Study looked at impact of text messaging on 9and 10-year-old children's literacy skills. Text messaging does not adversely affect the development of literacy skills - especially spelling. Children's use of mobile phone technology is increasing year on year.
Study looked at impact of text messaging on 9and 10-year-old children's literacy skills. Text messaging does not adversely affect the development of literacy skills - especially spelling. Children's use of mobile phone technology is increasing year on year.
Study looked at impact of text messaging on 9and 10-year-old children's literacy skills. Text messaging does not adversely affect the development of literacy skills - especially spelling. Children's use of mobile phone technology is increasing year on year.
Study looked at impact of text messaging on 9and 10-year-old children's literacy skills. Text messaging does not adversely affect the development of literacy skills - especially spelling. Children's use of mobile phone technology is increasing year on year.
phonological processing skillsjcal_398 28..36 C. Wood, E. Jackson, L. Hart, B. Plester & L. Wilde Psychology Department, Coventry University, Coventry, UK Abstract This paper reports on an intervention study that considered the impact of text messaging on 9- to 10-year-old childrens literacy skills. One hundred and fourteen children who had never owned a mobile phone before were recruited and randomly allocated to either the intervention or control conditions. All children were pre- and post-tested on a range of reading, spelling and phonological awareness measures. Children in the intervention group were given access to a mobile phone (enabled for text messaging only) for weekends and during half-term break for a 10-week period. It was found that there were no signicant differences between the two groups of children in terms of their literacy attainment during that period. However, within the mobile phone group, there was evidence that use of text abbreviations was positively related to gains in literacy skills. Moreover, after controlling for individual differences in IQ, and the childrens performance at pre-test, textism usage was able to account for a signicant amount of variance in post-test spelling scores. These results show that text messaging does not adversely affect the development of literacy skills within this age group, and that the chil- drens use of textisms when text messaging is positively related to improvement in literacy skills, especially spelling. Keywords literacy, mobile phones, phonological awareness, SMS, spelling, text messaging. Childrens use of mobile phone technology is increas- ing year on year. Not only are mobile phones considered to be must have technology by children, concerns about child welfare are also leading parents to give mobile phones to children at increasingly younger ages: a recent study found that some children in the United Kingdom were receiving their rst phones at the age of 5 years (Plester et al. 2009). The majority of children aged between 8 and 15 years in the United Kingdomand United States own mobile phones, and text messaging (SMS) is a popular function of the phones among this age group (LSE 2008; Ofcom 2008). Despite its popularity among young people, or perhaps because of it, there has been widespread concern in the media about the impact that text messag- ing may have on childrens literacy development. Such concerns have particularly focused on childrens use of text message abbreviations or textisms, such as CU L8R or anuva ng, when communicating with these devices (e.g. Thurlow 2006). However, recent studies have shown positive relationships between the degree of use of such spellings and childrens performance on standardized tests of reading and spelling. For example, Plester et al. (2008) found that there was a signicant positive correlation between the proportion of text Accepted: 28 October 2010 Correspondence: Clare Wood, Psychology Department, Coventry University, Priory Street, Coventry CV1 5FB, UK. Email: c.wood@ coventry.ac.uk All authors were based in the Psychology Department, Coventry University. Emma Jackson is now based in the Psychology Department, University of Worcester. Lucy Wilde is now based in the School of Psychology, University of Birmingham. doi: 10.1111/j.1365-2729.2010.00398.x Special section 28 2011 Blackwell Publishing Ltd Journal of Computer Assisted Learning (2011), 27, 2836 abbreviations used by 10- and 11-year-old children (in a standard English-to-text message translation exercise and spelling ability). In fact, use of the two most com- monly used types of textism were able to account for 32.9% of the variance in the childrens spelling scores. At this point, it seemed likely that this positive relation- ship could be explained by individual differences in phonological awareness in the children, as the textism types commonly used were phonologically based (i.e. they tended to be alternative phonetic spellings of words). So in a