Examining the Links between Parents’ Relationships with Reading and Shared Reading with their Pre-School Children

While much is known about the benefits of shared reading activity for children, and the role of the home in cultivating shared reading practices far less is known about the factors that can influence parents’ shared reading practices with their children. Given that many young people leave school with poor relationships with reading, this study explores the links between parents’ own relationships with reading and the shared reading they practice with their own children in the home. Drawing on deep-level interview data, this paper presents data from six parents of pre-school children, who reported that they have had a poor personal relationship with reading. These parents all developed positive shared reading relationships with their children, however the importance of this study lies in understanding the interplay between these reading relationships. The data strongly suggested that the construct of ‘reading’ was very different from the ways in which reading had previously been defined for these participants. Reading, within a shared reading context, was seen as a very flexible construct which included activities such as talking and telling stories. In some cases, parents’ own relationships with reading seemed to improve. Implications for intervention with other families are discussed.


IJEP
ngaging in regular shared reading activity is beneficial for children. Over the years a number of studies have shown how shared reading in the home supports young children's language development (Snow, 1994;Bus et al, 1995;Cline and Edwards, 2013). In 1995 Bus et al examined data from a number of studies on the frequency of book-reading to pre-school children and found that shared book reading had a moderate to strong relationship with measures such as emergent literacy, children's language growth and reading achievement. Importantly, this analysis revealed that even when children were living in areas of disadvantage, where levels of home literacy tended to be lower, frequent shared book reading had a positive impact on children's literacy skills. Not surprisingly, further studies have indicated that it is not just the frequency of shared reading activity that is important, but how the activity is implemented. For example, Mol et al (2008) conducted a meta-analysis of 16 studies, concluding that shared book reading was more likely to support children's language development when parents used specific techniques, such as evoking conversation with the child about the text. In other words, Mol et al (2008) were arguing that 'dialogic' reading was more effective than simply reading 'to' a child.
As a result of such research, many interventions have been driven by the intention to encourage parents to read regularly and more productively with their children. Yet such intervention is often built on the assumption that there is an optimum 'standard' of shared reading activity that all families should be aiming for, regardless of the individual and unique characteristics of the family in question. Yet research has also shown that it is important to recognise such characteristics. For example, a further meta-analysis by Manz et al (2010) supported Mol et al's (2008) findings, in that they reported an overall effect size for dialogic reading interventions that was consistent with that reported by Mol et al (2008). However, as effect sizes from emergent literacy interventions were found to be greater in studies involving participants of middle or high socioeconomic status, they concluded that interventions must therefore be sensitive to the possibility that they may not be consistent with the values, routines or resources of various families such as those of lower-income and ethnic minorities. The importance of recognising individual need within reading interventions was also highlighted by Reese and Cox (1999), who compared the impact of different reading styles on children's vocabulary development and found that the benefits of different reading styles on children's emergent literacy varied according to children's pre-intervention skill level.
This might go some way in explaining why reading interventions have not always been as successful as hoped. This was explicitly addressed by Justice, Logan and Damschroder (2015) who reported that interventions with caregivers often resulted in fewer sessions being implemented in the home than was recommended, a lack of adequate records detailing the implementation and, in some cases, the implementation of the intervention ceasing altogether (Justice, Skibbe, McGinty, Piasta, & Petrill, 2011;Lonigan & Whitehurst, 1998). This led Justice et al (2015Justice et al ( , p. 1852 to conclude: 'It is most certainly true that implementation of shared reading interventions by caregivers within the home environment does not always reach the levels intended by the intervention developers'. Together this suggests that if we are to be successful in encouraging more families to embed shared reading activity within their everyday lives, we must begin with families and understand the factors that may both motivate and discourage parents from different socio-economic, ethnic and cultural backgrounds from reading with their children. Interestingly, while many studies over the years have stressed the importance of understanding the role of home culture on children's literacy development (Pahl, 2002;Minns, 1997;Barton and Hamilton, 1998;Gregory et al, 2004), very little research has focused specifically on the factors that influence shared reading practices in homes. Seminal research by Heath in 1982 researched patterns of language use related to books, in three literate communities in the South Eastern United States, focusing specifically on 'literacy events' such as the bedtime story. This comparative study provided a unique insight into the different ways in which young children, from these three diverse communities, were able to 'take' from books and other written materials, and the extent to which this prepared them, or not, for success in the schooling system. Heath concluded that the findings from this study showed 'the inadequacy of the prevalent dichotomy between oral and literate traditions' (Heath, 1982: 49), an issue that is raised in the findings of this present study and reported later in this paper. Heath further concluded that study into language use in relation to written materials in the home and community 'requires a broad framework of sociocultural analysis', which again confirms the importance of acknowledging the sociocultural context within which shared reading exists.
