INITIATING SOCIAL RESEARCH
INTRODUCTION
Research process proceeds through a number of steps, beginning with the identification of the research topic, and moving through the process, until data collection, analysis and processing are completed. Here we shall address the first step of the research design, focusing on research topic and methodology.
1 SELECTION OF THE RESEARCH QUESTION
Researcher has to answer very clearly the questions: 'WHAT will be studied in this project?' and 'HOW will this be studied?' Although the latter is a very complex question, it is important to know at least whether the study will be qualitative or quantitative.
There are three major issues associated with the selection of the research topic. First, what can be studied in a research project? Second, who selects the research topic? Third, what factors influence the decision to study a certain research question?
1.1 The nature of the research question: 'What can be studied?'
In practice, social scientists can investigate virtually any social issue. The research topic can be related to people, groups, ideas, ideologies, attitudes and opinions, structures and processes, methods and practices, and the causes and effects of social events. Units of study can be taken from any level, that is, from the individual-to-individual level, individual-to-group level, or group-to-group level. A research project can operate at one or more levels.
The only restrictions of this freedom relate to issues of relevance, researchability, feasibility and ethics. Relevance refers to whether the study of the research topic is relevant to the purpose of the study. Researchability refers to whether the research topic is approachable methodologically. Topics, such as whether there is a God or whether there is life after death, are not suitable and cannot be studied empirically. Feasibility relates to whether the research is possible, that is, whether the researcher has access to the research subject as well as the means and resources that are required to complete the study Ethics, finally, refers to whether the proposed study is ethically justifiable and follows ethical standards and principles.
Second and third factor are directly relevant to the question: 'What can be studied?' There are two levels of rules: one relates to what researchers can study, and the other refers to what researchers should study.
1.2 Who chooses the research question?
The researcher: Researchers generally study issues that lie within the area of their personal expertise and interests. These interests are related to external factors, such as income, prestige or promotion, advancement of knowledge or improvement of social conditions (extrinsic motivation); or to internal factors, such as the research issue per se (intrinsic motivation).
Social conditions: In some cases research topic is determined by social conditions. Here, research has a social function and must be carried out irrespective of the personal preferences and interests of the investigator. It is argued here that the researcher is a part of the community, has a public responsibility and duty to serve, and that research is guided by the heeds of that community.
The sponsor: Complex and demanding research topics are usually funded by sponsors who not only provide the resources, but also determine the area of study, and often the research topic. In certain cases the researcher is employed to investigate a specific topic, and that remit cannot be changed.
1.3 Selection in a social context: important factors
| Factors affecting the choice of the research topic · Financial constraints: Topics funded by sponsors are more likely to be studied than those that receive no support. · Time: Studies that take up too much time are less likely to be chosen by researchers than others that are equally important but require less research time. · Availability of assistants and experts: Lack of research assistants may force researchers to opt for topics that can be studied without their help. · Research paradigm: Topics studied within popular paradigms (e.g. feminist research) may be preferred to topics investigated within other paradigms. · Expertise: Researchers normally study topics that are within their professional interest and expertise. · Ideology: Researchers study issues that are consistent with their ideological affiliation. Feminists study women, and Marxists study the status of workers in capitalist societies. · Access to the research subject: Issues that are difficult to access are less popular research topics than those that are easily accessible. · The need for data: The need for information on certain subjects attracts the interest of researchers not only through their own volition but also because these issues attract funds, and hence assistants and other resources. |
In many cases, the decision of a research topic is influenced by several factors, although one may be more influential than others.
1.4 Basic questions
After the research topic has been chosen, the investigator usually explores methodological issues. The extent of such an exploration depends on factors related to the nature of the topic; however, some questions are thought to be particularly important in this context.
· What is the research unit? Before embarking on the next stage, the researcher should make it clear that the research will deal with a certain unit, in a certain context, with certain characteristics. It is important to stress that this is the unit the investigator will refer to when conclusions, statements or generalizations are made.
· What is the level of research? Research can be conducted at several levels, for example:
1. first-level research, that is, the relation between individuals
2. second-level research, that is, the relation between individuals and groups
3. third-level research, that is, the relations between groups.
An investigation may be carried out at more than one level. It is the task of the researcher to define these levels adequately during the first step of the research design.
