Study on the influence of sports consumption decisions of college students in commercial fitness clubs in the post pandemic era of COVID- 19

Abstract

Accelerating the development of the sports industry and comprehensively promoting consumption has become a visionary goal for China’s national economy and social development. Based on the social background of the post pandemic era of COVID-19, this study mainly uses thematic analysis and analytic hierarchy process to analyse the influencing factors of sports consumption decisions of college students in commercial fitness clubs in Dongguan City, and to construct a sports consumption decision system. Initially, semi-structured interviews were conducted with 15 participants. Subsequently, thematic analysis was employed to identify 15 factors influencing college students in Dongguan City when making sports consumption decisions at commercial fitness clubs. Finally, the analytic hierarchy process was applied to determine the relative importance of each influencing factor. The results indicate that in the post pandemic era of COVID-19, college students in Dongguan City prioritised “consumption” ($D_2=0.4590$), followed by “consumption with a wait-and-see attitude” ($D_1=0.3193$), and lastly “no consumption” ($D_3=0.2217$) in the sports consumption decisions at commercial fitness clubs. All eigenvector values passed the consistency test, indicating that the judgment matrix is logically coherent. Among the factors affecting the sports consumption decisions of college students in commercial fitness clubs in Dongguan City, economic factors are the most significant, followed by time factors, self-need factors, and other factors. The findings of this study not only offer an objective understanding and fresh insights into the factors influencing college students’ sports consumption decisions in the post pandemic era of COVID-19 but also propose business strategies and practical recommendations for the sustainable development of the fitness industry and sports consumption market. The conclusions provide valuable practical guidance for promoting the healthy development of the sports consumption market and the broader sports industry in the post pandemic context.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-22805-1.

Introduction

Consumption is a social, cultural and economic process of choosing goods. This process reflects the opportunities and constraints of modernity [1]. In 2021, the Chinese government explicitly proposed in the plan for national economic and social development to focus on the needs of residents’ consumption upgrading and promote the development of sports consumption in order to promote the development of comprehensive consumption. This highlights both the developmental trajectory and the research domain of sports consumption, while also fostering the comprehensive advancement of consumption. However, the global economy was severely impacted by the outbreak of COVID- 19 [2]. Although on 5 May 2023, the World Health Organization announced that COVID- 19 no longer constitutes a public health emergency of international concern. However, this does not imply that COVID- 19 is no longer a global health threat, but rather that it remains an ongoing health issue. This signifies that the world has entered a “post pandemic era of COVID- 19” [3]. The post-epidemic era does not represent the complete eradication of the epidemic or a return to the status quo, but rather a period characterised by fluctuating outbreaks and the potential for small-scale resurgence at any given time. This will have a profound and lasting impact on various aspects of the economy and society [4]. Research has indicated that COVID- 19 has exerted a markedly negative impact on the sports competition and performance industry, the fitness and leisure sector, sports tourism, and the production and sale of sporting goods [5]. In addition, COVID- 19 has altered consumers’ product demand, shopping behaviour, purchasing behaviour, and their level of post-purchase satisfaction [3]. Scientifically, COVID- 19 has not yet disappeared, and it remains uncertain whether it will lead to a new epidemic [6]. The “post-epidemic era” represents a continuous phase, rather than a simple beginning and end. The impact of COVID- 19 on sports consumption and the fitness industry has emerged as a research focal point in recent years. In this specific social context, it is especially essential to examine the prominent social issues surrounding college students’ sports consumption, with particular focus on emerging influencing factors.

A considerable body of research has examined the factors influencing sports consumption decisions, primarily focusing on economic factors, demographic characteristics, and time allocation. For example, Thibaut et al. analysed the relationship between income groups and sports consumption from an economic perspective, demonstrating that income is a crucial factor influencing consumers’ sports consumption [7]. Similarly, the findings of Paár et al. indicate that economic income is one of the most important determinants of participation in sports and sports consumption [8]. However, Fernando et al., in their analysis of the socio-demographic and economic determinants of sports participation and consumer sports expenditure, found that sports consumption behaviour is influenced not only by economic factors (e.g., individual income level) and demographic characteristics (e.g., age, gender, ethnicity, and education level) but also by other factors such as time and occupation [9]. Furthermore, Steenhuis et al. highlighted that pricing strategy is also an intervention tool for promoting consumer participation in sports activities. Reasonable pricing facilitates increased sports consumption and participation, whereas inappropriate pricing acts as a barrier [10]. Shapiro’s research not only further confirmed the relationship between reasonable pricing strategies and consumers’ sports consumption but also demonstrated how pricing strategies influence consumer decision-making through perceived value, such as perceptions of fairness [11].

Previous research has also explored how consumers’ psychological and social factors are linked to sports consumption decision-making. For example, studies have shown that venue quality, promotional activities, service quality, product quality, and purchase satisfaction can all influence consumers’ psychological responses, thereby affecting their sports consumption decisions [12–16]. Additionally, Kamphuis et al. found that social factors (e.g., perceived insecurity and limited social networks) and personal cognition are strongly associated with consumers’ sports consumption and participation in physical activities [17]. Hightower et al. also investigated how the service environment can influence consumers’ purchasing decisions by affecting word-of-mouth and loyalty [18].

In addition, many scholars have conducted studies on sports consumption based on existing theories. For example, the findings of Chiu et al., grounded in the theory of planned behaviour, confirm that attitude, subjective norms, and both positive and negative anticipated emotions significantly influence consumers’ decision-making and willingness to purchase sports products online [19, 20]. The findings of Shapiro et al., based on perceived risk theory, indicate that identification with perceived value is positively correlated with consumers’ purchase intentions [21]. Similarly, Chou et al. also validated the application of this theory in sports consumption, demonstrating that consumers’ purchase motivation and perceived risk serve as significant predictors of their consumption decisions [22].

Ko et al., drawing on Maslow’s hierarchy of needs, analysed the influence of consumer personality and needs on consumption decisions. The results indicate that arousal and belonging needs have a significant impact on consumers’ sports participation behaviour [23]. Pawlowski et al., based on time allocation theory, explored the relationship between sports consumption demand and the distribution of sports facilities. Their findings reveal that the “distance to sports facilities” can positively influence consumers’ sports consumption decisions through the time required to reach these facilities [24]. The findings of Thibaut et al. further confirm that time is an essential factor influencing sports participation and consumption [25].

Moreover, Thibaut et al., drawing on social class theory, found that consumers from different social classes exhibit variations in their sports consumption behaviour [7]. Finally, Yoshida et al., based on services marketing theory, demonstrated that environment, facility convenience, product services, and brand can significantly enhance sports consumers’ satisfaction, thereby promoting a positive purchasing experience [26].