In recent years, it has become increasingly recognised that children's preschool reading experience now includes substantial access to digital texts, which has challenged traditional notions of the terms 'reading' and 'shared reading'. Indeed Bearne et al (2007, p. 11) noted that 'very young children show expertise in on-screen reading, even where homes may not have computers'. Others have argued that it is clear that 'young children are immersed in practices relating to popular culture, media and new technologies from birth' (Marsh et al, 2005). This suggests that reading, and indeed shared reading, may involve a wide variety of texts including those on paper and on screen. However, the literature suggests that books remain a valued and valuable resource, especially for young children (Evans, 2009;Lamme et al, 2004). What is more, even though the role of digital technology was discussed at length in the interviews in this study, parents overwhelmingly talked about their use of books in relation to shared reading with their children. For this reason, shared reading, in this paper, refers largely to a use of book texts While some have attempted to investigate the barriers to shared reading in the home, these have tended to be quantitative in nature, asking participants to select their responses from a fixed set of response choices (Harris et al., 2007;Lin et al, 2015). Yet if we are to understand the reasons why some parents do not read with their children, or read infrequently, then there is a need to talk to parents to understand how shared reading is perceived and how it does or does not fit within their individual constructions of family life. Given that a practice such as shared reading can be viewed as a 'convention', evident in the aims of reading interventions as discussed above, yet is clearly unique to each family, there is a need to explore this phenomenon from a perspective of 'cultural connectedness' (James andCurtis, 2010, p. 1164). In other words, as James and Curtis point out, it is this cultural connectedness that rests at the heart of how families display aspects of their everyday life; moreover it is this duality of social expectation and individual family culture that helps to explain the paradox of how 'families can be experienced as unique, while also reflecting social conventions and reproducing commonplace ritual and practices' (Smart, 2007, p. 51)'. 128 Levy, The research presented in this paper sought to understand factors that influenced parents' shared reading practices in the home. This study, which is part of a larger Economic Social Research Council (ESRC)-funded project, drew from interviews with 29 parents of pre-school children, in two cities in the UK. These interviews provided an insight into the families' day to day life, everyday family practices and relationships and the role of reading within this. While the data revealed a number of factors that influenced shared reading practices in the home, careful analysis of the interviews allowed us to specifically understand links between the parents' own relationship with reading and their shared reading practices with their children. Given that this has important implications for how reading interventions may be developed, this paper explains why it is important to understand this link and offers suggestions for intervention on the basis of findings from this study

The Role of Parents in Shared Reading Activity
For many years now, researchers have explored the role of the home and community in children's development of language and literacy skill (Heath, 1982;Minns, 1997), concluding that the home environment is critical in shaping children's literate identities (Cameron and Gillen, 2013;Perregaard, 2010). As discussed in the introduction, there is no shortage of literature supporting the view that shared reading in the home has a number of benefits for children including support for their language and emergent literacy skills (Bus et al., 1995, Mol et al., 2008Anderson, Anderson, Friedrich, & Kim, 2010) through to the promotion of social and emotional advantage (Vandermaas-Peeler, Sassine, Price, & Brilhart, 2011).
It is not surprising that the literature has paid close attention to parents' roles within their children's reading activity and much of this literature has focused on the ways in which parents read with their children. For example, researchers have argued that shared reading is more productive when parents construct conversations 'surrounding' the book (Gjems, 2010), and invite their children into these discussions (Snow, 1994). Further research has focused on activities that parents may include within their shared reading encounters, such as singing songs and reciting rhymes (Bennett et al. 2002;Owoki 2001), arguing that such activity promotes phonological awareness which is important in the development of early literacy skill.
