Survey research method

Surveys

The basic process of survey research can be outlined as follows:
1. define your research aims
2. identify the population and sample
3. decide how to collect replies
4. design your questionnaire
5. run a pilot survey
6. carry out the main survey
7. analyse the data

A crucial part of good research design concerns making sure that the questionnaire design addresses the needs of the research. To put this another way; somehow we need to ensure that the questions asked are the right ones. To move from the research aims (1) to deciding what are the right questions to put on a questionnaire (4) is a key aspect that needs to be addressed by the researcher.
Define Your Research Aims
Start your survey by setting down the aims for the survey. To define the aims for academic, as opposed to market, research you will need to review the relevant literature and you may need to do some preliminary research amongst your target subjects. Fulfilling these aims should then drive the design of your questionnaire and help select questions that are relevant, concise and efficient.
Most researchers make the mistake of asking too many questions. This often arises from an incomplete analysis of how to meet the survey aims. Your greatest enemy in survey research may well be poor response rate. Clear and concise questionnaires can help get the best response.
Questionnaire Design
Design of the questionnaire can be split in to three elements:
a) determine the questions to be asked,
b) select the question type for each question and specify the wording, and
c) design the question sequence and overall questionnaire layout.

Available software tends to focus on support for (b) and (c).
Determine the Questions to be Asked
This step is a key one that seems not to be sufficiently stressed in the literature or conducted in practice. A key link needs to be established between the research aims and the individual questions via the research issues. Issues and questions can be determined through a combined process of exploring the literature and thinking creatively. A simple illustration of the outcome of such a process is given below.
Survey aims: to explore the factors that might explain the reasons that Leeds University candidates give for undertaking a MBA programme:

Issue: What reasons might candidates give for undertaking an MBA?

Is the candidate looking for:
career change
career advancement
higher remuneration
etc.

Issue: Could past experience affect the reasons?
How many years work experience does candidate have?

Issue: Could gender differences affect the reasons?

Is the candidate male or female?

Issue: Could educational background and attainment affect the reasons?

What is highest educational qualification obtained?
What subject area(s) is this qualification in?
The above process generates the focus for individual questions that can then be designed in detail.
If you are relying on the respondent to complete the questionnaire, begin with questions that will raise interest. However, there are different views on sequencing of questions. For example, someone might argue that the easier questions to answer should be at the beginning to get the respondent in to the swing of things. However, someone else might suggest that questions about personal data, which are easy to answer, should be left until the end when the respondent has committed themselves to answering and they are less likely to object to giving such data. Whatever approach you choose you should try to have a logical sequence, e.g. group together all questions that relate to similar areas.
You should try to keep the flow through a questionnaire logical and very simple, i.e. avoid complex branching. Although some questions may be consequent upon earlier answers, keep the number of branches to the minimum. If necessary, use two or three versions of the questionnaire for respondents in different situations.

Analyse the Data
A precursor to analysis is the coding, entry and checking of data. Some comments were made earlier about the statistical analysis packages that are available (e.g. SAS, Minitab and SPSS). In all instances data can either be entered direct or imported from other packages such as Excel, provided the instructions for the receiving package are adhered to. In all cases a similar approach is used for coding and formatting data.
Usually the data is help on the computer in a rectangular data table where each row represents a ‘case’, i.e. a specific respondent and their data. Each column represents a specific variable, i.e. the data for that variable for all respondents. Note that a question on the questionnaire may require more than one variable to specify the data collected by that question.
A variable will have a unique title and a specific level of measurement. The measurement level of a variable is important because it determines the type of analysis that can be undertaken. In ascending order of sophistication these levels are:
Nominal, Ordinal, Interval and Ratio
(Note in SPSS the Inteval and Ratio levels are grouped together and called scale.)
For ease of data handling and analysis the
values that variables can take are usually designated by numeric codes, even when the variable is a nominal one. For example, gender can take the value male or female, but would usually be coded O and 1 (or 1 and 0) for ease of handling. Putting these data entry codes on the distributed questionnaire can help at data entry time, but obviously has the downside of putting numbers on the questionnaire that are of no relevance to the respondent and therefore could make the questionnaire look messier than it needs to.
Analysis packages usually make arrangements for missing values to be coded automatically; if they do not, this will have to be specifically taken care of when entering data.


Source: A general introduction to the design of questionnaires for survey research, http://remiss.politics.ox.ac.uk