Exploring the Preference of Corporations for Sponsorship Motives and the Impact of Sponsorship Motives on Sponsoring Intention in Post-Epidemic Era: Using Two Different Approaches—FPR and SEM

Abstract

The aim of this study is to explore the preference of corporations for sponsorship motives and the impact of sponsorship motives on sponsoring intention in the post-epidemic era of COVID-19. Taking a Taiwanese company as a case study, a total of 60 expert groups comprising 300 respondents (the management of the sampled companies) were surveyed in the post-epidemic period, with data being collected from 60 sampled companies in February–May 2022. Data were analyzed by using two different functional approaches, including fuzzy preference relations (FPR) for the first survey (study 1) and structural equation model (SEM) for the second survey (study 2). Results reveal that corporate image is the most preferential motive of sponsorship and also demonstrates the most significant and positive influence on sponsoring intention. Meanwhile, the measured factor of performing corporate social responsibility (CSR) appears the most correlated with the construct of corporate image. Based on the results, the study can fully fill the gap between sponsorship motives and sponsoring intention in sponsorship knowledge. Additionally, the conjunction of FPR and SEM can also create methodological synergies, namely, enhancing complementary effects and achieving better holistic analysis. Findings also suggest that special attention should be paid to CSR, which plays a pivotal role in affecting the decision of corporations for sponsorship motives and sponsoring intention and, in a post-epidemic era, continuing to develop CSR actions to enhance corporate image can be the best strategy while facing internal and external challenges of implementing sustainable development (SD).

1. Introduction

The epidemic of COVID-19 has lasted for more than two years since early 2020, along with the occurrence of variant virus. During the epidemic period, specific measures, such as quarantine control, home isolation, work suspension, schooling suspension, tour control, and even massive border restriction, have been adopted in many countries for preventing sweeping spread of the epidemic [1]. Under the implementation of these measures, the existing business models of some industries are being challenged, which have made various impacts on current business operation to some extent [1,2].

Sponsorship can be a soft power tool which not only conducts a positive trend for business operation, but also creates a win–win situation for both the sponsors and the sponsored institutions [3,4,5]. Chadwick et al. [6] even emphasized that soft power of sponsorship is the ability to promote the attractiveness of a sponsoring entity by altering the attitudes and behaviors of key target audiences. In such an epidemic, impact on business development, whether sponsorship motives of attempting to obtain win–win results have been changed or even the intentions of sponsoring have been adjusted by the sponsoring corporations, can be an important issue worth noticing.

As revealed from the COVID-19 recovery report of IMI International, a global research and consulting firm, it demonstrates that sponsorship continues to be the second largest marketing communication during the epidemic. In addition, IMI International discovers that intention to attend live events has been growing since pre-pandemic and, in the midst of live events, sports events are up 43%, community events are up 63%, and festival events are up 43% compared to those in the pre-epidemic period, while over 100,000 participants across 39 country markets responded in this survey [7]. This survey result explains that most people are enthusiastic about joining live activities in the post-epidemic era, which facilitates the conduct of sponsorship through live events, such as sports, culture and arts, social community, etc.

In this context of people loving to engage in live activities in a post-epidemic era, it is not surprising that sponsorship events will be evoked to a larger extent than pre-pandemic. Basically, a sponsorship event is explained to be either financial support or support through products and services, which is provided by a sponsoring company or organization. Meanwhile, sponsorship categories approximately include sports, arts, entertainment, festivals, annual fair events, social welfare, education, etc. [8,9,10]. In the past, a large body of literature explored the related issues of sponsorship. There has been, however, little research to investigate the actual sponsorship motives of the sponsors. Therefore, to explore what sponsorship motives corporations may prefer or decide, particularly in a post-epidemic era, is the main purpose of this study.

Moreover, in spite of a multitude of the literature exploring the sponsorship events on effectiveness or sponsorship objectives, no study has yet been carried out that has analyzed the relationships between sponsorship motives and sponsoring intention, which still exists a lacuna in the sponsorship literature up to the present. In order to fully fill the gap, apart from the exploration for the preference of sponsorship motives, this study will examine the impacts of sponsorship motives on sponsoring intention, specifically among Taiwanese companies in the post-epidemic era, which also constitutes the other important motivation for this study. For achieving anticipated results, two different approaches, including fuzzy preference relations (FPR) and structural equation model (SEM), will be used to analyze the preferential sponsorship motives and the causal relationships of sponsorship motives with sponsoring intention. In this study, two research questions (hypotheses) are also proposed to further demonstrate the potential issues we highlight, including what motive is the most preferential sponsorship motive and whether the most preferential motive has the most significant and positive influence on sponsoring intention.

Following the introduction, this study begins with a review on the possible motives generally discussed in the mainstream sponsorship literature. And, based upon literature review, we generalize and propose the noticeable constructs for categorizing sponsorship motives and, further, develop hypotheses. The Study 1 and Study 2 sections separately describe the participants, measuring process, and analyzed results for the two different approaches—FPR and SEM. Finally, this article concludes with the Discussion and Conclusion sections that discuss the importance and contribution of the research and propose suggestions for future research connected to the findings of this study.

2. Theoretical Foundation and Hypotheses Development

2.1. Sponsorship Motive

Motive is meant to be an inner force causing action or driving activation, which is also an intrinsic cause orientating someone to be engaged in the particular activity intended [11]. Following the definition above, sponsorship motive can be directly explained as the internal driving force for sponsors to decide to deploy sponsorship. And, according to the work of Smith et al., sponsorship motive can be regarded as the major reasons or the fundamental elements of the decision to sponsor [12].

As described above, there is a scarcity of research exploring the motive of sponsorship, and the representative work is the article of Slåtten et al. analyzing the sponsorship motive for sports events by qualitative research. In this article, a sponsorship motive matrix is presented, comprising two dimensions: the dimension of external versus internal motives and the dimension of opportunistic versus altruistic motives. And these two dimensions are combined into four quadrants: market motive, society motive, bond motive, and clan motive. Market motive denotes increasing sales among possible existing or new customers; society motive denotes showing responsibility for local community; bond motive denotes building ownership to turnover intention among collaborators and stakeholders; clan motive denotes expressing dedication and care for employees. In this regard, these four motives can be considered to belong to the areas of marketing instrument, media exposure, fulfilling accomplishment, and obtaining partiality from stakeholders [13].

Actually, in contrast to these four motives of sponsorship indicated by Slåtten et al., most articles exploring sponsorship usually focus on sponsorship objectives or sponsorship effectiveness. Essentially, to achieve objectives or effectiveness can be explained as the critical reasons for sponsoring or the motivation of corporations to sponsor. A group of scholars argued that corporations favoring sponsorship projects or events lie in intending to procure their specific objectives, including increasing corporate image and brand promotion, enhancing corporate awareness and corporate perception, strengthening the attention of the public, enlarging market share of products or services and reaching target market, blocking competition, fulfilling the accomplishment of sponsoring, building support and development relations, reinforcing employee relations and motivation, and raising involvement with the local community [14,15,16,17,18,19,20]. It can be seen that these sponsorship objectives almost involve five areas of motives, including corporate image, marketing instrument, media exposure, accomplishment, and obtaining partiality.

