Practices and trajectories: A comparative analysis of reading in France,Norway,the Netherlands,the UK and the USA

Abstract

Using time diary data to examine practices of reading, this article examines trajectories of change within five countries in the last quarter of the 20th century. It employs a conceptual framework derived from theories of practice to illustrate their application in a quantitative and comparative analysis of change in patterns of consumption. Analysing recruitment and defection, the multiplication and diversification of reading-related practices, and the presence of distinct enthusiast groups leads to the rejection of popular claims that the practice of reading is in decline and that this might be a universal process across societies characterized as having ‘advanced reading cultures’. Critiques of cultural homogenization in the context of global consumer cultures are also corroborated. Also, a conceptual and methodological framework for the application of practice theories to the analysis of consumption and social change is advocated.

Keywords

comparative analysis consumption patterns reading theories of practice time-use

Introduction

One of the most enduring and engaging objectives of studies of consumption is to account for patterns of change. This demands an understanding of shifts in how and what goods and services are acquired, how they are distributed across social groups and their uses socially organized, and in the meanings associated with specific forms and patterns of consumption. Steered by theories of globalization and consumer culture, most studies of changing patterns of consumption of commodified products tend to anticipate reduced variation and increased similarity of consumption experiences across cultural contexts. Then, when they examine specific cases they usually find many sources of stubborn resistance to the homogenization of cultural life. Ritzer (2004) captures this dilemma through the rather ugly neologisms grobalization and glocalization, the former being ‘centrally conceived, controlled, and comparatively devoid of distinctive substantive content’ (2004: 3), in which standardized, mass produced (and consumed) goods and services exchanged in disembedded markets represent the substance of empirical evidence. Glocalization, by contrast, captures tendencies that are ‘locally conceived and controlled and rich in substance’ (2004: 8), where personalized interpretations and uses of goods, services and places tend to be emphasized in empirical analysis. While Ritzer conceives of the two tendencies as co-existing, he argues that there ‘is a gulf between those who emphasize the increasing grobal influence of capitalistic, Americanized, and McDonaldized interests and those who see the world growing increasingly pluralistic and indeterminate’ (2004: 80). Most studies, however, for lack of data, are unable systematically to compare and contrast trajectories of consumption in different cultural contexts, especially when the focus is explicitly on what people ‘do’, rather than on their preferences for and purchases of particular goods, or on interpretations of the meanings of goods and services (see Hylland Eriksen (2011) for a more detailed discussion of consumption and globalization). This article approaches the issue of change by way of a comparative research design. We examine what people do by means of time diary data that permit comparison of trajectories in the practice of reading books, newspapers and periodicals in five countries in the last quarter of the 20th century.

We employ an explicit practice-theoretical conceptual framework, an approach currently flourishing in many disciplines and many domains of activity (e.g. Couldry, 2004; Gherardi, 2009; Jones and Murphy, 2011; Pennycook, 2010). Regarding consumption, the promise of theories of practice is that they challenge dominant accounts by exposing and analysing phenomena neglected by cultural theories and by providing an alternative to explanations based on the intentions, attitudes and choices of individuals (Warde, 2005; McMeekin and Southerton 2012). The next section briefly outlines some pertinent features of a practice theoretical approach to the study of consumption, paying special attention to conceptual and methodological issues associated with providing systematic accounts of the broad trajectories of change in practices over time and across space and recommends quantitative comparative analyses for their distinctive insight into the transformation of consumption.

The third section reviews research on reading. Print media are universal features of modern societies, in which the vast majority of people are literate. Reading is a mundane activity, used for many different practical purposes in the conduct of daily life, at home, at work and in public space. It is also a specialized recreational activity, where people devote time and concentration to the reading of books, newspapers and magazines. Not only is the ability to read highly regarded, some forms of reading carry high prestige (Griswold et al., 2005). In context of debates about the collapse of the distinction between high and popular culture, and in the debate about the homogenizing effects of a global consumer culture, reading is an interesting case. The reading of printed materials is widely believed to be being undermined by the spread of global consumer culture, by the emergence of competitive media sources (television, internet), and the growth of international publishing houses that standardize media formats across print and broadcast media. The consequence would be a significant restructuring of patterns of consumption.

The methodological design of the study is described in the fourth section. The fifth section reviews the findings in terms of three analytic dimensions of practice change: recruitment and defection of practitioners; the multiplication and diversification of reading-related practices; and, the presence of enthusiast groups. The last section discusses further the implications of the study for changes in reading and for ways of approaching the study of consumption through the lens of practice theory.

Theories of practice and changing forms of consumption

Theories of practice are, as widely recognized, very diverse (Schatzki, 2011). Emerging in the 1970s, they are identified with a heterogeneous set of authors, prominent among them (according to Ortner, 1984; Schatzki, 1996) being Bourdieu, Giddens, Sahlins, Foucault, Lyotard, Garfinkel, Charles Taylor and Judith Butler. The dominant view is that this development was a diverse response to fundamental and pressing problems of social theory such as structure and agency, the role of rules in social order, and the scientific status of social theories (Ortner, 1984; Rouse, 2006). Renewed attention was signalled by the announcement in 2001 of ‘the practice turn’ in contemporary social theory (Schatzki et al., 2001). One key feature of the reformulation in the later 1990s was to proclaim practices the fundamental unit of social analysis and to suggest that the recursive relationship between individual performances and social practices could provide a sound basis for empirical social science. More generally, interest in ontology has been growing, with practices located within the broad realist school. Schatzki (1996) was the most systematic exponent, but Reckwitz (2002) was also particularly important for the advancement of a sociological approach to consumption. Since the turn of the century, the concept of practice (and various derivatives) has circulated widely, and been applied variously in empirical studies that have described, interpreted and explained social processes and behaviour in specific domains (Postill, 2010). The domain of consumption is no exception, and the Journal of Consumer Research and the Journal of Consumer Culture (see especially Halkier et al., 2011) both have carried several articles since 2005. In economics, too, debates about institutional change have also focused on practices and routines and have begun to see norms in more complex theoretical frames than in earlier decades.

The idea that practices are social and shared is commonly agreed, though explanations of sharing are contentious, depending on whether practices are considered as entities in themselves or are merely similar performances repeated widely across time. When operationalized, practices are generally treated as configurations of recognizable, intelligible and describable elements that comprise their conditions of existence. While there is no single agreed typology of elements (compare options in Gram-Hansen, 2011; and Shove et al., 2012), some combination of material objects, practical know how, and socially sanctioned objectives is deployed, these often in context of socio-technical systems, social and economic institutions, and modes of spatial and temporal organization. Such elements both configure how practices are performed and make them identifiable to practitioners and non-practitioners alike. The relationship between practices and performances is recursive: practices configure performances, and practices are reproduced, and stabilized, adapted and innovated, through performances.

One common criticism of theories of practice is that they account poorly for change. This is partly because most empirical studies have not been designed for such a purpose, mostly being qualitative studies of how specific practices are currently organized (e.g. Watson and Shove 2008; Hargreaves, 2011; Magaudda, 2011). The specific agents which reproduce practices are many and diffuse. Performances are reported on the basis of interviews or observation, making it inherently difficult to extrapolate to, or generalize about, broader trends and processes. Broader questions related to the emergence, spread and decline of practices across cultures, and of the relationship between practices as entities and their performances by particular social groups, require methodological approaches capable of identifying underlying dimensions of processes that lead to the reproduction or transformation of the practice over time and across cultural space. Comparative analysis over time is a means for identifying trajectories of practices and the mechanisms and processes of change. Arguably, such a form of analysis poses and answers different but complementary research questions from those concerned to deliver rich description of the cultural meanings and performances.

