An investigation into the English reading comprehension of Grade 10 English first additional language learners at a senior secondary school

Poor levels of English first additional language (EFAL) reading comprehension among school learners at most public schools in South Africa are a great concern. To this end, reading interventions are mounted at different intervals not only here in South Africa (Donald & Condy 2003; Dotwana 2009; Pretorius & Matchet 2003), but also in places such as the United States of America (USA) (Asimov 2006; Baer et al. 2007; Grabe 2010) and the United Kingdom (UK) (Center 2005; Chall, Jacobs & Baldwin 2004; Heinz 2004; Silliman & Wilkinson 2004). There are sentiments expressed by some scholars to the effect that poor levels of English academic literacy, especially poor levels of English reading comprehension, prevalent in schools are a world-wide challenge (see Muhammad 2013). For example, Asimov (2006) points out that fewer than half of California’s students’ reading and numeracy competence is commensurate with that of grade level learners nearly a decade after the state began overhauling its public education. In the same breath, in Britain too, the year 2008 was declared a National Year of Reading because of the poor reading competence of some of the learners in English (Rankin 2013). Similarly, in South Africa, decisions to tackle low levels of literacy afresh were taken before the announcement of the results of the Progress in International Reading Literacy Study. In this case, in a major study of literacy skills of Grade 4 and 5 learners conducted in forty countries, South Africa came out at the bottom, notching a fortieth position (Baer et al. 2007).


Introduction
Poor levels of English first additional language (EFAL) reading comprehension among school learners at most public schools in South Africa are a great concern.To this end, reading interventions are mounted at different intervals not only here in South Africa (Donald & Condy 2003;Dotwana 2009;Pretorius & Matchet 2003), but also in places such as the United States of America (USA) (Asimov 2006;Baer et al. 2007;Grabe 2010) and the United Kingdom (UK) (Center 2005;Chall, Jacobs & Baldwin 2004;Heinz 2004;Silliman & Wilkinson 2004).There are sentiments expressed by some scholars to the effect that poor levels of English academic literacy, especially poor levels of English reading comprehension, prevalent in schools are a world-wide challenge (see Muhammad 2013).For example, Asimov (2006) points out that fewer than half of California's students' reading and numeracy competence is commensurate with that of grade level learners nearly a decade after the state began overhauling its public education.In the same breath, in Britain too, the year 2008 was declared a National Year of Reading because of the poor reading competence of some of the learners in English (Rankin 2013).Similarly, in South Africa, decisions to tackle low levels of literacy afresh were taken before the announcement of the results of the Progress in International Reading Literacy Study.In this case, in a major study of literacy skills of Grade 4 and 5 learners conducted in forty countries, South Africa came out at the bottom, notching a fortieth position (Baer et al. 2007).
It is worth mentioning that fluent oral reading in EFAL is an essential skill for learners at any level of their learning career.In fact, most language teaching experts agree that efficient and effective word recognition skills are a sine qua non for becoming a successful reader (Chard et al. 2009;Grabe 2009Grabe , 2010;;Mikulecky 2008).For learning to be successful and effective, learners need to be able to read well.It becomes clear, therefore, that EFAL is very important in every respect since it is used as a language of learning and teaching (LOLT) for most learning areas at schools in South Africa.It also serves as a lingua franca for many learners speaking different languages.Thus, it is a tool that enables them to communicate with one another.As a result, EFAL teachers should teach it bearing in mind the aforesaid factors.

