First, you might want to clarify the hypothesis, look at the variables in question and select a sample of data that you can look at. Then, you might decide on the type of data you'll need: quantitative or qualitative, and consider what would make your research accurate and reliable.
Making all these choices and contextualising them, you just created your research design. Depending on your decisions, it is either an experimental, a cross-sectional or a descriptive one.
- We will look at the definition of research design.
- Then we will consider research design examples in sociology.
- We will consider qualitative and quantitative research design.
- We will describe how to measure the quality of your research.
- Finally, we will mention three types of research design; experimental, cross-sectional and descriptive.
Definition of 'research design' in sociology
There can be no sociological research started without first creating a research design. But what exactly is research design about?
Research design refers to the overall strategy a researcher undertakes in planning and executing the data collection, analysis and interpretation of findings.
Methods and procedures of research vary from field to field, but there are a few identifiable features of general scientific research that distinguish it from other types of information collection. Normally, scientists - social scientists as well - make hypotheses and design a research process in order to test those hypotheses and make predictions.
Research design examples in sociology
Researchers consider the following points in their research design:
Let us consider some of these points in more detail.
Formulation of hypotheses in sociology
Hypotheses are statements predicting a certain outcome of an experiment or assuming there is a relationship between two phenomena.
Boys perform better in STEM subjects than girls.
Regular exercise increases life expectancy.
The job of the researcher is to test such a relationship and find out whether their hypothesis is true or false.
Variables to be measured in sociological research
A variable is a data item; a number, characteristic or quantity that can be measured. It often changes over time.
Examples of variables can be age, sex, income, class grades but also country of birth, eye colour and type of vehicle.
Fig. 1 - Eye colour is a variable.We distinguish between independent and dependent variables. Independent variables are the ones that are changed by the scholar. The dependent variable is the subject of the observation. The researcher is interested in knowing whether and how it is dependent on the value of the independent variable.
Sampling in sociological research
Would you believe a claim that on average, younger people spend three times more time on their phones than older people if the researcher asked only 10 participants?
Would you believe that there is no institutional racism in the UK if the only people who were asked were white?
Would you believe that the gender pay gap is not a thing if the only people asked were men?
These are all examples of why sampling is an important aspect that must be considered when designing a study.
We differentiate between representative sampling and non-representative sampling. Depending on whether your study is aiming to make wider statistical claims or present an in-depth story of a selected research subject, some sampling types will be appropriate and some not.
Let us move to the methods of data collection, which is also a crucial point to consider in the research design. Generally, we differentiate between qualitative and quantitative research design. We will briefly go through both types.
Fig. 2 - The nature of the sample is important when considering generalisability
Quantitative research design in sociology
Testing of a hypothesis is typically done through a systematic process using objective measurements, which involves quantitative methods.
Main forms of quantitative data collection:
Experiments
Surveys
Official statistics
Qualitative research design in sociology
In the research design, scholars decide whether they are going to implement quantitative or qualitative research methods.
Qualitative methods consist of detailed narratives about research subjects, and therefore are not subject to objective comparisons or replicability.
The equivalent of a hypothesis used in qualitative research is a 'problem statement' concerning the research subject. That said, qualitative research can lead to a formulation of a quantitatively testable hypothesis.
Main forms of qualitative data collection:
Interviews
Observational studies
Field diaries
Measuring the quality of research
When choosing a research design, the research must consider the quality of prospective research findings. This will be measured through considering 5 main points: validity, reliability, credibility, transparency and generalisability of the research findings.
Let us go through them one by one.
Measuring validity of a research design
Data can be described as valid when it truly reflects the phenomenon a researcher is trying to measure.
For example, measuring body weight of a person using scales.
Measuring reliability of a research design
Data can be described as reliable when another researcher gets the same outcome using the same methods.
For example, measuring the body weight of the same person using scales over time and getting a result of +/- 1 kg.
Measuring credibility of a research design
Data can be described as credible if we can trust the source of it and the findings.
For example, the Office of National Statistics in the UK is a credible source of socio-economic data.
