For a survey researcher, one of the highest-paying social science jobs, designing research is more than a means to an end. It’s a core job duty. Without well-designed research surveys, social scientists, policy analysts, business professionals and so many other occupations wouldn’t be able to acquire the data they need to make breakthroughs in their fields of inquiry or develop actionable plans to achieve their goals. Surveys are often a form of non-experimental, or observational, research but may also be used as experimental research methods in their own right.
Types of Survey Questions
Developing the actual wording of survey questions is a big part of the process of designing research surveys. In fact, the language skills that allow you to formulate clear questions and answers that really capture what respondents think and feel are the most important area of knowledge you need to have as a survey researcher, according to O*NET.
There are many different kinds of survey research questions you could choose to include in your questionnaire. Closed-ended questions require respondents to choose their answer from a list of options. However, these questions can take the form of a true/false question, a yes/no question, a multiple-choice question or some form of ranking or rating scale. The other major type of question is the open-ended question. Open-ended questions ask the participant to provide their own answer – as short as a few words or as long as several paragraphs. There are also funneling, factorial and unfolding questions, which may be closed-ended or open-ended but which involve a series of related questions or questions that go along with provided content, like a hypothetical situation.
Generally, closed-ended questions are able to provide quantitative data, while open-ended questions are better able to provide qualitative data. Both types of data are important in the social sciences, and many survey researchers combine both types of questions to acquire each type of data.
Randomized and Nonrandomized Research Designs
A big part of research design is planning your method for selecting participants. Except in unusual cases, such as studies of rare conditions or situations, it simply isn’t feasible for a researcher in any field to survey every single member of the relevant population. Instead, survey researchers must use statistical analysis methods, along with some essential critical thinking, to identify a way to obtain a sampling of participants that can reliably be assumed to represent the entire population. One way researchers may accomplish this is through randomization.
Choosing participants at random helps avoid the bias of, say, choosing participants in the same geographical area, socioeconomic class or level of education. However, making your survey population too random – not filtering by any factors – could make it difficult to get the specific data you need in certain areas of research, such as investigating certain types of sociological phenomena or mental health conditions.
Survey researchers can use many different types of sampling, from simple random sampling to systematic and multi-stage sampling techniques. Some research is designed to be completely randomized, others in a randomized block, and still others – sometimes considered quasi-experiment – are nonrandomized.
Students preparing to work as survey researchers may take numerous courses related to sampling and randomization of research design, including sampling theory, randomized and nonrandomized research design and applied sampling.
Other Types of Research Designs
Surveys are just one subtype of one category of research design. While surveys are the kind of research design your career will revolve around in this particular occupation, you should understand the different kinds of research categories that can be attempted and the advantages and disadvantages of each one. If your work is not strictly survey research, but instead research in a field like one of the social sciences, then you might use a combination of research methods to investigate your field of inquiry.
Survey typically falls under the category of a descriptive form of research design, along with case studies and naturalistic observations. Experimental research methods include controlled experiments and field experiments, while observational studies and case-control studies fit in the category of correlational research. Researchers can also learn from conducting literature reviews and meta-analyses.
Different forms of research can shed light on different aspects of a complex field in the social sciences. A controlled experiment shows what participants would do when exposed to stimuli, while a survey shows what participants are actually thinking.
For Further Reading: