Data-based instructional decision making related to basic early literacy skills in the Intermediate Phase

(1)

Data-based instructional decision

making related to basic early

literacy

skills

in the Intermediate Phase

M O'Connor

20438184

Thesis submitted for the degree Doctor Philosophiae in

Learner Support at the Potchefstroom Campus of the

North-West University

Pro motor:

May 2017

It all starts here TM

Prof C Nel

• NORTH-WEST UNIVERSITY ®

YUNIBESITI YA BOKONE·BOPHIRIMA NOORDWES·UNIVERSITEIT

(2)

(3)

Acknowledgements i Acknowledgements

I would like to give all the glory and the honour of this study to my Saviour. Nothing is impossible with God by your side. Thank you Lord that this never ending, seemingly impossible study has been completed. All I am is because of You. Without You Lord, none of this would have been possible. Thank you for never leaving me and loving me beyond my understanding. Thank you for giving me the opportunity to honour You through my studies. Help me to never discard what you have entrusted to me. Thank you for making me a lifelong learner and for placing me in a working environment where I can blossom, learn and enjoy every day.

I would like to express my deep indebtedness to my promoter, Prof. Carisma Nel. Thank you for being the best mentor, academic, role model, confidant, advisor and study leader any student could ever ask for. Your exceptional guidance, support, friendship and enthusiasm were critical for my studies and it will inspire me throughout my life. Your patience, flexibility, genuine caring and concern, and faith in me during this thesis process enabled me to attend to life while also working towards earning my PhD. You have motivated and encouraged me on my PhD journey. I am also thankful for your excellent example as a successful academic, woman, mother and professor. You are truly one of the greatest leaders and brightest minds in the academic field.

I would like to thank my husband, soul mate and best friend, Donald William O‘Connor, without whom this study would also not have been possible. Thank you for your endless love, support, help and encouragement. This is your study as much as mine and I am forever in your debt. I love you more than life itself and you are still my reason for waking up every day. Thank you for picking me up when I did not have the courage to continue. Thank you for loving me the way you do and for your continued encouragement especially during the time of the study. Thank you that you believed in me so much and for pulling me through this.

(4)

Acknowledgements ii

I would like to thank my family, especially my father Michael Britz, my mother Joey Britz, and my sister Yolandé van Eck. You have instilled many admirable qualities in me and given me a good foundation with which to meet life. You have taught me about hard work and self-respect and persistence. Thank you for always believing in me and encouraging me to follow my dreams and that you supported and helped me during every stage of my personal and academic life, and longed to see this achievement come true.

I would like to thank all my family and friends, especially my parents-in-law, for all their help, love and encouragement. My deepest appreciation is expressed to them for their love, understanding and inspiration.

(5)

Summary iii

Summary

Arising from the increasing demands of the twenty-first century workplace, concern over learner reading performance is at the forefront of national education. The increasing demands have raised the literacy bar for learners and subsequently, schools have been forced to accommodate instruction for these increased expectations. As teachers increasingly are held responsible for learner achievement, school teachers struggle to find ways to effectively document learner responsiveness to interventions and track progress towards important reading literacy outcomes. Data-based instructional decision making pertains to the systematic collection, analysis, examination, and interpretation of data to inform practice and policy in education settings. Effective use of assessment data to plan, judge, and modify teaching is a fundamental competency for good teaching. Teachers need to actively use the information collected via the assessments to critically evaluate their teaching in order to determine how it could be changed to better meet the learner‘s needs. Generally, schools collect enormous amounts of data on learners‘ attendance, behaviour, and performance. But when it comes to improving teaching and learning, it‘s not the quantity of the data that counts, but how the information is used.

The purpose of the study was to determine what assessment data a district (i.e., Fairy Tale district) requires schools (i.e., sample namely, Grade 4 in the Intermediate Phase) to submit; what a district expects from schools and/or teachers in terms of assessment data analysis, interpretation and instructional decision making; what assessments teachers use to assess basic early literacy skills such as oral reading fluency and reading comprehension; how teachers assess basic early literacy skills; what instructional decisions teachers make related to their learners‘ basic early literacy skills assessment results; and how the implementation of a data-based instructional decision making model (i.e., Outcomes-driven model) affected teachers‘ instructional decision making as related to basic early literacy skills.

This study was conducted with an interpretive research paradigm. A proactive action research design was chosen for this study. The aim was to collaborate with a subject

(6)

Summary iv

specialist (district level), school management team (school level) and two Grade 4 teachers (classroom level), in order to obtain an in depth understanding of assessment practice in general, and specifically the assessment support needs of teachers and learners.

Overall, the results indicated that use of data is part of an ongoing cycle of instructional improvement. After collecting multiple sources of data about learner learning, teachers should interpret data and develop hypotheses about how to improve learner learning then modifying instruction to test those hypotheses. The results also clearly indicated that data in a district and school context isn‘t powerful or valuable until tools, processes, and training or supports are put in place for users to accurately understand it and put it into action.

In this study, I developed a system-wide basic early literacy skill assessment and instructional support framework for the Intermediate Phase. The framework is intended to ensure that all partners in education accept collective responsibility for ensuring learner reading achievement by addressing data-based instructional decision making at all levels of the educational system related to reading literacy assessment practices and data-based instructional practices. All components of a coherent system are aligned with the key goals for learner‘s learning. A comprehensive assessment system addresses a full range of knowledge and skills expected by the curriculum (i.e., CAPS). It provides different users at different levels in the system (district, school, and classroom) with the right kinds of data, at the right level of detail, to help with decision making. A system that is continuous provides on-going streams of information about learner‘s learning throughout the year.

Key words: accurate and fluent reading, reading comprehension, vocabulary and

language skills, assessment, progress monitoring assessment, instructional decision making, data-literacy, data-driven decision making, data-driven instructional practices, assessment literacy, data-based instructional decision making.

(7)

Opsomming v

Opsomming

Voorspruitend uit die toenemende eise van die hedendaagse beroepsomgewing, ontstaan kommer oor leerders se leesprestasie. Die toenemende eise het die geletterdheidsvlak vir leerders gelig; derhalwe word skole genoodsaak om onderrig in ooreenstemming te bring met hierdie toenemende gelettersheidsverwagtinge. Onderwysers word toenemend verantwoordelik gehou vir leerderprestasie, en soek voortdurend nuwe maniere om leerders se reaksie op intervensie te dokumenteer en na te speur binne die uitkomste van leesonderrig. Datagebaseerde onderrig besluitneming verwys na die sistematiese insameling, ontleding, ondersoek en interpretasie van data om die praktyk en beleid toe te lig in die onderwysinstellings. Die doeltreffende gebruik van assesseringsdata om te beplan, beoordeel en onderrig aan te pas is ‗n fundamentele bekwaamheid wat onderwysers moet bemeester vir goeie onderrigpraktyk. Onderwysers moet die inligting aktief gebruik om hulle onderrig krities te evalueer ten einde te bepaal hoe dit aangepas kan word om te voldoen aan die leerders se behoeftes. Groot hoeveelhede inligting rakende bywoning, gedrag en akademiese vordering is daagliks in skole beskikbaar, maar vir die verbetering van onderrig en leer is dit nie noodwendig die hoeveelheid inligting wat van belang is nie, maar wel hoe die inligting gebruik word.

