2.1 Brand equity

Brand equity is defined as the differential effect of brand knowledge over consumers’ responses to organizational marketing activities (^{Keller & Brexendorf, 2019}). According to this definition, these responses are related to consumer perceptions, attitudes, and behaviors towards any brand, and they become more positive when the consumer is familiarized with a brand and has built unique, strong, and favorable associations with it (^{Keller & Brexendorf, 2019}). Many authors identify brand equity as a multidimensional construct (^{Buil et al., 2013}; ^{Mourad et al., 2020}). ^{Keller & Brexendorf (2019)} assert that brand knowledge is based on two dimensions: brand awareness and brand image. On the other hand, ^{Aaker (1991)} suggests four dimensions from the consumers’ perspective: brand awareness, brand associations, perceived quality, and loyalty.

^{Salinas (2009)} adds that brand equity can also be conceived from a firm’s perspective, summarizing definitions from various authors that propose that brand equity corresponds to the differential financial value generated from any branded good/service in comparison to the same good/service, but unbranded. Consumer and company perspectives are linked since favorable consumer perceptions, attitudes, and behaviors regarding a brand translate into more sales, earnings, and financial value for the brand and the firm that owns it (^{Kumar et al., 2021}).

^{Mourad et al. (2020)} explain that developing brand equity in universities, achieving a differential image and a high-quality service perception, reduces current students’ and applicants’ perceived risk regarding the service, positively affecting university enrollment. In addition, HEIs offering STEM degrees could achieve a differential perceived value among prospective students compared to other HEIs, as these programs prepare graduates for high-demand and better-paid jobs (^{Li, 2020}), which in turn are desired criteria for choosing a university (^{Mourad et al., 2020}). Furthermore, universities could benefit from more funding if they offer STEM degrees and achieve higher graduation rates in these fields, as governments in several countries have made it a priority to encourage and increase the number of STEM students in higher education (^{Li, 2020}).

Previous research has addressed various facets of brand equity and the selection process of higher education institutions, as summarized in Table 1. Scholars have analyzed the relationship of various brand equity dimensions with loyalty (^{Pinar et al., 2020}) and overall brand equity (^{Pinar et al., 2014}; ^{Khoshtaria et al., 2020}). Other research has been conducted to identify the determinants of preference, willingness to pay a price premium , and choice intention (^{Aggarwal et al., 2013}), as well as the determinants of attitudes and enrollment intentions (^{Nagoya et al., 2021}).

^{Royo-Vela and Hünermund (2016)} have examined how marketing efforts affect key aspects such as awareness, trust, and attitude formation toward HEIs. ^{Nuseir and El Refae (2022)} identify the dimensions underlying the selection of a university, although they do not provide a process model. Previously, a comprehensive framework has been proposed that encompasses the decision-ma-king process of international students, considering factors such as institutional image and program evaluation as determinants in this context (^{Cubillo et al., 2006}). ^{Mourad et al. (2020)} , in their comparative study between the USA and Egypt on brand equity dimensions, conclude that the determinants of brand equity reported in the literature may vary depending on the higher education industry maturity, as well as the country and cultural contexts. Therefore, more research on this topic is recommended in new contexts to identify those characteristics of the higher education service that are relevant for potential students’ decision making and the HEI’s brand value.

2.2 Brand valuation approaches and models

Brands have become very valuable assets for companies (^{Kumar et al., 2021}), representing, on average, 20% and 17% of equity and total firm market values, respectively (^{Knowles, 2017}).

There are three approaches to calculating a brand’s financial value: cost, market, and income approaches (^{International Organization for Standardization [ISO], 2010}; ^{Salinas, 2009}). The cost approach proposes that brand value corresponds to the cumulated historical costs derived from the brand’s creation, or the estimated costs that must be assumed to develop a similar brand currently. The market approach calculates a brand value from market values observed from similar brands traded in a relatively brief time. Finally, the income approach determines the brand’s value by calculating the present value of the economic income attributable to the remaining useful economic life of the brand, using a discount rate adjusted for the risk of the intangible asset.

