Start by formulating the question and the purpose of a planned survey. What do you want to achieve? Is there a real need for a survey or is it done because it is customary, or because it feels like a requirement? Most surveys conducted by breed clubs are so-called health surveys, containing questions about the possible presence of various diseases. However, some surveys have been more focused on mentality and function. You want to get as much information as possible when a questionnaire is sent out anyway, but how comprehensive can a questionnaire be before the recipient throws it in the bin immediately?
Some tips along the way:
Formulate the main purpose of the survey
Think about whether a survey is the best way to get information about what you want
Do not ask about everything but only things that are relevant to the purpose
It should not take more than 30 minutes to complete the questionnaire
Target group and selection
The next question that is fundamental to the outcome of the survey is: to whom should the survey be asked, ie who is the target group? All dog owners in the dog breed? Breeder of the dog breed? To owners of dogs of all ages, or to certain age groups?
Even if you make a first selection, you probably have to, for practical reasons, limit the number of individuals that can be included. The answer to the question of target group depends, among other things, on the purpose of the survey. If the purpose is to get an idea of the prevalence of health problems that usually occur in middle-aged or older dogs, it is hardly an idea to send out the questionnaire to owners of puppies and young dogs.
It is usually practical to select a certain time interval, for example all living and dead dogs born during a certain time period. Keep in mind, however, that if the time interval extends far back in time, the dropout rate can be large among the dogs that became ill or died at a young age. Too short a time interval can mean that the surface becomes small, depending on how many puppies are born within the breed. It is important to inform those who respond to the survey that the deceased dogs can provide very valuable information, both if the dog has
died at a young age or become old!
The vast majority of clubs cater to both active breeders and "regular dog owners", however, it is conceivable that you will miss those who are not members of the breed club.
To find dog owners of a certain breed, you can apply free of charge to get addresses from SKK's owner register. If the club wants access to addresses from the register, contact Helena Nyberg at SKK's member department firstname.lastname@example.org
Including those who are not affiliated with the club in their target group can be a great way to attract new members to the club! It is extremely rare to be able to do a so-called total examination. Instead, a sample is made of the people, or in this case dogs, from the entire population who are to be included in the survey.
Regardless of how the selection takes place, you want the individuals ("sample") that are included to be representative of the entire population. Making a random selection means that each individual in the population has an equal chance of being selected. However, this is not entirely easy to achieve in practice! One way can be to draw lots for which dogs to include after the criteria, for example with regard to age range or year of birth, have been set.
In order to achieve a high response rate, it is important to have a broad support in the club so that the survey feels like a matter for the whole club. Inform and advertise in the club magazine and on the club's website.
How many questionnaires should you send out?
Once you have decided who constitutes the target group, the next natural question is how many questionnaires to send out. This depends, among other things, on how far-reaching conclusions one wants to be able to draw from the survey, and what response rate one expects. Keep in mind that not everyone will respond to the survey. Economic trade-offs also come into play. A basic precondition for a survey to be carried out is that the breed's number is large enough for the survey to be compiled in a meaningful way. A rule of thumb is that the larger the sample, the higher the probability that the sample is representative of the population as a whole.
However, it is difficult to give an exact figure for how many surveys should be sent out, the number must be adapted to the purpose and to the different conditions prevailing in the clubs. If, for example, the purpose is to detect the occurrence of a rare disease, a larger number of questionnaires must of course be sent out compared to if you ask how to activate your dog.
Regular survey or web survey?
The choice between a paper survey or an online survey is largely a matter of taste among those who will work with the survey and is largely governed by the club's finances and the number of surveys to be distributed. Sending out a paper survey by post entails costs for materials and postage, but can be perceived as more personal by the recipient than an online survey.
A paper survey can be an advantage if the number of dogs to be included is small, but usually requires more work in compiling the answers than a well-designed online survey. In a numerically larger dog breed, a web survey may be preferable, however, web surveys can also entail high costs if you choose to use a commercial program.
The design of the survey
The quality of the information obtained through a survey depends to a large extent on how the survey is designed, "as you ask, you get answers". Below are some tips that can be good to take with you when designing the survey.
Anonymity or not
A basic question is whether to include the dog's identity (registration number) or whether the survey should be anonymous¹. In this decision, the purpose of the survey should be weighed. If the purpose is to get a general picture of the health status of the breed, an anonymous survey may provide more reliable answers.
If, on the other hand, you want to investigate whether a specific health problem is more common in certain lines or families in a breed, the dogs' identity is a necessity. The latter also applies if you want to be able to follow up individual dogs over time. Including the dogs' identity in the survey also provides, unlike an anonymous survey, an opportunity to evaluate how representative the sample of those who responded to the survey is.
It is important to inform that even if the dog's identity is stated, the answers will not be reported for individual dogs! The question of anonymity can also be solved by coding the questionnaire, which gives anonymity to the person compiling the answers, but an opportunity to identify the dogs if required.
Rules of thumb for survey design
When formulating the questions, there are some rules of thumb to consider:
Set up the questionnaire so that the questions have a logical order.
Feel free to start with questions that are easy to answer.
Feel free to number the questions and use the layout for grouping related questions.
Try to find appealing headlines. For example, write "Eyes" and "Skin" as headings rather than "Eye problems" and "Skin problems".