Moreover, most corporations launch sponsorship for reaching three kinds of goals on sponsorship effectiveness, including media goals that can create effective media benefit, marketings goal that can help corporations closely contact the target market, as well as extensive business goals that can build or maintain a trustable and popular company image [21,22,23,24]. These three goals appear to be concerned with the motives of media exposure, marketing instrument, and corporate image.

As discussed above, five preferential constructs or dimensions of sponsorship motives can be generalized from the exploration, namely, corporate image, marketing instrument, media exposure, accomplishment, and obtaining partiality. In addition, the relevant arguments or important viewpoints highlighted for each motive will be used as the major factors of measurement to assess each motive construct. Meanwhile, the measured factors will be simultaneously examined in data analysis for FPR and SEM approaches. These elements or measured factors in each motive construct will be discussed in the following sub-sections.

2.1.1. Corporate Image (CI)

CI typically involves reputation, credibility, and goodwill of a corporation or an organization, which is regarded as an intangible asset that can enhance the trust of stakeholders, including customers [10,25]. Earlier studies have confirmed that sponsorship can promote CI. For example, Wang et al. [26] argued that sponsorship events can alter individuals’ perceived event value and further increase the public’s perception and identification to the image of the sponsors. Rajabi et al. [27] demonstrated that different sponsorship-fit profiles influence respondents’ attitudes toward sponsors, especially in enhancing the visibility and credibility of the sponsors.

In addition, creating a better reputation of corporate social responsibility (CSR) can be an important measured factor for the motive of CI [25,28]. Following the importance of sustainable development (SD), more and more corporations combine their CSR strategies with SD to maintain competitive advantages and simultaneously deploy sponsorship to be a major program or project of implementing CSR and, further, to enhance business brand and shape stronger CI [9,29,30].

As previously explored, it can be seen that the measured factors generalized in CI motive construct encompass enhancing the visibility and credibility of the sponsors, increasing the public’s perception and identification to the sponsors and cultivating CSR to shape the corporate image.

2.1.2. Marketing Instrument (MI)

As previously mentioned, sponsorship is the second largest marketing communication, and launching sponsorship in the marketing mix has been developed to be an alternative of MI [31]. There has been consistent support for the factors of using sponsorship as the communication tool of MI. For example, Su and Kunkel [32] analyzed that the enterprises with lesser-known brands can enlarge market share of products or services through effective sponsoring events in a competitive environment. Rosário and Raimundo [33] argued that launching sponsorship can reach the connection with target consumers, which is also proved by Premananto and Zulkifli [34] in exploring the performance of sponsorship events. Zhang [35] indicated that sponsorship-linked marketing is correlated with return on investment, helping corporations maximize sponsorship investment effectiveness.

As previously discussed, it can be seen that the measured factors generalized in MI motive construct encompass enlarging market share of products or services, reaching the connection with target consumers, and maximizing sponsorship investment effectiveness (increasing return on investment).

2.1.3. Media Exposure (ME)

ME in this study represents the achievement of exposure in the media. Previous research has supported that using sponsorship as a tool for creating ME is an important factor that corporations evaluate. For example, Lin and Bruning [36] indicated that different sponsorship contexts can produce different results of media exposure and coverage, especially in the context of media and programing content sponsorship. Cornwell and Kwon [37] argued that sponsorship-linked marketing can influence audience response and further strengthen the attention of the public. Alonso Dos Santos et al. [38,39] argued that specific sponsorship formats can enhance different levels of customers’ recall and recognition of products or services.

Accordingly, the measured factors generalized in ME motive construct can involve increasing media exposure and coverage, strengthening the attention of the public, and enhancing customers’ recall and recognition of products or services.

2.1.4. Accomplishment (AT)

AT means that sponsorship can fulfill sponsors’ accomplishment, particularly in meeting the goals that sponsors orientate [40]. In other words, the goals of sponsorship might be meaningful or worth cherishing for sponsors [41], which is supported by previous studies. For example, Pontes et al. [42] emphasized that understanding the sponsorship articulation–fit relationship can help sponsors effectively develop their favorable events and, further, to show and convey the spirit of the sponsored project or event. Lee [43] explored the sustainable reciprocity mechanism of social initiatives in sport for supporting the continuity of development for the sponsored project or event. Jun and Kim [44] argued that sponsors can create better development relations with the sponsored they favor, further providing assistance to the sponsored project or event as much as they can.

Accordingly, the measured factors generalized in AT motive construct can involve showing and conveying the spirit of the sponsored project or event, supporting the continuity of development for the sponsored project or event, and providing assistance to the sponsored project or event as much as sponsors can for building development relations.

2.1.5. Obtaining Partiality (OP)

OP can be explained as sponsors delivering friendliness, co-operation, and goodwill to stakeholders, and the stakeholders may include employees, communities, suppliers, customers, and any other persons or groups influencing or being affected by the sponsors [45]. For example, Joglekar and Tan [46] argued that sponsorship events may stir employees to have a positive relationship with employers, further reinforcing employee’s centripetal force and satisfaction. Kim et al. [47] explored the challenges of conducting National Olympic Committee (NOC) sponsorship in small states and found that integrating into the local community is identified as one of the key themes for the importance of NOC sponsorship. Thomas et al. [48] analyzed that the perceived sincerity of the sponsoring brand has a significant positive direct effect on motivation of consumers for sports consumption, which demonstrates that showing sincerity and goodwill can establish better relationships with customers.

Accordingly, the measured factors generalized in OP motive construct can involve reinforcing employees’ centripetal force and satisfaction, integrating into the local community, and showing goodwill and establishing a better relationship with customers.

As explored on the previous studies above, some scholars noticed that CI may be the most important motive of sponsorship as the relationship of CSR with CI is particularly highlighted [9,13,29,30]. Considering such an argument of previous studies, one of the hypotheses of this study is formulated as follows:

Hypothesis 1 (H1).

CI is the most preferential sponsorship motive construct amongst these five constructs while assessed by FPR approach.

2.2. Sponsoring Intention

Sponsoring intention in this study can be explained as the behavioral intention to sponsor. Behavioral intention can be defined as an individual’s judgment with subjective probability for a specific behavior, which reflects the willingness of an individual or an organization to adopt or engage in a specific behavior or action [49,50]. As for the measurement of behavioral intention, previous studies argued that it can involve the intention to engage in a target behavior in the short term and the intention to continue to engage in that target behavior in the future [51,52]. In addition, Ajzen [51] emphasized that behavioral intention implies the willingness and eagerness to do something desired. Accordingly, the measured factors of sponsoring intention thereby encompass considering sponsoring again in the short term, continuing to choose sponsorship in the future, and being glad to sponsor.

In this study, the causal relationship between the five motive constructs and sponsoring intention will be examined by SEM approach, and we assume that, under the survey of the same samples, the influential result of the most preferential sponsorship motive will be in line with the assessment of FPR approach in preference priority. This study thereby proposes a second hypothesis below:

Hypothesis 2 (H2).