The income approach is the most accepted, as it includes the brand’s capability to generate future earnings (^{Macías et al., 2021}). In contrast, the cost approach is criticized as it does not contemplate any potential future benefit that the brand would generate nor consider any added value for the brand’s management besides historical incurred costs (^{Salinas, 2009}). On the other hand, the market approach’s criticism lies in the difficulty of finding transactions for brands with similar characteristics within a short period (^{ISO, 2010}).

Many models have been developed under the three mentioned approaches. ^{Macias et al. (2021)} compare the models detailed by ^{Salinas (2009)} with the one proposed by SENADI in Ecuador, concluding that most models are based on the income approach and contain a marketing component that seeks to estimate a brand strength factor, which is mainly based on CBBE. ^{Macias et al. (2021)} also found that few models include a brand’s legal risk assessment. According to the ISO norm, there must be a connection between marketing, legal and financial components. For example, the ^{Interbrand (2022)} model contains a factor named Brand Strength (BS), which measures CBBE variables (also called external factors) and internal factors, including brand legal protection. Interbrand asserts that for a higher BS, a lower discount rate must be applied to brand economic revenues, which results in a higher financial value. The ^{Brand Finance (2022)} model also contains a Brand Strength Index (BSI), which also includes a component related to brand equity. Brand Finance affirms that a stronger BSI leads to an increase in brand-attributable income (applying a higher royalty rate), reducing the discount rate and increasing brand value. The valuation model proposed by SENADI in Ecuador is also framed within the income approach and is described below.

2.3 Official brand valuation model in Ecuador and study variables

On February 27^th, 2015, SENADI issued the Methodological Manual for Intellectual Property Intangible Assets Valuation (MVAIPI for its acronym in Spanish) by resolution No. 095-2014-DE-IEPI, published on Ecuadorian Official Record N°277. The model follows the ISO 10668 norm, which in turn is conceptually grounded on brand equity literature and is composed of the following elements:

This study delves into the marketing component of the MVAIPI, analyzing whether the proposed variables are related to the applicants’ desired behaviors, such as the choice of the HEI and its recommendation to third parties. These behaviors are closely related to the revenues and cash flow that a brand can generate.

2.3.3 Choice process and its determinants

Generally, the consumer evaluates the competing brands within the consideration set preceding a purchase decision. Consumers can elaborate on the evaluations for that decision making, or the assessment can be based on pre-existing evaluations (^{Blackwell et al., 2006}). Some processes are more elaborated, while others may be impulsive, which could be explained by the level of product involvement (^{Macias et al., 2021}), defined as the product category’s relevance or importance for the consumer (^{Coulter et al., 2003}), determined by the extent the consequences related to the purchase or choice of that product connect with the personal needs, goals or values (^{Laaksonen, 1994}). And more involved consumers use more criteria when deciding, seek more information, accept fewer alternatives, and process relevant information in greater detail (^{Baker et al., 2002}).

When evaluating the MVAIPI in the agricultural sector, ^{Macías et al. (2021)} show that the evaluation of purchase determinants was significant in explaining repurchase intention in those products purchased less frequently and representing greater monetary investment, while in products of routine purchase and lower monetary outlay, the purchase decision was not based on a detailed evaluation of determinants, but was guided by the brand in the Top of Mind. According to the MVAIPI, the determinants of choice may vary according to the product or service category.

In the higher education sector, ^{Royo-Vela and Hünermund (2016)} describe the decision process of students in three stages: (1) predisposition to brands due to personal and environmental factors (family, peers), (2) search, where the marketing efforts of HEIs have an influence, and (3) evaluation, where the attributes or determinants of the choice of HEIs are analyzed. Among the main determinants commonly mentioned in several studies are:

• Academic quality: program quality and recognition by society and future employers (^{Aggarwal et al., 2013}; ^{Cubillo et al., 2006}; ^{Mourad et al., 2020}; ^{Nuseir & El Refae, 2022}), quality of teachers (^{Cubillo et al., 2006}), appearance in higher education rankings (^{Royo-Vela & Hünermund, 2016}).