Die doel met hierdie studie was om te bepaal watter assesseringsdata die distrik (naamlik die Fairy Tale distrik) van die skole (steekproef, naamlik Graad 4 in die Intermediëre Fase) vereis om in te dien; wat die distrik vanaf die skool sowel as onderwysers verwag in terme van assessering, data-analise, interpretasie en onderrigbesluitneming; watter assesseringstegnieke onderwysers gebruik om basiese vroeë geletterdheidsvaardighede soos mondelingse vlotlees en leesbegrip te assesseer; watter onderrigbesluite onderwysers neem met behulp van basiese vroeë geletterdheidsvaardighede assesseringsresultate; en hoe die implementering van 'n data-gebaseerde onderrig besluitnemingsmodel (bv. Uitkomsgebaseerde gedrewe model) onderwysers se besluitneming affekteer met betrekking tot die basiese vroeë geletterdheidsvaardighede.

(8)

Opsomming vi

Hierdie studie was uitgevoer met 'n interpretiewe navorsingsparadigma. 'n Pro-aktiewe aksie-navorsingsontwerp was geselekteer vir hierdie studie. In hierdie studie was die doel om met ‗n vakspesialis (distrikvlak), skoolbestuurspan (skoolvlak) en twee Graad 4-onderwysers (klaskamervlak) saam te werk ten einde ‘n grondige begrip te verkry van assesseringspraktyke in die algemeen, en spesifiek die assesseringsondersteuningsbehoeftes van onderwysers en leerders.

Die resultate het aangetoon dat die gebruik van data deel is van ‗n voortdurende siklus van onderrigbevordering. Na die insameling van verskeie bronne van inligting oor leerders se onderrig en leer, moet onderwysers data interpreteer en ‗n hipotese ontwikkel oor hoe leerders se onderrig verbeter kan word en onderrig dienooreenkomstig verander om sodoende die hipotese te toets. Die resultate het ook duidelik getoon dat die data in die distrik- en skoolkonteks nie van waarde is, totdat opleiding of ondersteuning in plek is om die data (interpretasie) suksesvol uit te voer nie.

In hierdie studie het ek ‗n stelselwye vroeë basiese geletterdheidsvaardighede assessering-en-onderrigondersteuningraamwerk vir die Intermediëre Fase ontwikkel. Die raamwerk is daarop gerig dat alle rolspelers in die onderwys verantwoordelikheid moet aanvaar vir die leerder se leesprestasie deur data-gebaseerde onderrigbesluitneming op alle vlakke van die onderwysstelsel te verseker wat verband hou met leesgeletterdheid assessering- en datagebaseerde onderrigpraktyke. Alle komponente van 'n samehangende sisteem is in lyn met die doelwitte vir die onderrig van leerders. ‘n Omvattende evalueringstelsel spreek die omvang van kennis en vaardighede binne die kurrikulum (bv. KABV) se verwagting aan. Dit voorsien verskillende opvoedkundige rolspelers op verskillende vlakke in die stelsel (distrik, skool, en klaskamer) met die regte soorte data op die regte vlak aan, om te help met die besluitnemingsproses. Die stelsel verseker dat inligting ten opsigte van leerders se vordering deurlopend beskikbaar is.

Sleutelwoorde: akkurate vlotlees, leesbegrip, woordeskat en taalvaardighede,

assessering, deurlopende assessering, onderrigbesluitneming, data-geletterdheid, data-gedrewe besluitneming, data-gedrewe onderrigpraktyke, assesseringsgeletterdheid, datagebaseerde onderrigbesluitneming.

(9)

Table of contents vii Table of contents Acknowledgements……… Summary………. Opsomming……….. Table of contents………. List of Tables……… List of Figures………..

Chapter 1: Introduction and problem statement………

1.1 Introduction………. 1.2 Key words and clarification……….. 1.3 Research problem and motivation for study………..………... 1.4 Purpose of the study………... 1.5 Theoretical perspective and literature review………... 1.6 Methodology………... 1.6.1 Literature review……… 1.6.2 Empirical investigation……….. 1.6.2.1 Research paradigm……… 1.6.2.2 Research approach……… 1.6.2.3 Research design……… 1.6.2.4 Sampling……….. 1.6.2.5 Data collection methods……… 1.6.2.5.1 Documents………. 1.6.2.5.2 Semi-structured interviews……….. 1.6.2.5.3 Observations……….. 1.6.2.5.4 Dibels Next assessments………... 1.6.2.5.5 Focus group interview……….. 1.6.2.6 Data collection procedure……..………. 1.6.2.7 Data analysis……….. 1.7 The role of the researcher………

i iii v vii xiii xv 1 1 1 4 9 10 19 19 19 19 20 22 23 23 23 24 25 26 26 27 27 28

(10)

Table of contents viii

1.8 Trustworthiness………. 1.9 Ethical aspects………... 1.10 Chapter division……….

Chapter 2: Theoretical perspective: Simple view of reading and the basic early literacy skills………...

2.1 Introduction……….… 2.2 Theoretical frameworks...

2.2.1 The simple view of reading... 2.2.2 The four part processing model... 2.2.3 Scarborough‘s rope model... 2.3 Basic early literacy skills...

2.3.1 Phonemic awareness……… 2.3.2 Phonics... 2.3.3 Fluency... 2.3.4 Vocabulary... 2.3.5 Comprehension... 2.4 Alignment of Dibels Next measures with the basic early literacy

skills assessment process... 2.5 Conclusion...

Chapter 3: Data-based instructional decision making………

3.1 Introduction... 3.2 Perspectives on data and assessment... 3.3 Use of data in education... 3.3.1 Curriculum... 3.3.2 Teacher improvement and teaching strategies... 3.3.3 School improvement...

3.3.3.1 Analysing data to determine the relationship between data and school improvement... 3.3.4 Use of data to inform instruction... 3.3.5 Professional knowledge and skills required...

29 29 30 31 31 32 32 41 47 52 52 54 56 58 60 63 66 68 68 69 72 72 73 74 75 76 77

(11)

Table of contents ix

3.3.6 Improving teaching and learning... 3.4 Types and purpose of assessment... 3.4.1 Screening assessment... 3.4.2 Diagnostic assessment... 3.4.3 Progress monitoring assessment... 3.4.4 Outcomes-based assessment... 3.4.5 Informal assessment... 3.5 Frameworks supporting data-based instructional decision making... 3.6 Factors that influence data use...

3.6.1 Context... 3.6.2 Collaboration... 3.6.3 Teachers‘ attitude... 3.6.4 Information management systems... 3.6.5 Leadership... 3.6.6 Teacher preparation... 3.7 Conclusion...

Chapter 4: Research methodology and design………..

