Aspects of the reading motivation and reading activity of Namibian primary school readers

Abstract This paper reports on the reading motivation and reading activity of 402, urban, Namibian learners in 6 schools in the central region of Namibia. From the fourth grade these Grade 7 learners received their instruction through the medium of English, and offered English as Second Language in addition to another Namibian language. They were enrolled in schools that performed above-average in the Namibian Standardised Achievement Test for English language and with reasonable access to reading resources. Employing adapted formats of instruments, developed by John T. Guthrie and colleagues, the relationships between various aspects of reading motivation, reading activity and achievement, as well as gender differences, were explored. Different from early adolescent readers in North America, the group showed moderate reading activity and high levels of motivation. Learners reported that they read slightly more often for pleasure than for academic purposes, and read fewer texts in digital than print format. Recognising the multifaceted nature of motivation, this study revealed that aspects, such as curiosity about specific topics, the importance of reading and reading for grades, were factors that highly motivated these Namibian learners. Positive relationships between motivation and reading activity, as well as between reading motivation and reading achievement, were established. Contrary to expectations, no statistically relevant correlation between academic achievement and reading activity was found. Relationships between these variables and gender, though modest, corroborate findings from previous research. These results may prove valuable in further research regarding the development of effective and inclusive classroom practices in the Namibian context.

ABOUT THE AUTHORS Emmarentia Kirchner is a senior lecturer in the Faculty of Education at the University of Namibia. Her main research interests are reading, reading motivation and literature education. As member of the Namibian Children's Book Forum, she has been active in promoting a Namibian reading culture via the publication of children's literature in all Namibian languages. She is currently enrolled at the Inland Norway University of Applied Sciences as doctoral student in teaching and teacher education.
Maria Louise Mostert is an associate professor in the Faculty of Education at the University of Namibia. Her interest in language acquisition, reading comprehension and reading motivation stems from teaching and supervising master's students in the area of Literacy and Learning.
Other key research areas are within the field of Educational Psychology, including school guidance and counselling.

PUBLIC INTEREST STATEMENT
This paper reports on the reading motivation and reading activity of 402, urban learners in 6 schools in the central region of Namibia. Employing adapted formats of three instruments, the relationships between various aspects of reading motivation, reading activity and achievement, as well as gender differences, were explored. This sampled group of pre-adolescents seemed highly motivated to read, which is different from the normal tendency that reading motivation and reading activity decline as readers get older. Even though reading resources are quite scarce in Namibia, these readers showed curiosity about different topics and regarded reading as important. Positive relationships between motivation and reading activity, as well as between reading motivation and reading achievement, were established. These findings, as well as learners' interest in reading stories and newspapers outside of school, should be used to improve reading conditions in Namibian classrooms and for all Namibian children.

Introduction
Reading remains one of the most important ways to connect with people and to make sense of the world. As societies are becoming more complex, literacy and reading are involved in many daily activities. In the journey to become "a lifelong reader who can engage with text for information, knowledge, aesthetics, and enjoyment" (Pearson, 2015, p. 8), reading competence is an essential and necessary skill (De Naeghel, Van Keer, Vansteenkiste, & Rosseel, 2012, p. 1018. The ability to read determines future career opportunities and is crucial for academic achievement (Chhabra & McCardle, 2004, p. 3). Learners cannot move through their school careers without interacting regularly with a variety of texts, such as prescribed text books, additional reading material and assignments. In addition, children can access information not only through printed text, but also via visual and audio media, and increasingly in digital format via mobile phones. Research indicates that the amount, frequency and breadth of learner reading activity affect various aspects of performance (Guthrie, Wigfield, Metsala, & Cox, 1999, pp. 232-233;Wigfield & Guthrie, 1997, p. 420). Learners who are willing and able to engage in these literacy-based activities (especially reading), has an advantage over less skilful and reluctant readers.
Attention to reader engagement and motivation, is therefore important to increase reading proficiency and achievement in school (Deci & Ryan, 1985, p. 245;Guthrie & Klauda, 2015, p. 47). Gambrell, Malloy, and Mazzoni (2007, p. 19) indicate that motivated readers "choose to read, read more and become better readers than their less motivated peers". In this respect, various researchers refer to the "Matthew effect" regarding reading, the ever-widening gap between proficient, skilful and less competent readers, as well as the importance of motivation to curb this outcome (Bates, D'Agostino, Gambrell, & Xu, 2016;Chang, Wang, & Ma, 2015;Malloy & Gambrell, 2010). How to motivate learners from diverse backgrounds and cultures is central in addressing these disparities. The fact that, from 1990-2015, percentages of "illiterate" youth in sub-Saharan Africa have increased to almost half of the world total is a cause for concern, given the role that various forms of literacy and reading can play in accessing education and creating more just and equal societies (Unesco, 2017, p.8). Research on reading activity and reading motivation, in these contexts is important. The impact of mobile technology and the cost of connectivity in African countries might also have marked influences on literacy in these contexts.
Regarding reading motivation and engagement and its relations to reading activity and reading achievement, most research has been conducted in Canada and the USA, and relatively little information on learners in African countries is available (Mucherah & Herendeen, 2013;Schiefele, Schaffner, Möller, & Wigfield, 2012). Research on the motivation of English Language Learners seems to focus more on motivation or investment in learning English as second or foreign language, and not on reading motivation per sec (Mori, 2002, p. 92). Research focusing on African contexts is, by comparison, relatively meagre, but does cover a wide range of topics ranging from creating reading cultures in schools and communities, researching reading attitudes amongst tertiary students and teachers, as well as, important for our study, the effect of social interaction on adolescent reading motivation (Bitz & Emejulu, 2016;Lukhele, 2013;Mucherah & Herendeen, 2013;Okebukola, Owolabi, & Onafowokan, 2013;Pretorius & Mampuru, 2007). In Namibia, no related studies have been conducted to date. In their review on research regarding reading motivation, Schiefele et al. (2012) did

