Experiencing Carnival in Brazil: participation, motivations and outcomes Open Access

Carnival in Brazil has become “arguably the most elaborate, widespread recreation of the logic of the festival anywhere in the contemporary world” (Parker, 2015, p. 215). This collective ritual enables Brazilians to celebrate important traditions and, following Bahktin (1968), escape from everyday life. Carnival has been the subject of many studies by historians, anthropologists and social scientists from a variety of perspectives, including the social, political, economic, race, identity, class and gender aspects of the event (Green and Scher, 2007). In Brazil, the analysis of Carnival has often detailed the form, function and effects of the event, particularly its relationship with different cities. However, as De Matas (2014, p. 120) notes: “Few scholars, however, have examined the possibilities of analysing the senses and emotions of the people who experience carnival, either as spectator, participant, or artist, or those who adopt other positions in relation to Carnival.”

One of the reasons for this downplaying of individual experience is possibly the overwhelmingly collective nature of the event in Brazil. Carnival is seen as being an essential part of national and local identity, a means of bringing people together and celebrating communality. However, as a number of authors have pointed out, Carnival is also a field of tension and differentiation. People attending Carnival take on different roles: as spectators, performers and co-creators of the Carnival experience, by becoming part of the Carnival crowd (Jaguaribe, 2013).

It is therefore germane to ask – Is the experience of Carnival also differentiated between participants, and do different experiences lead to different outcomes in terms of satisfaction and intention to participate in the future? Such questions are important for the future of Carnival, since the event cannot exist without participation and a willingness of participants to reproduce Carnival practices.

This paper presents an analysis of Carnival participant experiences in Brazil, drawing on survey data from different parts of the country in the post-COVID period (2023–2025). Analyses of post-COVID developments are vital in providing pointers to the future of Carnival in Brazil and elsewhere. The aim of the paper is to analyse the dimensions of the experiences of Carnival participants in Brazil, and how these differ between individuals, groups and cities. The analysis presented here can contribute to important debates about the nature of Carnival, including the effects of trends towards mediatisation and touristification, which have raised questions about the contemporary authenticity of the celebrations.

Few events are as iconic as Carnival in Brazil. As Abreu and Brasil (2020) commented: “Carnival is one of the most powerful images of Brazil in the contemporary world.” This collective ritual enables Brazilians to celebrate important traditions and, following Bahktin (1968), escape from everyday life. More recent interpretations have viewed Carnival as a lens through which to view the tensions and cohesions of Brazilian society, which are magnified during the festivities.

Da Matta (1982) addressed the question of whether Carnival in Brazil can be considered a unique event. Comparing Brazilian Carnival with the Carnival in New Orleans in the USA, he concluded that in many respects they are similar (for example, in the transformation of public space, the use of costume and erotic language) but beyond this, “everything is radically different” (p. 166). Research on Carnival in Europe also emphasises the deep social and religious roots of the festivities there, in contrast to the relatively recent development of Carnival in Brazil. DeWaal (2013) analyses the history of the Carnival in Cologne, for example, highlighting the use of symbols dating back to the Middle Ages.

Research on Carnival in Brazil tends to emphasise the social dimension of the event, with collective rather than individual experience playing the primary role. There are many studies of the evolution of Carnival and the relationship of the form and significance of festivities to social, economic and political factors. Less attention has been paid to the ways in which individuals experience Carnival, and how they articulate their experiences to the collective aspects of the event. Most academic studies of Carnival emphasise anthropological and sociological perspectives, highlighting the Bakhtinian conceptualisation of Carnival as a ritual of reversal, in opposition to daily life (Bahktin, 1968). From this perspective, Carnival becomes a time out of time (Falassi, 1987) in which people can play with alternative identities (Simons, 2020).

The extensive literature on Carnival in Brazil tends to focus on the interplay between Carnival celebrations and various forms of inequality, between classes, races and urban spaces. Abreu and Brasil (2020) argue that previous interpretations viewed Carnival “as a space in which Brazilians come together to celebrate commonly held cultural traditions or as an effective escape valve allowing common people to forget the woes of everyday life.” More recent historical and anthropological studies have seen Carnival as a window that offers glimpses of the tensions, rivalries and alliances of everyday life, “magnified and played out in public on the festival days” (Abreu and Brasil, 2020).