The conceptual frameworks of practice-theoretical approaches suggest some critical dimensions for analysing the emergence, spread and decline of practices over time and across cultural space (see Warde, 2005). The first relates to the recruitment or defection of practitioners, how many people participate and how much time is allocated (see also Shove et al., 2012). The second refers to processes of multiplication and diversification: the multiplication of related practices may undermine established orthodoxy, e.g. the multiplication of telecommunications practices undermining conventional telephone calls; diversification has the effect of reproducing the practice in modified forms, e.g. the emergence of diverse forms of text and voice communications through mobile devices. The third analytic dimension relates to the variation and distribution of commitment across participants. The structure of participation may involve a small number of heavily committed enthusiasts and a large number of casual and occasional members, or involvement may be more balanced. Since this may affect the status and the reproduction of the practice it is important to discern structural trends. Finally, in addition, and a dimension that cuts across the other three, the socioeconomic and demographic composition of participants with different levels of engagement has consequences for the trajectory of the practice. Patterns of social differentiation indicate, for example, the extent to which any practice is becoming more socially inclusive or increasingly specialized. We examine data pertaining to the first three dimensions, and consider the role of social differentiation in each to examine trends in the practice of reading printed materials across five countries between the early 1970s and late 1990s.

The decline of reading print material?

Accounts of reading often employ the terminology of it being a practice, although the precise conceptual features that qualify it as such remain unclear. Barton and Hamilton (1998) argued for its conceptualization as a practice, and not merely a description of an activity, because it is: central to diverse domains of life (e.g. work, leisure and domestic organization); institutionalized; and representative of power relationships. Griswold et al. (2011), in their review of empirical research, distinguish between reading for pleasure (recreational reading) and for the purposes of work or education, excluding the latter from their conceptualization of reading as a practice on the grounds that it presents primarily a technical requirement of occupational activity. Clarifications aside, they outline four sets of approaches employed to analyse reading as a practice. The first focuses on the material dimensions of reading practices, particularly the development of print technologies and how print media structure cultural texts, the symbolic value of print materials as an indicator of social status (see Chartier, 2002), and the embodied performance of reading books (Lerer, 2006). The second set of approaches refers to reading as a social practice. Particular attention is on examining social differentiation with respect to the types of texts (books, magazines, newspapers, comics, genres of fiction, and so on), and degrees of participation in the practice. The institutional organization of reading represents a third set of empirical concerns, with the vast institutional apparatus that supports and reproduces the practice (e.g. education systems, Religion, the workplace, the state, and formalized interest groups) identified as regulating literacy standards, contesting the boundaries between ‘high’ and ‘low brow’ content, and arbitrating the representation of culture(s). The final set of approaches seeks to classify societies according to the maturity of reading practices, with consensus formed in terms of a four-fold societal typology: advanced reading cultures with high rates of literacy and established reading practices; restricted reading cultures with high rates of literacy but low levels of participation in reading practices; emerging reading cultures that feature rapidly expanding reading practices; and potential reading cultures where literacy rates remain comparatively low.

It is in societies with advanced reading cultures where anxieties about the ‘death of reading’ are most pronounced, with concerns focused upon the implications for standards of literacy, cultural imagination and creativity, and access to information necessary for citizen participation in democratic processes (Griswold et al., 2011). Evidence that reading practices are in decline is largely derived from survey data that enquires into respondents self-reporting of the frequency and form of printed materials that they read. For example, the National Endowment of Arts (2004) compared reading of printed material between 1982 and 2002 in the USA and found a progressive decline in the number of participants, especially among young adults. Knulst and Kraaykamp (1998) reveal similar trends in the Netherlands. The principal explanation relates these trends to the multiplication of reading-related practices that compete for the attention (or time) of potential practitioners. Television, particularly for the period between the 1970s and late 1990s, has been cast as the major culprit, with surveys showing that younger generations, assumed to be more exposed to broadcast media, read less than older generations in societies with an advanced reading culture (Knulst and Kraaykamp, 1998; Roberts and Foehr, 2004).

The impact of television on reading printed material as a practice is more than a simple matter of practitioner defections to other practices. In these accounts the decline of reading is less a matter of total defection and more a diversification of how people perform media-related practices. In their study of Dutch preferences for different types of printed material Knulst and van den Broek (2003) also refer to inter-generational differences. They show that older survey respondents favoured literature and non-fiction books, whereas young adults preferred adolescent books, comics and thrillers that are more likely to present cultural texts in point form, spectacular fragments and dramatic highlights, which Postman (1992) attributes to the influence of standardized formulae of television for the transmitting of information. Other Dutch studies suggest that preferences for different types of reading material are socially differentiated according to levels of education. Van Eijck and Van Rees (2000), who analysed the changing relationship between reading magazines and newspapers and television preferences in the Netherlands between 1975 and 1995, revealed three groups: those with preferences for television ‘entertainment’ tended to be ‘non-readers’; those who disliked entertainment favoured reading ‘quality’ newspapers; and an emergent group of highly educated ‘omnivores’ whose preferences spanned practices across print and broadcast media.

A second explanation for declining self-reported participation in the reading of printed material is the proliferation of digital technologies. While their diffusion largely post-dates the period studied here, it is noteworthy that concerns about the impacts of digital media on reading practices are similar to those advanced about the impact of television. Not only do digital media compete with printed materials for the time and attention of practitioners, but the format of digital media is claimed to diversify the performance of reading practices in ways that lead Gomez (2008) to predict that printed materials such as novels are destined to become easily navigable, ‘short’ and ‘pithy’ narratives. Empirical research, however, shows a different form of diversification, with studies in the USA (Cole and Robinson, 2002), UK (Gershuny, 2003), and Netherlands (De Haan and Huysmans, 2002) indicating that internet users spend more time reading books, magazines and newspapers than non-users. Griswold et al. (2005) suggest two reasons why this might be the case. First, internet use supports other practices including reading, for it is a source of information about printed media, a means of purchase and a medium through which book enthusiasts have emerged around particular authors and genres. Second, and following Van Eijck and Van Rees’s (2000) Dutch study, a growing number of highly educated cultural omnivores participate in a wider variety of cultural activities; those most likely to read are also more likely to use the internet, a tendency apparently replicated across societies with advanced reading cultures (Griswold et al., 2011).

Together these studies demonstrate that it is unlikely that zero-sum substitutions between different media formats will cause defection from reading printed material. Rather than linear trajectories of decline, research suggests the multiplication and diversification of reading practices. Such processes lead Griswold et al. (2011) to distinguish between reading cultures and reading classes. Reading cultures are present in societies where ‘reading is expected, valued, and common’ (Griswold et al., 2011: 23) and performed on a casual basis by a large proportion of that society. A reading class, by contrast, refers to committed enthusiasts, described as ‘heavy’ or ‘avid’ readers, who share socioeconomic characteristics of being highly educated and affluent. In an earlier study, Griswold et al. (2005) concluded that a reading class who embrace ‘literature, serious non-fiction, and the quality press as an esteemed, cultivated, supported practice of an educated elite’ (Griswold et al., 2011: 139) has become increasingly visible in the USA. According to Griswold et al. (2011), this has been replicated in other societies with advanced reading cultures, with surveys of cultural activities consistently revealing between 10–15 percent of the populations of such societies as being heavy readers who share the ‘educated elite’ characteristics of a reading class (see also Ross et al., 2006).