The Namibian context
The relatively small but culturally diverse Namibian population of 2.1 million speak around 13 mother tongues (Namibian Statistics Agency, 2011). After Independence in 1990, English was declared the only official language, and, apart from the first four years of education, is also the medium of instruction in schools. According to the language policy for schools, the various Namibian Languages are expected to be used as instruction media in the first three years of schooling, with Grade Four regarded as a transition year. The Namibian languages and (mostly) English Second Language are offered as subjects throughout the school year until grade 12. Especially in urban areas, demographic diversity results in some learners having to opt for another local language in the place of their mother tongue as medium of instruction as well as school subject (Ministry of Education & Culture, Namibia, 2015, p. 30). The implementation of the language policy and the inadequate role or attention to mother tongue education has been seen as problematic in Namibia (see Harris, 2011;Tötemeyer, 2010).
While the value of learning and reading in the home language cannot be contested (see Brock-Utne, 2001;Cleghorn & Rollnick, 2002) it is also true that, despite efforts by government and the private sector, the provision of Namibian reading materials in all languages leaves much to be desired. The provision of reading material in all African languages, albeit desired, is fraught with challenges. In fact, in Namibia the provision of trade books has declined at an alarming rate over the past years (Tötemeyer, 2013, pp. 17-18). Except for Afrikaans, reading for pleasure in the mother tongue is not really possible in Namibia, and learners often have to rely on English material published outside of Namibia. Most newspapers are published in English with small sections in other Namibian languages. Adequate English reading skills under these challenging circumstances in Namibia is therefore important.
Despite massive urbanisation, the majority of the population (57%) still lives in rural areas (Namibian Statistics Agency, 2011), where English is not widely spoken. This fact, together with the large disparity between rich and poor (Schmidt, 2009, p. 4), makes the provision of equitable and equal, quality education particularly challenging. Academic achievement in Namibia remains a concern. For example, from 2005 to 2011, the promotion rate of Grade 10 learners remained unchanged at only 55.6% (Ministry of Education, Namibia, 2012, p. 58). The lack of reading proficiency in English is possibly one of the most important factors underlying this poor outcome. According to the reports of Southern and Eastern Africa Consortium for Monitoring Educational Quality (SACMEQ) the largest African, comparative study on learner literacy in English Second Language reading and Mathematics, Namibia's performance can be seen as weak. Although there was marginal improvement between 2005 and 2010, the reading proficiency score for Namibia in 2010 was 496.9, compared to a mean score of 511.8 for all 15 participating countries. Furthermore, the report shows that only 2.5% of learners reached the highest level, Level 8, signifying critical reading (Makuwa, 2005;SACMEQ, 2010). The 2015 results of the Namibian National Achievement Test (SAT) show a decline in performance and indicate that 87% of Grade 7 learners performed in the below basic and basic achievement categories. Relatively large differences in achievement exist across the regions which can partly be attributed to demographic and socio-economic differences (Mupupa, 2016). The two top performing regions, Khomas and Erongo, have larger urban proportions living in relatively better socio-economic conditions. While, similar to other African contexts, English can be regarded as a foreign language in poor rural areas and not used much outside of school (Cleghorn & Rollnick, 2002, pp. 349-359), the situation of urban learners is different. They live in communities where English is also spoken outside of school, and where teachers are also better qualified (Ministry of Education, Namibia, 2012, p. 69).
Another challenge facing Namibian schools is the lack of reading material in homes and schools and the absence of functioning libraries in schools and the community. Like elsewhere in Africa, parents are frequently not in a position to buy books for their children (Parry, 2003, p. 743;Tötemeyer, Kirchner, & Alexander, 2015, p. 16) and few Namibian learners thus have the privilege of exposure to a wealth of children's literature. Various studies indicate that Namibian school libraries, especially in rural areas, are generally in a very poor state (Makuwa, 2005;Nengomasha, Uutoni, & Yule, 2012;Siririka, 2007;Smith, Fouche, Muirhead, & Underwood, 2008;Tötemeyer et al., 2015). Schools are often very far from public libraries or resource centres. While a minority of advantaged learners from affluent homes and in well-resourced schools have access to books, this is not the case for the majority of Namibian children. Good reading habits cannot be developed in situations where there is little or nothing to read, and comprehensive reading instruction programmes and motivational support can only be successful when a variety of reading materials on different reading levels are available.
A Namibian study, comprising 1402 learners, found that only 22.4% of the sampled group could be regarded as readers who read for pleasure. Of these readers, only 3.6% indicated a preference for non-fiction while 18.8% preferred fiction. Furthermore, for most of these children, reading seemed to be a laborious and mechanical task, not associated with pleasure at all (Tötemeyer et al., 2015, p. 23). According to this study most leisure readers came from urban and well-resourced schools, and were female. In order to shed more light on the reading motivation and reading activity of Namibian learners who are in a position to read for leisure, the present study aims to focus particularly on the reading motivation and reading activities of Grade 7 urban early adolescent readers.