An investigation into the English reading comprehension of Grade 10 English first additional language learners at a senior secondary school
Poor levels of English first additional language (EFAL) reading comprehension amongst school learners at most public schools in South Africa are a great concern.In fact, for learning to be successful and effective, learners need to be able to read well in EFAL.This is more so as EFAL serves as a language of learning and teaching (LOLT) for most learning areas in South Africa's public schools.Against this background, this study set out in 2012 to investigate the English reading comprehension of Grade 10 EFAL learners at a senior secondary school.
Using purposive and voluntary sampling techniques, the study had 17 EFAL learners (M = 10, F-7) as its participants.It employed three reading measures, a recall task, a summary and a comprehension test, which were based on three English extracts, to assess participants' reading comprehension.It then assessed and scored participants' responses to the three tasks by using an oral reading rubric and two prepared marking memoranda.One of the findings of this study was that, of the three reading tasks administered, participants did slightly above average in the comprehension test, but performed below average in the two other tasks -the recall and summary tasks.
pre-reading, during-reading and post-reading strategies (see Cekiso & Madikiza 2014;Ghuma 2011).For example, prereading strategies help facilitate an initial understanding of a text.They do this by utilising text signals, formulating initial hypotheses or activating prior knowledge.For their part, during-reading strategies are used for identifying and interpreting the main information in the text.They encompass searching for linguistic cues; skipping unimportant or irrelevant linguistic items; comparing different main ideas; generating questions about the main ideas; and visualising, inferring, and making affective responses to the text.Lastly, post-reading strategies, as their name suggests, are employed after the reading process to evaluate a reading text.They entail, among other things: re-reading parts of the text to enhance its comprehension; clarifying hazy parts; evaluating the content of the text or its overall quality; summarising the text; reconstructing the hypothesised macro elements of the text; and making inferences (Ghuma 2011;see Yukselir 2014).
In addition to these three reading strategies, there is the fourth one = reading monitoring strategies = identified by Ghuma (2011).It is worth highlighting, in this case, that reading comprehension is construed as a complex activity that entails the conscious and unconscious use of multiple strategies (cognitive and meta-cognitive) that the reader has to bring to bear to a given text so as to be able to extract the meaning embedded in it (see Johnson 1983).Above all, the success of reading comprehension is largely determined by the characteristics of the text, the level of difficulty of the text, the context in which the text is embedded, the reader's prior knowledge, the level of the reading ability (see Johnson 1983;Pardo 2004) and the reader's learning styles.
Given the points highlighted above, the current study contends that the reading ability of EFAL learners to comprehend texts written in English is crucial in preparing such learners to be optimally functional in the social, political and economic arenas.Therefore, it is important that learners acquire the necessary reading comprehension in EFAL in order for them to be able to meaningfully interact with their reading texts at school.

Framing issues
Reading as an activity and as a literacy practice is theorised from different perspectives, depending on the purpose it is meant to achieve, and on the scholarly angle from which it is conceptualised.For example, Granville (2001) discusses three views of reading: the text-based view, the interactive view and the critical, socio-cultural view.She argues each of these views resonates with its respective pedagogy or approach to teaching reading.According to the first view, the meaning resides unproblematically in the text and the reader simply has to discover it as it is fixed and determinate.
It conceives reading as a unidirectional enterprise in which the reader is passive.In terms of classroom teaching, this view privileges such mechanical reading skills as word recognition, decoding, phonics, and reading aloud.Reading comprehension tasks, in keeping with this view, are generally biased towards literal meanings and low-order inferences (Granville 2001).
Informed principally by cognitive psychology, the second view of reading prizes the centrality of the interaction between readers' prior knowledge (background knowledge) and the text.It conceptualises the reader as an active meaning-maker as for it, the meaning is not unproblematically embedded in the text.In this sense, according to it, reading is a dynamic dual process between the reader and the text.Most significantly, this view subscribes to a plurality of meaning in as far as the text is concerned.However, the notion of plural meanings for given text or a given word (see Granville 2001;cf. Bakhtin 1981) as articulated by this view, can itself lead to semantic confusion, particularly when each reader ascribes his or her own idiosyncratic meaning to a given text or to a given word.
The third view of reading contends that textual meanings are not confined to the triad of the text, the writer and the reader.Rather, it argues, meanings are implicated in the social, political, cultural and historical contexts in which the text, the writer and the reader are embedded.Therefore, any meaning -including unbridled plural meanings attached to any text -is determined by social, political, cultural and historical exigencies.The criticality of this view lies in the ability of readers to be critical readers (Granville 2001; also see Shihab 2011;Talebinejad & Matou 2012).
Added to Granville's (2001) three views of reading are a schematic-theoretical view of reading and a model of a thoughtful, expert reader.The two are loosely interrelated (Pearson et al. 1990).A schematic-theoretical view of reading has some elements of Granville's (2001) second view of reading.Contrasting sharply with skills-based view of comprehension, this particular view conceives reading as an active process in which meaning is constructed by aligning old knowledge to new information communicated by the text.It contends that readers construct meaning by participating in a series of recursive interactions.In addition, it posits that new textual data serves as an invitation to confirm, reconsider or revise existing knowledge in line with the information at hand.Thus, iteratively, readers get to construct and reconstruct meaning (Pearson et al. 1990).
Aligned to the schema-theoretically based view of reading is a model of a thoughtful, expert reader.According to this model, active and expert readers are capable of: • Constantly searching for links between what they know and what they encounter as new information in the text.• Constantly monitoring the adequacy of the models of text meaning that they construct.• Taking steps to repair faulty comprehension once they realise they have failed to understand something.• Learning very early on to distinguish less important from important ideas in the text they read.• Synthesising information within and across texts and reading experiences.• Making inferences during and after reading to achieve a holistic, integrated understanding of what they read.• Consciously and unconsciously, asking questions about themselves, about the writers of texts and about the texts they read (Pearson et al. 1990).
Two types of reading, which occur in converse pairs, warrant mentioning in this section.They are serial/non-serial reading and single/repeated reading.Serial reading refers to reading a text in a linear manner from start to finish.In contrast, non-serial reading entails reading the text from one section to another in a non-linear or incomplete way.For their part, both single and repeated reading involve reading the text once and repeatedly, respectively (O'Hara 1996).
A point worth noting here is that the education system in South Africa allows learners to study in their mother-tongue during their first year of study, and then to switch to EFAL as a language of learning and teaching (LOLT) at a grade 4 level.Learners usually need between six and eight years to learn a second language before they can use it as an LOLT at school (Heugh 2005).In this instance, a study conducted by Cruickshank (2006) supports the idea that EFAL learners, on average, require at least five years of exposure to academic English to make up for their lack of native speaker norms.
In this regard, there are types of poor reading habits that are at the core of poor reading competence for most EFAL learners.These are: limited eye span; finger-pointing; vocalisation; and slow reading speed.
Here is what each of these reading habits is about: • Limited eye span refers to a reader's inability to read enough words in a line or in a sentence without moving the eyes from one end of the line or of the sentence to the other.• Finger-pointing is when a reader follows words in a passage using his or her finger.• Vocalisation is about a reader reading words aloud or verbalising what he or she is reading.• Slow reading speed is reading fewer than the expected number of words per minute (Heinz 2004;Mikulecky 2008;Perfetti 1985;Pretorius & Matchet, 2003;Taguchi, Takayasu-Maass & Gorsuch 2004).
All these factors negatively affect EFAL learners' development of reading skills.This is supported by Heinz (2004) and Mikulecky (2008), who state that for an efficient reader to decide on the most appropriate reading strategy, he or she must be sure of the purpose of a reading activity in which he or she is to engage.According to them, the lack of a reading purpose may lead to boredom, distraction and miscomprehension.In this context, Algozzine et al. (2009) are of the view that fluent oral reading is one of the essential skills for children from elementary schooling.At the same time, Katz and Carlisle (2009) maintain that children with a reading disability demonstrate slow progress in their knowledge and application of cognitive strategies during reading, compared to proficient readers of the same age.
Furthermore, there are, according to Pretorius and Matchet (2003), three distinct levels of reading comprehension at which readers operate at any given time.These are: the independent, instructional and frustration levels.The first level consists of highly skilled readers who can independently access information from texts through reading, and who can also effectively learn from texts appropriate for that specific maturation level.The second level relates to readers who have minor reading problems but who can comprehend the written information bit by bit.And the third level entails readers who have major reading problems, especially with regard to comprehending written information.In addition, readers at this level can read below their maturation level.
As such, they need intensive reading programmes to increase their reading skills level.Given all this, the current study sought to find ways through which poor reading performance of grade 10 EFAL learners could be identified at a given senior secondary school in Mthatha.