Use ordinary language and avoid difficult-to-understand technical terms, but formulate the question so that it cannot be misinterpreted due to too rough a simplification.
Avoid value-laden words and long wordings. Remember that words such as "often", "rarely", "almost never" mean different things to different individuals!
Ask about one thing at a time. For example, split the question "is your dog afraid of shots and fireworks?" in two questions.
Feel free to use numbers when you are looking to rank how often a behavior or activity occurs.
Feel free to build up the question with follow-up questions that make it easier for those who do not master medical terms, if these terms must be used. For example, "Does your dog have allergies?" is not the same question as "Has your dog been diagnosed with atopic dermatitis?".
Feel free to include footnotes or glossaries to clarify medical or other technical terms in case these terms are considered necessary.
If the questions relate to disease diagnoses, it is important that it is clear whether the diagnosis has been determined by a veterinarian or is based on the dog owner's own assessment.
Do not forget to leave an opening in the survey to capture things we do not know in advance. Ex. there may be health aspects of the breed that are not covered under any of the predefined headings.
Be careful not to construct questions that exclude multiple answer options if necessary. For web forms, it is extra important to think through the use of radio buttons that exclude another answer option. For the question "Do you have a" Bitch "or" Male ", it is good if one answer excludes the other. However, it can cause problems if the question is "Has your dog had a tumor disease?" and only one response option can be indicated, as the dog may have suffered from various tumor diseases.
Predefined response options facilitate the compilation and analysis of the survey responses, but also limit the information you receive. Do not forget the option "do not know" or "other". For some questions, it can be an advantage to also include space for free text / comments to enable a more nuanced picture.
Feel free to test the survey on a small group before it is fully launched. It provides an opportunity to capture ambiguities or misunderstandings. You often get many good views on both the technical form and the design of the questions.
Dispatch and reminder
Formulate a cover letter
Formulate a nice and informative cover letter that clearly explains the purpose of the survey, who it is addressed to and how the answers are handled and reported. The cover letter can be decisive for whether you want to answer the survey or not! Keep a positive tone and emphasize the common good of the dog breed. It is important to inform that no results for individual dogs or breeders will be reported. It can be an advantage if the cover letter is built into the form.
Choose a time with care
The timing of the mailing can be of great importance for the response rate. Avoid sending out the questionnaire in connection with, for example, summer holidays or Christmas weekends.
Anyone who has filled in a web form must of course receive a confirmation that the survey has reached its recipient!
Reminder gives higher response rate
For a high response rate, a reminder should be sent to those who did not answer the questionnaire after a certain time. A reminder via email can be a way to save costs. You may even be able to hand out some tickets or a prize among those who have answered. Compilation and accounting
Who compiles the survey results?
Before the survey is conducted, it should be clear who or who has the time and competence to receive and compile the survey results. The recipient of the survey responses should be an uncontroversial person, perhaps even someone outside their own breed club. Organize the work so that even the subsequent board or breeding committee can take over and continue the work. A well-thought-out design of the survey facilitates compilation and analysis.
What analyzes can be done?
A survey of the kind that can be carried out within a breed club seldom provides opportunities for far-reaching statistical analyzes. The material is usually too small and the statistical competence in the breed club is limited. It is not reasonable to require strict statistical significance in the way that can be done in a scientific study. These limitations are important to take into account in the interpretation of the questionnaire responses. Feel free to present facts and draw conclusions, but do not exaggerate the importance of individual answers!
Include in the cost calculation a possible purchase of computer software for compilation and processing of the answers. The results are usually more easily illustrated if they are presented as curves and bar graphs compared with text and tables. If average values are reported, please state the standard deviation around the average value. Be careful to express the results in percentages if the number of dogs is small; then rather print out the actual numbers, for example 3 out of 15 instead of 20%.
Reporting the results
It is of the utmost importance that those who have contributed to the survey are allowed to take part in the compilation of results within a reasonable time. The results can, for example, be reported in the club magazine and on the website. Through reporting on the web, dog owners and breeders who are not members of the breed club are also reached. Reporting and discussion of the results should also be included at breeders 'and members' meetings and at the annual meeting. Remember to archive the survey results and make the results easily accessible to your successors as well!
Follow-up and measures
The compilation of the questionnaire responses should form the basis for a discussion about what conclusions can be drawn from the new information and the need for possible measures. Examples of questions are: how do the results in the survey correspond to our expectations and to information from other sources, such as insurance data and SKK data? Do the questionnaire responses provide new knowledge that gives rise to urgent measures or indicate a need for more in-depth investigations?
Perhaps the new knowledge added by the survey suggests that the existing breeding strategies or priorities in the breeding work should be changed, provided that the response rate is sufficiently high. The compilation and conclusions of the survey should of course be included in the breed-specific breeding strategy, RAS.
For more detailed information on questionnaire studies, please refer to the books below:
The survey in practice - A handbook in survey methodology by Göran Ejlertsson
The questionnaire by Jan Trost: Enkätboken (in Swedish)- see free download @ https://pdf.zlibcdn.com/dtoken/87115528aa37feba09f115ada6584a2c/Enkätboken_by_Trost,_Jan)_2550531_(z-lib.org).pdf