4.1 Introduction... 4.2 Methodology... 4.2.1 Literature review... 4.2.2 Empirical investigation... 4.2.2.1 Research paradigm... 4.2.2.2 Research approach... 4.2.2.3 Research design...

4.2.2.3.1 Rationale for action research in this study... 4.2.2.4 Sampling... 4.2.2.5 Data collection methods... 4.2.2.5.1 Documents... 4.2.2.5.2 Semi-structured interviews... 4.2.2.5.2.1 Interview procedure... 4.2.2.5.3 Observations... 79 80 81 82 84 85 86 87 95 95 96 97 97 98 99 101 103 103 104 104 106 106 108 109 112 113 115 115 117 118 119

(12)

Table of contents x

4.2.2.5.4 Dibels assessments... 4.2.2.5.4.1 Dibels Oral Reading Fluency... 4.2.2.5.4.1.1 Administration directions... 4.2.2.5.4.2 DAZE reading comprehension...

4.2.2.5.4.2.1 Administration directions... 4.2.2.5.5 Focus group interview... 4.2.2.6 Data collection procedure... 4.2.2.7 Data analysis... 4.2.2.7.1 Qualitative content analysis... 4.2.2.7.2 Focus group analysis... 4.2.2.7.3 Observation analysis... 4.2.2.7.4 Dibels assessment analysis... 4.2.2.8 Data organisation and security... 4.3 The role of the researcher... 4.4 Trustworthiness... 4.5 Ethical aspects... 4.6 Conclusion...

Chapter 5: Results and discussion………

5.1 Introduction... 5.2 Cycle 1: List hopes and concerns... 5.2.1 District level... 5.2.2 School level... 5.2.3 Classroom level... 5.2.4 Documentation... 5.2.5 My reflection... 5.3 Try a new practice... 5.3.1 Training... 5.4 Collect data... 5.5 Check what the data mean... 5.6 Reflect on alternative or refined practices... 5.7 Fine-tune practice and start of cycle 2...

119 120 122 125 125 128 130 131 132 136 140 140 140 141 142 144 147 148 148 149 150 154 156 159 183 183 194 195 196 207 209

(13)

Table of contents xi

5.8 Cycle 2: Collect data and check what the data mean... 5.8.1 Dibels Next assessment administration... 5.9 Reflect on practice... 5.10 Fine-tune the assessment practice... 5.11 Conclusion...

Chapter 6: System-wide basic early literacy skill assessment and instructional support framework for the Intermediate Phase……….

6.1 Introduction... 6.2 Data-based instructional decision making framework for reading literacy...

6.2.1 Three essential cogs……….……….……. 6.2.2 Outcomes-driven decision making component………...…… 6.2.2.1 Identify need for support………. 6.2.2.2 Validate need for support………... 6.2.2.3 Plan and implement support……….. 6.2.2.4 Evaluate and modify support……….…… 6.2.2.5 Review outcomes………..….. 6.2.2.6 Communicating assessment results………. 6.3 Conclusion...

Chapter 7: Conclusion and recommendations for further research………

7.1 Introduction……….... 7.2 Literature review……… 7.2.1 Reading frameworks….……….………..……….. 7.2.2 Data-based instructional decision making as related to the basic

early literacy skills …..……… 7.3 Summary of research results……….. 7.3.1 District expectations………..………..… 7.3.2 Assessment and instructional decision making by teachers………….... 7.3.3 A data-based instructional decision making model………..…. 7.4 Contribution of the study……….. 7.5 Limitation of the study………...

209 209 222 223 224 225 225 227 227 233 240 245 248 254 256 258 259 261 261 262 263 265 267 267 269 270 271 271

(14)

Table of contents xii

7.6 Implications of results for teacher professional development……… 7.7 Recommendations for future research……….. 7.8 Conclusion………..

Bibliography………. Appendices1

1_{The full appendices can be found on the disk that is included.}

272 274 276

(15)

List of Tables xiii

List of Tables

Table 4.1: Overview of Dibels Oral Reading Fluency……….

Table 4.2: Overview of Dibels DAZE………..

Table 4.3: Matrix for assessing level of concensus in focus group…………....…….

Table 4.4: Matrix for documenting proxemic, chronemic, kinesic,

and paralinguistic information………..

Table 5.1: Focus group data………..

Table 5.2: Classroom report: Grade 4A………..

Table 5.3: Classroom report: Grade 4B………

Table 5.4: School (Grade 4 A) overview report………..….

Table 5.5: School (Grade 4 B) overview report………..

Table 5.6: Grade 3 DORF benchmark goals and cut points for risk………..….

Table 5.7: Grade 3 DAZE benchmark goals and cut points for risk……….

Table 5.8: Outcomes-driven model questions and data………

Table 5.9: Grade 4A school overview report……….…..

Table 5.10: Grade 4B school overview report………

Table 5.11: Various graphs……….. 121 125 138 139 185 197 198 199 200 202 203 211 213 214 215

(16)

List of Tables xiv

Table 5.12: Grade 4A classroom report – middle of the year………

Table 5.13: Grade 4B classroom report – middle of the year………

Table 5.14: Grouping suggestions………...

Table 6.1: Guidelines for planning, decision making supporting reading literacy...

Table 6.2: Patterns of reading difficulties and possible intervention needs………… 219

220

221

235

(17)

List of Figures xv

List of Figures

Figure 1.1: Simple view of reading……….

Figure 2.1: Four processing systems that support word recognition………

Figure 2.2: Four part processing system……….

Figure 2.3: Scarborough‘s rope model……….

Figure 2.4: Model of basic early literacy skills………

Figure 4.1: Reflective spiral……….

Figure 4.2: Proactive action research steps………

Figure 4.3: Dibels Oral Reading Fluency………..………...

Figure 4.4: Dibels DAZE……….………..

Figure 5.1: Outcomes-driven model steps……….………...

Figure 6.1: Data-based instructional decision making framework..………...

Figure 6.2: School overview report: An example..………

Figure 6.3: Classroom report……….………..

Figure 6.4: Benchmark guidelines of Dibels Next assessments..………..

Figure 6.5: Decision-making model for Grade R and 1……… 12 42 43 48 65 110 112 124 127 210 228 241 242 244 246

(18)

List of Figures xvi

Figure 6.6: Decision-making model for Grade 2-3……….

Figure 6.7: An Intermediate Phase literacy diet……….

Figure 6.8: Reading literacy dietary supplements………..

Figure 6.9: Progress monitoring report……….

