For some research topics, a temporary question cannot be formed. Therefore, the researcher has to gather data directly by doing research.
For instance, as mentioned above, for the research topic, a hypothesis that can be generated will be – is increase in poverty related to increase in number of crimes?
Stage 4: Making a research design
The fourth stage involves the planning of how the data should be collected. Various types of research methods exist and more importantly, the outcome of the research depends heavily on what method is used. For example, in the book ‘The Managed Heart’ (1983), author Arlie Hochschild carried out methods such as interviews and participant observation in her famous research with flight attendants. These two methods were suitable to the topic studied.
Stage 5: Carrying out the research
If all the four stages are followed, then this stage just involves carrying out the research process and record the information. However, obstacles can be faced by the researcher while collecting the data which may eventually affect the end result. For example, while carrying out interviews and questionnaires, it might be impossible to reach the respondent or while collecting secondary data such as government statistics, the researcher might not be allowed to have access to such confidential information. Unforeseen difficulties are bound to pop up while carrying out the research process, but a successful research is the one that overcomes all the difficulties and provides an unbiased, reliable and valid report.
Stage 6: Interpreting the results
Once the material to be analyzed has been collected, the researcher’s half job has been done. The most critical part of the process is to analyze the data collected and relate it back to the research problem.
Stage 7: Reporting the research findings
The final stage of the research process is to state what conclusions were drawn from the research process and questions such as – what is their significance? How do they relate to previous findings? , could be answered. The research report is usually published as a journal article or a book. In the case of Hochschild, her report was published in the book ‘The Managed Heart’.
Reliability and Validity
The term ‘reliability’ does not refer to the idea of common English usage. It does not mean trusting someone. Instead it refers to how far the data collected through a specific method is true or reliable. For instance, if the data, a researcher gathers claim to be ‘reliable’, it means that the other researchers should get the same results using the same method. For example, an experiment in a chemist lab should always give the same result, regardless of who is doing it, condition that it is carried out under the same conditions.
Some methods in Sociology are considered to be more reliable than other methods. Research methods such as interviews and surveys have a higher degree of reliability than participant observation. It is basically because participant observation involves a single researcher in a situation that cannot be repeated. When using any method of data collection it needs to be examined critically for reliability before use as to what extent it is going to be reliable and valid. Bell (1999) defines reliability as
"The extent to which a test or procedure produces similar results under constant conditions on all occasions." (p. 103).
This means that any method that is used needs to give the similar results each time it was used. Thomas and Nelson define validity (1996) as the:-
"Degree to which a test or instrument measures what it purports to measure; can be categorized as logical, content, criterion, and construct." (p.214)
Official statistics hold a very high degree of reliability in the sense that they are genuine records of what has happened in the past. For example, a researcher may use official statistics, recorded decades ago, in his research which may mislead him/her in this situation. It is because the researcher looking at those data may have asked different questions, which the data he/she used was not really intended to answer. Therefore, it can be said that the research has a low degree of validity. Validity here refers to the degree to which the data collected is a true picture of what is being studied. In another words, it can also be understood as, is the data gathered really the evidence of what it claims to be evidence of. ?
Concept and Indicator
Sociological research is based on objective and empirical data. One of the ways in which sociologists try to be more systematic and objective in collecting data is to be precise as possible by using clearly defined and explained concepts. The term ‘concept’ is an idea of how something is, or should be done. In reality, concept is a definition in short hand of a class or group of facts. For example, if a researcher is doing a research on class in a particular society, his concept can be determined as in terms of wealth or only in terms of heredity. Furthermore, in order to avoid confusion, how would one differentiate that Person A has more wealth than Person B? Well, where comes ‘concepts’ there come ‘indicators’. Therefore, the researcher can use different indicators of wealth such as income, assets (property, cars etc), savings and so forth. In conclusion, the term ‘concepts’ and ‘indicators’ do simplify the sociological research to a certain extent.
