Accepted Manuscript

Title: Challenges in wide scale implementation efforts tofoster higher order thinking (HOT) in science education acrossa whole school system

Author: Anat Zohar

PII: S1871-1871(13)00041-2DOI: http://dx.doi.org/doi:10.1016/j.tsc.2013.06.002Reference: TSC 207

To appear in: Thinking Skills and Creativity

Received date: 22-1-2013Revised date: 12-6-2013Accepted date: 15-6-2013

Please cite this article as: Zohar, A., Challenges in wide scale implementation efforts tofoster higher order thinking (HOT) in science education across a whole school system,Thinking Skills and Creativity (2013), http://dx.doi.org/10.1016/j.tsc.2013.06.002This is a PDF file of an unedited manuscript that has been accepted for publication.As a service to our customers we are providing this early version of the manuscript.The manuscript will undergo copyediting, typesetting, and review of the resulting proofbefore it is published in its final form. Please note that during the production processerrors may be discovered which could affect the content, and all legal disclaimers thatapply to the journal pertain.

Page 1 of 45

Acce

pted M

anus

cript

Challenges in wide scale implementation efforts to foster higher order

thinking (HOT) in science education across a whole school system

Abstract

This study explores the challenges involved in scaling up projects and in

implementing policies across the whole school system in the area of teaching higher

order thinking (HOT) in Israeli science classrooms. Eight semi-structured individual

interviews were conducted with science education experts who hold leading positions

pertaining to learning and instruction on the state level of the following school subjects:

elementary and junior- high school science and technology; high-school physics; high

school chemistry; and high school biology. Some of the challenges that the interviews

revealed are common to many types of educational change processes. The interviews

also revealed several challenges which are more specific to the educational endeavor of

teaching HOT according to the infusion approach across large numbers of classrooms:

challenges involved in weaving HOT into multiple, varied, specific science contents;

challenges involved in planning a reasonable and coherent developmental sequence of

thinking goals; the fact that content goals tend to have priority over thinking goals and

thus to disperse of the latter in policy documents and in implementation processes; and

finally, the considerable challenges (pedagogical and organizational) involved in

developing educators sound and deep professional knowledge in the area of teaching

HOT and metacognition on a large, nation-wide scale. The data shows that wide-scale

implementation of thinking in Israeli science classrooms often develops as an

evolutionary rather than as a revolutionary process. The implications for designing large

scale implementation programs aimed at fostering students reasoning are discussed.

Key words: Higher order thinking, thinking strategies, large scale

implementation, from policy to practice, teachers' knowledge

Page 2 of 45

Acce

pted M

anus

cript

1. Introduction

1.1 Issues involved in teaching HOT on a large scale

So teaching for thinking and understanding [across the whole school

system] We have not yet entirely deciphered the code of how to do it

[L2]

As the introductory citation indicates, the challenge of teaching thinking on a

large, national scale is a huge one. There is nothing new in acknowledging that a large

gap often exists between educational policy and the way it is implemented. This gap is

especially large in the context of policies that address changes in the core of education,

i.e., changes in learning and instruction, such as the change involved in teaching

thinking. Unfortunately, it is very difficult to change the core of education on a large,

system-wide scale.. Large scale efforts to improve teaching and learning focus more on

structural and administrative characteristics of reform than they do on fundamental

changes in the instructional core. Innovations that require significant changes in the core

of educational practice are usually not only limited in their effects to a small scale, but

also do not usually last very long.

Innovations addressing the teaching of thinking are definitely at the very core of

learning and instruction. Without delving into the challenges involved in defining higher

order thinking (e.g., Resnick, 1987; Schraw & al., 2011), I refer here to the latter

concept in its widest sense encompassing issues such as thinking skills/strategies,

critical thinking, argumentation, use of evidence, scientific reasoning, scientific literacy,

inquiry, problem-based learning and problem solving. During the past 30 years there

have been a substantial, and rapidly growing number of empirical studies supporting

models and theories that address teaching thinking in science classrooms. Consequently,

educators are currently familiar with many good models that work quite well for

teaching science by emphasizing students' higher order thinking rather than merely

memorization of facts.

Most of the models for teaching thinking in science education classrooms were

studied within small scale projects. In addition, there have been some pioneering

attempts to scale up such projects to scores of teachers and classrooms (e.g., Adey &

Page 3 of 45

Acce

pted M

anus

cript

Shayer, 1993; 1994; Blumenfeld, Fishman, Krajcik & Soloway, 2000; Osborne,

Erduran and Simon 2004; White & Frederiksen, 1998; Author, 2004). However, at the

turn of the 21st century, successful teaching of thinking on the level of small or even

large - scale projects is no longer sufficient. Policy documents from all over the world

highlight the need to teach 21st century skills. HOT is an important component of any

list of 21st century skills (Partnership for 21st Century Skills, retrieved July, 2011;

Pellegrino & Hilton, 2012). Resnick (2010) argues that scaling up the "thinking

curriculum" in a way that will foster proficiency for ALL students is currently a major

educational challenge:

"Today we are aiming for something new in the world: An elite standard for

everyone That is what the term 21st-century skills really means. The skills are

not new (some students have been successfully learning them in some schools

from the beginning of civilization). But the aspiration to successfully teach

knowledge-grounded reasoning competencies to everyone is still just thatan

aspiration. But the transformation of the institution of schooling that will be

needed to come close to making the aspirational goal a real achievement is huge

"(P. 184)

The goal of this paper is to examine the challenges involved in scaling- up

instruction of higher order thinking. The meaning of scaling up in this context is to take

ideas and practices educators are familiar with on the level of projects and to implement

them on a national level, i.e., across the state's whole school system. The paper

examines these challenges by studying the views of leaders who had been involved in

various large scale efforts to implement HOT in science instruction. Naturally, some of

the pertinent challenges are common to gaps between policy and practice in general, or

to scaling - up innovative, reform pedagogies in other areas (e.g., Blumenfeld, Fishman,

Krajcik, Marx, & Soloway, 2000; Dede, Honan and Peters, 2005; Elmore, 2004;

Fullan, 2007; Levin, 2008; Levin & Fullan, 2008; Lee & Krajcik, 2012). Yet, because of

the unique features of teaching higher order thinking, some of these challenges are

unique to efforts aiming at fostering students' thinking across hundreds or even

thousands of classrooms.

Since many of the challenges that will be described in the findings section

pertain to the development of teachers' knowledge, , the next section will discuss

relevant prior studies addressing teachers' knowledge in the context of teaching HOT.

Page 4 of 45

Acce

pted M

anus

cript

This will be followed by a section that will describe the educational context within

which the present study took place.

1.2 Teachers' knowledge in the context of teaching HOT

Instruction of HOT requires much more than adopting a new curriculum because

it requires a deep change in teaching practices. Like the teaching of other issues that

pertain to current educational reforms, it stretches and challenges teachers' capabilities.

In order to be able to respond to the unexpected events characterizing "thinking rich

classrooms", teachers must be able to teach in an intelligent, flexible and resourceful

way that cannot be scripted into a fixed set of technical instructional routines and skills

(Carpenter et. al., 2004; Loef-Frank et al., 1998). In order to teach thinking successfully

teachers need to replace the traditional view of teaching as transmission of information

and learning as passive absorption with more active, constructivist views of learning

and an intricate set of specific beliefs and knowledge about teaching. Let us take a

closer look at this knowledge.

1.2.1 Subject Matter Knowledge and Pedagogical Content Knowledge (PCK)

in the Context of teaching HOT

As many studies show, familiarity with whatever it is that one is supposed to

teach is a necessary condition for instruction. Another necessary condition for sound

instruction is familiarity with appropriate teaching methods. There is a large body of

literature that, following Lee Shulmans work, addressed various components of

teachers knowledge and distinguished (among other things) between subject matter

knowledge, general pedagogical knowledge and pedagogical content knowledge (PCK).

