1 Background

One of the basic principles of the University of Heerlen is to offer students the opportunity to qualify in one or more disciplines on the basis of their own independently pursued development. For the Open University of Heerlen (OU), involving students with widely differing backgrounds, variations in the levels of prior knowledge cause particular problems. This probably holds for all higher and adult education. Furthermore, one of the principles of open modular education is that the students should be free to choose the order in which they take the modules or courses. This leads to problems for course team leaders and experts in defining exact entrance requirements and for students with regard to correctly estimating their entrance level and their learning progress.

In addition, it is essential that the OU, having been recommended in the HERP (1990) as the future national centre for the development of modules, is able to offer research-based support, with regard to entrance indicators, and should be able to offer guidelines and a variety of learning paths to students. These indicators (like sex, age, educational background, etc.) do play a role in the registration behaviour of potential OU students. It is likely to inform and to help these students before and during their study, and thereby enable them to achieve sufficient success and a satisfactory completion rate in an effective way. Studies in educational psychology have indicated that the prior knowledge is a major variable in education. Thirty to sixty per cent or more of the variation in academic performance is explained by the prior knowledge of the students.

The yield of higher education does not match expectations (HERP,1990) and is capable of considerable improvement. Since prior knowledge is the factor with the greatest influence, reaching an understanding of it and developing ways of tackling the problems associated with it offers a solution to this problem which can provide results within the short term.

Until now, electronic interactive media have made up only a minimal part of the OU's total education package. It is expected that more use will be made of electronic interactive learning and testing systems in future. The potential of these testing systems for making the learning process more individually oriented is especially important, since it means that education can be adapted to the prior knowledge and requirements of individual students.

There can be no doubt that interest in the problem of prior knowledge has been reawakened since Ausubel (1968), in his Textbook on Educational Psychology, wrote: "The most important single factor influencing learning is what the learner already knows". Subsequently, dozens of authors have referred to the determining influence of prior knowledge on the learning process and on academic results (Glaser, 1984). Despite this attention, however, Peeck concluded in 1979 that prior knowledge has, until recently, hardly been chosen as an object of investigation. Still today, in our view there is too little attention paid to the subject.

The reported research project started from several specific problems, encountered at the University of Heerlen. Many of the problems encountered in the practice of open learning relate to the students' prior knowledge. In the first place, open learning assumes equal opportunities for all students, but seldom does the arrangement of the subjects with regard to time and content genuinely take account of the academic experience of the individual students (de Wolf, 1985). Secondly, a new polarity must be the starting point for making open learning as accessible as possible. On the one side are the students, with practical experience, specific prior knowledge, particular requirements, questions and intentions, who wish to widen their knowledge, understanding and skills; on the other side are knowledge bases, areas of subject matter with definite characteristics and content, which therefore call for a certain study behaviour (de Wolf, 1985). Good open learning involves the construction of theoretically and empirically supported bridges between these two.

Open, modular education works with building blocks, which in theory can be put together in any desired order. In practice, it is not always possible to realize this degree of flexibility in the sequence in which course modules are studied, a fact that is in conflict with the above principle. In a genuinely open learning situation, the course modules should be much more "switchable". The sequence a student chooses is determined by his prior knowledge. The question is, to what extent should this be dependent on, or independent of, the prior knowledge.

Although OU courses include information on the assumed prior knowledge of the student, this information is often fairly vague, so it is difficult to ascertain the degree to which a student fulfils the pre-requirements. The information is often based on vague and inconsistent theories of cognition and acquisition of knowledge by adults. Whether the prior knowledge is present remains uncertain; whether the knowledge, if it is indeed present, is available remains unclear. Until now, it is hardly known whether students themselves give consideration to this information on the assumed prior knowledge and whether they take it into account when taking the decision to study.

The essence of the research project can be formulated in five questions.

First, there is the question for a general theoretical context for investigating prior knowledge and the specific context for the present study.