Reading motivation
This study draws upon a number of theoretical frameworks of motivation. The expectancy value theory, as developed by Eccles and Wigfield (2002) postulates that the perceived value of a task and the expectancy of success in task engagement influences and directs behaviour (Malloy & Gambrell, 2010, pp. 164-165). The self-determination theory has contributed to the development of the concepts of intrinsic and extrinsic motivation, and indicates how feelings of autonomy vs. control affect motivation . An engagement perspective of reading, as developed by Guthrie and Wigfield (2000), holds that motivation, in combination with strategy use, will result in higher levels of reading achievement (Van Steensel, Van der Sande, Bramer, & Arends, n.d., p. 10). Engagement perspectives also account for the social reasons for engaging in activities, as well as stress the impact of cultural and sociocultural factors (Ellis & Coddington, 2013, pp. 232-235).
Based on the concept analysis of a vast number of articles, Conradi, Jang, and McKenna (2014) describes reading motivation as "the drive to read resulting from a comprehensive set of an individual's beliefs about, attitudes towards, and goals for reading" (p. 156). Motivation can thus be seen as an internal factor that causes or guides different types of behaviour and engagement as the observable and unobservable actions associated with reading activities (Unrau & Quirk, 2014, p. 272).
Comprehensive overviews on reading motivation and engagement (Conradi et al., 2014;Schiefele et al., 2012) indicate the complexity of motivation and its relationship to reading achievement, as well as to the amount and breadth of reading. The motivational profiles of learners differ across age groups, gender and cultures, and there are numerous individual differences in how these variables combine. Wigfield and Guthrie (1997) maintains that self-efficacy beliefs, reading goals and values, as well as social aspects of reading, are central to reading motivation, and that motivation should be viewed as multifaceted or multidimensional. Based on both qualitative and quantitative research, they argue that eleven dimensions of motivation could be identified. These dimensions have been debated and amended (Bates et al., 2016;Boerma, Mol, & Jolles, 2015), and employed in various combinations in follow-up studies (Baker & Wigfield, 1999;Guthrie et al., 1999;Schiefele et al., 2012;Wang & Guthrie, 2004). After careful consideration of the studies conducted with learners from diverse ethnic backgrounds (see 3.2.1), we opted to use all 11 dimensions in this study. Based on existing accounts of these dimensions (Baker & Wigfield, 1999;Schiefele et al., 2012;Unrau & Schlackman, 2006;Wigfield & Guthrie, 1997), they are defined as follows:

Dimensions of reading motivation
(1) Reading efficacy: the belief and confidence in one's reading ability; that one can be successful at reading; (2) Reading challenge: willingness to engage with complex reading material; the satisfaction of mastering complex ideas in text; (3) Reading curiosity: the desire to learn or read about a particular topic of interest; (4) Reading involvement: the enjoyment of experiencing different kinds of literary or informational texts; to "get lost" in a story; (5) Importance of reading: the desire to achieve important goals through reading, such as furthering one's education; (6) Reading work avoidance: the inclination to avoid reading-related activities, disliking reading; (7) Competition in reading: the desire to outperform others in reading and to reach higher levels of reading achievement than other learners; (8) Recognition for reading: the gratification or pleasure in receiving a tangible form of recognition for success in reading, such as rewards or praise for good reading performance by teachers, parents or friends; (9) Reading for grades: the desire to improve one's grades in school and to be evaluated favourably by the teacher; (10) Social reasons for reading: the process of sharing the meanings gained from reading with friends and family; (11) Compliance: reading because of an external goal, requirement or because of external pressure.

Intrinsic and extrinsic motivation
The motivation to engage in an activity, such as reading, can be integral to the activity itself; it can stem from the value attached, or the result which one wants to achieve by participating in the activity. Ryan and Deci (2000) used the following distinction as point of departure: Intrinsic motivation refers to doing something because it is "inherently interesting or enjoyable" (p. 55), while extrinsic motivation refers to doing something "because it leads to a separable outcome" (p. 55). In extrinsic motivation, the focus is thus not on the activity itself, but rather on the result, such as a reward (Gagné & Deci, 2005;Moran, Diefendorff, Kim, & Liu, 2012). A central issue in the classroom will thus be whether learning tasks are pursued because of the enjoyment of the activity and whether the motivation also stems from the possible outcomes or rewards and "educational" value attached to the activity (Conradi et al., 2014, p. 156).
Normally, the argument is that intrinsic motivation is more sustaining and that extrinsic motivation can impact negatively on intrinsic motivation. However, the impact of some forms of extrinsic reward remains a reality, and many actions are, in part, extrinsically motivated. According to the self-determination theory , reading motivation will differ, depending on the extent to which behaviour can be regarded either as autonomous and self-regulated or externally controlled.
Employing different theoretical perspectives, as well as results from research, Guthrie et al. (1999) and Wang and Guthrie (2004) relate their dimensions of motivation to these higher order categories of intrinsic and extrinsic motivation. We followed the model of Wang and Guthrie (2004) to build composite scales for measuring intrinsic and extrinsic motivation. The intrinsic motivation composite is made up of challenge, curiosity and involvement (dimensions 2-4) and the extrinsic motivation composite of recognition, grades, competition, social reasons and compliance (dimensions 7-11).