Figure 6.10: School overview report……… 247

253

254

255

(19)

Chapter 1: Introduction and problem statement 1

Chapter 1

Introduction and problem statement 1.1 Introduction

Data-based instructional decision making is not a new concept in education. Teachers have moved towards a culture of being data-driven, and have declared data use in schools to be significant for school improvement and accountability. Data-based instructional decision making pertains to the systematic collection, analysis, examination, and interpretation of data to inform practice and policy in education settings (Mandinach, 2012:71; Shen & Cooley, 2008). The South African Department of Education uses variable assessment data to evaluate the effectiveness of the educational system, educational districts use assessment data to monitor the success of the implementation of CAPS and classroom teachers use assessment data to determine learners‘ strengths and weaknesses in particular basic early literacy skills (i.e., phonemic awareness, phonics, fluency, vocabulary and reading comprehension). Yet, as school districts make great strides in creating a culture of data-based instructional decision making – collecting, analysing and interpreting data – little is known about how individual teachers make sense of data and how they use the data to form instruction. Data use can lead to school improvement in terms of increased learner achievement primarily when it can influence teaching in a meaningful manner (Carlson, Borman & Robinson, 2011; McNaughton, Lai & Hsiao, 2012). The purpose of this chapter is to discuss the research problem and motivation for study (i.e., data-based instructional decision making), to give an overview of the research methodology and design that is used in this study.

1.2 Key words and clarification

Learners‘ ability in the basic early literacy skills can distinguish between successful readers and those who are likely to struggle. Evidence shows that these skills are

(20)

the core components that every learner must master in order to become a proficient reader (National Reading Panel, 2000a; RSA Department of Education, 2008:14).

These components are phonemic awareness, phonics (alphabetic principle and basic phonics as well as advanced phonics and word attack skills), accurate and fluent reading of connected text, reading comprehension, vocabulary and language skills (National Institute of Child Health and Human Development, 2000). The following is a description of the terms that are used in this study, along with a brief explanation of each.

Accurate and fluent reading of connected text are critical components of reading and allows meaning to be gained from text. To read and understand the text easily, words must be decoded effortlessly (Ehri, 1998). Reading fluency depends on word attack skills (National Reading Panel, 2000a) such as efficient and automatic decoding of regular and irregular words, and the use of expression and phrasing (prosody) while reading aloud (Dowhower, 1991; Schreiber, 1987). Oral reading fluency is the link between accurate, automatic, word-level decoding and reading comprehension.

Reading comprehension is the envisaged outcome of instruction in the basic early literacy skills. These specific skills include accurate and fluent reading, monitoring while reading, and the ability to use cognitive strategies flexibly to gain meaning from text (Goldman & Rakestraw, 2000; Pressley, 2000). To decode words in the text, reading comprehension requires access to linguistic knowledge about syntax, semantics, and word morphology (Catts & Kahmi, 1999; McGuinness, 2005).

Data-based instructional decision making: An ongoing process of analysing and evaluating information to inform important educational decisions and actions, specifically related to basic early literacy skills. Instructional decision making is a

systematic process of using learner achievement and other data to guide instructional decisions (Stecker & Hintze, 2006).

(21)

Dibels Next: Dynamic Indicators of Basic Early Literacy Skills (Dibels) is a set of measures used to assess early literacy and reading skills for learners from Grade 1 to Grade 6. Dibels Next can be used to: identify learners who may be at risk for reading difficulties; help teachers identify areas to target instructional support; to monitor at-risk learners while they receive additional, targeted instruction; and examine the effectiveness of a school‘s system of instructional supports. Dibels Next measures are brief, powerful indicators of foundational early literacy skills that: are quick and efficient to administer and score; serve as universal screening (or benchmark assessment) and progress monitoring; identify learners in need of intervention support; and evaluate the effectiveness of interventions/support.

Assessment literacy: ―The ability to understand the different purposes and types of assessment in order to select the most appropriate type of assessment to meet a specific purpose‖ (Ainsworth & Viegut, 2006:53).

Data-driven instructional practices: A term referring to collecting and using learners‘ learning data to plan lessons and assessments at the classroom level.

Data-driven decision making: ―The processes of selecting, analysing, and making meaning of learner performance data to inform instructional decisions‖ (Bettesworth, 2006:4).

Data-literacy: The process of knowing how to collect, access, link, manipulate, report, analyse, and critique data for an intended purpose (Earl & Katz, 2006).

Basic early literacy skills: In this study, basic early literacy skills refer to the essential skills that every learner must master to become a proficient reader (National Reading Panel, 2000). Basic early literacy skills are also known as ―core components of foundational skills‖ that help distinguish learners who are on track to become successful readers from learners who are likely to struggle (Good & Kaminski, 2011:2).

(22)

1.3 Research problem and motivation for study

A number of assessment studies in recent years have shown that the educational achievement of learners in South African schools is unacceptably poor (Systemic Evaluations; SACMEQ II; PIRLS; Annual National Assessments; RSA DoE, 2003; Moloi & Strauss, 2005; Mullis, Martin, Kennedy & Foy, 2007; RSA DoE, 2010b). The Department of Education‘s (since 2001, the Department of Basic Education) systematic evaluations, conducted in Grade 3 (in 2001) show low level of literacy among learners. Scores for the Grade 3 learners averaged 39% for reading comprehension (RSA DoE, 2003). The second cycle of systemic evaluations conducted in 2007 reveals only a limited change in learners‘ achievement, namely 36% for literacy (Buhlungu et al., 2007). The 2011 Annual National Assessment results indicate a 35% literacy rate for South African learners in Grade 3 and a 30% literacy rate for learners in the North West Province of South Africa (DBE, 2010).

The Committee on the Prevention of Reading Difficulties in Young Learners (Snow, Burns & Griffin, 1998) provided compelling evidence that learners who do not learn to read fluently and independently in the early grades have few opportunities to catch up and virtually no chance to surpass their peers who are reading on grade level by the end of Grade 3. ―Contrary to the popular theory that learning to read is natural and easy, learning to read is a complex linguistic achievement‖ (American Federation of Teachers, 1999:11). As evidence mounts that reading difficulties originate in large part from difficulties in developing phoneme awareness, phonics, spelling skills, reading fluency, and reading comprehension strategies (Snow et al., 1998; NICHD, 2000; Nel & Malda, 2011), the need for informed instruction for the millions of South African learners with insufficient reading skills is an increasingly urgent problem.

According to the International Reading Association (2003) position statement, Investment in Teacher Preparation in the United States, teacher education programmes should ensure that teachers, amongst other aspects, ―know how to assess the progress of every learner and change teaching when it is not working;

(23)

know how to communicate results of assessments to various stakeholders, especially parents‖. The RSA Department of Education (2010a:5) states that two of the purposes of the Annual National Assessments (ANA) are to ―provide teachers with essential data about the baseline Literacy/Language and Numeracy/Mathematics capabilities of learners at the beginning of each grade and thereby help them make informed decisions when planning the year‘s programme; provide parents with a better picture of the levels of learner performance in the school so that parents are better informed when they become involved in efforts to improve performance, for instance through decision-making in the School Governing Body and support to learners in the home‖. Assessment is an important part of successful teaching because instruction needs to be calibrated according to learners‘ knowledge, skills, and interests. It is essential that teachers ―administer timely and valid assessments to identify learners lagging behind and monitor progress‖ (Crawford & Torgesen, 2006:1). These assessments help increase the quality, consistency, and impact of teaching by focusing directly on those areas in which learners need specific assistance.