The second question relates to the definition and operationalization of the concept 'prior knowledge', namely, how can prior knowledge be mapped?

The third question is, how can prior knowledge be measured? Consideration can be given to methods such as the use of variables as indicators, the construction of thinking-aloud protocols, knowledge tests, methods for the analysis of cognitive structures, such as free recall or cued recall, domain specific tests, work sample tests, portfolio assessment, etc.

The fourth question defines the specific relationship with the study results: What is the role and influence of the prior knowledge with relation to the learning process and the academic results? Here attention is directed not simply towards assessment; also of interest are the ways in which certain results or conclusions can be used. Are there different kinds of prior knowledge which influence studying and how can we use the assessment results to optimize the learning? Are variations in prior knowledge of different students interfering with the multifunctional design of courses?

The fifth question is: Is there evidence to assume that is it possible to construct a system or instrument based on prior knowledge that will support the student in acquiring knowledge in the fields of study planning, the selection of subject matter, and didactic tuition, so that the academic yield is increased?

The reported investigation will mainly focus on these questions. Further, we will try to interpret the research results in relation to recent instructional and cognitive psychological theories.

3 Overview of this study

This book is structured in relation to the above stated questions which contribute successively to the solution of the described problem. After a general introductory chapter that clarifies and identifies the problem and the subsequent questions facing this study, chapter 2 and 3 provide an answer to the first question i.e. a general theoretical framework and the specific context for the present study. Chapter 2 reviews the recent literature which supports the importance of prior knowledge. Different effects of prior knowledge are given and we try to explain these effects using current theories and research results. From an information-processing point of view, the relationship between the different explanations is worked out.

The important concepts of this study which seem to need a clear description, as stated in the previous chapters, are defined in chapter 3. This chapter provides an answer to the second question. Further, in order to construct a conceptual map of prior knowledge terminology, categories of and differences in prior knowledge are explicated. The final part of this third chapter introduces an indexation of the 'prior knowledge' concept based on literature and views of experts and overviews methods for the assessment of prior knowledge, as formulated in the third question. The final part of this chapter sustains the choice for investigating domain-specific prior knowledge by using recent research results.

After this theoretical framework, an answer is given to the fourth of the earlier stated questions in chapters 4 to 6.

An introduction to the empirical studies is given in chapter 4. This chapter starts an empirical search in order to answer the question concerning assessment methods. An exploratory study of students' views on prior knowledge is described and an overview is given of studies into students' prior knowledge in economics, the domain of the present study. Further, the population involved and the course which is subject to our investigations are described. The proceeding chapters search successively for an adequate answer to the fifth question. Chapter 5 discusses a first ex post facto research, investigating the impact of prior knowledge of Economics and Law students on their study results. This investigation looks for evidence for the earlier-stated problem, namely the multifunctional nature of modules and for the assumptions about differences in prior knowledge and study results of students.

In chapter 6, a second ex post facto research, we study the possibility of using direct measurable personal and contextual variables as indicators of prior knowledge. If this is the case, an easy and cheap method for determining students' prior knowledge levels is found. If this is not feasible, we will search for another useful method. The most promising method, with concrete relations to support and remediation of the learning process, seems then to be the so called prior knowledge state tests. Chapter 7 gives the background and description of the different prior knowledge state tests used in the next phases of investigation.

Related to the fifth question, chapters 8 to 10 discuss empirical investigations concerning the use of prior knowledge state tests with different student populations. A set of tests tries to assess the complex of prior knowledge state components and aims at searching for the most relevant components at the subject matter level. In a subsequent investigation (chapter 9), we look for additional support for the results of the preceding chapter and for our 'context-transfer approach', i.e. the transferability of results between student populations. Chapter 10 reports investigations in which the design and analysis go beyond the subject matter level. For different student populations the use of knowledge profiles as a means of determining students' prior knowledge state is tried out.

A concluding chapter 11 formulates all results of the investigations and discusses implications for instructional and cognitive psychological theory and for educational practice.