Summary

• Article Source: CONDUCT RESEARCH, The University of Texas at Austin, Sep 21, 2011.

Survey

Overview

A survey is an ordered series of questions or statements assessing attitudes, behaviors, or personal characteristics that is administered to individuals in a systematic manner. Surveys may be administered in a variety of mediums (e.g., paper, oral, electronic) using various delivery methods (e.g., face-to-face, telephone, mail, Internet).

Types of surveys

With the growth of the Internet and e-mail, electronic surveys are becoming more widely used. They can be distributed by paper, as e-mail messages, or posted as forms on the Internet. Both paper and electronic surveys have strengths and weaknesses.

Comparison of paper and electronic surveys
Comparison of electronic survey tools

Comparison of paper and electronic surveys

Suggested uses of surveys

Evaluating program performance. Gaining insight into attitudes and outcomes.
Assessing changes in practices, especially when used as part of a single-group experiment. Measuring the effects of an intervention when used as part of a single-group experiment.

Limitations of surveys

Not suitable for assessing individual performance. Not suitable for collecting in-depth information. Requires some knowledge or understanding of relevant issues in order to write appropriate questions and properly organize a survey.

Resource requirements

A moderate level of knowledge about survey design and question writing is required unless you are using previously validated questions or surveys. You should also understand how to use and interpret basic statistics (e.g. frequencies, means, weighting), and have experience or training in constructing and using spreadsheets databases (for very large classes or surveys). Data entry may be time consuming, requiring additional staff, although using scanable answer sheets or an electronic survey tool will greatly reduce the knowledge, training, and time required to enter and analyze survey responses. more

Plan your survey

STEP 1. Identify the educational research problem or topic

Pay attention to the feasibility of your research problem or topic and whether it can be researched systematically. Determine the resources needed to conduct the study, your interest level, its size and complexity, as well as the value of your results or solution for both theory and practice.

To thoroughly describe the research problem or topic, create a statement that includes the educational topic or specific problem and the justification for research.

STEP 2. Review prior research

The literature review will help you gain an understanding of the current state of knowledge pertaining to your research idea. It will inform you of what data collection methods have been used for similar research and to help make sense of the findings from these methods once data analysis is complete. Try to specifically explore previous research that has used content analysis for research problems or topics similar to your own.

To review prior research, the most effective and efficient way is to search educational journals through electronic computer databases such as ERIC, PsychINFO, or Google Scholar. Searching other library databases is also recommended.

STEP 3. Determine the purpose, research question(s) or hypothesis(es)

Identifying a clear purpose helps determine how the research should be conducted, what research design you will use, and the research question(s) or hypothesis(es) of your study.

Hypothesis – a predictive statement about what one would expect to find or occur if a theory is correct.

For surveys, the purpose of your study will generally be more explorative or descriptive in nature rather than testing a hypothesis. However, surveys can also be used to explain (and triangulate) findings obtained from other methods. For example, if we determine from an experiment that A is better than B, a survey can be used to help us understand why A is better than B.

Triangulation - using multiple research methods to gather information or multiple sources of information on one topic or research question usually with the intent of improving reliability and/or validity . Sometimes referred to as using “multiple measures.”

A more complex survey is a validated scale, which can be used to determine the effects of an intervention or to predict future behavior. more

Validated scale - a collection of questions intended to identify and quantify constructs based on educational theory that are not readily observable such as knowledge, abilities, attitudes, or personality traits.

Once you have created your research question(s) or hypothesis(es), specify or match which question(s) surveys will help to answer, or which hypothesis(es) surveys will help to triangulate or test.

STEP 4. Consider the research implications of survey findings

Implications are the practical ways your research will assist the field of education. These are the underlying goals, the rationales for, or the importance of your study. Implications are linked to your research problem or topic, research purpose, and research question(s) or hypothesis(es) of your study. Therefore, once you have matched the research question(s) or hypothesis(es) that survey findings will try to answer, determine the implications of these findings and how it will aid the field of education. This will help you keep focused and maintain a clear vision when planning surveys and interpreting results.

STEP 5. Create the survey

Clearly determine what you want to measure

From your research question(s) and hypothesis(es), you should have a clear idea of the attitudes, behaviors, or personal characteristics you want to measure using the survey. If you want to measure certain educational constructs (or quantify educational phenomenon) that is grounded in theory, you may want to create or use a certain type of survey often referred to as a validated scale. more

Write survey questions

Writing good survey questions is crucial to avoid compromising the validity of responses and limiting your ability to answer research questions. Rewrite questions until they are clear and succinct. more

Surveys can also contain related multiple questions that can be combined or summed to measure underlying constructs. These surveys are often referred to as summated scales. more

Determine question type

The information you want to obtain and how you plan to use it should dictate the question type or response scale you choose.

More about question types
More about response scales

Organize and format the survey

The survey format is very important because a poorly organized survey may confuse respondents and lead them to skip questions or not complete the survey. more

Conduct pilot testing

Test the survey on a small sample of individuals that resembles your target sample (but does not include it) to check if the questions are answered as you intended and how long it takes to complete the survey. Revise questions as necessary before administering them to the study sample.

References

[1] Babbie, E.R. (1973). Survey research methods. Belmont, CA: Wadsworth Publishing Company.
[2] Bordens, K.S. and Abbott, B.B. (1996). Research Design and Methods; A Process Approach. 3rd ed. Mountain View, CA: Mayfield.
[3] Punch, K. F. (2003). Survey research: The basics. London: Sage Publications Ltd.