Reading culture as shared ethos: A study of Finnish self-identified readers

This article advances understanding about book reading as a sociocultural phenomenon in the 2020s. We make a contribution to the cultural sociology of reading by investigating Finnish self-identified book readers by analysing the significance of sociodemographic variables (gender, education, age, and place of residence) in terms of reading activity and access to books. Our study is placed in the context of Finnish reading culture that is characterised by a particular appreciation of reading and measures promoting equal access to culture. Based on an online survey of 955 respondents conducted in 2021, our statistical analyses show that the social stratification of book reading activity that is prominent in population level does not recur within the specific group of people who identify themselves as readers. Among Finnish self-identified book readers, education, gender, and place of residence do not induce significant differences in reading activity. Our analysis that foregrounds inclination instead of quantity as a criterion for readers sheds light on reader equality from a different direction than previous research into nationwide reading habits or descriptive studies on avid readers.


Introduction
In Finland, a Nordic democracy, reading has been valued and socially constructed as enjoyment as well as self-improvement, and it constitutes one of the most popular leisure pastimes (Eurostat, 2018;Laine & Salmi-Niklander, 2018;Mäkinen, 2015;Statistics Finland, 2017).More generally, books and literature are highly appreciated in Western culture -reading has been considered as an intellectual virtue (Furedi, 2015;Manguel, 2014;Spjeldnaes & Karlsen, 2022).In recent decades, reading has been affected by digitalisation in particular, resulting in diversified book formats and new platforms for readers' interaction and changes in the prevalence and practices of reading (Knulst & van den Broek, 2003;Kurschus, 2015;Thompson, 2021).Therefore, new approaches in conceptualising reading and readers of books are called for.
In this article, we make a contribution to this discussion in the frame of cultural sociology of reading.We analyse the sociodemographic patterns of self-identified readers of books, i.e., the particular group of people who identify themselves as readers notwithstanding skills or number of books read.Thus, we have a shared premise with Merga (2017) as we authorise the participants to define for themselves whether they are readers of books or not.With this self-identification as a starting point, we analyse the significance of different sociodemographic variables in relation to our participants' reading activity and access to books.
Therefore, our vantage point deviates from that of previous empirical research on readers of books in sociology of literature and sociology of consumption in at least two respects.On one hand, the focus of these research traditions has been on reading habits and literary taste patterns in different social groups across nationally representative populations (e.g., Atkinson, 2016;Bourdieu, 1984;Bukodi, 2007;Purhonen et al., 2010;Verboord & van Rees, 2003).On the other hand, these studies have mainly identified readers based on external and measurable criteria such as the number of books read or time spent on reading.Consequently, research that specifically focuses on readers often touches on those who read the most according to national statistics, while our outlook on readers supposes a larger variety of factors.
There is a distinct value traditionally attached to reading worldwide (Griswold et al., 2005).Our study is placed in the context of the Finnish reading culture, where the appreciation for reading has its particular historical, cultural, social and even political significance that is above all connected to equality in society (see Section 3).Our research project Reading in change examines book reading habits and culture in Finland in the 2020s.In this article, we analyse the features and reading activity of 955 respondents who participated in an online survey in 2021.More specifically, we focus on voluntary reading, which is here defined as reading that is selected by the reader and that is read for the sake of reading (Krashen, 1993), in one's leisure.
Our research questions are as follows: 1: How does reading activity relate to the age, education, gender, and place of residence among self-identified readers of books in Finland?2: How does access to books relate to the age, gender, reading activity, and level of education among self-identified readers of books in Finland? 1   Reading activity is here defined as the amount of time the participants report spending on reading books daily (see Section 4.2).Access to books refers to the respondents' identification as book borrowers and book buyers as well as their inclination towards different book formats, i.e., printed books, e-books and audiobooks.
This article is structured as follows.Next, we discuss our key concepts (Section 2.1) and explain our contribution to previous research on reading in sociology as well as other disciplines (Section 2.2).In our review of Finnish reading culture (Section 3), we show how reading books in Finland forms a historical continuum as context to our analysis.We then outline the population and data collection of the study (Section 4.1) and explain the method of the quantitative analysis (Section 4.2).Further, the results are presented (Section 5), followed by conclusions (Section 6) and discussion (Section 7).

