When the writer studied for his Master’s degree in the late 1960s the only paradigm he was aware of was the "scientific". Yet, a decade later, when he worked for the Schools Council, the "scientific" paradigm’s relevance was being seriously challenged by researchers trained in sociology and anthropology who used approaches associated with the "socio-anthropological" (interpretative or qualitative) paradigm.

As Husén (1988) commented:

The twentieth century has seen conflict between two main paradigms employed in researching educational problems. The one is modelled on the natural sciences with an emphasis on empirical quantifiable observations which lend themselves to analyses by means of mathematical tools. The task of research is to establish causal relationships, to explain (Erklären). The other paradigm is derived from the humanities with an emphasis on holistic and qualitative information and to interpretive approaches (Verstehen). (Husén, 1988, p17)

The criticisms of the "scientific" paradigm were in essence that:

Statistical findings failed to provide the ‘qualitative’f information required to help explain why something happened (Parlett & Hamilton, 1976). Burstall et.al. (1974) found that the effects of teaching modern languages in primary school appeared to make little if any difference on final achievement in secondary school; but she failed to consider whether the secondary schools used the skills learned by the primary school pupils.

The constraints behind ‘controlled’ research designs (Parlett & Hamilton, 1976; Reid, 1979) were perceived as rigid and artificial; raising questions about the design’s suitability as the basis of valid and generaliseable comments. Much of the arguments criticising the use of the agricultural-botanical model are exemplified by the problems Barker-Lunn (1970) faced in her research into teaching pupils in non-streamed primary schools. Finally, as Cooley (1971) commented:

If there were no differences the innovators were convinced that the results were not valid and continued to show that their innovations were superior to other programmes available in schools. If the results of the comparison indicated that the innovation was superior to brand X, the innovators congratulated themselves and the evaluator on a job well done. Those who were skeptical of the innovation found flaws in the design and proceeded to debunk the validity of the proposition. (Cooley, 1971)

More recently, Gronn & Ribbins (1993) emphasised the inadequacy of the traditional science paradigm as a basis for underlying theoretical explanations within their field of educational administration:

Griffiths (1988 p.30), for example, accepts that its ‘demise came at the 1974 meeting of the IIP in Bristol’. Indeed the ‘coup de grâce was delivered by Greenfield who made an across the board denunciation of every aspect of the theory movement.’ (Gronn & Ribbins, 1993, p.161)

During the 1980’s and 1990’s the use of the "socio-anthropological" paradigm continued to increase. The change in paradigm emphasis was very evident when four general research texts on the writer’s bookshelf were examined; Bassey (1995) failed to mention the agricultural-botanical model and Blaxter et.al. (1996), Bell (1993) and Hopkins (1993) only referred to it critically:

the usual form of educational research, the psycho-statistical or agricultural-botany paradigm, has severe limitations as a method of construing and making sense of classroom reality. (Hopkins, 1993 p41)

Notwithstanding this, the writer has found that the "scientific" paradigm is a relevant and effective approach for tackling many curriculum questions. The insights of the "scientific" paradigm with respect to objectivity, validity, reliability, comparability and generaliseability combined with its speed of use have much to offer.

This view is supported by the increased involvement by the Government, since Callaghan’s Ruskin College Speech in 1977 and the 1988 Education Reform Act. In the national curriculum the two key elements shared by the "scientific" paradigm and systematic curriculum theory are the importance of clarifying what should be taught (aims and attainment targets) and the need to ensure that the teaching has been successful (publication of Standard Attainment Test’s, GCSE examination results and Ofsted inspection reports are all kinds of summative evaluation).

In the remainder of this web page, the initial emphasis is on clarifying  the "scientific" paradigm. It does this by considering three questions:

What is research?

What is a paradigm?

What is the "scientific" paradigm?