Although significant progress has been made in existing literature and related theories on sports consumption decisions. For example, perceived risk theory identifies six main types of perceived risk, all of which have the potential to influence consumers’ purchasing decisions. Additionally, Maslow’s hierarchy of needs emphasises the importance of consumers’ need for a sense of safety. However, these theories do not address the safety risks and health concerns associated with COVID- 19, nor do they explore in depth the impact of safety needs on sports consumption decisions in the context of the pandemic. Concerns about COVID- 19, increased health awareness, and heightened safety considerations may become new potential influencing factors in college students’ sports consumption decisions in the post pandemic era of COVID- 19. Therefore, it is worth further investigation whether the factors influencing college students’ sports consumption decisions at commercial fitness clubs will change in the post-pandemic context and whether new influencing variables will emerge. This study focuses on the sports consumption behaviour of college students at commercial fitness clubs in Dongguan during the post pandemic era of COVID- 19, addressing the lack of attention in existing research to this specific group and context. This serves to highlight the novelty of the present study.

This study is grounded in Maslow’s hierarchy of needs theory, service marketing theory, and time allocation theory, offering a theoretical framework to examine how various factors influence the sports consumption decisions of college students at commercial fitness clubs in the post pandemic era from multiple theoretical perspectives. The study primarily employs thematic analysis and the analytic hierarchy process (AHP model), offering a methodology that integrates both qualitative and quantitative analyses, as well as a multi-level decision-making tool, to provide new perspectives for addressing the complexities of sports consumption decisions. The aim of this study is to construct a sports consumption decision system, systematically analysing the key factors that influence college students’ sports consumption decisons at commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19. Additionally, the study seeks to explore the underlying reasons for the emergence of these influencing factors, with a view to proposing targeted strategies aimed at promoting sports consumption. This will provide a theoretical foundation for the development of commercial fitness clubs and sports consumption in the post pandemic era of COVID- 19, thereby contributing to the sustainable development of sports consumption and the sports industry.

Material & methods

Participants

The sample for this study comprised 15 college students from Dongguan City, selected randomly. These students represented various schools, majors, and grades.

Research method

The primary research methods employed in this study are “thematic analysis” and “AHP model”. Thematic analysis, the first key method used, is a qualitative approach for identifying, analysing, and reporting patterns, or “themes”, within data. This method is particularly suited to unstructured data sets, such as open-ended questionnaire responses, interview transcripts, and observation notes. In this study, thematic analysis was applied to the raw data from semi-structured interviews. The interview material was analysed through a three-level coding process to identify the factors influencing college students’ sports consumption decisions in commercial fitness clubs in Dongguan City. The specific three-level coding process is presented as follows.

1. In open coding: The original interviews were transcribed into transcripts using NVIVO 14 software. Each concept within the transcripts that held research value was named and assigned a free node. The specific steps involved obtaining 15 interviews using the interview outline (Supplementary Material 1), reading all 15 interviews, identifying and summarising the words and sentences that were emphasised, highlighted, or related to the interview themes, and summarising the relationships between these words and sentences. The concepts that best matched the original interviews were generalised through multiple comparative analyses and created as free nodes. In this study, a total of 34 concepts were distilled through open coding. For example, from the interview with respondent 005: “Income is limited and I need to control my expenditure budget”. The term “expenditure budget” is closely related to the topic of this study, so it was extracted as an initial concept in the open coding process.

2. In axial coding: In NVIVO 14 analysis software, a master node (also referred to as a core node) can be created using axial coding. Other nodes can then be associated with the master node to illustrate the relationships between them. During the axial coding stage, the 34 initial concepts can be consolidated into 15 categories. For instance, in the interview with respondent 005, the initial concept “expenditure budget” from the open coding stage can be categorized under “Personal income” in the axial coding stage.

3. In theme: By analysing the meanings of the various categories, four core categories were ultimately identified: economic factors, self-need factors, time factors, and other factors. For instance, in the interview with respondent 005, the category “personal income” from the axial coding stage can be classified under “economic factors” of the core categories in theme coding stage.

In this study, the AHP model serves as the second, and most crucial, research method. Initially introduced in 1971 by Professor T.L. Saaty in the United States, the AHP model has since seen widespread application across various fields of decision-making. As a result, the theory behind AHP model has been significantly deepened and developed. The use of AHP model in decision-making has consequently gained increasing popularity [27]. The AHP model decomposes the factors influencing decisions into multiple levels, such as objectives, criteria, and options, and conducts both qualitative and quantitative analyses based on these levels [28]. It is a research method that integrates qualitative and quantitative approaches, simplifying complex processes. However, a key characteristic of the AHP model is its inherent subjectivity. Despite this, the model remains highly practical for both qualitative and quantitative research. Currently, AHP is frequently applied in areas such as risk assessment and decision-making.

The fundamental principle of the AHP model is the “sorting” principle [29]. The detailed steps of the AHP model were as follows.

1. Construct a structural model diagram: The problem to be solved is designated as the target layer, the factors affecting the target layer are designated as the criterion and sub-criterion layers, and the final alternative is designated as the measure layer. The relationships between the factors in the hierarchy are also indicated, and the structural model diagram is constructed step by step.

2. Construct the judgement matrix: Experts or authorities are invited to construct the judgement matrix by assigning values to the importance of the influencing factors at each level according to the structural model diagram.

3. Hierarchical single ranking and hierarchical total ranking: Firstly, the indicators at the same levels are ranked according to the degree of influence they exert on the indicators of the preceding level. Finally, the total score of the alternatives is calculated and ranked based on the corresponding weight values obtained.

4. Overall priority ranking: The total score of the alternatives is determined by calculating the corresponding weight values layer by layer, and the alternatives are then ranked in descending order.

5. Consistency test: For the constructed judgement matrix table, the $C.R.$ value needs to be calculated. Only if $C.R.<0.1$ can the judgement matrix be considered logical and pass the consistency test.

Research Instruments

This study employs four primary research instruments: a voice recorder, NVIVO 14 analysis software, the fundamental scale, and the table of standard values of the average random consistency index ($R.I.$). Semi-structured interviews were conducted with 15 participants using a voice recorder. The original interviews were recorded using a voice recorder to provide primary data for subsequent thematic analysis. In this study, a Shinco A05 model was used, with dimensions of 1.2 × 1.6 × 9.7 cm and a weight of 41 g. The recorder supports WAV format, has a battery capacity of 130 mAh, a bitrate of 128kbps, and a Type-C/DC5 V (500–1000 mA) USB port. It offers 13 h of recording time and has a memory capacity of 64 GB. In the semi-structured interviews conducted for this study, the choice to use the Shinco A05 voice recorder was based on two key considerations. Firstly, compared to video recording and traditional notebooks, the Shinco A05 offers the advantages of a compact size and long battery life. It is highly portable and, at the same time, helps reduce tension between the interviewer and the interviewee, thereby facilitating a more natural and effective interview process. Secondly, the Shinco A05 supports WAV format recording, which not only ensures the completeness of the interview data but also allows the recorded content to be directly transcribed into text using a computer. This significantly enhances the efficiency of subsequent coding and data analysis. Research indicates that the use of voice recorders has become a well-established method for documenting interviews [30]. The voice recorder is a highly effective tool for qualitative research. (Figure 1 shows the voice recorder used in this study).