Key concepts: self-identified reader, reading as a social practice, and reading culture
In the centre of our research is the reader-led definition: the self-identified reader.There is a vast range of definitions of readers and reading in the existing literature (see e.g., Ross et al., 2018, 139-140), and our study especially touches on so-called avid readers or committed readers, terms that are used diversely in reading research (see e.g., Ross, 1999;Ross et al., 2018;Stebbins, 2013;Wilson & Kelley, 2010).The approach that we apply here emphasises people's own determination over the amount or frequency of reading: readers are perceived as those who are inclined towards reading books and who subjectively experience being readers (see Merga, 2017;Nolan-Stinson, 2008).In contrast, a key component of being an avid reader can also entail reading frequently or consuming large numbers of books.In the Finnish national statistics, active readers are defined quantitatively as those who have read at least 10 books over the past six months (Statistics Finland, 2017).These approaches are not utterly opposing, as attitude and activity are likely to overlap (see Stokmans, 1999).While this turned out to be the case in our data as well, our initial objective was to recruit participants who perceive themselves as readers of books according to their own personal definition as opposed to predetermined criteria (on the population of this study, see Section 4.1).
As Thumala Olave (2018, 449) has stated, predetermined categories deny the agency of the readers and tend to ignore the fundamental importance of their meaning-making (on reader-led criteria, see also Koolen et al., 2020).Even though the topic of our investigation is not the reception of literature, we share the essential point of departure: that of letting readers themselves define whether they belong to readers of books regardless of how many books they read during a certain time period.During the life-course of a reader, the actual number of books read may vary allowing times of non-reading.Thus, in our study, the self-definition as a reader as a form of cultural construction of subjectivity is forefronted (see DeNora, 1999, 54; on the self-definitions of Finnish readers of books, see Suomalainen, 2022Suomalainen, , 2023)).
This kind of agency represents reading as a social practice, the ways in which individual readers choose to pursue reading books in their everyday life (see Griswold et al., 2005, 127;Thumala Olave, 2022a).In addition to the traditional ways of reading printed books, reading practices have diversified in the 21st century due to digitalisation and online platforms (Vlieghe et al., 2015).These platforms allow readers to start online book clubs, write reader reviews or give peer-to-peer recommendations as prosumers ('producing consumers') (Jaakkola, 2019;Jenkins, 2006, 93, 106;Linko, 2023;Linko & Lindh, 2024;Verboord, 2010).By reading practices, readers of books enhance their own identities as readers but also participate in shaping the reading culture.As defined by Kurschus' study about the diverse European book cultures, two senses of the term 'culture' are relevant here: firstly, culture as a set of values, traditions and actions, and the institutions that support them, and secondly, culture as a sector that promotes cultural products and is affected by political decisions (Kurschus, 2015, 38-39).By using the concept of reading culture instead of book culture we want to highlight the role of the readers as its active participants.They are both producers and members of the Finnish reading culture.The reading culture is also shaped by history, educational systems and cultural politics which have created a certain kind of infrastructure for reading in Finland.This development is further elaborated in Section 3.
In the book trade, digital formats and subscription-based reading services have become more established.The increased production and use of audio books have significantly changed the publishing industry in Finland, following the trend in other Nordic countries (Finnish Publishers' Association, 2022;Hanifi, 2021;Lindh & Hiidenmaa 2023).Furthermore, the high importance of visibility on various digital platforms as well as the utilisation of user data and algorithms have all affected the book markets.All these phenomena are part of the global trends of consumerism, commercialisation and mediatisation of literary life and art in general (Alexander & Hägg, 2018;Erjavec 2018;Lauristin & Vihalemm, 2014;Thompson, 2021).