In her presidential address to the British Educational Research Association, Broadfoot (1988), an educational sociologist who works within the "socio-anthropological" paradigm, argued that the key underlying philosophical concepts in educational research were common to both paradigms, stating that:

As educational researchers we form part of that scholarly community which upholds and seeks to promulgate, the values of systematic enquiry and respect for evidence. ...most of us would nevertheless claim that we work within a set of ethical principles which are a distinctive characteristic of our calling. In the most general terms these ethical principles centre on the pursuit of truth. This does not mean that we embrace any absolute notion of truth since for most of us in the social sciences, this cannot be; but rather we seek to describe, illuminate, portray and hopefully sometimes even explain, that section of reality that serves as a focus for our particular enquiry. Recognising that we can never divorce ourselves from our own values - some of which we may not even be aware of overtly - we try to generate insights which are dictated as closely as we can make them by the data themselves. These data are in turn collected according to the canons of procedure which years of experience within the disciplines have consecrated as the most impartial and illuminating basis for scholarship in that field.

It is this scientific culture that we seek to teach our students when we urge them to distinguish between fact and opinion, evidence and value-judgment. When we teach them the skills of systematic enquiry, interpretation and analysis. Indeed from their earliest days in school, pupils are taught to observe, consider, report and interpret, distinguishing this process from the experiential, intuitive, emotional response which characterises other parts of their learning.

....It is clear that educational research, like the other so-called social sciences occupies an uneasy conceptual and methodological middle ground between naturalistic science on the one hand and interpretive disciplines such as history and literature on the other. Indeed it can and should build on the strengths of both. (Broadfoot, 1988, pp5-6)

Broadfoot’s view is that both empirical educational paradigms can claim that they are sciences. That does not mean to say, however, that the commonalities overcome the philosophical differences between them.

A paradigm is, therefore, perceived to be the underlying philosophical concept that structures thinking in disciplines. Since the research approaches initially used in education drew on that used in Psychology and Biology, the traditional research paradigm can be historically described as "scientific" (Tall, 1994a).

Inevitably, when a research paradigm is used in a new field of study like education, it would be surprising if methodological problems did not occur and even more surprising if the more traditional users of that paradigm didn’t criticise new developments. In some cases, however, the changes made are far less than those that developed in the natural sciences. Biology found it necessary to control its experiments (the agricultural-botanical model) but this did not create a new paradigm because what happened was perceived as a refinement, a development of the traditional research approach. When such changes are small, and do not contradict underlying belief systems, then inevitably over a period of time the developments become absorbed into the original paradigm. A case in point was that described by Merrett (1999) who told the writer that behaviourist psychologists faced such serious problems in initially getting their research published that they had to found a new journal. The behaviourists’ experiments (Merrett & Wheldall, 1986; Tall 1994b) differed from traditional research designs by comparing the same child’s behaviour over time rather than comparing the behaviour of different groups of children synchronously. But, perhaps the most important difference was that the behaviourists insisted that, because the changes in behaviour they were looking for had to be large, it was not necessary to check results using statistical tests. In retrospect, as far as the writer is concerned, the problems faced by Merrett and his behaviourist colleagues are difficult to believe. Philosophically their experimental approach is identical to those used by biologists and psychologists. Their decision not to carry out statistical tests to demonstrate a difference in behaviour is ultimately a foible. In the examples Merrett uses to illustrate his methodology (Merrett & Wheldall, 1978, 1988, 1992) the evidence is such that the null hypothesis would be rejected by the Sign test.

