Entire issue.pdf South African teacher pro�les and emerging teacher factors: �e picture painted by PIRLS 2006 SURETTE VAN STADEN Centre for Evaluation and Assessment, University of Pretoria SARAH HOWIE Centre for Evaluation and Assessment, University of Pretoria Abstract: !e Progress in International Reading Literacy Study (PIRLS) assessment is an international comparative study of reading skills of Grade Four learners. South Africa’s "rst participation in the study took place in the 2006 cycle (Mullis et al., 2007), with repeat participation planned to take place for PIRLS 2011. PIRLS 2006 results pointed to serious issues of under achievement among South African Grade Four learners, resulting in the adoption of the National Reading Strategy (Department of Edu- cation, 2008) and the Foundations for Learning Campaign. While some time has passed since the release of the PIRLS 2006 results, participation in PIRLS 2011 would highlight trends and possible progress made since the PIRLS 2006 study. !is paper reports on the analysis of the Grade Four learner achievement in the PIRLS 2006 assessment into the teacher characteristics, use of resources and instructional practices and analyses of the PIRLS 2006 Teacher Questionnaire data. !e main "ndings outlined by this paper re#ects the need for teachers’ continued professional develop- ment at Intermediate Phase, the need to employ strategies to retain young teachers and the importance of making available good quality reading materials to schools. Introduction In the last decade, South African teachers have faced extensive changes to the education system and South Africa has successfully produced many policy (including curricula) documents, but has been less successful in implementing them. To add to the di$culties of implementation, there are many di%erences between schools in South Africa, and, a&er thirteen years of democratic rule, schools that were previously designated for White learners only are still elite schools compared to those that were previously disadvantaged under the Apartheid system. Johnson, Monk and Hodges (2000) are of the opinion that RW& Article 48 Reading and writing South Africa e%ectively still has separate education systems operating within the country in light of the starkness of continuing di%erences in teacher edu- cation and educational provision. Fleisch (2007) also re#ects on this in his description of the country having two nations. Howie (2001) earlier described South Africa as containing features of both richer countries and poorer coun- tries with regard to its education system. Notwithstanding these continuing variances in education provision, the national Department of Education views teachers as key contributors to transformation in South Africa. In the ‘Revised National Curriculum State- ment’ (Department of Education, 2002), the Department presented a vision of teachers who are quali"ed, competent, dedicated and caring. Teachers’ roles and functions are seen to include being mediators of learning, interpreters and designers of Learning Programmes and materials, leaders, managers, administrators, scholars, researchers, lifelong learners, community members and citizens, and assessors and Learning Area or Phase specialists. In the next section, in the light of the above, the literature on teacher e%ectiveness is reviewed, on the assumption that teacher e%ectiveness is a critical component of the education system. Literature review on teacher e!ectiveness Ongoing concerns about the development of learners’ literacy skills in South Africa drive the literacy teaching and learning research landscape. !ese include: concerns associated with school students’ development of basic lit- eracy skills at the foundational levels of education (Bloch, 1999, Lessing and de Witt, 2005); concerns about students’ acquisition of more advanced literacy skills in high school (Matjila and Pretorius, 2004, Pretorius and Ribbens, 2005); and concerns about the development of the advanced literate language skills needed for tertiary level education (Pretorius, 2002). Such concerns are con- sistently re#ected in local research (see Howie et al., 2007). With regards to school-related factors impacting on the reading achievement of learners, Howie (2006) reports a number of factors speci"cally related to learners in South African classrooms. !ese factors include inadequate sub- ject knowledge of teachers, inadequate communication between learners and teachers in the language of instruction, lack of instructional materials, teachers’ abilities to manage classroom activities e%ectively and overcrowded class- rooms. !ese factors are also highlighted by the Department of Education (2008). 