Brought to you by the TQS Professional Services Team
You don’t need a research degree to create a survey that will produce the most accurate and unbiased results. Below are some tips to help you create a professional survey in minutes.
On the Define Page:
- Check your Survey Name vs Internal Name: The Survey Title is what the respondent sees before entering a survey. We suggest using a general topic like “Lifestyle” or “Beverages” to avoid bias. (You can actually hide this to avoid bias with our Hide Survey Title feature. The internal survey title can be more descriptive, and useful for you or your team. For example, you can say Soda_Questionnaire_Round1_US)
On the Build Page:
- Preview your survey. Previewing your survey is always good to work out logic, grammar, and more. In school, we were told to always double check our work, well, it’s true for surveys too.
- Check your Filter Logic: Is it correct? Will it render the correct results? Will the respondents who need to be screened out, get screened out? Test every situation possible.
- Check that you have randomized applicable response lists to avoid bias. See our blog on randomization for more information.
- Check that “None” answers are exclusive and/or anchored in choice questions. Anchoring “none” to the bottom of a choice list is proper survey etiquette. Making the choice exclusive in a multi-select makes sure that if the respondent cannot say “yes I exercise 2 times a week” and also “none of the above.” It’s one or the other.
- Check that scaled questions are set up as “Rating” questions instead of a choice question: If you are looking to ask something like “How likely are you to purchase a low fat dairy product in the next 3 months?”, instead of using a choice question, use a rating question so that you can scale your answers. This question type utilizes preset scales and customizable weights and is typically most useful for finding levels of agreement, likeability, interest, and more. Setting scaled questions as Rating will allow you to gather statistics like mean, median, standard deviation and standard error in just one click within the TolunaAnalytics report. This can be a huge time-saver! For a full list of question types, see this post.
On the Launch Page:
- Check that your Targeting has been set and is feasible. On the launch page, you can choose your demographics and targets to your needs. At the top of the pop-up, you can see feasibility for your target.
- Check that you have set the appropriate “Limit number of respondents” who will start the survey. Make sure that this number matches your respondent count so that the survey works fast and efficiently.
- Check that you have excluded previous surveys if needed. Perhaps you are launching two similar surveys on exercise, but you want to make sure to get different respondents for each. You launched Exercise Survey 1 on Thursday, and you want to launch Exercise Survey 2 on Friday. You can exclude Exercise Survey 1 respondents from Exercise Survey 2 by clicking the button below.
- Test the “Test Link” on the “Launch” Page under the “Link” section: Using a test link with a colleague will help you figure out things you might have missed in this checklist. See why it’s so important to preview and test before launching here.
- Check that you have turned your demographic questions on: If you are only using AskToluna, not AskToluna Pro, and you want to track age, gender, and more, click on the “Target Respondents” option to select which demographics you would like to track. These bits of information will appear in your TolunaAnalytics report, and can be very handy when analyzing data.
Finally, add credits (if using AskToluna) and Launch!
Craving more tips? We have comprehensive tutorials here.
Don’t have time to become a survey pro? Enlist ours.