Fig. 1.

The voice recorder utilised in this study

The second research instrument utilised in this study is NVIVO 14 analysis software. Primary material from the semi-structured interviews was coded at three levels in this study, primarily using NVIVO 14 analysis software. Influencing factors in the sports consumption decisions of college students at commercial fitness clubs in Dongguan City were summarised through thematic analysis. This process provides quantitative indicators for the subsequent application of the AHP model. The utilisation of NVIVO 14 analysis software in the coding process of thematic analysis offered notable advantages in three primary respects. Firstly, NVIVO 14 analysis software enabled a systematic identification and summarisation of core themes, thereby enhancing the efficiency and accuracy of the coding process. Secondly, it facilitated visual analysis, which provided an intuitive representation of the relationships between variables. Lastly, the software supported effective data management and retrieval, thereby improving the review and validation of analytical outcomes. It has been found that combining traditional tools for data analysis, such as coloured pens and paper, with NVIVO 14 software provides an effective and validated approach to qualitative research [31]. (Figure 2 illustrates the operational principle of NVIVO 14 analysis software).

Fig. 2.

Operational principle of the NVIVO 14 analysis software

The fundamental scale is the most crucial step in quantifying the data. This table serves as a key reference for comparing the relative importance of each influencing factor within the AHP model [32]. It demonstrates the importance of each factor on a scale of 1 to 9. (The fundamental scale is shown in Table 1).

Table 1.

The fundamental scale

Intensity of importance on an absolute scale	Meaning
1	Indicates that two elements are of equal importance compared to each other
3	Indicates that the former is slightly more important than the latter when comparing two elements
5	Indicates that the former is significantly more important than the latter when comparing the two elements
7	Indicates that when two elements are compared, the former is more strongly important than the latter
9	Indicates that the former is more extremely important than the latter when comparing the two elements
2,4,6,8	Indicates the intermediate value of the above judgement
countdown	If the ratio of the importance of element $i$ to element $j$ is $a_{\mathit{ij}}$, then the ratio of the importance of element $j$ to element $i$ is: $a_{\mathit{ji}}=1/a_{\mathit{ij}}$

In this study, the table of standard values for random index ($R.I.$) is the fourth crucial research instrument. This tool is essential for calculating the consistency ratio ($C.R.$). Only by referring to this table can the corresponding values of $R.I.$ be accurately obtained. The consistency of the judgement matrix can be assessed for reasonableness by obtaining the value of $C.R.$ [32]. (The table of standard values for random index ($R.I.$) is shown in Table 2).

Table 2.

Standard value for random index ($R.I.$)

matrix order	1	2	3	4	5	6	7	8	9	10	…
$R.I.$	0	0	0.52	0.89	1.12	1.26	1.36	1.41	1.46	1.49	…

Research procedure

The primary objective of this study is to utilise the AHP model to simplify the issue of sports consumption by college students in Dongguan City’s commercial fitness clubs in the post pandemic era of COVID- 19. This aims to construct a sports consumption decision system to analyse the factors influencing college students’ sports consumption decisions, thereby promoting the healthy development of sports consumption. The research procedure of this study is divided into two main parts.

In the first part, semi-structured interviews were conducted with 15 college students in Dongguan City using a voice recorder. Subsequently, thematic analysis was employed to code the original interview materials (three-level coding), ultimately summarising the “influencing factors” of college students’ sports consumption desicions in commercial fitness clubs in Dongguan City in the post pandemic era of COVID- 19. The detailed coding process was carried out using NVIVO 14 analysis software.

In the second part, the influencing factors identified in the first part were quantified using the AHP model. Initially, the AHP model defines the problem to be addressed as the target layer. Through successive simplifications, the importance of each sub-factor influencing the target layer is determined and ranked. Finally, alternatives are ranked according to their scores, and consistency is tested. If the consistency test is passed, the results can serve as a basis for decision-making.

Accordingly, this study adopts the AHP model as the primary research framework. Through a three-level coding process, a decision system for college students’ sports consumption is constructed. The model is employed to analyse the factors influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City. (The research procedure is illustrated in Fig. 3).

Fig. 3.

Research Procedure of this study

Data collection and analysis/Statistical analysis

This study focuses on the application of the AHP model to sports consumption decisions in commercial fitness clubs. Initially, primary information on the factors influencing college students’ sports consumption decisions of commercial fitness clubs in Dongguan City was collected through semi-structured interviews. Subsequently, thematic analysis was employed to perform a three-level coding process—comprising open coding, axial coding and theme—to summarise the “influencing factors” of sports consumption decisions among these students. Finally, experts were invited to assign scores to the identified influencing factors using the AHP model. Thus, the data in this study predominantly stems from the scores assigned by the experts to each of the sub-factors. The analysis of this data is primarily based on the formulas involved in the AHP model, all of which are detailed in the “Results” section.

Results

Part I

Results of the coding matrix

Research has demonstrated that thematic analysis is a widely utilised technique for analysing qualitative data. This method can be divided into six key steps: transcription, selection of keywords, coding, theme development, conceptualisation through interpretation of keywords/codes/themes, and the development of a conceptual model [33]. In the research method, this study employed semi-structured interviews with 15 college students. The process of three-level coding of the raw interview data was systematically demonstrated using the thematic analysis. Therefore, this study identified the factors influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City in the post pandemic era of COVID- 19. These factors were categorised into 4 themes, 15 subcategories, and 34 initial concepts.

The four identified themes encompassed a total of 15 categories. For instance, the theme “economic factors” comprised three categories: personal income, fitness pricing, and additional expenditures. The theme “self-need factors” included four categories: interests and hobbies, entertainment, physical fitness and well-being, and social interaction. The theme “time factors” consisted of four categories: travel duration, study duration, leisure time, and fitness duration. Lastly, the theme “other factors” comprised four categories: indoor environmental comfort level, safety, service quality, and popularity.