However, since the classic discourse in this area usually applies to teaching concepts

rather than to teaching thinking, the meaning of these components of teachers

knowledge is not straight forward when we try to apply it to the context of teaching

HOT. It therefore requires further clarification.

The term used in the literature for whatever it is that one is supposed to teach is

subject-matter knowledge (e.g., Cocharn & Jones, 1998; Shulman, 1986, 1987; Wilson

et al., 1987). But because of the unique nature of thinking strategies this concept is

confusing when the focus of our attention is on teaching thinking rather than on

teaching facts and concepts. Although according to Shulman subject matter knowledge

includes substantive knowledge (the explanatory structures or paradigms of the field)

Page 5 of 45

Acce

pted M

anus

cript

and syntactic knowledge (the methods and processes by which new knowledge in the

field is generated), content knowledge (the knowledge of specific facts and concepts) is

an essential component. When we focus on teaching thinking the traditional meaning of

content knowledge is not at the core of our educational agenda. Therefore, in order to

avoid confusion and to delineate the unique nature of teaching thinking, I prefer in this

context to substitute the term subject-matter knowledge with the term knowledge of

elements of thinking. This includes: a. knowledge of individual thinking strategies such

as making comparison, formulating justified arguments or drawing valid conclusions; b.

knowledge of genres of thinking such as argumentation, inquiry learning, problem

solving, critical thinking, scientific thinking or creative thinking (Schraw & al., 2011;

Ministry of education, 2009); c. knowledge of metacognition (for elaboration, see

below); and d. knowledge of a variety of additional issues which are important for a

successful "thinking classroom" such as thinking dispositions or habits of mind, and an

appropriate "culture of thinking" (Perkins et. al., 1993; Swartz et. al., 2008). It is

important to note that several previous studies show that in-service and pre-service

teachers initial knowledge of thinking strategies are often not sound enough for

purposes of instruction (e.g., Bransky et al., 1992; Brownell et al., 1993; Jungwirth,

1987, 1990, 1994; Paul et al., 1997; Zembal-Saul et al., 2002; Author, 2004).

A second component of teachers knowledge which is significant for the present

paper is pedagogical content knowledge (PCK). PCK is a blend of pedagogical

knowledge and subject-matter knowledge that is specific to each teaching topic (e.g.,

Adams & Krockover, 1997; Cocharn & Jones, 1998; Gess-Newsome & Lederman,

1999; Kennedy, 1990; Loughran et al., 2000; Shulman, 1986, 1987; Van Driel et al.,

1998; Wilson et al., 1987; Zeidler, 2002). In the context of teaching higher order

thinking, the classic conceptual distinction made in the literature between pedagogical

content knowledge and general pedagogical knowledge is fuzzy and unclear. Part of the

difficulty in aligning teachers knowledge in the context of teaching thinking with the

prevalent concepts used in the literature is related to the debate among scholars

regarding the question of whether thinking strategies are general or content specific.

Teaching thinking according to the infusion approach i.e., integrating the teaching of

thinking with the teaching of specific contents (Ennis, 1989; Swartz & al., 2008;

Abrami, 2008) assumes that thinking skills have some elements that are general and

other elements that are content specific (Perkins and Salomon, 1989). This notion

presents an innate difficulty in referring to the pedagogical knowledge teachers have in

Page 6 of 45

Acce

pted M

anus

cript

this field as either pedagogical content knowledge (that tends to be embedded in

specific subject-matters), or as general pedagogical knowledge (that tends to be

independent of specific subject-matters). It seems that because of the special nature of

the type of knowledge under consideration the existing constructs are problematic. I had

therefore suggested addressing teachers pedagogical knowledge in relation to

instruction of higher order thinking by using a special term: pedagogical knowledge in

the context of teaching higher order thinking (Author, , 2004; 2008 ). This term fits

well with the term knowledge of elements of thinking explained earlier, and

highlights the fact that pedagogical knowledge in this field has some unique

characteristics. At the same time this term does not imply a commitment to treat this

knowledge as either content-specific or general.

1.2.2 Teachers' knowledge in the context of metacogniton

A third component of relevant teachers knowledge pertains to

metacognition. One of the most widely used definitions of metacognition was proposed

by Flavell and his colleagues (Flavell, 1979; Flavell, Miller, & Miller, 2002),

distinguishing between two major components of metacognition: metacognitive

knowledge and metacognitive monitoring and self-regulation. The latter component is

also named in current literature as metacognitive skills (e.g., Veenman, Van Hout-

Wolters, & Afflerbach, 2006). Metacognitive knowledge includes three sub-categories:

knowledge about persons, tasks, and strategies. In the metacognitive skills branch of

metacognition, Flavell et al. (2002) elaborate on monitoring, self-regulation, and also

describe planning and evaluating. These sub-categories are also used in other prominent

frameworks. For example, Schraw and his colleagues describe the regulation component

of metacognition as comprising of processes of planning, monitoring, and evaluation

(Schraw, 1998; Schraw & Moshman, 1995). More recently, Whitebread and his

colleagues have proposed a framework that adopts Flavells definitions of

metacognitive knowledge and defines regulatory skills of planning, monitoring,

evaluating, and control (Whitebread et al., 2009). Many studies show that using

metacognition in the classroom may improve learning in general (see Veenman 2011 for

review), and learning of problem-solving, inquiry and higher order thinking in particular

(e.g., Chen & Klahr, 1999; Lin & Lehman, 1999; Ross, 1988; Schoenfeld, 1992; Toth,

Klahr, & Chen, 2000;White & Frederiksen, 1998, 2000; Author, and XXXA, 2008;

Author, and XXXB, 2008).

Page 7 of 45

Acce

pted M

anus

cript

Yet, despite its strong effect on learning, recent studies highlight the fact that

teachers find enacting a pedagogy for metacognition difficult. It is not trivial for them to

take up research-based ideas in this field and to translate them into practical

recommendations (Leat and Lin, 2007). However, despite the prominent role of

metacognition in student success, only limited research has been conducted to explore

teachers' and pre-service teachers' metacognitive knowledge, their pedagogical

knowledge, and their ability to make progress in these types of knowledge following PD

(Abd-El Khalicka and Akerson, 2009; Kramersky and Mihcalsky, 2009; Wilson and

Bai, 2010). More specifically, a review of journal articles about metacognition revealed

that between 2000-2012 only three studies were conducted to explore science teachers'

learning of issues pertaining to the pedagogy of teaching metacognition in science

classrooms (Author and XXX, submitted).

One of the components of metacognition which is particularly significant

for teaching and learning thinking strategies is meta-strategic knowledge, or MSK for

short. MSK consists of general knowledge about thinking strategies, i.e., what is the

strategy and when, why and how it should be used. In order to apply MSK successfully

in the classroom, teachers need to have sound MSK, as well as a variety of specific

elements of pedagogical knowledge consisting of relevant teaching strategies such as:

modeling the use of a thinking strategy in a variety of specific contents; providing

opportunities for students to articulate the thinking strategies they apply during

thinking; to introduce the "language of thinking" into the classroom; to design and to

teach careful and thoughtful learning activities in which thinking goals are made

explicit; and to engage in long term and systematic planning of thinking activities across

several sections of the science curriculum. Research findings show= that science

teachers' initial knowledge regarding MSK is lacking and unsatisfactory for the purpose

of instruction and that teachers are not aware of the pertinent pedagogical knowledge

before they study about it in PD courses designed to address that knowledge (Author,

1999; 2006; 2008).

1.2.3 Teachers' knowledge pertaining to pedagogies of knowledge

construction

1.3 Educational Context: relevant facets of the Israeli educational

school system and a brief history of teaching HOT in science classrooms in Israel

Page 8 of 45

Acce

pted M

anus

cript

In order to understand the context of the processes discussed in this paper, some

relevant background information about the Israeli educational school system and about

pertinent efforts to teach thinking over the years is called for.