Related research
Our study is related to two traditions of sociological research in reading books.We draw on the research area that Thumala Olave (2020; 2022a; 2022b) refers to as the cultural sociology of reading, in which readers' own meaning making processes and the reflection and development of self-identity by reading are emphasised (see also Eskola, 1990a;Linko, 1998;Long, 2003;Radway, 1984;Taylor, 1989;2019).The reception and preferences of specific genres among readers is a part of this line of research and has also been studied extensively in Finland (e.g., Eskola, 1990a;Eskola & Linko, 1986;Jokinen, 1994;Launis et al., 2018).
The focus of another relevant tradition of the research into the sociology of readers of books is on the structure of literary taste and its connection to accumulation of cultural capital and further, its connection to social class.This research area typically analyses various forms of cultural capital, not only the reading of books.Since Pierre Bourdieu's (1984) ground-breaking work the literary taste of different social groups and the distinction strategies including the acceptance or rejection of different literary genres has been analysed within several cultural contexts (Atkinson, 2016;Bennett et al., 2009;Prieur et al., 2008;Purhonen et al., 2010).
Within this field of study, the omnivorousness of cultural taste patterns first analysed by Peterson (1993) and further developed by Peterson and Kern (1996) has inspired empirical research on the manifestation of cultural taste in different social classes and moreover, theoretical analysis on the relationship between cultural taste and social class.While literary taste is left beyond the scope of this article, our approach is connected to this research tradition by the notion that reading books adheres to social structures and functions as an intermediary or form of cultural capital.
There is a shared view that reading books in Western societies has declined in general (Bennett et al., 2009, 95;Toivonen, 2013) and a growing share of young generations especially belong to a group of non-readers (Heikkilä, 2022).Therefore, the debate on the components of elite taste in literature, whether it can be described as omnivorous taste (Peterson 1993;Peterson & Kern 1996) or highbrow taste (e.g., Bennett, 1999, Wright, 2006), has been partly replaced by the view that reading in itself is a source of cultural capital, a legitimate form of activity (Bennett et al., 2009, 111;Purhonen et al., 2014;Heikkilä, 2022).In this regard, distinctive features of readers and non-readers have been studied (Purhonen et al., 2014, 71-82;Verboord & van Rees, 2003, see also Heikkilä, 2021;Heikkilä, 2022).Typically, reading books or belonging to a group of non-readers is interpreted in the framework of class structure and inequality.
In the context of the sociology of literature, there is a rather solid understanding of some principal trends in leisure reading: on a large scale, characteristics of those who belong to the group of readers are high social status and high level of education.This also applies to any kind of cultural activity.Further, women are more likely than men to participate in traditional high status leisure activities.(Bennett et al., 2009;Bihagen & Katz-Gerro, 2000;Lizardo, 2006;Heikkilä, 2022.)This group of diversely privileged people, above all highly educated and mainly women, have also been referred to as the reading class (Griswold, 2001;2008;Griswold et al., 2005).
In addition to sociology, reading of books has been studied in e.g., information and media studies, cultural studies and literary studies.These domains share the perception that is also typical of the cultural sociology of reading, namely that of reading books as a meaningful leisure activity that provides ingredients for forming and maintaining self-identity in changing circumstances (e.g., Merga, 2017;Pyrhönen & Kantola, 2018;Ross, 1999;Ross et al., 2018;Usherwood & Toyne, 2002).In contrast to these predominantly qualitative approaches, there is only a limited amount of quantitative research examining the organisation of background variables and diversity within the group of people who have a keen interest in reading and to whom it is part of their everyday life.This topic is addressed by the present article.