subsequent study, which was designed to look at both reading and spelling, Plester et al. (2009) found that phonological awareness did account for much of the concurrent relationship between literacy skills and textismuse in 10- to 12-year-old children (this time using an scenario type task that asked the children to imagine that they were in a particular situation, and to write the text message that they would send). However, this study also found that even after controlling for age, short-term memory, vocabulary, phonological aware- ness and how long they had owned a mobile phone, textismuse was still able to account for signicant addi- tional variance in reading ability. It is not immediately clear what the nature of the additional contribution of textism use to literacy development might be. It could be that the contribution is a motivational one, as textism creation and use is something that is playful and enjoy- able. Alternatively, it could be that the extra contribution is simply the contribution of the additional exposure to print that children who text message are likely to expe- rience as a result of daily practice at sending and reading text messages. The results of such studies are promising as they indi- cate the potential benets that such technology may have for childrens literacy development, given their widespread use by increasingly younger children. However, these previous studies did have some limita- tions. For example, for practical reasons, they relied on text messages elicited during contrived tasks rather than on text messages actually sent by the children during their leisure time. Also, the data in these two studies were concurrent, and therefore no direction of causality may be inferred from the associations reported between textismuse and literacy skills. As much as we may wish to infer that textism use is contributing positively to lit- eracy skills, it seems equally likely that literacy skills may contribute to textism use. It is therefore essential that the direction of causality is established. In order to address the issue of causal inuences, and to overcome the limitations of previous work in this area, this study evaluated the impact that text messaging has on UK childrens literacy skills by giving children who had never owned a phone before the chance to use one, for text messaging only, at weekends and during half-term break, for one academic term. The childrens phonological skills, reading and spelling were assessed at pre- and post-test, and compared with a control group who participated in identical assessment activities and levels of contact with the researchers, but who did not have access to a phone during the intervention period. It was anticipated that the children with access to the mobile phones would show signicantly greater increases in literacy performance relative to the control group, after individual differences in IQ had been con- trolled. It was further predicted that use of textisms by the children in the mobile phone group would be posi- tively related to gains in literacy performance during the course of the study. A unique feature of this present study is that, as a result of its design, we were able to track the volume of text messages that the children sent and received each week during the intervention period. This also enabled us to consider the contribution that these usage data may be able to make to understanding how text messaging might benet the childrens literacy development. Method Participants One hundred and fourteen children aged between 9 and 10 years [mean age 9; 10, standard deviation (sd) = 5.6 months] participated in the study. They were recruited from 12 schools in the Midlands region of the United Kingdom. All children within the specied age range who attended these schools, and who did not already own a mobile phone, were invited to participate in the study. Once written parental consent was obtained, half the children within each class were randomly allocated to the mobile phone group (intervention group) and the other children were allocated to the control group. Where an uneven number of children were recruited within a class, the extra child was allocated to the control condition. This allocation procedure resulted in 56 children in the mobile phone group, and 58 children in the control group. Effect of text messaging on literacy skills 29 2011 Blackwell Publishing Ltd Test battery The Wechsler Abbreviated Scale of Intelligence (Wech- sler 1999) was used in its short formto provide a simple measure of the childrens IQ. The vocabulary and matrix reasoning sub-tests were administered and scored according to the standardized instructions. The standardized IQ score for each child was used in the analysis. The British Ability Scales II Word Reading Subtest (Elliot et al. 1996) was used to measure the chil- drens reading ability. This task required the children to read from an A4-sized card that showed 90 words that become progressively more difcult. No