Studies with older children have indeed revealed that children may be more motivated to read if they see a parent reading for pleasure. For example, Clark and Hawkins (2010) noted from their study that nearly twice as many young people who report that they see their mother and father read a lot, also report that they read outside of school every day, in comparison with the participants who stated that they do not see their parents read. Similar findings were reported by Love and Hamston (2004) in Australia, who investigated the recreational reading practices and attitudes towards reading of two cohorts of teenage boys and their parents, however Love and Hamston also concluded that the boys reacted to their families' reading cultures in unique and individual ways. Further study has shown that gender may also be a factor in the parent-child reading relationship. In a large study of households with children aged 8 and over, Mullan (2010) analysed time-diary data on the reading of parents and young people in the households. Mullan concluded that while parents do have a positive impact upon young people's reading, there was a 'pronounced gendered dimension in that mothers' reading is positively associated with girls' reading and fathers' reading is associated with boys' reading' (p. 427).
To bring this together, we know that parents play a key role in supporting their children's reading and that shared reading activity between parent and child is highly beneficial for children on a number of levels, but we know very little about the factors that motivate and indeed prevent parents from reading with their children. Few studies have explicitly explored the barriers to shared reading with young children; among those that have parents were restricted to selecting from a set of fixed responses, which may not measure the things that are most relevant to parents (Harris et al, 2007). More recently, Lin et al (2015) considered a wider range of potential barriers, in an attempt to investigate the factors that prevented mothers from reading to their 18-monthold children. This study revealed that mothers are more likely to report childcentred barriers than any other type, however the authors acknowledged the limitations of restricting the barriers that parents could report.
What is clear from this literature is that we know very little about the relationship between parents' own reading and their reading relationships with their children. Research into the relationship between parents' reading and their children's reading has not only largely focused on older children but has 130 Levy, Relationships with Reading also tended to try and establish whether parents' reading influences what their children read (Chandler, 1999), and how often their children read (Clark and Hawkins, 2010). But very little is known about the ways in which parents' own relationships with reading have an impact, positively or negatively, on their desire to read with their own children. This may be particularly important for parents who have not had strong relationships with reading themselves.
Research indicates that many young people leave school with an identity of being a 'poor' or 'struggling' reader (Alvermann, 2001). This is particularly alarming when we consider that in the United States, for example, the Census Bureau reported that out of 3.9 million eighth graders who took the National Assessment of Educational Progress (NAEP) in 2007, twenty-six percent 'did not attain basic levels of literacy, and only 31% reached proficiencymeaning that roughly 1 million eighth graders were stalled at basic literacy levels and another 1.7 million were not proficient' (Greenleaf & Hinchman, 2009, p5). Further research has suggested that for many of these young people, difficulties with reading have plagued them since their early years in school (Wanzek, Wexler, Vaughn, & Ciullo, 2010). Hall (2016, p. 56) therefore warns that as a result of this constant 'failure', 'these students may refrain from engaging in classroom reading practices because of the belief that they cannot succeed, have little to offer, and cannot change their situation'.
While this is clearly a situation that warrants attention, it is not the purpose of this paper to explore this phenomenon in any further detail, but rather to raise the following question; as many of these young people go on to become parents themselves, does this poor relationship with reading have a negative impact on their reading relationships with their own children? There are a number of ways in which this question could be explored, however, given that very little is known about the relationship between parents' own reading and their reading relationships with their children, there is a need to begin with exploratory research that invests time in talking with parents about their own reading, their shared reading practices with their children and the ways in which reading does and does not fit within the context of their everyday lives. The next section describes the design of this study, which was based upon indepth semi-structured interviews with 29 families living in two cities in the north of the UK IJEP -International Journal of Educational Psychology, 7(2) 131

The Study
As Lewis and Fabos (2005, p.474) noted, understanding literacy practices means moving beyond 'events' with texts to encompass the interweaving of events with broader social and cultural norms, beliefs and attitudes. This has significant implications for understanding shared reading in the home and the relationships that are developed within the context of these practices, and very much influenced the design of this research. This study, which is part of a larger program of research to explore the impact of shared reading on children's language development, recruited participants from inner-city areas that were considered as relatively disadvantaged on the Indices of Multiple Deprivation 1 . The research was advertised to parents at various nursery schools, and participants were also recruited in playgroups, health visitor drop-ins and children's centres in low and mixed-income areas. We checked that each potential participant had a child who had not yet started school.
