2.1 Theoretical foundation

The S-O-R (Stimulus-Organism-Response) theory is derived from the S-R (Stimulus-Response) theory of behavioral psychology [27]. The S-O-R model was first proposed by Woodworth, and he explained that stimuli elicit different responses depending on the state of the organism [28]. Environmental psychologists use the S-O-R paradigm to describe the external stimulus, intervention variables, and behavioral outcomes [29]. The salient feature of the S-O-R theory is that consumer decision making goes through three distinct stages, i.e., external stimuli (S), affecting individual emotional states (O), and finally triggering behavioral responses (R) [30]. S-O-R theory is widely used to study offline and online consumer psychology and behavior. Currently, popular online consumption scenarios include: Fashion e-commerce retail [31], TikTok store [32], Live streaming e-commerce [33].

Kotler (1973) defined atmosphere as an environment designed to elicit emotional and behavioral responses from customers [19]. In 1974, Mehrabian and Russell pioneered the use of the S-O-R paradigm to examine the influence of atmospheric factors on consumer behavior [29]. In the positivist paradigm (S-O-R), numerous scholars argue that atmosphere is typically external stimulus (S) that influences the consumers’ internal organism (O), subsequently leads to behavioral response (R) [34–36]. In recent years, the S-O-R framework has been employed extensively to examine the online shopping atmosphere in a multitude of contexts. For example, the visual complexity of shopping websites [37] and the website informativeness and effectiveness of information content [38] have been identified as key factors. Chen and Li (2019) posited that the atmosphere of shopping festivities is the primary factor influencing consumer purchasing intentions [24]. Xie et al. (2023) investigated the influence of the atmosphere of an online shopping festival as a stimulus factor on consumers’ intention to engage in continued participation [26]. In accordance with the S-O-R theory, Cheng and Chen (2023) devised a creative atmosphere scale, and its process and methodology provide a very valuable reference for this research [39]. This study puts forth the concept of shopping festival atmosphere (SFA) from the vantage point of consumption perception stimulus, with a foundation in the S-O-R framework. SFA is evidenced by the implementation of extensive product promotions, the configuration of online platforms, the incorporation of entertainment games, the facilitation of social interaction, and other forms of expression. SFA is a stimulating element (S), consumers experience the joy, stimulation, and excitement associated with shopping carnivals (O), ultimately exhibiting a tendency to engage in continuous participation (R).

2.2 Definition of atmospherics

Atmospherics refers to the conscious designing of retailing environments to arouse buyers’ specific emotional responses and enhance their purchase possibility [19]. In other words, the term “atmospherics” is defined as the intentional control and structuring of environmental cues [40]. Since the early 1960s, researchers have investigated the effects of environmental elements (e.g., music and shelf space) on purchasing behaviors [40]. Kolter (1973) argued that the environmental cues rather than the products themselves play an important role in consumers’ decision making because of competition and only small differences among products [19]. Almost at the same time, environmental psychologists started to use the Stimulus-Organism-Response (S-O-R) paradigm to describe environments, intervening variables, and behavioral outputs [29]. In the context of retailing, consumers’ emotional response (O) plays a mediating role in the relationship between atmospheric stimuli (S) and behaviors (R). As consumers are influenced by the physical and network surroundings perceived at the point of purchase, the practice of establishing the intended atmosphere should be an effective marketing tool for retailers. How many dimensions do atmospherics contain? How are atmospherics supposed to take effect? And how do consumers perceive and respond to specific atmospherics? To answer these questions, atmospherics research has experienced several improvements during the past forty years. We classify the atmospherics literature into four categories: 1) atmospherics of physical stores; 2) atmospherics of online stores; 3) cultural atmospherics; and 4) festival atmospherics.

2.3 Atmospherics of physical stores

In his seminal work, Kotler (1973) posited that the store atmosphere encompasses four dimensions: visual, auditory, tactile, and olfactory [19]. In the early atmosphere literature, atmospherics almost always specifically referred to “store atmospherics”. Baker (1994) classified in-store elements into ambient (e.g., music, lighting, smell), design (e.g., color, wall covering, cleanliness, layout), and social factors (i.e., salespeople) [41]. Their study posited that environmental cues are the antecedent of merchandise quality and service quality, while the latter two factors are linked to store image. Referring to an earlier review of atmospheric effects on shopping behavior [40], the store atmospheric variables were synthesized into five categories: external variables (14 items), general interior variables (14 items), layout and design variables (14 items), point-of-purchase and decoration variables (10 items), and human variables (5 items). Similarly, the store atmosphere can be divided into three dimensions: social categories (3 items), physical categories (3 items), and ambient features (2 items) [42].