corrective feedback was given to the children during the study and the childrens raw scores were converted to ability scores prior to analysis. The British Ability Scales II Spelling Subtest (Elliot et al. 1996) was administered individually to the children according to the standard- ized instructions, which indicated specic start and stop points for each child depending on their age and ability. No corrective feedback was provided at any point during the study and, as with the reading scores, the childrens raw scores were converted to ability scores prior to analysis. Specic sub-tests from the Phonological Assessment Battery (PhAB; Frederickson et al. 1997) were used to assess different aspects of childrens phonological pro- cessing skills. A broad range of measures was used so that a comprehensive assessment of phonological skills could be provided and examined in relation to textism use, as past research has shown that phonological awareness seems to be linked to textism use (Plester et al. 2009). Non-word reading was administered to provide a measure of decoding ability (which contrasts with reading real words, which may be read by a sight word approach). The maximum score possible on this test was 20. Rhyme detection was used to provide an assessment of broad phonological awareness and required the children to say which two words out of a set of three sounded the same at the end (maximum score = 21). The spoonerisms sub-test was used as a more specic measure of phonemic awareness. In this test, the children were asked to substitute the onsets of specic words for either specied phonemes or to swap the onsets of two spoken words to produce a true spoo- nerism (maximum score possible = 30). The rapid picture naming test was used here as a measure of rapid phonological retrieval, and rapid naming is a measure that has been linked to reading disability (e.g. Denckla & Rudel 1976; Bowers & Wolf 1993). The children were required to name the items presented in a grid of 50 as quickly as they could. This was administered twice and the total time (in seconds) taken to complete the task was noted. The uency measures from the PhAB were also included as measures of lexical retrieval as they assessed how rapidly the children could access and produce object names from memory. There were three kinds of uency test. The rst was a measure of allitera- tion uency, which involved the children being timed for 30 s during which they were asked to name as many words as they could that started with specied sounds. During the rhyme uency test, the children were asked to say as many words as they could think in 30 s of that rhymed with specied words. In the semantic uency test, the children were given 30 s to name as many words as they could that related to certain topics. In each version, there was one practice trial and two assessed trials. The scores obtained represented the total number of words produced in each sub-test. General procedure Following ethical approval for the study from the uni- versitys Ethics Committee, and permission to conduct the study from the head teachers of the schools con- cerned, letters and consent forms were sent to the parents or guardians of all the children who were aged 910 years at the school who did not have a mobile phone. The children were then individually briefed and asked if they would like to participate in the study; we made it clear to all children that there was the chance that they would not be selected to receive one of the mobile phones if they did choose to participate. No chil- dren withdrewfromthe study as a result of this brieng. Within each class of children, the eligible children were randomly allocated to either the mobile phone group or the control group. All children completed the full test battery at the beginning of the study (pre-test phase). Once the pre-testing was complete, the children in the phone group received a brieng on the Friday afternoon about how to use their new phone. The mobile phones were all functionally simple Nokia 1112 (Nokia, Finland) models. Basic handsets were selected for this study to minimize the desirability of the phones to other children who were not participating in the intervention, and also to make them easy for the children to learn to 30 C. Wood et al. 2011 Blackwell Publishing Ltd use. The children were told howto send and receive text messages, and practised this in front of the research assistant by sending test messages to each other. They were told they were to have the phones for the next 10 weekends and during the week-long half-term break to text their friends. As the children were novice texters, they were provided with the phone numbers of other children taking part in the project at the same school, so that they had someone to communicate with. However, during the study, it became apparent that most children (although not all) had a good network of friends who already owned phones with whom they could text, and parents were also noted as recipients of