Many studies concentrate on themes such as identity and performance (Pravaz, 2012; Guimarães, 2025), cultural authenticity, sexuality and race (Williamson, 2012; Da Silva, 2005). The different cultural and social elements of Carnival, including music (Crook, 2013), popular culture (Henfrey, 2024), religion (Oosterbaan, 2017) and protest (Magalhães and Queijo, 2015), have frequently been analysed. Carnival is also an event with many organisational aspects, including governance (Cabral and Krane, 2018), security and crime (Kane, 2010). Carnival also produces a range of effects on Brazilian society, including the representation of Carnival in politics (Sheriff, 1999) and the media (Pravaz, 2012), attracting tourism (De Rosa, 2013), generating income and jobs (Moreira and Gonçalves, 2020).

Many studies have also highlighted the historical evolution of Carnival, charting the shift from the 19th century Entrudo, or proto-Carnival, to the development of the Grandes Sociedades (Grand Societies) by the middle classes in the late 19th century and the emergence of samba schools in the 1930s. Parker (2015) draws the distinction between Carnaval de rua, literally “street carnival” or the carnival of the poor, and Carnaval do Salão operated by the Grandes Sociedades, with organised parades and floats. Blocos, or Carnival groups, originally formed among the poorest sections of society. It was in the blocos that the influence of samba was first felt, and samba schools arose out of larger blocos linked to specific neighbourhoods. This formed a strong bond between the urban fabric and Carnival, which is still reflected in the styles of different cities. Miguez (2012 p. 90) argues:

For example, Guimarães (2025) reviews the different styles of Carnival, highlighting the Southeast, with São Paulo and Rio de Janeiro for their spectacular parades and samba schools for their international projection. In the Northeast, Carnaval de rua is characterised by vibrant music and Afro-Brazilian traditions, particularly in Salvador and Recife/Olinda. In Minas Gerais, Carnival is more nostalgic, with historic cities such as Ouro Preto and Diamantina acting as a backdrop for the blocos. Many studies therefore concentrate on a particular city in Brazil, such as Rio de Janeiro (Pereira and Ferreira, 2009), São Paulo (Dozena, 2009), Salvador da Bahia (Moura, 2009) or Recife (Silva, 2019). The evolution of Carnival has also been studied more recently in cities like Belo Horizonte, where Fernandes and Quintão (2022) identify four recent periods of development: firstly, from 2009 emancipatory occupation of the public space through the organisation of blocos; secondly, public sector intervention from 2010 to 2012 attempting to limit disturbances in the city centre; thirdly, conquest of the emancipated space, from 2013 to 2014, with an increase in social movements and a period of expansion and strengthening of the carnavalesco movement; and, finally, attempts at commercialisation (with public sector support) from 2015 to 2018, marked by a large brewery using graphic materials produced by the blocos.

Many cities actively monitor the tourism and economic impacts of Carnival to justify investment in the event and tourism promotion. For example, Belo Horizonte has carried out regular Carnival surveys since 2013, including questions on visitor profile, participation, tourism activity and expenditure (Falcão and Isayama, 2021). The results show a steady increase in participation, economic impact, job creation and attraction of tourism (Moreira and Gonçalves, 2020). The number of blocos participating in the Carnaval de Rua in Belo Horizonte grew from 24 in 2012 to over 500 in 2019. This reflects a generalised process of recovery of street Carnival, which was in danger of disappearing in major cities such as Rio de Janeiro (Farias, 2014) until a revival after the Millennium. In 2025 Belo Horizonte recorded 460 blocos (Belotur, 2025), also indicating a high level of post-COVID recovery.

This review indicates that the majority of Carnival research in Brazil, as elsewhere, has focussed on collective issues rather than the individual experience of Carnival. This is changing, however, as links are made with the emerging field of event studies (Marques et al., 2021). Indeed Frost (2015) makes the case for a separate field of “festival studies” within the event studies field, underlining the specificities of events such as Carnival. Frost (2015, p. 573) also notes that “because experiencing carnival is fundamentally a process of individual perception, a generalised account will always be found wanting.” This also highlights the fact that Carnival includes both personal and social dimensions of experience and identity formation, which are considered in the research conducted in this paper.

