surveys and questionaires
Survey research involves the collection of information from a sample of individuals through their responses to questions.. It is an efficient method for systematically collecting data from a broad spectrum of individuals and educational settings
One important point when using surveys and questionnaires is to have some quantitative data. One way of doing this is to have responses that are open ended and then to categorize them into groups according to the investigation.
Must be:
- A set of questions specifically designed to address a particular issue
- A printed form that participants complete themselves.
- A printed form that the interviewer completes or Online
- Delivered via email.
- Must be given to a statistically valid number of participants (minimum 30 people).
- Can be used to collect standardized data.The survey consists of eight questions. All are either multiple choice (participants were only allowed to choose one option), or free response question (no word limit)
Things to consider:
Make a table with the Question/possible responses/justification for asking the question and how it will help answer RQ.
Be careful of leading questions (one’s that hint at the way you want them to answer) and false binary questions. Usually an agree through disagree response options (1-5) does the trick in avoiding this. Plus if there are many 1s or 5s, it gives you some things to talk about in your RAC and DEV- why you think the respondents answered the way they did.
Here is a great link in how to design surveys - Survey Research
One important point when using surveys and questionnaires is to have some quantitative data. One way of doing this is to have responses that are open ended and then to categorize them into groups according to the investigation.
Must be:
- A set of questions specifically designed to address a particular issue
- A printed form that participants complete themselves.
- A printed form that the interviewer completes or Online
- Delivered via email.
- Must be given to a statistically valid number of participants (minimum 30 people).
- Can be used to collect standardized data.The survey consists of eight questions. All are either multiple choice (participants were only allowed to choose one option), or free response question (no word limit)
Things to consider:
- How much data do you need to collect? A general rule of thumb is that a minimum of thirty surveys are needed for investigating a correlation; at least thirty (ideally fifty) surveys per independent variable - so if gender was the independent variable then you would need 30 females and 30 males completing the surveys.
- What is the question you are trying to answer? What is your independent variable? What data will help answer your question?
- What type of data will be more meaningful and allow you the ability to analyze the data statistically? Questions allowing an answer on a scale of 1-5 or 1-10 are generally better than yes/no questions.
- How will you avoid bias in your data and how will you cope with "no response"?
- Will you trial your questions? How might you do this and refine your questions?
- How will you encourage people to participate in your survey?
- How will you allow participants to know your scale of reference, e.g. 1 is bad, 5 is good or 1 is disagree, 5 is strongly agree ?
- What tool will you use for your survey?
- What ethical considerations do you need to make?
Make a table with the Question/possible responses/justification for asking the question and how it will help answer RQ.
Be careful of leading questions (one’s that hint at the way you want them to answer) and false binary questions. Usually an agree through disagree response options (1-5) does the trick in avoiding this. Plus if there are many 1s or 5s, it gives you some things to talk about in your RAC and DEV- why you think the respondents answered the way they did.
Here is a great link in how to design surveys - Survey Research