text messages. Every phone was given to the children fully charged and with texting credits pre-loaded onto them. The phones were given to the children on Friday afternoons at the end of the school day to reduce any unnecessary disruption to the school. The phones were handed back at school early on Monday morning. These were col- lected by the research team, and the text messages that were sent by the children were transcribed by hand exactly as they were written on the phones. The research assistants also copied the number of messages sent and received each week from the phones call logs. The call log was then reset, the phone charged, and new text credit put on the phone, ready for the next weekend. During the half-term break, the children were given additional credit and were given the charger for their phone. All children in the study (control and mobile phone groups) were assessed on their reading and spelling only each week. This was done to enable the research team to monitor the childrens progress and to see if there were any signs that the phones might be adversely affecting the childrens literacy development. Had this been found to be the case, the study would have been terminated for ethical reasons, but there was no sign of the mobile phone group showing declining levels of lit- eracy during the intervention. By testing both groups of children, we were able to ensure that both groups had a broadly similar level of contact with the research team. It should be noted that the same reading and spelling assessments were used throughout the study, which could have resulted in practice effects on these mea- sures. However, as no feedback was given during the tasks, there was little apparent evidence of substantial practice effects, and these effects would have been present in both the intervention and control groups, as both groups were tested on a weekly basis during the study. After the 10-week period had elapsed, all the assess- ments, with the exception of the IQ test, were re-administered to the children in the study (post-test phase). Once the post-testing was complete, all the chil- dren were thanked for their participation in the study. The children in the control group were also given access to the phones for a brief period, so that they could also experience using them. Results Table 1 shows the childrens performance at pre- and post-test on the literacy measures assessed in the study. It can be seen that the two groups of children were com- parable in terms of their pre-test performance on the measures, and there is also little difference between the two groups in terms of their post-test improvement. This was borne out by the results of analysis of covariance in which IQ and pre-test performance on the measures were entered as covariates before comparing the two groups on their post-test performance on each measure. The results of these analyses are summarized in Table 1, which shows that there were no signicant differences between the groups on any of the literacy measures taken. As mentioned earlier, in this study, we tracked the number of text messages that the mobile phone children sent and received during the intervention, and consid- ered whether there was any inuence of these variables on the mobile phone groups literacy development during the period of intervention. The reason for this was because these measures may be seen as a proxy for exposure to print in the context of mobile phone use, and this is something that we would expect to benet the childrens literacy development. Table 2 provides summary statistics for the mobile phone groups phone usage data over the course of the study. It can be seen that there was very enthusiastic use of the phones at the beginning of the study, with participants sending an average of almost 45 messages in that week, but this dropped steadily over the course of the study to just under six by the nal week of use. The degree of varia- tion in the numbers of messages sent also reduces steadily over time, and a similar pattern is observed for the number of messages that the children received during the study. With respect to their use of textisms, Effect of text messaging on literacy skills 31 2011 Blackwell Publishing Ltd we can see that the textism ratio is roughly the same at the beginning and end points of the study, being some- what lower than the overall average for the study as a whole. Table 3 summarizes the correlations between the degree of improvement observed in the mobile phone groups literacy skills (using composite measures con- structed from the childrens pre- and post-test scores), and the number of messages the children sent and received in the rst, middle and last weeks of the inter- vention period. The reason for looking at these time points was because, as shown in Table 2, the children initially experienced a hallelujah effect in which they sent very high numbers of text messages to each other in their initial enthusiasm with the technology. By the end of