Reading and the early adolescent
The interpretation of research results regarding reading and reading motivation is dependent on age. Ideally, the early adolescent reader should have developed as a learner who is fluent in reading and engaging in reading by choice. However, most research indicate a decline in the amount of reading, as well as in the intrinsic motivation and engagement with reading and literacy activities, specifically by learners in the senior primary phase (elementary school) and onwards (Cabral-Márquez, 2015;De Naeghel et al., 2012;Edmunds & Bauserman, 2006;Guthrie et al., 2004). For the early adolescents this occurs when they are faced with a myriad other changes pertaining to their development and social worlds. The change to a new school brings a new environment, new social responsibilities and exposure to multi-literacies, resulting in changes to their world view (Ryan, 2010, p. 100). From the many leisure activities they could choose, such as television and spending time on social media, reading is not necessarily selected as a preferred activity by adults and children alike (Van der Voort, 2001, p. 113). In today's technological era, specifically older children value reading less than their younger peers. The awe and excitement with which the pre-schooler regards reading and the reading of numerous texts fast disappear as learners relate reading to schoolwork and boredom. Adolescents also have lower beliefs in their abilities to perform well in reading tasks, and are less self-efficacious. Malloy and Gambrell (2010, p. 4) refer to "decreasing positive beliefs" regarding reading abilities. Adolescent girls have been found to be more positive about reading, both academically and recreationally, than boys. This attitudinal gap seems to widen over time, especially in terms of recreational reading (Bozack, 2011;De Naeghel et al., 2014).

Population and sample
The population for this study was Grade 7, Namibian learners of the Khomas region, one of the 14 educational regions in central Namibia. These early adolescents (Ryan, 2010, p. 100) were in the final year of the senior primary phase, and should be independent readers, intrinsically motivated to read widely.
Strategic, criterion and random sampling techniques were used in this study. To strategically include learners who were in a reasonable position to read and who could report realistically on their reading behaviour and motivational levels, only schools within the region that had scored an average of 50% and above in the 2014 SATs for English Grade 7, and with access to some reading resources, were included in a list. This list was employed for the further random sampling of schools. The final sample (see Table 1) included 402 learners (mostly between 12 and 14 years old) from six urban schools, in and around the capital, Windhoek. All schools were from previously disadvantaged communities in the Khomasdal and Katutura neighbourhoods and, according to school principals, the majority of learners came from average to below-average income groups. Of the participants, 54% were female and 46% male. The learners were representative of 12 language groups. The majority spoke English as second or third language, and all of them studied English as first or second language at school from Grade 1, together with their home language or another local language. All of these learners received their instruction through the medium of English for more than three years. Similar to the Namibian demographics, participants from the Oshiwambo language group comprised the majority. The participants could be regarded as relatively good readers, with an above basic mean score of 63.6% for Reading Achievement (as measured with the SAT).

Instruments
Three instruments were administered to all Grade 7 learners from the selected schools. These were the Reading Activity Inventory (RAI), the Motivation for Reading Questionnaire and the Namibian Standardised Achievement Test.

Reading activity inventory
A self-report instrument was designed, based on the RAI developed by Guthrie, McGough, and Wigfield (1994), as a measure of the frequency and breadth of children's reading. While no traditional reliability for this measure is available (Wigfield & Guthrie, 1997), Cox and Guthrie (2001, p. 241) have reported the adequate, predictive validity of the RAI. Normally shortened versions of the instrument were used. After the pilot study, and based on the need for further studies on reading amount as formulated by Schiefele et al. (2012, p. 458) it was decided to once again broaden the scope of the instrument to include questions on breadth (genres read), frequency (how often a genre was read) and purpose (in and out of school reading: for school work, academic purposes vs. own interests or for pleasure). Items regarding the mode (reading fictional and non-fictional texts in traditional print vs. digital modes of reading, using computers and mobile phones) were also included. This added a new dimension to the original RAI.
A questionnaire of 20 items was subsequently developed. Internal consistency, when using Cronbach's alfa, was measured at 0.782. The items were structured as follows: learners were first asked whether they had read a certain type of text in the previous week, and they had to respond with yes or No. If they responded with yes, they had to complete a follow-up question requesting specific information, such as the topic about which they had read. Lastly, a second set of questions required learners to give an indication of how frequently they read the different types of texts or