Furthermore, the 15 categories encompass a total of 34 initial concepts. For instance, the category of “personal income” includes two initial concepts: “budget control” and “source of income”. The category of “fitness pricing” comprises two initial concepts: “cost-effectiveness” and “budget range”. The category of “additional expenditures” encompasses a single initial concept: “additional expenses for fitness”. The category of “Interests and hobbies” consists of one initial concept: “personal interests”. Similarly, the category of “entertainment” comprises two initial concepts: “enjoyment” and “relaxation”. The category of “physical fitness and well-being” includes four initial concepts: “enhancing physical immunity”, “health awareness”, “improving physical condition”, and “enhancing mental well-being”. The category of “social interaction” comprises three initial concepts: “group classes”, “exercising with friends”, and “expanding social circles”.

The category of “travel duration” encompasses two initial concepts: “travel distance” and “commuting time”. The category of “study duration” includes two initial concepts: “heavy academic workload” and “study time management”. Regarding fitness time-related aspects, the category of “leisure time” consists of the initial concept “leisure time arrangement”, whereas the category of “fitness duration” comprises the initial concept “scheduling of fitness time”.

Environmental and safety considerations are also reflected in the categorisation. The category of “indoor environment comfort level” includes four initial concepts: “hygienic condition”, “spacious fitness environment”, “positive fitness atmosphere”, and “well-equipped facilities”. The category of “safety” encompasses four initial concepts: “emergency measures”, “environmental safety”, “population density”, and “equipment safety”.

Additionally, market-related factors are captured in two categories: “popularity” and “service quality”. The category of “popularity” includes two initial concepts: “brand awareness” and “brand influence”. Meanwhile, the category of “service quality” comprises three initial concepts: “staff attitude”, “professional guidance”, and “equipment maintenance”.

Theoretical saturation verification

In qualitative research, sample size is primarily determined based on data saturation [34]. This study adopted the method for assessing data saturation proposed by Francis et al. (2010), using a naturally occurring theoretical saturation approach to analyse the interview data [35]. Specifically, a total of 15 participants were interviewed for this study. During the coding process of the first 12 participants’ interview data, no new concepts or relationships emerged. Subsequently, the same coding process was conducted with an additional three participants for verification. The results confirmed that no new concepts were identified beyond those found in the coding of the initial 12 participants. Therefore, it can be concluded that the concepts in the study had reached saturation. This further demonstrates that a sample size of 15 participants was sufficient to achieve data saturation, covering the influencing factors required for this study and meeting the standards of qualitative research.

Based on this, a three-level coding structure was established in NVIVO 14 analysis software for this study, ultimately leading to the determination of the main matrix (as shown in Fig. 4). All influencing factors emerged from the semi-structured interviews. The identification of these factors provides foundational support for the application of the AHP model in the second phase of the study.

Fig. 4.

Main matrix. Note: prisms represent selective coding; ovals represent axial coding; rectangles represent theme

Part II

After completing the first part of the research process, the factors influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City, in the post pandemic era of COVID- 19, were accurately summarised. In the second part of the study, these influencing factors were incorporated into a structural model diagram using the AHP model. The specific steps and results are outlined below.

Construct a structural model diagram

In response to the issue of sports consumption decisions among college students in commercial fitness clubs in Dongguan City, a hierarchical model diagram is constructed. The target layer is clearly delineated in the constructed hierarchical structure diagram, representing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City in the post pandemic era of COVID- 19. The factors influencing these decisions are multifaceted. The hierarchical levels of each factor are determined based on their relationship to the target. The established recursive hierarchy is analysed below:

• The first layer is the target layer A: the highest decision target is set as the sports consumption decisions of college students in commercial fitness clubs in Dongguan City in the post pandemic era of COVID- 19.

• The second layer is the criterion layer B: the general factors influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19 include the following four major aspects: economic factors, self-need factors, time factors, and other factors. These factors basically encompass the determinants affecting the sports consumption decisions of college students in Dongguan commercial fitness clubs.

• The third layer is the sub-criteria layer C, which evaluates and measures the influencing factors and indicators from the criterion layer. This includes personal income, fitness pricing, additional expenditures, Interests and hobbies, entertainment, physical fitness and well-being, social interaction, travel duration, study duration, leisure time, fitness duration, indoor environment comfort level, safety, service quality and popularity, totalling 15 sub-factors.

• The fourth layer is the measures layer D, which represents the options within the evaluation system, namely the choices available: wait-and-see attitude, consumption, or non-consumption.

Based on the evaluation content and criteria at each level, a decision system for the sports consumption of college students in commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19 has been constructed, as presented in Table 3 below.

Table 3.

A decision system for sports consumption of college students in commercial fitness clubs in Dongguan City in the post pandemic era of COVID- 19

Target layer $A$	Criterion layer $B$	Sub-criteria layer $C$	Measures layer $D$
The sports consumption decisions $A$	Economic factors $B_1$	Personal income $C_1$	Wait-and-see attitude $D_1$ or Consumption $D_2$ or Non-consumption option $D_3$
		Fitness pricing $C_2$
		Additional expenditures $C_3$
	Self-need factors $B_2$	Interests and hobbies $C_4$
		Entertainment $C_5$
		Physical fitness and well-being $C_6$
		Social interaction $C_7$
	Time factor $B_3$	Travel duration $C_8$
		Study duration $C_9$
		Leisure time $C_{10}$
		Fitness duration $C_{11}$
	Other factor $B_4$	Service quality $C_{12}$
		Popularity $C_{13}$
		Safety $C_{14}$
		Indoor environment comfort level $C_{15}$

Construct a judgement matrix and invite experts to assign scores

A critical step in the AHP model is the construction of the judgement matrix, which forms the foundation for assessing the importance of factors and serves as the basis for calculating weight values. Based on the criteria of the judgement matrix, experts were repeatedly consulted to determine which of two elements was more important when compared pairwise. They were then asked to assign a value reflecting the level of importance on a scale of 1 to 9 (the fundamental scale is presented in Table 1).

Suppose the filled-in judgement matrix is $A={\left({\text{a}}_{\mathit{ij}}\right)}_{\text{n}\times n}$, which possesses the following properties:

$$\left(\begin{array}{c}{\alpha }_{\mathit{ij}}>0\\ {\alpha }_{\mathit{ji}}=\frac{1}{{\alpha }_{\mathit{ij}}}\\ {\alpha }_{\mathit{ii}}=1\end{array}\right)$$ 1

• Step 1: Construct the judgement matrix as outlined below (note: all subsequent calculations are uniformly rounded to four decimal places). The data for the judgement matrix was primarily derived from the questionnaire (Supplementary Material 2). Figure 5 shows the judgement matrix from layer $C$ to layer $A$.

Fig. 5.

Judgement matrix from layer $C$ to layer $A$

• Step 2: For each judgement matrix, calculate the “column sum” for layer $A$,$B$ and $C$, then perform standardisation and normalisation to derive the “normalised matrix”. Figure 6 illustrates the “normalised matrix” from layer $C$ to layer $A$.