Participants were aged between 21 and 36+ with the majority falling into the 26 -35 bracket. Of the 29 families participating in the study, 14 had two children. Children were mainly aged between 3 and 5 years of age. Around half of the sample described their ethnicity as White British/Irish (n = 14); the remaining participants described themselves as Asian/Asian British (n=7), Mixed White and Other (n=4), Arab (n=3) and black (1). In terms of qualifications, 12 participants were educated to degree level or higher, 8 to GCSE (General Certificate of Education) and 5 did not possess any formal qualifications. It should be noted that although we were keen to talk to fathers and mothers, it was mothers who responded to our request for participation in the vast majority of cases. As a result, interviews were conducted with 28 mothers and one father, however most participants spoke about the whole family during the course of the interview.
We drew upon concepts of 'narrative inquiry' (Clandinin and Connelly, 2000) in designing the interviews, to encourage participants to present the story of their experience. Given the need to access that 'the wisdoms, beliefs, assumptions, and lay theories of sociocultural practices' (Lim and Renshaw, 2001, p.15), in-depth semi-structured interviews were designed so as to facilitate this access. This was important because we did not want to talk about reading practices in isolation, but rather we sought to ascertain a picture of family life and understand how reading practices were situated within everyday family structures, routines and activities. Interviews were conducted in participants' homes. The length of the interviews varied substantially; we very much took our lead from the participants in this respect as some participants clearly wanted to talk more than others.
The interviews were audio recorded and transcribed verbatim; to preserve anonymity, pseudonyms have been used and all potentially identifying information has been removed. The data were initially analyzed within Nvivo, which is a software package designed for qualitative data. Analysis followed the principles of grounded theory analysis, with three distinct stages taking place; open coding, clustering of codes around categories and thematic coding (Braun and Clarke, 2006). Open coding and thematic coding were conducted by three researchers independently. The resulting analyses were compared and the researchers agreed a set of emergent core themes. We then conducted a further phase of data reduction in order to understand the link between parents' own personal relationships with reading and their shared reading relationships with their children. This resulted in the following five overarching themes: • Parent attitude to school • Parent reading as a child (including being read to) • Parents' reading as an adult • Shared reading relationship with child • How parents and child read together Analysing the data within these themes allowed us to understand the parent's own personal relationship with reading, as a child and as an adult, and situate this alongside the shared reading relationship that they had developed with their child. Given constraints on time, we decided that while our interest was in exploring this connection, it was important to firstly categorise the families, so we could focus our attention on the data that would be most useful in informing reading intervention. Based on the themes described above, we then placed the families into one of these four categories (this was agreed by two researchers): • DID read as a child/ young adult and DOES read to own child now • DID NOT read as a child/ young adult and DOES read to own child now • DID NOT read as a child/ young adult and DOES NOT read to own child now • DID read as a child/ young adult and DOES NOT read to own child now It should be noted that the purpose of using this categorisation was not to make a quantified statement about the amount of families that fell into each category, but to help us to focus on the data that allowed us to understand if parents who had/have a poor relationship with reading themselves were able to develop positive reading relationships with their own children. As one would expect, participants within each category varied substantially; for example, some participants in the 'DID NOT read as a child/ young adult and DOES read to own child now' group reported that they disliked reading at school and do not read as an adult, while others started to enjoy reading at a later stage in life. This again underlines the fact that while the categorisations alone did tell us something about the links between parents own relationship with reading and their shared reading relationships with their children, the depth of understanding came from analysing the data within the individual interviews.
