Educational Research (1)
Written by T. Neville Postlethwaite, Institute of Comparative Education University of Hamburg. from Educational Research: Some Basic Concepts and Terminology. I. Introduction Research is the orderly investigation of a subject matter for the purpose of adding to knowledge. Research can mean
‘re-search’ implying that the subject matter is already known but, for one reason or another, needs to be studied again.
Alternatively, the expression can be used without a hyphen and in this case it typically means investigating a new problem or phenomenon. Within the realm of educational planning, many things are always changing: the structure of the education system, curriculum and textbooks, modes of teaching, methods of teacher training, the amount and type of provisions to schools such as science laboratories, textbooks, furniture, classroom supplies, and so on.
These changes may lead to an improvement, or a worsening, in the quality of an educational system. Sometimes they may result in no impact upon quality - in which case major government expenditures on such changes have been wasted. The educational planner working within this kind of environment must be able to undertake assessments of the effects of major changes and then provide policy advice that will consolidate and extend the post productive courses of action, and also intercept and terminate existing practices that are shown to be damaging and wasteful.
II. Types of educational research There are many types of educational research studies and there are also a number of ways in which they may be classified. Studies may be classified according to topic whereby the particular phenomena being investigated are used to group the studies. Some examples of educational research topics are: teaching methods, school administration, classroom environment, school finance, etc. Studies may also be classified according to whether they are exploratory or confirmatory.
An exploratory study is undertaken in situations where there is a lack of theoretical understanding about the phenomena being investigated so that key variables, their relationships, and their (potential) causal linkages, are the subject of conjecture. In contrast a confirmatory study is employed when the researcher has generated a theoretical model (based on theory, previous research findings, or detailed observation) that needs to be tested through the gathering and analysis of field data.
A more widely applied way of classifying educational research studies is to define the various types of research according to the kinds of information that they provide. Accordingly, educational research studies may be classified as follows:
1. Historical research generates descriptions, and sometimes attempted explanations, of conditions, situations, and events that have occurred in the past. For example, a study that documents the evolution of teacher training programs since the turn of the century, with the aim of explaining the historical origins of the content and processes of current programs.
2.Descriptive research provides information about conditions,situations, and events that occur in the present. For example, a survey of the physical condition of school buildings in order to establish a descriptive profile of the facilities that exist in a typical school.
3. Correlational research involves the search for relationships between variables through the use of various measures of statistical association. For example, an investigation of the relationship between teachers’ satisfaction with their job and various factors describing the provision and quality of teacher housing, salaries, leave entitlements, and the availability of classroom supplies.
4. Causal research aims to suggest causal linkages between variables by observing existing phenomena and then searching back through available data in order to try to identify plausible causal relationships. For example, a study of factors related to student ‘drop out’ from secondary school using data obtained from school records over the past decade.
5.Experimental research is used in settings where variables defining one or more ‘causes’ can be manipulated in a systematic fashion in order to discern ‘effects’ on other variables. For example, an investigation of the effectiveness of two new textbooks using random assignment of teachers and students to three groups - two groups for each of the new textbooks, and one group as a ‘control’ group to use the existing textbook.
6. Case study research generally refers to two distinct research approaches. The first consists of an in-depth study of a particular student, classroom, or school with the aim of producing a nuanced description of the pervading cultural setting that affects education, and an account of the interactions that take place between students and other relevant persons.
For example, an in-depth exploration of the patterns of friendship between students in a single class. The second approach to Case Study Research involves the application of quantitative research methods to non-probability samples - which provide results that are not necessarily designed to be generalizable to wider populations. For example, a survey of the reading achievements of the students in one rural region of a particular country.
7. Ethnographic research usually consists of a description of events that occur within the life of a group - with particular reference to the interaction of individuals in the context of the sociocultural norms, rituals, and beliefs shared by the group. The researcher generally participates in some part of the normal life of the group and uses what he or she learns from this participation to understand the interactions between group members.
For example, a detailed account of the daily tasks and interactions encountered by a school principal using observations gathered by a researcher who is placed in the position of ‘Principal’s Assistant’ in order to become fully involved in the daily life of the school.
8. Research and development research differs from the above types of research in that, rather than bringing new information to light, it focuses on the interaction between research and the production and evaluation of a new product.
This type of research can be ‘formative’ (by collecting evaluative information about the product while it is being developed with the aim of using such information to modify and improve the development process). For example, an investigation of teachers’ reactions to the various drafts and redrafts of a
new mathematics teaching kit, with the information gathered at each stage being used to improve each stage of the drafting process.
Alternatively, it can be ‘summative’ (by evaluating the worth of the final product, especially in comparison to some other competing product). For example, a comparison of the mathematics achievement of students exposed to a new mathematics teaching kit in comparison with students exposed to the established mathematics curriculum.
Educational Research (2)
Written by T. Neville Postlethwaite, Institute of Comparative Education University of Hamburg from Educational Research: Some Basic Concepts and Terminology III. Three Types of Research Questions in Educational PlanningIn research on issues concerned with educational l planning, the main educational research questions can be subsumed under three categories: descriptive, correlational, and causal.
1. Descriptive questions In the field of educational planning, the research carried out on descriptive questions is often focused on comparing t he existing conditions of schooling with: (i) leg isolated benchmark standards,(ii) conditions operating in several other school systems, or (iii) conditions operating in several sectors of a single school system. Some examples are:
• What is the physical state of school buildings in the country?