• Overall brand assessment: perceived brand value by prospective or current students (^{Aggarwal et al., 2013}).

• Location and infrastructure quality (^{Aggarwal et al., 2013}; ^{Cubillo et al., 2006}; ^{Nuseir & El Refae, 2022}).

• Cost, funding, or perceived value for the cost (^{Aggarwal et al., 2013}; ^{Cubillo et al., 2006}; ^{Nuseir & El Refae, 2022}).

The MVAIPI proposes that brand awareness interacts with brand associations to determine the BF. This means that higher brand awareness will result in a stronger effect of brand association on consumer behavior. Then, the consumers’ evaluation of choice determinants would directly influence their behavior (Figure 1).

Figure 1 MVAIPI Brand Factor (^{SENADI, 2015}). 

2.4 Enrollment and recommendation intention

Deciding to enroll in an HEI and recommending it to others increase the brand’s consumer base, impacting its revenues, cash flow, and, ultimately, its financial value. Therefore, it is argued that these behaviors would be good proxies of brand value (^{Macías et al., 2021}). In this study, the intention of applicants to carry out these behaviors is analyzed, considering that intention is a good predictor of behavior (^{Li et al., 2020}).

Brand awareness plays an important role in consumer decision making for several reasons, according to ^{Keller (1993)} . First, the greater the brand awareness, the more likely it is to be spontaneously remembered and part of the consideration set during the decision to choose an HEI or to recommend it to others. Second, there may be situations in which brand awareness can influence the choice of a brand within the consideration set, even when there are no other brand associations. ^{Macias et al. (2021)} found this phenomenon in brands of agricultural inputs with high purchase frequency and low cost, where being Top of Mind was the predominant factor in choosing a brand, while associations had no significant effect. However, we do not consider this to be as likely in the decision to enroll in an HEI, as it is described as a complex decision process involving cognitive, emotional, and rational aspects that demands effort and time from the applicant during the consideration and evaluation of alternatives (^{Stein et al., 2011}).

On the other hand, brand awareness functions as a signal that the brand has been present in the market for a long time and has achieved success (^{Aaker, 1991}). ^{Aggarwal et al. (2013)} show that brand familiarity (de-fined as brand awareness) influences the willingness to pay a premium in the cost of an HEI. Other studies report a positive and significant association between brand awareness and an overall brand equity construct reflecting the HEI reputation among students (^{Pinar et al., 2014}; ^{Khoshtaria et al., 2020}), as well as with attitude to being loyal (Pinar et al., 2020). With this background, the following hypotheses are proposed:

H1.1: Brand awareness is positively associated with the intention to enroll in an HEI.

H2.2: Brand awareness is positively associated with the intention to recommend an HEI.

Another argument in favor of brand awareness, agreed upon by ^{Keller & Brexendorf (2019)} and ^{Aaker (1991)} , is that it influences consumer decision making by affecting the formation and strength of associations. The better established the brand node is in the memory, the more accessible the brand associations are to the consumer. Precisely, this is what MVAIPI argues to propose an interaction effect between brand awareness and associations, which determines favorable behaviors. Therefore, the hypotheses are:

H2.1: The interaction between brand awareness and associations positively relates to the intention to enroll in an HEI.

H2.2: The interaction between brand awareness and associations positively relates to the intention to recommend an HEI.

Brand associations shape the image of the HEI in the consumer’s mind. These associations may refer to service attributes, sensations, or consumer experiences (^{Aaker, 1991}; ^{Keller & Brexendorf, 2019}). The more associations are stored in the consumer’s mind, the stronger the brand is and the more likely it is to recall it in different situations, such as when choosing an HEI or recommending it to others (^{French & Smith, 2013}). ^{Nagoya et al. (2021)} show that the perception of quality and the overall image of the university influence the affective attitude as a mediator of the intention to enroll in the HEI. ^{Walter et al. (2022)} conclude that a positive brand image leads potential students to perceive a differential value with respect to competitors and attracts them to their classrooms. According to these theoretical and empirical antecedents, it is hypothesized that:

H3.1: Better brand associations are related to a greater intention to enroll in an HEI.