The Israeli educational system is centralized. With approximately 2 million

students (k-12th grade) and over 4000 schools there is basically one mandatory

curriculum prescribed by the Ministry of Education that covers a large percentage of

what is taught in most schools. At the end of high school students take the matriculation

exams that consist of exams in 7 mandatory core subjects: Language (Hebrew/Arabic),

English (as 2nd language), Mathematics, History, Bible, Literature and Civics.

Additional subjects are mandatory in elementary and junior high schools (Science and

technology, Geography, 2nd foreign language, etc.). In addition, many other subjects

are electives in high school (e.g., Biology, physics, chemistry, communication, arts,

computers, etc.). Each subject has a steering committee which consists of academics,

Ministry of Education officials and teachers. The steering committee is usually chaired

by a distinguished professor from the pertinent academic discipline who is interested in

contributing to educational practice in his or her field. For each subject there is also a

National Subjects Superintendent (NSS) who is responsible (in collaboration with the

steering committee) for policy making (i.e., for defining the goals of the curriculum)

and for the practical sides of instruction, including teachers PD and assessment in that

particular subject. NSSs work with a team of instructors who help to coordinate and to

lead the above activities in each subject. Instructors also provide teachers with

pedagogical support through PD courses, visits in classrooms and schools and by

meetings with small groups of teachers to discuss professional matters. As we shall see,

instructors have a prominent role in the implementation processes described in what

follows.

Teaching thinking is not a new goal in science education in Israel. In fact, it has

quite a long history. Although in the first half of the 20th century students had

sometimes studied science by inquiry and through observations that took place in field

excursions, science education during that period had been, for the most part, descriptive

and fact-based. The first comprehensive science education reform in this area took place

in the late 1960's, highlighting inquiry and higher order thinking. It centered on a policy

decision to change biology education into inquiry teaching by adopting the inquiry-

oriented Yellow version of BSCS (BSCS- Biological Science Curriculum Study,

1963), and adapting it to the context of the Israeli school system. This reform which

Page 9 of 45

Acce

pted M

anus

cript

had been extremely innovative at the time, assumed that the way to create a long-

lasting, sustainable educational change must be holistic (Tamir, 2006). It was assumed

that the route to success must pass through working simultaneously in three main areas:

curriculum materials, professional development, and assessment. The new assessment

methods that had been developed as part of that change process in Israel were

innovative and unique in the 1970s'. The biology team under the leadership of Pinchas

Tamir understood that a lack of consistency between the inquiry-goals of the new

curriculum and the assessment will put the reform effort at risk, because teachers will

continue to teach for the test in terms of content, rather than follow the inquiry-

oriented learning environment of the new curriculum. The matriculation exam was

therefore transformed in a radical way to reflect the goals of the new inquiry-based

curriculum. In effect, Tamir and his team changed the traditional matriculation

examination into an assessment that by todays terminology can be defined as

alternative assessment. The "new" matriculation examination consisted of multiple,

varied means of assessment designed to assess science knowledge as well as inquiry

skills: a written test, a school-based research project accompanied by an oral exam, a

school-based laboratory test, and a field test. It is remarkable to note that many of the

main facets of the inquiry-based biology curriculum and assessment are still practiced in

the biology national curriculum even today, more than 40 years after the reform had

been initiated. Consequently, biology teaching in Israel had been more thinking oriented

than physics or chemistry even in the early 2000's (Author and XXX, 2005).

The next noteworthy reform policy is the "Tomorrow 98" reform. Following the

nomination of a public committee, a comprehensive report was written in 1992 about

math, science and technology education. The report made a list of recommendations

aimed at the improvement of education in these areas with an eye to preparing students

for life in the 21st century (Ministry of Education and Culture, 1992). Among other

goals, the report stated that the development of students' thinking in science is an

important educational goal. The report was followed by a generous budget devoted to

the improvement of k-12 science education. Among the report's practical consequences

were the following: a. Two new curriculum documents (for grades 1-6 and for grades 7-

9) which included explicit HOT goals to be taught in science lessons. In those

documents lists of thinking skills are delineated in a special section of the curriculum

that is separate from the sections delineating science topics and concepts b. Several

Page 10 of 45

Acce

pted M

anus

cript

separate projects that created learning materials designed to integrate the teaching of

science topics with the teaching of thinking strategies and to promote project-based

learning.

Another important facet of these projects were PD courses aimed at developing

teachers' ability to use the new learning materials in a proficient way (Eylon & Bagno,

1997; Spector-Levy, Eylon & Schertz, 2008; r, 2004). Changes aimed at fostering HOT

and inquiry learning were also introduced to high school chemistry (See Avargil & al.,

this volume).

Finally, in January 2007 the Israeli Ministry of Education (MOE) adopted a new

general (i.e., not limited to science education) national educational policy: Pedagogical

Horizon (PH) Education for Thinking (Office of Pedagogical Affairs, MOE, 2007;

Author, 2008; Gallagher, Hipkins and Zohar, 2012). The novelty of the PH policy was

is in addressing the teaching of thinking as an explicit, major and universal educational

goal; and in planning practical means for wide-scale implementation throughout the

school system. The emphasis of the Pedagogical Horizon policy was on pedagogy

rather than on content: on how to rather than on what to teach. The policy adopted an

infusion approach to teaching Higher Order Thinking (HOT): thinking was integrated

into school curricula rather than taught as an independent subject. Therefore, the policy

advocated thinking within conceptually rich domains of knowledge. An ideal lesson

according to the PH policy consisted of both content goals and thinking goals each of

which were addressed in an explicit way. The lesson was rich in cognitively challenging

questions and tasks that made intense usage of thinking strategies such as

argumentation, problem solving, asking questions, comparing and contrasting, making

decisions, controlling variables, drawing conclusions and identifying assumptions. All

these thinking strategies were being used within the lessons subject matter and were

thus embedded in rich conceptual contents. The classroom learning environment

fostered discourse that wa rich in the language of thinking. Inquiry learning was

encouraged. Finally, lessons also fostered metacognitive thinking, including intensive

engagement with MSK. This means that meta-level elements of thinking strategies were

taught in an explicit way. Principles pertaining to the "what", "when" and "why" of

thinking strategies were explicitly addressed in class discussions, in active individual

and group-work and in learning materials (textbooks and computerized learning

materials. Many studies show that such explicit engagement with general meta-level

Page 11 of 45

Acce

pted M

anus

cript

features of thinking have a large positive effect on students' level of thinking (e.g.,

Abrami, 2008; Dwyer & al., 2012; Author and XXXA, 2008; Author and XXXB,

2008).

Following the model of the BSCS program that had taken place 4 decades earlier

the PH was implemented through working simultaneously on the development of three

areas: Curricula and learning materials, PD and assessment.

1.4. Research Question

In sum, this brief review shows that several system-wide substantial efforts to

promote HOT had taken place in science education in Israel over the years. These

efforts may provide a fruitful context to study system-wide change processes in the area

of teaching HOT. Numerous informal conversations conducted over the years with

some of the people who were deeply involved in leading such implementation processes

indicated that they had gained rich knowledge and insights through their experiences. A

set of interviews was thus designed with the overall goal of uncovering that knowledge

and insights in order to turn them into explicit common knowledge. The interviews

covered too many topics to be addressed in a single paper (see below for more details).

The present paper centers on one of these topics- the challenges involved in scaling up

instruction of HOT across the whole school system. Consequently, the present paper

explores the following research question: What do science education leaders view as the

main challenges in scaling up instruction of higher order thinking across the whole

school system?

2. Methodology

This study is based on a set of eight semi-structured individual interviews

conducted with science education leaders in Israel. These leaders were chosen because

of their prominent roles and first-hand experiences in leading activities and processes on

the national level in the following school science subjects: elementary general science,

junior high school general science, high school biology, high school physics, and high

school chemistry. Four interviewes were conducted with National Subject Supervisors

(NSS's), i.e., with people who have had leading positions in directing and supervising

the pertinent school science subjects within the Ministry of Education, Four additional

interviews were conducted with university science education professors who, in

Page 12 of 45

Acce

pted M

anus

cript

addition to their academic roles, had been active and highly influential in national level

science education committees in the Ministry and in university-based science education

projects. They served as chairs or members of steering committees who directed the

various school science subjects, as directors of large-scale science projects, and/or as

directors of teachers' national PD centers.