CI has the most significant and positive influence on sponsoring intention while examined by SEM approach.

3. Study 1 (FPR Approach)

3.1. Method

3.1.1. Pilot Study

Prior to conducting the formal survey of FPR approach, we invited 30 experts (the managers and sub-managers of 10 companies in Kaohsiung city, Taiwan) to launch a pilot test. The results show that Cronbach’s alpha (α) values of all five motive constructs were greater than the acceptable threshold of 0.70 [53], ranging from 0.816 to 0.908 and showing high reliability. According to the results, the contents for each question in the questionnaire are reliable for measurement and do not need to be modified.

3.1.2. Participants

This study conducts a questionnaire method for interviewed experts in the FPR approach, sampling 60 expert groups comprising 300 respondents of the 60 sampled companies in Taiwan. These companies include 20 firms of service industry, 20 firms of manufacturing industry, and 20 subsidiaries from abroad. For each sampled company, five respondents (executives, managers, sub-managers, or section managers) will be selected as an expert group, namely, an expert group is regarded as an evaluator, and an important priority item (one of five motive constructs and its associated measured criteria, see

Table 1. Motive construct and evaluation criteria.

Construct	Evaluation Criteria Description	Sources
CI	• Cultivating CSR to shape better image	[9,29,30]
	• Changing the public’s perception and identification	[9,26]
	• Enhancing the visibility and credibility	[9,27]
MI	• Enlarging market share	[13,32]
	• Reaching the connection with target consumers	[31,33,34]
	• Maximizing sponsorship investment effectiveness	[13,35]
ME	• Increasing media exposure and coverage	[13,36]
	• Strengthening the attention of the public	[13,37]
	• Enhancing customers’ recall and recognition	[13,38,39]
AT	• Conveying the spirit of the sponsored event	[13,42]
	• Supporting the development for the sponsored event	[28,40,43]
	• Providing assistance to the sponsored event	[13,44]
OP	• Reinforcing employees’ centripetal force	[13,46]
	• Integrating into the local community	[13,47]
	• Establishing better relationship with customers	[13,28,48]

) in the questionnaire will be judged and decided together by this expert group. All respondents have been involved in and were knowledgeable about the decision-making process for sponsorship events their companies ever conducted. Data were collected from 60 sampled companies in February–May 2022, and each sampled company was visited one by one to collect back all the questionnaires in person.

3.1.3. Fuzzy Preference Relations (FPR)

For AHP [54] or Fuzzy AHP [55] decision approach, evaluators always need to provide n(n − 1)/2 judgments for a preference matrix with n elements as calculated in judging procedures. To reduce the judgment times, the additive consistent FPR is employed in this study, requiring only n − 1 judgments from a set of n elements. The consistent FPR is proposed by Herrera-Viedma et al. [56] for establishing pairwise comparison preference decision matrices using the so-called reciprocal additive transitivity property. This method not only enables decision makers to express their degree of preference for a set of attributes or alternatives, but also avoids checking the inconsistency in the decision-making process. In addition, for the appropriateness of equation used in FPR, the motive construct of CI is amended to be M₁; MI is amended to be M₂; ME is amended to be M₃; AT is amended to be M₄; and OP is amended to be M₅. Meanwhile, the software of Microsoft Office Excel is used for data processing in the FPR approach.

3.2. Measures

3.2.1. Definition of Linguistic Terms

The ambiguity problems for the assessed elements may arise while judged by an evaluator. For clarifying the linguistic terms, AHP utilizes pairwise comparisons with a 1–9 ratio scale to analyze the relative importance of an element to another [54], and, meanwhile, Fuzzy AHP (FAHP) specifies the reciprocal values according to AHP’s 1–9 ratio scale [55]. Based on the scales of AHP and FAHP (see

Table 2. The AHP and FAHP scales.

Definition of Linguistic Terms	AHP Scale	FAHP Scale
	Intensity of Importance	Reciprocal Values
Equally important (EQ)	1	1
Weakly important (WK)	3	1/3
Strongly important (ST)	5	1/5
Very strongly important (VS)	7	1/7
Absolutely important (AB)	9	1/9
Intermediate values	2, 4, 6, 8	1/2, 1/4, 1/6, 1/8

), nine linguistic terms are presented in this study to provide the evaluators’ simple linguistic terms quantified on a scale of [¹/₉, 9] to express their strength of preference among sponsorship motives.

The nine linguistic terms, namely ‘‘absolutely more important’’, ‘‘very strongly more important’’, ‘‘strongly more important’’, ‘‘weakly more important’’, ‘‘equally important’’, ‘‘less weakly more important’’, ‘‘less strongly more important’’, ‘‘less very strongly more important’’, and ‘‘less absolutely more important’’, are provided for comparing neighboring preferential sponsorship motives corresponding to a real number (see

Table 3. Linguistic terms for priority weights of sponsorship motives.

Definition	Intensity of Importance
Absolutely more important (AB)	9
Between AB and VS—Intermediation (AV)	8
Very strongly more important (VS)	7
Between VS and ST—Intermediation (VT)	6
Strongly more important (ST)	5
Between ST and WK—Intermediation (SW)	4
Weakly more important (WK)	3
Between WK and EQ—Intermediation (WE)	2
Equally important (EQ)	1
Between EQ and LWK—Intermediation (ELW)	1/2
Less weakly more important (LWK)	1/3
Between LWK and LST—Intermediation (LWLS)	1/4
Less strongly more important (LST)	1/5
Between LST and LVS—Intermediation (LSLV)	1/6
Less very strongly more important (LVS)	1/7
Between LVS and LAB—Intermediation (LVLA)	1/8
Less absolutely more important (LAB)	1/9

).

3.2.2. Definitions of Reciprocal Additive Transitivity FPR

The major basic definitions below are utilized in this study, described as follows:

Definition 1.

Assume a fuzzy preference relation M on a set of alternatives X is denoted by a matrix $M\subset X\times X$, which is meant by a membership function: ${\mu }_m$:$X\times X\to \left[0, 1\right]$,$M=\left(M_{ij}\right)$,$M_{ij}={\mu }_m\left(x_i,x_j\right)$ $\forall i,j\in \left\{1,\dots ,n\right\}$.$M_{ij}$ is explained as the preference degree of the alternative $x_i$ over $x_j$. If $M_{ij}=\frac{1}{2}$, this demonstrates that there is no difference between $x_i$ and $x_j\left(x_i~x_j\right)$;$M_{ij}=1$ indicates $x_i$ is absolutely preferred to $x_j$; similarly $M_{ij}=0$ indicates $x_j$ is absolutely preferred to $x_i$; $M>\frac{1}{2}M>\frac{1}{2}$ indicates that $x_j$ is preferred to $x_j(x_i>x_j)$. M is assumed to be additive reciprocal, that is:

$$M_{ij}+M_{ji}=1 \forall i,j\in \left\{1,\right.\cdots \left.,n\right\}$$

(1)

Definition 2.

