Types of Research
All of us engage in actions that have some of the characteristics of formal research, although perhaps we do not realize this at the time.
We try out new methods of teaching, new materials, new textbooks. We compare what we did this year with what we did last year. Teachers frequently ask students and colleagues their opinions about school and classroom activities. Counselors interview students, faculty, and parents about school activities. Administrators hold regular meetings to gauge how faculty members feel about various issues. School boards query administrators, administrators query teachers, teachers query students and each other. We observe, we analyze, we question, we hypothesize, we evaluate. But rarely do we do these things systematically. Rarely do we observe under controlled conditions. Rarely are our instruments as accurate and reliable as they might be. Rarely do we use the variety of research techniques and methodologies at our disposal.
The term research can mean any sort of “careful, systematic, patient study and investigation in some field of knowledge. Basic research is concerned with
clarifying underlying processes, with the hypothesis
usually expressed as a theory. Researchers engaged in
basic research studies are not particularly interested
in examining the effectiveness of specific educational practices. An example of basic research might be an attempt to refi ne one or more stages of Erickson’s psychological theory of development.
Applied research, on the other hand, is interested in examining the effectiveness of particular educational practices. Researchers engaged in applied research studies may or may not want to investigate the degree to which certain theories are useful in practical settings. An example might be an attempt by a researcher to fi nd out whether a particular theory of how children learn to read can be applied to first graders who are non-readers. Many studies combine the two types of research. An example would be a study that examines the effects of particular teacher behaviors on students while also testing a theory of personality.
Many methodologies fit within the framework of research. If we learn how to use more of these methodologies where they are appropriate and if we can become more knowledgeable in our research efforts, we can obtain more reliable information upon which to base our educational decisions. Let us look, therefore, at some of the research methodologies we might use.
QUANTITATIVE AND QUALITATIVE RESEARCH
Another distinction involves the difference between quantitative and qualitative research. Although we shall discuss the basic differences between these two types of research more fully in its own icon attached in my web page, we will provide a brief overview here. In the simplest sense, quantitative data deal primarily with numbers, whereas qualitative data primarily involve words. But this is too simple and too brief. Quantitative and qualitative methods differ in their assumptions about the purpose of research itself, methods utilized by researchers, kinds of studies undertaken, the role of the researcher, and the degree to which generalization is possible. Quantitative researchers usually base their work on the belief that facts and feelings can be separated, that the world is a single reality made up of facts that can be discovered.
Qualitative researchers, on the other hand, assume that the world is made up of multiple realities, socially constructed by different individual views of the same situation. When it comes to the purpose of research, quantitative researchers seek to establish relationships between variables and look for and sometimes explain the causes of such relationships. Qualitative researchers, on the other hand, are more concerned with understanding situations and events from the viewpoint of the participants. Accordingly, the participants often tend to be directly involved in the research process itself.
Quantitative research has established widely agreedon general formulations of steps that guide researchers in their work. Quantitative research designs tend to be preestablished.
Qualitative researchers have a much greater fl exibility in both the strategies and techniques they use and the overall research process itself. Their designs tend to emerge during the course of the research.
The ideal researcher role in quantitative research is that of a detached observer, whereas qualitative researchers tend to become immersed in the situations in which they do their research. The prototypical study in the quantitative tradition is the experiment; for qualitative researchers, it is an ethnography.
Lastly, most quantitative researchers want to establish generalizations that transcend the immediate situation or particular setting. Qualitative researchers, on the other hand, often do not even try to generalize beyond the particular situation, but may leave it to the reader to assess applicability. When they do generalize, their generalizations are usually very limited in scope.
Many of the distinctions just described, of course, are not absolute. Sometimes researchers will use both qualitative and quantitative approaches in the same study. This kind of research is referred to as mixed-methods research. Its advantage is that by using multiple methods, researchers are better able to gather and analyze considerably more and different kinds of data than they would be able to using just one approach. Mixed-methods studies can emphasize one approach over the other or give each approach roughly equal weight.