Reading in Finnish cultural context
While we draw a distinction between voluntary reading of the self-identified readers and that of the general public, we argue that both are founded on mutual cultural, historical and political developments and consequently, the distribution of cultural and social capital.The main institutional apparatuses that support reading and readers are education, religion, commerce and state (Griswold et al., 2011, 24-25).As we show in this section, all four quarters have had a significant role in the development of Finnish reading culture.
In Finland, reading books has crucial significance as shared cultural capital, and the imagery of Finns as a nation that particularly P. Hiidenmaa et al. values reading is persistent (e.g., Eskola, 1990b;Hanifi, 2021, 113).Below, we explore two historical developments relevant to our analysis: the process of reading becoming a civic duty, and equal access to books for all citizens of a modern Nordic democracy.We thus draw on the notion of a distinct albeit not utterly uniform Finnish reading culture that provides a specific context to our analysis.
In parallel with other reformed societies, the roots for Finnish reading culture have been traced to the Protestant Reformation in the 16th century.As of then, the church and religious movements began to support and test literacy, which was a prerequisite for studying the Bible and thus represented a personal endeavour and devotion to one's faith (Laine, 2019;Laine & Salmi-Niklander, 2018;Sinnemäki, 2020).Due to the Lutheran values, literacy gained ground not only as a skill of the individual but as a civic duty and a means of self-development.In the 1700s Finland, the purposes of reading books began to expand from social acceptance and utility towards enjoyment and personal interpretation (af Forselles, 2011, 95).During the 1800s national movement, the love of reading, which derives from the European discourse and stands for a spontaneous and even irrational motivation to read, was harnessed for policies advancing reading.Reading as well as reinforcing the Finnish literary institutions played an important role in the building of a collective identity.(Mäkinen, 2015.)By the late 1800s, the possibility to acquire basic reading skills transpired at all levels of society and was fostered by several social movements such as the labour movement (Laine & Salmi-Niklander, 2018).
In 20th century Europe, mass literacy was one of the prerequisites for a society that was making the transition from tradition to modernity.Literacy among all residents was generally thought to be beneficial as it required schools, produced individual mobility, reduced social inequality and was politically and ideologically neutral.(See Lyons, 2010.)Along with the modernisation, the level of education in Finland rose dramatically, and the book market was liberated.Consequently, reading became an even more relevant activity across all areas of life.
In the 1960s, Finland's development towards a so-called Nordic welfare state (see Esping-Andersen, 1999) included public investments not only in healthcare and social services but also in culture and education.As in other Nordic countries, cultural policies were developed with the aim being all-encompassing and providing equal opportunities for all to participate in cultural activities (Kvist 2013;Sokka et al., 2022, 8).Also, a comprehensive basic education system was established.The welfare state project also led to the strengthening of the public library system (Kurschus, 2015, 220;Mäkinen, 2009).
The above described measures aiming for equality form the basis of contemporary Finnish society where the lack of economic resources is, in principle, not an obstacle to participation in legitimate culture but can instead be complemented with cultural and social capital (see Purhonen et al., 2014, 82-83).There is, however, empirical evidence that in Finland reading books is organised according to age, education, and gender, along with other sociodemographic factors.For instance, educated women are the most active readers in light of the number of literary works read in leisure.(Purhonen et al., 2014, 72-73.)This pattern of social stratification also applies to participation in any kind of cultural activity (Heikkilä 2022, 39-40;Heikkilä & Lindblom, 2022).
Moreover, in recent decades there has been a consistent decrease in voluntary reading of books: when active readers are defined as those who have read at least 10 books over the past six months, in 1981, as many as a third of people aged at least 10 years met this criterion.In 1991 and 2002, the amount of avid readers had shrunk to a fourth, and in 2017 to only 13 %.(Statistics Finland, 2017.)Simultaneously, digitalisation has cast a two-fold outlook on the prospects of voluntary reading of books.While the effects on the book industry as well as on people's cognitive abilities have raised concerns, recent studies have suggested that technologies also hold the potential of increasing reading motivation and enabling new ways of participating in the reading culture and its subcultures (e.g., Aaltonen, 2019;Jaakkola, 2019;Kajander, 2015;Linko, 2023;Linko & Lindh, 2024;Logrén, 2020;Saresma, 2013).
To summarise, when reading time or number of books read defines who is considered to be a reader, research on leisure reading in Finland presents at least two conclusions.On the national scale, reading books is becoming less popular, and furthermore, this change occurs unevenly in light of variables related to social structure.Similar developments take place in other Western societies despite their cultural policies aiming at equal access to culture (Griswold et al., 2005, 131-132;Lizardo & Skiles, 2015;Purhonen et al., 2011Purhonen et al., , 2014;;Scherger & Savage, 2010).These results depict the relations between various groups on the overall scale from non-readers to readers, hence leaving the internal diversity among certain groups on the reading activity spectrum unobserved.
As for the Finnish welfare state and its cultural policies, reading books can be interpreted as a form of social capital which Putnam (2000) describes as bridging: it connects people from different backgrounds and standpoints.Our study aims to highlight the importance of discussing not only the decrease and differentiation in reading books but also its vitality: there are people to whom reading books is a meaningful part of everyday life and identity, and by examining the diversity of readers we strive to add to the knowledge on reading in general.

