Survey and Questionnaire Design

Surveys give a snapshot of people’s thinking at a given point in time and collecting factual information relating to groups of people (Denscombe, 2010). Questionnaires are one of the most widely used means of collecting data (Rowley, 2014).

A sample population is a sub-group of the entire population (Denscombe, 2010). In my survey the entire population would be all staff in the Department of which I will survey line managers; this represents a sample of the population. My survey will be online, Bryman and Bell (2011) state that a researcher can achieve a higher response rate with online surveys. Saunders et. al. (2012) states that voluntary responses are more likely to be truthful.

The questionnaire is the medium of communication between the researcher and the subject (Brace, 2004). The questionnaire will have open questions and some based on the likert scale, where respondents are asked to indicate how strongly they agree or disagree (Rowley 2014). Denscombe (2010) states one advantage of online questionnaires is that there are no transcription errors.

 

Questionnaire Design

The design of the questionnaire will be self administered, internet-mediated one (Sauders, 2012;Bryman and Bell, 2011;). Questionnaires show maintain the focus of the research (Greener, 2008) and contain a series of questions types. The layout of the questionnaire should be logically laid out and easy to read (Saunders et. al., 2012; Bryman and Bell, 2011;Bell 2010).

The questionnaire should start with easy questions that do not require a lot of thought to ease the respondent into it (Saunders et. al., 2012).

 

Types of questions

• Open
• Close
  • Dichotomous
  • Likert scales
  • Multiple Choice
  • Ranking
  • Prompting
• Demographic questions

Open questions

Useful for collecting more in-depth insights, and allow respondents to use their own language and express their own views (Rowley, 2014). Open questions give richer data but are more difficult to analyze and code (Greener, 2008) but may produce unexpected results (Bell, 2010). Cohen et. al. (2011) state open questions are an invitation to the respondent to ‘write what one wants’.

I will need to be careful using open questions as I will need to analyse all the responses and it could be time consuming. Simply put, Open questions invite respondents to provide data or offer comments in a non-structured fashion that can be difficult to analyse later.

Closed questions

Have a limited number of responses that the surveyor can predicts. Closed questions facilitate the collection of quantitative data that can be measured, compared and examined against other sets of data (Saunders et. al., 2012).

Dichotomous questions are a type of closed questions which have only two possible answers, i.e. yes and no (Saunders et. al, 2012; Brace, 2004).

Likert scales give respondents a series of options to choose from , usually five or points and should include a not applicable (Rowley, 2014; Saunders et. al, 2012). Likert scale questions increases the level of accuracy of the data being collected. A summated score of the different options can be generated for each question

Multiple choice questions present respondents with a pre-determined series of possible answers (Saunders et. al., 2008) and are quick to analyse (Cohen et. al., 2011) however phrasing of the question is extremely important to ensure the correct data is being collected.

Other types of closed questions are categorised, i.e. answers can be broken into categories, and ranking, i.e. the respondent ranks a series of statements from one to five (Rowley, 2014; Brace, 2004).

Prompting gives a set of option to the respondent (Brace, 2004) for them to choose one or even more than one.

Demographic Questions

I chose not to use demographic questions, i.e. age-gender-location, as I believed it could be used to identify individuals. For example if a respondent replied they were a female Veterinary Inspector in the west of Ireland they could be easily identified and would not respond. Cohen et. al. (2011) states that respondents find this type of question threatening and they should be kept to a minimum.

 

Piloting

After the questionnaire is design it is important to pilot test it with a group that is similar to the target group for the questionnaire to give you a sense of whether the questions are straightforward and whether the questionnaire is easy to complete (Rowley, 2014; Saunders et. al., 2012). Piloting can help tease out technical matters; give clarity about layout and appearance, timing, length, ease or difficulty of completion; test questions for validity helping eliminating those that ambiguous; ensures the right balance of questions (e.g. multiple choice, open-ended, closed) (Bryman and Bell, 2011; Cohen et. al., 2011; Scott & Usher, 2011; Greener, 2008).

 

References

Bell J., (2010), Doing your research project: a guide for first-time researchers in education, health and social science, Open University Press, Maidenhead.

Brace, I. (2004), Questionnaire design : how to plan, structure and write survey material for effective market research, Kogan Page, Sterling, VA

Bryman A., and Bell E. (2011), Business Research Methods 2nd Edition, Oxford University Press, Oxford

Cohen L., Manion L., & Morrison K., (2011), Research Methods in Education (7th ed.), Routledge, Falmer

Denscombe M., (2010), Title The good research guide: for small-scale social research projects 4th Edition, Maidenhead, Open University Press, 2010.

Greener S. (2008), Business Research Methods, Ventus Publishing ApS eBook

Rowley, J. (2014). Designing and using research questionnaires, Management Research Review, 37(3), 308-330

Saunders, M.N.K., Lewis, P., and Thornhill.,A. (2012). Research Methods for Business Students (6th Edition), Rotolito Lombarda, Italy

Scott D. and Usher R., (2011), Researching Education Data; Methods and Theory in Educational Enquiry 2nd edition, Replika Press, India