the study, this had abated and text messaging had become a mundane activity with much lower usage levels, to the extent that a few children were sending no messages at all at that point in the study. By considering the relationships between literacy and sending and receiving messages as these time points, we can con- sider whether any effects observed might be consistent, or are linked to abnormally high levels of text messag- ing. The composite outcome measures were constructed Table 1. Descriptive statistics on measures by group (standard deviations in parentheses). Assessment Control group Mobile phone group F (P) Reading Pre 142.1 (24.9) 143.9 (23.2) Post 153.0 (25.4) 156.4 (24.3) 0.775 (0.381) Spelling Pre 109.8 (19.8) 109.8 (22.7) Post 116.2 (19.2) 117.8 (20.9) 0.143 (0.706) Rhyme detection Pre 17.9 (3.7) 17.5 (4.2) Post 18.2 (4.0) 18.2 (3.7) 0.480 (0.490) Spoonerisms Pre 20.4 (7.0) 20.4 (6.5) Post 21.4 (5.7) 22.5 (5.4) 1.381 (0.243) Non-word reading Pre 16.1 (4.2) 16.0 (4.1) Post 16.5 (2.9) 17.2 (3.3) 2.105 (0.150) Rapid picture naming Pre 97.6 (20.3) 96.5 (20.4) Post 94.6 (21.4) 94.3 (22.2) 0.008 (0.927) Alliteration uency Pre 11.3 (3.8) 11.4 (4.4) Post 11.5 (4.2) 11.7 (4.0) 0.002 (0.883) Rhyme uency Pre 11.8 (4.5) 9.6 (3.2) Post 11.8 (4.1) 10.9 (4.6) 0.760 (0.385) Semantic uency Pre 22.2 (5.2) 20.6 (5.03) Post 20.2 (4.5) 19.4 (4.4) 0.349 (0.556) Table 2. Descriptive statistics for the mobile phone groups use of the mobile phones (standard deviations in parentheses). Week 1 mean Week 5 mean Week 10 mean Overall mean No. of messages sent 44.9 (38.4) 14.5 (19.9) 5.9 (13.8) 19.5 (16.7) No. of messages received 51.4 (41.2) 16.1 (25.4) 5.2 (5.2) 19.7 (17.0) Textism ratio 0.129 (0.121) 0.157 (0.174) 0.120 (0.183) 0.156 (0.133) 32 C. Wood et al. 2011 Blackwell Publishing Ltd as follows: the childrens pre-test scores were sub- tracted from their post-test scores to give a difference score. These difference scores were converted to z-scores so that performance on each test was equally weighted when the scores from the various tests were summed. The improvement in reading and spelling measure comprised the sum of the reading and spelling difference z-scores. The improvement in phonological awareness measure comprised the sum of the rhyme and spoonerism difference z-scores. Finally, the improvement in uency measures score comprised the sum of the alliteration, rhyme and semantic uency dif- ference z-scores. Table 3 shows that there were no signicant relation- ships between literacy development during the interven- tion and number of messages sent and received at the beginning of the study when the abnormally high levels of textism use were evidence, or at the midpoint after 5 weeks. There was some evidence of a signicant asso- ciation between the number of messages sent at the end of the study and improvement in phonological aware- ness. Interestingly, both the number of messages sent and received at the end of the study were linked to improvement on the uency subtests of the PhAB. Next, we looked at the associations between mean textism use and literacy performance at pre- and post- test, and we also looked at mean number of messages sent and received over the course of the intervention (see Table 4). It should be noted that the mean textism ratio observed among this group of novice phone users is much lower than has been observed in other samples; 0.156 (sd = 0.134) compared with 0.34 reported in Plester et al. (2009). This is in line with what we might expect from a group of children relatively inexperi- enced in this medium, and data from a cross-sectional study of childrens text messaging shows that it does usually increase with age/experience (Wood et al. 2009). Mean textism use during the study was signi- cantly associated with most of the literacy skills at pre- and post-test, which is consistent with other studies of literacy and textism use (e.g. Plester et al. 2009). This pattern also contrasts strongly with the data on the average number of messages sent and received during the study, where only uency measures were signi- cantly associated with the average number of messages sent and received. Finally, we considered whether mean textism use might be able to predict literacy improvement longitudi- nally after controlling for individual differences in IQ and pre-test performance on the given measures. It was found that textism use could predict a signicant amount of the variance in spelling development during the intervention period, R 2 change = 0.086, F = 10.488, P = 0.002, b = 0.307. This relationship remained even after also controlling for the mean number of text mes- sages sent and received during the intervention, R 2 change = 0.083, F = 10.218, P = 0.002, b = 0.330. None of the other literacy measures were predicted by mean textism use after controlling for IQ and auto- regressors. Discussion The main nding from this study was that the children who were given access to mobile phones for the purpose of text messaging did