Population and data collection
The intended population of this study consists of those Finnish people who identify themselves as readers.In other words, we aim at investigating reading activities and the related variables among the reading people in Finland, not in the whole Finnish population.By the previous research (e.g., Purhonen et al., 2014;Heikkilä, 2022), we know that the gender distribution within our population is skewed: a clear majority of the Finnish readers are female.
Our data were collected through an online survey distributed via several social media channels in order to reach Finnish readers as P. Hiidenmaa et al. broadly as possible.A link to the questionnaire with an invitation to participate was published on the blog site of our project. 2The blog post was distributed through various literary groups on Facebook, and the researchers' own accounts on Facebook, Twitter, LinkedIn and Instagram.Suitable hashtags were used to help the message spread.For instance, the group called Kirjallisuuden ystävät (Friends of Literature) had more than 30 000 members at the time of the questionnaire (currently over 40 000).The questionnaire was open for three weeks in August-September 2021.
In the cover notes, we emphasised self-identification by inviting participants with expressions like "friends of books", "people who like to read" and "all kinds of readers", which indicated that we wanted to reach people who have a positive attitude towards reading whether they read often or just occasionally.The questionnaire was rather long, but we emphasised that all questions were voluntary except for the demographic questions, i.e., questions regarding age, gender, education, occupation and place of residence.It was also possible to choose the option "I don't want to answer".The answering times that were reported to us varied from twenty minutes to two and a half hoursdepending on how much time the respondent wanted to use on the open-ended questions.
The questionnaire consisted of closed questions (yes/no), open-ended questions, multiple-choice questions and Likert scale questions.All in all, we got 955 responses, and most respondents replied to all of the closed questions.For the open-ended questions, the average response rate was approximately 65 %.
According to background variables, we can generalise that women with a higher education were the most active participants in our survey.Eighty-four percent of the respondents were women, 14 % were men, and 1 % chose the option "other" and 1 % did not want to disclose or define their gender. 3This corresponds quite well to previous studies on reading activity in the whole Finnish population, so our sample can be said to represent the reading population well with respect to gender.The distribution is also rather even across age, especially between ages 35 and 65.The youngest (from 15 to 24) and oldest (over 75) age groups were however somewhat underrepresented Figs. 1, 2.
In general, the respondents were well-educated.Half of them held a higher academic degree, and 21 % had a lower academic degree.In addition, 8 % held a doctoral degree and 12 % had a vocational degree.
Considering that the questionnaire was targeted at book readers and distributed via social media platforms, we can assume that the respondents represent people who not only have a positiveeven committedattitude towards reading but who are also familiar with the use of the Internet and social media.The majority of respondents had very active readingincluding listening to audiobookshabits: 86 % read every day, and over 80 % read or listened to over 20 books a year, which is significantly more than the reading activity of the Finnish population as a whole.By comparison, only 13 % of Finns read 10 books in six months according to statistics from 2017 (Statistics Finland, 2017).

Quantitative analyses of data
The quantitative analyses of our data were performed using the statistical software SPSS version 28.The five-step Likert scales contained in our questionnaire were interpreted as continuous, and we therefore used Student's t-tests and one-way analysis of variance (One-way ANOVA) with Bonferroni's post hoc test to study the mean differences between various groups and variables.We used Pearson correlation analysis to study the covariance of these scales.In addition to studying the statistical significance, we used Cohen's d to measure the effect size of interesting mean differences.In the interpretation of the values of d, we applied the critical values given by Cohen (1988).
The participants' reading activity was measured on an ordinal scale and categorised into four categories as follows: 1= "Occasionally", 2= "Less than one hour every day", 3= "1-2 h every day", and 4= "More than two hours every day".Similarly, the participants were divided into four groups with respect to their highest educational degree.The categories are 1= "Secondary or vocational education", 2= "Bachelor", 3= "Master", and 4= "Postgraduate".Whenever these variables were included in the quantitative analyses, we used the Chi-Square test to study eventual dependencies.

Results
We begin answering the first research question by summarising in Table 1 how the participants in this study fall into different categories with respect to their reading activity (cf.Section 3.2).Two of the participants did not reveal how much they spend time reading on average; hence N = 952.
Table 1 also shows the crosstabulation of the participants' reading activity across gender and residence.The Chi-Square test did not reveal any dependence between the activity and gender (χ 2 (3) = 3.07, p > .05).In other words, the variation of the activity in Table 1 is similar for female and male participants.Likewise, the variations of the reading activity do not significantly differ from one another with regard to the regions that represent different residential areas of Finland (χ 2 (18) = 16.46,p > .05).
In this analysis, we have excluded eighteen participants who did not reveal their gender or who expressed it as something other than male or female.Similarly, thirteen participants did not reveal their residence, and they were excluded from the latter analysis.
In Table 2, the number of participants included in the analysis of the relation between reading activity and the level of education is 927 because 27 respondents did not clearly indicate their level of education.This table shows that reading activity is clearly related to age and level of education.Indeed, the evidence for these interdependencies is strong as the p-value in the test is less than 0.001 in both cases.Moreover, the relation (χ 2 (9) = 34.92,p < .001) between reading activity and level of education is somewhat surprising.Namely, the participants with a vocational degree or who have passed only a secondary school seem to be the most active readers.In Table 2, the two highest reading activity categories are both modes, i.e., equally most common, for this group, whereas the second highest reading activity category is the only mode for the other educational level groups.The highest reading activity category comprises 36 % of those in the lowest educational level group, whereas this category represents only 17-26 % in the other groups.
When it comes to the relationship between reading activity and age, the result (χ 2 (18) = 76.97,p < .001) is clear but, perhaps, not as surprising as in the previous case: younger participants express that they do not read as much as the older participants indicate they do.This effect is most clear for the group consisting of people older than 74 years.However, since this group represents only 2 % of the sample, it is more reasonable to focus on the location of the median.For the participants who are 25-34 in age, the median is in the second lowest reading activity category.If two observations from the third category had been classified in the second category, then the same would apply also for the participants who are 15-24 in age.For the other age groups, the median is clearly in the third

