, Complexity and Confidence in Secondary Students’ Decision-Making. Educ. Futures; 2019; 10, pp. 31-58.

28. Hendriana, E.; Awang, K.W.; Yusof, R.N.R. The roles of the country’s cognitive, affective, and personality aspects in international students’ decision-making. J. Mark. High. Educ.; 2021; 30, pp. 1-18. [DOI: https://dx.doi.org/10.1080/08841241.2021.1900486]

29. Sarstedt, M.; Hair, J.F., Jr.; Ringle, C.M. “PLS-SEM: Indeed a silver bullet”—Retrospective observations and recent advances. J. Mark. Theory Pract.; 2022; 30, pp. 1-15. [DOI: https://dx.doi.org/10.1080/10696679.2022.2056488]

30. Hair, J.F.; Risher, J.J.; Sarstedt, M.; Ringle, C.M. When to use and how to report the results of PLS-SEM. Eur. Bus. Rev.; 2019; 31, pp. 2-24. [DOI: https://dx.doi.org/10.1108/EBR-11-2018-0203]

31. Henseler, J.; Ringle, C.M.; Sarstedt, M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. J. Acad. Mark. Sci.; 2014; 43, pp. 115-135. [DOI: https://dx.doi.org/10.1007/s11747-014-0403-8]

32. Chin, W.; Cheah, J.-H.; Liu, Y.; Ting, H.; Lim, X.-J.; Cham, T.H. Demystifying the role of causal-predictive modeling using partial least squares structural equation modeling in information systems research. Ind. Manag. Data Syst.; 2020; 120, pp. 2161-2209. [DOI: https://dx.doi.org/10.1108/IMDS-10-2019-0529]

33. Sarstedt, M.; Ringle, C.M.; Cheah, J.-H.; Ting, H.; Moisescu, O.I.; Radomir, L. Structural model robustness checks in PLS-SEM. Tour. Econ.; 2020; 26, pp. 531-554. [DOI: https://dx.doi.org/10.1177/1354816618823921]

34. Henseler, J.; Ringle, C.M.; Sarstedt, M. Testing measurement invariance of composites using partial least squares. Int. Mark. Rev.; 2016; 33, pp. 405-431. [DOI: https://dx.doi.org/10.1108/IMR-09-2014-0304]

35. Shamsudin, M.F.; Ali, A.M.; Ab Wahid, R.; Saidun, Z. Factors influence undergraduate students’ decision making to enroll and social media application as an external factor. Humanit. Soc. Sci. Rev.; 2019; 7, pp. 126-136. [DOI: https://dx.doi.org/10.18510/hssr.2019.7116]

36. Fornell, C.; Larcker, D.F. Structural Equation Models with Unobservable Variables and Measurement Error: Algebra and Statistics. J. Mark. Res.; 1981; 18, pp. 382-388. [DOI: https://dx.doi.org/10.1177/002224378101800313]