H3.2: Better brand associations are related to a greater intention to recommend an HEI.

As previously explained, choice determinants are those relevant factors of a product or service category that consumers evaluate to make their purchase decision (^{Royo-Vela & Hünermund, 2016}). If a brand is well evaluated on choice determinants, a preference is generated in the consumer’s mind, which triggers desired behaviors such as choice, willingness to pay a price premium or recommending the brand to others (^{Buil et al., 2013}).

^{Cubillo et al. (2006)} propose choice factors such as the prestige of the HEI, the quality of the teachers, the quality and fit of the curriculum to the student’s needs, the infrastructure of the HEI, and the tuition cost. Considering that the focus of Cubillo et al.’s study is on international students, factors such as cultural proximity, the image of the city, and the cost of living are also added. In an empirical study, ^{Pinar et al. (2014)} show positive and significant correlations between factors such as HEI reputation, environment, HEI infrastructure and services, and expected professional development with an overall brand equity construct. Along the same lines, ^{Khoshtaria et al. (2020)} show that, beyond brand awareness and brand image, specific attributes of the HEI infrastructure (physical infrastructure and services such as library and canteens) and the expected future development of graduates influence the overall evaluation of the HEI. ^{Aggarwal et al. (2013)} show that curriculum recognition is a relevant determinant of choice in most of the universities included in their study and, to a lesser extent, value for money paid. Finally, ^{Nuseir and El Refae (2022)} state that academic reputation, funding, location and proximity, and university facilities and services are factors students analyze when choosing a university. Therefore, the hypotheses are:

H4.1: There is a positive relationship between the evaluation of choice determinants and the intention to enroll in an HEI.

H4.2: There is a positive relationship between the evaluation of choice determinants and the intention to recommend an HEI.

3.1 Research design and methods

A mixed design was used: exploratory and conclusive. The purpose of the exploratory phase was to identify the determinants of HEI choice according to the target audience, defined as potential students of STEM pro-grams in the coastal region of Ecuador. For this purpose, interviews were conducted with a small sample of thirteen prospective students.

The purpose of the concluding phase was to test the study hypotheses by applying regression models to the dependent variables, intention to enroll and intention to recommend the HEI, using as independent variables brand awareness (H1.1 and H1.2), the interaction between brand awareness and associations (H2.1 and H2.2), brand associations (H3.1 and H3.2), the evaluation of the choice determinants (H4.1 and H4.2), and the following control variables to capture the heterogeneity of the respondents: type of school, city of residence, monthly family income, gender, and age.

To collect the data, an Internet survey was applied to high school graduates aspiring to enroll in an HEI in the coastal region. Convenience sampling was performed, obtaining a final sample of 227 respondents. Convenience sampling is justified in this case, considering the limitation to determining the sampling frame, given that there was no access to high school graduates’ databases. The HEI used as the target brand is a polytechnic school located in the coastal region.

3.2 Variables measurement

Following MVAIPI guidelines, brand awareness was measured with three consecutive questions: (1) "Mention the first higher education institution that comes to your mind" (if the respondent does not mention the HEI under study, the next question is shown). (2) "Mention other higher education institutions that also come to your mind" (if the respondent does not mention the HEI under study, the third question is shown). (3) "Do you know that [HEI brand name] is a higher education institution? " (YES/NO). There were no observations at the fourth level of brand awareness, i.e., there were no respondents who did not mention the HEI in the first two questions and answered NO in the third question.

For the estimation of the models, we used two dichotomous variables or dummies (D_TOM and D_RECALL) representing the first three levels of brand awareness: D_TOM=1 if HEI was mentioned in the first question, D_RECALL=1 if HEI was mentioned in the second question. When both dummies take the zero value, the observation corresponds to the base case defined as recognition (third level of awareness). Since there was no observation at the fourth level, a third dummy was not necessary.