· Will objectivity be observed? The attitude of the researcher to objectivity should be made clear at the outset. If measures to ensure objectivity have been taken, these should be stated. If objectivity is considered irrelevant, this should be made known. Researchers must state clearly whether they take the role of a 'detached observer', an 'empathetic observer', a 'faithful reporter', a 'mediator of languages', a 'reflective partner' or a 'dialogic facilitator'. Each role entails a different degree of objectivity. Hence, the degree of commitment to objectivity should be defined clearly at this stage of the project.
· Can observance of ethical standards be guaranteed? Ethical issues should be considered at all stages of research.
When all questions have been answered satisfactorily, the researcher will proceed to the next step.
| Ethical considerations when choosing a research topic · Can this topic be addressed without violating ethical standards? · Can confidentiality and anonymity be guaranteed? · Can it be ensured that the study of this topic will not endanger the respondents and their families? · Can this study hurt the researcher? · Can it be ensured that all members of the research team will adhere to ethical standards? · Will the study of this topic necessitate any type of unfair dealing, illegal or unethical activity? · Can the research arouse outside interest (e.g. from the authorities) that could harm the participants? · Is it likely that ethics approval from the proper authorities will be obtained? (Benini, 2000: 27-8) |
2 SELECTION OF THE RESEARCH METHODOLOGY
2.1 Introduction
The manner in which research will proceed depends on many factors, but the underlying methodology is the most important. In some cases just stating that the investigation will be based on a quantitative or qualitative framework might suffice (be enough). In other cases more information may be required. For instance the case when researchers need to state the paradigm that guides their theory and ideology.
The choice of methodology is usually made after the research object is selected, but in many cases it is chosen earlier. It is common for researchers to work mostly within a specific paradigm, for example they may be Marxists, feminists or structuralists, and hence their methodological affiliation is a constant and not a variable.
2.2 Criteria of choice
Social research can be conducted within a quantitative or qualitative context, and that both types of research are equally legitimate. Hence, when we are talking about the choice of methodology, the question is not about its quality but about its suitability. Normally, although ideology plays an important role, the choice of a suitable methodology is directed by theoretical principles.
| Factors affecting the choice of a methodology Factors that may affect choice include: · the underlying theoretical paradigm · the appropriateness of the method for the theoretical goals · the adequacy of the method for the research object · the overall purpose of the project · the nature of the expected outcomes · the realization of methodological rules which determine its structure, possibilities and limitations · examination of the prerequisites and conditions that must be considered when performing mathematical-statistical tests |
Quantitative methodology is employed when the project is interested in:
· observable phenomena
· quantification and measurement
· objectivity
· large samples
· validity and reliability
· description, relationships and causality
· statistics as a tool of data analysis
· representativeness and generalizability
· replication, precision and accuracy
Factors in favour of a qualitative methodology
A qualitative methodology is chosen when:
· The standard of knowledge in the area of the research subject is inadequate and provides no sound basis for a quantitative study. The qualitative research in this case takes the form of an exploratory study.
· There is a need to study reality from the inside, that is, to understand it from the point of view of the subject.
· The study object is so complex that a quantitative method is of little use.
· There is a need to capture reality 'as it is', that is, in interaction.
· The researcher intends to present the information gathered, not as numbers or formulae but verbally, in a detailed and complete form.
· The researcher wishes to approach reality without preconceived ideas or pre-constructed designs and patterns.
· The investigator perceives researcher and researched as elements of the same situation and the research design as a whole unit.
· The researcher wishes to capture the meaning and regularities of social action.
3 METHODOLOGICAL CONSTRUCTION OF THE RESEARCH TOPIC
Following the choice of methodology the researcher will focus on the research question and work to refine and adjust it so that it can be approached methodologically. In quantitative research, during this step, additional information is gathered to further clarify the object of study; to refine the research question, making it more specific and preparing it for investigation; to develop a framework for the research project; and to link the research question with research methods and procedures.
More specifically, when constructing the research topic the following tasks are to be considered:
· definition of the topic and associated variables
· exploration
· operationalization
· formulation of hypotheses
Quantitative researchers usually consider all four tasks in their research. For some, these tasks are developed during the research. For others, the design should be flexible and not commit them in advance.