In earlier research, environmental psychologists applied pleasure-arousal-dominance (PAD) to predict consumers’ approach-avoidance behaviors in store atmospherics [43]. Chebat, Chebat, and Vaillant (2001) found that background music fit rather than music pleasure or music tempo-induced arousal helps salespersons attract and persuade shoppers [44]. Later, researchers revealed that retail atmospherics can be useful in customer relationship management by creating both utilitarian and hedonic shopping values for customers [45]. In fact, store ambiance is a conscious effort to create a retail environment that makes retail more attractive to consumers than the product itself [46,47]. As retailers look to differentiate their service offerings, the atmospherics of music and aroma can be deliberately adapted to cater to young fashion shoppers’ preferences [48]. Their experimental study showed that the volume of music and the presence of a vanilla aroma both significantly impact on shoppers’ pleasure levels and satisfaction levels, which subsequently lead to increased time and money spent in the store.

In recent years, the field of market research has recognized color, lighting and music as atmospheric tools that can elicit better multi-sensory congruent cues (such as visual, auditory and olfactory) and ultimately influence purchase intentions [49,50]. There have been many interesting findings on the study of store atmosphere on consumer behavior. Burns, Dato-on, and Manolis (2015) reported that Hispanic consumers in the U.S. do not like crowded stores [51]. Haj-Salem, Chebat, Michon, and Oliveira (2016) investigated more than 900 North American shoppers and found that female shoppers’ loyalty to malls is more affected by shopping atmospherics than that of male shoppers [52]. In young shoppers’ sections and pop-up stores, playing modern and stylish music can increase their desire to stay in the mall [50]. In global apparel retailer’s stores, customers in a non-English speaking country are more likely to buy when music is played in English [53]. The olfactory congruence of a shopping mall has a positive effect on bridging the mall atmosphere and brand image [54]. In young shoppers’ sections and pop-up stores, playing modern and stylish music can increase their desire to stay in the mall [31]. The visual atmosphere of a coffee shop can have an impact on consumers’ expectations of different coffee flavors [55].

2.4 Atmospherics of online stores

In the first years of the 21^st century, with the emergence and booming of electronic commerce, online retailing became a new retailing format. Eroglu, Machleit, and Davis (2001) took the first step to investigate the atmospheric nature of online retailing [56]. As virtual shopping outlets, online stores in that era had no opportunity to apply VR/AR (Virtual Reality/Augmented Reality) technology, so that the online environment lacked some of the sensory appeal of traditional retail atmospherics. The entire online store environment is reduced to a computer screen. Eroglu et al. (2001) suggested that online atmospheric cues should be classified into two groups: high task-relevant cues and low task-relevant cues [56]. High task-relevant cues include descriptions, prices, terms of sale, delivery and return policies, and navigation aids (e.g., site map, guide bar), which are defined as the dimensions of site informativeness and effectiveness of information content. Low task-relevant cues include colors, borders, layout, background patterns, type styles and fonts, animation, music and sounds, which are defined as the site entertainment dimension. These atmospheric cues can create a mood or an image for the site.

Following the work of Eroglu et al. (2001) [56], Dailey (2004) proposed that navigational cues are another important dimension of web atmospherics, as restrictive navigational cues hinder consumers from using a web site, which, in turn, arouses psychological reactance and leads to avoidance behavior [57]. Similarly, Mummalaneni (2005) argued that navigation design is an important atmospheric cue for a virtual store to help consumers proceed smoothly through the purchasing stages [58]. Similar to brick-and-mortar stores, web-based stores have the atmospheric dimension of storefront designs. The online shoppers react more positively to web-based stores with a thematic and picture-based design than those with a non-thematic and text-based design [59]. A picture-based information display provides consumers a convenient store image and leads to their expectations of merchandise quality. Some scholars differentiate between the various elements that comprise the atmosphere of online stores, including information, navigation quality, as well as platform design [60,61].