not perform differently from the children who were not given mobile phones in terms of their literacy development. This suggests that although the children with the phones did not benet signicantly from access to the technology, their literacy develop- ment was also not adversely affected. The lack of sig- nicant positive benets is striking given the previous literature on this subject, which has shown signicant concurrent associations between textism use and Table 3. Spearman correlation coefcients showing the strength of association between literacy improvement and numbers of text mes- sages sent and received at the beginning and end of the study. Week 1 sent Week 1 received Week 5 sent Week 5 received Week 10 sent Week 10 received Improvement in reading and spelling 0.242 0.171 0.149 0.096 0.132 0.074 Improvement in phonological awareness 0.057 0.059 -0.076 -0.060 0.324* 0.142 Improvement on uency measures 0.158 0.225 -0.009 -0.053 0.408** 0.390** Improvement in rapid naming -0.219 -0.154 0.157 0.100 -0.243 -0.145 *P < 0.05; **P < 0.01. Effect of text messaging on literacy skills 33 2011 Blackwell Publishing Ltd literacy development (e.g. Plester et al. 2008, 2009). However, it seems likely that the duration of the present intervention was not long enough for the benets of text messaging to become apparent. The interim results of a longitudinal study that studied the development of lit- eracy over the course of an academic year do suggest that textism use impacts causally on spelling develop- ment (Wood et al. 2009). Also, it was noted that levels of textism use in this study were much lower than those previously reported in other studies because of the chil- drens relative inexperience with the technology. It seems likely that if a longer period of intervention was undertaken, that would enable the children to become more prolic users of text abbreviations and there may have been stronger evidence of impact. Similarly, in order to get support from schools, it was necessary to restrict the childrens access to the devices: the children were only allowed to use the phones during weekends and the half-term break, which was just 1 week long. This does not reect childrens usual pattern of access and use when they own their own mobile phones, and is likely to have restricted the impact of the technology on the childrens learning. It is therefore important to note that this study underscores the message that it is not having a mobile phone per se that is benecial, but rather the use of textisms when text messaging that is linked to benets in literacy development. This suggestion is borne out by the data we obtained fromthe children in the mobile phone group. That is, for this group, despite more limited levels of textism use, there was still evidence of a signicant contribution of textism use to the childrens spelling development during the study. This nding is signicant as not only were individual differences in IQ controlled in this analysis, but pre-test performance on the spelling measure was also a covariate; the period of intervention was only10weeks induration, andsothe degree of asso- ciation between the spelling scores taken at pre- and post-test was high, which makes this result noteworthy. So there was some evidence that the childrens text mes- saging behaviour had the potential to impact signi- cantly on their literacy skills, but relative to the control group these advantages were not sufciently marked in the present study. Similarly, the correlational data showed that mean textism use was signicantly related to literacy outcomes within the mobile phone group. Table 4. Spearman correlation coefcients to examine the relative associations between textism usage, number of messages sent and received, and performance on the literacy measures. Mean textism usage Mean no. of messages sent Mean no. of messages received Pre-test variables Reading 0.307* 0.086 0.132 Spelling 0.281* -0.060 0.006 Rhyme 0.295* 0.013 0.061 Spoonerisms 0.438** 0.033 0.095 Non-word reading 0.371** 0.074 0.124 Rapid naming -0.398** -0.156 -0.112 Alliteration uency 0.292* 0.104 0.141 Rhyme uency 0.324* 0.252 0.224 Semantic uency 0.095 0.288* 0.268* Post-test variables Reading 0.227 0.109 0.141 Spelling 0.390** 0.017 0.119 Rhyme 0.200 0.011 0.093 Spoonerisms 0.401** 0.003 0.033 Non-word reading 0.281* 0.007 0.134 Rapid naming -0.438** -0.096 -0.052 Alliteration uency 0.144 0.125 0.099 Rhyme uency 0.324* 0.362** 0.368** Semantic uency 0.244 0.131 0.063 *P < 0.05; **P < 0.01. 