Consider an example. It is often common in surveys to use closed-ended questions that lend themselves to quantitative analysis (such as through the calculation of percentages of different types of responses), but also open-ended questions that permit qualitative analysis (such as following up a response that interviewees give to a particular question with further questions by the researcher in order to encourage them to elaborate and explain their thinking).
EXPERIMENTAL RESEARCH
Experimental research is the most conclusive of scientific methods. Because the researcher actually establishes different treatments and then studies their effects,results from this type of research are likely to lead to the most clear-cut interpretations.
Suppose a history teacher is interested in the following question: How can I most effectively teach important concepts (such as democracy or colonialism) to my students? The teacher might compare the effectiveness of two or more methods of instruction (usually called the independent variable) in promoting the learning of historical concepts. After systematically assigning students to contrasting forms of history instruction (suchas inquiry versus programmed units), the teacher could compare the effects of these contrasting methods by testing students’ conceptual knowledge. Student learning in each group could be assessed by an objective test or some other measuring device. If the average scores on the test (usually called the dependent variable) differed, they would give some idea of the effectiveness of the various methods. A simple graph could be plotted to show the results, as illustrated in Figure 1.
In the simplest sort of experiment, two contrasting methods are compared and an attempt is made to control for all other (extraneous) variables—such as student ability level, age, grade level, time, materials, and teacher characteristics—that might affect the outcome under investigation.
Methods of such control could include holding the classes during the same or closely related periods of time, using the same materials in both groups, comparing students of the same age and grade level, and so on. Of course, we want to have as much control as possible over the assignment of individuals to the various treatment groups, to ensure that the groups are similar. But in most schools, systematic assignment of students to treatment groups is difficult, if not impossible, to achieve. Nevertheless, useful comparisons are still possible.You might wish to compare the effect of different teaching methods (lectures versus discussion, for example) on student achievement or attitudes in two or more intact history classes in the same school. If a difference exists between the classes in terms of what is being measured, this result can suggest how the two methods compare, even though the exact causes of the difference would be somewhat in doubt.
Another form of experimental research, single-subject research, involves the intensive study of a single individual (or sometimes a single group) over time. These designs are particularly appropriate when studying individuals with special characteristics by means of direct observation.
CORRELATIONAL RESEARCH
Another type of research is done to determine relationships among two or more variables and to explore their implications for cause and effect; this is called correlational research. This type of research can help us make more intelligent predictions. For instance, could a math teacher predict which sorts of individuals are likely to have trouble learning the subject matter of algebra? If we could make fairly accurate predictions in this regard, then perhaps we could suggest some corrective measures for teachers to use to help such individuals so that large numbers of “algebra-haters” are not produced. How do we do this? First, we need to collect various kinds of information on students that we think are related to their achievement in algebra. Such information might include their performance on a number of tasks logically related to the learning of algebra (such as computational skills, ability to solve word problems, and understanding of math concepts), their verbal abilities, their study habits, aspects of their backgrounds, their early experiences with math courses and math teachers, the number and kinds of math courses they’ve taken, and anything else that might conceivably point to how those students who do well in math differ from those who do poorly.
We then examine the data to see if any relationships exist between some or all of these characteristics and subsequent success in algebra. Perhaps those who perform better in algebra have better computational skills or higher self-esteem or receive more attention from the teacher. Such information can help us predict more accurately the likelihood of learning difficulties for certain types of students in algebra courses. It may even suggest some specific ways to help students learn better. In short, correlational research seeks to investigate the extent to which one or more relationships of some type exist. The approach requires no manipulation or intervention on the part of the researcher other than administering the instrument(s) necessary to collect the data desired. In general, one would undertake this type of research to look for and describe relationships that may exist among naturally occurring phenomena, without trying in any way to alter these phenomena.
CAUSAL-COMPARATIVE RESEARCH
Another type of research is intended to determine the cause for or the consequences of differences between groups of people; this is called causal-comparative research. Suppose a teacher wants to determine whether students from single-parent families do more poorly in her course than students from two-parent families. To investigate this question experimentally, the teacher would systematically select two groups of students and then assign each to a single- or two-parent family—which is clearly impossible (not to mention unethical!). To test this question using a causal-comparative design, the teacher might compare two groups of students who already belong to one or the other type of family to see if they differ in their achievement. Suppose the groups do differ. Can the teacher definitely conclude that the difference in family situation produced the difference in achievement? Alas, no. The teacher can conclude that a difference does exist but cannot say for sure what caused the difference Interpretations of causal-comparative research are limited, therefore, because the researcher cannot say conclusively whether a particular factor is a cause or a result of the behavior(s) observed.