In the rest of this article we mobilize a form of evidence not yet much exploited in studies of reading to assess some of these empirical claims about patterns of change, that of time diaries. We attempt to present this evidence more specifically in terms of concepts and categories of theories of practice that target change over time. By showing that comparative, quantitative evidence of performances can chart trajectories in national practice we can evaluate hypotheses about both changes in reading and tendencies to convergence in consumer culture.

Time diary data and analysing the trajectories of reading practices

Time diary surveys present a particularly useful tool for comparative analysis of practices given that they provide a relatively consistent measure of what people do in their everyday lives across societies. They have, however, had a rather enigmatic role in sociological analysis. On the one hand they offer a level of micro-measurement detail, almost equivalent to money in economics, because an activity can be calibrated in small units, typically minutes. On the other hand there are many suspicions, not totally misplaced, about the validity and reliability of calculations based on people’s recording and reporting of their time use that always threaten to invalidate deductions made from such apparently precise data. Concern regarding the systematic under-reporting of certain activities by different social groups is, however, largely dispelled by data testing against other survey instruments (see Gershuny, 2000). In any case, for the purposes of this article we require time diary surveys to do no more than provide broad brush maps of the organization of daily life. Of course, the data tells us little about subjective experiences of reading or about changing tastes and preferences in reading genres; this is in line with the main valid criticism of time diary data, that it does not account for experiences of time or the meanings attributed to the activities that it measures (see Southerton, 2009). Nevertheless, analysing what people report doing rather than relying on self-reported tastes gives more reliable indications of cultural activities, such as reading, than does attitude and cultural values data (Gronow and Southerton, 2009). As a source of comparative data, while not perfect, time diaries are the most reliable for exploring, in micro-detail, participation within everyday activities.

The countries analysed in this study were selected primarily in light of data availability. These were the only countries with suitable harmonized data available at two similar time points 20 or more years apart. Nevertheless, the five countries do provide theoretically strategic cases. They include two strongly contrasting cases of levels of participation where the changing practice of reading has already been extensively studied (the USA and Netherlands), two countries where cultural differentiation by social class has been prominent (France and the UK), and a more egalitarian and social democratic Nordic country (Norway). As societies that would each fit the criteria of having advanced ‘reading’ and ‘consumer’ cultures they constitute some test of trajectories towards cultural homogenization. The year of study was selected on the basis of the earliest and latest years in each country where comparable data were available (with each case taking a year from the early-mid 1970s and mid-late 1990s). In order to take account of over-sampling of specific sub-groups, non-response, and variations of sample size and time slots recorded (from 10 to 30 minute slots) weights computed by the Multi-National Time Use Survey (MTUS) were used. Each country was analysed separately in order to isolate trajectories of reading, which were then compared across countries.

Each of the surveys distinguished the reading of books from newspapers and periodicals. Book reading includes manuals, factual and fictional material printed in book format. Reading conducted for the purposes of work or education is excluded, so the data refers to reading for recreation (consistent with Griswold et al.’s (2011) definition of reading practices). It was only possible to analyse reading as a primary activity because the earlier studies do not record secondary activities. Descriptive statistics of mean minutes and participation rates were compiled to provide a broad overview of cross-national trends. Multiple regression analysis was then employed to analyse the socio-demographic basis of the amount of time devoted to the practice. Each step was then repeated separately for reading books and for magazines and newspapers. For the purpose of comparison, demographic variables were limited to those contained in the least comprehensive of the datasets, and include: employment status, gender, age, educational qualifications, single households and living as married, and the presence of children in households. Other variables such as income, social class, and occupational group were not directly recoverable from all the surveys and so were dropped. In such instances, education and employment status tend to act as proxies for social class and income levels.¹ However, such regression equations (one per country per period) are particularly useful for calculating mean differentials in time use, after controlling for all the measured variables, and for estimating the relative importance of socio-demographic variation.

The three dimensions of practice change (introduced in the second and third sections above) were operationalized as follows. Recruitment and defection of practitioners was analysed according to whether the practice has gained or lost practitioners, as a proportion of the total survey population, between the two years, and additionally by the reduction of the average amounts of time devoted to the practice by practitioners per day. Multiplication and diversification is explored through an analysis of the relationship between time spent reading books and time spent reading magazines and newspapers. Multiplication of practices that undermine the performance of reading books or magazines and newspapers can be indicated by decline in the number of participants and the mean minutes devoted to the practice by those participants. Diversification, where practitioners still read but with less commitment, would be indicated by stable participant rates accompanied by declining time devoted to the practice. (It should be noted here that the data amenable for this analysis largely pre-dates the widespread use of digital media in the form of e-books and digital platforms, and as such our analysis is indicative of processes related to multiplication and diversification up to 2000.) Finally, groups of enthusiasts can be identified by estimating the prevalence of persons devoting a significantly greater proportion of their time per day to the practice in comparison with numbers of ‘casual’ participants. Patterns of social differentiation are analysed across each dimension. We estimate degree of cultural homogenization simply by comparing trajectories of each society in terms of the three dimensions of practice change.

Findings

In this section we deal in turn with the three dimensions of practice change: recruitment, defection and decline in overall engagement in reading; changes in the relationship between the two printed media forms that we can distinguish; and levels of enthusiasm. The findings described are discussed further in the subsequent section.

Recruitment and defection

The first dimension for analysing trajectories of the practice of reading focuses on the extent that practitioners have been recruited to, or defected from, the practice, and whether variations across social groups are indicative of social differentiation in the performance of the practice. Taking each country in turn, starting with the Netherlands and USA, existing studies of which suggest have seen significant decline in reading, general patterns of trajectory are examined according to changing participation rates and changing mean minutes devoted to the practice (see Table 1).

Table 1.

Mean minutes and participation rates, for an average day, for sum of book reading, and magazine and newspaper reading, five countries, various dates, respondents aged 16+ (weighted data)

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Mean minutes, survey population (book) (magazines/papers)	24 (10) (14)	25 (18) (7)	54 (13) (41)	42 (12) (30)	28 (5) (23)	39 (9) (29)	22 (3) (20)	27 (7) (20)	31 (5) (25)	22 (7) (14)
Minutes, average predicted, all ages 16 and over	28	16	58	40	30	37	24	23	27	14
Minutes, average predicted for those aged 16–61only	26	12	55	33	28	30	19	18	22	8
Mean minutes, participants only (book) (magazines/papers)	54 (60) (43)	71 (73) (48)	56 (27) (43)	47 (29) (35)	48 (56) (42)	62 (59) (52)	34 (21) (31)	46 (39) (39)	77 (95) (68)	87 (87) (75)
Participation rate (book) (magazines/papers)	44% (16%) (33%)	35% (25%) (14%)	97% (49%) (95%)	90% (41%) (87%)	60% (10%) (55%)	63% (16%) (57%)	66% (13%) (64%)	58% (17%) (52%)	40% (5%) (37%)	25% (8%) (19%)

Note: The base case for regression is a hypothetical inactive married man, whose education is below secondary level and who has no children. However, in the third row, we present the average predicted (from the regression) for all people age 16 to 61. In this way, the third row omits the bias that is created in Row 1 by the presence of retired people in some of the surveys.

The Netherlands exhibited a significant decline of participation (from 97% to 90% of the survey population) and time spent reading (-12 minutes), and those who did participate in 1995 did so for less time than did readers in 1975 (-9 minutes). It is, however, important to note that the Netherlands has a considerably higher participation rate than any other country.² Results of regression analysis (Table 2) were also consistent with previous studies.³ Education was the most significant source of variation; the highly educated read much more than the less educated, and this had increased marginally between the years of survey. Employment status, gender, marital status and dependent children were not statistically significant variables.

Table 2.