This tension between the social and individual experience of events is an area of research that has been explored by the ATLAS Events Experience Project (de Geus et al., 2016; Richards, 2017; Coetzee and Pourfakhimi, 2020). This collective research initiative was initiated by the Association for Tourism and Leisure Education and Research (ATLAS) to analyse and compare the experiences of event participants worldwide. The project developed an Event Experience Scale (EES) as a standardised measurement tool, which has been applied at events around the world (de Geus et al., 2016). From 2015 onwards the EES has also been applied to Carnival events in Brazil, generally concentrating on major Carnival cities, including Recife, Rio de Janeiro, São Paulo and Belo Horizonte. This has also enabled comparisons with Carnival events and festivals in other countries (Richards and Marques, 2024). This research has revealed that Brazilian Carnival is characterised by relatively high levels of excitement and atmosphere, although active involvement levels were often low, thanks to the spectating role of many carnival goers in Brazil (Richards, 2017).

This study of the experience of Carnival in Brazil attempts to answer the following research questions: What factors are important in shaping the experience of Carnival in Brazil? What influence does the social context of participation have on the individual experience of Carnival? What outcomes does the experience of Carnival create, for example, in terms of satisfaction and intention to return? To address these questions, we analyse survey data on the experience of residents and visitors as Carnival participants, including those participating in blocos.

Surveys of Carnival participation were conducted via online survey links distributed to a convenience sample of individuals in Brazil in the post-Carnival period in the years 2023, 2024 and 2025. These years were chosen to reflect the post-COVID-19 period. The results therefore arguably give a picture of relatively “normal” patterns of Carnival participation. Although there are official dates for the celebration of Carnival in most places in Brazil, the timing and extent of the celebrations vary from one location to another. Many people also celebrate Carnival over a relatively long period of time, rather than just the few days of the official celebrations. This may make it hard to compare experiences that are spread over a greater length of time and, in some cases, across different locations. To minimise this problem, the dates indicated in the survey related to the week before Ash Wednesday. The sample was gathered through a snowball method, starting with individuals known to the researchers who indicated they participated in Carnival. They were asked to distribute the survey links to other people they knew who participated in Carnival. In each year of the survey, an initial mailing was followed by two reminders to increase the response.

The questionnaire included questions on the profile of respondents, their Carnival attendance, motivations for participation, social context and experience. The concept of identity was operationalised in terms of both social and personal dimensions, following the argument of Vignoles (2018), and recognising the importance of both dimensions in the Carnival literature. A 10-item version of the EES was used to measure participant experience, using a 7-point Likert scale. The order of items on the EES was randomised for each participant to avoid repetitive answers. The EES, first developed by de Geus et al. (2016), is based on the three “classic” experience dimensions – cognitive engagement, physical engagement (conative) and affective engagement, widely recognised in the literature (see review by Getz, 2008) – and an additional “novelty” dimension identified through empirical research. Following the development of the EES, the scale has been extensively tested at a range of different types of events (cultural events, leisure events and sporting events, festivals and carnivals) in various locations (different countries and types of venues, both indoor and outdoor) (see Coetzee and Pourfakhimi, 2020; Richards, 2020; Marques et al., 2021; Oklevik et al., 2022; Tran-Pham et al., 2024). These studies have largely confirmed the dimensions identified in the original EES formulation, which indicates that the scale is robust across different settings.

A total of 1,553 responses were received in the three survey years, with a fairly even distribution between the years. A screening question was asked about participation in events during the Carnival period in each year. This revealed that the participation rate was 67% in 2023, 64% in 2024 and 69% in 2025. These data indicate a fairly high and relatively stable level of Carnival participation following COVID. A comparison of participants and non-participants in the sample indicates that respondents who participated in Carnival were more likely to be younger, with lower incomes and lower education levels than non-participants. There was no significant difference in participation in terms of gender.

Data from participating respondents were screened for missing values, with those not responding to the experience evaluation questions being removed from the analysis. The final dataset was therefore composed of a total of 1,081 valid responses from Carnival participants, which were analysed with SPSS statistical software.