In the example presented here, the teacher cannot be certain whether (1) any perceived difference in achievement between the two groups is due to the difference in home situation, (2) the parent status is due to the difference in achievement between the two groups (although this seems unlikely), or (3) some unidentified factor is at work. Nevertheless, despite problems of interpretation, causal-comparative studies are of value in identifying possible causes of observed variations in the behavior patterns of students. In this respect, they are very similar to correlational studies.
SURVEY RESEARCH
Another type of research obtains data to determine specific characteristics of a group. This is called survey research. Take the case of a high school principal who wants to find out how his faculty feels about his administrative policies. What do they like about his policies? What do they dislike? Why? Which policies do they like the best or least? These sorts of questions can best be answered through a variety of survey techniques that measure faculty attitudes toward the policies of the administration. A descriptive survey involves asking the same set of questions (often prepared in the form of a written questionnaire or ability test) of a large number of individuals either by mail, by telephone, or in person. When answers to a set of questions are solicited in person, the research is called an interview. Responses are then tabulated and reported, usually in the form of frequencies or percentages of those who answer in a particular way to each of the questions.
The difficulties involved in survey research are mainly threefold: (1) ensuring that the questions are clear and not misleading, (2) getting respondents to answer questions thoughtfully and honestly, and (3) getting a sufficient number of the questionnaires completed and returned to enable making meaningful analyses.
The big advantage of survey research is that it has the potential to provide us with a lot of information obtained from quite a large sample of individuals. If more details about particular survey questions are desired, the principal (or someone else) can conduct personal interviews with faculty. The advantages of an interview (over a questionnaire) are that open ended questions (those requiring a response of some length) can be used with greater confidence, particular questions of special interest or value can be pursued in depth, follow-up questions can be asked, and items that are unclear can be explained.
ETHNOGRAPHIC RESEARCH
In all the examples presented so far, the questions being asked involve how well, how much, or how efficiently knowledge, attitudes, or opinions and the like exist or are being developed. Sometimes, however, researchers may wish to obtain a more complete picture of the educational process than answers to the above questions provide. When they do, some form of qualitative research is called for. Qualitative research differs from the previous (quantitative) methodologies in both its methods and its underlying philosophy.
Trying to reconcile the two approaches lets take into consideration the subject of physical education. Just how do physical education teachers teach their subject? What kinds of things do they do as they go about their daily routine? What sorts of things do students do? In what kinds of activities do they engage? What explicit and implicit rules of games in Physical Education (PE) classes seem to help or hinder the process of learning? To gain some insight into such concerns, an ethnographic study can be conducted. The emphasis in this type of research is on documenting or portraying the everyday experiences of individuals by observing and interviewing them and relevant others. An elementary classroom, for example, might be observed on as regular a basis as possible, and the students and teacher involved might be interviewed in an attempt to describe, as fully and as richly as possible, what goes on in that classroom.
Descriptions (a better word might be portrayals) might depict the social atmosphere of the classroom; the intellectual and emotional experiences of students; the manner in which the teacher acts toward and reacts to students of different ethnicities, sexes, or abilities; how the “rules” of the class are learned, modified, and enforced; the kinds of questions asked by the teacher and students; and so forth. The data could include detailed prose descriptions by students of classroom activities, audiotapes of teacher-student conferences, videotapes of classroom discussions, examples of teacher lesson plans and student work, sociograms depicting “power” relationships in the classroom, and flow charts illustrating the direction and frequency of certain types of comments (for example, the kinds of questions asked by teacher and students of one another and the responses that different kinds produce).