Motivation for reading questionnaire (MRQ)
The MRQ, a self-report questionnaire, developed by Guthrie (1995, 1997), assesses the 11 different dimensions of reading motivation, grouped around three theoretical constructs of motivation: self-efficacy; intrinsic-extrinsic motivation and learning goals; as well as social motivation for reading (Wigfield & Guthrie, 1997, p. 422). In their study, reliabilities for the different dimensions ranged from 0.43 (Grades) to 0.81 (Competition), and factor analyses indicated evidence of construct validity for all 11 factors. Baker and Wigfield (1999) and Bozack (2011) reported similar reliability results.
According to research conducted by Conradi et al. (2014), Wigfield and Guthrie's instrument is the most frequently utilised tool for measuring reading motivation. Adapted and abbreviated versions of this instrument were employed in various other studies, including a variety of contexts and cultures (e.g. Baker & Wigfield, 1999;Bozack, 2011;Chang et al., 2015;Cox & Guthrie, 2001;Hedges & Gable, 2016;Lau, 2009;Louick, Leider, Daley, Proctor, & Gardner, 2016;McElhone, 2012;Neugebauer, 2014;Unrau & Schlackman, 2006;Wang & Guthrie, 2004). Baker and Wigfield (1999) applied it on an African-American and White student population, while Unrau and Schlackman (2006) compared the motivation levels of Asian and Hispanic six to eight graders. Chinese versions of the MRQ was developed by Wang and Guthrie (2004), Chang et al. (2015) and Lau (2009), and also used outside of the US in Taiwan and Hong Kong. Mucherah and Herendeen (2013) used the MRQ in upper level primary Kenyan schools. The version of the MRQ utilised in this study was adapted slightly for the Namibian context (see below), but is closely aligned to the one used by Wigfield and Guthrie (1997), Baker and Wigfield (1999) and Wang and Guthrie (2004). This enabled us to compare our results with those from other contexts and countries (Baker & Wigfield, 1999;Unrau & Schlackman, 2006).
During the process of adapting the MRQ for the Namibian context, literature on the compilation of questionnaires for children was consulted and various accommodations to assure validity were considered (Sireci & Faulkner-Bond, 2015, pp. 232-234). Because our urban population had a relatively good command of English, but might not have had equal reading skills in their home languages/ mother tongues, as they did not necessarily studied it as school subject at all, it was decided not to opt for the translation of the MRQ into other Namibian languages. However, accommodations which included simplifying the language and aligning items to the Namibian reality, were made. For example, because not all parents are in a position to be involved with the reading abilities and homework of their children, the item "My parents often tell me what a good job I am doing reading" was substituted with "I like it when the teacher says I read well".
The adaptation of the MRQ was executed in collaboration with reading experts in Namibia. To address issues of validity in a diverse context, the "Namibian" version of the questionnaire, together with the RAI (see 3.2.2) was pilot-tested on four Grade 7 learners with different home languages at the end of 2015. This included students rephrasing items in their own words and indicating items that they found difficult to understand. The revised and shortened MRQ finally consisted of 45 items presented in random order. The items were tested for internal consistency reliability during data gathering and a score of 0.864 (Cronbach's α) was obtained.
Similar to the original MRQ, all the items related to a specific statement (motivational stance) to which respondents could reply on a four-point Likert scale: 1 = very different from me; 2 = a little different from me; 3 = a little like me; and 4 = a lot like me. Mean scores for every dimension, as well as a mean motivation score (averaging the scores over the 11 dimensions) per participant and for the total sample, were calculated. https://doi.org/10.1080/2331186X.2017.1411036

Namibian national standardised achievement test (SAT)
The SAT was conducted to verify the reading achievement of the sampled group. The SAT is a measure that was developed by the Ministry of Education in Namibia in 2009, and implemented and utilised until 2015. This test monitors learner achievement in English Second Language for Grades 5 and 7 (Mupupa, 2016). The researchers obtained permission to use one form of the 2014, Grade 7, English Second Language Test as a measure of reading achievement. The test consisted of 50, standardised, multiple-choice items based on a variety of short texts, reflecting key competencies in the Namibian English Second Language syllabus. Individual scores were recorded out of 50 (see Table 1).

Procedure
Ethical research procedures, as stipulated by the Centre for Research and Publications, University of Namibia, and the Data Protection Official at the Norwegian Social Science Data Services, were strictly adhered to. This specifically included informed consent from all learners, as well as the confidentiality of individuals' data and data storage. After obtaining the necessary permission from the Ministry of Education and Culture in Namibia, the purpose of the research was explained to the principals at the selected schools, and written consent from parents and learners was obtained. Data were collected in March 2016. The SAT was administered first, followed by the MRQ and the RAI, after a short break. With these questionnaires and tests, the reading motivation, reading activity, as well as reading achievement of learners, were established. All data were entered into a data base and SPSS 24 was used for all statistical procedures.

Results
First, the descriptive statistics of both reading activity and motivation will be discussed. This is followed by a presentation of the relationship between these two variables, as well as the relationship with achievement and with gender.
While all participants completed the instruments, complete data sets for all 402 were not obtained, as some did not complete all the items. As recommended by Pallant (2011), the "exclude cases pairwise option" (p. 58) was, where relevant, utilised to exclude cases where data were missing, only for the dimension of motivation, mode or purpose of reading. This accounts for differences in the n-statistic. Only cases with complete data were included to calculate the mean MRQ and RAI scores.

Reading activity
For this study, reading activity refers to the breadth (text type or genre), frequency, purpose and mode of reading. Both reading during the previous week and the general frequency of reading a genre were gauged. Composite scores were created to measure the purpose of reading (reading for own interest or reading for school purposes), as well as the mode of reading (reading of traditional print text or reading in digital format).

Reading activity during the previous week
For the first set of questions in the RAI (reading texts for own interest or for school work during the past week), the vast majority of participants who responded affirmatively, also gave additional, narrative information on the topics read. Table 2 compares the responses for each of the text types, modes of, and purposes for reading.
Although participants demonstrated a fair amount of reading, they read more for their own interest than for school. They read the various text types in print format (which excluded school text books) more for their own interest than for school purposes. Both digital and print modes were also used more for their own interests than for school. Newspapers were read far more for their own interests (63%) than for school (22%). The reading of information books seemed unpopular, for both own purposes as well as for school (32 and 28%), in comparison to the reading of story books. About half of the participants indicated that they used the Internet for reading -equally for school work and for their own interests. Mobile phones were mostly used for personal communication.