The adapted version of the EES used to capture attendee experience included items covering elements of cognitive engagement, affective engagement, physical engagement and novelty (de Geus et al., 2016; Richards, 2020). The Cronbach's Alpha score for the scale was 0.798, indicating a high degree of reliability.

In addition to the standard survey questions, respondents were also invited to comment on their Carnival experience via an open question at the end of the survey, allowing more detailed information to be supplied.

The respondents were relatively highly educated and had relatively high incomes by Brazilian standards, although those participating in Carnival during the study period had lower incomes and education levels than non-participants. Respondents celebrating Carnival were most likely to be resident in the location where they celebrated Carnival (59%) and were significantly more likely to be female (69% of residents responding) (Table 1). Respondents resident in the Carnival location were significantly more likely to be female and had a younger age profile. There were no differences in education or income between residents and visitors.

The experience items that scored highest overall were “I was excited”, “I was aware of my own values” and “I was active”. This underlines the experience of Carnival as an activity that generates a sense of excitement and which underpins collective values. There were no significant differences in the experience of visitors and residents in terms of the top four experience items. Only two items scored significantly higher for visitors: “I felt a sense of adventure” and “I used my intellect”. These results might be explained by the novelty and unfamiliarity of being in a different setting. Being outside of your usual environment may bring a heightened sense of adventure and can also be more demanding cognitively. People may make more mental effort and encounter more stimuli. This may explain the positive correlation between “I felt a sense of adventure” and “I thought this event was unique” (r = 0.42).

These results underline the high levels of excitement generated by active participation in Carnival, which also underlines the personal values and identities of participants. The fact that there are few differences in the experience for visitors and residents highlights the generalised nature of the Carnival experience in Brazil, which enables visitors to experience the event in much the same way as locals. The vast majority (97.5%) of the visitors are domestic tourists, which increases the shared nature of the experience. There was also no significant difference in the total experience score for the 10 elements between visitors and residents (Table 2).

The generalised excitement of the Carnival experience was also underlined in many of the comments from participants, many of which emphasised the idea of Carnival as an event for all.

The idea of an event for all is also supported by the experience of participants, which shows relatively few differences by gender, education level, age or income. There was no significant difference in Carnival experience or outcomes by gender.

In terms of location of Carnival participation, the highest average experience scores were found in Recife, Rio de Janeiro and Belo Horizonte (Table 3). Recife is renowned as a city with a distinctive style of Carnival, combining Frevo and Maracatu music and dance and the immense Galo da madrugada parade, and Rio de Janeiro is widely viewed as the epitome of Brazilian Carnival with its famous samba schools (Guimarães, 2025). Belo Horizonte has developed its Carnival into a major festival in recent years by actively encouraging the formation of blocos and investing in marketing, infrastructure and security. The relative security of Belo Horizonte was mentioned as a motivation by some visitors:

Some respondents mentioned small-scale events as being attractive, also because of greater security:

Levels of animation were highest in the bigger cities of Recife, Rio and Belo Horizonte. In smaller locations, the variety of entertainment options is likely to be more limited. One participant said there was a need to “To foster the culture of traditional carnival in smaller municipalities. Currently in the smaller municipalities there is only funk … Nothing conducive to families and children”. This was also reflected in the motivations of participants, which were much more related to entertainment in large cities such as Recife (72%) and Rio (68%) than in smaller cities in Minas Gerais (55%).

In larger cities the reproduction of Carnival as a spectacle for tourists was also commented on by some respondents:

But not all respondents saw these developments as positive:

The idea that Carnival can also reflect social divisions was noted by other participants. One teacher said they were losing purchasing power relative to other groups and that

The results indicate each location can produce a differentiated experience of Carnival dependent on the audience, the programming and the setting. Interestingly, São Paulo has relatively low experience scores for a large city, which may relate to the relatively late development of Carnival traditions there, as well as the perceived lack of safety (Guimarães, 2025). Carnival participants there reported being less active, whereas in Rio de Janeiro, groups of participants were smaller, with fewer children, and this seemed to be linked to higher experience scores.

