General Information About a Research Design

Composing a research design as a strategy of answering a research question is an important thing for the start of any research project. You will use empirical information in your work, so to collect it efficiently, you need to consider the following:

• the aim and approaches to your research work;
• the most appropriate type of research design to choose;
• methods and criteria for picking out samples;
• methods of collecting data;
• procedures for collecting information;
• methods of data analysis.

Methods and aims of your research design should match each other, and it is important to choose the best type of analysis of data. A research design can be a separate assignment or a part of a research proposal or a large research project. Be careful about choosing the most effective methods to answer the research question.

Let’s regard the main parts of a research design and stages of composing it.

Stage 1. Choose the Aims and Approaches

Start with making up a research question. It means defining the subject of your investigation. For example:

✔️ How does remote learning influence students’ ways of doing assignments?

Start with aims and priorities meaning what you want to see in your research. Choose either a qualitative or quantitative approach.

A qualitative approach is about understanding the main concepts, subjective experiences, or beliefs. All of them can be related to the culture or certain context. Look in-depth at the main problems and produce some new ideas. The qualitative approach is a flexible approach since you can change your initial ideas and opinions within the process of research. To answer the question mentioned above, you should consider the main challenges students face while doing their assignments remotely.

A quantitative approach is about measuring variables and describing correlations. It is also about defining averages and frequencies correctly to support the hypothesis about variables and their relationships. You will also need to check the efficiency of the approach, program, or product under consideration. Quantitative research is a deductive approach. The information is well-defined and fixed because you use definite variables and measurements within your data collection process. To answer the question under consideration, you can use this approach for testing the effectiveness of remote learning, which influences students’ grades and further progress.

You can also use a mixed approach, which combines qualitative and quantitative analyses for obtaining a more complete picture and supporting the rightness of conclusions.

Stage 2. Opt For the Appropriate Type of Research Design

It does not matter what kind of approach you will choose for your research. Each of them uses certain types of research design to shape the framework of the research process.

Quantitative Design Types

You can choose among the four types of quantitative designs. They are experimental, quasi-experimental, correlational, and descriptive. You can use experimental and quasi-experimental types to check relationships between the cause and effect. Descriptive and correlational designs are meant for measuring variables and understanding their relationships.

An experimental design type allows for testing causal relationships. It includes operations with independent variables to measure their influence on the dependent variables. The groups and subjects are random but the experiments need a controlled environment, such as a laboratory.	A quasi-experimental design also tests causal relationships but it cannot be conducted randomly. The results obtained from pre-existing groups are important here. You need to compare and contrast them. The experiments should take place in a natural environment.
A correlational design implies testing the relationships between variables. You cannot influence them additionally to make measurements.	A descriptive design needs the measurements free from additional influences as well. You use it to describe trends or specific characteristics of objects.

You cannot make conclusions about cause and effect when you use descriptive and correlational designs but you get a clear picture of relationships between objects and their characteristics.

Experimental and quasi-experimental designs are the best ways to check the cause-and-effect relationships. However, you need to be attentive about the choice between the controlled and natural environment if you want to obtain the most reliable results.

Qualitative Design Types

You do not need strict definitions for the types of a qualitative design. They are used for obtaining a more detailed picture for the specific phenomenon or event. You can be more flexible here. The main types of a qualitative design include a case study, ethnography, a grounded theory, and phenomenology.

A case study provides an analysis of some concrete places, organizations, or events. You can collect data from different sources and obtain the overall understanding of this very case.	Ethnography implies the detailed research of ways and traditions of a specific group of people. You should get immersed extendedly into the community and closely observe the details. Then, you describe and interpret their social dynamics, views, or customs.
When you analyze qualitative data systematically to develop a certain theory using induction, it means that you use a grounded theory design.	You describe the experiences of some person or group to understand a phenomenon or event. It means that you use phenomenology as your research design.

Stage 3. Choose the Population and Methods of Sampling

Define what your research deals with and what the main focus of it is. Decide on how you are going to pick out subjects or participants. Think about the population of research meaning the choice of the whole group and about the sample which is a smaller group or separate individuals you want to collect your data from.

How to Determine the Population

The population can include anything you want to research - organizations, countries, plants, texts, animals, etc. it can be a group of people from a specific background, area, culture, or demographic. They can be consumers of a product, or people with some health condition.

Anyway, you need to define the population precisely to obtain the clearest representative samples.

Methods of Sampling

Even if you have defined your population precisely, it does not mean that you will be able to collect the needed data from every member of it. You need a narrower sample. To select it, you should opt for either probability or nonprobability samples.

Probability samples are selected randomly. They can be used in quantitative research to make strong statistical inferences.

Nonprobability samples are chosen in a controlled way, and they can be used either in qualitative or quantitative research. However, there is more risk of facing bias here.

If you choose probability sampling, your statistics will be more valid. However, these statistics are difficult to obtain if your population is small or poorly accessible. Most types of research use non-probability sampling, but you need to consider all the limitations and bias possibilities. That is why try to make your sample as representative as possible.

Using Case Selection in Qualitative Research

Sometimes, sampling cannot be relevant to the purpose of your research. When you use ethnography, you need to understand the context but not the overall characteristics of the population. So, you need to gather as much data as possible.

Think about the right choice of community or case. Make sure that this very case is suitable for your purposes of answering the research question. For instance, you may opt for several cases to compare them in the context of your research.

Stage 4. Opt For an Appropriate Method of Collecting Data

You need to gather the necessary information and/or measure the variables. They will allow you to get deeper insights into the problem. That is why, think about the most appropriate data collection method. Or you may opt for some methods for the same study.