49Surette Van Staden and Sarah Howie Sailors, Ho%man and Matthee (2007), in their evaluation of schools that pro- mote literacy learning in low-income communities, summarized the work of a number of researchers (such as Weber, 1971, Ho%man and Rutherford, 1984) who identify themes across e%ective schools that could guide reform e%orts in failing schools that operate in resource-poor environments. !e common themes in these schools that in#uenced learner achievement posi- tively included: a clear school mission, e%ective instructional leadership and practices, high expectations for learners to achieve and perform at their best, a safe, orderly, positive physical environment, ongoing curriculum improve- ment, maximum use of available instructional time, frequent monitoring of learner progress and positive home-school connections. Following their work on high-achieving schools from low-income environ- ments in a sample of South African primary schools, Sailors et al. (2007) identi"ed their own set of themes and factors that seem to impact positively on learner achievement. Dovetailing with the work of other researchers, they cite the presence of a safe, orderly learning environment as having a posi- tive in#uence on learner achievement. Another factor identi"ed includes the presence of strong leadership that guides the school in terms of academic guidance, community relations and shared decision making. In describing teachers as ‘excellent’, a third factor is identi"ed that impacts learner achieve- ment positively, where teachers are characterized as committed, competent, caring and collaborative. A fourth identi"ed factor is that of a shared sense of competence, pride and purpose in schools that function e%ectively in low- income communities. Lastly, community participation and engagement with the school constitutes a factor associated with higher achievement among learners (Sailors et al., 2007). Classroom teaching for reading instruction needs to be considered as the critical factor in preventing reading problems and must be the central focus for change (Moats, 1999). Teachers’ acquisition of the teaching skills necessary to bring about the development of literate language competency are critical, (Zimmerman, 2008) especially as, in South Africa, many assumptions have been largely unquestioned about how to teach reading and writing, which languages to use and what counts as high quality practice in classrooms (Bloch, 1999). Stoller and Grabe (2001) emphasize that the requirements for the devel- opment of reading #uency necessitate that teachers as well as curriculum developers determine what instructional options are available to them and how to go about the optimal pursuit of instructional goals in various contexts. 50 Reading and writing Data sources and analysis !e main data source for this paper is the PIRLS 2006 database for Grade Four (Mullis et al., 2007). !e PIRLS 2006 assessment consisted of a read- ing assessment, and in addition to that, contextual questionnaires that were administered to school principals, teachers, parents and learners. For the purposes of reporting results, this paper utilized the data obtained from the teacher questionnaires, as completed by teachers of Grade Four learners who participated in the PIRLS 2006 assessment. Data analysis in this paper is limited to examining results of descriptive sta- tistics to Grade Four learners’ teachers’ responses. !ese results are provided against the background of South Africa’s overall performance in PIRLS 2006 compared internationally, as is reported in the PIRLS 2006 International Report and Grade Four learner overall performance in particular. South Africa’s overall reading achievement compared internationally A total of forty countries and forty-"ve education systems participated in PIRLS 2006. !e International Association for the Evaluation of Educational Achievement (IEA) released the PIRLS 2006 international reading literacy achievement results on 28 November 2007 at Boston College in the United States of America. !e results provided overall reading averages achieved by each participating country. !rough the use of Item Response !eory (IRT) scaling, the PIRLS 2006 average is set at a "xed 500 points with a standard deviation of 100 points. Participants’ achievements are therefore placed rela- tive to the international mean of 500. Appendix A provides the distribution of reading achievement as taken from the PIRLS 2006 International Report (Mullis et al., 2007). Internationally, PIRLS 2006 required the assessment of learners who have had four years of schooling and for most countries this requirement translated to Grade Four learners. !e South African PIRLS 2006 study assessed this "rst population of Grade Four learners, but also included a second population of Grade Five learners as a national option included in the study. Appendix A indicates that South Africa achieved the lowest score of the forty-"ve partici- pating education systems. Appendix A also indicates that the international report only provides results for South Africa’s Grade Five population. With an 51Surette Van Staden and Sarah Howie average age of 11.9 years, the South African learner population was the oldest across all participating countries. Grade Four learners achieved on average 253 points (SE=4.6)1, while Grade Five learners achieved on average 302 (5.6). Average achievement for both these Grades is well below the "xed interna- tional average of 