With the development of web technology, multimedia controls are embedded in the online marketplace. Wu (2008) revealed that both background music and color factors have a significant effect on consumers’ emotional response, which subsequently affects their intention to purchase in online retailing settings [62]. Specifically, fast-tempo background music is linked to online shoppers’ arousal emotions and pleasure more than slow-tempo music. A warm background color is linked to higher levels of arousal, whereas a cold background color is linked to higher levels of pleasure. Gao (2014) investigated the impact of website atmospheric cues (informativeness, effectiveness, and entertainment) on consumers’ purchase intention and satisfaction [63]. Their empirical study indicated that flow fully mediated the atmospheric cues with regard to purchase intention and satisfaction. Khan, Wang, Ehsan, Nurunnabi, and Hashmi (2019) classified web atmospheric cues into five categories: visual cues, information cues, ethics cues, social cues, and security cues [64]. Ettis (2017) found that the blue-hues background of an online store induces more consumers’ flow (Enjoyment and concentration) than yellow [65]. The unique shopping atmosphere of e-commerce live streaming is reflected in three aspects: bullet screen, guide information as well as parasocial interaction [66]. In live streaming e-commerce, participants generate high arousal when they are in an atmosphere with complex background images and fast-paced background music, and fast-paced music seems to better meet the needs of consumers [33].

2.5 Cultural atmospherics

Cultural factors influencing shopping atmospherics have two aspects: 1) shopping atmospherics are adapted to cultural atmospherics; and 2) shopping atmospherics adopt cultural elements. As Kolter (1973) noted in the very early stage of atmospherics research, people of different cultures have different ideas about colors, noisy environments, and aromas [19]. The establishment of shopping environments must make the target shoppers feel agreeable and comfortable. For instance, in the Chinese culture, red is a “happy” color. A red background color of a web-based retailing environment may result in a more pleasant feeling for Chinese consumers than a blue one [62]. According to the distinctiveness theory, each ethnic group has a unique community with their own cultural values. Ethnicity plays an important role in Hispanic-American consumers’ shopping preferences and habits. Ethnic cues of store atmospherics are highly related to Hispanic consumers’ patronage behavior [67]. Sometimes, culture creates a “must-have” for a specific segment. For example, the atmospherics for the Christmas shopping season must have the cultural cues of Christmas music, Christmas trees, and Santa Claus [20].

Second, cultural atmospherics are adopted to create differentiation. Ha and Jang (2010) reported that Korean-themed restaurants in the U.S. attract not only Korean but also American customers [68]. The Korean ethnic atmospherics, such as Korean-style interior design, music, mood, and layout, were verified to have moderating effects on the links between service and food quality and customer satisfaction and loyalty. A recent study by Naletelich and Paswan (2018) showed that toward two different art genres (realist vs. abstract), consumers’ purchase intention has different antecedents [69]. The purchase decision making of consumers who are exposed to the retailing settings of realist art genres is significantly associated with product aesthetics and symbolism, whereas the purchase decision making of consumers who are exposed to the retailing settings of abstract art genres is significantly associated with hedonic and utilitarian motivations and openness to art.

2.6 Festival atmospherics

There are a few scattered studies in the literature on festival atmospherics and their effects on consumer behavior. However, we collect relevant references from three domains: festival shopping (and holiday shopping), shopping festivals, and large online promotions. Combined with the above review, we identify four categories of festival atmospheric cues:

3.1 Initial item Generation

Following DeVellis (2021) ’s recommendation, the authors apply three techniques to generate a sample of items: 1) A comprehensive literature review is conducted on shopping atmospherics, including not only festival atmospherics but also atmospherics of physical stores, online stores, and culture (see Sec. 2) [79]. A list of possible measurement items is generated. 2) The authors conduct semi-structured in-depth interviews to confirm possible items and add new items to the list. 3) Finally, the coarse list of measurement items is submitted to a panel of five experts with research expertise in electronic commerce. The five experts distinguish the coarse items and select the most applicable items to include in the draft of the questionnaire.

Content analysis.

Two research assistants corrected errors and then checked the machine transcription. Two experts coded the transcriptions, assessing the content validity and relevance of the items. They also further assess and review the applicability of the classifications attributed to each item. This resulted in 43 coarse items after synthesizing all atmospheric cues reported by respondents. These items have been translated into English and are presented in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Shopping festival atmospherics reported by respondents in the interviews.

https://doi.org/10.1371/journal.pone.0324989.t002

3.2 Purifying the measurement scale

Data collection.

In the third and fourth steps, the present study purifies the measures. The data collection is outsourced to a third-party company, wjx.cn (https://www.wjx.cn/), one of the largest online survey companies in China. There is such a large participant population involved in SDSFs that the technique of random sampling is needed. The company has the ability to do so because it has surveyed a great deal of respondents and volunteers over several years, including college students, white-collar workers, and urban residents.