In addition to ethnographic research, qualitative research includes historical research and several other, less commonly used approaches. Casey (1996), for example, has identified 18 types of “narrative” methods. Among them, four are the most distinctive. These include biography, where the researcher focuses on important experiences in the life of an individual and interacts with the person to clarify meanings and interpretations (e.g., a study of the career of a high school principal). In phenomenology, the researcher focuses on a particular phenomenon (such as school board conflict), collects data through in-depth interviews with participants, and then identifies what is common to their perceptions. A third approach is the case study, in which a single individual, group, or important example is studied extensively and varied data are collected and used to formulate interpretations applicable to the specifi c case (e.g., a particular school board) or to provide useful generalizations. Lastly, grounded theory emphasizes continual interplay between raw data and the researcher’s interpretations that emerge from the data. Its central purpose is to inductively develop a theory from data (e.g., a study of teacher morale in a particular school beginning with interviews and other types of data).
HISTORICAL RESEARCH
You are probably already familiar with historical research. In this type of research, some aspect of the past is studied, either by perusing documents of the period or by interviewing individuals who lived during the time. The researcher then attempts to reconstruct as accurately as possible what happened during that time and to explain why it did. For example, a curriculum coordinator in a large urban school district might want to know what sorts of arguments have been made in the past as to what should be included in the social studies curriculum for grades K–12. She could read what various social studies and other curriculum theorists have written on the topic and then compare their positions. The major problems in historical research are making sure that the documents or individuals really did come from (or live during) the period under study and, once this is established, ascertaining that what the documents or individuals say is true.
ACTION RESEARCH
Action research differs from all the preceding methodologies in two fundamental ways. The first is that generalization to other persons, settings, or situations is of minimal importance. Instead of searching for powerful generalizations, action researchers (often teachers or other education professionals, rather than professional researchers) focus on getting information that will enable them to change conditions in a particular situation in which they are personally involved. Examples would include improving the reading capabilities of students in a specifi c classroom, reducing tensions between ethnic groups in the lunchroom at a particular middle school, or identifying better ways to serve special education students in a specifi ed school district. Accordingly, any of the methodologies discussed earlier may be appropriate.
The second difference involves the attention paid to the active involvement of the subjects in a study (i.e., those on whom data is collected), as well as those likely to be affected by the study’s outcomes. Commonly used terms in action research, therefore, are participants or stakeholders, reflecting an intent to involve them directly in the research process as part of “the research team.” The extent of participation varies from just helping to select instruments and/or collect data to helping to formulate the research purpose and question to actually participating in all aspects of the research investigation from start to finish.
EVALUATION RESEARCH
There are many different kinds of evaluations depending on the object being evaluated and the purpose of the evaluation. Evaluation research is usually described as either formative or summative. Formative evaluations are intended to improve the object being evaluated; they help to form or strengthen it by examining the delivery of the program or technology and the quality of its implementation. In contrast, summative evaluations seek to examine the effects or outcomes of an object by describing what happens after the delivery of the program or technology in order to assess whether the object caused the outcome.
An example of a formative evaluation product is a needs assessment report. A needs assessment determines the appropriate audience for the program, as well as the extent of the need and what might work to meet the need. Summative evaluations can be thought of as either (a) outcome evaluations, which investigate whether the program or technology appeared to have caused demonstrable effects on specifi cally defi ned target outcomes, or (b) impact evaluations, which are broader and attempt to assess the overall effects (intended or unintended) of the program or technology as a whole.
Evaluators ask many different kinds of questions and often use a variety of methods to address them. For example, in summative evaluations, evaluators often use quasi-experimental research designs to assess the hypothesized causal effects of a program. Formative evaluations that examine program implementation may also include analysis of existing data sources, surveys, interviews, observational data, and focus groups.
ALL HAVE VALUE
It must be stressed that each of the research methodologies described so briefl y above has value for us in education. Each constitutes a different way of inquiring into the realities that exist within our classrooms and schools and into the minds and emotions of teachers, counselors, administrators, parents, and students. Each represents a different tool for trying to understand what goes on, and what works, in schools. It is inappropriate to consider any one or two of these approaches as superior to any of the others. The effectiveness of a particular methodology depends in large part on the nature of the research question one wants to ask and the specifi c context within which the particular investigation is to take place. We need to gain insights into what goes on in education from as many perspectives as possible, and hence we need to construe research in broad rather than narrow terms.