Having analysed the data according to the themes presented (See Figure 1 below), we found that 18 out of the 29 parents fell into the category 'DID read as a child/ young adult and DOES read to own child now'. 10 of the parents were assigned the category 'DID NOT read as a child/ young adult and DOES read to own child now' while only 1 parent fell into the category 'DID read as a child/ young adult and DOES NOT read to own child now'. This parent spoke about wanting to read to her children, but felt unable to, due to several health issues. No participant fitted the category 'DID NOT read as a child/ young adult and DOES NOT read to own child now'. While the data set as a whole gave a unique insight into the factors that motivate parents to read with their children, including those who had positive relationships with reading themselves, (these findings have been reported elsewhere -see Preece and Levy, in press), this paper was more concerned with understanding the data of the parents who had reported poor relationships with reading themselves. For this reason the next section will focus on the participants who fell into the category 'DID NOT read as a child/ young adult and DOES read to own child now'. While all ten of these participants provided valuable data that supports the arguments presented in this paper, we have only drawn on data from six, given the limitations of space. The following table provides information about these six participants whose data are reported in the findings below.

Findings
It is certainly encouraging that within this sample, all of the participants who reported that they did not have a strong relationship with reading when they were children, and in many cases this continued into adulthood, went on to develop positive shared reading relationships with their own children. Of course it is not possible to generalise from this sample, and this was not the purpose of the study. Rather, it is important to understand these reading relationships as this could help inform reading interventions with other parents who do not have a good relationship with reading themselves, given that it shows the potential for reading relationships to change throughout the life course.
Analysis of this data revealed that for these parents, the concept of 'reading', within a shared reading relationship, was quite different to the ways in which parents perceived reading within their own lives. What is more, in a few cases it was clear that through the context of developing a positive reading relationship with their children, some parents became more confident with reading themselves. These findings, and their implications for intervention will now be discussed.

Shared Reading Is Different from 'Reading'
The data strongly suggests that one of the reasons why these parents, who reported that they had not enjoyed reading themselves, went on to enjoy reading with their children, pertained to the fact that the concept of 'reading' was seen differently by the parents within these two contexts. In other words, these parents appeared to believe that reading to a child did not seem to carry the same connotations about reading as was evident in their own experiences with reading.
Not surprisingly, these participants often spoke about school when asked to talk about their own interactions with reading, however these experiences were often negative. For example, Bina reported that she was expected to read aloud in class in secondary school, but she 'wasn't confident' and 'didn't like doing it' but 'had to do it', concluding that the process made her feel 'shy'. Similarly when asked to talk about any reading that happened when she was a child, Fiona spoke about 'English Literature' and the fact that 'the teacher would read some and then he'd pick people to read out of the book'. In relation to this she continued: 'Yeah, I just remember that nervous feeling of, it went -oh no, when he comes to me to read, is there gonna be… you'd be trying to skim through to see if there are any big words that you might not be able to pronounce, that kind of thing. I remember that anxiety a little bit'.
The anxiety connected to being made to read aloud in class was a familiar theme amongst these parents and was almost always situated in a fear of not being able to read particular words in a book. This was described in detail by Natalie who reported: I've got certain memories in English where I couldn't read certain words and I used to have to ask the teacher and obviously you don't want to have to ask the teacher in front of kids. I can remember it was a big long word and I remember thinking, I didn't know what it was and I got the question wrong. I knew I'd got the question wrong because I couldn't understand what the word was but I didn't want to ask the teacher.
What is more, when asked to talk about her own experiences of reading, Kylie directly connected the fact that she never read to herself, with being made to read aloud in class. She stated: 'I never read myself. I can remember reading out in class and I wasn't confident around them and I'm still not, don't enjoy reading, not out loud'.
It is not difficult to see how these reading events were not positive experiences for these participants. In each of these cases the participant seemed to be focusing on the act of decoding print so they could 'pronounce' the word correctly or make sense of the word for the purpose of getting a question right. Rather than enjoying the experience of reading a book together as a class, these participants were describing reading as a compulsory experience which was testing their ability to decode print. As difficulty in decoding could result in public humiliation, these reading experiences were clearly stressful.
The perceptions and behaviours of others was also mentioned by two of the participants in relation reading being a mark of social acceptance. For example, when asked to talk about reading as a child, Bina stated: 'My experiences of reading were having a Tracy Beaker book 'cos I thought that 138 Levy, was cool at the time, because my friends had them, that might have been beginning of secondary school, but I never actually read them'.