One of the university science education professors was involved intensively in

leading educational activities in two different school science subjects and was therefore

interviewed twice a separate interview took place for each of the school subjects s/he

was involved in. Consequently there were a total of eight interviews but only seven

different interviewees.

All the potential candidates that were asked to participate in this study consented

to do so (100% consent). The researcher had prior professional acquaintance with all

seven interviewees. In order to encourage free expression of ideas the interviewer

promised anonymity. Because the interviewees are public figures who are well known

in the educational arena in Israel all personal details and potentially identifying

information are omitted from the study. Therefore, additional information about the

interviewees cannot be provided here in order to preserve the promised anonymity.

Interviews asked about the present state of teaching for thinking and

understanding in the relevant school science subjects, about past and current change

processes related to wide-scale implementation of thinking and about challenges and

assessments related to these processes (for the full interview protocol please see

Appendix A). Interviews were between one and two hours long. Interviews were semi-

structured: Following a general presentation of the goal of the study and the definition

of higher order thinking in the interview, the researcher presented the same set of 10

questions to all interviewees. However, following their diverse areas of expertise and

personal experiences, each interviewer was allowed to elaborate on different sections of

the interview protocol. Different follow-up questions were thus presented to each of the

interviewees. The interviews were transcribed and analyzed using content analysis, i.e.,

"a process of qualitative data reduction and sense - making effort that takes a volume of

qualitative material and attempts to identify core consistencies and meanings" (Patton,

2002, p. 453). The interview transcripts were read several times and several short ideas

emerging from the text were written for each section of the interview, using a word

Page 13 of 45

Acce

pted M

anus

cript

processor table. Sections consisting of ideas that had a similar meaning were tassembled

together to create patterns.

Each pattern was then read and re-read to create codes and sub-codes. Sections

consisting of the same codes and sub-codes were assembled together to create groups of

citations. Finally, each of the patterns chosen for the analysis presented in the present

paper (for elaboration on the choice process see below) were analyzed by trying to

interpret the meaning of the citations grouped under the same sub-code and code. The

analysis thus focused on trying to make sense of citations grouped in each pertinent

pattern using a narrative approach.

3. Results

As mentioned earlier, due to the interviewees' high expertise, the interviews were

extremely rich in ideas and insights and the text contained too many significant patterns

to be included in one paper. In order to reduce the number of patterns for the present

analysis, it was decided to limit the topic of the present paper to the challenges involved

in large- scale implementation of teaching HOT. Within this wide topic, two major

groups of patterns were identified: (a) patterns addressing challenges in large-scale

educational change in general (i.e., challenges that are visible in many areas of

educational change, see section 3.1); and, (b) patterns addressing challenges pertaining

to pedagogical issues that are more specific to large-scale implementation of HOT (see

section 3.2). In what follows the first group of patterns is described in short and the

second group is described in detail.

3.1 Challenges which are visible in many areas of educational change

The first group of patterns addressed issues involved in large-scale educational

changes in general (i.e., changes that are not specific to the implementation of HOT).

Although the term HOT appeared in the text describing these patterns, this term could

have theoretically been replaced with terms describing other educational goals without

any loss of meaning. The ideas expressed in these patterns can be found in numerous

previous studies describing educational change processes in varied contexts. These

patterns were judged as less central to the present study and are thus summarized and

described briefly in Figure 1.

--------------------------------

Page 14 of 45

Acce

pted M

anus

cript

Insert figure 1 about here

--------------------------------

. The data presented in Figure 1 confirms that successful large scale

implementation of teaching HOT encounters the same general system- wide challenges

as many other large scale educational change processes in numerous areas that appear in

the literature concerning educational change (e.g., Elmore, 2004; Fullan, 2007)

3.2 Challenges that are unique to large scale implementation of HOT

The second group of patterns addressed pedagogical issues which are more

specific to the challenges involved in large scale implementation of HOT (see Figure 2).

These patterns were judged as central to the present study and therefore they are

described in detail, using a narrative approach. Since the borderline between the two

groups of patterns is not always clear-cut, my rule of thumb was that boarder - line

patterns were judged according to whether or not I estimated that the challenges

described in the patterns contributed in a significant way to the emerging discussion

about large scale implementation of HOT.

In the case of patterns reflecting assessment however, an exception had been made

to this rule of thumb. Although issues pertaining to assessment are central to large scale

implementation of HOT they were not described here in detail (but instead were

described briefly in Figure 1). Since space restrictions made it necessary to leave

something out, I chose not elaborate on assessment in this particular paper because: (a)

addressing this significant topic required an elaborate theoretical and data-based

analysis that was beyond the scope of this paper; (b) Since assessment was the focus of

several of the other papers in this Special Issue I assumed that the relative contribution

of an additional detailed discussion of assessment would be smaller than that of other

topics which were not as prominent in the other papers.

The second group of patterns are described in detail in the following sections and

summarized briefly in Figure 2.

Page 15 of 45

Acce

pted M

anus

cript

--------------------------------

Insert figure 2 about here

--------------------------------

3.2.1 The challenge of teaching thinking in varied, specific science contents

Integrating the teaching of content and thinking goals to create a coherent

curriculum is a considerable challenge even in small scale projects. However, the

teaching of HOT strategies in varied, yet specific science contents across the whole

science curriculum is a much larger challenge. The underlying theoretical framework is

that instruction needs to integrate these two dimensions (i.e., content and thinking)

according to the infusion approach to teaching HOT (e.g., Ennis, 1989; Swartz et. al.,

2008). Accordingly, teaching thinking should be woven into the teaching of

conceptually rich science content. Moreover, the theoretical approach is that thinking

goals should be conceived of as explicit, distinct educational goals that should be

discussed explicitly in the classroom while engaging with authentic, rich science topics.

L1 formulated this theoretical approach by saying: "We are talking about direct

teaching of thinking, about teaching according to the infusion approach - content

knowledge together with procedural knowledge". By the time this study took place (the

interviews took place during the summer of 2012), this approach is no longer an

innovation in Israeli science classrooms, because, as L2 states: "Explicit reference to the

need for infusing thinking skills has been on the agendasince 1996". Yet,

implementing this approach across the whole school system encounters difficulties on

many levels:

"This whole business of anchoring the thinking in the science topics was not

entirely clear, nor was it emphasized There were a lot of discussions around this issue

but there was no clear policy and no clear indication of how to actually carry it out.

Eh In the context of the new syllabus [i.e., the syllabus written in 2009- 2010] the

notion was that thinking skills should be combined [with the science topics] in a

systematic way. But ... at least at the beginning it was not clear where [i.e., in what

science topics] to combine what [thinking skills] and how to perforn it in a spiral way

This whole issue of science content and thinking skills, and of teaching them in an

integrated way, this is the discourse you hear on the policy level. You would like to

Page 16 of 45

Acce

pted M

anus

cript

construct it in a way that the thinking skills will support content knowledge and the

content knowledge will support the thinking skills. But nobody knows exactly how to

do it... [L2).

