Share:

Share on Facebook
Share on Twitter
Share on Google Plus
  • Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

Research methods main concepts

Learn more about the main concepts and ideas, from Primary and Secondary Data to Sampling - all the bases are covered here.

Quantitative and qualitative data

The information collected by sociologists is known as data. There are two main types of data. Quantitative data takes a statistical or numerical form; This may be expressed as tables or charts with appropriate statistical analysis. Methods such as questionnaires usually generate quantitative data, often using large samples in order to ensure that the data is representative of the target population. Quantitative data is produced from closed questions where respondents have to opt for one or more answers provided by the researcher.

Methods such as interviews often produce data in a different form – the respondent talking in their own words which is then written down (transcribed) by the researcher. Similarly the findings of observation are sometimes the researcher’s personal notes on the situation they are observing. This form of written data is known as qualitative data. Qualitative data often provides a detailed account of people’s thoughts, interpretations and behaviour, expressed in their own words.

Primary and secondary data

Sociologists may conduct their own original research or they may make use of data that already exists. The information produced from the sociologist’s original research is known as primary data. Methods such as questionnaires, interviews and observation produce primary data.

Often it is simpler and more cost effective for the researcher to use data that already exists – secondary data. This can take a wide variety of forms, some of the most common being official statistics, historical records and personal documents such as diaries.

Types of research: Surveys and longitudinal studies

Sociologists may conduct their own original research or they may make use of data that already exists. The information produced from the sociologist’s original research is known as primary data. Methods such as questionnaires, interviews and observation produce primary data.

Often it is simpler and more cost effective for the researcher to use data that already exists – secondary data. This can take a wide variety of forms, some of the most common being official statistics, historical records and personal documents such as diaries.

Types of research: Ethnography and case studies

Sometimes sociological research needs to provide an overview of people’s views or behaviour. In these situations large-scale quantitative data needs to be collected in the form of a survey. Surveys aim to make broad generalisations based on a large sample. One example of a survey is the government’s Crime Survey for England and Wales which asks around 50,000 people each year about their experiences of crime and deviance.

A problem with conventional surveys is that they only provide a ‘snapshot’ of one moment in time. To overcome that problem longitudinal studies are sometimes used. In this form of research the sample becomes a ‘panel’ who are surveyed regularly over a long period of time, often years. This type of research is very expensive and there is a problem of maintaining contact with members of the panel (losing participants is known as sample attrition). However, longitudinal studies can provide great insight into social change. The British Household Panel Survey, for example, has been following a sample of 5,000 households and 10,000 individuals since 1991, providing a range of insights into social change for a variety of organisations.

Validity and reliability

The terms validity and reliability are helpful when evaluating the quality of research data.

Validity refers to the extent to which data actually measures what it sets out to measure. Have respondents told the truth? Might there have been some interviewer bias leading to the respondent telling the interviewer what they think they want to hear rather than what they really think? Some sociologists see qualitative data as stronger in validity than quantitative as respondents can express themselves in their own words without having to ‘force’ their views into categories set by the researcher.

Data is said to be reliable if another researcher using the same method and a similar approach to sampling would receive the same results. Quantitative data is sometimes seen as strong in reliability as the set questions of a structured interview or questionnaire are easy to replicate. Qualitative approaches such as participant observation rely on a unique relationship between the researcher and participants so cannot be replicated easily.

Crucial to ensuring the validity and reliability of data is the process known as operationalising concepts. This involves defining the key terms of the research in such a way that the research can measure them accurately. Operationalising concepts usually involves identifying indicators that can provide a basis for questions and observation schedules. For example, if research set out to measure students’ ‘educational achievement’ it might use indicators such as test scores, exam results and homework grades.

Sampling

Sociological researchers are not usually able to conduct research with every member of a particular target population. So they take a sample of the group they are interested in and conduct their research using that sample. If the sample is selected carefully it can accurately represent the whole target population and generalisations can be made beyond the sample.

The process of sampling is made easier if a list of names of the target population is available. Known as a sampling frame this list can be used to select the sample. Often sociologists break the sampling frame down into groups such as men and women or people of different ages so that a more representative sample can be selected. Once the sample has been stratified in this way the names of the actual sample can be selected randomly, ensuring that there is no bias in the selection.

If no sampling frame is available researchers might use quota sampling whereby interviewers are allocated certain types of people to interview in accordance with their representation in the target population. Sometimes members of the target population are so hard to locate (for example, deviant groups) that a snowball sample is used. One individual is identified for the sample, they then suggest others and so on until the sample is large enough. Snowball sampling does not create a representative sample but it is sometimes the only option available to the researcher.