Bina is suggesting that she liked the idea of reading certain books as a teenager because this was seen as a 'cool' thing to do, however, she did not enjoy reading the books and therefore did not manage to conform to this norm. Similarly, when describing the reading that took place during her secondary school years Latika also reported: I did do a little bit of, erm, start reading because my friends were reading, and I said 'ok, I'll start reading novels and stuff', because my friends are doing it, just …because they are doing it. I find it so boring' What is interesting to note here is that both participants are describing reading as something they felt they ought to be doing, as determined by an external structure. They wanted to read novels as this fitted a social norm, however they did not enjoy it. It is not clear why these participants did not enjoy reading; in fact it is possible that even the participants did not know why. For example, Bina spoke of the fact that she just 'wouldn't choose to get a book and read'. She went on to state that she 'wished' she was 'like that' (being someone who enjoyed reading), but she has 'just never been like that'. Similarly, Latika spoke of other people who would say 'I love reading! I love reading', which made her think 'I'm the only one on this Earth who doesn't like reading'. These words were also echoed by Kylie who spoke about her lack of reading in relation to others when she reported 'You know when people go on holiday and get into a book? I just can't get into one'.
What is emerging from this data is that reading, for these participants, was not a comfortable activity during their school years. Together they are describing reading as a messy construct which for them combined difficulty with decoding with a lack of interest in reading books. What is more, reading was also fraught with expectations, emanating from both the school discourse and their social context, however these participants were unable to meet these expectations. Much of what they are reporting can be summed up in the words of Hadra who actually became an avid reader later in life but did not enjoy reading at school. Hadra told us: 'I used to struggle a lot in school with reading… I wasn't really into it I, I suppose, as in actually reading books, so, yeah, I used to struggle a lot'. This concept of reading being a 'struggle' and something that they are not 'into' was very evident in these participants' accounts of their own reading but did not feature within their narratives pertaining to their shared reading encounters with their children. In contrast, reading to their children was described as being easy, relaxed and enjoyable. For example, Kylie who told us that she 'never read', 'wasn't confident' and did 'not enjoy reading…out loud' went on to describe reading to her son as being 'so easy'. She reported: I think it's one of the easiest things you can do, because you can sit and read for half an hour or five minutes, it's just something that you can fit in. There's no cleaning up afterwards. Just put the book back and get on. You can do it while the tea is getting cooked, things like that Coupled with being seen as an easy thing to do, was the suggestion that reading to a child was enjoyable. Moreover these participants were not only saying that they enjoyed reading to their children, but were clear about the fact that they would most probably not read to their children if they did not enjoy it. This was particularly evident in Latika's account. Speaking of reading to her daughter Latika stated: Having claimed that she wished she enjoyed reading for herself but did not read because she found it 'so boring', it is striking to see that Latika is talking about being completely engaged with the activity of reading to her daughter. What is particularly noteworthy is the fact that she then went on to say that although she does believe that it is good to read to your child, she does it because 'I enjoy reading for her, it's not that I'm forced'. This stands in sharp contrast with the data relating to these participants' own school-based reading, which indicated that they did indeed feel forced to read in ways that made them uncomfortable and read texts that they did not enjoy.
Clearly, these participants saw the act of reading to their children as being very different to their own experiences of reading for themselves. What is more, they actively wanted the process to be different for their own children; this is salient given that these participants seemed to have been especially traumatised by their experiences of making mistakes when reading aloud in class. This was evident in Natalie's interview when, speaking of her older children, she reported: I want them to enjoy it and want to do it. So I try and get them to enjoy it. So I was a bit disappointed in the school when they said they were focusing more on phonics and things While not all participants specifically spoke about phonics, many spoke about wanting to prioritise enjoyment when reading with their children above the act of 'learning to read'. For example Hadra told us: To be honest, at the moment, importance in terms of how her reading is and how good she is with it, it doesn't come into it. I think, that's not why we do it. We do it because she enjoys it and it's something we can do together' This again underlines the fact that these participants were not only enjoying reading to their children but appreciated the fact that the purpose of this reading was to enjoy the experience together, rather than focus on the accuracy of print decoding. In fact, further data suggests that for many of these parents, 'reading' included a whole variety of features that extended beyond the actual text. A recurring theme throughout the data was that shared reading provided parents with an opportunity to spend 'protected' time with their child, and this included time to talk. Natalie spoke of their shared reading as: It's our little bonding time really, that time together. It goes off books as well, just because we've got books in our hand, we talk about the rest of the day, we get a story in, but then that's our time Natalie is here articulating that the book is just part of the process; while the book was included in the interaction, the