Health surveys - creating a survey

Dog standing on rocks
Once you have decided on the type of breed health survey you will use, there are four essential steps that you will need to follow.
  1. Plan - set your objectives and design the survey
  2. Issue - communicate and motivate father responses
  3. Analyse - analyse the responses and publicise the results
  4. Improve - implement your priority actions, review and repeat the survey
This toolkit will take you through the first two steps and give you practical advice on how to carry out a survey. For steps 3 and 4, please refer to our analysis and reporting toolkit.

Step 1: Plan

Decide what you want to achieve

Start with the end in mind! You need to be clear what information you are trying to obtain and how you will use the data from your survey. Thinking about the questions is only part of the design process. You also need to consider how you want to analyse and display the results.

Tip: Try to identify the types of graph/chart you’ll want to present in your final report, so that it’s clear in your mind what questions you need to ask.

You may want to analyse the results for different categories of response e.g. dog/bitch, entire/neutered, bred from/not bred from, Standard/Miniature, or age.

To achieve this you will need to include an “identifier” page of questions, usually at the beginning of the survey.

Tip: Create these identifier categories as lists of options rather than allowing free-form answers, to ensure consistency.

This also raises the question of whether or not the survey should allow anonymous responses. If you can guarantee anonymity, then it may be appropriate to ask for each respondent's name or e-mail address, particularly if you need to target people for any follow-up investigations. You may decide that you need to be able to identify the registered names of the dogs whose data are being collected. This will enable you to carry out any subsequent analysis of pedigrees, but it may put some people off responding to your survey.

Survey design principles

  • Keep the survey focused on your objectives and be concise (avoid asking too many questions about too many topics)
  • Make it easy for people to answer the questions (they need to be clear and unambiguous)
  • Ask the right type of question (avoid leading and multiple questions – more on this later)
Tip: The shorter your survey, the more likely you are to get a high response rate. High response rates will give you more confidence in the validity of the answers.

Every question should add value and move you closer to meeting your objective. For each question, ask yourself “what would I do if I knew the answer to this question?”

Tip: Try to avoid mixing too many different types of question. Respondents will find the survey hard to follow and may get confused by the different types of response expected of them. Think about the logical flow of questions; don’t jump around from topic to topic.

Types of questions

The main types of question you may wish to use are:
  • Multiple choice
  • Rating
  • Ranking
  • Open
  • Closed
Multiple-choice questions
The respondent chooses one or more options from a list.

e.g. Which of the following eye conditions has your dog been diagnosed with?
  • Entropion
  • Ectropion
  • Distichiasis
  • Progressive retinal atrophy
  • Other; please specify
Make it clear if people are allowed to select more than one option and give them the choice of adding an alternative to those on your list. If you only want people to select one item, ensure the list contains mutually exclusive choices.
Rating questions
The respondent is asked to select one point on a rating scale (e.g. from strongly agree to strongly disagree). It is good practice to allow "don’t know" or "not applicable" responses.

To what extent do you agree with the following statements? Strongly agree Agree Neither agree, nor disagree Disagree Stongly disagree Don't know
My dog has an outgoing temperament.            
The breed club is doing a good job at communicating health issues in the breed.            

There is endless debate about whether rating scales should have even or odd numbers of choices and there’s probably no right answer!

The middle option of a rating scale often attracts a substantial number of respondents who are unsure of their opinion, or who want to “sit on the fence”.

A variation on this type of rating scale is to use a question that has two opposite statements at the ends of a rating scale, and the respondent is asked to choose from a numerical scale between the two ends.

Top: Make sure that you create statements that are truly “opposites”, e.g.

This dog's temperament is nervous 1 2 3 4 5 This dog's temperament is outgoing
Ranking questions
Respondents are asked to place a number of options in their order of preference.

e.g. Please put the following methods of communication used by the breed club to disseminate health information in order of its effectiveness (1 = most effective; 5 = least effective):
  • Email newsletters
  • Club Facebook page
  • Breed seminars
  • Club website
  • Club newsletter (paper)
Tip: Don’t make a ranking list too long – no more than 5 or 6 items.

Bear in mind that ranking questions can be quite difficult to analyse and present results. You’ll probably end up with something along the lines of: “80% of respondents rated email newsletters as the least effective method”.
Open questions
These are free-form questions asking for a text response; e.g.

“Please tell us what you think is the most significant health issue we need to address in the breed at the moment.”

Tip: Leave plenty of space for responses to open questions.

Remember, these types of answers are difficult to analyse, but they can give some useful insights into reasons behind responses to the other types of question. Don’t have too many of them otherwise you’ll spend a considerable amount of time trying to get any meaningful data. It’s usually a good idea to have a single, final open question to allow respondents to provide you with any other information that they feel would be helpful.

Tip: If these closed questions are clearly defined, you may not need to give people a “don’t know” option.
Closed questions
These questions are used to get “yes” or “no” answers; e.g.
  • Has this dog been bred from? Yes/No
  • Has this dog been spayed/neutered? Yes/No
  • Has this dog had a Cough in the past 12 months? Yes/No
Tip: If these closed questions are clearly defined, you may not need to give people a "don’t know" option.

Wording of questions

It’s very easy for words and instructions to be misunderstood and this is particularly the case in surveys because there is no opportunity for the respondent to ask questions.

Make sure that your instructions are absolutely clear. Keep sentences short, use lots of white space on the page and avoid jargon or words that may not be understood by everyone. You may also need to provide a separate glossary of health conditions, where you explain briefly what each condition is and its symptoms.

Tip: Ask somebody with a fresh pair of eyes to read your draft survey to check for errors and make sure it’s all clear and easy to understand.
If you are able to run a pilot (trial) of your survey, this can help you identify aspects of its design that might cause problems for respondents and therefore you can then make changes where necessary. A pilot might highlight questions that respondents are misinterpreting, or where you have missed possible answer options. It will also give you an idea of how long the survey will take for people to complete and you can get their reactions to the overall design and approach of the survey.

Tip: If you want to pilot your survey, take it to a club/council committee meeting and get people to try it out. If you’re holding a health testing day, or a seminar, these can also be good places to get initial reactions.

Other things to be aware of

Multiple questions should be avoided
e.g. the following should actually be two different questions. One is about exercise and one is about grooming.
  • How often does your dog get exercised and groomed?
Or this one, which should also be two questions:
  • Does your dog have an outgoing temperament and is it good with children?
Leading questions should be avoided
  • Do you prefer to give your dog food X or food Y?
The respondent may prefer neither of these options!
One small word can alter the meaning of a question radically
  • “My dogs are vaccinated” may get a very different response to “My dogs are vaccinated annually”.
Tip: Begin with easy/safe questions. If your survey is going to touch on sensitive issues such as inherited conditions that have occurred in the respondent’s dogs, start with questions that are “safer” (e.g. Has this dog been bred from?) and build towards the more difficult ones. You don’t want to put people off right at the start.

Tip: Check spelling, punctuation and grammar. If you’re not confident in your spelling, punctuation or grammar, ask someone who is for help. Always proof read your survey before it is issued.

Administrative details

Recipients will need to know:
  • The purpose of the survey – what incentive is there for people to respond?
  • Specific instructions for completing the survey – how to fill it in and how long it is likely to take
  • What you plan to do with the results – including whether they will get to see them
  • How you will deal with confidentiality?
  • Where, when and who to return paper-based surveys to – you may want provide a reply-paid envelope for paper-based surveys
  • How to access and complete the survey, if it’s being done online
  • Who to contact for help, or to answer queries?
Generally, you will need to allow people two to three weeks to respond (which makes allowances for holidays). Any longer and they may forget. Try to avoid issuing surveys around major holiday periods (e.g. summer/Christmas).

Finally, don’t forget to include a “thank you” at the end of the survey to let people know that their time and effort is appreciated.

Tip: Include a cover letter or statement from your breed’s health sub-committee chairman explaining why the survey is important, how it will be used and how it is endorsed by all your breed’s clubs.