Surveys

They allow you gather necessary data about experiences, opinions, features, or behaviors. You ask people immediately what they think about the problem. There are two ways of doing it - questionnaires and interviews.

Questionnaires imply quantitative research. They can be provided online, by phone, email, or directly to people. You usually use closed questions, which need the choice among certain options. The questionnaires allow you collect data from many different people.

Surveys are commonly used in qualitative research. A researcher conducts a survey in person, online, or by phone. Participants answer open questions in their own words. You can use surveys for smaller groups to obtain more in-depth results.

Observation

Observations are used to get data by observing social interactions or features without involving the participants in direct contacts. Observations are made in real time. You can take notes or record audios and videos for further analysis. They can also be qualitative and quantitative.

Quantitative observations include permanent measuring and counting in accordance with the criteria set up in advance.

Qualitative observations need detailed notes and descriptions. They should always be recorded.

Other Methods

Some other methods of data collection can be chosen in relation to the field and topic of research. They involve using media and communication channels, psychology, education, or physical sciences. You can collect samples from speeches, blog posts, or articles to make sure about certain narratives of cultural beliefs.

When you use psychology for your research, you may need some knowledge of eye-tracking, neuroimaging, or computer-based tasks to research reactions, attention span, or emotions. When you need to gather data on skills or qualifications, you can use educational assignments or tests. When the task is to collect data about chemical structure, blood pressure, or body parameters, you can use physical sciences.

You may doubt what type of methods to use in your research. So, you may need to read some literature on data collecting methods, which are specific for your field of study.

Using Secondary Data

It can happen that you are limited on time and resources to gather data from samples and populations. So, you can use secondary data from other researchers in this field, for example, results of official surveys or previous studies. This data is raw but you can always make your analysis of it to answer your specific research question.

Often, using secondary data is quite helpful because you can access more varied sources and samples than when you collect the data yourself. However, you should be careful about the choice of variables to measure and how to compare them. The conclusions made on the basis of such data may not always be relevant.

Stage 5. Think About Relevant Procedures for Collecting Data

Planning is helpful in data collection alongside the methods and approaches. Remember that you need unbiased, consistent, and exact data for your best research results. Planning works well for quantitative research where you need to be accurate about defining variables and getting reliable measurements.

Effects of Operationalization

While it is quite easy to measure such variables as age or weight, it is challenging to deal with abstract concepts. For example, you need to measure competence or levels of satisfaction, so you cannot use physical parameters here.

That is why you need operationalization to substitute the ideas by indicators which can be measured.

Importance of Validity and Reliability

Validity implies measuring the exact concept you are interested in while reliability refers to the consistent results which can be easily reproduced. When you think about the reliability, consider whether you have researched the same object within a certain period of time and whether you have obtained the same results for various contexts, or even if all the questions you asked relate to the same concept.

When you want to check how valid your research is, test whether your measurements involve all the concept aspects and how well they coincide or diverge from other measurements for the same concept.

You need to plan all the research procedures carefully to be sure that all the measurements are done properly and during all the procedures, you have followed the same steps for every participant.

Run a pilot study or trial for every questionnaire or other measurement tool you have developed on your own. You can test the reliability and validity of your research procedure before you start it.

Procedure for Sampling

Think about how you are going to involve the sample you have selected in the research procedure. Consider how many participants you will need in one sample, what criteria you would like to include or exclude for choosing the most reliable participants, and the ways of communication with your sample (by mail, phone, or in person).

If you select your sample randomly for the probability method, check whether all the selected members have taken part in the study. That is why you need to opt for the way to ensure the highest participation rate. With a non-probability approach, think how to avoid bias.

Managing Data

Develop a special plan for managing data to organize and store it properly. Think about ways of protecting any confidential data you may obtain. Make sure that your data is consistently backed up. You may also need to think about the ways of transcribing interviews or making observation entries.

Your data should be properly organized to avoid its loss. You will have more time for making its analysis, validating it, and developing findings.

Stage 6. Choose the Proper Strategies for Data Analysis

Even if you have collected as much data as possible, it is raw, so it cannot answer your research question. You need to develop a concrete plan of how you are going to analyze your data. That is the last stage in your research design.

Quantitative Analysis

If your research is mostly quantitative, you can use statistical analysis to sum up your sample data, test hypotheses, or make estimates.

Descriptive statistics will help you to:

• obtain data distribution (for example, accordance to the frequency);
• define the main tendency (or average score);
• observe the data variability (or standard deviation).

You can also make certain calculations depending on your variables’ measurements.

Inferential statistics can help you to estimate the population on the basis of the date received from the sample and test the hypotheses about the interaction between your variables.

You can also use correlation and regression tests to find associations between several variables. Or opt for comparison tests (for instance, ANOVAs or t-tests) to find the difference between the groups. Your choice can depend on the type of variables or data distribution.

Qualitative Analysis

If you use qualitative research, your data will be rich in information and ideas. You cannot simply sum up the numbers here, but you have to focus on details, interpret meanings, and define specific patterns, or extract the most relevant parts that correspond to your research purpose and question.

You can use either thematic or discourse analysis here.

Thematic analysis implies the focus on the content that allows to code and arrange the data for certain key topics (themes).

Discourse analysis allows you focus on data functioning in the context involving different communication levels (structure, language, tone, or others).

You can analyze qualitative data in different ways. It depends on the purpose of your research.

Final Thoughts

Now when you know the structure of a research design, you can make your research project more consistent and well-organized from the beginning. Remember that the structure and main steps of the design should correspond to the research questions and the overall aim of your work.