Suppose there is a set of alternatives $X=\left\{x_1,\cdots ,x_n\right\}$, which is associated with a multiplicative preference relation $A=\left(a_{ij}\right)$ with $a_{ij}\in \left[1/9,9\right]$. Then, the corresponding reciprocal additive fuzzy preference relation $M=\left(M_{ij}\right)$ with $M_{ij}\in \left[0, 1\right]$ to $A=\left(a_{ij}\right)$ is defined as follows:

$$M_{ij}=g\left(a_{ij}\right)=\frac{1}{2}\left(1+{\log }_9{a}_{ij}\right)$$

(2)

Definition 3.

For a reciprocal additive fuzzy preference relation M = (M_ij), the following statements are equivalent:

$$M_{ij}+M_{jk}+M_{ki}=\frac{3}{2}\begin{array}{cc}& \end{array}\forall i<j<k$$

(3)

$$M_{i\left(i+1\right)}+M_{\left(i+1\right)\left(i+2\right)}+\cdots +M_{\left(j-1\right)j}+M_{ji}=\frac{j-i+1}{2}\begin{array}{cc}& \end{array}\forall i<j$$

(4)

If the preference matrix contains any values that are not in the interval $[0,1]$ but, in an interval $[-a,1+a]$, a linear solution is required to preserve the reciprocity and additive transitivity, that is, preference of sponsorship motive M: $[-a,1+a]$→$[0,1]$. Based upon the above statements, a consistent fuzzy preference relation can be constructed by using the following equation:

$$M_{ji}=\frac{j-i+1}{2}-M_{ii+1}-M_{i+1i+2}\cdots -M_{j-1j}$$

(5)

3.2.3. Procedures of Obtaining the Priorities of the Preference for Sponsorship Motives

The following describes the procedures of the reciprocal additive consistent fuzzy preference relation for obtaining the priorities of the preference for sponsorship motives.

(1) Constructing pairwise comparison matrices amongst the preference of motives $M_i\left(i=1,2,\dots ,n\right)$. The evaluators (interviewed expert groups) $E_k\left(k=1,2,\dots ,m\right)$ then are inquired to select which is more important of each two preferential motives for a set of $n-1$ preference values $\left\{a_{12},a_{23},\dots ,a_{n-1n}\right\}$, for example, illustrated below:

$$\begin{gathered} \begin{array}{ccccc}{M}_1& M_2& M_3& \cdots & M_n\end{array} \\ {A}^k=\begin{array}{c}{M}_1\\ M_2\\ M_3\\ ⋮\\ M_n\end{array}\left[\begin{array}{ccccc}1& a_{12}^k& \times & \times & \times \\ \times & 1& a_{23}^k& \times & \times \\ \times & \times & 1& a_{34}^k& \times \\ ⋮& ⋮& ⋮& \ddots & a_{n-1n}^k\\ \times & \times & \times & \times & 1\end{array}\right] \end{gathered}$$

(6)

The $a_{ij}^k$ denotes the preference intensity toward motives i and j assessed by kth evaluator, $a_{ij}^k=1$ means indifference between motives i and j, $a_{ij}^k=3,5,7,9$ expresses that motive i is relatively important to motive j, while $a_{ij}^k=3^{-1},5^{-1},7^{-1},9^{-1}$ indicates that motive i is less important than motive j. The sign “x” means the remaining $a_{ij}^k$, which can be evaluated by inverse comparison methods.

(2) Transforming the preference value $a_{ij}^k$ into $M_{ij}^k$ in an interval scale $[0,1]$, then calculating the remaining $M_{ij}^k$ by using the reciprocal transitivity property, which is illustrated as follows:

$$\begin{gathered} \begin{array}{ccccc}{M}_1& M_2& M_3& \cdots & M_n\end{array} \\ {A}^k\stackrel{\hspace{1em}\frac{1}{2}\left(1+{\log }_9{a}_{ij}\right)\hspace{1em}}{\to }{M}^k=\begin{array}{c}{M}_1\\ M_2\\ M_3\\ ⋮\\ M_n\end{array}\left[\begin{array}{ccccc}0.5& M_{12}^k& \times & \times & \times \\ 1-M_{12}^k& 0.5& M_{23}^k& \times & \times \\ \times & 1-M_{23}^k& 0.5& M_{34}^k& \times \\ ⋮& ⋮& ⋮& \ddots & M_{n-1n}^k\\ \times & \times & \times & \times & 0.5\end{array}\right] \end{gathered}$$

(7)

The $M_{ij}^k=\frac{1}{2}$ denotes indifference between motives i and j, $M_{ij}^k=1$ indicates that motive i is absolutely important to motive j, $M_{ij}^k=0$ represents that motive i is absolutely less important to motive j, and $M_{ij}^k>\frac{1}{2}$ further demonstrates that motive i is preferred to motive j. If this transforming matrix contains any values that are not included in the interval $[0,1]$ but in an interval $[-a,1+a]$, then a transformation function is required to retain the reciprocity and additive transitivity. The transformation function is calculated by the following equation:

$$f\left(M_{ij}^k\right)=\frac{M_{ij}^k+a}{1+2a}$$

(8)

In this equation, “a” refers to the absolute value of the minimum in this transformation preference matrix.

(3) Drawing out the judgments from evaluators to procure the aggregated weights of preferential motives and conducting $M_{ij}^k$ to denote the transformed fuzzy preference value of evaluator k in the process of assessing the motives i and j. The average value equation is used to integrate the judgment values of m evaluators, which is shown below:

$$M_{ij}=\frac{1}{m}\left(M_{ij}^1+M_{ij}^2+\cdots +M_{ij}^m\right)$$

(9)

(4) Normalizing the aggregated fuzzy preference relation matrices and using $r_{ij}$ to denote the normalized fuzzy preference values of each preferential motive. The calculation equation is shown below:

$$r_{ij}=\frac{M_{ij}}{{{\displaystyle \sum }}_{i=1}^n{M}_{ij}}$$

(10)

(5) Given that the $\varpi$ denotes the priority weight of preferential motive i, the priority weight of each motive can be obtained, which is demonstrated below:

$${\varpi }_i=\frac{{{\displaystyle \sum }}_{j=1}^n{r}_{ij}}{{{\displaystyle \sum }}_{i=1}^n{{\displaystyle \sum }}_{j=1}^n{r}_{ij}}$$

(11)

3.3. Results

3.3.1. The Preference Weights for Sponsorship Motives

(1) For clearly explaining the computational process, the assessment of evaluator 1 (T1) is extracted as an example for data analysis. The original fuzzy preference pairwise comparison matrix of evaluator 1 is listed in

Table 4. Original fuzzy preference pairwise comparison matrix of evaluator 1 (T1).

T1	M1	M2	M3	M4	M5
M1	1	AV	X	X	X
M2	LVLA	1	VS	X	X
M3	X	LVS	1	LAB	X
M4	X	X	AB	1	LVS
M5	X	X	X	VS	1

T1: denotes the first interviewed expert group (evaluator) of the sampled companies.

and the linguistic terms of evaluator 1 can be further transferred into corresponding numbers as listed in

Table 5. Translated linguistic terms into corresponding numbers of evaluator 1.