Step 2: Issue

The focus of this section of the toolkit is to offer ideas on how to improve your survey’s response rate. This applies during the period leading up to the survey and while responses are coming in before the closing date.

Communicate and motivate

Unless you can get a good response rate to your survey, the results you collect may not be of much use. A communication campaign using multiple channels (e.g. email, website, newsletters) over the weeks and months before and during the survey will greatly increase your chance of having a good response rate.
Produce a user-friendly questionnaire
Make sure the questions you ask are kept simple, concise and easy to complete. Create a logical flow throughout your survey and do not ask too many questions. Respondents may be overwhelmed by the sight of a long or complicated survey and may be less likely to complete it.
Perform a pilot test
Test your survey with a small group of friends or colleagues before finally distributing it. Ask them to read through the questionnaire, or to try and answer the questions themselves. Getting a second opinion gives you the reassurance that all of the questions are short and well phrased. People are less likely to respond if the questions do not make sense or are confusing.
Give pre-notification
Let people know that you will be carrying out a questionnaire well in advance. Make sure you publicise when you will be carrying it out and why it’s important. People will be more likely to complete the survey if they've already heard about the questionnaire and are aware of why they should respond. Make sure to tell as many different people in as many different ways as you can think of.
Personalise any communications
If respondents feel that the questionnaire, or any communication regarding the questionnaire, has been directed towards them, they will be more likely to respond. If possible, try writing any envelopes by hand, addressing any letters or emails to the intended recipient (i.e. “Dear Mr. Smith”) or hand signing any letters. Unfortunately, if you are sending out large numbers of questionnaires this may not be possible or practical.
Distribute your questionnaire to a number of different people
Assuming that you would like health information from a wide range of owners (not just breed club members) your communication campaign will need to use as many different approaches as possible.

First of all, think about all the types of owners you need to make aware of the survey, e.g.: 
  • Club and council committee officers/members
  • Breed club members
  • Pet owners
  • Owners who work their dog(s), or who take part in non-show activities
Distribute your questionnaire in a number of different ways
To increase the number of responses to your survey, try increasing the number of ways in which people hear about, or access, your questionnaire. Your questionnaire could be paper based or electronic, or to gain maximum coverage, why not use both?

Paper based

Paper-based questionnaires can be posted out to potential participants, or you could hand them out at shows, trials, matches, seminars or club/council meetings. If you are distributing them by hand, make sure to include a return postal address.


Regardless of whether your questionnaire is an electronic document (e.g. Word or PDF) or uses an online survey tool (e.g. Survey Monkey or Kwik Surveys), it can be made available to participants on your breed website(s) as a link or a download. Once you have an electronic link to the questionnaire, this can be distributed through many channels, e.g.:
  • on forums/discussion groups (e.g. Yahoo Groups, Champdogs, Our Dogs Forum)
  • on social networking sites (e.g. Facebook and Twitter)
  • in the canine press (e.g. Our Dogs, Dogs Today, Dogs Monthly, and K9 Magazine)
  • in electronic newsletters or via emails
Alternatively, you could distribute leaflets at shows, trials, matches, seminars or club/council meetings, which let people know how they can access your questionnaire. The more places your survey is available, the more people are likely to see it.

Tip: If you can brief people face to face, e.g. at a breed seminar or show, you will be able to explain the survey’s purpose and answer any questions.
Keep your initial email or letter short and to the point
The more information you have in your opening invitation to complete the questionnaire, the less likely your participants are to read it. Make sure you keep it short and to the point, but do include:
  • who you are
  • why you are carrying out the survey
  • what you will do with the data
  • how long the questionnaire is likely to take to complete
  • what the deadline for completion is
  • your contact details (for returning completed surveys or for any feedback or questions)
Ensure confidentiality
The types of questions you ask in a health survey may be of a sensitive nature. Some participants may therefore be hesitant about answering these questions. Giving reassurance that responses will be kept confidential, or that participants can opt out of identifying themselves, may make some people more open to completing your survey.
Consider offering incentives
The offer of gifts or prizes, no matter how small, may encourage people to complete your questionnaire. You may want to offer people the chance to enter into a prize draw when they complete your questionnaire or even offer to make a small donation to charity for every survey completed. Although this can be an effective way of improving your response rate, the size of the giveaway will be strongly dependent on availability of resources and funds.