Factors influencing the time spent reading, respondents aged 16+

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Full-time	−10.7 (2.1)	−4.2 (0.9)	−0.9 (3.0)	−6.4 (2.2)	−12.8 (2.4)	−12.6 (2.7)	−8.2 (1.8)	−7.8 (1.2)	0.5 (4.1)	−11.6 (6.1)
Part-time	−4.8 (2.8)	1.3 (1.5)	n/a	−1.8 (2.4)	−5.1 (1.9)	−7.4 (3.0)	−4.6 (1.7)	−3.5 (1.3)	−3.0 (4.6)	−9.8 (8.1)
Retired	dropped	6.8 (2.1)	−2.0 (9.4)	5.6 (4.9)	dropped	dropped	23.2 (5.1)	11.0 (2.6)	3.8 (6.6)	−5.8 (16.0)
Unemployed	8.9 (5.6)	7.9 (2.0)	25.0 (8.8)	1.5 (4.1)	dropped	2.3 (7.5)	19.5 (8.8)	6.6 (4.0)	3.3 (7.6)	7.7 (13.6)
(inactive)
Female	−10.7 (1.6)	−2.4 (0.9)	1.3 (3.0)	−0.4 (1.9)	−14.5 (2.1)	6.4 (1.8)	−10.7 (1.8)	−3.3 (1.1)	2.4 (3.8)	6.4 (3.5)
(male)
Age	0.1 (0.5)	−0.1 (0.2)	0.3 (0.8)	1.2 (0.5)	0.2 (0.5)	0.7 (0.4)	0.3 (0.3)	0.3 (0.2)	0.1 (0.6)	−0.7 (1.0)
Age^2	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Secondary education	9.0 (1.3)	9.0 (1.0)	14.3 (3.9)	8.6 (2.1)	n/a	6.2 (2.3)	4.9 (1.8)	5.2 (1.3)	16.6 (3.7)	18.4 (4.8)
Tertiary education	19.3 (2.5)	21.7 (1.3)	16.0 (4.0)	18.4 (2.3)	n/a	16.8 (2.3)	6.8 (1.9)	13.3 (1.3)	23.8 (4.0)	24.2 (4.5)
(below secondary)
Single	3.7 (1.8)	4.2 (1.1)	0.1 (4.0)	0.8 (2.1)	−0.3 (2.7)	−0.2 (2.4)	2.8 (2.9)	−0.4 (1.1)	2.1 (4.4)	1.0 (3.6)
(married)
Young kids	−7.5 (1.8)	−4.8 (1.0)	−3.6 (3.6)	0.5 (2.3)	−2.1 (2.4)	−7.1 (2.3)	−2.2 (2.3)	−5.9 (1.2)	−6.9 (3.7)	−2.9 (3.2)
Older kids	−4.6 (1.6)	−1.3 (0.9)	−2.0 (4.3)	1.6 (2.2)	−3.4 (2.0)	−6.7 (2.9)	−3.8 (1.3)	−4.1 (1.0)	−5.1 (3.7)	−7.9 (5.8)
(no kids)
Constant	28.8 (8.9)	−0.3 (3.8)	29.7 (13.7)	−15.2 (8.3)	37.1 (10.0)	2.7 (7.8)	15.1 (7.1)	7.4 (3.8)	−3.8 (13.1)	6.0 (22.7)
R-squared	0.060***	0.105***	0.134***	0.258***	0.031***	0.162***	0.200***	0.163***	0.078***	0.107***
N	4633	15083	1189	3045	4291	3018	2620	8527	2405	1151

Note: Numbers in brackets are standard errors. A linear regression is used and coefficients are measures in units of minutes per day. Multiple regression within each country is conducted separately for each point in time.⁵

As expected, the USA also revealed a decline in the mean minutes devoted to reading across the survey population (-9 minutes). Most striking however, was the declining participation rate of its population (from 40% to 25%). While it is clear that many participants defected from the practice, those that remained increased their commitment by devoting, on average, an additional 10 minutes per day in 1998 than did readers in 1975.⁴ As for the Netherlands, regression analysis indicates that higher education is the only statistically significant variable for predicting greater durations of time devoted to the practice.

A marginal overall increase of time spent reading in France (+1 minute) masks significant participant defections from the practice (from 44% in 1974 to 35% in 1998). Consequently, those respondents who read during the days surveyed read for longer durations in 1998 than they did in 1974 (+17 minutes). Regression analysis reveals that those in full-time employment read less than any other employment group in both 1974 and 1998. Women read less than men, although the gap narrowed by 1998. Education was the most significant variable, with the highly educated devoting the most time to the practice in both years. Single-person households read more than those who are married or living as married, having older children had a negative effect, and having young children reduced time spent reading in both years.

Norway was the only country to reveal a significant increase in time devoted to reading (+11 minutes) and the recruitment of additional participants between the years surveyed (+3%). Furthermore, practitioners devoted an additional 14 minutes to the practice in 2000 when compared with readers in 1971. Regression equations demonstrate that being in full- and, to a lesser extent, part-time employment significantly reduced time devoted to the practice. Women spent significantly less time reading in 1971 and significantly more time in 2000; the result of a 10 percent increase in women readers over the period compared with a decline of 3 percent in men. Education was important and the higher one’s education the more time spent reading. The presence of children, especially young children, in a household significantly decreased time spent reading in 2000 but made little difference in 1971.

The UK trajectory is similar to that of France. Despite a marginal decline of participation from 66 percent to 58 percent, a small increase of time devoted to the practice (+5 minutes) can be identified and, as in France, the fewer participants increased the amount of time they spent reading by 12 minutes. Those in full- and part-time employment read significantly less than economically inactive respondents and the reverse is true for the retired and unemployed. These differences remained consistent over the period. Women read less than men but the statistical significance diminished a little by 2000. Education was very significant for both years and, as was the case in Norway, having children had an increasingly negative effect on time spent reading.

From a country-by-country analysis of general trajectories of reading practices no clear cross-cultural patterns emerge. Norway and the Netherlands present the two extreme cases, reading increasing on all measures in the former and declining by all measures in the latter. Only the USA shared a trajectory of apparent decline with the Netherlands, where a substantial defection of practitioners contributed to an overall general decline in the mean minutes devoted to the practice. And yet, by contrast to the Dutch, those Americans who did participate devoted more time in the later year. It is also important to take account of the absolute numbers: reading might be on the decline in the Netherlands but time devoted and the number of participants in the practice are greater than in any other country; and, when countries are compared reading remains widespread in the Netherlands but is becoming a minor activity in the USA. Perhaps similarity of trajectory are clearest in the two least remarkable cases, where both France and the UK have seen modest increases in mean minutes spent reading but a general defection of participations.

Regression analysis for each country separately shows that level of education was the most significant socio-demographic variable influencing time spent reading, and this was the case in all years. Employment status and gender were significant, for both years, in France, Norway and the UK, as was the presence of children in households. In the Netherlands and USA, education remained the only statistically significant variable in both years of study. Thus reading was more socially differentiated in the three countries where time spent reading has increased. However, overall, levels of differentiation remained stable.

Despite some shared tendencies, the initial positions in the 1970s were so varied that reading remains very different. Notably despite a significant measure of defection from the practice in the Netherlands, reading remains more prevalent there than elsewhere. Overall, as regards recruitment and defection, there is no compelling evidence of overall cross-cultural similarity in trajectories.