T1	M1	M2	M3	M4	M5
M1	1	8	X	X	X
M2	1/8	1	7	X	X
M3	X	1/7	1	1/9	X
M4	X	X	9	1	1/7
M5	X	X	X	7	1

.

(2) Subsequently, Equation (2) is used to transform the elements which are listed in Table 5 into an interval [0, 1], yielding the following values:

$$\begin{array}{l}{M}_{12}=\frac{1}{2}\left(1+{\log }_{9}8\right)=0.9732, M_{23}=\frac{1}{2}\left(1+{\log }_{9}7\right)=0.9428,\\ M_{34}=\frac{1}{2}\left(1+{\log }_9\frac{1}{9}\right)=0.0000, M_{45}=\frac{1}{2}\left(1+{\log }_9\frac{1}{7}\right)=0.0572.\end{array}$$

The remaining values then can be calculated by Equations (1) and (5). For M₂₁, M₃₁, and M₅₂ as examples:

$$\begin{array}{l}{M}_{21}=1-M_{12}=1-0.9732=0.0268,\\ M_{31}=\frac{3-1+1}{2}-M_{12}-M_{23}=1.5-0.9732-0.9428=-0.4160,\\ M_{52}=\frac{5-2+1}{2}-M_{23}-M_{34}-M_{45}=2-0.9428-0.0000-0.0572=1.0000.\end{array}$$

The fuzzy preference relation matrix of the preference for five sponsorship motives assessed by evaluator 1 can be built in

Table 6. Transformed fuzzy preference values of evaluator 1.

T1	M1	M2	M3	M4	M5
M1	0.5000	0.9732	1.4160	0.9160	0.4732
M2	0.0268	0.5000	0.9428	0.4428	0.0000
M3	−0.4160	0.0572	0.5000	0.0000	−0.4428
M4	0.0840	0.5572	1.0000	0.5000	0.0572
M5	0.5268	1.0000	1.4428	0.9428	0.5000

. Table 6 expresses that four elements have not been listed in the interval [0, 1] and, for ensuring the reciprocity and additive transitivity of the preference relation matrix, a linear transformation stated in Equation (4) is employed to calculate the transformation matrix as listed in

Table 7. Preference values transformed by linear solution for evaluator 1.

T1	M1	M2	M3	M4	M5
M1	0.5000	0.7510	0.9858	0.7206	0.4858
M2	0.2490	0.5000	0.7348	0.4697	0.2348
M3	0.0142	0.2652	0.5000	0.2348	0.0000
M4	0.2794	0.5303	0.7652	0.5000	0.2652
M5	0.5142	0.7652	1.0000	0.7348	0.5000

.

(3) Likewise, the same computational procedures (1) and (2) demonstrated above can calculate the fuzzy preference relation matrices of the other 59 evaluators (expert groups); therefore, using Equation (9), the aggregated pairwise comparison matrix of 60 evaluators can be obtained as listed in

Table 8. Aggregated pairwise comparison matrices of 60 evaluators.

	M1	M2	M3	M4	M5
M1	0.5000	0.5598	0.6270	0.6766	0.6311
M2	0.4402	0.5000	0.5672	0.6169	0.5713
M3	0.3730	0.4328	0.5000	0.5497	0.5041
M4	0.3234	0.3831	0.4503	0.5000	0.4545
M5	0.3689	0.4287	0.4959	0.5455	0.5000
Total	2.0055	2.3044	2.6404	2.8887	2.6610

.

(4) Equation (10) is used to normalize the aggregated pairwise comparison matrix. Taking r₂₁ as an example:

$$r_{21}=0.4402 / 0.5000+0.4402+0.3730+0.3234+0.3689=0.2195$$

(12)

(5) The priority weight of each sponsorship motive can then be obtained by Equation (7). The priority weight and rank of each sponsorship motive assessed by 60 evaluators (expert groups) are listed in

Table 9. Normalized matrix of priority weight and rank of motives for a total 60 evaluators.

	M1	M2	M3	M4	M5	Total	Weight	Rank
M1	0.2493	0.2429	0.2375	0.2342	0.2372	1.2011	0.2402	1
M2	0.2195	0.217	0.2148	0.2136	0.2147	1.0796	0.2159	2
M3	0.1860	0.1878	0.1894	0.1903	0.1894	0.9429	0.1886	3
M4	0.1613	0.1662	0.1705	0.1731	0.1708	0.8419	0.1684	5
M5	0.1839	0.1860	0.1878	0.1888	0.1879	0.9344	0.1869	4
						4.9999	1.0000

.

Therefore, the rank of weight for each sponsorship motive is illustrated as:

$$M_1\left(0.2402\right)\succ M_2\left(0.2159\right)\succ M_3\left(0.1886\right)\succ M_5\left(0.1869\right)\succ M_4\left(0.1684\right)$$

(13)

3.3.2. Final Result

The results demonstrate that the most preferential sponsorship motive is M₁ (CI) (0.2402); M₂ (MI) (0.2159) is the second; M₃ (ME) (0.1886) is the third; meanwhile, the two least preferential sponsorship motives are M₅ (OP) (0.1869) and M₄ (AT) (0.1684). Evidently, the obtained results confirm our first hypothesis.

4. Study 2 (SEM Approach)

4.1. Method

4.1.1. Pilot Study

Prior to conducting the formal survey of SEM approach, we also invited 45 experts (the managers and sub-managers of 15 companies in Kaohsiung city, Taiwan) to launch a pilot test. The results show that Cronbach’s alpha (α) values of the five motive and sponsoring intention constructs were greater than the acceptable threshold of 0.70 [53], ranging from 0.844 to 0.927 and showing high reliability. According to the results, the contents for each question in the questionnaire are reliable for measurement and do not need to be modified.

4.1.2. Participants

This study also conducts a questionnaire method for all the respondents in the SEM approach, sampling 300 respondents from the 60 sampled companies, which are with the same samples (respondents) as the FPR approach. The questionnaire was designed according to the motive constructs and associated measured criteria used in the FPR approach, including six constructs posited by this study, and using 7-point Likert scales (1 = strongly disagree, 7 = strongly agree): M₁ (CI) for three items, M₂ (MI) for three items, M₃ (ME) for three items, M₄ (AT) for three items, M₅ (OP) for three items, and sponsoring intention for three items. All the items of six constructs are observed indicators that will be used to test the relationship between five constructs and sponsoring intention. The original scales of the questionnaire were compiled by Chinese language and forwarded to respondents. Five respondents (regarded as an expert group or an evaluator in FPR survey) of each sampled company were requested to fill out the SEM questionnaires individually. All 300 copies of valid questionnaires were collected in person, while simultaneously conducting FPR questionnaires in February–May 2022.

4.1.3. Measures

Structural equation modeling (SEM) was used to examine the impact of these five sponsorship motives (M₁–M₅) on sponsoring intention and, further, to explore whether the influential result or the impact sequence of the most preferential sponsorship motive will be in line with the assessment of FPR approach in preference priority of study 1. This study uses the software of AMOS Graphics 21.0 version completely operating SEM analysis in a graphical way for data processing, and the confirmatory factor analysis for these six constructs is shown as

Table 10. Confirmatory factor analysis results.