Table 1
The percentage distribution of reading activity (N = 952) and activity across gender (N = 934) and residence (N = 941).category, except for the oldest participants, whose median is in the highest reading activity category.
It is a well-known fact that the level of education among Finns rose significantly during the last quarter of the 20th century.Therefore, one may suspect that the high reading activity of the participants at the lowest level of education can be explained by the fact that this group consists mostly of retired people who have more leisure time for reading.Therefore, we conducted further analyses to find out whether this assumption is true or false.The answer is "false".Table 3 shows the distribution of the age of the participants at the lowest educational level.The table shows that the case is preferably the opposite: the younger generations are slightly overrepresentednot underrepresentedin this group.This finding only emphasises the unexpectedness of the results discussed above.
For the second research question, we asked the participants how much they agree with the claims 5a= "I buy printed books for myself" and 5b= "I borrow printed books from the library or my friends".Here it is reasonable to focus only on printed books since, in practice, all participants consumed printed books, and less than a half of them consumed digital books to a significant degree.Moreover, printed books are much more expensive than digital books.The participants answered by using a five-step Likert scale orientated as 1= "not at all" and 5= "completely".
Fig. 3 shows how agreement with the claims decreases systematically with respect to age: the older the Finns are who participate, the lower the mean value.When measured by using Cohen's d, the mean difference between the youngest and oldest categories is of almost a large effect size (d = 0.74) for buying books and of a large effect size for borrowing books (d = 0.82).
Gender is similarly a significant factor when it comes to agreement with 5a (buy) and 5b (borrow).However, men and women answered in different ways.Table 4 reveals that male participants identified themselves more strongly as buyers of books (t(933) = 2.45, p < .01)and female participants as borrowers (t(180) = − 3.18, p < .001)(the degree of freedom is 180 in the latter test, which indicates that equal variances were not assumed).
We also studied how buying printed books for oneself is related to buying printed books as gifts to other people.The Pearson correlation between these two Likert scales was 0.62 (p < .001),which indicates a very large effect.Moreover, r = 0.12 (p < .001)for 5a and 5b.In other words, identifying oneself as an active buyer of books is related to identifying oneself as an active borrower of books, but this relation is rather weak.
Fig. 4 shows that respondents with different reading activity do not express different behaviour in buying books for themselves (F (3947)= 0.23, p > .05).However, there is a statistically significant difference in their tendency to use libraries and in borrowing books from friends (F(3947)= 5.65, p < .001).Respondents who read almost daily or 1-2 h per a day had the highest mean in 5b (borrow), whereas the participants who read over two hours daily had the lowest mean in 5b.In other words, the readers with the highest reading activity did not identify as book borrowers as strongly as their less-reading peersquite the opposite.This outcome may sound paradoxical, but it is not necessarily so.One remarkable reason for this outcome is the fact that according to our data, the most active readers are also the most active listeners of audiobooks: 30 % of those who read more than two hours each day also reported that they listen to audiobooks every week, whereas of all other readers, only 5 % said that they listen to audiobooks every week. 4Moreover, according to our data, the Spearman correlation between reading activity and the consumption of audiobooks (measured by an ordinal scale with six different categories) is ρ = 0.24 (p < .001).
In general, the participants agreed to a high degree on the claim about borrowing printed books.This is seen also in Fig. 5, which indicates that the participants that represent the higher levels of education had a remarkably higher mean in their assessment of buying printed books (5a) than the participants from lower educational levels (F(3924)= 6.12, p < .001).The difference here is significant also from the perspective of the effect size.Now, Cohen's d = 0.55, which can be interpreted to be, at least, medium size.Interestingly, there are no significant differences in borrowing books across the groups (F(3924)= 1.59, p > .05).
We complete our answer to the second research question by investigating how the use of digital and audiobook subscription services relate to the participants' age, gender, and level of education.For this purpose, we inquired whether the participants had paid for a subscription service during the previous twelve months.Tables 5 and 6 show the percentage of the subscribers in different categories.