"I think that in the [biology] curriculum content goals appear separately. Actually,

I think it's the same in the junior high school science curriculum. Yes, there was an

effort to connect the thinking skills into the blueprint of what needs to be taught [in

terms of content], but there is no indication of how to integrate them" (L4)

"So these needs [i.e., the needs to combine the teaching of content and thinking]

came up, but people did not really know how to handle them" (L6]

These excerpts confirm that although the underlying theoretical framework was

clearly that instruction needs to integrate the content and thinking dimensions, in effect,

over the years content goals and HOT goals were detailed in curriculum documents in

two separate lists. There was a list of content goals and a separate list of thinking goals,

with no indication for how to combine them. Moreover, it seems that this lack of

indication was not incidental. Rather, it reflected gaps in pedagogical knowledge even

among academic experts. It seems that this type of experts' pedagogical knowledge has

been developed over the years so that they knew more about it in recent years

compared to several years ago. Yet, it seems that there is still a long way to go because

even at the time the interviews took place experts did not feel they had a clear and

systematic method for how to integrate the teaching of thinking and of science content:

"The fact that we now have a better methodology [for integrating the teaching of

thinking with science content] than we used to, does not mean that we now have a good

methodology. I don't think it [i.e., a good methodology] really exists. It's something that

people are currently developing, and will continue to develop in the future [L6]

In conclusion , it seems that the challenge of integrating the teaching of thinking

in multiple, varied science content had not yet been resolved even in the minds of the

experts who were leading the implementation processes.

3.2.2 Planning a reasonable and coherent developmental sequence

Matching the thinking goals of a small-scale project to students' age and

developmental stage is a challenging yet manageable task. This may be done based on a

Page 17 of 45

Acce

pted M

anus

cript

combination of research-based recommendations and educators' experience and on their

intuition regarding the type of learning which is appropriate for students of various ages

and cultural backgrounds. Planning a whole curriculum encompassing varied ages and

school subjects, however, is a much more complex endeavor. In considering the "when"

dimension of teaching thinking, Alexander & al. (2011) briefly summarize the vast

developmental psychology literature indicating that age can have significant effects on

mental processing. Yet, Alexander and her colleagues emphasize that the level of

thinking competence depends upon a variety of additional reasons, such as the amount

of scaffolding provided to students during instruction, the learners' prior experiences

with HOT, their level of subject-matter expertise and more. One particular post-

piagetian model that deserves mentioning in this context is Fischer's Dynamic Skill

Theory which postulates that a simple age-related stage theory is too simple to explain

the vast variability observed in human psychological traits. Instead of viewing stages of

cognitive development fixed like the steps of a ladder, this theory assumes a more

dynamic metaphor for development that of a constructive web. Unlike the steps in a

ladder, the strands in a web are not fixed in a determined order but are the joint product

of the web-builder's constructive activity and the supportive context in which it is built.

Therefore, the support (e.g., types of learning experiences and amount of engagement)

that the educational system may provide to develop children's thinking abilities in

younger ages may have detrimental effect on the types of thinking tasks that would be

appropriate for them in later ages (Fischer and Bidell, 2006). Under these complex conditions it is therefore not surprising that the issue of creating a coherent

developmental trajectory of thinking goals across the curriculum stood out as a huge

challenge for the experts who participated in the interviews.

One major challenge is that it is not easy to determine the levels of difficulty of

science thinking tasks. This challenge came up both in elementary and high school

science. L1 expressed this challenge when describing the work she did with elementary

teachers on asking/posing questions:

"We have this exercise in which we invite teachers to ask questions on a [given]

text. Sometimes the questions are shallow and low-level, but sometimes we are

beginning to receive complex, high-level questions which are really challenging and

demanding. So how can we create a reasonable progression?" [L1]

Page 18 of 45

Acce

pted M

anus

cript

Similarly, L5 expressed the challenge involved in defining the meaning of a new

high school physics curriculum that will be more demanding in terms of its' levels of

thinking: "When you declare a new program with a higher level of thinking, you need to

define what does a higher level of thinking mean.". L5 emphasizes that over the years

there was no explicit theory for "why a certain task is simpler or more complex, or what

does a student need to do in terms of her thinking in order to provide an answer [to a

given task]". This is because in physics, the complexity of determining the thinking

level of a task is affected by the interaction of several confounding factors such as the

depth and quantity of mathematical understanding required for the task, the difficulty

involved in analyzing the visual information entailed in the task, and the degree to

which non-explicit assumptions are involved. In biology, L3 expressed a similar

concern regarding the level of inquiry projects in biology. She stated that it is perfectly

OK with her if her high school students choose to investigate the effects of salt

concentration on seed germination for their inquiry projects. But such a research project

could also apply to elementary school, and there needs to be a clear difference between

the requirements for these two age groups. For instance, in high school, she would like

students to get into the micro, cellular level rather than stay on the general, macro level.

However, L3 reports that she finds that defining those requirements is not a simple task.

Consequently, it is not easy to explain it to teachers in a clear way.

Another, related, challenge is involved in creating a systematic sequence of

thinking strategies that will be taught across ages and subjects. L2 describes how in the

1996 junior high school syllabus thinking goals were addressed explicitly in the first

time as a list of thinking skills:

"People always talked about the need to do it, but this was the first time that the

syllabus had a page with a list of thinking skills, for better or for worse..[15 years

later] I still dont think we have anything much better than that. I remember we said

that, at least in the syllabus, we should address thinking skills in an orderly manner,

rather than only state that they should be included" [L2].

When the junior high school science syllabus had been updated six years later,

the issue of how to address thinking skills in a systematic way was brought up once

again: "But em at least at the beginning, it was not clear where to address which

thinking strategies and how to do it in a spiral way" [L2]. By the end of the first

Page 19 of 45

Acce

pted M

anus

cript

decade of the 21st century, this challenge has not yet been resolved. By now, states L2, it

has been widely accepted by the experts who plan the curriculum that the teaching of

thinking skills must be integrated into the science content. However, when the experts

attempted to actually execute this idea in the 2009 updated syllabus, they once again

encountered a huge challenge:

"You must think [simultaneously] about several content areas as well as about

quite a large repertoire of thinking strategies you need to teach during those three years

[grades 7-9]. The coordination and the spiral, and repeating the same thinking

strategy in several different contents [which is significant for students' ability to achieve

transfer], and the various levels of thinking, and the adaptation to a variety of learners-

this is a huge challenge even for experts. We certainly can't expect teachers do be able

to do it on their own" [L2].

Interestingly, at the time of the interviews (which took place 16 years after the

1996 syllabus first included an explicit reference to thinking skills), the development of

a draft of a detailed sequence of science inquiry thinking skills for grades 1-9 has been

in progress. L1 and L8 both described this draft document stating that it is being

constructedt in a spiral way, detailing for each grade level 3-4 thinking strategies which

needs to be taught in an explicit, focused way. In addition, the document attempts to

map the curriculum and prevalent textbooks, suggesting content areas and activities in

which it is appropriate to engage in specific thinking strategies. However, since at the

time of the interviews this document was still only a draft, a deeper discussion of this

effort was precluded.

3.2.3 The priority of content goals over thinking goals in policy documents and in

their implementation

The fact that on the pedagogical level it was not entirely clear even to academic

experts what are the best methodologies for integrating HOT goals into the teaching of

science contents, was perhaps one of the factors contributing to the ambiguous

messages sent by policy makers. In the earlier years, it seemed that the policy

documents themselves suffered from inconsistency in this area:

Page 20 of 45

Acce

pted M

anus

cript

The process of trying to write a syllabus for junior high school took many years,

and it [i.e., the inclusion of thinking skills] went back and forth. Thinking skills were

put in, but then they were taken out again.(L2)

However, even after the message in the policy documents had become more

coherent so that curriculum documents emphasized HOT skills as explicit and clear

goals, policy makers as well as teachers still devoted much more resources to teaching

content goals then to teaching thinking goals. When describing the process by which the

schools were instructed by the NSS office how to teach the prescribed curriculum, L8

(who is a practitionaire working in the Ministry) said the following:

"We took out the most important science topics from the curriculum of each

school grade. We counted the number of hours allocated to each topic in the curriculum

and practically divided the school year in terms of what should be taught at each part of

the year. Now as you can see here [pointing to a document written for teachers and

found on-line] - it's all content. There is nothing here about skills. It's true that the

Introduction to the curriculum says that "skills will be studied in an integrated way"

[reading from the introduction]. But there is no policy document showing that [the NSS

office] supported the teaching of thinking in any structured or explicit way" [L8].