Construct/Items	β	CR	α	AVE
M1—Corporate Image (CI)
CI1: For cultivating CSR to shape the corporate image	0.835 ***	0.864	0.862	0.678
CI2: For changing the public’s perception and identification to my company	0.816 ***
CI3: For enhancing the visibility and credibility of my company	0.820 ***
M2—Marketing Instrument (MI)
MI1: For increasing market share of products or services	0.563 ***	0.796	0.792	0.548
MI2: For fulfilling the connection with target audiences	0.939 ***
MI3: For maximizing sponsorship investment effectiveness (return on investment)	0.667 ***
M3—Media Exposure (ME)
ME1: For increasing media exposure and coverage	0.721 ***	0.875	0.873	0.679
ME2: For attracting the attention of the public	0.802 ***
ME3: For strengthening customers’ recall and recognition of products or services	0.924 ***
M4—Accomplishment(AT)
AT1: For showing and conveying the spirit of the sponsored project or event	0.819 ***	0.854	0.851	0.662
AT2: For supporting the continuity of development for the sponsored project or event	0.853 ***
AT3: For providing assistance to the sponsored project or event as much as we can	0.766 ***
M5—Obtaining Partiality (OP)
OP1: For reinforcing employees’ centripetal force and satisfaction	0.840 ***	0.875	0.873	0.701
OP2: For integrating into the local community	0.840 ***
OP3: For showing goodwill and establishing better relationship with customers	0.831 ***
Sponsoring intention
SI1: I will consider sponsoring again in the short term	0.863 ***	0.887	0.884	0.724
SI2: I will continue to choose sponsorship in the future	0.888 ***
SI3: Overall, I am very glad to do sponsoring	0.800 ***

Note: β = CFA factor loading, CR = composite reliability, α = Cronbach’s alpha, AVE = average variance extracted, *** = p-value less than 0.001.

.

4.2. Results

4.2.1. Measurement Model

This study employed confirmatory factor analysis to assess the validity of the multiple items measuring the constructs of M₁, M₂, M₃, M₄, M₅, and sponsoring intention. Using the maximum likelihood estimation (MLE) method to estimate all parameter values from the data observed, the measurement model represents the six constructs with a good fit to the data (n = 300, Chi-Square/df = 479.201/120 = 3.993, CFI = 0.926, TLI = 0.906, IFI = 0.927, RFI = 0.878, NFI = 0.904, RMSEA = 0.100, see

Table 11. Test for CFA indices of goodness of fit.

Fit Index	Criteria	Test Results	Judgment
χ2	The smaller, the better	479.201
χ2/df	<5	3.993	Conforming
CFI	≥0.90	0.926	Conforming
TLI	≥0.90	0.906	Conforming
IFI	≥0.90	0.927	Conforming
RFI	≥0.80	0.878	Conforming
NFI	≥0.80	0.904	Conforming
RMSEA	≤0.10	0.100	Conforming

). Though root-mean-square error of approximation (RMSEA) appears slightly higher, it is still on the upper acceptable boundary of 0.1 [31,57].

The measurement model by AMOS Graphics is illustrated in Figure 1.

As shown in Table 10, all items were loaded significantly on the constructs (p < 0.001), showing that no offending estimates exist in this measurement model. They ranged from 0.816 to 0.835 for M₁, 0.563 to 0.939 for M₂, 0.721 to 0.924 for M₃, 0.766 to 0.853 for M₄, 0.831 to 0.840 for M₅, and 0.800 to 0.888 for sponsoring intention, and all exceeded the acceptable threshold of 0.50 [58]. In addition, three indicators were generally used to measure convergent validity: Cronbach’s alpha, composite reliability (CR), and average variance extracted (AVE). As shown in Table 10, Cronbach’s alpha (α) values of all six constructs were greater than the acceptable threshold of 0.70 [59], ranging from 0.792 to 0.884 and showing high reliability; CR values of all six constructs were greater than the acceptable threshold of 0.70 [59], ranging from 0.796 to 0.887; AVE values of all six constructs were greater than the acceptable threshold of 0.50 [59], ranging from 0.548 to 0.724. The results indicate that the measurement model has convergent validity and high internal consistency.

For discriminant validity, a group of scholars suggested that if the correlation between latent variables is above the absolute value of 0.7, it is recommended to use the confidence interval (CI) method for estimation [60,61,62]. In this study, M₁, M₂, M₃, M₄, M₅, and sponsoring intention (SI) were all latent variables, and the correlation values of all latent variables in

Table 12. The discriminant validity of 95% CI.

Pairs of Correlation	Estimate	95% CI
		BC		PC		BC/PC p Value
		Lower	Upper	Lower	Upper
M1 <--> M2	0.797 **	0.696	0.875	0.702	0.879	0.002/0.001
M1 <--> M3	0.864 **	0.795	0.914	0.801	0.920	0.002/0.001
M1 <--> M4	0.835 **	0.753	0.891	0.759	0.897	0.002/0.001
M1 <--> M5	0.769 **	0.667	0.843	0.676	0.848	0.002/0.001
M1 <--> SI	0.861 **	0.750	0.937	0.758	0.942	0.002/0.001
M2 <--> M3	0.771 **	0.686	0.837	0.691	0.840	0.001/0.001
M2 <--> M4	0.758 **	0.672	0.827	0.679	0.831	0.002/0.001
M2 <--> M5	1.081 **	1.058	1.110	1.057	1.109	0.001/0.001
M2 <--> SI	0.820 **	0.751	0.874	0.755	0.878	0.002/0.001
M3 <--> M4	1.027 **	0.994	1.058	0.994	1.057	0.001/0.001
M3 <--> M5	0.772 **	0.688	0.836	0.695	0.840	0.001/0.001
M3 <--> SI	0.729 **	0.638	0.799	0.646	0.804	0.002/0.001
M4 <--> M5	0.763 **	0.679	0.834	0.687	0.839	0.002/0.001
M4 <--> SI	0.718 **	0.621	0.790	0.628	0.797	0.002/0.001
M5 <--> SI	0.805 **	0.733	0.859	0.741	0.863	0.002/0.001

Note: ** = p-value less than 0.01.

were greater than 0.7. Thus, CI method was adopted to test the discriminant validity for the measurement model. In AMOS software, it provides two confidence interval estimation methods, including bias-corrected percentile method (BC) and percentile method (also called percentile correlation method, PC). And the estimation criterion (the ideal discriminant validity) of CI is to calculate the correlation coefficient between the latent variables within 95% confidence level of CI by repeating 2000 times calculation for bootstrap sampling and, meanwhile, the estimated correlation values are required to range between the upper and lower limits of the confidence intervals and the range cannot contain a value of 1.0. As shown in Table 12, all the estimates (correlation values) for all 15 pairs of correlation appear significant, and each pair of correlation under 95% CI (except the range of M₃ <--> M₄ containing 1.0) corresponds with the estimation criterion of CI. The test of discriminant validity may not be totally ideal for the measurement model but it still can be acceptable based upon the 14 pairs of correlation having discriminant validity [60].