Table 2
The percentage distribution of reading activity across age (N = 948) and level of education (N = 929).

Table 4
The descriptive statistics of access to printed books across age (N = 948) and gender (N = 951).The statistically significant differences in Tables 5 and 6 indicate that younger generations and especially female participants are more active users of subscription services than older generations and male participants.On the other hand, the participants' level of education does not play a significant role here.

Conclusions
Our first research question aimed at investigating how age, education, gender and place of residence explain the reading activity among self-identified readers in Finland.In the analysis, the participants were divided into four categories based on their reading activity, and our analysis shows that this variation is related to some of the variables mentioned above.In our data, older and less educated people are overrepresented in the groups that displayed the highest reading activity (Table 2), whereas gender and place of residence did not seem to play a significant role in reading activity (Table 1).At this point, it is good to note the composition of the target group of the study: a group of people who identify themselves as readers, not all Finns.From the viewpoint of equity, these findings are noteworthy especially regarding education, gender and place of residence.
First, with regard to the level of education, our study shows that among self-identified readers, high education is not a prerequisite for high reading activity.One possible interpretation can be found in the accumulation of activities transcending leisure activities (Griswold et al., 2005;Oinas et al., 2010) and taste (Peterson & Kern, 1996;Purhonen et al., 2010) where the widest repertoires accumulate to certain groups of people, namely the socioeconomically advantaged.Research shows that active users of digital media are also the likeliest to participate in other reading activities (Purhonen et al., 2021).Moreover, it would be beneficial to further explore if the differences in leisure reading between highly educated and those with vocational training are explained by content of work, i.e., knowledge work and screen time.
Second, according to the evidence at hand, there is no gender difference in reading activity among self-identified readers.Based on our analysis, we conclude that men are equally active readers as women when it comes to people reading books in Finland.These are remarkable findings because in the general population, reading is a more common leisure activity among highly educated people than among less educated people, and a majority of those who read the most books are female (Purhonen et al., 2014;Heikkilä, 2022).
As most of our respondents are highly educated women, the sociodemographic distribution in our data is consistent with previous research on the reading habits in Finland in general (e.g., Purhonen et al., 2011;Heikkilä, 2022), as well as the idea of a specific

Table 5
The percentage of consumers who had paid for a subscription service across age (N = 936).F(6929)= 5.66, p < .001.

Table 6
The percentage of consumers who had paid for a subscription service across gender (N = 923) and level of education (N = 914).reading class (Griswold, 2008, 160).However, our findings indicate that being a reader is similar across different groups.
In light of social equity, another important finding is that reading activity does not depend on the respondents' place of residence.Perhaps this allows the interpretation that there are no significant differences in attitudes towards reading between the metropolitan area and rural areas in Finland.This result is also related to our conclusion regarding the second research question, which takes Finnish readers' access to books into account.Namely, it suggests that books are still available throughout the country due to a comprehensive network of public libraries even though the number of bookstores has diminished considerably during the last decades, especially in smaller municipalities in Finland. 5 Another interesting finding related to our second research question is the difference between males and females in their habits of obtaining books (Table 4): Finnish men identify themselves more clearly as buyers of books than Finnish women do, whereas women identify to a higher degree with the claim about borrowing books.The elderly tend to buy or borrow books less than those of younger age-groups (Fig. 3).This may indicate that they have a large collection of books in their homes or receive books as gifts, and therefore they are less interested in acquiring new books.Their reading habits may also be different from that of younger generations with respect to genre, functions of reading or ageing.
The result shown in Fig. 4 is a surprising and counterintuitive feature in our portrait of a self-identified reader: among the reading population, the most active readers seem to identify as the most passive library users and borrowers.However, as already stated (Section 5), this finding can be explained by the fact that the most active readers are also the most active consumers of audio books.Furthermore, they probably already have a large collection of books at home, or they consume, to some degree, longer or more demanding books than other readers, which might explain why they do not buy or borrow books as frequently.We did not inquire about the number of books bought or borrowed, but the respondents were asked to what extent they identified themselves as buyers and borrowers of books.
There were no significant differences between the groups that represent different educational levels when it comes to borrowing books (Fig. 5).This is further evidence of the important and concrete role of public libraries in maintaining social equality.In contrast, it is hardly surprising that the participants with the highest education were the most active buyers of books according to our data (ibid.).This finding can be related to the possession of economic capital since new printed books are considered quite expensive in Finland.
Lastly, our findings do not indicate any remarkable differences between different groups when the digitalisation of reading is taken into account.Females and younger generations in Finland are more active users of digital subscription services than males and older generations.However, the participants' level of education (and hence economical status), for example, does not have a significant effect on the use of such services (see Tables 5 and 6).
The summarising observation of the findings on this topic is that the digitisation of books does not seem to be a threat to the Finnish readership.Rather, it is a complement to more traditional reading habits.An overwhelming majority of Finnish readers still prefer to read printed books as only 17 % expressed that they listen to an audiobook daily and less than 7 % said that they read an e-book every day.According to our survey, in a hypothetical situation in which all book formats are available, as many as 70 % of the participants would choose print.This, in part, suggests that digital formats could in some situations be the practical, albeit not the ideal, choice.Earlier research also indicates that despite digital formats, printed books remain significant to readers in many ways (see e.g., Herkman & Vainikka, 2014;Kajander, 2020, Kajander, 2022;Thumala Olave, 2020).