Problem statement
Poor reading amongst school learners as a factor inhibiting literacy acquisition has been a great cause for concern for both teachers and education authorities worldwide (see Against this background, one of the researchers in this study observed during her long teaching stint at the school in question, that grade 10 EFAL learners lacked the necessary reading skills.This was a problem that tended to affect their overall reading comprehension adversely and persisted unabatedly. http://www.rw.org.zadoi:10.4102/rw.v6i1.62 And it was particularly pronounced among most grade 10 EFAL learners who struggled to read and fully comprehend most of the EFAL reading material meant for them.

Rationale of the study
The perennial difficulty grade 10 EFAL learners encountered in reading EFAL material meant for their grade prompted the researchers concerned to conduct the current study.For example, when grade 10 EFAL learners were given comprehension texts to read with a view to answering questions based on them, they tended to transcribe aspects of the texts verbatim.They also took a very long time to finish reading texts.That is, their reading speed was low for their grade.Based on the points highlighted here, the purpose of the study was to find ways which would determine the reading comprehension of grade 10 EFAL learners at the school in question.

Research methodology
This study used a case study design.The latter = not to be mistaken for an experimental single-case design = embodies at least two features: a limited but specific focus and an indepth investigation of a variable or phenomenon (Griffee 2012).The research methodology it adopted was grounded in a qualitative research paradigm.The choice of this research methodology was informed by the types of data the study elicited from participants.

Research questions
The study addressed the following three research questions: • How much can grade 10 EFAL learners at a senior secondary school in Mthatha recall and summarise from given reading extracts?• How do these learners fare in a reading comprehension test?• How does these learners' reading performance compare in each of the reading tasks given to them?