With respect to associations (ASSO), the MVAIPI reflects the conceptualization presented in this study, contemplating that consumers’ open-ended expressions about a brand can be favorable, neutral, or unfavorable and can be specific or general. In this sense, an open-ended question where the respondent is asked to express what comes to mind when hearing the brand name was included. A qualitative analysis was conducted to identify the valence and extremism of the responses, based on the ^{Rocklage and Fazio (2015)} study. Two researchers coded each response between 1 and 5, where 1 means “Extremely negative” and 5 means “Extremely positive” (3=Neutral). When coders disagreed, the response was submitted to a third researcher’s opinion.

HEI’s choice determinants identified in the exploratory study (n=13) were cost, quality of education, proximity to the applicant’s home, future employment opportunities, international reputation, and physical infrastructure. In the survey, respondents were asked to assess the brand on each determinant with a score ranging from 1 to 4 (1=Very bad; 4=Very good). For the regression models’ estimation, a variable EVAL was coded, averaging the scores that each respondent gave to the six choice determinants.

Intention to enroll and recommend the HEI were measured with the questions "Would you apply to study at [HEI brand name]?" and "Would you recommend someone you know to apply to study at [HEI brand name]?", respectively, using 5-point Likert scales, where 1=Definitely no, and 5=Definitely yes.

3.3 Regression models

The regression models were estimated using generalized least squares (GLS) with robust standard errors for heteroscedasticity using the statistical soft-ware STATA (version 15.1). Equations (1) and (2) show the empirical models for testing this study’s hypotheses:

(1)

(2)

Where β₁ and β₂ represent H1.1. For H1.1 to be accepted, both β₁ and β₂ must be significant and positive, and β₁ ≥ β₂. β₃ and β₄ represent H2.1 (β₃ and β₄ positive; β₃ ≥ β₄). β₅ and β₆ represent H3.1 and H4.1, respectively (β₅ and β₆ positive). In Eq. (2), γ₁ and γ₂ represent H1.2 (γ₁ and γ₂ positive; γ₁ ≥ γ₂); γ₃ and γ₄ represent H2.2 (γ₃ and γ₄ positive; γ₃ ≥ γ₄); γ₅ and γ₆ represent H3.2 and H4.2, respectively (γ₅ and γ₆ positive). β and γ are two coefficient vectors describing the relationships of the control variables (X) with the dependent variables in Eqs. (1) and (2), respectively. Finally, µ_i and e_i are the error terms for both models.

4.1 Descriptive statistics of the sample

Table 2 shows the main descriptive statistics of the variables used in the study. More than one third of the sample has the focal brand in the TOM. Overall, brand associations, evaluation of choice determinants, and intentions have good average levels. It is observed that the intention to enroll in the HEI is lower than the intention to recommend it. Regarding the control variables, the average age of the applicants is close to 18 years, most of them come from private schools, the family income is mainly in the two lower ranges, and about 80% of the applicants are from Guayaquil, the main city of the Ecuadorian coast and location of the HEI, and 47% of the respondents are women.

4.2 Models’ estimation

Table 3 provides the results of the model estimates for each dependent variable. To analyze the robustness of the estimates, outcomes with and without control variables are compared. The results are similar for the two dependent variables, and there are no significant changes once the control variables are included. The brand awareness dummies (D_TOM, D_RECALL) are not significant, so H1.1 and H1.2 are rejected. Associations (ASSO) and the interaction of brand awareness with associations are also not significant, contrary to hypotheses H2.1, H2.2, H3.1, and H3.2. In contrast, the evaluation of choice determinants (EVAL) did show a positive and significant relationship with intentions to enroll and recommend HEIs (p < 0.01), supporting H4.1 and H4.2. Additionally, among the control variables, only gender was significant for intention to recommend. Specifically, women are more likely to recommend compared to the rest of the sample. The next section, therefore, moves on to discuss these findings.