General reading activity: Purpose and mode
For the second set of questions in the RAI the composite scores on frequency in terms of mode and purpose of reading were used, i.e. how of often they read in digital vs. print mode, and how often for academic vs. pleasure purposes. Participants responded to a four-point Likert scale. All statistics were calculated out of 4, with 1 indicating low reading activity and 4 indicating high activity. A mean reading activity score was calculated based on all the items in this second set of questions. These results are presented in Table 3.
All the scores, except those for reading for academic purposes in digital mode, were above the midpoint of 2.5, and showed little variance. Participants read texts in traditional print mode more frequently than digital texts on cell phones or computers. When the digital and print modes were combined, reading for pleasure (own purposes) had a higher score than reading for academic (school) purposes. This held true for texts in print and digital formats. While the highest reading activity score recorded was for reading printed texts for pleasure (2.82), the score for digital reading for pleasure, was also fairly high, and included reading text messages from friends and conducting internet searches.

Reading motivation
Descriptive statistics regarding the results from the Motivation for Reading Questionnaire (MRQ) are presented in Table 4. Sample size, means and standard deviations are presented separately for each dimension of motivation; for the mean of all 11 dimensions (further referred to as Mean Motivation Score); as well as the mean score for extrinsic and intrinsic motivation. All scores are out of four, with one indicating weak motivation.
In general, participants reported relatively high motivation levels. The Mean Motivation Score for all participants with complete data was 3.24. However, clear differences existed between dimensions rated high and those rated low. The dimensions of reading motivation rated highly were importance of reading (3.62), reading curiosity (3.54) and reading for grades (3.48). Lower means were obtained for competition in reading (2.83), social reasons for reading (2.81) and reading-work avoidance (2.8). The sampled readers thus seemed highly motivated to read, but less so for social reasons or in competition with others. They also claimed not to be avoiding reading tasks.
Furthermore, participants were slightly more intrinsically (3.34) than extrinsically motivated (3.08) to read. Higher scale scores came from both intrinsic (Curiosity) and extrinsic composites (Grades). In the next sections, correlations between these findings and other variables will be discussed.

Relationships between motivation and reading activity
The relationship between reading motivation and reading activity was investigated by employing Spearman's correlation. The Mean Motivation Score, as well as means for composites of Intrinsic and Extrinsic Motivation were correlated with various composites of reading activity. Statistically significant, positive correlations were found between Reading Motivation and the various composites of Reading Activity. Results are given in Table 5.
The Mean Reading Motivation correlated moderately with Mean Reading Activity (r = 0.274; p < 0.01). The highest correlation was found between Mean Reading Motivation and the reading of print text for pleasure (r = 0.405; p < 0.01). Slightly lower correlations were found between reading motivation and reading for pleasure (digital and print) (r = 0.374; p < 0.01), as well as reading of print (academic and pleasure) (r = 0.326; p < 0.01). Low, yet statistically significant, correlations were evident between motivation and reading print texts for academic purposes, as well as for all forms of digital reading. Other correlations were negligible and/or statistically not significant. The same trends were found when reading activity, intrinsic and extrinsic motivation were correlated. Higher correlations were found between pleasurable reading and all 3 motivation constructs (r = 0.374; r = 0.313; r = 0.305; p < 0.01) than between these and academic reading. The reading of printed text also correlated positively with the motivation scales throughout. Noteworthy is that a higher correlation between reading activity and extrinsic motivation (r = 0.236; p < 0.05), as opposed to intrinsic motivation (r = 0.212; p < 0.05), was found for this sample.

Relationships between motivation and reading achievement
The mean scores of the different dimensions of reading motivation, as well as for intrinsic and extrinsic motivation, were correlated with reading achievement and reading activity. No statistically significant correlations between Reading Activity and Reading Achievement in English Second Language were found. However, several small, positive and significant correlations between Reading Achievement and dimensions of motivation (including the composite scale of Intrinsic Motivation) were established (see Table 6).
A statistically significant, positive correlation was found between Intrinsic Motivation and Achievement (r = 0.186; p < 0.01). Similar to correlations with Challenge, and Compliance, this is a fairly weak correlation and thus these variables can only be regarded as slightly related. Higher correlations between Reading Achievement and the dimension of Involvement emerged. A small, negative correlation between achievement and Social Reasons for Reading was found.

Table 5. Correlations between motivation scale scores and reading activity
Note: Ns range from 266-380.

Gender differences regarding motivation and reading activity
The mean scores of boys and girls were very similar in terms of motivation, reading activity and achievement, with girls obtaining marginally higher scores. Girls scored higher in reading achievement (33.18 against 30.22). They also obtained higher mean scores for digital reading (2.7 compared to 2.5) and reading activity in general (2.76 compared to 2.61). Girls also had slightly higher scores in motivation, with the biggest difference in the intrinsic motivation score (3.4 compared to 3.03). All these differences were statistically significant.

Discussion
The purpose of the study was to describe the nature of the reading activities and reading motivation of a sample of Namibian, early adolescent, urban learners with adequate reading comprehension skills in their second language. A further aim of the study was to ascertain whether there were significant correlations between the different aspects of motivation and reading activity related to this sample, as well as to determine whether gender played any role in these variables. The suitability of the MRQ in an African context was explored, as well as a new instrument for Reading Activity utilised.