As far as we are concerned, research in education should ask a variety of questions, move in a variety of directions, encompass a variety of methodologies, and use a variety of tools. Different research orientations, perspectives, and goals should be not only allowed but encouraged. The intent of this book is to help you learn how and when to use several of these methodologies.
K. Casey (1995, 1996). The new narrative research in education. Review of Research in Education, 21: 211–253.
Applied research, on the other hand, is interested in examining the effectiveness of particular educational practices. Researchers engaged in applied research studies may or may not want to investigate the degree to which certain theories are useful in practical settings. An example might be an attempt by a researcher to fi nd out whether a particular theory of how children learn to read can be applied to first graders who are non-readers. Many studies combine the two types of research. An example would be a study that examines the effects of particular teacher behaviors on students while also testing a theory of personality.
Many methodologies fit within the framework of research. If we learn how to use more of these methodologies where they are appropriate and if we can become more knowledgeable in our research efforts, we can obtain more reliable information upon which to base our educational decisions. Let us look, therefore, at some of the research methodologies we might use.
QUANTITATIVE AND QUALITATIVE RESEARCH
Another distinction involves the difference between quantitative and qualitative research. Although we shall discuss the basic differences between these two types of research more fully in its own icon attached in my web page, we will provide a brief overview here. In the simplest sense, quantitative data deal primarily with numbers, whereas qualitative data primarily involve words. But this is too simple and too brief. Quantitative and qualitative methods differ in their assumptions about the purpose of research itself, methods utilized by researchers, kinds of studies undertaken, the role of the researcher, and the degree to which generalization is possible. Quantitative researchers usually base their work on the belief that facts and feelings can be separated, that the world is a single reality made up of facts that can be discovered.
Qualitative researchers, on the other hand, assume that the world is made up of multiple realities, socially constructed by different individual views of the same situation. When it comes to the purpose of research, quantitative researchers seek to establish relationships between variables and look for and sometimes explain the causes of such relationships. Qualitative researchers, on the other hand, are more concerned with understanding situations and events from the viewpoint of the participants. Accordingly, the participants often tend to be directly involved in the research process itself.
Quantitative research has established widely agreedon general formulations of steps that guide researchers in their work. Quantitative research designs tend to be preestablished.
Qualitative researchers have a much greater fl exibility in both the strategies and techniques they use and the overall research process itself. Their designs tend to emerge during the course of the research.
The ideal researcher role in quantitative research is that of a detached observer, whereas qualitative researchers tend to become immersed in the situations in which they do their research. The prototypical study in the quantitative tradition is the experiment; for qualitative researchers, it is an ethnography.
Lastly, most quantitative researchers want to establish generalizations that transcend the immediate situation or particular setting. Qualitative researchers, on the other hand, often do not even try to generalize beyond the particular situation, but may leave it to the reader to assess applicability. When they do generalize, their generalizations are usually very limited in scope.
Many of the distinctions just described, of course, are not absolute. Sometimes researchers will use both qualitative and quantitative approaches in the same study. This kind of research is referred to as mixed-methods research. Its advantage is that by using multiple methods, researchers are better able to gather and analyze considerably more and different kinds of data than they would be able to using just one approach. Mixed-methods studies can emphasize one approach over the other or give each approach roughly equal weight.
Consider an example. It is often common in surveys to use closed-ended questions that lend themselves to quantitative analysis (such as through the calculation of percentages of different types of responses), but also open-ended questions that permit qualitative analysis (such as following up a response that interviewees give to a particular question with further questions by the researcher in order to encourage them to elaborate and explain their thinking).
EXPERIMENTAL RESEARCH
Experimental research is the most conclusive of scientific methods. Because the researcher actually establishes different treatments and then studies their effects,results from this type of research are likely to lead to the most clear-cut interpretations.