Fig. 6.

“Normalisation matrix” from layer $C$ to layer $A$

• Step 3: Calculate the mean value of each row in each matrix to obtain the “preference matrix” (i.e., the weight values for each layer) for layers $A$,$B$, and $C$, respectively, as presented in Tables 4, 5 and 6.

Table 4.

The “preference matrix” for Layer $C$ (i.e., the weight values for Layer $C$)

	$D_1$	$D_2$	$D_3$
$C_1$	0.5889	0.2519	0.1593
$C_2$	0.2519	0.5889	0.1593
$C_3$	0.2519	0.5889	0.1593
$C_4$	0.1416	0.5247	0.3338
$C_5$	0.1062	0.6333	0.2605
$C_6$	0.1062	0.6333	0.2605
$C_7$	0.5247	0.1416	0.3338
$C_8$	0.5247	0.1416	0.3338
$C_9$	0.1062	0.6333	0.2605
$C_{10}$	0.1416	0.5247	0.3338
$C_{11}$	0.1416	0.5247	0.3338
$C_{12}$	0.4905	0.3119	0.1976
$C_{13}$	0.4905	0.3119	0.1976
$C_{14}$	0.2519	0.5889	0.1593
$C_{15}$	0.5247	0.1416	0.3338

Table 5.

The “preference matrix” for Layer $B$ (i.e., the weight values for Layer $B$)

	$B_1$	$B_2$	$B_3$	$B_4$
$C_1$	0.5889	0	0	0
$C_2$	0.2519	0	0	0
$C_3$	0.1593	0	0	0
$C_4$	0	0.2473	0	0
$C_5$	0	0.1820	0	0
$C_6$	0	0.4736	0	0
$C_7$	0	0.0971	0	0
$C_8$	0	0	0.0971	0
$C_9$	0	0	0.4736	0
$C_{10}$	0	0	0.2473	0
$C_{11}$	0	0	0.1820	0
$C_{12}$	0	0	0	0.2473
$C_{13}$	0	0	0	0.0971
$C_{14}$	0	0	0	0.4736
$C_{15}$	0	0	0	0.1820

Table 6.

The “preference matrix” for Layer $A$ (i.e., the weight values for Layer $A$)

	$A$
$B_1$	0.4690
$B_2$	0.1484
$B_3$	0.2786
$B_4$	0.1040

Tables 5 and 6 clearly illustrate the weight of various influencing factors in layers B and C. In layer C, the weights assigned to personal income, fitness pricing, additional expenditures, interests and hobbies, entertainment, physical fitness and well-being, social interaction, travel duration, study duration, leisure time, fitness duration, service quality, popularity, safety, and indoor environment comfort level are 0.5889, 0.2519, 0.1593, 0.2473, 0.1820, 0.4736, 0.0971, 0.0971, 0.4736, 0.2473, 0.1820, 0.2473, 0.0971, 0.4736, and 0.1820, respectively.

It is noteworthy that the weights in layer C preliminarily indicate that personal income, physical fitness and well-being, study duration, and safety are the most significant sub-factors affecting college students’ sports consumption decisions in commercial fitness clubs during the post pandemic era of COVID- 19, with their respective weights exceeding 0.4. Meanwhile, the weights of fitness pricing, interests and hobbies, leisure time, and service quality each exceed 0.2, which implies that these sub-factors also have a considerable impact on students’ sports consumption decisions in the post pandemic era of COVID- 19.

In layer B, the weights of economic factors, self-needs factors, time factors, and other factors are 0.4690, 0.1484, 0.2786, and 0.1040, respectively. These findings clearly indicate that during the post pandemic era of COVID- 19, economic factors are the primary determinant influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City. This is followed by time factors, self-needs factors, and other factors. The importance of these influencing factors and the underlying reasons for their significance will be further explored and analysed in the discussion section.

Next step: Multiply the Preference Matrix of Layer $C$ by the Preference Matrix of Layer $B$ obtain a new “3 * 4” New Preference Matrix.

2

Next step: Calculate the total score for options $D_1$,$D_2$ and $D_3$.

3

Obtained:

$$\begin{array}{ccc}{D}_1=0.3193,& D_2=0.4590,& D_3=0.2217\end{array}$$

Overall prioritization

The calculated scores for $D_1$(option 1), $D_2$(option 2) and $D_3$(option 3) are 0.3193, 0.4590, and 0.2217, respectively. Based on these scores, it can be concluded that Option 2 represents the optimal choice for the sports consumption decisions of college student in Dongguan's commercial fitness clubs during the post pandemic era of COVID- 19. Option 1 ranks second, followed by Option 3 as the least preferred choice.

Consistency test for hierarchical single and total ranking

The judgement matrix completed by the expert is considered valid and consistent only if it passes the consistency test. This is determined using the following formula:

$$W_i=\frac{1}{n}\sum _{j=1}^n\frac{{\alpha }_{\mathit{ij}}}{\sum _{k=1}^n{\alpha }_{\mathit{kj}}}$$ 4

The consistency index ($C.I.$) for the single and total ranking of each layer is calculated using the following formula. Table 7 shows the principal eigenvalue (${\lambda }_{max}$) and values of $C.I.$ for each layer:

$$C.I.=\frac{{\lambda }_{\mathit{max}}-n}{n-1}$$ 5

Table 7.

${\lambda }_{max}$ and $C.I.$ of Layer $C$ to Layer $A$

Layer	${\lambda }_{max}$	$C.I.$
$C_1$	3.0539	0.0270
$C_2$	3.0539	0.0270
$C_3$	3.0539	0.0270
$C_4$	3.0538	0.0269
$C_5$	3.0387	0.0194
$C_6$	3.0387	0.0194
$C_7$	3.0538	0.0269
$C_8$	3.0538	0.0269
$C_9$	3.0387	0.0194
$C_{10}$	3.0538	0.0269
$C_{11}$	3.0538	0.0269
$C_{12}$	3.0537	0.0269
$C_{13}$	3.0537	0.0269
$C_{14}$	3.0539	0.0270
$C_{15}$	3.0538	0.0269
$B_1$	3.0539	0.0270
$B_2$	4.2174	0.0725
$B_3$	4.2174	0.0725
$B_4$	4.2174	0.0725
$A$	4.2165	0.0722

The weights of the sub-criterion layer are determined. By consulting the table of standard value for random index ($R.I.$) (as shown in Table 2), the value of $R.I.$ is obtained and substituted into the formula:

$$C.R.=\frac{C.I.}{R.I.}$$ 6

The consistency ratio ($C.R.$) of the sub-criterion layer is calculated. The consistency test for the remaining judgement matrices follows the same steps, ensuring that all single rankings satisfy the condition $C.R.<0.1$. The process for calculating and testing the total hierarchical ranking is identical. The CR values for the target layer $A$, criterion layer $B$, and sub-criterion layer $C$ are all computed, with results indicating $C.R.<0.1$ (as shown in Table 8). These results demonstrate that the overall consistency of the judgement matrix is acceptable, and all eigenvector values pass the consistency test. This confirms that the constructed judgement matrix is logical and that the eigenvector values are valid and feasible for use as weight values.