4 DEFINITION OF THE TOPIC
The first point to be considered at this stage is the definition of the topic and of the related variables. Quantitative researchers usually define the topic accurately and specifically; qualitative researchers prefer to define it loosely and in general terms.
For example, if the research question is about 'Family and delinquency', quantitative researchers are expected to explain what they mean by both 'family' and 'delinquency', and define these concepts accurately. In doing so, the researcher will not necessarily develop a new definition for each concept, since there are many well-accepted definitions available for both concepts. However, since there are many types of families and of delinquency, researchers will specify the kind of issues they intend to investigate.
For instance, whether they will study: (a) one-parent families, de facto families, homosexual families, reconstituted families, extended families, ethnic families, city families or country families; (b) families of married couples, unmarried couples or both; (c) families with parents of all ages or of specific age groups; (d) families with children or without children; (e) families with children of a specific age, and so on.
When defining the topic, researchers will reduce it significantly. They may, for instance, reduce the concept 'family' into 'families in which at least one partner is over 25 years of age', 'families with children under five years of age', or 'families with teenage children'. Quite frequently, researchers begin their definition with the statement that 'for the purpose of this study family is defined as... and then describe the type of unit they intend to deal with in their study. In a similar fashion, the researcher will define 'delinquency'.
In qualitative investigations, definitions are loosely structured; it is expected that additional information will be collected during the study and will help to refine concepts and define them more clearly in a more concrete and specific manner. It must be kept in mind that qualitative researchers are interested in people's interpretations of objects and events, and this includes their definitions.
5 EXPLORATION
5.1 Introduction
In some cases exploratory work is elementary, and serves as a guide for the formulation of hypotheses and/or for the operationalization of the concepts. In other cases, exploration is a major part of the research study. The latter is, for instance, the case in certain forms of qualitative research.
Exploratory studies are most frequently carried out when there is insufficient information about the research topic, so that the formulation of hypotheses and the operationalisation of the question are difficult or even impossible.
| What is exploration for? · Feasibility. Exploration will show whether a study of the issue in question is warranted, worthwhile and feasible. · Familiarization. Exploration will familiarize the researcher with the social context of the research topic, with details about relationships, values, standards and factors related to it, and with methods. · New ideas. An exploratory study may help to generate ideas, views and opinions about the research object, which are useful when constructing the research design. · Formulation of hypotheses. Exploration will show whether variables can be related to each other, and if so in what way, direction and degree. · Operationalization. Exploration can help to operationalize concepts, by explaining their structure and by identifying indicators. |
5.2 Types of exploratory studies
Exploration is usually found to take one or more of the following forms:
· review of available literature
· expert surveys
· analysis of case studies
Most projects employ more than one type of exploratory study, with literature research being predominant. This also constitutes the first step of exploration, when more types of exploration are employed.
Literature review: This involves a secondary analysis of available information already published in some form with the aim of collecting information about its structure, process and relationships, thus increasing the familiarity of the researcher with the research object. In addition, it can consider previous research for a historical or comparative analysis of the issue in question so that the current study can be placed in a historical context. Finally, it may review a theory or the methods and techniques most suitable for the study.
Expert surveys: Expert surveys involve interviews with experts who have substantial knowledge and experience in the research area, although their views might not have been published. This unpublished information is often very relevant to the research object, and can only be obtained through such interviews.
Case studies: When information collected through literature review and expert surveys is insufficient, researchers collect first-hand information through case-study research. Single cases relevant to the issue are selected and studied, in order to collect information for the main study.
6 OPERATIONALISATION
Often, even the best definitions do not prepare concepts clearly enough to make further research possible. Using definitions in research can be misleading and cause confusion, bias and distortions. Concepts such as social class are difficult to use directly in research because they incorporate many elements, and can therefore be understood and interpreted differently by the respondents. Asking people straight questions about which social class they belong to is found to produce biased and inaccurate answers, with a large proportion of respondents ranking themselves as middle class.