Many experience items exhibited significant differences between locations (Table 4). In general, the highest experience item scores were in Recife, although Rio scored significantly higher for “I learned something” and “I acquired new knowledge”. This seems to suggest that the extensive and spectacular programming found in Rio particularly enhances the cognitive aspects of Carnival experience. For Recife and Rio de Janeiro, which also had the highest overall experience scores, “I was excited” was the highest scoring item, and Recife had the highest ratings for excitement and activity. This seems related to the vast crowds attracted to these cities during Carnival and the spectacular programming.

Belo Horizonte did not score highest on any individual experience item, although it was ranked third after Recife and Rio in terms of total experience score. Data from the tourism administration of Belo Horizonte also indicate that the city has been improving in terms of visitor satisfaction in recent years, with a score of 7.3 on a 10-point scale for visitors in 2017, rising to 8.8 in 2025 (Belotur, 2025). The city claims that this improvement is related to greater public investment in Carnival which, according to one respondent, influences the experience and structure of the event: “Belo Horizonte differs from other street carnivals due to its essentially political character.”

In addition to issues related to crowds and security, there is a feeling among some respondents that the promotion of tourism leads to a loss of identity:

The mode of participation in Carnival seems to have an influence on experience (Table 5). For example, participation in Carnaval de Rua is positively correlated with total experience score (r = 0.19) and intention to return (r = 0.24). Celebrating at home was linked to less positive experiences, and participation in private parties or watching on TV had no significant relationship with experience. However, even those celebrating at home generally had a better experience if they also engaged in making preparations for Carnival (making decorations, practicing choreography, etc.).

If we compare those participating in Carnaval de Rua only, then they are significantly more likely to be animated and have a sense of adventure, reflect on ideas and think the experience was unique (Table 6). There was no difference in total experience score, or the size of the group, but they spent less (particularly residents) and had a higher intention to return, emphasising the importance of this routine practice for many. In terms of motivations, the atmosphere of Carnival was more important for those attending Carnaval de Rua, as were entertainment, being with family/friends, seeing a specific element/performer and feeling part of a community. Meeting new people was generally less important. This seems to confirm the community base of Carnaval de Rua being important for the experience. Those participating in Carnaval de Rua were also more likely to have celebrated all their lives or to be professionally involved in Carnival.

Over half the respondents participating in Carnival indicated that they were members of a bloco or samba school, which reflects the strong growth of these groups in recent years (e.g. Belotur, 2025). Those participating in these groups, which are integral parts of the Carnaval de rua, also led to significantly higher experience scores (mean total 5.2 for participants vs 4.8 for non-participants). Group participation was also linked to higher levels of Carnival ritual, including playing Carnival music, making decorations, dancing, singing and participating in group activities. Belonging to a Carnival group may strengthen the “Interaction Ritual Chain” of the event, generating greater emotional energy and producing more positive outcomes for participants (Bargeman and Richards, 2020). Although there was no difference in the cognitive experience of bloco members and others, those in a Carnival group reported stronger conative, novelty and physical experience dimensions. Those in a group also reported higher experience item scores in all areas except those linked to thinking, reflection and intellect.

Carnival generates a large amount of excitement, or what Collins (2004) termed “emotional energy”, which stimulates people to participate in the ritual again. In line with this theory, there was a very high level of correlation between Carnival participants who experienced excitement and satisfaction with the experience (r = 0.60) and with their intention to return (r = 0.53). Revisit intention is also strongly correlated with the mean total experience score (r = 0.45). However, it should be noted that these correlations do not establish causality, which should preferably be the subject of further research.

The differences in Carnival experience are also reflected in the outcomes for different locations. Levels of satisfaction were highest for Recife and Belo Horizonte, and lowest for Sao Paolo and smaller cities in Minas Gerais. Large cities offer more Carnival programming and often invest large sums of public money in events. But there are also more negative aspects of large city Carnival experiences, which include higher levels of crime and insecurity (Guimarães, 2025).

There was no significant relationship between income levels and satisfaction. However, those with lower incomes were less likely to intend to return in the future, indicating that more mobile groups with higher purchasing power may be more likely to travel to other locations as tourists.

Intention to return was highest for Recife and Belo Horizonte, which also had the highest satisfaction levels (Table 7). Interestingly, there was no significant difference in intention to return between residents and visitors.