500 points. Closest to South Africa in reading achievement was Morocco, the only other African country that participated in PIRLS 2006, with a Grade Four average of 323 points (5.9). According to PIRLS 2006, South Africa had the highest infant mortality rate (53 per 1000 live births), the lowest life expectancy (46 years) and the highest learner-to-teacher ratio of all the participating education systems. In terms of budgetary expenditure on education as a percentage of the Gross Domestic Product (GDP), South Africa is ranked average in comparison to the expen- diture on education by other countries (Howie et al., 2007). South Africa’s overall achievement for Grade Four learners nationally by language !e PIRLS 2006 reading assessment was administered to a sample of 16,073 Grade Four learners in all eleven o$cial languages in South Africa. It has to be kept in mind that the results for each language are in terms of the lan- guage of the test, not home language. Children were tested in the language in which they had been receiving instruction for the "rst three years of formal schooling. Average achievement scores that are provided here per language are therefore for learners who completed the assessment in the language of the test (the language that should have coincided with the language in which the learner has been receiving instruction for the "rst three years of schooling), and may therefore be di%erent from the learner’s home language. Figure 2 indicates that the learners who wrote the PIRLS 2006 assessment in Afrikaans achieved the highest average score of 351.70 (12.04), closely followed by learners who completed the assessment in English (346.82, SE=17.46). Learners who completed the assessment in African languages achieved well below 300 points, with Setswana learners achieving the highest scores, while IsiNdebele and IsiXhosa learners achieved the lowest average scores at 176.80 (9.02) and 189.97 (6.50) respectively. All South African achievement scores were well below the international average of 500 points. 1 SE stands for Standard Error and refers to the extent of deviation of the sampling distribution from the standard. e.g., where SE is given as 4.6, the extent of deviation can be interpreted as small relative to the sample size. Note that the numbers in brackets that follow also refer to SE scores. !e initials SE are le& out a&er the "rst instance for convenience sake, in keeping with common practice in statistical reports. 52 Reading and writing Figure 2: Average achievement according to language for Grade Four Teacher characteristics, use of resources and instructional practices In light of the dismally low reading achievement scores for Grade Four learn- ers, further enquiry is needed into teacher characteristics, use of resources and instructional practices. !e teachers who taught the Grade Four South African sample students completed the PIRLS 2006 Teacher Questionnaire. !ese teachers had on average "&een years of teaching experience, and, more speci"cally, 6 years (0.4) of Grade Four teaching experience (Howie et al., 2007). In terms of gender, Grade Four learners were taught by a majority of female teachers at 84.19% with the remaining 15.81% of the sample made up of male teachers. Of concern is that there was a very small percentage of young, up-and-coming teachers. Only 1% (0.6) of learners were taught by teachers who were younger than twenty-"ve years old, with only another 4% (1.4) being between the ages of twenty-"ve to twenty-nine years (see Table 1 below). 53Surette Van Staden and Sarah Howie While the number of students studying to be teachers is fairly substantial at tertiary teacher education institutions across South Africa, the high incidence of teacher attrition rates is problematic. Some (o&en White) young teachers prefer to leave South Africa for more lucrative teaching positions overseas, or merely choose to leave the profession a&er only a short period of teaching. !e average achievements for Grade Four learners taught by teachers under the age of twenty-"ve years (461, 52.4) and teachers over the age of sixty years (432, 64.7) were higher than the average achievement obtained for any other groups of learners based on their teachers’ ages (Howie et al., 2007). !e two groups of youngest and oldest teachers, whose Grade Four learners achieved the highest average scores, represent the smallest percentages of Grade Four teachers for this sample, and this can be seen as a matter of concern. Larger percentages of teachers in the remaining age groups achieved the lowest aver- age achievement on the PIRLS 2006 assessment. Table 1 provides a breakdown of average achievement by Grade Four learners on the PIRLS 2006 reading assessment categorising the age of their teachers: Teachers Age N % SE Average PIRLS 2006 Reading Achievement Under 25 94 0.93 0.56 461.03 25-29 years 291 3.67 1.40 261.32 30-39 years 5537 36.68 2.90 237.11 40-49 years 6792 41.15 2.86 255.09 50-59 years 2420 17.47 2.33 255.75 60-69 years and older 77 0.11 0.09 432.07 Table 1: Average PIRLS 2006 achievement for