When, however, methodological changes are based on widely different philosophical positions and the changes in methodology are so great that researchers from both sides criticise each other over a long period of time, then accommodation to include both approaches only occurs slowly if at all. This was clearly the case when educationalists with a psychological background considered the work of researchers from a socio-anthropological background and vice versa. The underlying philosophical difference between the two paradigms is based on the "socio-anthropological" belief, that education could not be studied effectively using the traditional "scientific" paradigm because it requires a holistic stance. The "scientific" paradigm’s concept that everything could be reduced to quantitative information was also rejected. The difference in positions, and their effect on research methods, are so great that the differences between the two groups of researchers can be classified as being of a paradigmatic nature. Whilst the term paradigm is not accepted by all researchers: Davies (1994) insists on using the term ‘qualitative’ research rather than "socio-anthropological" paradigm. The differentiation of the two types of research remains evident, Davies (1994) herself described the reality of the two paradigms as follows:

While the metaphor for quantitative research might be the flight simulator (accurate sampling or reality to test phenonema), the metaphor for qualitative research is more like the hot air balloon - hovering over a particular area with no necessarily clear idea of where to land, and jettisoning old theories like sand-bags. You will realise that choice of methodology is not just a technical concern, but relates to something deep in the psyche of the researcher: you are either comfortable with qualitative research or you are not. A qualitative approach will demonstrate the variability of behaviour according to the context and the subjective interpretations of the actors in that context. It is messy, untidy and subject to constant revision and to constant criticism. While it is possible and indeed desirable in ‘quantitative’ research to begin with a hypothesis or to test stated beliefs, in qualitative research one is often required to ‘suspend beliefs’ altogether - to go in knowing one’s biases, but being prepared to act at least initially, as a blotting paper pad to absorb whatever splotches of data come your way. (p.5)

In the writer’s view however, the term paradigm is essential, because the difference is attitudinal and not simply the collection of different kinds of data. Indeed, the writer commonly collects ‘qualitative’ data in his curriculum research although his research approach lies strongly within the "scientific" paradigm.

As Rist (1977) argued:

Ultimately, the issue is not research strategies per se. Rather, the adherence to one paradigm opposed to another predisposes one to view the world and events within it in profoundly different ways. (p43)

Bassey (1995) agreed with Broadfoot (1988) concerning paradigm commonalties, but argued for their separation:

Both of these forms of research entail systematic, critical and self-critical enquiry which aims to contribute to the advancement of knowledge, but the divide between them is so great that until recently the practitioners of the former were inclined to deny that the latter is research. (Bassey, 1995, p7)

Whilst educational researchers differ about the existence of additional empirical paradigms in educational research (some colleagues who use Action Research at Birmingham argue that is a paradigm in its own right though others, such as O’Hanlon (1997), disagree; the research module’s External Examiner told the writer that he believed gender researchers have a distinct paradigm {Scott, 1997}), there is a general acceptance of the existence of the "scientific" and the "socio-anthropological". Representatives of both groups of researchers also accept the need for being eclectic in the evidence they collect when the research question warrants it. As Bell (1993) commented:

Different styles, traditions or approaches use different methods of collecting data, no approach prescribes nor automatically rejects any particular method. Quantitative researchers collect facts and study the relationship of one set of facts to another. They measure, using scientific techniques that are likely to produce quantified and, if possible, generaliseable conclusions. Researchers adopting a qualitative perspective are more concerned to understand individuals' perceptions of the world. They seek insight rather than statistical analysis.... Yet there are occasions when qualitative researchers draw on quantitative techniques and vice versa. (Bell, 1993, 4)

Reichardt and Cook (1979) commented:

researchers in no way follow the principles of a supposed paradigm without simultaneously assuming methods and values of the alternative paradigm (p232)

This position of distinguishing between the philosophical concept/attitude/belief system of the researcher and the methods generally used is emphasised by the writer when teaching the University of Birmingham’s MEd research module. To help students understand this Figures 1 and 2 are used.

Figure 1 illustrates that from a paradigm viewpoint researchers looks at questions with a mind set that tends to result in the use of some methods and the avoidance of others. The Figure goes on to indicate that researchers within a paradigm can choose methods normally used by the other.

Figure 1

In Figure 2, the commonality of the two paradigms is emphasised by the scientific aspects they share, namely in Broadfoot’s (1988) words, "The pursuit of truth" (p5) an acceptance that data must be both valid and reliable, that a literary search is necessary and an acceptance of Occam’s Razor# . Given the common research platform, then the identification of an individual’s paradigm can normally be recognised by the balance of different research methods used. Only when the choice of methods is truly eclectic, is the balance approximately horizontal and the identification of the researcher’s paradigm less possible and arguably less important.

Figure 2

Further, whilst the two paradigms can be used to research virtually any question, it is evident that the traditional "scientific" paradigm is most relevant where the information required to answer the research question can be easily specified e.g. whether the:

examination results with the new modular examination are comparable to those obtained previously? Or, when the questions asked are concerned with simple data collection (monitoring).

The "socio-anthropological" paradigm appears more relevant when there is uncertainty about the information required or even the question being studied and a holistic approach is logically required.

In applied research it is evident that the underlying research question and the audience for the report are at least as crucial in determining research methods as the researcher’s own paradigm. Parlett & Hamilton (1976) demonstrated this clearly when they argued that whilst illuminative evaluation was based on the insights of the socio-anthropological paradigm, it was eclectic and hence when needed would collect ‘quantitative’ information. Similarly in Tall & Langtree (1988) the research is clearly based on the "scientific" paradigm but included open-ended questions in the questionnaire and ‘respondent’ interviews (Powney & Watts, 1988).

b) providing clear information to allow their research to be repeated.

 The writer carried out such a re-analysis of Reid’s (1967) work as part of his study for his M.Ed (see Tall (1994b)). Several of the writer’s students have similarly contacted researchers so that they could reanalyse the original work.

The limitation of Comte’s analysis is that he ignored the fact that scientific research itself requires insight - conceptual understandings that are rarely immediately explicable in terms of empirical science and rational methods. Charles Darwin was well aware of the concept of evolution, a theory that was espoused by his grandfather, Erasmus. The problem was that the only mechanism postulated for evolution, Lamarkism, was not supported by experiments.

Though criticised by philosophers, such as Kuhn (1970), for not accurately reflecting the untidy reality of scientific research, the quotation from Nuffield biology illustrates the simplified mechanism that future scientists are taught in schools.

Being a scientific detective means that you must be able to observe and record your observations accurately, make your own hypothesis from your records and invent experiments to test your hypothesis.

We could represent the stages of enquiry thus

PROBLEM -------------» HYPOTHESIS -------------» EXPERIMENT -------------» RESULT

(Nuffield Foundation, 1966 p.32)

Figure 3 summarises scientist’s concept of their approach, emphasising inductive and deductive aspects.

What is not given sufficient emphasis is that scientific research is impossible without an understanding and knowledge of what is being studied and an insight into what should be looked for. Charles Darwin demonstrated in his Origin of the Species that the "scientific" paradigm inevitably begins with, or is based upon, extensive qualitative observation (data collection) and subsequent induction (creation of hypothesis or possible explanation).

Figure 3

As Toulmin commented:

However full a notebook one kept of the phenomena one came across…. It would in all probability be of no use to physicists at all. In physics, its no use even beginning to look at things until you know exactly what you are looking for: observation has to be strictly controlled by reference to some particular theoretic problem. (Toulmin, 1953, p54)

Certainly, in the writer’s own curriculum research, he has had to begin with open-ended interviews (section 5a and 5b) or work with colleagues who had the necessary information (sections 7, 10 and 11) so that relevant instruments could be designed.

In education, teachers develop or propose a new grouping method or teaching approach. The hypothesis is that the grouping method or teaching approach will be more effective than what was used before. It is the hypothesis that is the basis for the experimental design.

The inductive-deductive sequence specified in Figure 3 reflects the "scientific" paradigm in educational experiments. When MEd students and others assume that "scientific" experiments begin with a hypothesis, they have ignored the reality that what they wish to investigate is based on data and induction already carried out. Relatively rarely, in a practical field like curriculum studies, are the hypotheses tested developments of general theories.

Dewey (1933) generally argued that the "scientific" paradigm attempts to ensure it can make legitimate generalisations by:

a) Carefully defining the problem that needed to be solved.

b) Clarifying potential limiting factors and attempting to ‘control’ them through the research design (see subsequent discussion on the agricultural-botanical model). This may involve using a large number of examples.

c) Observing, collecting and classifying evidence relevant to the problem.

e) Formulating tentative hypotheses (inductive logic) on the basis of the observations.

f) Assessing often using statistical tests, the probability of obtaining the observed results by chance.

g) Formulating conclusion(s) based on the evidence collected in the light of the defined problem and context of the research.

A negative description, though one with some truth in it, concerning the "scientific" paradigm would be "If it moves count it" (Tall, 1994b). The underlying belief system of the "scientific" paradigm is the assumption that if something is important it can be clearly described, and that once described it can be measured. The question is how, and it is in that aspect that traditional users of the "scientific" paradigm would disagree with the writer, who would accept combinations of relatively subjective attitude questions:

What are creativity, love altruism? Unlike variables such as weight, blood pressure and temperature, they cannot be measured on a scale, sphygmomanometer, or thermometer, in units of pounds, millimetres of mercury or degrees Fahrenheit. Instead they can be thought of as unifying constructs or labels that characterise responses to related groups of variables. For example, answers of "strongly agree" to items such as "sends me flowers", "listens to my problems", "reads my manuscripts", "laughs at my jokes," and "gazes deeply into my soul" may lead you to conclude that love is present. Thus love is not a single measurable entity but a construct which is derived from other, directly observable variables. Identification of such underlying dimensions - factors - greatly simplifies the description and understanding of complex phenomena like social interaction      (Norusis, 1992, p47)

Norusis’s explanation for factor analysis illustrates the fact that it is the "scientific" paradigm’s equivalent to "socio-anthropological" researchers’ holistic analysis.

Major differences between the "scientific" and the "socio-anthropological" paradigms relate to the fact that, the "scientific" paradigm attempts to be as objective as possible along the ‘Subjective Objective’ dimension. "Scientific" researchers’ try to stand back and not affect their research findings. They attempt to identify precisely the key elements that need to be measured, demonstrate the validity of so doing, by checking the elements relevance with experts in the field. The emphasis is on obtaining quantifiable data and where relevant using experimentation as an integral part of the design. Both the "socio-anthropological" and the "scientific" paradigm deem reliability to be essential, but the latter assesses this primarily through repeat measures and by assessing internal consistency. By specifying clearly what has been done "scientific" researchers make replication possible. Because of the care taken in sampling the "scientific" paradigm researcher believes that an estimation can be made, using statistical tests, of the probability with which the conclusions drawn can be generalised.

Whilst researchers in both paradigms recognise the importance of research validity and reliability, Davies (1994) argued that a weakness of the "scientific" paradigm is that its approach reduces its internal validity, that for research to be valid one cannot simply study particular elements. The difference between the two paradigms is highlighted in Figure 4 which emphasises the need for all researchers to be eclectic. "Scientific" researchers have to ensure that the information they collect is as valid as possible, whilst "socio-anthropological" researchers need to recognise the value of using quantitative measures to increase reliability.

With such an emphasis on objectivity, it is hardly surprising that "scientific" research findings can be very persuasive to policy and decision makers. However, the lack of the ‘human’ element found in descriptive and persuasive case-studies makes it less attractive and persuasive to classroom teachers. The argument that "scientific" research is objective has been criticised by researchers from the "socio-anthropological" paradigm on the grounds that researchers cannot avoid affecting their research findings (Harber, 1996). Valid criticisms have been made that tests devised in a specific culture may disadvantage individuals from a different ethnic group. The "scientific" paradigm response is that objectivity is not based on the researcher being an automaton, but on open statements of hypotheses, specification of evidence to be collected, making research data collected available, all of which enable readers to check each aspect of the research, the methods used and the validity of the conclusions drawn.

