A survey is a research method used for collecting data from a pre-defined group of respondents, to gain information and insights, on a topic of interest. Depending upon the methodology chosen and their purpose; surveys have a variety of functions and can be executed in many ways. A survey involves asking people for information through a questionnaire. You can distribute this on paper, although with the arrival of new technologies, it is more common to distribute them digitally via email or social networks. They can also be administered by telephone or face-to-face. Mail and electronically administered surveys have a wide reach. They are relatively cheap to administer, information is standardised and privacy can be maintained. These approaches do, however, have a low response rate and cannot be used to investigate issues to any great depth. They also require that the target group is literate and do not allow for any observation.
Surveys are just one way of gathering information but in the peer support space, their use has been successful and wide-ranging. They are usually asking people to answer the questions on ‘a level playing field’ to avoid biased opinions that could influence the evidence we are collecting. As surveys are self-reported by participants, it is vital they are designed and tested for validity/reliability with the target groups who will be completing them. If your resources do not allow for this, then careful attention must be given to the design of the survey. If possible, the use of an already designed and validated survey instrument will ensure that the data being collected is accurate. If you design your own survey, it is necessary to pilot test the material on a sample of your target group to ensure it is appropriate for the target group. Make sure that wherever possible, you use easy English and pictures, particularly, if your respondents are likely to struggle with more complex or wordy communications.
You are able to ask survey questions in several ways. These include: closed questions, open-ended and scaled questions, and multiple-choice questions. Closed questions are usually in the format of yes/no or true/false options. Closed questions give a limited choice of responses, but they are quick and easy to process and collate. Open-ended questions leave the answer entirely up to the respondent and therefore provide a greater range of responses. While open questions enable the respondent to answer freely and gives greater choice of responses, the data is then difficult to collate or group. Surveys can also utilise scales to assess attitudes. Semantic scales (where responders are asked to rate subjectively something from 1 to 5) are also widely used. For example, ‘How connected do you feel with your peer group’ on a scale of 1 to 5 (when 1 is not at all connected, and 5 is extremely connected)? You can also utilise multiple-choice questions. For example, asking respondents to indicate their favourite topic covered in the peer group, or their preferred location.
When constructing a survey, there are a large range of considerations. These consist of: question sequence, layout and appearance, length, language, together with, an introduction and cover letter. The length of the questionnaire will depend on your aims: 7-10 questions (no more than 1-3 pages) is usually an appropriate number. The layout of the questionnaire is equally important. Start by asking relevant background information and then lead into more specific and/or complex questions. It is a good idea to place any sensitive questions near the end of a survey, rather than at the beginning. You should only ask questions, where responses are relevant/required, as well as always being polite, neutral and sensitive to people who might not feel comfortable sharing some information such as age, gender or cultural background. It is important that you “road-test” your survey with similar responders, to those, you plan to survey. Their feedback will help you modify questions, which might be difficult, poorly worded or confusing.