Suppose a history teacher is interested in the following question: How can I most effectively teach important concepts (such as democracy or colonialism) to my students? The teacher might compare the effectiveness of two or more methods of instruction (usually called the independent variable) in promoting the learning of historical concepts. After systematically assigning students to contrasting forms of history instruction (suchas inquiry versus programmed units), the teacher could compare the effects of these contrasting methods by testing students’ conceptual knowledge. Student learning in each group could be assessed by an objective test or some other measuring device. If the average scores on the test (usually called the dependent variable) differed, they would give some idea of the effectiveness of the various methods. A simple graph could be plotted to show the results, as illustrated in Figure 1.
In the simplest sort of experiment, two contrasting methods are compared and an attempt is made to control for all other (extraneous) variables—such as student ability level, age, grade level, time, materials, and teacher characteristics—that might affect the outcome under investigation.
Methods of such control could include holding the classes during the same or closely related periods of time, using the same materials in both groups, comparing students of the same age and grade level, and so on. Of course, we want to have as much control as possible over the assignment of individuals to the various treatment groups, to ensure that the groups are similar. But in most schools, systematic assignment of students to treatment groups is difficult, if not impossible, to achieve. Nevertheless, useful comparisons are still possible.You might wish to compare the effect of different teaching methods (lectures versus discussion, for example) on student achievement or attitudes in two or more intact history classes in the same school. If a difference exists between the classes in terms of what is being measured, this result can suggest how the two methods compare, even though the exact causes of the difference would be somewhat in doubt.
Another form of experimental research, single-subject research, involves the intensive study of a single individual (or sometimes a single group) over time. These designs are particularly appropriate when studying individuals with special characteristics by means of direct observation.
CORRELATIONAL RESEARCH
Another type of research is done to determine relationships among two or more variables and to explore their implications for cause and effect; this is called correlational research. This type of research can help us make more intelligent predictions. For instance, could a math teacher predict which sorts of individuals are likely to have trouble learning the subject matter of algebra? If we could make fairly accurate predictions in this regard, then perhaps we could suggest some corrective measures for teachers to use to help such individuals so that large numbers of “algebra-haters” are not produced. How do we do this? First, we need to collect various kinds of information on students that we think are related to their achievement in algebra. Such information might include their performance on a number of tasks logically related to the learning of algebra (such as computational skills, ability to solve word problems, and understanding of math concepts), their verbal abilities, their study habits, aspects of their backgrounds, their early experiences with math courses and math teachers, the number and kinds of math courses they’ve taken, and anything else that might conceivably point to how those students who do well in math differ from those who do poorly.
We then examine the data to see if any relationships exist between some or all of these characteristics and subsequent success in algebra. Perhaps those who perform better in algebra have better computational skills or higher self-esteem or receive more attention from the teacher. Such information can help us predict more accurately the likelihood of learning difficulties for certain types of students in algebra courses. It may even suggest some specific ways to help students learn better. In short, correlational research seeks to investigate the extent to which one or more relationships of some type exist. The approach requires no manipulation or intervention on the part of the researcher other than administering the instrument(s) necessary to collect the data desired. In general, one would undertake this type of research to look for and describe relationships that may exist among naturally occurring phenomena, without trying in any way to alter these phenomena.
CAUSAL-COMPARATIVE RESEARCH
Another type of research is intended to determine the cause for or the consequences of differences between groups of people; this is called causal-comparative research. Suppose a teacher wants to determine whether students from single-parent families do more poorly in her course than students from two-parent families. To investigate this question experimentally, the teacher would systematically select two groups of students and then assign each to a single- or two-parent family—which is clearly impossible (not to mention unethical!). To test this question using a causal-comparative design, the teacher might compare two groups of students who already belong to one or the other type of family to see if they differ in their achievement. Suppose the groups do differ. Can the teacher definitely conclude that the difference in family situation produced the difference in achievement? Alas, no. The teacher can conclude that a difference does exist but cannot say for sure what caused the difference Interpretations of causal-comparative research are limited, therefore, because the researcher cannot say conclusively whether a particular factor is a cause or a result of the behavior(s) observed.