Table 8.

$C.R.$ of Layer $C$ to Layer $A$

Layer	$C.R.$
$C_1$	0.0518
$C_2$	0.0518
$C_3$	0.0518
$C_4$	0.0517
$C_5$	0.0372
$C_6$	0.0372
$C_7$	0.0517
$C_8$	0.0517
$C_9$	0.0372
$C_{10}$	0.0517
$C_{11}$	0.0517
$C_{12}$	0.0517
$C_{13}$	0.0517
$C_{14}$	0.0518
$C_{15}$	0.0517
$B_1$	0.0518
$B_2$	0.0814
$B_3$	0.0814
$B_4$	0.0814
$A$	0.0811

Discussion

The above empirical research reveals that the factors influencing the sports consumption decisions of college students in commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19 are multifaceted. These factors can be grouped into four main categories: economic factors, self-need factors, time factors, and other factors. Although the results of the study provide a detailed ranking and weighting of individual influencing factors, this section further engages in a comparative analysis of these factors in relation to relevant theories and contemporary studies, aiming to offer fresh insights.

1. Economic factors: central role and consumer spending capacity

The study revealed that “personal income” emerged as the most significant factor influencing the sports consumption decisions of college students at commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19. Our findings are consistent with Borooah’s conclusions, which indicate that an individual’s capacity for consumption is primarily determined by their income level [36]. In practice, it is evident that higher income levels are positively correlated with disposable expenditure and the ability to sustain consumption. Social class theory also strongly supports our findings, indicating that economic income significantly influences individuals’ choices in sport participation and consumption [37]. Research has shown that consumers’ income levels negatively moderate the positive relationship between fear and impulse buying behaviour [38]. This is associated with variations in household income, which result in corresponding differences in the disposable incomes of college students. Higher income levels are linked to a stronger willingness to engage in sports consumption. However, insights gained from the interviews revealed that the direct impact of COVID- 19 on college students’ household income was primarily reflected in a decline in their parents’ earnings, which, in turn, indirectly affected their personal disposable income. In contrast, prior to the outbreak of COVID- 19, college students’ disposable income was neither directly nor indirectly influenced by the pandemic. This distinction accounts for the differences observed between the pre- and post- pandemic eras of COVID- 19. This further underscores the pivotal role of economic factors in shaping college students’ sports consumption decisions.

In addition, our findings suggest that price, as a secondary sub-factor within the economic dimension, was similarly supported by research conducted during the post pandemic era of COVID- 19. This is consistent with the findings of research prior to the occurrence of COVID- 19, which identified price and product attributes as significant factors influencing consumer purchasing decisions [39]. If commercial fitness clubs are overpriced, they are likely to deter sports consumption behaviour, thereby influencing the outcome of consumption decisions. Conversely, reasonable pricing can encourage both consumption behaviour and decisions. Commercial fitness clubs that establish reasonable and stable pricing strategies based on prevailing market economic conditions are likely to attract more fitness enthusiasts, thereby fostering increased participation in sports consumption and promoting the healthy development of the fitness industry [40]. Therefore, a new perspective can be drawn: the influence of rational pricing on the decisions of sports consumers is not affected by the pandemic but rather depends on whether the price corresponds to the consumer’s economic capacity at a given time.

Finally, the findings of this study also indicated that college students, categorized as a “no-income group”, had their level of disposable income directly determined by the living expenses provided by their families. The findings from our interviews indicated that if additional fees were required for participation in sports consumption in commercial fitness clubs, this could impose not only a financial burden but also psychological strain on college students. These factors may directly or indirectly influence their consumption decisions. This further substantiates the central role of economic factors in the sports consumption decisions of college students at commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19.

• 2.Time factors: the conflict between academic pressure and sports Consumption

The findings indicated that “study duration” had the most significant impact on college students’ sports consumption decisions. This is consistent with the perspective of time allocation theory, which posits that time, as a scarce resource, plays a central role in individuals'economic decisions when allocated rationally [41]. The results of the semi-structured interviews revealed that academics were the top priority for college students, with their primary time allocation largely devoted to studying. However, variations in majors, grades and individual circumstances result in differences in the amount of “leisure time” available to college students. Time availability is a significant barrier to expanding participation or increasing the intensity of engagement in sports. Furthermore, consumer expenditure on sports is influenced by factors such as gender, education, income level, and occupation [9]. The results of this study align with pre-COVID- 19 research findings, which indicate that leisure time influences sports participation rates among different gender groups [42]. Our interviews also revealed that, in the post pandemic period, although COVID- 19 continued to exert a certain degree of negative influence and challenges on college students’ study methods and habits, it did not impose any restrictions on their study duration. This further demonstrates that although time is the second most influential factor in college students’ sports consumption decisions, it is not affected by COVID- 19. Regardless of the occurrence of COVID- 19, time remains a significant factor influencing college students’ sports consumption decisions.

Additionally, the findings of this study indicated that college students perceived the “fitness duration” as having a greater influence on their decision to participate in commercial fitness clubs compared to the “travel duration”. Our findings can be validated by previous research, which highlight that among the services provided by commercial fitness clubs, the time and effort consumers invest in sports activities are key dimensions influencing sports consumption [43]. A deeper analysis revealed that the majority of college students believe that, as customers, having paid to access commercial fitness clubs, they cannot afford to waste their time. Consequently, they tend to derive satisfaction and a sense of acquisition from the services they consume. This research finding is also validated, indicating that immersion experiences serve as an intermediary link between demand preferences and consumption behaviour, providing a crucial basis for evaluating individuals’ willingness to repurchase and engage in recommendation behaviour [44]. The impact of COVID- 19 intensified the satisfaction that college students derived from pursuing health or other objectives through fitness consumption. They tended to seek both satisfaction and a sense of achievement from sports consumption services. This, in turn, explained why greater emphasis was placed on “fitness duration” in their sports consumption decision-making process during the post pandemic era of COVID- 19.