Reading activity
In most studies, reading activity has been investigated in terms of the time spent on reading activities, as well as the breadth of reading (types and genres read.) We have included purpose and preferred mode as part of Reading Activity.
When one looks at the recent reading activities of the participants (past week), it is shown that newspapers and stories were read most. The high frequency of reading fiction (similar to the results of Birr Moje, Overby, Tysvaer, & Morris, 2008), could possibly be linked to the own preferences of these adolescent readers, as they read fiction for pleasure as well as for academic purposes; however, it could also be linked to what teachers expect from them. The fact that the newspaper is an important source of information in Namibian homes possibly accounted for the high reading frequency of newspapers for own interest. It is worrying that, apart from reading story books, reading for academic purposes seemed low. The percentage of participants indicating that they read information books was low, and newspapers were also not read for academic purposes, such as for school assignments. This is a cause for concern, given the demands of the curriculum regarding the reading of non-fiction in subsequent school years. Together with the participants' possible preferences, the shortage of non-fiction in print form was most likely a causing factor (Tötemeyer, 2013). Given the availability of this resource in Namibia, the internet played a significant role in reading both for pleasure and academic purposes for this group of early adolescents. Mobile phones were more frequently used for own, recreational purposes, such as chatting and playing games. It was also used for clarifying information about school assignments or to connect to the internet by means of smart phones.
In general, participants indicated moderate reading activity. Over a week, they had read more for pleasure than for academic reasons, confirming the reading patterns reported above. Texts in print were most frequently read. The use of digital modes for reading was less frequent than that of printed texts.
It seems important to respond to these findings in the compilation of reading material and reading programmes. Reading frequency could be maintained by offering interesting stories to read. However, the low interest in non-fiction should be addressed urgently, as this can be regarded as a prerequisite for school success. It seems that, apart from motivating the reading of non-fiction texts in various ways, newspapers could be a valuable teacher resource, given the participants' high interest in reading newspapers. Therefore, newspapers should be made available for reading during school hours. The growing awareness and use of digital modes of reading, including the use of the internet, should further be explored in the Namibian context. In this regard, McGeown, Duncan, Griffiths, and Stothard (2015, p. 565) points out that digital literacy activities are increasingly replacing traditional book-reading in adolescents' reading activities.

Reading motivation
Reading motivation scores were relatively high, which is typical in self-report-questionnaires in Namibia (Tötemeyer, Kirchner, & Alexander, 2014;Wikan et al., 2007). The most important dimensions of reading motivation that these early adolescents identified with, were curiosity about topics, the importance of reading and reading for grades. These dimensions correspond closely with those rated highest by Kenyan learners (Mucherah & Herendeen, 2013, p. 581). This finding is also similar to that of Greaney and Neuman (1990) who indicated escape, enjoyment and utility as central functions of early adolescent reading across a range of cultural settings. This implies that these participants read because of the inherent nature of the text, but also because they could identify with the value of reading. External reward in the form of grades also played an important role. In line with Self-Determination and Expectancy Value theories of motivation, these dimensions of reading motivation represent motivational stances across a continuum-from being intrinsically motivated to read to having integrated and internalised the value of the activity. This implies that teachers should adapt strategies to reply to these varying, motivational stances (Cole, 2010). While texts should be introduced based on the interests of the individual, sharing information about the importance and value of reading to fulfil academic and career goals is also essential. Our findings support the work on reading preferences previously conducted in Namibia (Tötemeyer et al., 2015). The fact that these Namibian readers seemed to be more intrinsically than extrinsically motivated to read, indicates that support of relevant intrinsically motivated behaviour should facilitate the improvement of reading competence.
The relatively low importance of reading for social reasons and competition in this group is surprising, given the fact that current research indicates that the sharing of reading experiences impacts positively on motivation, also in the African context (Bright, Emmanuel, & Tuboulayfa, 2015). We can only speculate on causes for this phenomenon. It is quite possible that these results reflect the Namibian school culture where social discussion about reading can sometimes be met with gossip and ridicule (Kirchner, 2017), and where only the top achievers usually engage in interaction around texts, as well as compete to outperform peers in reading tasks. Further research to explore the different ways that classroom talk can be used to strengthen discussions around texts seems important. In discussions on the challenges in Namibian and African classrooms, the continued emphasis on teacher-led discussions, as well as the challenges to implement learner-centred approaches to teaching and learning, are often alluded to (Cleghorn & Rollnick,2002, p. 356;O'Sullivan, 2004, p. 600).
The high motivation levels were quite surprising and encouraging for a sample of early adolescents and might point to differences between the Namibian and other contexts. The high, self-reported motivation levels could be seen as a result of the status of reading in the Namibian community, where illiterate parents are not uncommon. Another reason might be, as Mucherah and Herendeen (2013, p. 589) postulated, that reading motivation levels could be high during the final year of primary school, based on learners' aspirations towards higher grades and access to good secondary education. These high motivation levels may not be generalisable to other, early-adolescent, Namibian readers where reading levels are lower and where reading material is not as available as in this urban region.

