Adobe Connect User Community








Bookmark and Share

Home >  Learning Center >  Tutorials >  10 Tips for Writing and Delivering Effective Poll and Survey Questions

10 Tips for Writing and Delivering Effective Poll and Survey Questions

Randah McKinnie, Sr. Product Manager eLearning Solutions, Adobe Acrobat Connect Pro

February 2009

Expertise Level: Beginner

  
0 Votes

Polls and surveys are an easy way to learn more about your audience and their experience level, check whether they can recall important information youve presented, or wish to gather feedback on the efficacy of your sessions. Adobe Acrobat Connect Pro includes a Polling Pod as an easy-to-use live audience survey tool. To create more advanced surveys, evaluations, and examinations, you may also wish to use the Quiz Manager in Adobe Presenter. As you formulate standard poll and survey questions, refer to the guidelines below to build effective polls and surveys.

As always, start with your goal in mind. If you plan to ask a lot of questions, before you start writing them, remember that you are asking your audience to do some work. Request of the attendees as little as possible, so that your response rate stays high. Be clear about the goals you have for gathering the information qualitative or quantitative and what you will do with this information once you have it. The answers you obtain are only as reliable as the clarity and simplicity of the questions you pose.

Now you are ready to start writing your questions. Below is a list of six general survey-writing suggestions, and four more that are specific to using the Poll pod in Connect Pro Meetings.

  1. Make each question as simple and direct as possible.
    Ask yourself whether a typical group of people in your intended audience would understand the question. Would they all interpret the question to mean the same thing, or are there multiple possible interpretations? Keep it is simple and clear as you can, to ensure your question is easily understood without the need to read it over multiple times.

  2. Avoid nuances and colloquialisms.
    Whenever possible, select words that have a single meaning. Be specific and try not to attribute your own background knowledge to your audience. For example, to determine the level of experience of the audience at using a software package, rather than ask if it is used most of the time, frequently, infrequently or never, ask whether it is used hourly, daily, weekly, or monthly. If even more specific information would be useful such as the number of hours per day of active use, then the question should clearly ask that question.

  3. Offer a reasonable and consistent number of rankings for Likert-style questions.
    A maximum of 5 rankings should suffice for a rating type question. If a list of items needs to be ranked, and requires a specific per-item ranking, a user may opt out of the question or the entire survey at this point because this is just too much work. If the list is longer than 5 or 6 items, consider splitting a ranked list into separate questions.

    Look at your entire set of rating or ranking questions, and if possible, use consistent language across all of the questions. Could your goal be accomplished using only level of agreement questions, or all level of satisfaction questions, or do you need to have both? If so, try to group like questions together, so there is less work for respondents to pay attention to rating scale changes every time they approach a new question.

  4. Ask only one question at a time.
    Questions that ask a combination of elements at once may leave the respondent torn, as she may feel strongly in favor of one part of the question, and not in favor of another part. For example I find online meetings are easy to run and a great way to build my remote teams. They may be easy to run, but not at all of assistance with team building if not run appropriately for that purpose. The question should be split, or just the piece you need the answer to should be asked.

  5. Allow neutral or opt out responses.
    People may take their responses to your survey questions very seriously, and will not want to answer if the question does not apply or align with their views. Dont force them to pick an answer, as they may opt out of completing the entire survey. Provide options such as neutral, or does not apply so the survey taker will continue to the end.

  6. User-test your survey.
    Run the survey questions past a sample set of people who are representative of your audience, and ask each person what the question means to him. If anything is unclear or confusing, make the necessary changes and try again.

  7. Connect Pro: Use the Poll pod as a way to spark conversation.
    In the Connect Pro poll pod you have a simple checkbox option to broadcast the results of the poll live to respondents. Use the feedback received as a way to enrich a conversation, or encourage participants to chime in. For example, in a team meeting where youve asked what percentage of the quarterly goals have been met, you may share the results with the room, then ask those who said theyd met 100% of their goals what was working for them this quarter that might help others. (Note that in a large, marketing/webinar style presentation this level of open conversation is not a recommended approach.)

  8. Connect Pro: Use multiple-answer when it will garner useful information
    There are two Connect Pro poll formats: multiple-choice and multiple-answer. Use multiple-choice when you wish to narrow down the selection to only one possible answer, such as a question to indicate your company size in the majority of cases this has a single, specific answer for each participant.



    Use multiple-answer when you are trying to gather varied and related information, such a request to uncover one or more forms of transit attendees regularly use at least one day per week to commute to work.



  9. Connect Pro: Opt not to divulge the raw number of respondents
    In Connect Pro you have the option to display the raw numbers of people who have responded (this is the default setting), or to display the percentage of respondents who have selected certain options. Use the percentage setting for presentations when you dont necessarily wish to give away a sense of the true number of attendees in your session. Such a case may exist during a public marketing seminar, particularly if you have poor attendance and dont want to advertise that information.



  10. Connect Pro: Consider using Polls alongside graphics or other presented information.
    Adobe Presenter is a great tool to use if you wish to display rich information right beside the questions you are asking in your poll or survey. However, you may have just one poll you want to post which tests an opinion or other knowledge about a slide or graphic and so may not wish to make a presentation in this case. Using the flexibility of the layouts and pods in Connect Pro, simply organize your room so that participants can clearly see both the graphic in question and the poll at the same time, and so it is clear what you are asking. You may set this up ahead of time on its own layout so you have it ready when presenting, or may think of a question in the middle of your presentation and pull up a poll pod to ask for feedback on the spur of the moment. Take advantage of this flexibility to increase the clarity of the information exchanged with your audience.