Discussion: the ethos of book reading
The contribution of this study is to look into the specific group of those who read in their leisure.The analysis delves into the ethos of a self-identified reader, an active member of the digital information society who also values traditional print culture and the public library network.The statistics in Finland and elsewhere show the trends of decline and the social stratification in reading activity on national levels.With reader-led meaning making in focus, our findings are not in contrast with earlier statistics on a macro level, but they give a more detailed overview of readers of books in the diversifying reading culture of the 2020s.
Although leisure reading among the population as a whole has decreased, those who read are committed readers for whom reading is an integral part of their everyday life.In our interpretation, this may be due to what can be described as a shared ethos of reading.We argue that this ethos is mediated by, among other things, public educational and cultural institutions and reading culture as part of the Finnish welfare state and its prime objective of equality, of which we provided context for our analysis in Section 3.
We are aware of the limitations of our data in terms of the diversity of reading.However, our focus has been on readers of books, either fiction or nonfiction, in all formatsas we invited respondents that identify themselves as readers and friends of books.Therefore, our point of view of reading culture excludes reading activities that mainly consist of participating in "browsing culture", interactional writing platforms, or reading of newspapers, magazines and information seeking both offline and online.Our results call for subsequent analyses especially in terms of the digitality of reading and stratification of literacy.While our data shows that the use of digital formats does not correlate with age, the long-term impacts of digital reading environments on reading time, choice of genre and ways of sharing reading experiences are yet to unfold.As our questionnaire was disseminated via social media and conducted online, it is possible that it did not reach readers that are less familiar with digital environments and whose input could have thus changed the ways our data reflects the usage of digital reading formats.Still, according to our survey data, the majority of the participants prefer printed books when all formats are available (see also Lindh & Hiidenmaa, 2023).
We conclude that in today's reading culture, readers do not only make independent choices when it comes to reading for pleasure or entertainment, but they increasingly master their own reading at large.The prestige or necessity of reading books no longer stem from institutional dominance or relatively uniform reading culture.This shift has entailed the democratisation of reading and the empowerment of readers.However, we suggest that the agency and independence of readers does not only derive from individual choices but instead demand enabling structures: a reading culture that is determinedly built to ensure convenient infrastructures for reading, the availability and visibility of books in public domains, and equal opportunities of participation.We believe that our study has relevance for subsequent research on book reading as a personally chosen practice yet touched by the questions of social environment equity.
The channels and platforms for the creation and distribution of books are in transition, which will continue to impact the whole book trade: literary genres, publishing policies and business models, on one hand, and the practices and the meaning of reading in the lives of readers on the other.Although reading books and its materiality are undergoing change, what remains is the appeal of stories, knowledge and ideas and the advantages of sharing them across all social groups.We trust that by actively fostering the preconditions of equality, reading culture will succeed in attracting readers of books time and again.And vice versa: a vital reading culture maintains equality in society.

Declaration of competing interest
None.

Fig. 1 .
Fig. 1.The distribution of age among the respondents.

Fig. 2 .
Fig. 2. The educational level of the respondents.

FemaleFig. 4 .
Fig. 4. Buying and borrowing books across levels of reading activity.

Fig. 5 .
Fig. 5. Access to printed books across the level of education.

Table 3
The percentage distribution of the age of the participants at the lowest level of education (N = 176).
Fig. 3. Buying and borrowing books in different age groups.