34 C. Wood et al. 2011 Blackwell Publishing Ltd These results are in line with those of past work in the area (e.g. Plester et al. 2008, 2009; Wood et al. 2009). The reason for this association between spelling and textism use is partly explained by the highly pho- netic nature of the textisms that are popular within this age group, as the phonological and alphabetic aware- ness that is required for the construction and decoding of these textisms also underpin successful reading development (e.g. Adams 1990). However, it is also possible that textism use adds value because of the indirect way in which mobile phone use may be increasing childrens exposure to print outside of school. One way of assessing this idea that print exposure might be contributing to literacy skills was to examine the number of text messages sent and received by the children during the intervention period, as these data are an appropriate proxy for phone-based print expo- sure. This study is the rst to collect and analyse such data in relation to educational outcomes of children. The correlations in Tables 3 and 4 show that the only literacy variables that were related to number of mes- sages sent and received were the uency sub-tests of the PhAB. It will be recalled that these measures were included as measures of lexical retrieval. It would therefore seem appropriate to nd that improvements in lexical retrieval are linked to mobile phone behav- iours that involve reading and composing text mes- sages. It would seem that such behaviours enhance childrens word-nding skills. This suggests that while the exposure to print explanation of the relationship between texting and literacy is not supported, exposure to print through mobile phones does impact on other language skills that relate to the representation of lexical items in memory. Further research into lexical processes and text messaging is required to understand the exact nature of this relationship. The relative lack of use of the mobile phones at the end of the study was also worth commenting on when considering the lack of a signicant difference in out- comes between the groups in this study. That is, the children were quick to explore and personalize their new phones as far as the basic models that we provided them with would allow. The children were provided with very basic phones because we were interested in their use of the text messaging function and we did not want this effect to be affected by other aspects of the phones functionality. However, the limited functional- ity of the handsets we provided did result in some lack of engagement with them over time. In summary, this study has shown that allowing chil- dren access to mobile phones for text messaging over a 10-week period does not signicantly advantage or dis- advantage the children. However, textism use during texting was linked to spelling development and the number of messages sent and received was linked to lexical retrieval skills. Acknowledgements This work was supported by a research grant from Becta. The authors also gratefully acknowledge the support of the children, parents and schools who col- laborated on this project. We would also like to thank Nicola Ayres, Kathryn Emm, Joanne Hicks, Nikki Holiday, Melissa McBean, Neelam Nagra, Becky Nicholls, Katie Sketchley-Kaye, Luisa Tarczynski- Bowles, Helen Veater and Carly Woods for their assis- tance with data collection and data transcription. References Adams M.J. (1990) Beginning to Read: Thinking and Learn- ing about Print. MIT Press, Cambridge, MA. Bowers P.G. & Wolf M. (1993) Theoretical links among naming speed, precise timing mechanisms and ortho- graphic skill in dyslexia. Reading and Writing 5, 6985. Denckla M.B. & Rudel R.G. (1976) Rapid automatized naming (R.A.N.): dyslexia differentiated from other learn- ing disabilities. Neuropsychologia 14, 471479. Elliot C.D., Smith P. & McCulloch K. (1996) British Ability Scales: Second Edition (BAS II). NFER Nelson, Windsor, UK. Frederickson N., Frith U. & Reason R. (1997) Phonological Assessment Battery. NFER Nelson, London, UK. LSE (2008) Mobile Life Report. Available at: http://www. mobilelife2007.co.uk/Mobile_Life_2008.pdf (last acces- sed 20 January 2009). Ofcom (2008) Media Literacy Audit: Report on UK Chil- drens Media Literacy. Available at: http://www.ofcom. org.uk/advice/media_literacy/medlitpub/medlitpubrss/ ml_childrens08/ (last accessed 20 January 2009). Plester B., Wood C. & Bell V. (2008) Txt msg n school lit- eracy: does texting and knowledge of text abbreviations adversely affect childrens literacy attainment? Literacy 42, 137144. doi: 10.1111/j.1741-4369.2008.00489.x. Plester B., Wood C. & Joshi P. (2009) Exploring the relation- ship between childrens knowledge of text message abbre- Effect of text messaging on literacy skills 35 2011 Blackwell Publishing Ltd viations and school literacy outcomes. British Journal of Developmental Psychology 27, 145161. doi: 10.1348/ 026151008X320507. Thurlow C. (2006) From statistical panic to moral panic: the metadiscursive construction and popular exaggeration of new media language in the print media. Journal of Computer-Mediated Communication 11, 667701. doi: 10.1111/j.1083-6101.2006.00031.x. Wechsler D. (1999) Abbreviated Scale of Intelligence (WASI) Manual. Psychological Corporation, San Antonio, TX. Wood C., Plester B. &Bowyer S. (2009) Liter8 lrnrs: is texting valuable or vandalism? British Academy Review14, 5254. 36 C. Wood et al. 2011 Blackwell Publishing Ltd