teachers’ age groups Teachers’ formal education and training !e highest percentage of South African learners (41%, 3.7) was taught by teachers with a three-year college diploma. Of these learners, 60% (3.2) were taught by teachers with a teaching certi"cation from the former teacher train- ing colleges or were in possession of a post-matric certi"cate (Howie et al., 2007). A little over 14% of learners had language teachers who reported to have postgraduate degrees – these learners achieved a higher average compared to those learners whose teachers had no postgraduate quali"cations. When asked about knowledge domains or subject areas of specialization, teachers of Grade Four learners reported exposure to training that focused 54 Reading and writing on language, literature, pedagogy, teaching reading, psychology, children’s language development and second language learning. Approximately 50% (3.2) of South African learners had teachers who reportedly received training where second-language learning received major emphasis. Amongst special education and remedial reading learners, teachers reported to have little to no training in these areas. For 40% (3.2) of the teachers of Grade Four learners remedial reading was only covered in an introductory fashion during training, while an additional 39% of teachers reported to have had no exposure to remedial reading at all. Given the students’ low achievement in reading literacy and the extent of learner variation at classroom level, the dearth of remedial reading training received by teachers is a cause of concern. Instructional materials Amongst the teachers of Grade Four learners, 57% (2.6) reported using text- books everyday or almost everyday. Only 12% (1.8) of Grade Four learners had teachers who reported using a variety of children’s books for reading instruc- tion every day or almost every day (Howie et al., 2007). A notable anomaly was the higher average achievement of learners whose teachers reported to never using textbooks (375.32, 45.9) compared to their counterparts who reported monthly (339.16, 37.53) or daily use of textbooks (235, 5.2). !e reliance on textbooks by a large percentage of teachers of Grade Four learners must be seen within the context of teaching in many schools in South Africa. Not only are textbooks o&en the only source available to the teacher as an aide to teaching reading, but the quality of these books is sometimes a concern. Anecdotal evidence gathered speci"cally during school visits in rural areas points to the fact that many outdated Afrikaans and English textbooks have been handed down to rural schools. In some cases, these books can be found unused on shelves, but equally alarmingly, cases exist where these out- dated books are being put to use. In addition, teachers o&en rely on preserving limited resources of textbooks to the extent that learners are not permitted to take these books home for fear of damage or loss. !us, learners’ only exposure to books in poorly resourced schools is o&en only in the form of textbooks, and then only for the limited time the learner is present in class. Instructional strategies and activities Most children in South African schools are reportedly exposed to certain read- ing skills only, during the Foundation Phase (Grades 1 to 3), such as decoding strategies and understanding vocabulary. According to Pretorius (2002), the 55Surette Van Staden and Sarah Howie Intermediate Phase (Grades Four to Six) a%ords learners the opportunity to use reading as a language- and information-processing skill, as learners are largely expected to be able to decode text. At Grade Four, learners should also begin the switch from learning the ‘lower level’ skills involved in learning to read, and to adapting those skills in order to use reading as a tool to learn. Twenty "ve per cent (2.4) of Grade Four learners in the PIRLS 2006 report have teachers who engage them daily in decoding strategies, compared to 69% (0.5) of teachers of Grade Four learners internationally. In terms of time allocated to reading activities in the classroom, more than half the Grade Four learners (61%, 2.9) had teachers who reported reading aloud to the whole class every day or almost every day. For these learners, this reading is the most prominent activity listed by teachers. Of concern is the fact that reading aloud to the class is a teacher-centred approach where learners are only involved passively and where the teacher mostly assumes that learners are able to follow and able to understand what is being read. Also of concern are the low frequencies at which learners are a%orded the opportunity to read inde- pendently. Indeed, 7% (1.7) of teachers of Grade Four learners indicated that the learners never or almost never engaged in independent reading in class, with only 27% (2.7) reporting reading independently as little as once or twice a month (Howie et al., 2007). Learner achievement was highest (299.3, 12.5) for those learners’ teachers who reportedly a%orded learners the opportunity to read silently everyday or almost everyday. Given the lack