In the example presented here, the teacher cannot be certain whether (1) any perceived difference in achievement between the two groups is due to the difference in home situation, (2) the parent status is due to the difference in achievement between the two groups (although this seems unlikely), or (3) some unidentified factor is at work. Nevertheless, despite problems of interpretation, causal-comparative studies are of value in identifying possible causes of observed variations in the behavior patterns of students. In this respect, they are very similar to correlational studies.
SURVEY RESEARCH
Another type of research obtains data to determine specific characteristics of a group. This is called survey research. Take the case of a high school principal who wants to find out how his faculty feels about his administrative policies. What do they like about his policies? What do they dislike? Why? Which policies do they like the best or least? These sorts of questions can best be answered through a variety of survey techniques that measure faculty attitudes toward the policies of the administration. A descriptive survey involves asking the same set of questions (often prepared in the form of a written questionnaire or ability test) of a large number of individuals either by mail, by telephone, or in person. When answers to a set of questions are solicited in person, the research is called an interview. Responses are then tabulated and reported, usually in the form of frequencies or percentages of those who answer in a particular way to each of the questions.
The difficulties involved in survey research are mainly threefold: (1) ensuring that the questions are clear and not misleading, (2) getting respondents to answer questions thoughtfully and honestly, and (3) getting a sufficient number of the questionnaires completed and returned to enable making meaningful analyses.
The big advantage of survey research is that it has the potential to provide us with a lot of information obtained from quite a large sample of individuals. If more details about particular survey questions are desired, the principal (or someone else) can conduct personal interviews with faculty. The advantages of an interview (over a questionnaire) are that open ended questions (those requiring a response of some length) can be used with greater confidence, particular questions of special interest or value can be pursued in depth, follow-up questions can be asked, and items that are unclear can be explained.
ETHNOGRAPHIC RESEARCH
In all the examples presented so far, the questions being asked involve how well, how much, or how efficiently knowledge, attitudes, or opinions and the like exist or are being developed. Sometimes, however, researchers may wish to obtain a more complete picture of the educational process than answers to the above questions provide. When they do, some form of qualitative research is called for. Qualitative research differs from the previous (quantitative) methodologies in both its methods and its underlying philosophy.
Trying to reconcile the two approaches lets take into consideration the subject of physical education. Just how do physical education teachers teach their subject? What kinds of things do they do as they go about their daily routine? What sorts of things do students do? In what kinds of activities do they engage? What explicit and implicit rules of games in Physical Education (PE) classes seem to help or hinder the process of learning? To gain some insight into such concerns, an ethnographic study can be conducted. The emphasis in this type of research is on documenting or portraying the everyday experiences of individuals by observing and interviewing them and relevant others. An elementary classroom, for example, might be observed on as regular a basis as possible, and the students and teacher involved might be interviewed in an attempt to describe, as fully and as richly as possible, what goes on in that classroom.
Descriptions (a better word might be portrayals) might depict the social atmosphere of the classroom; the intellectual and emotional experiences of students; the manner in which the teacher acts toward and reacts to students of different ethnicities, sexes, or abilities; how the “rules” of the class are learned, modified, and enforced; the kinds of questions asked by the teacher and students; and so forth. The data could include detailed prose descriptions by students of classroom activities, audiotapes of teacher-student conferences, videotapes of classroom discussions, examples of teacher lesson plans and student work, sociograms depicting “power” relationships in the classroom, and flow charts illustrating the direction and frequency of certain types of comments (for example, the kinds of questions asked by teacher and students of one another and the responses that different kinds produce).
In addition to ethnographic research, qualitative research includes historical research and several other, less commonly used approaches. Casey (1996), for example, has identified 18 types of “narrative” methods. Among them, four are the most distinctive. These include biography, where the researcher focuses on important experiences in the life of an individual and interacts with the person to clarify meanings and interpretations (e.g., a study of the career of a high school principal). In phenomenology, the researcher focuses on a particular phenomenon (such as school board conflict), collects data through in-depth interviews with participants, and then identifies what is common to their perceptions. A third approach is the case study, in which a single individual, group, or important example is studied extensively and varied data are collected and used to formulate interpretations applicable to the specifi c case (e.g., a particular school board) or to provide useful generalizations. Lastly, grounded theory emphasizes continual interplay between raw data and the researcher’s interpretations that emerge from the data. Its central purpose is to inductively develop a theory from data (e.g., a study of teacher morale in a particular school beginning with interviews and other types of data).