| What is operationalization? Operationalization is the process of converting concepts into their empirical referents, or of quantifying concepts for the purpose of measuring their values, such as occurrence, strength and frequency. It is employed when concepts are vague, unclear or abstract, and thus involves a process of translating abstract concepts into synonymous empirical referents; in this way it facilitates a precise measurement. |
Central procedures
Operationalization contains four major elements: identifying the dimensions of the concept, selecting indicators, identifying empirical references, and quantifying the variable.
| Major elements of operationalization · Identification of dimensions that reflect the nature and complexity of the concept. Dimensions refer to aspects of the concept. · Selection of indicators that reflect the presence/absence and strength of the dimensions of the concept. · Identification of empirical referents, that is, the range of values the indicators can assume, and assignment of scores that represent the degree of presence or absence of the concept or variable. · Quantification of the variable, that is, identification of the continuum of values the variables can assume, and assignment of scores as above - here, for the main variable. |
Process of Operationalisation
| Step 1 | Theoretical CONCEPT |
| Step 2 | Identification of dimensions |
| Step 3 | Selection of indicators |
| Step 4 | Identification of empirical referents |
| Step 5 | Quantification of the concept |
| Step 6 | Empirical CONCEPT |
Examples of Operationalisation
| Operationalising social class | |||
| Concept | Dimensions | Indicators | Empirical referents |
| Social class | 1. Economic status 2. Occupational status 3. Educational status | 1. Income 2. Occupation 3. Education | 1. Amount of money 2. Type of job 3. Years of formal study |
The researcher who investigates the distribution of people within the class system of their community will not ask the respondents directly about which class they think they belong to, but rather about the amount of money they earn, the type of job they have and tile years of study they completed. Indicators and referents serve here as a bridge between theoretical and empirical concepts. This not only makes it easier for the respondents to answer; it also facilitates more accurate and more valid responses.
Basic questions in operationalisation
The basic questions that need to be addressed during the process of operationalization are:
· What is the concept to be studied?
· What are the dimensions that need to be addressed?
· What are the most appropriate indicators that describe each dimension fully?
· Does opcrationalization measure what it is supposed to measure? '
· Are the instruments of measurement reliable?
7 TRIANGULATION
Triangulation refers to the practice of employing several research tools within the same research design. This procedure allows the researcher to view a particular point in research from more than one perspective, and hence to enrich knowledge and/or test validity. Triangulation can be applied in all aspects of the research process. The following are the most commonly used types of triangulation:
Method triangulation: This combines several methods in the same study. It employs a mixed-method design to investigate different aspects of the same phenomenon. It can employ methods of different methodological affiliation (inter-method triangulation), or of the same methodological affiliation.
Time triangulation: This method entails the use of research at different time for example, surveying students during the first and last week of their first season.
Paradigm triangulation: Here a number of different paradigms (e.g. positivist and interpretive) are employed to study the same phenomenon. A qualitative study, for instance, may be employed in a manner that produces quantifiable data, and after the data collection the data is quantified, analyzed and interpreted within a quantitative perspective.
Investigator triangulation: In this form, triangulation combines the expertise of more than one investigator in the same study. This is a useful method in qualitative research where flexibility and openness are accepted. In quantitative research, investigator triangulation can be applied during the construction of the research design, especially during the interpretation of findings.
Sampling triangulation: Here two or more samples are employed within the same project. Such triangulation is found in experiments.
Purpose of triangulation
Triangulation is employed for a number of reasons. It allows the researcher
· to be thorough in addressing all possible aspects of the topic
· to increase the amount of research data, and hence increase knowledge
· to enrich the nature of research data
· to facilitate a study, where one procedure serves as a stepping-stone for the other
· to allow comparison (e.g. longitudinal study)
· to achieve a higher degree of validity, credibility and research utility
· to overcome the deficiencies of single-method studies
8 FORMULATION OF HYPOTHESES
A hypothesis is an assumption about the status of events or about relations between variables. It is a tentative explanation of the research problem, a possible outcome of the research, or an educated guess about that outcome. For some methodologists any logically justifiable assumption can be a hypothesis; for others such an assumption should be scientifically justifiable, and created on the basis of sufficient theoretical or empirical evidence. Hypotheses are generally answers to research questions and can be generated in many ways. They can be developed through existing theories, through research findings of other studies, evidence, commonly held beliefs or intuition. Its purpose is to offer a clear framework and guide when collecting, analyzing and interpreting the data. More specifically, hypotheses are expected:
· To guide the social research, by offering directions to its structure and operation
· To offer a provisional answer to the research question
· To facilitate statistical analysis of variables in the context of hypothesis testing
In many cases hypotheses serve as a testing tool for the relationships between variables. In this sense, a hypothesis contains a possible solution to the research problems validity will be tested by the evidence gathered by the study. At the time of construction, hypotheses cannot be described as true or false; they can only be relevant or irrelevant to the research topic. If we were to study, for instance, the effects of education on religiosity, a relevant hypothesis could be ‘a high level of education is associated with low religiosity'; equally possible could be: education is conversely related to religiosity'; 'education is positively correlated with religiosity'; and 'there is no relationship between education and religiosity. Every stance taken within the hypothesis is correct, as long as it is relevant to the purpose of testing.