Multiplication and diversification

Distinguishing between two different forms of the practice, reading books as opposed to magazines and newspapers, presents the opportunity to examine if the trajectories of the general practice mask changes in form. As existing research demonstrates, reading books is a different cultural activity from reading magazines and newspapers, the former largely being understood as a practice related to entertainment, the latter often associated with the search for information. The key empirical question is whether trajectories of multiplication or diversification can be identified in the two forms of the practice. On the one hand, the multiplication of reading-related practices might be expected to generate competition for the attention of practitioners, such that fewer participants devote less time to the practice. Diversification, on the other hand, would suggest that practitioners perform a broader range of reading-related practices, although not to the exclusion of reading printed material, such that participation rates remain consistent despite practitioners devoting less time. Whether such processes, if identified at all, are consistent across the forms of the practice will be revealing about whether reading for entertainment or for information represent similar patterns of change.

Table 1 shows small increases in time spent and in the proportion of participants reading books everywhere except the Netherlands, which despite being the only country to see book reading participation decline over the period still had the highest number of book readers at the end of the period. Interestingly, those who read books were doing so for longer durations than earlier, a finding consistent for all except the USA. Book reading is not becoming marginalized.

Regression analysis (see Table 3) reveals that everywhere the most educated spent more time reading books, and in all countries the difference has become more statistically significant over the period; a finding consistent with Griswold et al’s (2011) claim that book readers in societies with advanced reading cultures disproportionately consist of the highly educated. The most significant shifts were, however, related to gender. More women than men participated in reading books and devoted more time to the practice by the end of the period (see Table 4). Participation of men either declined or remained relatively static, and in the USA the time spent by men fell precipitously. In France, Norway and the UK the gender differences reversed: men had spent more time reading books than women in the 1970s, but not in the late 1990s. Finally, France and the UK exhibit greatest social differentiation: employment status was significant and, in France, people living alone read books more, while those with young children read less.

Table 3.

Factors influencing the time spent reading books, respondents aged 16+

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Full-time	−5.7 (1.5)	−3.2 (0.8)	−1.4 (1.8)	−3.4 (1.5)	−2.6 (1.5)	−4.0 (1.7)	−2.3 (0.7)	−1.7 (0.7)	2.8 (2.3)	−4.2 (3.1)
Part-time	−2.7 (2.1)	−0.2 (1.4)	n/a	1.0 (2.1)	−1.5 (1.0)	−1.6 (1.8)	−1.6 (0.6)	−1.8 (0.7)	0.3 (2.0)	−5.4 (3.9)
Retired	dropped	2.6 (1.9)	−4.1 (3.9)	0.2 (3.1)	dropped	dropped	3.4 (1.5)	3.8 (1.5)	3.3 (3.5)	1.4 (8.6)
Unemployed	5.5 (5.1)	6.0 (1.8)	13.7 (7.9)	2.1 (3.2)	dropped	8.4 (6.0)	4.9 (3.8)	4.5 (3.2)	3.7 (3.4)	−1.9 (6.4)
(inactive)
Female	−2.9 (1.2)	2.5 (0.8)	3.5 (1.8)	3.2 (1.2)	−3.3 (1.4)	3.8 (1.1)	−1.3 (0.6)	2.0 (0.6)	2.4 (2.7)	5.4 (1.8)
(male)
Age	−0.6 (0.3)	0.0 (0.2)	−0.4 (0.3)	0.3 (0.4)	−0.4 (0.3)	0.4 (0.2)	0.1 (0.1)	0.0 (0.1)	−0.3(0.4)	−0.2 (0.5)
Age^2	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Secondary education	5.6 (1.0)	8.3 (0.9)	4.6 (2.5)	3.2 (1.2)	n/a	2.8 (1.6)	1.7 (0.7)	1.7 (0.7)	4.9 (2.2)	4.9 (2.8)
Tertiary education	11.3 (1.90)	20.8 (1.1)	5.7 (2.7)	9.0 (1.6)	n/a	5.9 (1.4)	1.7 (0.6)	3.0 (0.7)	3.6 (2.4)	7.0 (2.5)
(below secondary)
Single	4.4 (1.3)	4.4 (1.0)	1.6 (2.2)	3.6 (1.5)	−0.6 (1.5)	2.5 (1.4)	0.2 (0.9)	0.7 (0.6)	4.8 (3.2)	1.5 (2.0)
(married)
Young kids	−4.1 (1.3)	−3.2 (0.9)	−1.5 (2.0)	−0.1 (1.5)	−1.6 (1.4)	−3.3 (1.4)	−0.8 (0.6)	−1.0 (0.7)	−3.1 (2.2)	0.3 (1.7)
Older kids	−1.7 (1.1)	−0.2 (0.8)	−3.5 (2.2)	−0.5 (1.5)	−1.1 (1.2)	−2.5 (1.7)	−0.7 (0.4)	0.0 (0.6)	−1.8 (2.1)	−2.4 (3.1)
(no kids)
Constant	25.5 (6.7)	−5.1 (3.4)	17.6 (6.6)	−5.6 (6.3)	16.9 (5.8)	−5.3 (5.0)	2.6 (1.8)	1.7 (2.2)	2.6 (7.9)	0.4 (12.0)
R-squared	0.043***	0.066***	0.049**	0.071***	0.006	0.028***	0.042***	0.029***	0.018	0.041***
N	4633	15083	1189	3045	4291	3018	2620	8527	2405	1151

Table 4.

Mean minutes and participation rates for book reading by gender, respondents aged 16+ for an average day

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Mean minutes, survey population Men Women	10 9	15 20	11 15	9 14	6 5	7 12	3 2	5 8	4 6	5 10
Mean minutes, participants only Men Women	67 55	75 71	27 27	29 29	65 47	62 57	25 17	43 37	112 87	80 90
Participation rate Men Women	15% 17%	21% 28%	42% 55%	33% 49%	10% 9%	12% 20%	12% 14%	12% 20%	4% 7%	6% 11%

Only in the Netherlands was there any defection from reading books. In the other countries participation rates increased. Also, only in the USA did participants spend less time reading books at the second survey; elsewhere mean minutes per day devoted to reading books increased. Hence, multiplication of media-related practices seems not to have undermined the reading of books. Furthermore, even if book readers have diversified their reading-related practices this has not been at the expense of their commitment (measured by time allocation to the practice of book reading). Perhaps rather than book reading being in decline, with the exception of Dutch and American men, the tendency is for additional recruitment, particularly of women and the highly educated, and also marginal increases in their commitment.

Table 1 shows that time and participation rates devoted to reading magazines and newspapers fell everywhere except Norway. Decline is most dramatic in France and the USA, where only a small minority reported reading magazines and newspapers latterly. However, although participation rates declined, those who did read such material did so for longer durations in later years, except in the Netherlands. In contrast to book reading, the general trend is towards defection, but with greater commitment by the remaining participants.

Regression analysis (see Table 5) indicates that reading magazines and newspapers is a more socially differentiated practice than reading books. Education remains highly significant, and in the UK grew in significance. In France, where magazine and newspaper readership has become a minority activity, the significance of education disappears. Men spend more time reading magazines and newspapers in France and the UK in both years; in the Netherlands this only became the case in 1995; in Norway men read more than women in 1971 but the difference disappears by 2000; while in the USA gender is never not significant in either year. As with book reading, the UK and France are the most socially differentiated, with full-time employment remaining negatively correlated, and in both countries living with children has negative and increasing statistical significance. There are also indications of an age effect in the Netherlands – older people are more likely to read magazines and newspapers in 1998 than younger people.

Table 5.