According to the assessment for measurement model, the results indicate that the measurement model has acceptable quality, possessing high reliability, convergent validity, discriminant validity, and suitable goodness of fit, which confirms that the structural model can be further developed.

4.2.2. Structural Model

Using the MLE method to estimate all parameter values of the structural model from the data observed, the structural model of the six constructs also demonstrates a good fit to the data (n = 300, Chi-Square/DF = 393.703/118 = 3.336, CFI =0.943, TLI = 0.926, IFI = 0.944, RFI = 0.898, NFI =0.921, RMSEA = 0.088, see

Table 13. Test for SEM indices of goodness of fit.

Fit Index	Criteria	Test Results	Judgment
χ2	The smaller, the better	393.703
χ2/df	<5	3.336	Conforming
CFI	≥0.90	0.943	Conforming
TLI	≥0.90	0.926	Conforming
IFI	≥0.90	0.944	Conforming
RFI	≥0.80	0.898	Conforming
NFI	≥0.80	0.921	Conforming
RMSEA	≤0.10	0.088	Conforming

), showing that the structural model is acceptable in the test of model fit.

Figure 2 illustrates that, among the simultaneous influences of five motive constructs on sponsoring intention, M₁ with standardized path coefficient 0.696 (p = 0.000 < 0.001) and M₂ with standardized path coefficient 0.200 (p = 0.009 < 0.01) express positive and significant influences on sponsoring intention, whereas the other three constructs, M₃ with 0.050 (p = 0.812 > 0.05), M₄ with −0.185 (p = 0.350 > 0.05), and M₅ with 0.155 (p = 0.051 > 0.05), cannot demonstrate significant influences on sponsoring intention. In this study, we propose the hypothesis for the relationship between M₁ and sponsoring intention under the structural model, and its meaning in hypothesis testing is to reject H₀ (not significantly correlated) according to the analysis above.

In addition, the R² value in this whole structural model reaches 0.790, which indicates that sponsoring intention can be explained by these five constructs to the high extent of 79%. That is, the structural model is confirmed to be highly explanatory or convincing.

Figure 3 depicts the whole structural model of these five sponsorship motives (M₁–M₅) with sponsoring intention. As can be seen from Figure 3, M₁ construct demonstrates the most significant and positive influence on sponsoring intention. In M₁ construct, it consists of three measured factors: CI1, CI2, and CI3, and, of the measured factors, CI1 (with standardized regression weight 0.833, p = 0.000 < 0.001) is the strongest factor; CI3 (with standardized regression weight 0.821, p = 0.000 < 0.001) is the second; and CI2 (with standardized regression weight 0.817, p = 0.000 < 0.001) is the third. Meanwhile, the R² values (squared multiple correlations) of these three measured factors are 0.694, 0.667, and 0.674, respectively, and all of the R² values are greater than the acceptable threshold of 0.40 [58,60], reflecting high explanatory power on M₁ construct. The results indicate that, in respondents’ perception of M₁ (corporate image), all the three measured factors (CI1, CI2, and CI3) are important and influential for M₁ construct, and CI1 (for cultivating and performing CSR to shape the corporate image) appears the most correlated with M₁ construct. That is, in this structural model, the construct of CI is confirmed to have the most significant and positive influence on sponsoring intention and, simultaneously, CI1 is proven to be the most influential factor in the construct of CI. Therefore, based on the above, the results obtained confirm the second hypothesis we proposed.

5. Discussion

An important contribution of this study is to utilize FPR approach to explore the preference of corporations for sponsorship motives and then to use SEM approach to examine the impact of sponsorship motives on sponsoring intention in a post-epidemic era, while the same samples were surveyed for these two approaches. In Study 1 (FPR approach), we adequately utilize the advantages of the reciprocal additive consistent fuzzy preference relation to construct more appropriate pairwise comparison matrices, rather than using conventional multiplicative preference relation, further providing a faster way to execute and analyze the priorities of the preference for sponsorship motives. The exploration is based on taking a firm perspective that is similar to the article of Slåtten et al. [13], and the results of FPR analysis show that the most preferential sponsorship motive is M₁ (corporate image) and the second most preferential motive is M₂ (marketing instrument) among the judgments of 60 expert groups.

Evidently, corporate image is the motive of sponsorship that corporations most favor in the post-epidemic era, in spite of different industry backgrounds (service industry, manufacturing industry, and subsidiaries from abroad), which seems to appear consistent with the arguments highlighted by a certain number of previous studies. Namely, Chen et al. [25] stated that corporate image is a crucial component to increase customers’ trustworthiness for creating potential corporate profitability, particularly in times of crisis. Also, Wang et al. [26] showed that corporate image can be the most critical factor in increasing the public’s perception and identification.

Additionally, the preference results of corporate image and marketing instrument support the arguments of previous studies. Corporate image implies performing CSR to shape better image, changing the public’s perception and identification, and enhancing the visibility and credibility, which highlights similar arguments reflected by a group of previous studies [9,13,25,26,27,28,29,30]. On the other hand, marketing instrument focuses on enlarging market share, reaching the connection with target consumers, and maximizing sponsorship investment effectiveness, which also involves similar arguments indicated by other previous studies [31,32,33,34,35].

Furthermore, the study has contributed to the exploration of motive constructs in the context of encompassing a wide range of sponsorship types or categories. Theoretically, by conducting an extensive review of the literature on sponsorship, five constructs are postulated for sponsorship motives, which are different from the framework of four sponsorship motive categories (market motive, society motive, bond motive, and clan motive) proposed by Slåtten et al. [13] and the new latest published work of Li et al. [63] presenting the structure of two motives (altruistic sponsorship motivation and egoistic sponsorship motivation) similar to the classification of Slåtten et al. Basically, these two articles have made important contributions to categorize the sponsorship motives, even though focusing on sports events. Based on the basis of these two articles’ categorization, the study also integrates relevant arguments of sponsorship objectives and effectiveness arising from other specific sponsorship types or events (e.g., arts or festivals) into the categorization of sponsorship motives. In this respect, this study anticipates providing a comprehensive perspective for sponsorship motive, expanding the research base in this field and benefiting future researchers to enlarge available classification.

Having the preferential choice for sponsorship motives does not mean having the intention to sponsor, which is the reason for using SEM approach in Study 2. And, in this way, to explore the measurement of sponsoring intention and the relationships between sponsorship motives and sponsoring intention, which still exists a lacuna in the literature related to sponsorship up to the present, is another important contribution of this study. Comparatively, the relevant issues about the relationship of sponsorship with customer purchasing intention or customer behavioral intention were continuously explored over time by numerous previous studies [5,9,20,21,31,48,63,64], in accordance with a customer perspective. By taking a firm perspective, this study opens up the opportunity to provide better understanding for the relationships between sponsorship motives and sponsoring intention, offering not only constructive value to the existing literature, but also persuasive arguments that can be quoted.