One outcome that is particularly important for cities is the spending generated by Carnival, particularly by visitors. The average spend per person for Carnival participants was BRL 559 (approximately €91 or $95 during Carnival, 2025). Levels of spending were not surprisingly much higher for visitors (BRL 850) than for residents (BRL 336). This is a similar ratio to that reported by Carnival surveys in Belo Horizonte in 2025, for example (BRL 217 for residents, BRL 682 for visitors – Belotur, 2025). There was no correlation between the amount spent on attending Carnival and satisfaction, but there was a significant negative correlation between spending and intention to return (r = − 0.081). This may reflect a common complaint in recent years that the cost of attending Carnival is becoming very high, particularly in relation to incomes.

Our analysis of the Carnival experience highlighted several important aspects, including the profile of participants and the mode of participation, the location of Carnival attendance and the public support for Carnival celebrations, and the role of identity and groups in Carnival.

The idea of Carnival as a collective festive experience for Brazilians is reflected to a large extent in the responses of participants. There are few significant differences in overall experience scores between different groups of participants, supporting the idea of Carnival as a generalised celebration for all. The data also suggest high levels of Carnival participation in the post-COVID years 2023–2025, with high levels of excitement and emotional energy. This is also supported by other research in cities such as Belo Horizonte, where mean satisfaction scores reached 8.8 on a 10-point scale in 2025, close to the level in 2020 (9.0) (Belotur, 2025).

These results underline the resilience of the Carnival ritual in Brazil, which shows high levels of post-COVID participation in Carnaval de Rua and blocos and samba schools. These forms of collective participation are important to the individual experience, which is heightened as a result. However, we still found significant differences in the individual experience of Carnival. As De Matas (2014) has argued, there have been few studies related to the experiences of those with different roles in Carnival, such as spectator, participant, or artist, for example. Our research indicates that there are many different positions from which to experience Carnival, and the cognitive, conative and emotional aspects of experience differ with these positionalities. Beyond the crowd, individual experience can be shaped by positionality and form of participation as well as the rituals and practices linked to Carnival. Those respondents who engaged in routines related to Carnival also tended to have more positive experiences of the event. This strengthens the argument of De Matas (2014) that more attention needs to be paid to the experience of individuals within the Carnival collective.

As Jaguaribe (2013) the collective Carnival crowd is composed of individuals, who each participate by dancing, swaying and playing, blending in according to their own desires, abilities and physical resistance. But the collective rituals and energies of the crowd also shape how the individual merges into the bodily maze. The interplay of individual and collective experience is also mediated by the presence of groups in the Carnival crowd. Those participating in Blocos, one of the important organisational aspects of Carnival de Rua, tend to be more active and experience a greater sense of adventure. The fact that public authorities often play an important role in organising the spaces of Carnival means that they have an influence on these elements of experience as well. Participation in Carnaval da Rua and blocos varies by location, often depending on the active encouragement of the authorities. The presence of such groups tends to generate more positive experiences, seemingly justifying attempts to increase participation by blocos in many cities. The relationship between social or collective experience and the individual experience of Carnival is also an important point in terms of theoretical development. Previous studies have emphasised the social experience of Carnival, usually paying less attention to individual experience. However, our findings indicate that both social and personal aspects of experience are important to outcomes, both for the individual Carnival participants and for the social context that they interact with and help to shape (Bargeman and Richards, 2020). Using theoretical insights from practice theory might help to unravel the dialectic relationship between the Carnival practices of the individual and the broader cultural and social structures that give Carnival its distinctive form.

One important finding of the current study is that the location of participation may have an important influence on the experience of Carnival. We found that larger cities with a significant and distinctive Carnival tradition, such as Recife and Rio de Janeiro, also tended to generate the highest experience scores. This has potentially important implications, since it suggests that the strategies engaged in by these cities to create a specific type of Carnival experience can have some influence on the outcomes for participants. If, as our research suggests, this leads to a higher level of repeat visitation, it can help the cities achieve greater economic impact and a higher level of social cohesion. Only the reflexive dimension of experience did not show significant differences. This underlines the importance of context and the specific form of Carnival in the experience. This seems to indicate that the different styles of Carnival identified in the literature (e.g. in Recife, Rio de Janeiro and Salvador) produce different experiential outcomes.