Relationships between reading activity, reading motivation and reading achievement
We found positive relationships between reading motivation and reading activity supporting other research findings (Cox & Guthrie, 2001;Edmunds & Bauserman, 2006). Similar to Morgan and Fuchs's (2007, p. 177), this correlation can be seen as reciprocal, where high reading activity increases motivation, and the resulting higher motivation again gives rise to higher reading activity.
Specifically, the reading of (print) texts for pleasure correlated relatively highly with motivation. This relationship may indicate the important impact that a good story has on reading motivation, and that the reading of fiction stimulates intrinsic reasons for reading. The reading of extended texts, in particular the reading of books, was found to impact positively on reading achievement (McGeown et al., 2015, p. 565). Low correlations between the reading of academic texts and motivation could be explained by Namibian readers' relatively poor exposure to interesting, non-fiction material in print. The same may be true for reading in digital format.
Since readers were more intrinsically motivated to read, the significant correlation between external motivation and reading activity was surprising; however, at this stage this finding is uncorroborated by other research. This could be explained by the fact that Namibian readers were aware of those values and benefits of reading that were linked more directly to aspects of external motivation, such as the educational value of reading and reading to be successful in school (Tötemeyer et al., 2015, p. 22).
Research evidence indicates that the amount of leisure, out of school reading; and the reading of fiction; contribute significantly to reading achievement (De Naeghel et al., 2012, p. 1019McGeown et al., 2015, p. 566;Schiefele & Schaffner, 2016, p. 2). However, we did not find any correlation between reading activity and reading achievement. Nevertheless, the amount of out of school reading, specifically the frequent reading of fiction, found in the current study with relatively competent readers, remains relevant.
Research consistently points to the fact that reading motivation and reading achievement are interdependent and correlated (Conradi et al., 2014;Edmunds & Bauserman, 2006). Klauda and Guthrie (2015, p. 262) found stronger relationships between these two variables for advanced readers, compared to struggling ones. In line with these findings, various positive though modest correlations between reading motivation and reading achievement were found in this study. Readers enjoyed the involvement and challenge of a variety of texts. Various studies demonstrated this correlation between reading achievement, as measured by standardised tests, and intrinsic motivation, for both elementary and middle school learners (Baker & Wigfield, 1999;Guthrie, Coddington, & Wigfield, 2009;Unrau & Schlackman, 2006). In line with Self Determination Theory intrinsic and autonomous motivation, in contrast to external motivation, normally correlates positively with reading achievement (De Naeghel, Van Keer, Vansteenkiste, Haerens, & Aelterman, 2016, p. 233).
The complexity of these relationships is highlighted by the slight correlations between achievement and compliance as well as social reasons for reading. External pressure to achieve thus do impact on the motivation of these learners. The negative correlation with social reasons for reading was surprising, and may indicate that high achievers did not find discussions about reading in Namibian classrooms or with peers enriching or valuable. In this respect, the Namibian classroom atmosphere and activities might have differed substantially from those investigated in other, American and European contexts. However, providing opportunities for social interaction remains important as individuals may differ in their need for discussing texts (Cole, 2010, p. 161).
The modest correlations between the reading of digital texts and other variables possibly indicate an unexplored, though promising, field of study in Namibia. The current research should be followed up as the increasing availability of internet connectivity, also at schools, could shape these results in the near future, and may conceivably lead to a change in classroom strategies where information and communication technology will be used increasingly in academic spaces.
Our research confirms most of the findings regarding gender differences, in terms of early adolescents. As was found by Taylor (2004), girls were better readers than boys, and also seemed to demonstrate more breadth in their reading activities. Similar to De Naeghel et al. (2014), Bozack (2011) and Schiefele and Schaffner (2016), we found that girls, in general, showed more positive attitudes and higher motivation towards reading than boys, and seemed to be more intrinsically motivated than boys. The finding that girls used digital modes of reading more often than boys was surprising, and is not supported by other gender research. It could be possible that girls spent more time on social media than boys. Therefore, our findings are thus somewhat related to the findings of Svensson (2014, p. 347), who found that female learners spent more time on social media than males. The gender differences observed in our study were small. Our findings, therefore, can neither confirm nor refute those of Tötemeyer et al. (2015), who reported no differences in reading preferences between Namibian girls and boys, as well as Mucherah and Herendeen (2013) who also found no gender differences regarding reading motivation amongst African learners.

Conclusion
It seems abundantly clear that attention to reading motivation in schools and the reading classroom is not an optional extra, but should be included in all reading curricula. Guthrie and Klauda (2015, p. 48) posits that "… the benefits of motivation for achievement growth are not a mere marginal luxury. Reading motivation may stand as the strongest psychological variable influencing achievement". Motivated readers, moreover, have a good chance of becoming life-long readers (Wang & Guthrie, 2004).
While valuable within the African context, this study also had limitations. Relatively small in scale, the results should be generalised with care, and only within similar contexts, such as to schools with some basic, reading resources, and to urban African contexts only. Furthermore, the data of this study were gathered quantitatively, by means of self-report questionnaires measuring reading motivation as general construct, which inherently have limitations in explaining variation (Neugebauer, 2014) and may not be totally free from some influence of social desirability.
The Namibian, early adolescent readers included in this study seem highly motivated to read. Intrinsic, as well as extrinsic, motivation played a role in the reading profiles of these readers. Given the challenges to improve reading and general school achievement in Namibia, educationalists should use these positive attitudes to advance the reading agenda. The fact that the attitudes of this group may differ from rural and deprived communities should urge stakeholders to reverse the inequalities, unequal opportunities and lack of access to resources that may still exist in Namibia. The provision of reading material for classroom collections and school libraries remain crucial, and should include stories but also information texts, aimed at including the preferences of boys as well.
These results, furthermore, add to the research on reading motivation in Africa. It also indicates that quantitative instruments, such as the MRQ and RAI, are useful within the African context, albeit not without some basic adaptations. Further research in Namibia on reading and reading motivation, specifically regarding early adolescents, should include different kinds of intervention studies that document reading strategies and reading programmes developed for this specific context. These findings should also be expanded further by including studies of teachers' views and beliefs about reading. Considering the development of programmes (similar to McGeown et al., 2015) that include digital opportunities for reading and instruction to navigate this complex reading process effectively also seem important, given the rollout of internet connectivity in Namibia also to schools without electricity.