L8 then described how following low achievements in the TIMSS and PISA

international tests the Minister of Education and the General Director decided to

allocate more resources to improve science achievements. She stated that the goal of

that program was defined as improving "science knowledge and skills". Despite this

definition, said L8, the field work that followed the policy statement centered almost

exclusively on improving students' knowledge:

"We focused more on content, on what to teach, which topics, principles,

concepts, phenomena and scientific processes [and] much less on the skills, even

though the whole policy move was oriented towards raising achievements, i.e.,

improving knowledge and skills. During the first two years we were working almost

exclusively on the content After two years we said, OK, the teachers are already

teaching the required science topics, we now have to start working on the science skills"

[L8]

Page 21 of 45

Acce

pted M

anus

cript

This report was corroborated by an academic who had been extensively involved

in developing learning kits for teachers (developed as part of implementing that same

policy):

"The [recommended] time lines [for teaching] were arranged around contents.

Now you need to teach energy, now this and now that. Developing the teaching and

learning kits followed a similar pattern. The time-line for their development was so tight

that it was not possible to treat the skills in an appropriate way. I mean, more work was

done on the conceptual dimension. The treatment of the skill dimension was much less

serious from now on, in the next several years, there will be an opportunity to do this

because the learning materials will begin to integrate thinking skills in a more serious

way [L2].

These excerpts show that a similar picture was portrayed by an academic and a

practitioner: although policy documents called for the integration of HOT skills and

science contents, work on implementing this policy across the school system centered

initially on the science content, shying away from the HOT skills. This indicates that the

content goals were perceived as more important and tangible. Only after two years of

extensive work on implementing the new science contents, attention turned to

integrating the thinking into the science contents. During those first several years, a

substantial budget was allocated to raising science scores. This budget was, however,

temporary. By the time this paper is being written, the special budget had already been

reduced in order to attend to newer educational policies in other areas. The big question

is whether by the time the system would be ready to treat the thinking skills seriously,

enough of this special budget will still be available for the substantial processes required

for the implementation of HOT across the school system.

3.2.4 The challenge of human capital: the intricate nature of large scale

professional development (PD) in the context of teaching HOT

Like many recent scholars worldwide (e.g., Barber & Mourshed, 2007), all the

interviewees in the present study viewed teachers as a crucial element in successful

implementation of HOT in science classrooms. They brought numerous examples of

teachers' participation in committees that engaged in policy making, in planning

curricula and in planning the implementation process. These examples affirm that

Page 22 of 45

Acce

pted M

anus

cript

representation of teachers in various national committees responsible for designing

these processes has become a common practice. Yet, similar to the findings of the

previous studies discussed in the literature review , the interviewees see the knowledge

teachers need in order to become proficient in teaching HOT as extremely challenging

and complex. The development of that knowledge is in essence a pedagogical process

which is not easy to achieve even on a small scale project. When this pedagogical

challenge encounters the constraints inherent to large scale implementations, it becomes

even more demanding. The interview data point to two main organizational constraints

characteristic of large scale implementation that affected the feasibility of the deep and

long-term learning teachers need: (a) the amount of time required for deep learning of

the relevant complex knowledge relative to the small number of hours allocated to that

learning in teachers' workshops (see section 3.5.1); This problem was corroborated by

shortage of resources for school-based support for teachers; and, (b) a shortage in a

stable group of highly trained and professional instructors who could lead the work with

teachers (see section 3.5.2)

3.2.5 Conflict between scope and complexity of knowledge goals for teachers and

duration of PD workshops

The enormity of the pedagogical challenge involved in developing the needed

teachers' knowledge conflicted with the organizational structure that was available for

teachers' PD: the administrative infra-structure for PD simply did not allow enough time

to address this complex pedagogical challenge in a satisfactory manner. This idea was

expressed in the interviews in many different ways. L1 who was intensively engaged in

leading teachers' PD workshops elaborated on this issue:

"Getting to know the [thinking] strategies is not enough. You must also learn to

understand their significance, to know how to adapt them to your own classroom's

needs, to experiment with them to be able to lead reflection processes with students

and with other teachers, to evaluate students' work, to be able to give them feedback.

All these elements which are entailed in teaching HOT are not part of the culture we are

used to for example, what feedback will a teacher write to her student when she is

grading a test? There is a lot of work here, you know How do you work with

teachers on acquiring all these tools? Its really really not that simple and we don't

have the time to do it".[L1]

Page 23 of 45

Acce

pted M

anus

cript

In elementary school, an agreement with the teacher union dictates that teachers'

participation in PD would be limited to 30 hours per year. Some of these 30 hours must

be devoted to content knowledge. The remaining hours are viewed as insufficient for

meaningful learning of the complex knowledge teachers need in order to make the

transition to a thinking rich classroom:

"Throughout their professional lives they are used to teaching declarative

knowledge. The curriculum always addresses declarative knowledge. When they write

lessons plans, it's always about declarative knowledge. What do I have to teach

tomorrow? Take a look at teachers' lesson plans and you will see: The respiratory

system, or this or that system Their head is tuned in this direction. Suddenly they are

hearing from us: "No. Not only declarative knowledge. From now on you also need to

address procedural knowledge". This is a paradigm shift. If a science teacher attends

[a workshop] for thirty hours of PD thirty hours means only seven or eight meetings,

and part of the time is devoted to science content. You can't expect a meaningful

learning process in thirty hours What also seems to me extremely important in

assimilating a culture of thinking is to work on habits of mind and on thinking

dispositions so that they would become part of the classrooms culture. I believe it is

extremely important to do so rather than to just be satisfied only with constructing

distinct thinking strategies [L1].

L1 then continue to describe the complex demands of inquiry learning. The

knowledge teachers need to be able to lead inquiry learning is not only complex in itself

compared to the duration of the workshops, but it becomes even more difficult to

address it appropriately in the face of competing and rapidly changing policies:

It's also important to work on the development of complex thinking processes

[such as inquiry] and to give them tools to engage in them.eh but we don't have

enough time to do all that- You can't do all that in thirty hours. And there are always

new policies. This year the NSS declared that inquiry learning is a priority.[quiet for a

while].But along with inquiry they had also decided to promote the subject of health.

You can't do it all at once. Ehh.. in many of the courses this year we worked with

teachers on how to carry out a complete inquiry process from beginning to end. I mean,

starting with encountering a phenomenon, asking questions this is a very difficult

process. We try to make them aware of the complete sequence of inquiry teaching. But

Page 24 of 45

Acce

pted M

anus

cript

this is not enough. We are missing the ability to support them once they are back in their

classrooms. They come to the workshop and they study but I don't know exactly what

they absorbed. The fact that they nodded and told me how exciting this is I don't

knowI am afraid there is a large gap here." [L1]

3.2.6 The challenge of securing a highly-proficient, system-wide, stable

infrastructure of instructors

The second significant challenge in this section pertains to the problem of

securing a highly-proficient, system-wide, stable infrastructure of instructors who can

assist in PD workshops and provide school-based pedagogical support for teachers. The

roles of these professionals include teaching in teachers' PD workshops; participation in

teams that develop model learning materials and assessments; and visiting classrooms to

provide school based feedback and support for teachers. This level of work obviously

requires knowledge that needs to be even more proficient than the knowledge teachers

need for classroom instruction. It therefore takes quite a long time to prepare the high-

quality practitioners needed to fulfill these roles. In all 4 areas (general science, physics,

biology and chemistry) the interviewees reported that they see instructors as an

extremely important link in the implementation process and that considerable resources

were invested in their PD:

"Yes. A second strategy which I really believe in pertains to instructors. Those

who went through the PD courses" (L5)

"Implementation, in all areas, but also specifically in the area of teaching HOT,

takes place on several levels. The first level is that of the NSS and her instructors...