Increasingly, researchers are finding that classroom-based assessments are an effective and important part of being a successful reading teacher. Effective teachers constantly monitor each learner‘s reading skills and provide instructional scaffolding to help the learner move to the next stage. This same information is the foundation for communicating with parents about the learner‘s progress (Morrow, Tracey, Woo & Pressley, 1999; Pressley, 2002; Pressley, Wharton-McDonald, Raphael, Bogner & Roehrig, 2002). In addition, learners in classrooms that use classroom-based reading assessments have greater gains in achievement than those in classrooms that do not focus on classroom-based assessments (Ross, 2004; Stecker, Fuchs & Fuchs, 2005). According to Kanjee (2008a), there is a growing trend in South Africa towards the use of assessment to improve learning and also an increased focus on classroom assessment. However, Kanjee (2008a) mentions that there is limited guidance, support and information for teachers on ―how‖ to use assessment to improve learning. Kruger (2008) quotes the National Director of assessment of the FET Mr Poliah in Rapport, where he states that doubt exists about the effectiveness

(24)

and reliability of continuous assessment. Mr Poliah qualifies his statement by mentioning some challenges with regards to assessment in schools, namely:

 Teachers lack of training to implement assessment effectively.

 The allocation of marks differs from teacher to teacher.

 Assessment increases the workload on learners and teachers.

 Different schools follow different methods of assessment.

 Learners could be assisted in their assessment tasks by parents and friends.

The RSA DoHET (2011:53) states that one of the competencies that newly qualified teachers should have is the ability ―to assess learners in reliable and varied ways, as well as being able to use the results of assessment to improve teaching and learning‖.

Assessment that can be used to adapt teaching to meet learner needs is called formative assessment (Kaminski & Cummings, 2008). Because the primary purpose of formative assessment is to support learner learning, it may arguably be considered the most important assessment practice in which teachers engage. The RSA DoE (2010a:12) states that ―[D]ecisions and plans on what, when and how to teach must be informed by the evidence that comes out of the assessments, both school-based and ANA assessments‖. Effective use of assessment data to plan, judge, and modify teaching is a fundamental competency for good teaching (Hosp & Ardoin, 2008; Hosp, 2010). A reason for linking assessment and teaching is that teachers need to make screening, diagnostic, progress, and outcome decisions, and those decisions need to be accurate; if they are not, valuable teaching time could be lost using teaching strategies that do not address the learners‘ needs. When it comes to planning teaching practices for learners, the best way to maximise the accuracy of teachers‘ decisions is to base them on data (Shepard, Hammerness, Darling-Hammond & Rust, 2005). Research indicates that when teachers use assessment data to make their teaching decisions, learner performance increases (Fuchs & Fuchs, 1986; Black & Wiliam, 1998b; Wohlstetter, Datnow & Park, 2008). The learners whose teachers collect systematic progress-monitoring data, and use it

(25)

to make decisions, score on average one standard deviation higher than their peers whose teachers do not collect and use these data (Stecker & Fuchs, 2000). In addition, teachers using systematic progress-monitoring data more frequently make changes in their teaching for those learners who are experiencing difficulties (Fuchs, Fuchs, Hamlett & Stecker, 1991). Teachers need to actively use the information collected via the assessments to critically evaluate their teaching in order to determine how it could be changed to better meet the learner‘s needs (Fuchs et al., 1991). Generally, schools collect enormous amounts of data on learners‘ attendance, behaviour, and performance. But when it comes to improving teaching and learning, it‘s not the quantity of the data that counts, but how the information is used (Hamilton, Halverson, Jackson, Mandinach, Supovitz & Wayman, 2009). The learning-assessment process can be framed by three questions, namely: Where are you trying to go? Where are you now? How can you get there? (Atkin, Black & Coffey, 2001). Shepard et al. (2005:278) state that, ―By answering the assessment question 2 in relation to the instructional goal question 1, and specifically addressing what is needed to reach the goal question 3, the formative assessment process directly supports improvement‖.

After a learner‘s performance has been measured, a key component to making decisions about his/her performance and planning teaching is the teacher‘s ability to make comparisons to a standard for performance. Three ways of determining standards are typically used in education: normative, criterion, and ipsative (Hosp, 2010:5). Normative standards involve comparing a learner‘s performance on the assessment to that of other learners in a comparable peer group (e.g., learners in the same grade). Criterion standards involve comparing a learner‘s performance to an empirically derived level of proficiency. For example, the performance levels for the ANA‘s range from level 1 (0 to 34%) labelled as ―Not Achieved‖ to Level 4 (70% and above) labelled as ―Outstanding‖. Ipsative standards involve a learner‘s prior performance as the basis for comparison of his/her current performance. Ipsative standards are often considered when monitoring learner progress because the learner‘s current performance can be compared to prior performance as well as, later, to future performance.

(26)

The establishment of benchmark goals is a challenging, but important task. For teachers knowing which skill areas are crucial for early literacy is an important first step, but equally important is knowing how proficient learners are in these critical skills. An effective benchmark goal should be specific, measurable, ambitious, and target a critical indicator of learner performance (Fuchs et al., 1991). ―An indicator is a brief, efficient index that provides a fair degree of certainty about a larger, more complex system or process‖ (Dynamic Measurement Group, 2011:2). An indicator is not intended to be a comprehensive, in-depth assessment of each and every component of a basic early literacy skill. Instead, indicators, such as the Dibels, are designed to measure key components that are representative of that skill area, and predictive of overall reading competence (e.g., an indicator of accurate and fluent reading of connected text is oral reading fluency – correct words per minute and accuracy). A benchmark goal for oral reading fluency could be 47 words per minute correct at the end of Grade 1.

Barnett, Elliott, Graden, Ihlo, Macmann, Nantais and Prasse (2006) note the need for formative assessment tools that are linked with a well-defined, decision-making model such as the Outcomes-driven model (Kaminski & Good, 1998; Tilly, 2008). The Outcomes-driven model was developed to address specific questions within a prevention-oriented framework designed to pre-empt early reading difficulty and ensure step-by-step progress toward outcomes that will result in established, adequate reading achievement. The Outcomes-driven model accomplishes these goals ―though a set of five educational decisions: (1) identify need for support, (2) validate need for support, (3) plan support, (4) evaluate and modify support, and (5) review outcomes‖ (Kaminski & Cummings, 2008:3). In order for formative assessment tools to be used effectively to link assessment to teaching, they must ―(a) accurately identify risk early, (b) provide meaningful and important goals, (c) evaluate adequate progress toward those goals, and (d) provide a way to evaluate both the overall system of support as well as the learners‘ response to that support‖ (Kaminski & Cummings, 2008:5). The nature of the problem that inspired me to conduct this research was that most schools use the collection of data to satisfy administrative duties, rather than to assess and evaluate learner improvement.

(27)

The following research questions will be addressed in this study:

Primary research question:

How is data-based reading literacy instructional decision making approached in a district, school and by teachers in Grade 4 classrooms?