8.2 Types of hypotheses
There are many forms and types of hypotheses, mainly depending on their structure, goals and nature. A few examples of types of hypotheses are described briefly below.
Working hypotheses: A working hypothesis is a preliminary assumption about the research topic, most commonly made when there is not sufficient information available to establish a hypothesis, and as a step towards formulating the final research hypothesis.
Statistical hypotheses: A statistical hypothesis is a statement or set of statements developed by means of statistical principles related to the probable distribution of certain criteria of the population. Statistical hypotheses are used as part of the process of verification, lend themselves to statistical testing, and are expressed in the context of a null hypothesis and an alternative hypothesis. Statistical testing should determine whether a statistical hypothesis is accepted or rejected.
Null hypothesis (H0): This is one of a set of two hypotheses (the other is the alternative hypothesis) formulated by the researcher to be used in the context of hypothesis testing.
Alternative hypothesis: This is the other of the set of two hypotheses referred to above (see ''null hypothesis"), and states the opposite of the null hypothesis. In statistical tests of null hypotheses, acceptance of Ho means rejection of the alternative hypothesis; and rejection of Ho means acceptance of the alternative hypothesis.
8.3 Criteria of hypothesis construction
Hypotheses can be presented in any form except that of a question. However, they have to meet a number of standards, which are listed below. In general, hypotheses are required to:
· describe variables or establish a relationship between variables
· be empirically testable (capable of being proven right or wrong)
· focus on one issue only
· describe variables or relationships between variables
· be clear, specific and precise
Hypotheses can be formulated in a descriptive or relational form; in the former they describe events, in the latter they establish relations between variables. They may also be directional or non-directional, depending on whether or not they make a concrete suggestion about the research question.
Types of hypotheses
| Types of hypothesis | Examples of hypotheses |
| Descriptive | ‘Many single parents live in poverty' |
| Relational | ‘The gender of single parents affects their quality of life' |
| Non-directional | 'Marriage is associated with happiness' |
| Directional | 'Married people are happier than unmarried people' |
MAIN POINTS
· Research initiation entails the selection of methodology, selection and definition of the research topic, the decision to conduct an exploratory study, operationalization and formulation of hypotheses.
· The step of research initiation is undertaken by all researchers, although qualitative researchers are less strict about detailed procedures than quantitative researchers.
· The research topic is usually chosen by the researcher but can also be determined by social circumstances or the sponsor.
· The choice of methodology is guided by the underlying theoretical paradigm, and the nature of the research topic.
· The goals of exploratory studies are to explore the feasibility of a study, to familiarize the researcher with the research topic and the respondents, to bring new ideas to the research, and to facilitate operationalization and the formulation of hypotheses.
· Operationalization is the process of quantifying variables for the purpose of measuring their strength and frequency. It entails selection and quantification of indicators, and quantification of the variables.
· The rules of operationalization are the rule of empirical relevance, the rule of correspondence, the rule of empirical adequacy and the rule of quantification,
· Triangulation is the procedure in which data collection is accomplished through more than one avenue.
· A hypothesis is an assumption about the possible outcomes of the study. Hypotheses are expected to be clear, specific, precise and empirically testable, to describe one issue at a time, and not to contain statements that are contradictory.
· Qualitative researchers do not employ operationalization, and do not construct hypotheses prior to the commencement of the research.
No comments:
Post a Comment