Factors influencing time spent reading magazines and newspapers, respondents aged 16+

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Full-time	−5.1 (1.2)	−1.0 (0.4)	0.5 (2.2)	−3.0 (1.7)	−10.2 (2.1)	−8.5 (2.1)	−6.0 (1.5)	−6.1 (1.0)	−2.3 (3.2)	−7.4 (4.7)
Part-time	−2.1 (1.8)	1.5 (0.6)	n/a	−2.9 (1.6)	−3.6 (1.7)	−5.8 (2.4)	−3.0 (1.6)	−1.7 (1.0)	−3.2 (4.0)	−4.4 (6.3)
Retired	dropped	4.2 (1.0)	2.1 (7.3)	5.5 (3.8)	Dropped	dropped	19.8 (4.6)	7.2 (2.1)	0.5 (5.6)	−7.3 (9.0)
Unemployed	3.4 (3.1)	1.9 (0.8)	11.3 (5.1)	−0.6 (2.8)	Dropped	−6.1 (4.1)	14.6 (7.2)	2.1 (2.1)	−0.4 (5.6)	9.7 (10.5)
(inactive)
Female	−7.8 (1.0)	−4.8 (0.4)	−2.2 (2.2)	−3.6 (1.5)	−11.2 (1.8)	2.6 (1.5)	−9.3 (1.6)	−5.3 (0.9)	0.0 (2.7)	1.0 (2.8)
(male)
Age	0.6 (0.3)	−0.1 (0.1)	0.7 (0.5)	0.9 (0.3)	0.6 (0.4)	0.3 (0.3)	0.2 (0.3)	0.2 (0.2)	0.4 (0.4)	−0.5 (0.6)
Age^2	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Secondary education	3.4 (0.9)	0.7 (0.6)	9.7 (2.6)	5.5 (1.7)	n/a	3.4 (1.7)	3.2 (1.5)	3.5 (1.0)	11.8 (2.7)	13.5 (3.6)
Tertiary education	7.9 (1.6)	0.9 (0.6)	10.3 (2.8)	9.3 (1.7)	n/a	10.9 (1.8)	5.1 (1.7)	10.3 (1.1)	20.2 (3.2)	17.2 (3.5)
(below secondary)
Single	−0.7 (1.2)	−0.3 (0.5)	−1.5 (3.1)	−2.8 (1.6)	0.3 (2.3)	−2.7 (1.9)	2.5 (2.6)	−1.2 (0.9)	−2.7 (2.9)	−0.5 (2.9)
(married)
Young kids	−3.4 (1.1)	−1.6 (0.4)	−2.1 (2.6)	0.6 (1.9)	−0.5 (2.0)	−3.7 (1.8)	−1.4 (1.9)	−4.9 (1.0)	−3.8 (2.8)	−3.3 (2.7)
Older kids	−2.9 (1.0)	−1.1 (0.3)	1.5 (3.1)	2.1 (1.6)	−2.3 (1.6)	−4.2 (2.3)	−3.1 (1.1)	−4.1 (0.8)	−3.3 (2.8)	−5.5 (4.7)
(no kids)
Constant	3.3 (5.9)	4.8 (1.6)	12.1 (10.1)	−9.7 (5.5)	20.2 (8.4)	8.0 (5.9)	12.5 (6.6)	5.7 (3.0)	−6.4 (9.6)	5.6 (13.7)
R-squared	0.047***	0.087***	0.172***	0.259***	0.031***	0.168***	0.192***	0.162***	0.091***	0.082***
N	4633	15083	1189	3045	4291	3018	2620	8527	2405	1151

To sum up, in each country, people spent more time reading magazines and newspapers than books in the 1970s, but the gap narrowed by the late 1990s as time spent reading books increased (except in the Netherlands), while time spent reading magazines and newspapers declined (except in Norway). France exhibited the biggest shift, with time spent reading books exceeding that of magazines and papers in 1998, a reversal of the 1974 situation. The trajectories of each country suggest that any multiplication and diversification of media-related practices produced defection from magazine and newspaper reading practices more than for book reading (with the exception of Norway, which reveals recruitment to all forms of the practice of reading printed materials). Magazine and newspaper reading is also more socially differentiated across the countries – education, age and the presence of children in households having greater effect. For both forms of reading, education (with the exception of magazine and newspaper reading in France) becomes increasingly statistically significant. Gender is also consistently significant: latterly men read magazines and newspapers more than women and women read books more than men. While there is no clear similarity of cross-national trajectory, the socio-demographic composition of the reading public shows some international commonalities.

Commitment and enthusiasm

Given that participation rates for reading have mostly declined and that participants mostly are reading longer, we can consider whether this generates distinct groups of reading enthusiasts. As operational indicators of commitment we distinguish between casual readers (those who read for less than 15 minutes per day), light readers (who read for between 15 and 60 minutes per day) and heavy readers (who read for over 60 minutes per day). The emergence of a group of enthusiasts would be signalled by polarization in time commitment between casual and heavy readers. Previous studies, particularly those that suggest the emergence of a distinct reading class in societies with advanced reading cultures (e.g. Griswold et al., 2005, 2011), would anticipate that enthusiasm would vary in relation to education, generation/age, and gender.

Table 6 demonstrates that heavy readers represent minorities of the samples everywhere except the Netherlands, where casual readers were in the minority. Only for the two countries in which a general decline of the practice of reading can be identified (Netherlands and USA) are there signs of an emerging group of reading practice enthusiasts – in both countries the proportion of casual readers has increased relative to light and heavy readers. In Norway and the UK there has been a general shift towards greater commitment, while France reveals little change between the years of study.

Table 6.

Proportion of casual, light and heavy readers by country by year, respondents aged 16+ (percentages by column)

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
All reading
<15 minutes	64	66	12	32	56	42	62	59	62	74
15–59 minutes	24	21	58	47	33	37	31	27	21	13
>60 minutes	11	13	30	21	11	21	7	14	17	13
Books
<15 minutes	86	76	72	76	92	78	94	88	95	90
15–59 minutes	9	15	24	15	5	17	5	9	3	5
>60 minutes	5	9	4	9	3	5	1	3	2	5
Magazines and newspapers
<15 minutes	75	86	20	43	61	49	65	65	64	82
1–59 minutes	20	11	63	46	32	37	30	25	21	10
>60 minutes	5	3	17	11	7	14	5	10	15	8

A contrast between reading books and magazines and newspapers is again evident. The number of light and heavy book readers increased, except in the Netherlands, suggesting greater enthusiasm for reading books. This pattern is reversed for commitment to magazines and newspapers: in France, the Netherlands and USA there has been a marked increase in casual readership, with little change in the UK and growing commitment in Norway. Overall, as the gap between casual and heavy readers widened the latter became an increasingly small minority. With respect to claims that in societies with advanced reading cultures a distinct reading class emerges, especially with respect to books, the patterns are unclear. Griswold et al. (2011) claim that a reading class reflects a declining commitment to the practice of reading across the general population (representing a progressive shift from heavy toward casual readers), leaving a small but clearly identifiable minority of heavy readers. While heavy book readers are a minority they are in fact a growing minority in all countries studied. Only for reading magazines and newspapers does the proportion of heavy readers reduce. Evidence for emerging groups of committed practice enthusiasts that could resemble a distinct reading class are not evident in any country other than, perhaps, the USA.

A second criterion for the presence of a distinct reading class relates to social differentiation. Griswold et al. (2005) identify such groups as consisting of the highly educated, while Knulst and Kraaykamp (1998) suggest that in the Netherlands enthusiasts tend to be drawn from older generations. Table 7 presents a logit model of the likelihood of being a heavy reader. They are most likely to be older in all countries except the USA and, latterly, highly educated, while gender became less significant than in the 1970s. In sum, only in the USA can an emerging and distinct group of enthusiasts be identified, one largely confined to magazine and newspaper reading and comprising a minority of older and highly educated people. Patterns are less clear in European countries beyond the general observation that highly committed enthusiasts are a minority in all countries, which alone seems insufficient to support a claim that societies with advanced reading cultures are characterized by a diminishing minority of committed enthusiasts that constitute a distinct reading class.