Secondary research questions:

1. What assessment data does the district require schools to submit?

2. What does the district expect from schools and/or teachers in terms of assessment data analysis, interpretation and instructional decision making? 3. What assessments do teachers use to assess basic early literacy skills such

as oral reading fluency and reading comprehension? 4. How do teachers assess basic early literacy skills?

5. What instructional decisions do teachers make related to their learners‘ basic early literacy skills assessment results?

6. How does the implementation of a data-based instructional decision making model (i.e., Outcomes-driven model) affect teachers‘ instructional decision making as related to basic early literacy skills?

1.4 Purpose of the study

The purpose of the study is to (determine):

How data-based reading literacy instructional decision making is approached in a district, school and by teachers in Grade 4 classrooms.

In addition, the purpose is to determine:

1. what assessment data does the district require schools to submit.

2. what does the district expect from schools and/or teachers in terms of assessment data analysis, interpretation and instructional decision making. 3. what assessments do teachers use to assess basic early literacy skills such

as oral reading fluency and reading comprehension. 4. how do teachers assess basic early literacy skills.

(28)

5. what instructional decisions do teachers make related to their learners‘ basic early literacy skills assessment results.

6. how does the implementation of a data-based instructional decision making model (i.e., Outcomes-driven model) affect teachers‘ instructional decision making as related to basic early literacy skills.

1.5 Theoretical perspective and literature review

The assessment of learners‘ early literacy skills, and in particular their reading skills, is a core task for teachers as assessment scores are used by teachers to determine which learners will receive intervention or instructional adaptation. The simple view of reading serves as the main theory-driven framework by which to interpret the assessment results collected from various assessment instruments (i.e., classroom assessment and Dibels). In chapter 2, two additional theoretical models are discussed that highlight the relationship between the core reading skills.

The simple view of reading as originally proposed by Gough and Tunmer (1986) and Hoover and Gough (1990) posited that both decoding (D) and the comprehension of language (LC) comprise reading comprehension (R), and can be represented using the formula R = D x LC. Hoover and Gough (1990:130) specifically defined decoding as ―simply efficient word recognition‖ and the comprehension of language, as ―the ability to take lexical information … and derive sentence and discourse interpretations‖. The simple view of reading is especially important to the field of literacy because word recognition and language comprehension are both ―teachable skills‖ that can lead to improvements in reading comprehension (Kirby & Savage, 2008:79).

The simple view of reading is a perspective of how a beginner reader develops into a proficient reader. Different methods of teaching are needed to develop skills, as a representation of reality that is simplified in order to allow for improving understanding of a complex phenomenon (Oakhill, Cain & Bryant, 2003). According to the simple view of reading, there are two sets of abilities that contribute to reading:

(29)

word recognition abilities (the ability to read and understand the words on the page) and language comprehension ability (the ability to understand language we hear and language we read). These two sets of abilities are seen as continuous dimensions: people can vary independently on each (Spooner, Baddeley & Gatharcole, 2004; Catts, Adlof & Weismer, 2006; Grigorenko, Klin & Volmar, 2003). It is a fully interactive model but one which delineates the two different dimensions of reading and thus better allows teachers to specify their teaching objectives and engage learners in relevant activities to foster development towards achieving those objectives: the separation is in the teacher‘s mind, for pedagogic purposes, not in the learner‘s mind.

The simple view of reading was presented using a standard formula to illustrate the model: RC = LC x D. Each variable can range from 0 to 1, a score of 1 being an indication of accomplishment. If either decoding or language comprehension receive a 0, reading comprehension will also be 0, regardless of the results on the other variable. These two aspects depend on each other. The simple view of reading implies that as decoding increases, approaching the score of 1, the importance of language comprehension in predicting reading comprehension will increase. If scores for either decoding or language comprehension are available, the third variable can be determined using the simple view of reading formula. This model can be illustrated in quadrants as shown in Figure 1.1 (Wyse & Jones, 2007:44-45).

(30)

Figure 1.1: Simple view of reading

(Stuart et al., 2008:62).

Teachers need to be clear about learners‘ performance and progress in each of the two dimensions of decoding and language comprehension. Learners‘ performance can be identified in the four quadrants. The four patterns of performance reflect the differences in the balance of decoding and language comprehension (Gough & Tunmer, 1986) abilities as both dimensions are continuous; learners can vary continuously on each. Thus, as teachers assess the learners in the two dimensions, it is important to identify the learners‘ particular needs for effective intervention.

Word recognition is depicted on the horizontal axis and language comprehension is depicted on the vertical axis. As Figure 1.1 shows, a reader‘s reading competence depends upon both of these variable skills. Fluent readers are shown in the upper right quadrant with good word recognition and language comprehension skills, while readers who struggle with word recognition but are good in language comprehension

(31)

are shown in the upper left quadrant. Learners with language comprehension difficulties fall in the lower half of the quadrants. Poor reading comprehension can occur either in combination with poor word recognition or when word recognition skills are well developed. If a learner cannot decode a word accurately, the learner cannot comprehend that word. Accurate decoding of words is necessary for access to meaning.

These two variables (i.e., word recognition and language comprehension) differ from good to poor in each quadrant. The simple view of reading model proposes that there are three categories of reading comprehension difficulties: poor in recognising printed words (decoding), poor in comprehending spoken language (language comprehension), or a combination of both (Tunmer & Greaney, 2010). Learners with poor word recognition ability may also have poor language comprehension skills. According to Catts et al. (2003), when there are deficits in reading comprehension, there are always deficits in either language comprehension or decoding skills. In some cases, there are learners whose reading comprehension suffers because while they are very good at understanding spoken language, they are poor at decoding text. There are other learners whose reading comprehension suffers because, while they are good at decoding text, they have difficulties understanding spoken language. When a learner has difficulty with reading comprehension, and they also experience difficulty with decoding and language comprehension, these learners likely have reading difficulties. Research suggests that approximately 5% to 10% of school aged learners may show this pattern of reading difficulty (Nation & Snowling, 1997; Yuill & Oakhill, 1991). Contrary to this, there are learners who can decode automatically and fluently and have adequate language abilities but who continue to experience difficulty coordinating the processes needed to comprehend more complex text (Nation & Angell, 2006).

Clearly, when learners are good in both of the variables (for example: word recognition and language comprehension), they will be good reading comprehenders. The simple view of reading refers to the fact that word recognition as well as language comprehension is necessary for a learner to become a proficient

(32)

reader. Learning to read involves increasing automaticity in processing word units, processing these units into recognisable words, and connecting the words while reading a text (LaBerge & Samuels, 1974). All the word-, sentence- and text-level skills involved in reading comprehension are, therefore, clarified as a subdivision of one of these two main components.

