Surveys. Survey. A structured way to collect standardized information from individuals using a questionnaire. Surveys may be conducted once; at repeated intervals, or concurrently with multiple samples They may be used to collect information from a few or many.
A structured way to collect standardized informationfrom individuals using a questionnaire.
This set of slides draws on the booklet seen at left and other resources. We will cover:
Types of surveys
Pros and cons of surveys
Steps in conducting a surveyResponse rateCover letter
What do you think? Answer YES or NO to
each of the following
Check your answers
You want to collect information from individuals (vs. a group or collective)
You want standardized information from everyone
Potential respondents can read and write
You want information from many people
Privacy is important or independent opinions and responses are needed
You have resources to send, track, analyze and interpret the questionnaires
Things to consider:
Tradition of reading, writing
Not best choice for people with oral tradition
Translation (demands more than just literal translation)
How cultural traits affect response
How to sequence the questions
Pretesting the questionnaire when it may be viewed as intrusive
Computer access and use if an electronic survey
Recommendation: use a mix of modes to ensure that everyone can and does respond
“It depends”… upon
If you’ve determined that a survey is the best and most appropriate way to collect information, then take some time to plan your survey.
Determine who should be involved in conducting the survey - - engage them
Define what information you will collect
What do you want to know?
How will use the information?
Identify the respondents
Determine sampling strategy, if a sample is to be used
Select how the survey will be distributed: telephone, mail, hand-out, email, web-based
Think about data analysis – what will the end product/final report include (keep ‘the end in mind’!)
For tips and examples for ‘getting your survey done’, see pages 9-15 in the Survey booklet
See the ‘Questionnaire Design’ section of this web site for help with developing your survey questionnaire.
Anonymous means that NO ONE can identify who provided the information
Confidentiality means that you are able to identify the person but you guarantee that the information will not be identified with the person
The proportion of people who respond:
Example: If you distribute 50 questionnaires and you get 25 questionnaires back, your response rate is 50%.
# that answered = response rate
# you contacted
High response rate promotes confidence in results.
Lower response rate increases the likelihood of biased results.
See pages 15-17 in the survey booklet
Generate positive publicity for your survey.
Appeal to people’s helping tendencies – ask them to help.
Make the topic salient - seem important
Ensure that respondents see the value of participating.
Point out personal connection to the topic
Tailor, personalize communications
Make the questionnaire interesting-short and easy to complete AND easy to return
Use language that is suggestive rather than decisive.
For example: “The data suggests” vs. “The data prove”; “It appears” vs. “It shows”
Don’t generalize findings to the entire group.
Clearly describe who responded, i.e., who the data represents.
The purpose of the survey and its importance
Survey sponsor − use letterhead if a mail survey
Why the respondent was selected to participate
How the information will be used
Assurance of anonymity or confidentiality - Human Subjects Protection
Instructions for returning the survey
Date - when to respond by
How results will be made available
Your contact information
Personalize the letter in salutation or signature
Connect with the respondent
Hand-sign the letter
Express appreciation for their participation
Keep it short – not wordy
For mail survey, include pre-addressed, stamped return envelope
Remember to pilot -test your
cover letter just as you pilot
-Test your questionnaire!
May be a colleague or collaborator or other stakeholder will check your cover letter for you.
Is the cover letter likely to motivate the person to respond?
Spend a few minutes reflecting
on what we covered in these slides