HISTORICAL RESEARCH
You are probably already familiar with historical research. In this type of research, some aspect of the past is studied, either by perusing documents of the period or by interviewing individuals who lived during the time. The researcher then attempts to reconstruct as accurately as possible what happened during that time and to explain why it did. For example, a curriculum coordinator in a large urban school district might want to know what sorts of arguments have been made in the past as to what should be included in the social studies curriculum for grades K–12. She could read what various social studies and other curriculum theorists have written on the topic and then compare their positions. The major problems in historical research are making sure that the documents or individuals really did come from (or live during) the period under study and, once this is established, ascertaining that what the documents or individuals say is true.
ACTION RESEARCH
Action research differs from all the preceding methodologies in two fundamental ways. The first is that generalization to other persons, settings, or situations is of minimal importance. Instead of searching for powerful generalizations, action researchers (often teachers or other education professionals, rather than professional researchers) focus on getting information that will enable them to change conditions in a particular situation in which they are personally involved. Examples would include improving the reading capabilities of students in a specifi c classroom, reducing tensions between ethnic groups in the lunchroom at a particular middle school, or identifying better ways to serve special education students in a specifi ed school district. Accordingly, any of the methodologies discussed earlier may be appropriate.
The second difference involves the attention paid to the active involvement of the subjects in a study (i.e., those on whom data is collected), as well as those likely to be affected by the study’s outcomes. Commonly used terms in action research, therefore, are participants or stakeholders, reflecting an intent to involve them directly in the research process as part of “the research team.” The extent of participation varies from just helping to select instruments and/or collect data to helping to formulate the research purpose and question to actually participating in all aspects of the research investigation from start to finish.
EVALUATION RESEARCH
There are many different kinds of evaluations depending on the object being evaluated and the purpose of the evaluation. Evaluation research is usually described as either formative or summative. Formative evaluations are intended to improve the object being evaluated; they help to form or strengthen it by examining the delivery of the program or technology and the quality of its implementation. In contrast, summative evaluations seek to examine the effects or outcomes of an object by describing what happens after the delivery of the program or technology in order to assess whether the object caused the outcome.
An example of a formative evaluation product is a needs assessment report. A needs assessment determines the appropriate audience for the program, as well as the extent of the need and what might work to meet the need. Summative evaluations can be thought of as either (a) outcome evaluations, which investigate whether the program or technology appeared to have caused demonstrable effects on specifi cally defi ned target outcomes, or (b) impact evaluations, which are broader and attempt to assess the overall effects (intended or unintended) of the program or technology as a whole.
Evaluators ask many different kinds of questions and often use a variety of methods to address them. For example, in summative evaluations, evaluators often use quasi-experimental research designs to assess the hypothesized causal effects of a program. Formative evaluations that examine program implementation may also include analysis of existing data sources, surveys, interviews, observational data, and focus groups.
ALL HAVE VALUE
It must be stressed that each of the research methodologies described so briefl y above has value for us in education. Each constitutes a different way of inquiring into the realities that exist within our classrooms and schools and into the minds and emotions of teachers, counselors, administrators, parents, and students. Each represents a different tool for trying to understand what goes on, and what works, in schools. It is inappropriate to consider any one or two of these approaches as superior to any of the others. The effectiveness of a particular methodology depends in large part on the nature of the research question one wants to ask and the specifi c context within which the particular investigation is to take place. We need to gain insights into what goes on in education from as many perspectives as possible, and hence we need to construe research in broad rather than narrow terms.
As far as we are concerned, research in education should ask a variety of questions, move in a variety of directions, encompass a variety of methodologies, and use a variety of tools. Different research orientations, perspectives, and goals should be not only allowed but encouraged. The intent of this book is to help you learn how and when to use several of these methodologies.
K. Casey (1995, 1996). The new narrative research in education. Review of Research in Education, 21: 211–253.