Participants and sampling technique
Participants in this study were seventeen Grade 10 isiXhosaspeaking learners doing EFAL at a senior secondary school in Mthatha.Their consent was first sought and their voluntary participation was explained before the study was conducted.Of these participants, 10 were male learners, and 7 were female learners.These participants were selected from a target population of 181 EFAL learners through purposive and voluntary sampling techniques.These two sampling techniques were used in order to involve participants known to have had some reading problems.

Data elicitation instruments and data elicitation procedures
Three instruments were used to elicit data from participants: a reading recall task; a reading summary task; and a reading comprehension test.All the three reading tasks were administered to participants on three days -with each task on each day -within one week in March 2012.For the first task, participants were given an extract adapted from the article from the 2011 SA National Geographic Traveller magazine to read for 20 minutes.They were then required to orally recall, for five minutes, as many words as they could remember from the extract.For a reading summary task, participants were given an extract, About bees, adapted from the article from the 2011 Home and Garden magazine, to read for 20 minutes.Thereafter, they were asked to spend 15 minutes listing eight main points = in not more than 80 words = of the extract in full sentences.For the last task, participants were required to read an extract adapted from the 2011 March Bona magazine article.They had to spend 25 minutes reading the extract and 15 minutes responding to ten reading comprehension items.
All the three tasks were marked according to three prepared reading memoranda (an oral recall rubric for the recall task, and two marking memoranda for both the summary task and the reading comprehension test).The reading recall and reading comprehension tasks were assessed out of a total mark of 15 each, while the reading summary task was assessed out of a total mark of 10.

Data analysis
Participants' oral and written responses to the three reading tasks were assessed using an oral recall rubric (for the oral recall task) and two prepared marking memoranda (for the other two tasks).Thereafter, the responses were graded accordingly using both raw scores and percentages.The raw scores and their corresponding percentages (including overall percentage scores and mean percentage scores) were duly tabulated (see Table 1).Most importantly, the findings of this study were interpreted narratively and descriptively.but also none of them managed to score more than 47% in the oral recall task, and again none of them obtained more than 40% in the written summary task (see Table 1).In fact, in the former task, 7 participants (41% of the participants) obtained 47%; 1 participant scored 44%; 2 participants (12% of the participants) scored 40%; 6 participants (35% of the participants) obtained 33%; and 1 participant scored 13%.Likewise, in the latter task, only 2 participants (12% of the participants) obtained 40%; 8 participants (47% of the participants) scored 30%; 3 participants (18% of the participants) obtained 20%; 1 participant received 10%; and 3 participants (18% of the participants) scored 0% each.Now, taking participants' overall performance in the two tasks into consideration, it became evident that all the participants did not do well in these two tasks.Actually, participants' performance in these two tasks was not only below average in all the items comprising the two tasks, but there was also a zero percent performance in the summary for some participants.So, based on their performance in these two tasks, and taking into consideration that reading can be assessed through various measures, two of which are recalls and summaries of reading texts (see Bernhardt 2011;Gibbons 2002;Grabe 2009Grabe , 2010;;Han & Anderson 2009;Mikulecky 2008), all the participants scored varyingly well below average in these two measures of reading (see Table 1).This means that they hardly exhibited elements of expert readers as theorised by Pearson et al. (1990).
In keeping with the three reading levels cited earlier (also see Pretorius & Matchet 2003), participants' performance in the reading comprehension test indicates that half the participants tended to operate at the second level = the instructional level.Readers functioning at this level have minor reading problems but can comprehend the written information piecemeal after engaging in repeated reading, as O'Hara (1996) contends.The same seems to be the case with half the participants in the reading comprehension test.However, the other participants (47% of the participants) tended to operate at the third level = the frustration level.That is, these participants had major reading problems and performed below their maturation reading level in this task.Thus, they only grappled with the mechanical reading skills typifying the text-based view of reading.
By contrast, all the participants tended to function at the frustration level = and by analogy, well below their reading maturation level = in the other two reading tasks: an oral recall and a written summary.Most importantly, they did so variedly in both tasks, and far below an average reading performance in the latter task than in the former task (see Table 1).In fact, some participants (18% of the participants) seem to have performed well below even the frustration level given that they failed to obtain any score in this task.Another point worth noting here is that, apart from one participant, no other participant in the three tasks managed to function at the independent reading level or at the expert reader level.The latter level comprises, according to Pretorius