Different factors predict word reading from those that predict comprehension (Muter et al., 2004; Oakhill et al., 2003), for example, there are learners who have language comprehension difficulties in the absence of word recognition difficulties (Grigorenko et al., 2003; Nation, 2005). Likewise, there are learners who have word recognition difficulties in the absence of language comprehension difficulties (Catts et al., 2006; Spooner et al., 2004). According to Paul and Kellas (2004), there exist differences between the effects of context that operate at word and at text levels. The word-level skills that support comprehension include aspects such as accuracy, rate and prosody (National Reading Panel, 2000b). The rate makes it evident that accuracy alone does not predict comprehension (Stahl & Hiebert, 2004; Torgesen, Rashotte & Alexander, 2001). According to Stahl and Hiebert (2004:182), comprehension is built on a foundation of words. Thus, there is a strong link between word recognition and comprehension (Catts et al., 2006; Stahl & Hiebert, 2004; Torgesen et al., 2001).

Some learners, who are good at understanding spoken language, but poor at decoding text, are characterised as dyslexics. Other learners, who are good at decoding text, but experience difficulties understanding spoken language, are called hyperlexics. Readers experiencing difficulty in both decoding and language comprehension, are sometimes referred to as experiencing a variety of reading difficulties. After teachers categorise learners in the different quadrants through diagnosis, intervention can be planned as well as effective instruction (Rapp, Van den Broek, McMaster, Kendeou & Espin, 2007).

That is, it provides a clearer framework for teachers to focus their teaching clearly towards learning objectives for learners. Clear differentiation within the mind of the teacher between the two dimensions provides a conceptual framework that:

(33)

a) Encourages teachers not necessarily to expect that the learners they teach will show equal performance or progress in each dimension.

b) Offers the possibility of separately assessing performance and progress in each dimension, to identify learning needs and guide further teaching.

c) Makes explicit to teachers that different kinds of teaching are needed to develop word recognition skills from those that are needed to foster the comprehension of written and spoken language.

d) Emphasizes the need for teachers to be taught about and to understand the cognitive processes involved in the development of both accurate word recognition skills and of language comprehension (Stuart et al., 2000:12).

It also makes apparent that learners can experience various degrees of ease or difficulty in developing either word recognition or language comprehension or both, and invites teachers therefore to consider a learner‘s progress and ability in each of the two dimensions.

A learner‘s ability to understand and read the text can be predicted, thus only if a learner can decode words and has the ability to understand spoken language. When a learner can decode text automatically (pseudowords for preference), and when the learner has no problem understanding spoken language (implicit understanding of more formal language), only then is it possible to predict that the learner will not have difficulty with independent reading in comprehension (Stuart, Masterson & Dixon, 2000).

This perspective proposes that skilled reading entails a set of processes by which the words on the page are recognised and understood, and development of increasingly difficult language comprehension processes, by which texts as well as spoken language are understood and interpreted. Learning to read involves setting up processes by which the words on the page can be recognised and understood (word recognition), and continuing to develop the language comprehension processes that improve spoken and written language comprehension. Both sets of processes are necessary for reading, but neither is adequate on its own. Learners

(34)

who cannot sufficiently recognise the words on the page are by that fact alone prevented from fully understanding the text; however, recognising and understanding the words on the page is no guarantee that the text will be understood (Stuart et al., 2008).

Across the country, there is growing awareness of the dividends of early reading success and the stark consequences of early reading failure (DoE, 2002). A number of assessment studies in recent years have shown that the educational achievement of learners in South African schools is unacceptably poor. The South African Department of Basic Education has set a 75% achievement rate for Grade 3 learners performing in the country‘s Annual National Assessments (literacy) in 2019 as articulated in the Action plan to 2019: Towards the realisation of schooling 2030

(DBE, 2015).

In the eyes of many educational professionals, an extraordinary variety of classroom-targeted initiatives has been unleashed on schools over the past decade and more, all with the same general aim: the improvement of pupil learning. Assessment by teachers, whether formative or summative, is one of these developments that is considered to offer significant potential for improving pupils‘ learning (Black & Wiliam, 1998; Harlen, 2005). Teachers who integrate assessment into their teaching do so in order to identify where their pupils are in their learning and the steps they need to take for improvement and progress. Today, it is widely recognised that teachers should target teaching based on reliable evidence of what learners know and are ready to learn. Using evidence of learning to target teaching underpins two of the most powerful teaching strategies identified by researcher John Hattie in his landmark study, Visible Learning, which investigated more than 800 meta-analyses built on 50,000 individual studies (Hattie, 2009). Using evidence of learning to target teaching also underpins researchers Paul Black and Dylan Wiliam‘s seminal work on formative assessment (Black & Wiliam, 1998).

Like a doctor trying to identify what treatment patients need to improve their health, teachers need to identify what teaching their learners need now to improve their

(35)

learning. Good doctors use modern tests and procedures to understand their patients‘ symptoms and identify the underlying causes. Similarly, teachers should use proven strategies to develop a precise, evidence-based understanding of what their learners already know, and of what they are ready to learn next. When patients have an on-going condition, doctors follow up with them over time to assess symptoms, check on progress and adjust treatment if required. Similarly, teachers should observe and assess how learners respond to teaching, track their progress and adjust their teaching strategies accordingly.

A picture may be worth a thousand words, but in education, information speaks volumes. Data analysis can provide a snapshot of what learners know, what they should know, and what can be done to meet their academic needs. With appropriate analysis and interpretation of data, teachers can make informed decisions that positively affect learner outcomes. Research has shown that using data in instructional decisions can lead to improved learner performance (Wayman, 2005; Wayman, Cho & Johnston, 2007; Wohlstetter, Datnow & Park, 2008). No single assessment can tell teachers all they need to know to make well-informed instructional decisions, so researchers stress the use of multiple data sources. Generally, schools collect enormous amounts of data on learners‘ attendance, behaviour, and performance, as well as administrative data and perceptual data. But when it comes to improving instruction and learning, it‘s not the quantity of the data that counts, but how the information is used (Hamilton et al., 2009).

Making decisions about instruction is as core a component to teaching as providing the instruction itself. When providing services to learners at risk for poor educational outcomes, it is especially salient to ensure that the decisions that are made have the highest likelihood of accuracy as possible and lead to improving those outcomes. The learners with the greatest needs require the most accurate and effective decisions. In addition, recent increases in the need for accountability have put additional pressure on teachers to document their decisions and decision making processes. Now more than ever, effective use of assessment data to plan and judge instruction is a fundamental competency for good teaching.

(36)

Teachers need to make instructional decisions frequently. Estimates have put the number of decisions teachers make each day at 1,300 (Jackson, 1968) or about 10 interactive decisions per hour (McKay, 1977), but empirical work has also identified that teachers make 9.6 to 13.9 instructional decisions per lesson (Morine, Dershimer & Vallance, 1975). However, Peterson and Clark (1978) reported that decisions were only made when instruction was not effective and changes were only made in half of the situations in which learners were not learning sufficiently. Much of the research on the frequency of teacher decision making was conducted in the 1970‘s and 1980‘s (cf. Shavelson & Stern, 1981; Clark & Peterson, 1986). Since that time the focus of research has changed. Research on teacher decision making since the early 1980‘s has often focused on the outcomes of those decisions. The most common outcome is that when teachers use assessment data to make their instructional decisions, learner performance increases (Fuchs & Fuchs, 1986; Black & William, 1998a). In addition, teachers who collect systematic progress monitoring data (and use it to make decisions) make instruction changes more frequently for their learners who are experiencing difficulty (Fuchs, Fuchs, Hamlett & Stecker, 1991). Given the current focus on accountability and outcomes in education, training pre-service and in-service teachers to more effectively and efficiently collect and use assessment data to make instructional decisions for their learners and classes should be a core component of any professional development.