Figure 4: Validity and Reliability in the two paradigms.

What is particularly relevant in the "scientific" paradigm is the fact that by using statistical insights in populations and samples, it can both focus-down and generalise its findings. In Tall & Langtree (1988) the whole population of teachers was given the questionnaire and, in that instance, teachers were identified for further study who had had problems with the pastoral education process (see section 9). Focusing down could equally be to a ‘fair’ sample, one that represented the views of the whole staff. Generalisation, with the aid of statistical tests, can be used on data from a ‘fair’ sample to discover the views of the population. Both approaches inevitably help to reduce the amount of time needed to carry out curriculum research.

"Scientific" research is about probability rather than proof. The only definite statement that can be made about ‘quantitative’ data is that where there is diversity there will always be exceptions; but that the existence of exceptions must not be used to indicate that valid generalities cannot be meaningfully drawn. While statistics summarise individual data, it is inaccurate to state that such data is necessarily lost. Schools calculate the numbers of students with five or more GCSE grade C and above, that does not mean that they do not know and cannot discuss with candidates what they should do next.

"Scientific" proof is not an absolute guarantee. As Popper (1959) commented theories in science cannot be finally proved, all that science can do is discover enough evidence to support them (chapter 10).

In the writer’s view, the particular strength of the "scientific" paradigm is that it encourages doubt, it emphasises the necessity of experimentation or comparison to obtain the necessary evidence to challenge the proposed hypothesis. Nothing about "scientific" research should be used to state that all teachers must teach in a particular way; generalisations are made to provide advice - mixed ability teaching in secondary schools (section 6) and Active Tutorial Work (section 9) can be successfully introduced but, they require something more than headteacher determination.

Finally, it is generally argued, that the experimental skills used by "scientific" researchers are more easily learned and the researchers more interchangeable than with "socio-anthropological" research. That the methods used by the "scientific" paradigm allow greater numbers of individuals to be studied, questioned or tested in the same amount of time. All encourage the use of the scientific paradigm where it is relevant; none mean that it can easily replace the "socio-anthropological" paradigm when in-depth open-ended investigations are required.

Figure 5 The "Scientific" Research Paradigm

Description is the basic characteristic necessary to portray what exists. Such findings inevitably led by inductive reasoning to the development of hypotheses and experiments - whilst experiments such as Galileo’s dropping of objects of very different weights from the top of the Leaning Tower of Pisa might have been demonstrations rather than true experiments; Galileo must have carried out earlier experiments to check his hypothesis that gravity acted similarly on light and heavy objects.

Figure 6: Controlled Experiments in Education

 Figure 6 is an attempt by the writer to reclassify non-descriptive research carried out in the "scientific" paradigm. The underlying theme of the classification is that there is a continuum between the different kinds of experiment. The external control classification represents the research most commonly described as the agricultural-botanical model. The development and criticisms of the externally controlled experiments (Figure 7) and criticisms of their use will be referred to shortly. Scriven (1967) described such research as Summative Evaluation and the model’s biological roots are evident in its name:

Figure 7 Agricultural/Botanical Model: Compares Two or More Groups Simultaneously.

The reasons for the development of synchronous ‘external control’ designs were two-fold:

Changes in education are not easy to measure. And

Education takes time and in that time, variables other than those being assessed can possibly affect what happens. The logical solution is to teach the experiment and the control at the same time.

The three major criticisms aimed at the model are:

The ‘Hawthorne Effect’

Summative evidence which demonstrated what happened but failed to explain it - see Burstall’s (1974) research. And

The presence of uncontrolled variables.

All three criticisms are based on large scale national evaluations and are, in the view of the writer, much less relevant in small scale research where pupil groups can be carefully ‘matched’ and the same teacher can be used for both experimental and control groups. It should not be assumed that the above criticisms are only relevant with the "scientific" research paradigm; the Hawthorne effect and uncontrolled variables, are problems that occur whenever changes in practice take place. Indeed, they may be heightened when the researcher’s involvement is substantial as commonly happens in Action Research.