• Do some districts or regions have better or worse school buildings than others? (Behind these two questions are the implications that the Ministry of Education wishes to ensure that all schools have a minimum standard of school building while at the same time ensuring t hat there are not large differences among schools with respect to the state of their buildings.)
What is the level of achievement in the core subject areas at a particular level of schooling? Does such achievement accord with the Ministry’s view of what should have been learned by all students or particular sub-groups of students? A further question is often raised about student achievement – is it better,worse, or the same as last year’s achievement for the particular grade group? And, again, are there differences between regions,or urban and rural children, and so on? (By the early 1990s there was a thrust from some systems of education – especially that of the United States of America – to have comparisons of achievement (at the same age level) of different national systems of education for countries at a similar level of economic development.)
2. Correlational questions Behind these kinds of questions, there is often an assumption that if an association is found between variables then it provides evidence of causation. However, care must be exercised when moving between the notions of association and causation. For example, an ‘association’ may be discovered between the incidence of classroom libraries and average class reading scores. However,the real ‘cause’ of higher reading scores may be that students from high socio-economic backgrounds, while they tend to be in classes with classroom libraries, read better than other students because their home environments (in terms of physical, emotional, and intellectual resources) facilitate the acquisition of reading skills.Some examples are:
• Do students in poorer school buildings have lower achievement scores than those in better buildings?
• Do students in better equipped classrooms have better achievement scores than those in less well-equipped classrooms? Do students in schools where the teachers have better teacher housing have higher achievement than students in schools where teachers have poorer teacher housing?
3. Causal questions Causal questions are usually the most important to educational planners. For example, in some schools it is considered normal for children to have a desk at which to sit. In other schools the children sit on the ground and write on their laps. It is important to know if schools (with a particular socio-economic background of children) with a shortage of desks and seats achieve less well than schools (with a similar socio-economic background of children) with an adequate supply of desks and chairs.
Or, to put the question in a different way, is it the desks and chairs, or something else, which
really cause the better achievement? It may be a better supply of books or better qualified teachers or, or, or.... It is, therefore, important to disentangle the relative influence of each of the many input and process factors in schools on achievement. As will be seen from another module in this series on ‘Research Design’ both survey and experimental designs can be used to assess the relative influence of many factors on educational achievement.
It is unusual in education to find only one factor influencing student educational achievement. It is rather the case that several, or even many, factors from outside and inside the school influence how well or poorly students achieve in school. Thus, causal questions take one of two forms. Some examples are:
• All other factors being equal do students with Textbook A achieve better than students with Textbook B? What is the relative effect on school achievement of the following factors:
• the socio-economic level of students in the school;
• the general parental help given to the children with their homework;
• peer group pressure;
• the condition of the school buildings;
• the supplies and equipment in the classroom;
• the curriculum;
• the quality of teaching, etc.
IV. Sequential Stages in The Research Process In order for the research to proceed in a focused and systematic manner, these questions must be refined to form more specific research questions that indicate exactly which target populations and which variables or factors should be included in the research study.
1. Literature review The review of literature aims to describe the ‘state of play’ in the area selected for study. That is, it should describe the point reached by the discipline of which the particular research study will form
a part. An effective literature review is not merely a summary of research studies and their findings. Rather, it represents a ‘distillation’ of the essential issues and inter-relationships associated with the knowledge, arguments, and themes that have been explored in the area. Such literature reviews describe what has been written about the area, how this material has been received by other scholars, the major research findings across studies, and the major debates in terms of substantive and methodological issues.
2. Research design Given the specific research questions that have been posed, a decision must be taken on whether to adopt an experimental design for the study or a survey design. Further, if a survey design is to be used, a decision must be taken on whether to use a longitudinal design, in which data are collected on a sample at different points of time, or a cross-sectional design, in which data are collected at a single point of time.
Once the variables on which data are to be collected are known, the next questions are: Which data collection ‘units’ are to be employed? and Which techniques should be used to collect these data? That is, should the units be students, the teachers, the school principals, or the district education officers. And should data be collected by using observations, interviews, or questionnaires? Should data be collected from just a few hand-picked schools (case study), or a probability sample of schools and students (thus allowing inferences from the sample to the population), or a census in which all schools are included? For a case study, the sample is known as a ‘sample of convenience’ and only limited inferences can be made from such a sample.
For research that aims to generalize its research findings, a more systematic approach to sample selection is required. Detailed information on the drawing of probability samples for both experimental and survey design is given in another module in this series entitled ‘Sample Design’.
3. Instrumentation Occasionally, data that are required to undertake a research study already exist in Ministry files, or in the data archives of research studies already undertaken, but this is rarely the case. Where data already exist, the analysis of them is known as “secondary data analysis”. But, usually, primary data have to be
collected. From the specific research questions established in the first step of a research study it is possible to determine the indicators and variables required in the research, and also the general nature of questionnaire and/or test items, etc. that are required to form these.
Decisions must then be taken on the medium by which data owe to be collected (questionnaires, tests, scales, observations, and/or interviews). Once these decisions have been taken, the instrument construction can begin. This usually consists of the writing (or borrowing) of test items, attitude items, and questionnaire items. The items should be reviewed by experienced practitioners in order to ensure that they are unambiguous, and that they will elicit the required information. The broad issue of ‘Instrumentation’ (via both tests and questionnaires) has been taken up in more detail in several other modules in this series.
to be continued