Findings and discussion
The following are the findings and their related discussion integrated into one section.Participants' consolidated scores (raw and mean scores) and percentages (overall and mean percentages) are reflected in Table 1.According to this table, of the three reading tasks, participants did slightly above average (53%) in the comprehension test (better than they did in the other two tasks).In contrast, they did far below average in the other two tasks = the recall and summary tasks.For example, their mean percentage scores in these two tasks were 39.05% and 22%, respectively.
In relation to the recall task, no participant achieved a pass or an average mark in it.In addition, out of a total mark of 15 in this task, the highest and lowest marks were 7 and 2, each.Similarly, but displaying a much lower performance, the summary task (which was assessed out of a total mark of 10) had no participant scoring a pass or an average mark in it, while its highest and lowest marks were 4 and 0, respectively.By contrast, in the reading comprehension test, which was assessed out of a total mark of 15, there were 9 passes.This means that 53% of the participants managed to do well in this task.For all the three reading tasks put together, the highest and lowest percentage scores were 50% and 24%, correspondingly.Moreover, the mean percentage score for the three reading tasks was 38%.
As pointed out in the preceding paragraphs, of the three reading tasks administered in this study, participants did slightly better in the comprehension test.Their mean percentage score in this task was 53% (see Table 1), which was better than their mean percentage scores in the other two reading tasks.In addition, of the 17 participants who took part in the reading comprehension test, 9 (53%) of them passed the test, while 6 (35%) of them scored between 40% and 47%, and 2 (12%) of them scored 20% and 27%, respectively.Of those that passed the test, 1 participant obtained 73%.Looking at the participants' overall performance in this activity, it becomes clear that more than half of them did well in it while the rest did not.This means that half the participants managed to comprehend the items or questions meant for this activity as opposed to less than half who did not.On this basis, and taking into account that reading can be assessed through a variety of measures, one of which is a reading comprehension test (see Bernhardt 2011;Gibbons 2002;Grabe 2009Grabe , 2010;;Han & Anderson 2009;Mikulecky 2008), half the participants did perform above average in this measure, albeit in varying degrees (see Table 1).It nonetheless needs to be pointed out that participants' performance in the reading comprehension test tends to indicate that only half the participants operated at the rudimentary level of the text-based view of reading identified by Granville (2001).This means that participants barely functioned within the schematic-theoretical view of reading espoused by Pearson et al. (1990).
However, and also as highlighted earlier, all participants did not do well in the other two reading tasks: a recall and a summary.Not only did they not do well in these two tasks, and Matchet (2003), highly skilled or expert readers who can independently access information from texts through reading, and who can also effectively learn from texts appropriate for this specific maturation level.Pearson et al. (1990) regard such readers as thoughtful, expert readers.

Conclusion and recommendations
The current study set out to find ways through which poor reading performance of grade 10 EFAL learners could be identified at a given senior secondary school in Mthatha.It did so through employing three reading tasks: a reading recall task, a reading summary task and a reading comprehension test.In this regard, one of its observations is that, of the three reading tasks, participants did slightly above average in the comprehension test.By contrast, it also observed that participants did far below average in the other two tasks = the recall and summary tasks.Additionally, it noted that, overall, most participants operated at the frustration level.
There are many recommendations that can be proposed in this study on the basis of its findings.However, only a few are shared here.Reading lessons need to be incorporated into weekly mainstream grade 10 EFAL classes at this senior secondary school so as to accustom learners to reading as a practice, and to help improve learners' reading comprehension.This is necessary as reading cannot be taught as a one-off instructional activity, but has to be taught as an on-going activity into which EFAL learners have to be immersed.In engaging these EFAL learners in reading lessons, different reading strategies can be explored, one of which is the SQ3R (survey, question, read, recite, and review) strategy (see Dube 2011).Infused into this reading strategy can be related pre-and post-reading strategies such as scan, skim, speed, and recall reading and summarizing.Other essential reading strategies that these EFAL learners can be exposed to are: • automatic decoding -being able to recognise a word instantly • visualising -mentally making a picture of what one is reading about • paraphrasing -restating the contents of a text using one's own words • adjusting reading rate according to purpose = making one's reading speed correspond to the purpose of a reading text • drawing conclusions -being able to make conclusions from a reading text • inferencing -being able to make inferences from a reading text (Mikulecky 2008;cf. Harreveld, Baker & Isdale 2008).
Another recommendation is that, since the current study involved a small sample, future studies need to incorporate bigger samples.
In conclusion, this study investigated the reading comprehension of grade 10 EFAL learners = whose mothertongue was isiXhosa = by using three reading measures: a recall, a summary and a comprehension test.It recommends, therefore, that future reading studies investigate EFAL learners' reading comprehension using other measures such as those highlighted in the preceding paragraph, or combining some of these measures with the three employed in this study.

TABLE 1 :
Consolidated scores and percentages for the 3 reading tasks.