They form a support group. A group that brings up new ideas andtries them out in

experimental pilot projects. This is the group of people with whom I do much of my

work I have 22 instructors. Plans are made together with the instructors who form

the "implementation fan". Then they meet with groups of teachers in the various

districts and disseminate the decisions and plans that we made. Usually we first try it in

pilot projects and then disseminate it to all teachers. [L3]

Page 25 of 45

Acce

pted M

anus

cript

Leading teachers and instructors are usually chosen from teachers who are much

above average in terms of their pedagogical abilities, i.e., they are known as "star"

teachers. Yet, ample resources are often devoted to their long term PD:

"In order to actually implement this policy document [i.e., about teaching HOT],

we first needed to train instructors. We had a complete training system. We trained a

whole group of instructors in fostering HOT" [L1]

"Some groups of instructors met regularly for 6-7 years, other groups met only for

3-4 years.[L2] ".

Following such long term PD workshops, instructors were indeed capable to

support teachers in developing the ability to teach for thinking and understanding:

"Wherever we had projects that supported teachers in their field work we found

really interesting things For instance, [in a Ph.d study conducted under the

supervision of L2], we followed the development of teachers with whom we had

worked for several years. We found a huge development. Then we had quite a few

courses for leading teachers where we also saw a substantial development. It's not true

that you cannot help teachers make progress in this area. But we came to the conclusion

that it's very difficult to do that on a large scale"[L2]

In sum, the basic rationale for building the infra-structure of instructors was to

support the ability of scaling up. The assumption was that this group of professionals

would be able to work with relatively large numbers of teachers as leaders of PD

workshops and as mentors of school-based instruction. Experience as well as a formal

assessment study indicated that it was possible to help this group of excellent teachers

Page 26 of 45

Acce

pted M

anus

cript

make a huge development in their pedagogical knowledge in the context of teaching

thinking. Why then, the pessimistic conclusion regarding the scalability of teachers'

development in this area? The answer came up in many of the interviews when the

interviewees described examples of how various organizational obstacles blocked the

flow of knowledge from instructors to teachers: First, it took several years to prepare

high-quality practitioners who had enough knowledge and experience to guide teachers

in the complex PD processes that needed to take place. Then, there were several

organizational reasons that made this group of experts unstable with large turn-over,

including the following: a. Frequent change of policies required frequent transitions to

other areas of teacher support. Therefore, often after making the long-term investment

in the development of instructor's knowledge in the context of teaching HOT they were

then assigned to work with teachers on other areas (such as ICT or content goals) and

the investment in the development of their knowledge in the context of teaching HOT

was lost; b. Inherent difficulties involved in working with teacher combined with low

salaries for instructors (relative to the time they needed to invest in carrying out their

role), caused a rapid turn-over of instructors. Their less than optimal working conditions

encouraged many of these excellent people to leave their position after a relative short

period, thereby contributing to a disrupted flow of knowledge between policy

documents and classrooms.

3.2.7. Challenges pertaining to metacognitive knowledge

Another important yet complex knowledge element that was found to be difficult

to address during the PD workshops in a satisfactory manner is metacognitive

knowledge. As explained earlier, metacognition is "thinking about thinking" . This

means that in order to be able to engage in metacognitive teaching about HOT, teachers

must first be familiar with the pertinent cognitive processes i.e., with the strategies

involved in the thinking that needs to take place in their classrooms. In addition,

teachers must also acquire the special pedagogical knowledge that pertains to

metacognitive teaching. L1 described how addressing the various components of

metacognition during the PD workshops presented additional challenges:

"Teachers must first experience metacognition in their own learning, they need to

experience learning processes that include metacognition. Then we need to help them

construct two types of meta-level knowledge: First, the meta- strategic knowledge of

Page 27 of 45

Acce

pted M

anus

cript

the thinking strategies which they are not familiar with. Second, the meta-strategic

knowledge necessary for teaching the pedagogy of how to teach metacognition. If

a teacher is not familiar with the "thinking map" [i.e., with the meta-strategic

knowledge of a thinking strategy] it will be very difficult for her to construct a

metacognitive teaching strategy because these two things go together. [quiet]. And

we don't have enough time to do that. We have only 30 hours and we can't fit this in

in a meaningful way"[L1]

Over the years L1 had substantial experience managing numerous elementary

science teachers' workshops. Based on that experience she described how she had

repeatedly witnessed teachers' gaps in knowledge of thinking strategies. She also noted

that the workshops' limited number of hours hindered teachers' ability to acquire the

necessary pedagogical knowledge for teaching metacognition. L1 had been aware of

these limitations. Yet, she sadly stated that the PD workshops she was leading could not

adequately address these knowledge gaps. The problem from her point of view, i.e.

from the point of view of a highly qualified professional who planned the curriculum for

teachers' PD, was that the complexity of the metacognitive knowledge teachers needed,

the fact that it must be built on a prior solid familiarity with thinking strategies , and the

fact that the duration of the PD was limited, contributed to the difficulty of addressing

metacognition in an appropriate way. She then continued to describe how in some of the

courses, they did try to address metacognitive teaching. Yet, she was not happy with the

result. She viewed metacognitive teaching as so challenging that she was skeptical as to

the effect that a limited duration of engaging with the relevant knowledge could have on

teachers' practice. The fact that there was no budget for classroom supervision and

mentoring mades her even more skeptical as to whether the workshop indeed made it

possible for teachers to be able to actually apply metacognition in their classrooms.

A similar gap was also reported in the case of the ability of high school chemistry

teachers to apply metacognitive teaching in their classrooms:

"Metacognition appears in our instructional unit as "Time to Think".., i.e., the

unit includes a section which shows students how to think, a kind of reflection But

not all teachers can work with this section. I just gave a lecture in a workshop for

chemistry teachers and I was astounded to discover that many teachers, even though it

appears in the textbook, do not really understand it. It's not only that they do not teach

Page 28 of 45

Acce

pted M

anus

cript

the metacognition in the relatively simple way it appears in the instructional unit, but

they dont even know how to use it for monitoring their own thinking and for

monitoring their students' responses.even though it appears in the textbook. Well, we

know that the fact that it is written in the textbook does not yet mean that it is present in

the classroom" [L7].

4. Summary and discussion

This study illuminates the challenges involved in large scale implementation of

HOT in science education as they are viewed by a group of leaders who were engaged

in implementation processes in an intensive way. However, since this paper focuses on

challenges, it may give a misrepresentative image. At the outset of the discussion

section it is important to emphasize which conclusions should NOT be drawn from this

study. Focusing on the challenges does not mean that the processes we are studying

were not effective or that large scale implementation efforts in this area are doomed to

failure. The analysis presented here is NOT an evaluation study of the various attempts

to teach HOT in science education in Israel. Although a comprehensive evaluation study

of these efforts had not yet taken place, studies and reports examining sections of these

efforts show considerable effects and progress (see for example Avargil & al., this

volume, Office of Pedagogical Affairs, MOE, 2009; Spector-Levy, Eylon & Schertz,

2008;Gallagher, Hipkins and Zohar 2012). Despite these positive effects, however, the

implementation processes under consideration are far from being completed. Although

considerable developments in the desired direction have been taking place, it is clear

that if the aim is for thinking- rich instruction to become a routine in all science lessons,

there is still a long way to go. The gradual nature of these processes therefore should

allow us to recognize progress but still be concerned by the challenges hindering further

improvements.

Indeed, the data indicate that the route to thinking rich instruction in all

classrooms is neither short nor smooth. The challenges described in this study stress the

fact that introducing HOT to science classrooms does not entail "all or nothing"

processes because implementation processes of new pedagogies are often evolutionary

rather than revolutionary (Dede, 2006; Cohen, 2010)).Cohen (2010) explains that the

answers provided by educators regarding the feasibility of sweeping pedagogical

Page 29 of 45

Acce

pted M

anus

cript

changes are not satisfying. He argues that the question regarding the likelihood of

pedagogical change and its nature needs to be asked from a broad historical perspective,

rather than from a limited view of one isolated link in the chain of time. If we consider

the magnitude of the pedagogical change required to make the transformation to a

thinking-rich curriculum, we must consider the possibility that we are only at the

beginning of a long journey, and that learning from implementation in the field will be

slow. In Cohen's opinion, it is likely that those who strive to promote instruction that he

calls adventurous, are in effect trying to bring about the beginning of a great, slow

change in the perception of knowledge, learning and instruction. This future change,

however, is still in its infancy and needs many more years to materialize. Cohen argues

that the early stages of such huge system-wide changes are characterized by examining

alternatives, inventing new patterns of which only few will survive, and developing

ideologies and strategies for change. Thus, complex pedagogical changes that involve

deep changes in teachers' knowledge and beliefs regarding learning and instruction, are

more correctly viewed as long-term, slow, evolutionary processes, rather than

revolutionary processes. According to this view, it is important to study the challenges

involved in such long-term change processes because understanding them may be

informative for improving future system-wide implementation efforts.