of opportunity a%orded to Grade Four learners to read independently, it would be under- standable that the format of the PIRLS 2006 reading assessment (consisting of reading booklets composed of reading passages for each child individually) might have been an intimidating and alien experience for many South African Grade Four learners. Instructional time Despite South African Grade Four learners’ poor performance in the PIRLS 2006 reading assessment, the majority of teachers of Grade Four learners (50.37%, 3.06) indicated that they regarded the reading levels of their learn- ers to be average. Only 6.04% (1.72) of teachers reported reading levels to be below average. !e PIRLS 2006 international report indicates that on average internationally teachers allocated 30% of instructional time to language instruction and 20% to reading instruction (Mullis et al., 2007). On average, internationally, Grade Four learners were taught explicit reading instruction for more than 6 hours a week. Teachers’ reports in the South African study reveal that 10% (1.9) of 56 Reading and writing Grade Four learners received reading instruction for more than 6 hours per week, 18% (2.7) for between 3 and 6 hours and 72% (2.7) for less than 3 hours per week. From these data, it becomes clear that South African reports fall far below the international averages in terms of time spent on reading instruction. A third of the learners’ teachers reported engaging in reading instruction every day, but little direct relationship can be found between reported time spent on reading instruction and reading achievement. A multitude of variables in#u- ence learners’ reading achievement and time spent on reading instruction is not necessarily a stand-alone indicator of the quality of the activities in which learners are engaged. Class size and instructional activities !e average Grade Four class size included in the South African PIRLS 2006 study was 46 (0.12) learners, the highest of the participating countries and higher than the international average class size of 24 learners. Table 2 indi- cates South African class sizes in comparison to overall average reading achievement: Class Size % SE Average Achievement 1 - 10 Learners 0.14 00 165.57 11-20 Learners 2.89 0.86 277.26 21-30 Learners 13.68 1.99 351.13 31-40 Learners 27.41 2.52 258.92 41-50 Learners 32.87 3.13 235.96 51-60 Learners 15.26 1.89 219.30 61-70 Learners 3.68 1.13 213.96 71-80 Learners 1.20 0.62 176.90 81 Learners and above 2.89 1.05 186.39 Table 2: Overall average achievement in comparison to class size As indicated by Table 2, the highest overall average achievement was obtained by those Grade Four learners in classes with between 21 and 30 students. With more than 40 students per class, achievement drops markedly to as low as 186.39 for learners in classes of extreme sizes of 81 or more children. In terms of reading activities in the classroom, teachers of Grade Four learners reported most commonly reading aloud to the class when teaching reading 57Surette Van Staden and Sarah Howie instruction or engaging with the learners in reading activities. Figure 3 shows the percentages of activities teachers reported to engage with learners every day: Figure 3: Teacher engagement in everyday reading activities Figure 3 supports "ndings discussed below concerning teachers’ instructional strategies and activities, and illustrates that teachers reportedly spent most time in reading aloud to the class themselves (61%, 2.9). Alarmingly, silent reading took place least frequently of all the listed everyday reading activi- ties in this sample of South African classrooms. Top performing countries in PIRLS 2006 report much higher frequencies of silent reading activities, with 84% (2.9) of teachers from the Russian Federation and 85% (3.2) of teachers from Canada, Alberta, reporting silent reading as a daily activity. Conclusions and implications South African Grade Four learner-achievement in the PIRLS 2006 study accentuates the need for reading instruction practices aimed at addressing the di$culties South African learners encounter in both the Foundation and Intermediate Phases. As stated in the International Reading Association’s (2007: 1) synthesis on their research into teacher preparation for reading instruction “putting a quality teacher in every classroom is key to addressing the challenges of reading achievement in schools. Knowledgeable, strategic, adaptive, and re#ective teachers make a di%erence in student learning.” 