Finally, the findings of this study also indicated that if commercial fitness clubs were located too far away or in remote areas, the excessive time required for commuting might diminish individuals’ motivation to work out, thereby directly influenced their fitness consumption decisions. This result is supported by the perspective of Song et al., which suggests that factors such as travel distance and enjoyment positively influence consumers’ consumption decisions [45, 46]. In the post pandemic era of COVID- 19, “travel duration” also affects college students’ sports consumption decisions. The interviews revealed that the impact of travel duration on college students’ fitness consumption decisions was primarily indirect, manifesting through its influence on commuting time, which, in turn, affected their motivation for engaging in fitness activities. This indirectly introduces a new perspective: the influence of time factors on college students’ sports consumption decisions is not dependent on the occurrence of COVID- 19.

• 3.Self-needs: increased health awareness in the post Pandemic era of COVID- 19

Socialisation, diversion, and entertainment are key motivations influencing sports consumption [47]. Personal interests play a crucial role in determining consumption behaviour, which subsequently influences consumption decisions [48]. In addition, community fitness and recreational services have become a significant avenue for residents to cultivate their hobbies. Findings from one study indicated a high likelihood of residents engaging in fitness activities to enhance interpersonal interactions [49]. However, social distancing measures have deprived sports consumers of many of their usual fitness activities during the COVID- 19 pandemic [50]. Previous studies have shown that “interests and hobbies”, “entertainment”, and “social interactions” can directly influence consumers’ sports consumption decisions. However, our findings have revealed that in the post pandemic era of COVID- 19, college students’ interests and social activities were restricted by COVID- 19 control measures. This finding offers a new perspective: COVID- 19 can indirectly influence college students’ sports consumption decisions by directly restricting their needs for “interests and hobbies”, “entertainment”, and “social interactions”.

Additionally, this study also found that in the post pandemic era, COVID- 19 significantly heightened health awareness among college students, leading to an increased demand for fitness consumption. This indicates that the need for “physical fitness and well-being” is a key factor influencing college students’ self-needs. This finding is highly consistent with the conclusions of Sun et al., who argued that self-needs and social influence are the primary drivers of fitness consumption among university students [51]. This is consistent with Maslow’s hierarchy of needs theory, which suggests that once individuals’ basic needs, such as physiological well-being and safety, are fulfilled, they seek higher-level needs, including social belonging and self-actualisation [52]. However, the aforementioned studies did not address the impact of COVID- 19 on the “self-needs” factor. Our findings confirmed that “self-needs” were indeed influenced by COVID- 19, with “physical fitness and well-being” being the most prominent factor. This represents a notable difference from previous research.

• 4.Other factors: the multifaceted influence of safety, environmental comfort, and related Aspects

This study found that in the post-pandemic era of COVID- 19, college students ranked “safety” > “service quality” > “indoor environment comfort level” > “popularity”. This ranking indicates that college students are particularly concerned about the safety of commercial fitness clubs in the post pandemic era of COVID- 19. This aligns with the SERVQUAL model proposed by Parasuraman in service marketing theory. The model suggests that when consumers choose to purchase services, they focus not only on reliability (e.g., service quality) but also on assurance (a sense of security and trust) and tangibles (the physical environment and experience) [53]. However, the “safety factor” in the SERVQUAL model does not take into account the impact of COVID- 19. Nonetheless, the findings of Song et al. support the results of this study, suggesting that during the COVID- 19 period, safety concerns and risk perceptions have become key factors influencing consumer decision-making [45]. This result can be explained by two main reasons: on the one hand, the impact of COVID- 19 has significantly heightened college students’ awareness of health and safety protection. On the other hand, it is the outcome driven by COVID- 19 prevention and control requirements.

Our interviews revealed that, in the post pandemic era of COVID- 19, college students tended to place particular emphasis on the comfort of the indoor environment in commercial fitness clubs (e.g., hygiene conditions, a spacious fitness environment, a positive fitness atmosphere, and well-equipped facilities) as well as service quality (e.g., staff attitude, professional guidance, and equipment maintenance). If the comfort of indoor environment or the service quality of a commercial fitness club did not meet their expectations, it could lower their psychological anticipation and ultimately influence their sports consumption decisions. Both the physical environment and the quality of service have been shown to influence consumer behaviour and decision [18]. The quality of the indoor environment (e.g., air quality, layout, lighting, acoustic conditions, and thermal environments) plays a moderating role in consumers’ willingness to repurchase [54]. Furthermore, “service quality” significantly affects customer satisfaction, which in turn shapes their consumption decisions [55]. Commercial fitness clubs employ various promotional strategies, such as endorsements by sports stars, to enhance their popularity and competitiveness. These promotional efforts enable consumers to access detailed information about the commercial fitness clubs and compare their offerings more effectively, thereby serving as a reference point for making purchase decisions. Sports stars possess a distinct appeal to sports consumers, positively influencing their interest in sports-related activities [56]. These research perspectives also strongly support the findings of this study, indicating that in the post pandemic era of COVID- 19, “indoor environment comfort level”, “service quality”, and “popularity” of commercial fitness clubs have varying degrees of influence on college students’ sports consumption decisions. Therefore, this study proposes a new perspective: the influence of “indoor environment comfort level”, “service quality”, and “popularity” on college students’ sports consumption decisions in commercial fitness clubs in Dongguan is not dependent on the occurrence of COVID- 19. However, in the post pandemic era, COVID- 19 can raise college students’ awareness of health, safety, and hygiene, thereby increasing their focus on the fitness environment and service quality when making consumption decisions.

In summary, this study has comprehensively examined the results and provided an in-depth analysis of how various factors influence the sports consumption decisions of college students in commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19 by constructing a sports consumption decision system. Additionally, new insights have been offered through a comparison of the changes in each influencing factor before and after the pandemic of COVD- 19. Future research will focus on monitoring whether these factors continue to evolve and give rise to new variables in the post pandemic era, as well as on analysing the changes in influencing factors after the pandemic of COVID- 19 has fully subsided. In particular, investigating potential differences in these factors across different regions or cultural contexts will provide a good sustainability for future research. This approach aims to contribute to the body of knowledge on sustainability in sport consumption and offer more robust theoretical support for sustainable development in this field.

Conclusion and recommendation

This study utilises the AHP model to analyse the factors influencing college students’ decisions in sports consumption at commercial fitness clubs and establishes a corresponding decision system. The findings reveal that economic factors are the primary determinants of sports consumption among college students, followed by time factors, self-need factors, and other factors. The study demonstrates the application of the AHP model in constructing a sports consumption decision system, aimed at simplifying the complex issues surrounding the sports consumption decisions of college students in commercial fitness clubs in Dongguan City during the post pandemic era of COVID- 19. This approach facilitates an objective understanding of the key factors influencing college students’ sports consumption decisions and enables a more rational analysis of related social issues. It holds positive effect for promoting the comprehensive recovery of the sports consumption market and advancing the further development of the sports industry in the post pandemic era of COVID- 19. Particularly, it is reflected in the post-pandemic era of COVID- 19 in terms of how to tap into the market potential of university students, optimise the operational models of commercial fitness clubs, and promote the diversification of the sports industry. Based on these findings, this study proposes feasible business strategies and recommendations to further clarify the practical implications of the research conclusions for sports consumption and the development of the sports industry, thereby enhancing the academic value of the study.