Table 7.

Log odds of being a heavy reader versus being a casual/light reader by country by year. Respondents aged 16+; logit regression analysis

	France		Netherlands		Norway		UK		USA
	1974	1998	1975	1995	1971	2000	1975	2000	1975	1998
Books
Female	−0.5 (0.2)	0.1 (0.1)	0.5 (0.4)	0.2 (0.2)	−1.4 (0.4)	0.3 (0.2)	−2.3 (1.2)	0.3 (0.1)	0.3 (0.4)	0.8 (0.4)
(male)
Age	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Secondary education	1.0 (0.2)	0.8 (0.1)	0.7 (0.6)	0.3 (0.3)	n/a	0.0 (0.3)	1.4 (0.8)	0.5 (0.2)	1.2 (0.5)	0.8 (0.8)
Tertiary education	1.2 (0.2)	1.6 (0.1)	1.5 (0.5)	0.9 (0.3)	n/a	0.8 (0.2)	0.6 (1.1)	0.7 (0.2)	0.9 (0.5)	1.6 (0.8)
(below secondary)
F-statistic	9.12***	44.44***	3.69***	7.46***	3.85**	4.71***	11.47***	9.80***	3.13**	7.02***
Magazines/papers
Female	–0.8 (0.2)	–1.1 (0.1)	0.0 (0.3)	–0.4 (0.2)	–1.3 (0.2)	0.2 (0.1)	–1.1 (0.3)	–0.5 (0.1)	–0.3 (0.2)	0.2 (0.3)
(male)
Age	0.0 (0.0)	0.0 (0.0)	0.1 (0.0)	0.1 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)	0.0 (0.0)
Secondary education	0.2 (0.2)	0.1 (0.2)	0.6 (0.3)	0.5 (0.2)	n/a	0.1 (0.2)	0.4 (0.3)	0.3 (0.1)	0.7 (0.2)	1.8 (0.8)
Tertiary education	0.9 (0.2)	0.3 (0.2)	0.8 (0.3)	0.8 (0.2)	n/a	0.7 (0.1)	1.3 (0.3)	0.8 (0.1)	1.0 (0.2)	1.9 (0.8)
(below secondary)
F-statistic	7.63***	27.18***	7.64***	22.52***	5.90***	23.50***	12.28***	57.52***	9.47***	5.20***
N	4632	12698	1161	3027	4106	3016	1834	8418	2355	1074

Note: Each odds ratio can be interpreted as the multiplicative change in the odds of heavy reading for a single-unit change in the predictor variable. Each odds ratio is the exponentiated coefficient e^b. The standard errors are shown for each odds ratio.

Discussion and conclusions

Cross-national comparative analysis is challenging theoretically and methodologically. Time diary data represents one of a limited set of suitable resources available to social scientists for systematic comparative analysis because of its capacity to measure micro-detail of cultural participation. This article has focused on the trajectories of practices of reading printed material by analysing three dimensions of practice change that are recoverable from time diary data. It has shed light on three related issues. First, it provides substantive insights into the patterns and dynamics of a highly valued cultural practice that has caused debate and anxiety lest it disappear in the face global cultures of consumption. Second, systematic comparison has proved instructive regarding familiar debates surrounding the homogenizing effects of global consumer cultures. Finally, by developing concepts of change consistent with practice-theoretic approaches to consumption, it has demonstrated the relevance of quantitative evidence to the analysis of changing practices.

A principal concern among those who study reading is the extent to which the practice, particularly in its printed media formats, is in decline. According to critics, this matters because of the need to maintain literacy standards, to excite and ignite creative cultural faculties, and as sources of organized information necessary for democratic deliberations by citizens. Of course, there is no reason to suspect that reading materials in broadcast and digital forms make any less of a contribution to democratic public discussion, although a substantive decline in reading for recreational purposes may well have adverse effects on everyday standards of literacy. Analysing time diary data for reading as a recreational practice (and as a primary activity) over a 25-year period for five countries discloses aspects of its trajectories, even though the available data regrettably does not cover the recent period of proliferation of digital media. Cross-cultural trajectories prove generally dissimilar, with only the USA and Netherlands experiencing significant decline in the practice of reading, and that similarity must be placed in the context of very different absolute levels of participation.

While the countries selected for this study were constrained by available data, the number of cases could now be extended through the use of the Harmonised European Time Use Survey (HETUS). Notably, HETUS data is harmonized at the point of collection, meaning that a greater variety of daily activities can be analysed comparatively, although these will not be immediately compatible with early national time use studies. Perhaps more important is its scope for analysing reading practices as secondary activities. Exploration of the reading that occurs simultaneously with other activities would throw light on the synchronization of multiple practices and the question of the total amount of time spent reading. Analysis of digital media poses different problems. While it would be relatively straightforward to ask respondents to record time use related to the reading of e-books and e-magazines and newspapers, general browsing of digital media or internet use raises more challenging measurement issues given the diversity of activities that relate to the internet (e.g. online shopping, financial planning websites, and so on). Digital media multiply and diversify practices that extend far beyond those of reading printed material. Nevertheless, the challenge is to develop empirical techniques to show how multiple practices diversify, interact, undermine or support each other. The conceptual framework advanced in this article would, at the very least, provide some analytical dimensions against which to examine the trajectories of multiple practices.

This study is another which throws doubt upon accounts of the homogenizing effect of global consumer culture. There are some commonalities of trajectory across the five countries – participants appear to devote more time in the later years examined, which is indicative of greater commitment to the practice, but there are generally fewer readers, especially of magazines and newspapers (and no where do we see a resurgence of reading books or magazines and newspapers). Instead we see path-dependent development country by country. This is clearest in the Netherlands, whose population, despite an observed decline on most measures, still reads far more than in any other country studied. The USA, consistent with the evidence presented by previous studies, similarly experienced decline in reading, especially among those who read little. Yet, despite repeated claims that distinctive reading practice enthusiast groups are emerging across societies classified as having ‘advanced reading cultures’ (e.g. Griswold et al., 2011), we did not find such universal patterns. Trajectories show little similarity, let alone providing evidence of global cultural convergence and homogenization.

The final contribution of this article is its attempt to apply the concepts and methods of theories of practice to trajectories of change. Existing applications of practice theories to changing forms of consumption have overwhelmingly employed interview and observational techniques to provide detail about variation in cultural meanings, interpretations and performances in specific groups or societies. Few have provided a systematic comparative analysis of trajectories of change across several societies. This study identified concepts of dimensions of change that could be operationalized in terms of comparison of aggregate time devoted to the practice by the survey population and, for participants only, shifting rates of participation revealed patterns of recruitment and defection, the multiplication and diversification of reading-related practices and degrees of commitment to the practice. This permits analysis of substantive debates about the changing practices of reading, and provides some conceptual clarification of important dimensions of the practice of reading. Perhaps, more importantly, the conceptual and analytical process has demonstrated the capacity of quantitative comparative analyses to aid understanding of trajectories of practice. Further application of the framework to time diary data could shed light on other pressing substantive issues of change, for example, on the effect of changes in practices of eating, mobility and laundry on sustainable consumption.

Footnotes

Acknowledgements

We are grateful for funding from the ESRC and AHRB for the grant RES-143-25-0002 for a project under the research programme ‘Cultures of Consumption’. This article presents results drawn from the Multinational Time Use Survey (MTUS), and we are grateful to staff of The Centre for Time Use Research, Oxford University, UK, who have facilitated this research. The interpretations of this data and other views expressed in this text are those of the authors and do not necessarily represent the views of the MTUS team or any agency which has contributed data to the MTUS archive.