experience allowed for parent and child to spend time talking and indeed 'bonding'. Very similar sentiments were expressed by Hadra, however what is particularly interesting here is that Hadra reported that it is her husband who tends to read to their daughter; however unlike Hadra, who became interested in reading at a later stage in life, the husband was described as being someone who 'doesn't read at all'. Describing the shared reading interaction between her husband and daughter, Hadra reported: So he gets her changed and reads her her book, and then just tells her some stories and asks her about her day. So they talk -I wouldn't say rubbish (laughs), they talk randomly, and then he'll tell her a story, and I take her up to bed Not only is Hadra's husband another example of a parent who appeared not to have a strong personal relationship with reading, but had developed a strong shared reading relationship with his child, but this again illustrates how reading, within a shared reading context, includes time for talking and for parent and child to enjoy one another's company. This strongly supports Heath's (1982) assertion that the 'prevalent dichotomy' between what has become known as 'oral and literate traditions' is inadequate. Certainly this study has shown that any attempt to separate 'oral' and 'literate' practice within the context of shared reading activity is unhelpful. Together this shows how shared reading is a very different experience from these parents' own interactions with reading, especially from when they were at school themselves. These parents described a range of proficiency judgements that were attached to reading, which included judgements from teachers and peers on their ability to decode complex text with accuracy and read aloud with fluency. They also judged themselves against other people who enjoyed reading ('I'm the only one on this Earth who doesn't like reading') claiming that they wished they were 'like that' but they just couldn't get 'into it'.
Parents' reading with their children was not only a very different experience from their own schooled reading, but actually seemed to carry a different definition of what 'reading' actually is. These parents were describing an activity that was not forced, that did not carry proficiency judgement, where enjoyment was prioritised and, importantly, included factors such as talking to their child and enjoying spending time together. This may very well explain why these parents succeeded in developing a positive reading relationship with their children despite the fact that they had a poor relationship with reading themselves. This is a very positive finding, however, the data also suggested that in some cases, the parents not only developed a good reading relationship with their children, but that their own relationships with reading improved as a result of this. This is discussed next.

Shared Reading Supported Parents' Reading
It was clear from the data presented in the previous section that for almost all these parents, shared reading was not only something that they did regularly with their children, but that they enjoyed this reading. We have already argued that reading within a shared reading context was very different to the reading that these parents had previously experienced, however the data also showed that parents became more engaged with certain aspects of reading as a result of their shared reading interactions. For example, Latika, who had spoken about finding all books 'boring' while at school and as an adult, told us that she is now getting to know lots of stories. She reported: 'And with her, I'm learning now, there's so many things, there are so many stories I've read now, which I have never read in my life before.' Latika is making the point that she is an active participant in the shared reading relationship and is 'learning' alongside her child. She then goes on to make the point that it is because they read 'every day' that she herself has come to read 'so many stories', which she sees as a benefit to herself as well as her daughter. Similarly Fiona also spoke of the fact that it was the routine of everyday life with young children that encouraged her to read. Having initially reported that when she was in her twenties, it 'didn't enter my spectrum to think about reading books', Fiona spoke about the fact that having children meant she was 'at home a lot more' and this encouraged her 'to read a bit more', both with her children and for herself.
Finally Natalie who had found reading in school particularly stressful, talked in detail about the ways in which her own relationship with books had developed as a consequence of her shared reading activity. She stated: I think I enjoy reading more now with the kids because I'm excited to do different things with them, thinking of things I could do with the book, things I could get them doing while I'm reading the book. Which is good because I wasn't really a book reader.... And I'm learning about new authors as well, like it wasn't something that I was that interested in so it's nice, now I look at the stories and I'm looking for more books that that author's wrote that I think he might enjoy. I never used to do that -just got books that looked good on a cover There are so many points being raised in this statement that illustrate the complex ways in which Natalie's relationship with reading has grown from sharing books with her children. Having been someone who 'wasn't really a book reader', Natalie is now using the kind of strategies that experienced readers use to select books, such as 'learning about new authors' and finding more books that a particular author has written. In addition Natalie speaks of being 'excited' by the different things that she can do with her children with the book, suggesting that she now has a very different relationship with books compared to when she was younger. Interestingly, when asked directly about her own confidence in reading now, Natalie was in no doubt that reading with her children had improved her own confidence in reading. She reported: 'I feel okay. I've picked it up more as I've gone along. I'm more confident now with the kids than I was when I was younger. I think I'm good at reading now, I hope so.'