In regard to the findings analyzed by SEM approach, corporate image and marketing instrument express significant and positive influences on sponsoring intention, and corporate image appears most significant and positive among the simultaneous influences of the five motives, which confirms that the sequence of influences on sponsoring intention for corporate image and marketing instrument is in line with the assessment of FPR approach in preference priority. This finding explains that SEM approach can assist FPR approach to develop and examine a causal relationship of a preferential construct assessed by FPR with a potential dependent variable. However, the sequence of influences for five motives in SEM is not totally the same as the sequence of preference in FPR. The sequence of preference and the sequence of influence for M₅ (obtaining partiality) can be used to compare. In SEM analysis, the significance of obtaining partiality (p = 0.051) is close to the significant value (0.05) and its influence on sponsoring intention is the third, but its preference assessed by FPR is the fourth. This result might be considered as a gap between the decision of a group (FPR) and the decision of an individual (SEM) in measurement, even as the same samples were surveyed.

As discussed above for the combination of FPR and SEM approaches, it can be seen that FPR, an effective and helpful decision-making approach in evaluating preference judgment [56,65], demonstrates its contribution on the preference results for sponsorship motives while using the technique of constructing appropriate pairwise comparison matrices, and SEM, a statistical multivariate analysis approach utilized to examine the causal relationships of multiple construct factors with dependent variables [31,64], also makes its contribution to examine the causal relationship between sponsorship motives and sponsoring intention. In other words, FPR can help SEM judge potential or preferential constructs that can be used as independent latent variables, and SEM can assist FPR to develop and examine a causal relationship of a preferential construct with a potential dependent variable. Thereby, the conjunction of FPR and SEM can create the methodological synergy of achieving better holistic analysis and, meanwhile, the combination also can enhance the synergy of complementary effects between the decision-making approach (FPR) and the multivariate analysis approach (SEM). Some previous studies also reflect parallel methodological synergies by using two different approaches. For example, Kim et al. [66] used SEM and fuzzy set-quality comparative analysis (fsQCA) to explore the success factors of the K-pop industry and demonstrated that fsQCA can help SEM further confirm the key factors impacting tourist behavioral intention, which can achieve a complementary synergy. Also, Aw et al. [67] used partial least squares-structural equation modeling (PLS-SEM) and fsQCA to understand the combined effects of content attributes, interaction strategies, and parasocial relationships on purchase intention and found that fsQCA can help PLS-SEM indicate different combining solutions for improving the relationships between the constructs and purchase intention, creating the synergy of holistic analysis.

In addition, the specific finding that CI1 (performing CSR to shape better image) is the strongest measured factor in corporate image construct explains that CSR is the most important factor to promote corporate image, particularly in a post-epidemic era. Appreciably, the respondents of these 60 sampled companies clearly perceive that, for them, corporate image is the most preferential sponsorship motive and the impact of corporate image on sponsoring intention is the most significant, as well as CSR being the most influential factor for corporate image. It is certain that CSR genuinely becomes the most critical component for sponsorship motive and sponsoring intention while assessed and examined by these two approaches. This finding supports the arguments of previous studies that combining CSR strategies with sustainable development (SD) to maintain competitive advantages and simultaneously deploying sponsorship as a major program or project of implementing CSR can create an effective connection with the enhancement of corporate image [9,29,30]. Zaidi and Jamshed [68] even emphasized that developing CSR actions to implement SD can strengthen “brand power” of corporations and further enhance corporate image. In this regard, this finding points to a potential link among CSR, SD, and corporate image (CI), which implies that CSR can contribute towards SD, along with increasing corporate image, particularly for the perception of respondents.

Especially noteworthy is the fact that the COVID-19 pandemic continues to shake up societies and economies around the world in unprecedented ways, bringing internal and external challenges of implementing SD for corporations. Facing this ongoing disaster, almost all the international communities are now standing at the crucial moment for achieving Sustainable Development Goals of the United Nations. However, the results of this study show that a constructive link among CSR, SD, and CI is affirmed by the respondents we interviewed. Importantly speaking, even societies are facing the crucial moment for achieving SD, but the strong cognition of creating a positive relationship of CSR with SD to enhance CI is still confirmed in this study. Thus, we suggest that special attention should be paid to CSR, which plays a pivotal role in affecting the decision of corporations for sponsorship motives and sponsoring intention and, in a post-epidemic era, continuing to develop CSR actions to enhance corporate image can be the best strategy while facing internal and external challenges of implementing SD.

6. Conclusions and Future Research Recommendations

The aim of this study was to explore the preference of corporations for sponsorship motives and the impact of sponsorship motives on sponsoring intention in the post-epidemic era of COVID-19 by using two different approaches: FPR and SEM. Typically, these two popular approaches embody specific qualities and various features for statistical measurement in the application of methodology. In academia, FPR and SEM approaches, respectively, represent different research methods used for exploring distinct research fields or different research purposes, which were always applied separately in many articles published in the past.

Originally, we planned to deploy two separate draft writings: one for FPR approach and the other for SEM approach. However, in the period of conducting the investigations, we found that these two independent approaches could be complementary to each other, especially in some specific situations. Apparently, this study is a typical case to interpret the specific situations, which includes whether the intention to engage in a target behavior can be stirred while the preferential results are confirmed by using the decision approach (FPR) and whether the impact of preference constructs on intention can be proved (by using SEM approach) to be in line with the assessment of FPR approach in the sequence of preference. Mulling over those specific situations and relevant arguments that may be discussed, this study therefore integrates these two independent approaches into one draft we decide to submit. Indeed, the results of this study confirm that the FPR approach can combine with the SEM approach to present more comprehensive analysis under the same samples and measurement criteria. In the demonstration of analysis, both approaches express complementarily. The combination of these two different methodologies may not be an innovation but it is worth inspiring and reflecting in quantitative research. Just as explained in the discussion section, the conjunction of FPR and SEM in this study creates methodological synergies, namely, enhancing complementary effects and achieving better holistic analysis. Appreciably, if we use these two approaches separately, these synergies cannot emerge from analysis. Therefore, the methodological synergies can be regarded as important advantages, which deserve to be highlighted for future studies. In this regard, the pragmatic application of these two different approaches to quantitative analysis demonstrates both methodological and practical implications for this study.

Although this study attempted to provide valuable insights into the preference of sponsorship motives and the impact of sponsorship motives on sponsoring intention, some limitations of this study should be acknowledged. First, for the severe pandemic constraints, the number of respondents is somewhat limited, even with segmentation of different industry backgrounds. This may still limit the generalizability of the findings. A larger sample size with more diverse industry backgrounds is recommended for future studies to improve the generalizability. Second, the measured factors (criteria) for each motive construct are extracted based on empirical evidence relevant to sponsorship literature. Although the measured factors for the five sponsorship motives have been effectively examined, these factors may not sufficiently represent the whole implication to measure all the motives. Future research can incorporate other possible underlying factors into the motive construct, which may have the potential to contribute to different preferential results. Third, this study focuses on examining the causal relationship between sponsorship motives and sponsoring intention. Although the findings provide a clear picture of indicating the significant and positive influences of two sponsorship motives on sponsoring intention, it is important to note that different sponsorship types or events may also be posited to have significant and positive influences on sponsoring intention. In this respect, future research can try to examine the relationships between sponsorship types and sponsoring intention, which may further expand the research base in the field of sponsoring intention.