Assessing the transparency of a research design
Research can be described as transparent when all the products of research are available for scrutiny.
Measuring generalisability of the research
Research is generalisable if its results are true or relevant for a wider population or context.
Generalisability is particularly important when you cannot collect data from the whole population and are made to choose a sample.
If you wanted to know how many people in the UK are happy, you would need to ask more than 66 million people. That is impractical if not impossible to achieve. In these situations, researchers choose a sample of people that is representative of the population, namely, that accurately reflects the characteristics of the wider group. If the sample is representative, then researchers can generalise or transfer their conclusions to the whole population because they can make a reasonable assumption that if it is true for a representative sample, then it must be true for the population.
Now that we have gone through the factors determining research design, let us briefly look through the most common types of research design.
Types of research design in sociology
There are many ways to classify research designs. Sociologists usually differentiate between three main types of research designs: experimental, cross-sectional and descriptive.
Experimental research design
Experimental research design is built on measuring the effects of an intervention. The research process following the experimental research design consists of measuring the characteristics of something before and after the intervention and trying to assess the differences between the two.
In experimental research, the variables can be controlled by the researchers.
Method connected to experimental research design: Trial.
A good example of a question for an experimental research design would be 'Do free breakfast at school improve the academic achievement of children?'. One can approach this research question by examining academic success of children when they do not have access to free breakfast and also when they do. Then, the results of the two situations can be compared.
Fig. 3 - An example of a question for an experimental research design would be 'Do free breakfast at school improve the academic achievement of children?'. Cross-sectional research design
A cross-sectional research design is used to gather information from large, representative samples of the population. It suggests the assessment of the prevalence of the average and its comparison to various groups of the population. Then, conclusions can be made about each social group in comparison to the whole.
Method connected to cross-sectional research design: Social survey.
An example of a research question that can be discussed through cross-sectional research design would be 'Are there any gender differences in income in the UK?'. In this research, one can compare the income of men and women to the average income of UK citizens.
Descriptive research design
Descriptive research design is good for research that aims to identify trends, categories and characteristics of a larger phenomenon. It is interested in the how and the what, rather than the why. Therefore, it is often used as the basis for further research. One needs to understand the circumstances and have accurate data about the research problem before they can move on to consider why it is a problem in the first place.
Descriptive research design uses both quantitative and qualitative research methods.
In descriptive research, the variables cannot be controlled by the researchers.
Method connected to descriptive research design: Observation.
Research Design - Key takeaways
- Research design refers to the overall strategy a researcher undertakes in planning and executing the data collection, analysis and interpretation of findings.
- Researchers consider the following points in their research design: the type of the study, the research question, the hypotheses, variables (independent or dependent), sampling, the methods of data collection (quantitative or qualitative).
- We differentiate between quantitative and qualitative research design.
- When choosing a research design, the research must consider the quality of prospective research findings. This will be measured through considering 5 main points: validity, reliability, credibility, transparency and generalisability of the research findings.
- There are many ways to classify research designs. Sociologists differentiate between three main types of research designs: experimental, cross-sectional and descriptive.
How we ensure our content is accurate and trustworthy?
At StudySmarter, we have created a learning platform that serves millions of students. Meet
the people who work hard to deliver fact based content as well as making sure it is verified.
Content Creation Process:
Lily Hulatt is a Digital Content Specialist with over three years of experience in content strategy and curriculum design. She gained her PhD in English Literature from Durham University in 2022, taught in Durham University’s English Studies Department, and has contributed to a number of publications. Lily specialises in English Literature, English Language, History, and Philosophy.
Get to know Lily
Content Quality Monitored by:
Gabriel Freitas is an AI Engineer with a solid experience in software development, machine learning algorithms, and generative AI, including large language models’ (LLMs) applications. Graduated in Electrical Engineering at the University of São Paulo, he is currently pursuing an MSc in Computer Engineering at the University of Campinas, specializing in machine learning topics. Gabriel has a strong background in software engineering and has worked on projects involving computer vision, embedded AI, and LLM applications.
Get to know Gabriel