1. Optimise pricing strategies to fully explore the college student sports consumption market

   1. Commercial fitness clubs can diversify their pricing strategy portfolios by introducing various “student fitness discount packages” tailored to college students, such as “couple packages” and “freshman packages”.
   2. Commercial fitness clubs are encouraged to collaborate with different colleges to distribute fitness vouchers, thereby reducing the financial burden on college students and increasing their participation in fitness activities.

• 2.
  Optimise the business model of commercial fitness clubs to enhance market competitiveness

  1. Promote the self-service fitness model in smart fitness clubs: for example, brands such as LEFIT and SUNPIG offer 24-h self-service fitness facilities, catering to the diverse time demands of college students.
  2. Optimise the location selection of commercial fitness clubs: establish well-known chain branches near colleges or within campus areas to reduce students’ commuting time and cost.
  3. Implement an online-offline integrated model: develop digital fitness courses to improve convenience for college students in the post-pandemic era of COVD- 19 through online fitness assistance, instructional support, and data tracking.

• 3.
  Leverage COVID- 19 as an opportunity to optimise the service environment and improve service quality to promote the diversified development of the sports industry

  1. Seize marketing opportunities and create consumption hotspots: commercial fitness clubs can use COVID- 19 as a health marketing opportunity by promoting the benefits of exercise in boosting immunity and relieving stress, thereby attracting more college students to participate in sports and related consumption.
  2. Enhance service quality and strengthen user retention: in the post pandemic era of COVID- 19, commercial fitness clubs can optimise the fitness environment by offering contactless services such as smart fitness equipment and self-service training guidance to ensure a safe experience for consumers.
  3. Increase health awareness among college students through government and college initiatives: COVID- 19 can serve as an opportunity to offer health education courses that raise awareness of physical activity and improve participation, helping college students develop a proper understanding of sports consumption.

Novelty of research

Firstly, the research methodology offers a novel perspective. In this study, we employed a combination of thematic analysis and the AHP model to examine the social issues surrounding sports consumption among college students. This approach enables the rapid simplification of complex sports consumption decision-making issues, offering new insights and approaches that could potentially address broader social problems within the realm of sport.

Secondly, the study addresses a gap in the literature. This study primarily focuses on the unique societal context of the post pandemic era of COVID- 19 and the specific group of college students engaging with commercial fitness clubs. Distinct from previous studies that have relied on a single theoretical approach, this study constructs a sports consumption decision system to provide a comprehensive analysis of the factors influencing college students’ sports consumption decisions. By integrating multiple theoretical perspectives, this study offers a more holistic understanding of the underlying determinants. Consequently, the findings of this study provide a theoretical basis for the healthy development of commercial fitness clubs and sports consumption in the post pandemic era of COVID- 19. In particular, it can provide valuable experiences and insights into the development of sports consumption for human societies that may once again face a social context similar to the COVID- 19 pandemic in the future.

Thirdly, the results of the study offer new insights. In comparison with previous research, this study stands out by employing weighting and ranking methods to highlight the relative significance of different influencing factors. This is a key innovation of the current study. Additionally, the findings introduce new research perspectives, which not only enrich the theoretical landscape of sports consumption decisions but also offer new thinking for future research in the field.

Limitation and prospectives

Firstly, with regard to the research methodology, although thematic analysis can effectively summarise the influencing factors of college students’ sports consumption decisions in the post-pandemic era, it may be subject to a degree of subjectivity. For instance, the objectivity of the research findings could be influenced by the participants’ subjective responses and the researchers’ biases during the processes of coding and interpretation. Similarly, the Analytic Hierarchy Process mainly relies on the subjective judgment and experience of experts when assigning weights to various factors. While this method simplifies complex issues related to sports consumption effectively, the objectivity of the results may still be affected by the personal expertise and cognitive biases of different experts. In future research, we plan to adopt more advanced quantitative methods in combination with big data analysis to enhance the objectivity of the findings.

Secondly, the semi-structured interviews conducted for this study primarily involved 15 college students. While this sample size has been theoretically saturated and validated, it adheres to the standards of qualitative research. Nevertheless, the findings may still be influenced by individual variability, particularly due to the diversity among college students in Dongguan in terms of their economic status, family backgrounds, personal perceptions, and patterns of sports consumption. Such variability may impact the representativeness of the study’s results. In future research, we will consider increasing the sample size while broadening the inclusion criteria for participants. For instance, consideration could be given to including all healthy adults rather than restricting the sample to college students alone. Such an approach would likely enhance the generalisability and robustness of the findings.

Thirdly, the geographical scope of this study is limited to Dongguan City, Guangdong Province, which may restrict the generalisability of the findings. Although Dongguan is a representative city with a developed economy, a large fitness market, and a significant student population, it cannot fully represent the entire country. Therefore, caution is required when applying these findings to cities with different fitness market scales and levels of economic development. However, in future research, we aim to extend the scope to other cities with similar economic conditions and sports consumption markets to improve the universality of our conclusions.

Authors’ contributions

Conceptualization, ZHIWEI CHEN; Methodology, ZHIWEI CHEN; Semi-structured interviews, NAIRUN ZHOU; Data Processing for Thematic Analysis, MAOQIANG XU; Creation and modification of figures and tables, YANQING YAN, JIFENG DONG; Writing—Original Draft Preparation, ZHIWEI CHEN; Writing—Review and Editing, ZHIWEI CHEN, ZAINUDDIN Z, A; Supervision, ZAINUDDIN Z, A. All authors have read and agreed to the published version of the manuscript.

Funding

There is no financial support or other benefit from commercial sources for the work reported in the manuscript.

Data availability

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. The authors can be contacted by e-mail upon request. Correspondence: chenzhiwei@graduate.utm.my.

Declarations

Ethics approval and consent to participate

The ethical aspects of this study received approval from the Research Ethics Committee of Universiti Teknologi Malaysia (approval No.: UTMREC- 2024–93), the institution to which the researchers are affiliated. The study adhered rigorously to ethical values governing human research, aligning with the fundamental principles outlined in the Helsinki Declaration. These principles include obtaining informed consent, protecting personal data, ensuring confidentiality, and adhering to relevant educational management regulations. In this study, all participants provided informed consent after receiving a comprehensive explanation of the research protocol. This robust ethical framework ensured the integrity of the study while safeguarding the rights of the participants.

Competing interests

The authors declare no competing interests.