New teachers tend to lack the practical understanding and experience needed to conduct assessment effectively (Kanjee, 2008b; Kruger, 2008). They are also underprepared for interpreting assessment data and adapting their teaching in response to it – skills that are vital both to measuring progress and helping learners to succeed (Craven et al., 2014). This study aims to address these issues by means of a proactive action research study.

(37)

1.6 Methodology 1.6.1 Literature review

To identify relevant and recent sources for the purposes of the literature review, the data reference bases EBSCOHost Academic Search, RSAT, NEXUS and SABINET were utilised to search the following keywords: accurate and fluent reading, reading comprehension, vocabulary and language skills, assessment, progress monitoring assessment, instructional decision making.

A second EBSCOHost Academic Search was carried out to provide additional relevant and recent information obtained from theses, journals and other primary and secondary sources of information with the same keywords as in the paragraph above.

1.6.2 Empirical investigation

This section provides an overview of the methodology employed in this study. A detailed discussion is presented in chapter 4.

1.6.2.1 Research paradigm

This study is conceptualised within the interpretive paradigm. Interpretivists are concerned with understanding the meanings which people give to objects, social settings, events and the behaviours of others, and how these understandings in turn define the settings. In order to retain the integrity of the phenomena under study, interpretivists approach research differently from positivists. First, they study people in their natural surroundings (Connole, Smith & Wiseman, 1995). Second, they use methods of data collection that allow the meanings behind the actions of the people under study to be revealed. Commonly used methods in interpretivist studies are interviewing, observations, and analysis of documents of all kinds (Gephart, 1999). In addition to these, other methods of investigation may be integrated into the study

(38)

such as focus group interviews. With regard to the analysis of data, interpretivists carry out the task in tandem with data collection.

Potgieter (2008) sees the interpretative paradigm as being important because social contexts, conventions, norms and standards are of essence regarding a specific person or community if one wishes to understand human conduct. Maree (2009) points out that interpretivism aims at giving a perspective on a specific situation; to analyse this situation and give insight into the manner in which certain people, or a group of people, attach meaning to the situation. In this manner, a deeper meaning can be exploited – the situation can be understood – after which recommendations can be made. In this study, the aim is to collaborate with a Head of Department – a representative of the School Management Team (school level), teachers (classroom level), and Curriculum Subject Specialist (district level) in order to obtain an in depth understanding of data-based instructional decision making practices as well as the instructional support needs of teachers based on their collected data from various assessments. The collaborative aim is to establish a school-wide basic early literacy skill assessment and instructional support framework that will not only enhance the assessment practices of teachers, but also the system-wide decisions that need to take place so that effective instructional decisions can be made at all levels, and most importantly at the classroom level. Reponses from all stakeholders are analysed and interpreted within the context of an action research design.

1.6.2.2 Research approach

There are several considerations when deciding to adopt a qualitative research methodology. Strauss and Corbin (1990) claim that qualitative methods can be used to better understand any phenomenon about which little is yet known. They can also be used to gain new perspectives on things about which much is already known, or to gain more in-depth information that may be difficult to convey quantitatively. For example, in the present study little is known about the data-based instructional decision making practices of teachers. In addition, little is known about the data-based decision making support requirements on a system-wide level. The ability of

(39)

qualitative data to more fully describe a phenomenon is an important consideration not only from the researcher's perspective, but from the reader's perspective as well. "If you want people to understand better than they otherwise might, provide them information in the form in which they usually experience it" (Lincoln & Egon, 1985:120). Qualitative research reports, typically rich with detail and insights into participants' experiences of the world, "may be epistemologically in harmony with the reader's experience" (Stake, 1978:5) and thus more meaningful.

The particular design of a qualitative study depends on the purpose of the inquiry, what information will be most useful, and what information will have the most credibility. There are no strict criteria for sample size (Patton, 1990). "Qualitative studies typically employ multiple forms of evidence....[and] there is no statistical test of significance to determine if results count‖ (Eisner, 1991:39). Judgments about usefulness and credibility are left to the researcher and the reader.

By using qualitative research, a situation is investigated in a natural environment in which it occurs (Mayan, 2001). Generalisation must not take place; the deeper meaning that comes to the fore needs to be searched for (Mayan, 2001). The qualitative research approach is an approach by means of which an attempt is made to obtain, analyse and understand rich descriptive data pertaining to a specific subject or context. The idea is to understand individuals or groups in the social and cultural context they live in. Flick (2009) points out that qualitative research is applied to understand and describe a social phenomenon and to attempt to explain it; this study describes the phenomenon of data-based instructional decision making, and how it features in contexts such as a district, school and classrooms. An attempt is made to analyse the experiences of individuals, through the interaction between the individual and the specific phenomenon, namely data-based instructional decision making.

(40)

1.6.2.3 Research design

According to Babbie and Mouton (2001) and Neuman (2006), a research design is a plan, a protocol or a structured framework of how the researcher intends to conduct the research process so as to solve the research problem(s) or question(s). The research design therefore describes the nature and the pattern that the research intends to follow (Creswell, 1998).

A proactive action research design was chosen for this study. Action research has been described as a ―small-scale intervention in the functioning of the real world and a close examination of the effects of such an intervention‖ (Cohen & Manion, 1995:186). Action research in education seeks to blur the boundary between the researcher and the subject. It is conducted either by or with insiders to the institution or community being studied – teachers, parents, learners, administrators, etc. Similarly, action research seeks to bridge the gap that sometimes arises between research and practice by designing research that is explicitly centred around action, and aimed at changes in the institutions studied, and/or in the researchers themselves (Herr & Anderson, 2005). Action research might be conducted by an individual teacher seeking to analyse and further develop her practice, or by a whole team of stakeholders including learners, teachers and administrators, using their research to design district-wide changes. Knowledge creation and action take place in a cyclical, iterative process, akin to Freire‘s (1970) concept of praxis. Action research can include almost any method of data collection or analysis, and can be conducted based on diverse paradigms. At the same time, it is founded on some basic epistemological assumptions. First of all, action research privileges the use of insider knowledge of institutions and social systems, challenging ideas of the ―expert‖ and the need to observe from an ―objective‖ distance. Secondly, it privileges the production of localized knowledge; although action research can create knowledge that is transferable to other locations, it is first and foremost interested in site-specific knowledge that can be used in the location in which the research takes place (Herr & Anderson, 2005) (e.g., insider knowledge of data-based instructional decision making and how it features in a specific district, school and classrooms).