According to the survey instructions, the respondents decide whether to fill in the questionnaire voluntarily. They can scan the QR code or open the link to fill in the questionnaire through their mobile phones, and receive the reward after completing the questionnaire. The total sample size is 738. After deleting uncertain samples (e.g., incomplete answers, all the same answers, minors), the valid sample size is 686 (Six samples involving minors were deleted from the original data). Table 3 displays the profile of the respondents. As shown in Table 3., more than 98% of the respondents have participated in SDSFs, while approximately 80% participated in SDSFs almost every time. Approximately 75% of the respondents are from the young generation aged 20–40. However, there are still a quarter of the respondents who are middle-aged. In addition, more than half of the respondents have frequent chat experiences (5–19) during SDSFs. The demographics show that SDSFs is indeed an influential sales event attracting a large participant population.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Descriptive Statistics of Respondents (Sample 1 and Sample 2).

https://doi.org/10.1371/journal.pone.0324989.t003

In order to ensure the scientific and normative nature of the research, the collected data can be randomly divided into two parts. According to the research method and process of Dimitrova, Ilieva, and Stanev (2022), EFA and CFA were performed on the two parts of data respectively [80]. The proportions of gender, age, income, education, occupation, participation counts, and chat counts were maintained in the subsample. The demographic characteristics of the participants in the two randomly grouped sub-samples meet the requirements of the independence chi-square test. The numerical characteristics (mean and standard deviation) of the survey sample are shown in Table 4.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. The measure of the sample data (n = 686).

https://doi.org/10.1371/journal.pone.0324989.t004

Purifying the measurement.

The present study employs exploratory factor analysis (EFA) with a varimax rotation to determine the dimensions of the scales in sample1(n = 343). Items with factor loadings lower than 0.4 or cross-loaded on more than one factor are eliminated to ensure each attribute loads only on one factor [81]. In addition, the internal reliability of each factor is measured using Cronbach’s alpha. Churchill (1979) suggests that a low alpha coefficient indicates the item has a low impact on the measurement of the construct [78]. Therefore, factors with Cronbach’s alphas of less than 0.7 are eliminated [82].

The EFA suggests that the entire 21 items load on 3 factors just as classified above: social cues, cultural cues, and online cues. The first factor consists of 6 items with a Cronbach’s alpha of 0.920. However, the item “Alibaba’s shopping theme gala” is deleted from analysis due to its cross-loading on another factor. The second factor consists of 6 items with a Cronbach’s alpha of 0.903. However, the item “Impulsive purchasing is tolerated that day” is deleted from analysis due to its cross-loading on another factor. In addition, the third factor consists of 7 items with a Cronbach’s alpha of 0.899. By using Statistical Package for Social Sciences 21.0, the appropriateness of factor analysis is verified. The Kaiser-Meyer-Olkin (KMO) value is 0.902, which is expressed as “marvelous” (statistic>0.9) by Kaiser (1974) [83]. Bartlett’s test of sphericity yields 4452.192 (p < 0.001), indicating the factor analysis is proper. The remaining 19 items are translated into English as listed in Table 5. The items include factor 1 (social cues, 6 items), factor 2 (cultural cues, 6 items), and factor 3 (online cues, 7 items). The Cronbach’s alpha coefficients for individual SFA factors range from 0.899 to 0.920, indicating highly reliable results.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. Exploratory Factor Analysis Results: Sample 1 (n = 343).

https://doi.org/10.1371/journal.pone.0324989.t005

3.3 Reanalysis of measures

Criterion-related validity and cross-validation.

Table 5 shows the CFA results of the sample 2 (n = 343). After purification once more by the sample 1 (n = 343), the measurement model of the validation sample is found to be statistically significant. The results indicate a good model fit to the data, =2.927, p < 0.001, CFI = 0.939, TLI = 0.926, and RMSEA = 0.075. In addition, as shown in Table 6., construct and discriminant validity concerns are not found (i.e., composite reliabilities>0.7, average variance extracted (AVE)>0.5) [87]. The square root of AVE (the bold diagonal number in the table) is higher than the correlation coefficient between structures, indicating that discriminant validity is satisfied [88], as shown in Table 7.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 6. Performance of the measurement scale from CFA (sample 2, n = 343).

https://doi.org/10.1371/journal.pone.0324989.t006

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 7. Correlations of the measurement scale from CFA (sample 2, n = 343).

https://doi.org/10.1371/journal.pone.0324989.t007

4.1 Theoretical implications

This research has a number of theoretical implications. First, according to the S-O-R theory, which focuses on “environmental stimuli can trigger individual emotional responses”, combined with previous research findings, a shopping festival atmosphere scale was developed. In recent years, with the widespread attention of scholars on online shopping atmosphere, scholars have described the connotation of shopping festival atmosphere from different perspectives [24,26]. The research findings of this paper not only establish the concept of shopping festival atmosphere, but also expand its theoretical connotation and application.