58 Reading and writing !e teacher data presented in this paper indicates the need for Intermediate Phase teachers’ continuous professional development, as Grade Four learners’ low overall achievement scores, in relation to teacher quali"cations, suggests that these teachers have not been adequately prepared to teach reading literacy. Of concern is the high incidence of low achievement among learners who are taught by teachers aged between thirty and "&y-nine years. !e data highlights the need for strategies to retain the younger demographic of up-and-coming teachers, whilst tapping into the knowledge and experience of teachers closer to retirement and exiting from the system. !e data presented here leads to questions concerning the quality and avail- ability of reading materials and how these are used in the Intermediate Phase. Speci"cally, it seems that more investigation is needed into the quality and content of textbooks and the quality of teaching where textbooks are used as only or main source of reading instruction. While the e%ect of class sizes on reading achievement is illustrated by the PIRLS 2006 data, of concern is also the little time that is reportedly spent on explicit reading instruction. Compared internationally, South African Grade Four teachers spend on average far less time on frequent reading instruction. Coupled with this low frequency of time on task is also the relative low quality of associated reading activities. While the majority of teachers report to spend- ing a lot of time reading out aloud to the class as a whole, the nature of this activity is not an e%ective gauge of the levels of understanding of what is being read, nor does it e%ectively allow for learners to become independent readers. While more interrogation into the PIRLS 2006 achievement data and ques- tionnaire data is required, this paper provides preliminary evidence of teacher reports about aspects such as the use and availability of resources, and the nature of reading activities and instructional practices that most o&en tend to take place in Grade Four classrooms around South Africa. !e inadequa- cies of Intermediate Phase teacher preparation and lack of explicit time spent on reading instruction may well re#ect inadequacies at Foundation Phase and perhaps even at the pre-primary stage, as well as the lack of pre-primary education, in many cases. References Bloch, C. 1999. Literacy in the early years: Teaching and learning in multilingual early child- hood classrooms. International Journal of Early Years Education, 7(1): 39-51. Department of Education. 2002. Revised National Curriculum Statement, Grades R-9. Government Gazette, 443, 23406, May. 59Surette Van Staden and Sarah Howie Department of Education. 2008. National Reading Strategy. Pretoria: Department of Education. Fleish, B. 2007. Primary education in crisis: Why South African schoolchildren under- achieve in reading and mathematics. Cape Town: Juta and Company. Howie, S.J. 2001. Mathematics and Science Performance in Grade 8 in South Africa 1998/1999: TIMSS R. HSRC Report: Pretoria. Howie, S.J. 2006. Multi-level factors a%ecting the achievement of South African pupils in mathematics. In: Howie, S.J. and Plomp, T. (eds). Contexts of Learning and Science. Oxon: Routledge, 71-9. Howie, S.J., Venter, E., van Staden, S., Zimmerman, L., Long, C., Scherman, V. and Archer, E. 2007. PIRLS 2006 Summary Report: South African Children’s Reading Literacy Achie- vement. Pretoria: Centre or Evaluation and Assessment. International Reading Association (IRA). 2007. Teaching Reading well. A synthesis of the International Reading Association’s research on teacher preparation for reading instruction. Newark, DE: International Reading Association. Downloaded in March 2008 from www.reading.org. Johnson, S., Monk, M. and Hodges, M. 2000. Teacher Development and Change in South Africa: A Critique of the Appropriateness of Transfer of Northern/Western Practice. Compare: A Journal of Comparative Education, 30(2): 179-192. Lessing, A.C. and de Witt, M.W. 2005. An investigation into the early literacy skills of Grade R second-language (L2) learners in South Africa. Africa Education Review, 2(2): 242-257. Matjila, D.S. and Pretorius, E.J. 2004. Bilingual and biliterate? An exploratory study of grade 8 reading skills in Setswana and English. Per Linguam, 20(1): 1-21. Moats, L.C. 1999. Teaching reading is rocket science. What expert teachers of reading should know and be able to do. Report for the American Federation of Teachers. Item no. 39-0372. Downloaded in April 2007 from http://www.a&.org/pubs-reports/ downloads/teachers/rocketsci.pdf. Mullis, I.V.S., Martin, M.O., Kennedy, A.M., and Foy, P. 2007. PIRLS 2006 International Report: IEA’s Study of Reading Literacy Achievement in Primary Schools. Chestnut Hill: Boston College. Pretorius, E.J. 2002. Reading ability and academic achievement in South Africa: Are we "ddling while Rome is burning? Language Matters, 33: 169-196. Pretorius, E.J. and Ribbens, R. 2005. Reading in a disadvantaged high school: Issues of accomplishment, assessment and accountability. South African Journal of Education, 25(3): 139-147. Sailors, M., Ho%man, J.V. and Mathee, B. 2007. South African schools that promote literacy learning with students from low-income communities. Reading Research Quarterly, 42(3): 364-387. Stoller, F.L. and Grabe, W. 2001. Action research as re#ective teacher practice in the context of L2 reading classrooms. Tydskrif vir Taalonderrig, 35(2&3): 97-109. Zimmerman, L. 2008. Unpublished PhD proposal. PhD Assessment and Quality Assurance, Faculty of Education, University of Pretoria. 60 Reading and writing Appendix A