Places can influence the experience of Carnival through the design and equipping of spaces and facilities, by stimulating the presence of collective activities, such as blocos, by providing programming, marketing activities and ensuring security. Our research shows a differentiated experience of Carnival across Brazil as a result of the interaction of these factors. This also indicates that the efforts of cities such as Belo Horizonte to increase security during Carnival may have had a positive effect on visitor experience, and therefore on the outcomes of Carnival participation. Cities should also be aware of the potential for increased security and more structured programming to undermine the spontaneity and “authenticity” of Carnival. However, it is also the case that laments about loss of authenticity are not new (e.g. Taylor, 1982), and no doubt Brazilian Carnival will also be able to re-invent itself creatively in the future.

This analysis enables us to provide answers to the research questions posed in this paper. We firstly found a range of factors to be important in shaping the experience of Carnival in Brazil. These included demographic factors, such as age and income, the location of participation and attendance as a tourist versus resident. Individuals who were professionally involved in Carnival also tended to have a more positive experience, which may relate to their greater levels of involvement and their position at the centre of the Carnival ritual. Location also played a significant role in influencing the experience of Carnival. Not only do different cities in Brazil provide distinctive styles of Carnival performances, but they also have an important influence on the programme, Carnival facilities and levels of security.

In terms of the second research question, we also found that the social context of participation, such as being part of a bloco or samba school, can influence the experience of Carnival. Bloco members tended to report higher levels of excitement and activity, acquiring new knowledge and a sense of novelty. The emotional energy of group participation was generally strengthened by a higher level of Carnival routines, including singing, dancing, making decorations and joining in group activities. Following such routines also tended to be linked to higher experience scores, except in terms of the item “I was thinking”. This may indicate that group members are more likely to experience a sense of flow during Carnival (Lohman, 1999).

Thirdly, in terms of the outcomes created by the experience of Carnival, we found higher experience scores to be generally related to positive outcomes such as satisfaction and intention to return. Satisfaction was significantly positively correlated with most experience items, and this pattern was repeated for intention to return. In terms of the experience dimensions identified by de Geus et al. (2016), satisfaction and intention to return were most strongly correlated with conative and physical involvement, and less strongly correlated with cognitive experience. This seems to indicate that the desire to be involved in Carnival, and the physical activities that this generates, are most likely to create positive outcomes.

Our research aligns with the argument of Da Matta (1981), who sees Brazilian Carnival as a collective ritual in which individuals submit to the collective in order to strengthen their social identity. The event is still experienced at the individual level, although clearly mediated by collective structures, such as the blocos and Carnival practices and routines.

There are a number of limitations that should be noted for this research. The use of convenience sampling helped to generate a large number of respondents who had participated in Carnival in the research time frame. However, convenience sampling of event audiences also has the drawback that the sample cannot be considered representative of the event population as a whole. The use of snowball methods to locate respondents also tends to affect the results. In this case, the sample was relatively highly educated because of the distribution of some survey links through education networks. The use of a one-shot post-event survey also limits the view of Carnival participation.

Quantitative research on Carnival has the potential to explore the many differentiations of the ritual in Brazil, as well as offering opportunities for comparisons with other countries. Is Carnival a global celebration or a series of very different experiences in different places? Further quantitative studies might also help us to answer questions about the “Brazilianness” of Carnival in Brazil. Which experiences and outcomes are unique to Brazil, and which experiences are repeated in different Carnival celebrations around the world?

One interesting finding from the current research was a strong link between motivations, experience and outcomes. It could be fruitful to explore these links further, particularly through qualitative studies that might be used to explore possible causality.

It is also important to develop bridges between the quantitative insights generated here and the extensive body of Carnival literature based on qualitative research. The current paper does this to some extent by analysing the comments of participation, but more in-depth interviews with the participants and organisers of Carnival in different locations might also help to generate richer insights into some of the issues raised here. It would be particularly interesting, for example, to examine the relationship of collective and individual experience, exploring the dynamics and tensions between the collective experience of the Carnival crowd (Jaguaribe, 2013) and the individual experience of the Carnival event.

We are grateful to many people for help with the data collection, including Magnus Emmendoerfer, Mariana Bueno, Carla Borba, Celiane Camargo-Borges, Carolina Jordão, Luana Silva, Marcos Knupp, Marina Simião and many others.