The Panels Tab contains setting used for proper rotation of previous test-takers (panels). You can select whether or not to set current test-takers for a callback (to take another test) in the future, and if so, specify how many days between tests you desire, the panel composition of each test cycle (set in percentage), and the maximum number of times each respondent may take the test.
Turn Panels On/Off
Turn Panels On/Off is a toggle. When it is checked, all people who complete the test and say it's OK that you call them again in the future will be tagged with a recall date (based on the number of days you have entered in the next field, Call Back Days).
If you don't wish to build panels from those that you're recruiting this week, simply make sure the Turn Panels On is not checked.
Panel building is not retroactive.
That is to say, if you have panels off (Turn Panels On is not checked), and then you decide to begin building panels today, only respondents that take the test beginning today will be assigned a schedule callback record (added to panels) for the future. The only way you can bring back previous test-takers who are not paneled is through the use of Reactive Respondents.
Call Back Days
Call Back Days is the number of days you wish to elapse before someone completing the test today is called back to take another test. Generally, we recommend when you're just starting to use ComQuest that this be set to 30 or 40 days. Then, as the months go by, and you have begun to build some pretty good sized panels, you can begin to spread out the interval between tests, to 60, 90 or even 120 days.
Percent From Panels
Percent From Panels allows you to control how much of your weekly sample is comprised of previous test-takers (the balance of the sample would presumably be comprised of new, first-time test-takers). This value, while set here as a percentage, is actually converted to a real number, based on the Maximum Target setting you have in Set Recruitment Quotas.
For example, if you set Percent From Panels to 40%, and you have 120 as your Maximum Target, that means as soon as you get 48 previous test-takers in your current research cycle (40% of 120), no more panels will be allowed. Before each respondent is selected by ComQuest, and evaluated as to whether they should be sent to an interviewer's Workstation, one of the criteria that is reviewed is their past test history. If they have previously taken a test (whether they finished or not), and you have already reached your Percent From Panels cap for the week, they will not be sent to an interviewer's Workstation. Instead, their schedule record is moved 5 days from today, which in effect will move them into the next week for consideration (in hopes they will come up earlier in the next test cycle, when you still need panels. That's why they aren't moved 7 days from today).
If you set your Maximum Target unrealistically high, such as to 500 or 999, the Percent From Panels logic will not work.
If you set your Maximum Target to 999, for example, and you have Percent From Panels set to 40%, that means ComQuest will allow approximately 400 previous test-takers into this week's sample. Of course, you'll probably never get 400 completes in a week, which means, when you end the week with 120 completes, for example, all 120 of those (or a good many, at least) will be panels. ComQuest worked just as you asked it to! It never allowed more than 400 panels into your sample, just as you had requested! The lesson here is, set your Maximum Target to a realistic number, such as 100 or 120 or 75, etc.
You can set a limit of how many total times a respondent is allowed to take your music test. This value can be anything between 0 and 99. 0 (zero) is actually Off; meaning there is no real limit. We recommend setting the values on this screen in a manner that will provide the most efficiency from your panels, without unjustly using the same respondents over and over in short intervals.
Again, this is all a tradeoff. If you spin panels at 10 days, allow 100% of your sample to be comprised of panels, and allow each person to take the test 50 times, you will no doubt hit your quotas each week. However, you also have to ask yourself what the value of getting information from the same small group of people each week is.
On the other hand, if you take a purist's approach; only allow people to take the test twice, and set Panels to 5% with a 180 day interval, it's unlikely you'll ever hit your quotas. As with many things in life, the best option is somewhere in the middle. Always work to recruit new, first-time test-takers, but also allow a respectable quantity of previous test-takers, who have been adequately rested, back into the sample.
You can skip the "May we call you again" question at the end of Song Survey.
In this case, all respondents completing the test are automatically scheduled for a call back as a panelist each time they complete the test, without giving them the choice to opt-out. If you wish to enable this feature, please contact ComQuest Technical Support.