The distinction between ‘external’ and ‘internal’ control designs made in figure 6 is that the ‘internal control’ designs ensure matching of experimental and control groups by using the individual or group as its own control. This approach is used in before and after experiments (section 7 of this thesis) and in behaviourist research (Merrett & Wheldall, 1978) etc.

The final experimental classification used by the writer recognises the reality that there is an ‘invisible control’ design. Such designs occur where comparisons are made with recognised standards. In Figure 6 they are placed adjacent to descriptive designs. The only underlying difference between "invisible controls" and descriptive research is that comparisons are made in the former to an external standard. Such comparisons commonly occur in formative and summative evaluations when the researcher compares outcomes using the new on the basis of past experience. Invisible control is evident in national curriculum assessments; criticisms of school effectiveness in SATs are based on whether pupils achieved the ‘levels’ identified for them by experts on national curriculum working parties. Formative evaluation (see Figure 8) was a crucial component in USA. and UK national curriculum projects in the sixties and seventies and used a variety of both ‘quantitative’ and ‘qualitative’ techniques (Schools Council, 1973). The range of evidence collected inevitably varied in different projects, but the underlying reason for collecting qualitative as well as quantitative data was that it was necessary to know whether the teaching process envisaged was usable ‘in the real world’. In other words, formative evaluation required data which demonstrated that the product was being achieved (quantitative) and information which ensured that it was likely to be used (qualitative).

It is evident therefore that data obtained using qualitative methods can fit within the "scientific" paradigm as long as every attempt is made to use effective sampling approaches and to collect the information as objectively as possible. Put simply, it would be ridiculous for a traditional scientist to consider that open-ended interviews, open-ended questions in a questionnaire and unstructured observations, could not be used by researchers coming from the "scientific" paradigm if they helped fill perceived gaps in the quantitative evidence collected.

Figure 8 National Project Evaluations in 1960’s to 1980’s

 Formative evaluation, when used by individual institutions, was initially called school/self evaluation. That developed by "socio-anthropological" researchers is commonly called Action Research that used in schools by researchers from the "scientific" paradigm is likely to be called Monitoring and Evaluation.

The paradigm’s strengths are that it emphasises the necessity of:

careful sampling;

specifying exactly what data is/needs to be collected and how that data should be analysed;

collecting evidence objectively;

checking hypotheses by experimentation; i.e. it is cautious of accepting new ideas simply because they seem appear successful;

the need for cross checking, comparability; and, the use of methods which provide a rapid means of collecting valid and reliable information.

In the writer’s opinion the "scientific" paradigm includes:

the need to eclectically obtain relevant qualitative data

and the need to present quantitative information in an easily assimilated format.

The paradigm’s value, in national terms, is its emphasis on validity, reliability, objectivity and its belief in comparability; its value for schools lies in the relative speed in which data can be collected and analysed. For schools and colleges ‘quantitative’ evidence is essential for monitoring effectiveness, identifying precisely what has happened and in reporting such findings to external decision-makers. Its particular advantage is that with careful sampling the findings can be generalised and comparisons can be made. It is, accordingly, relevant for small-scale evaluations where the emphasis is on whether aims have/have not been achieved. That does not mean that the "scientific" paradigm is the only or the even the most suitable approach to tackle a particular issue, but it is an insistence that the paradigm is a valid way of tackling many of the questions raised in schools and curriculum studies and has much to offer schools and teachers in evaluating their work.

In the remainder of this thesis the writer illustrates the potential of the "scientific" paradigm in curriculum studies using Figure 9 overleaf (the combined classification of Figures 5 and 6 above) which illustrates the concept that descriptive and experimental research are ultimately part of a continuum:

Figure 9 The "Scientific" Research Paradigm