Notes

Dale Southerton is Professor of Sociology and Research Director of the Sustainable Consumption Institute at the University of Manchester, UK. He is also Director of the ESRC co-funded Sustainable Practices Research Group. His research focuses on accounts of social change, particularly with reference to consumption. Recent research has addressed consumption in relation to sustainability, time use and temporality, theories of practice, innovation, and everyday life.

Wendy Olsen is Senior Lecturer in Socio-Economic Research at the University of Manchester, UK. Her research focuses on the sociology of economic life. She has interests in methodology cutting across the whole range from quantitatively based to qualitative research and discourse analysis. She specializes in the study of economic institutions from sociological and moral economy vantage points. Her research has included case studies of Indian and UK labour markets, the credit market involvement of the poor in India, and other topics in economic sociology including the study of cultural consumption.

Alan Warde is Professor of Sociology and Professorial Fellow of the Sustainable Consumption Institute at the University of Manchester, UK. His research addresses theories of consumption, the application of theories of practice, conventions theory, social network theory and field theory.

Shu-Li Cheng is a Research Fellow at the Institute of Statistical Science, Academia Sinica, Taiwan, and remains an Honorary Fellow of the Manchester Institute of Innovation Research at the University of Manchester, UK. Her research focuses on labour markets, innovation and cultural consumption, with a particular emphasis on developing novel and innovative statistical techniques of analysis.

References

Barton

Hamilton

(1998) Local Literacies: Reading and Writing in One Community, London: Routledge.

Chartier

(2002) ‘Labourers and voyagers: From text to reader’. In: Finkelstein

McCleery

(eds). The Book History Reader, Routledge: London, pp. 47–58.

Cole

Robinson

(2002) Internet Use, Mass Media and Other Activity in the UCLA Data. IT&Society 1(2): 121–33.

Couldry

(2004) Theorising media as practice. Social Semiotics 14(2): 115–132.

De Haan

Huysmans

(2002) Differences in Time Use between Internet Users and Nonusers in the Netherlands. IT&Society 1(2): 67–85.

Gershuny

(2000) Changing Times: Work and Leisure in Postindustrial Society, Oxford: Oxford University Press.

Gershuny

(2003) Web-use and Net-nerds: A Neo-Functionalist Analysis of the Impact of Information Technology in the Home. Social Forces 82(1): 139–66.

Gherardi

(2009) Introduction: The critical power of the “practice lens”. Management Learning 40(2): 115–128.

Gomez, J. (2008) Print Is Dead: Books in Our Digital Age. Macmillan, London.

10.

Gram-Hanssen

(2011) Understanding change and continutity in residential energy consumption. Journal of Consumer Culture 11(1): 61–78.

11.

Griswold

McDonnell

Wright

(2005) Reading and the Reading Class in the Twenty-First Century. Annual Review of Sociology 31: 127–41.

12.

Griswold

Lenaghan

Naffziger

(2011) Readers as Audiences. In: Nightingale

(ed.) The Handbook of Media Audiences, Oxford: Wiley-Blackwell, pp. 19–40.

13.

Gronow

Southerton

(2009) Consumption and Leisure in Europe. In: Therborn

Immerfell

(eds). Handbook of European Societies: Social Transformations in the 21st Century: A Reference to EU Societies and their Transformations in the 21st Century, London: Springer, pp. 355–384.

14.

Halkier

Katz-Gerro

Martens

(2011) Applying practice theory to the study of consumption: Theoretical and methodological considerations. Journal of Consumer Culture 11(1): 3–14.

15.

Hargreaves

(2011) Practice-ing behaviour change: Applying social practice theory to pro-environmental behaviour change. Journal of Consumer Culture 11(1): 79–100.

16.

Huysmans F, van den Broek A and de Haan J (2005) Culture-lovers and Culture-leavers: trends in interest in the arts & cultural heritage in the Netherlands. The Hague: Social and Cultural Planning Office.

17.

Hylland Eriksen

(2011) Globalization. In: Southerton

(ed.) The Encyclopedia of Consumer Culture, Thousand Oaks, CA: Sage, pp. 670–678.

18.

Jones

Murphy

(2011) Theorising practice in economic geography: Foundations challenges and possibilities. Prog Hum Geog 35(3): 366–92.

19.

Knulst

Kraaykamp

(1998) Trends in leisure reading: Forty years of research on reading in the Netherlands. Poetics 26(1): 21–41.

20.

Knulst

van den Broek

(2003) The readership of books in times of de-reading. Poetics 31: 213–33.

21.

Lerer

(2006) Epilogue: Falling asleep over the history of the book. PMLA 121(1): 229–234.

22.

Magaudda

(2011) When materiality “bites back”: Digital music consumption practices in the age of dematerialization. Journal of Consumer Culture 11(1): 15–36.

23.

McMeekin

Southerton

(2012) Sustainability transitions and final consumption: Practices and socio-technical systems. Technology Analysis & Strategic Management 24(4): 345–361.

24.

National Endowment of Arts (2004) Reading at risk: a survey of literary reading in America. Research Division Report 46, Washington DC: National Endowment of Arts.

25.

Ortner

(1984) Theory in anthropology since the Sixties. Comparative Studies in Society and History 26: 126–66.

26.

Pennycook

(2010) Language as a Local Practice, London: Routledge.

27.

Postman

(1992) Technopoly, New York: Knopf.

28.

Postill

(2010) Theorising media and practice. In: Bräuchler

Postill

(eds). (2010) Theorising Media and Practice, New York: Berghahn Books, pp. 1–34.

29.

Reckwitz

(2002) 'Toward a theory of social practices: A development in culturalist theorizing'. European Journal of Social Theory 5(2): 243–63.

30.

Ritzer

(2004) The Globalization of Nothing, London: Sage.

31.

Roberts

Foehr

(2004) Kids and Media in America, Cambridge: Cambridge University Press.

32.

Ross CS, McKechnie EFL and Rothbauer PM (2006) Reading Matters: What the Research Reveals about Reading, Libraries, and Community. Libraries Unlimited, Westport, CT.

33.

Rouse, J (2006) ‘Practice theory’, in Handbook of the Philosophy of Science, vol.15: Philosophy of Anthropology and Sociology, (eds. S.Turner and M Risjrod), Elsevier, 500–540.

34.

Schatzki

(1996) Social Practices: a Wittgensteinian approach to human activity and the social, Cambridge: Cambridge University Press.

35.

Schatzki

(2011) Theories of practice. In: Southerton

(ed.) Encyclopedia of Consumer Culture, Thousand Oaks, CA: Sage, pp. 1447–52.

36.

Schatzki

Knorr Cetina

von Savigny

(2001) The Practice Turn in Contemporary Theory, London: Routledge.

37.

Southerton

(2009) Changing Times: Quantitative analyses of time use. In: Devine

Heath

(eds). Doing Social Science: Evidence and Methods in Empirical Research, second edn. London: Palgrave, pp. 59–79.

38.

Shove

Pantzar

Watson

(2012) The Dynamics of Social Practice, London: Sage.

39.

Van Eijck

Van Rees

(2000) Media orientation and media use: Television viewing behavior of specific reader types from 1975 to 1995. Communication Research 27: 574–616.

40.

Warde

(2005) Consumption and theories of practice. Journal of Consumer Culture 5(2): 131–153.

41.

Watson

Shove

(2008) Product, project, competence and practice. Journal of Consumer Culture 8(1): 69–89.