Conclusion: Implications for Intervention
This paper has reported the encouraging, yet somewhat unexpected finding, that the parents in this study who had identified themselves as having a poor personal relationship with reading, went on to develop positive shared reading relationships with their children. While this is a welcome discovery, it would be very naïve to assume that this is the case for all parents who have struggled with reading themselves at school or who do not enjoy reading for themselves. This is particularly important given that the literature suggests that vast numbers of young people leave school each year with a poor identity of themselves as a reader (Alvermann, 2001), and/or have been identified as 'unsuccessful' in literacy-based assessment (Greenleaf & Hinchman, 2009). However, by taking time to carefully understand the nature of these shared reading relationships, this study has significant implications for interventions with families, especially where the parents have not enjoyed a positive relationship with reading themselves.
While there is a substantial amount of literature on reading in the home that has explored what parents do, and why it is beneficial (Clark and Hawkins, 2010;Gjems, 2010;Mullan, 2010), very little research has attempted to understand the nature of these shared reading relationships. Yet this study has shown that these practices are highly individual and unique to the social and cultural context of the family. Moreover, these participants reported that their reading activities were often led by their own children, rather than by the parents themselves. This has very important implications for reading interventions in general. Having established that many reading interventions are not successful (Justice, Logan and Damschroder, 2015;Justice, Skibbe, McGinty, Piasta, & Petrill, 2011), this study suggests that this may be because reading interventions rarely acknowledge the unique nature of family reading practices, and the factors that motivate and discourage parents from reading with their children. This paper has begun to address this issue by ensuring that families, and not the intervention, are at the starting point for this discussion.
Data from this study showed that the construct of 'reading', within their shared reading relationships with their children, was very different from the ways in which reading was defined in school for these participants. For many of these parents their own reading carried a set of proficiency judgements, determined by the educational discourses and societal expectation, however this was not the case with their shared reading activity with their children. This was not 'forced' and did not carry proficiency judgement. Reading was seen as a very flexible construct which included talking and telling stories. Above all parents seemed to value the space it provided for them to enjoy protected time with their children and for all parties to enjoy the activity. This raises two particular implications for intervention. Firstly, this suggests a need to encourage parents, and particularly those who may have had a poor personal relationship with reading, to embed shared reading into their everyday lives in ways that suit them and their child. The emphasis should be on enjoying time with their child and ensuring that the activity is enjoyable for themselves and their child. While it may be helpful for some parents to receive recommendations for books, parents should be encouraged to find texts (and these could include digital texts as well as paper) that they personally enjoy interacting with. It is important that parents understand that shared reading does not need to be about rigorously decoding print, or sticking faithfully to the sequence of printed text, but can be an opportunity to talk about, or around the text, sing songs, talk about their days and just enjoy spending time together.
This leads to a second major implication. This study has shown that these parents were very comfortable in developing their own shared reading relationships with their children, where 'reading' carried a broad and inclusive definition that was free from the regulation that had governed their own experiences with reading. However, what we do not know is whether these shared reading relationships were maintained once the children started school. Previous research has shown that children can lose confidence in themselves as readers, from their earliest years in school, due to the dominance of a schooled discourse in reading (Levy, 2008). This same research showed that parents were also concerned about getting reading 'right' once their children started school, given the emphasis on decoding print and mastering the reading of reading scheme books (Levy, 2009). This suggests an urgent need to talk to parents about the value of their own reading interactions with their children. In particular it is important that parents have the confidence to keep on prioritising enjoyment within their shared reading relationships with their children and know that they are doing the 'right' thing for their children by maintaining these reading activities in ways that suit their own families. We already know that the maintenance of shared reading practice is beneficial for children (Snow 1994;Bus et al, 1995;Cline and Edwards, 2013) however this study has shown that such activity can also help some parents to improve their own personal relationship with reading. 146 Levy,