Objectivity & Subjectivity
An important school of thought in sociological research is objectivity. It means that the sociologist seeks to understand a phenomenon in terms of its factual nature, rather than reacting to it simply in terms of opinion, bias or preconceived prejudice. May (2001) identifies that if our own values do not interfere with the research it is objective. This objective view predetermines what type of research to carry out, which tends to be quantitive techniques. Experiments will be set up so that particular effects are being looked at from the outset, this predetermination of what to look for is reproduced in other research methods in order to obtain an objective outcome. But can social research be objective? As mentioned, social research is an investigation into social life and we (the researchers) are part of this social life.
Subjectivity, on the other hand, is personal and focuses on the meanings that people give to their environment. Researchers are said to have a better understanding, as they are part of this environment. Data derived from observational research methods are seen as being more personal and idiosyncratic than answers to questions on a schedule. It is argued that such subjective research relies heavily on the ability of the researcher and therefore influences the results of the study.
Comparative Method
‘Comparative sociology is not a branch of sociology. It is sociology itself’ (Durkheim 1895). According to Durkheim, comparative method was the most important method of sociological analysis (Patrick McNeill’s Research Methods 1985). Adding to that, comparative method was largely used in sociological research by many early sociologists when making an attempt to spot the major historical patterns of social change. For example, Emilie Durkheim, in his studies of suicide, took suicide rates of Catholics and Protestants, and asked questions like why are the rates different?
Comparative method works in two alternative ways: either it draws similarities among two different cases which has similar outcomes or it draws differences among two cases which has different outcomes.
Quantitative and Qualitative Data
Qualitative data is recorded which is not in the form of a numerical answer. Capel (1997) defines qualitative research as “any method used to gain insight rather than statistical analysis" (p.194).
This type of research allows for a deeper personal insight to the subject area, and allows for greater flexibility. Qualitative research has the disadvantage of being difficult and time consuming to analyze as there is often large quantities of data, some of which maybe meaningless to the study.
This method gives more specific information than quantitative data. Data can be obtained by research methods such as participant observation, in-dept interviews and research surveys.
On the other hand, quantitative data is data that provides the researcher with numerical values. Capel (1997) states that quantitative research is "any method that produces data which can be reduced to a numerical form and can be analyzed statistically" (p.193). Quantitative data looks for causal correlation. A correlation refers to two things appearing together. A casual correlation refers to the notion that they appear together because one causes another.
Quantitative research data has an advantage as it allows statistics to be produced, which can prove or disprove hypothesis. Quantitative research has drawbacks as it does not have the depth that qualitative research has, nor does it allow for the respondents feeling to be fully investigated. Using the survey method it is possible to collect both data sets along side each other. It can be in the form of social surveys, for example, yes/no answers. In questionnaires or formal interviews using yes/no or maybe answers. One can find quantitative data in statistical reports, statistics by government departments and Internet surveys.
In sociological research, the methods that the researcher carries out to obtain data depend heavily on what kind of sociological problem is being dealt with. Therefore some problems like: what is the crime rate in a society, may give numerical data whereas questions like why crime rate in a society is very high, may generate non-numerical information.
.
Primary and Secondary Data
Primary data is research, which is original, carried out by the researcher at first hand and can be used as a valuable secondary source of data for other researchers. Whereas secondary data is information generated by somebody else but used by a researcher in support of his/her work.
Examples of primary data are – participant observation, surveys, interviews, questionnaires, direct observation and life histories and so on.
Examples of secondary data are – official statistics, government documents, personal documents (e.g. diaries), police, medical and other government records, historical documents; research data collected by other scientists, newspapers can also be a source but are highly subjective and bias.
Advantages and disadvantages
Secondary data provides useful background information, which the researcher can read up on before starting his/her own research. Reliable secondary research can save time and money. Secondary sources are a cheap and readily available source of information. Furthermore, it also helps in longitudinal studies. Longitudinal refers to the analyzing of data over time. Adding to the advantages, a researcher can access all the literature and can cover all parts of the world by reading published books and internet articles.
But the disadvantages are that the information may not be reliable and could be bias. It is impossible to double check their accuracy, problems of validity can crop up as the data collected could be for a different set of reasons. Steve Taylor describes the difficulties of using secondary data in his article ‘Measuring Child Abuse’. He points that the police records carry a certain degree of inaccuracy in the figures because of under-reporting (not all cases are reported to the police) and non-recording.