Part of the challenges encountered by the leaders who were interviewed for this

study are common to many types of educational change processes (see figure 1). In

addition, this study revealed several meaningful challenges that are specific to the

educational effort of applying HOT according to the infusion approach across large

numbers of classrooms (see figure 2): challenges involved in weaving HOT into

multiple, varied, specific science contents; challenges involved in planning a reasonable

and coherent developmental sequence of thinking goals; the fact that content goals tend

to have priority over thinking goals and therefore to disperse of the latter in policy

documents and in implementation processes; the challenges involved in supporting

teachers' ability to teach metacognition; and finally, the huge challenges involved in

developing educators sound and deep professional knowledge in the area of teaching

HOT.. Although challenges related to PD are by no means unique to the area of

teaching HOT, the specific nature of these challenges in this area is deeply rooted in the

unique nature of teaching thinking. These challenges are aggravated by lack of adequate

organizational support.

Page 30 of 45

Acce

pted M

anus

cript

Several reasons make metacognition an exceptionally difficult topic to implement

in system-wide PD workshops which almost always suffer from a shortage of hours: the

fact that metacognitive knowledge is itself quite complex, that chronologically speaking

it can only be addressed after participants had acquired a certain degree of knowledge

concerning the cognitive level of using thinking strategies, and the fact that

metacognitive teaching requires particular teaching strategies. Previous studies show

that when a small-scale project addresses metacognitive teaching in a focused way, it

can indeed be effective (Author,, 2006; Veenman, 2011). But when moving to a system-

wide scale, the challenges involved in metacognitive teaching (relative to the many

other important goals that teachers' workshops need to address during a limited period)

make it an especially difficult issue to address in an appropriate way.

More generally, the data indicate that the interviewees viewed the teaching of

HOT according to the infusion approach as requiring complex teachers' knowledge.

They believed it requires a high level of content knowledge, sound knowledge of

thinking strategies, knowledge of complex thinking processes such as problem- based

and inquiry learning, knowledge about the culture of thinking, knowledge about

thinking dispositions, and acquaintance with a variety of pedagogical tools that are

specific to the teaching of HOT.

Due to the magnitude of the shift required in teachers' knowledge while they

make the transition to teaching thinking, one of the interviewees stated that it merits the

label of a "paradigm shift". Addressing this paradigm shift in an appropriate way

requires long-term learning with highly trained instructors. Organizational constrains

which are typical to large scale implementation made it difficult to support the required

long- term teachers learning and therefore hindered the development of teachers

knowledge. These organizational constrains were most apparent in the number of hours

devoted to teachers' learning and in the limited support for instructors which hindered

the feasibility of creating a stable infrastructure of highly professional instructors.

These findings have several practical and research-oriented implications. The

challenges brought up in the interviews point to the directions that future

implementation efforts and future research should center on (in addition to the activities

already conducted in the implementation processes described earlier). First, several of

the findings show gaps in experts' knowledge, indicating that even the experts do not yet

Page 31 of 45

Acce

pted M

anus

cript

have a coherent systematic model and practical plan for weaving HOT into the topics of

the science curriculum over the span of the school years. Consequently, the leaders of

the implementation need to invest time as well as Research and Development funds in

order to: a. conceptualize theory-driven models for weaving HOT systematically into

multiple, varied, specific science contents and, b. for planning a reasonable and coherent

developmental sequence for teaching thinking goals across different ages and student

populations. Subsequently, various such models may be tried out and evaluated

empirically so that decision making as to the optimal model of implementation in

various educational contexts and for different student population would be evidence-

based. At the moment very little research has been conducted around these issues.

Second, leaders and teachers have to learn how to view thinking goals in a way

that would render them as important and tangible as content goals.

Third, the findings have important implications concerning metacognition.

Although ample research conducted in laboratory or small- scale studies in schools

show that metacognition has tremendous effects in improving students reasoning,

hardly any research has been conducted on the effect of metacognitive teaching across

the whole school system. The difficulties shown here concerning teachers' knowledge

and PD in the area of metacognition point to a dangerous potential hazard in scaling up

the teaching of metacognition. More research is needed in order to understand what

elements of metacognitive teaching may be more amenable for relatively short-term PD

and thus less sensitive to the "dilution" (i.e., loss of focus and meaning) which may take

place in large scale implementation. Such research is crucial for a conceptualization of

just what components of metacognition can we hope to teach in a sound way across a

large number of classrooms and just how to conduct wide scale PD in the area of

metacognition.

Finally, and perhaps most importantly, the findings have important implications for PD.

in the context of teaching HOT. The mismatch between the scope of the required PD

according to pedagogical considerations and the administrative and organizational

support for its execution seems to be a major impediment in the ability to apply policies

in this area in a successful way. Recommendations for overcoming this impediment

center on increasing the weight of PD as a determining factor in any implementation

process in this area. The details of the recommendations depend upon the political level

Page 32 of 45

Acce

pted M

anus

cript

supporting the relevant educational policy. If the educational policy advocating

thinking-rich instruction is supported by a stable (in terms of longevity), high-level

political entity that sees education for thinking as a primary goal and has the power to

assign sufficient organizational and financial resources to the implementation process,

then a systematic plan of implementation can be based upon the pedagogical needs for a

high quality and lengthy PD. In this case, pedagogy may drive the planning of

administrative and organizational structures and of an appropriate budget that can

support the complex pedagogical needs. Examples of such comprehensive and long

term support for PD may be seen in the successful large scale efforts of pedagogical

reforms that took place in locations such as Finland or Ontario (Darling-Hammond,

2010; Fullan, 2007). In addition, under such favorable conditions, rational and

systematic models for taking innovations to scale are in place (Lee and Krajcik, 2012)

In many cases however, conditions are less favorable and high level political

support in policies advocating instruction of HOT does not exist. In the case of the

Israeli efforts to foster policies advocating thinking rich instruction in science education,

the thinking curriculum had been embraced by the highest levels within the Ministry of

Education, but it had never been a primary goal of a high level and stable (in terms of

longevity) political entity. I suspect that efforts to implement the thinking curricula in

other countries often operate under similar, less than optimal, conditions. In such cases

it is unlikely that the optimal organizational and administrative conditions needed for

profound PD will be met. Consequently, a gap between the desired and actual teacher

knowledge in the context of teaching thinking will be created and at least some of the

pedagogical goals of the thinking curriculum would have to be compromised.

Since this gap in teacher knowledge is likely to affect the implementation process

of teaching HOT in a fundamental way, it is important to recognize it and take it into

consideration when planning large scale efforts in this area. Some possible directions

for addressing this problem include the following: a. Along with the recommendations

of numerous previous reports (e.g., Darling-Hammond, 2010; Elmore, 2004; Fullan,

2007) , this study calls to give PD priority over other goals; In particular, the findings

point to the need to improve the structural and organizational infra- structure necessary

for supporting deep and long term teacher learning. ; b. An implementation plan cannot

be deeper than the depth of the PD it can support. When a realistic assessment of the

organizational infra-structure for PD processes indicates that PD would necessarily be

Page 33 of 45

Acce

pted M

anus

cript

limited, the thinking goals of the large scale implementation should be re-cons