Tip: The best incentive may be as simple as letting people know how the survey’s results will contribute to improving the breed’s health.

Gather responses

Once your survey has been announced, make sure you continue to be proactive and do things to ensure people remember to complete the survey and get it back to you on time.
Send reminder emails or letters
Sending out reminder emails or letters is likely to be the most effective way of increasing your response rates.

Some people may be very good at responding to you straight away, but many people will quickly forget about the questionnaire, or may put it on their ever-expanding to-do list. A few gentle reminders will be useful in jogging their memory. However, try not to send too many reminders and, if possible, try to send reminders only to those who have not yet replied. If you have used any forums, social media etc. to publicise your questionnaire, then don’t forget to issue reminders via these tools as well.
Monitor your response rates
If collecting paper-based surveys, you’ll probably only know near to the closing date what sort of response you’re going to have, as most people will leave it to the last minute. In the worst case, if you have a low response rate, you can always announce an extension for latecomers and perhaps allow another one or two weeks. If you do announce an extension, make sure there is enough publicity to help stimulate more replies.

If you are using an online survey tool, you will usually be able to see reports of how many surveys have been completed and take account of this information to trigger further publicity. These reports will sometimes show you if people are fully completing the survey, or if they are giving up part way through, or missing particular questions. Again, this information can help you identify further communications to improve response rates.

Tip: If, for example, you allow people a three-week period to complete the survey, it can be helpful to send weekly reminders by email and through social networking sites (e.g. “two weeks to go”, "only one day left to complete your survey”, etc.).
After the closing date
As soon as you have responses to your survey you can start recording them in whichever software tool you have chosen (e.g. MS Excel/MS Access, or one of the online services such as Kwik Surveys). Don’t underestimate how much time all this data entry can take - another good reason to opt for an online survey where the respondent fills in the information and you don’t have to.

Some final thoughts on how to increase your survey's response rate

People who responded to your questionnaire will want to see what the outcome was and what your findings are. Make sure that you publish your results online and, if possible, send your findings to those who participated. The sooner you can share some results with the people who completed the survey, the more they will feel their effort has been worthwhile.

Tip: Showing people how they helped will encourage them to complete any surveys that you carry out in the future.

Online software to create your survey

Options available to you
Providing you have a reliable broadband internet connection, online software can often be used for analysing your survey responses and to allow respondents to submit their survey responses. This gives you the potential to have your respondents do the data entry themselves, so all you have to worry about is analysing the results. However, you have to remember that not everyone has internet access and you might deter people from responding if they don’t have the option of sending you a paper survey response.

Tip: If you want to run a continuous health survey, rather than a one-off, you really need to use one of the online survey options.

If you do a google search for “online survey software” you will find many examples, but the most popular ones are Survey Monkey and Survey Console.

These are dedicated survey tools that allow you to create your survey online and produce a variety of simple, standard reports quite easily. You do need to check out the limitations of any “free” options; e.g. some tools only allow you to collect up to 100 responses.

Most online services will enable you to design surveys with a variety of question types, such as:
  • multiple choice with a single answer allowed
  • multiple choice with multiple answers allowed
  • ranking
  • star rating
  • text boxes
Tip: Don’t get carried away by using too many different question types; keep your survey simple for people to complete.

One alternative to the dedicated online survey software is Google Forms, which enables you to design a survey and store the results in a Google spreadsheet.

Whichever online survey software you choose, you will have the option to send a link for the survey in an email or embed your survey in a website for people to submit their responses.

Most of the online survey tools have fairly limited reporting capabilities which means that you will have to download the survey responses as a spreadsheet and then do your detailed analysis offline.


We wish to thank the following people for their help in developing this toolkit:

Main author

Ian Seath

The project team

Judith Ashworth, Sheila Atter, Archie Bryden, Brian Hill, Dorothy McIntyre,
Shula Shipton and Marion Wilks