Secondly, the SFA scale developed in this paper is different from previous measures of atmosphere because it contains three components: cultural, social, and online atmosphere, and it is described in detail. The shopping festival atmosphere not only adds the concept of “festival” to the original online shopping atmosphere, but also takes into account China’s unique cultural background and the most extensive online and offline social interaction. This concept extends the application of atmosphere research to the phenomenon of “online shopping festival”. The connotation of the shopping festival atmosphere identified in this study is discussed as follows:

Cultural clues refer to the particularity and sense of ritual that must be consumed in online shopping festivals. SDSFs was initiated from Singles Day, an unofficial festival. However, the festival creates a custom for the SDSFs. It is held on November 11 every year. This is a strong cultural hint in many participants’ eyes. They treat SDSFs as a special day. They have the custom or habit of accomplishing what must be done. Just as Zhao and Wan (2017) found in their study that participating in SDSFs is a ritual event [35]. In addition, they need to prepare in advance for these “must-do” behaviors. On the other hand, the consumerist culture is booming, associated with the SDSFs in China. The Confucian values of “keeping face” and “listening to others” exert a moderating influence on the effect of the atmosphere of the online shopping festival on purchase intention [24]. The happily crowded environment (“Renao” in the Chinese language) is a positive atmospheric cue for Chinese participants. This type of cultural cue gives participants a symbolic link between the SDSFs and the “Buy! Buy! Buy!” slogan.

Social cues mean that participants actively participate in the most extensive topic communication and knowledge sharing during the SDSFs, both offline and online. The extensive social interaction on the shopping day (such as topic discussion, shopping communication, experience sharing, etc.) has gradually become the biggest feature of Singles Day shopping day [9,25]. Through online and offline advertising, single consumers can establish a sense of identity with Singles Day [35]. Furthermore, Chen and Li (2020) demonstrated that the festive atmosphere exerts an influence on consumers, which is shaped by the social medias and the people around them [9]. All of these social environments should ultimately result in a crowded atmosphere that is safe, secure and happy. Otherwise, participants may experience negative emotions and misbehave [15,18].

Online cues are the evaluation of the comprehensive service of the network platform by the consumers participating in the online shopping festival. As an online sales event, the SDSFs had to make changes for special days. This study found that in addition to the traditional online climate of information, efficiency and entertainment, the fair climate was also a new feature perceived by participants. They need very good deals [25]. They don’t want to be cheated by artificially inflated prices and they don’t want to go to great lengths to play the coupon game. In addition, the authors found that size itself is an attribute of online atmosphere. Studies by Tzeng, Ert, Jo et al. (2020) show that product information and product quality are prerequisites for consumers to feel satisfied after participating in online shopping festivals [10]. The website informativeness and effectiveness of information content can help to improve consumer satisfaction and purchase intention [38]. Participants’ perception of the promotional scale was that they expected more opportunities to get what they really wanted at a lower price than usual, more opportunities to avoid restrictions on the discounted items they wanted, and ultimately more benefits from SDSFs.

4.2 Managerial implications

This study has established a reliable scale for measuring the “shopping festival atmosphere”, and has also provided managers with a tool for objectively understanding consumers’ perception of the shopping festival atmosphere. The results of this study have significant implications for the practice of management, which can be grouped into many areas: Firstly, managers are able to utilize this scale to ascertain whether the shopping festival atmosphere has been successfully created, particularly when the SFA is low. Subsequently, managers are able to modify their strategies in accordance with the aforementioned findings. Secondly, the utilization of cultural, social and online cues can also serve as a valuable means of evidence for managers seeking to establish a shopping festival atmosphere. Thirdly, these cues can be integrated with actual circumstances to develop innovative designs that enhance consumers’ purchase intention during the shopping festival. Ultimately, the findings of this study offer valuable insights that can inform the successful implementation of online shopping festivals in other countries, particularly in light of the growing prevalence of such events in China and the global convergence of e-commerce platforms.