|
WebQ - Help |
|
WebQ - Help |
WebQ-sorting differs from other online-questionnaire formats in that it allows to interactively rank-order and sort items on screen. The psychological significance of Q-sorting consists in the evaluation of items relative to each other rather than independently. By this way, it is assumed, a more 'holistic' rendering of the respondent's view can be provided than would be possible by asking and aggregating one bit of information after the other.
This page describes rational and procedure of using WebQ in some detail. For a quick hands-on instruction you may click the 'WebQ Tutor' button above.
Imagine that the IS department of an organization wants to assess its internal customers' attitude to service quality. It asks these customers to Q-sort 4 statements using a forced distribution of the following form:
Most important +1 
0 -1 Least important
In other words, each customer is required to place 1 statement in the +1 category (most important), 2 in the middle or neutral category, and 1 in the -1 category (least important). Initially, the statements are randomly ordered and all placed in the middle pile.
When WebQ is started, the statements are displayed in random order. All statements are assigned to the '0' category ('undecided'), with all sets of radio buttons checked for the center category. The other categories or piles are still empty, and your task will be to sort the appropriate number of statements into each of them.
Symbols are used to indicate the condition of the number of statements in each category. For example:
|
Category provides place ('slots') for 1 item. However, no item is sorted into this category yet. |
 |
There is 1 item sorted into this category, 1 other slot is still empty. |
 |
Category should be filled with 3 items, but there are at least 3 too many. |
|
Still 1 item too much in a category that provides 3 slots. |
|
Exact match, both slots filled with 1 item. |
Notice that the statements are in order from most important to least important and see how the symbols have changed.
Keep sorting and re-sorting until you get all green, and until your view is represented in the statement ordering, then click on the Send button
If there are many statements to sort (and more categories than in the above example), you may want to first divide them into three piles. Use the three central buttons (-1, 0, and +1) to create three piles (e.g., more important, undecided, and less important). Click on the Update button to sort these statements. Turn now to the upper (+1) pile and select the required number of statements for the topmost category (e.g., most important) by clicking the rightmost button (e.g., +4). Then turn to the lower (-1) pile and select the required number of statements for the bottommost category (e.g., least important) by clicking the leftmost button (e.g., -4). Don't forget to click on the Update button to sort statements into place. For the next steps, the basic idea again is to continue selecting the next-most extreme statements from the +1 and -1 piles, i.e., filling the piles from outside to center.
An alternative strategy would consist in using all categories (radio buttons) for 'pre-sorting' in the first place, and then refining the sorting more and more. It's up to you which strategy fits you best, as long as your final q-sort represents the way you think about the issue.
WebQ uses email for sending your data to the researcher. This is the method which is least demanding for setting up an online survey but it does not easily provide the possibility that you could hide your identity from the researcher. On the other hand, WebQ has been designed to secure maximal transparency, which would not be possible with other methods for submitting data. Your own email program is used for sending the message containing the data - you see to whom it is addressed, what is in the email, and you can add your own comments, before sending the message off (see, however, the section on browser versions). If you want to see how that works, click the 'WebQ Tutor' button at the top of this Help page. In this case, instead of the researcher's address, a 'dummy' email address will be used, so that you can freely experiment with it.
When you click on the Send button, the number of statements in each pile is checked, and if there are categories with too many or too few statements, you are asked to continue with sorting. When your Q-sort is OK, you will next be asked for a code word which will be used for processing your data anonymously. If you participate in a project that consists of several parts, please use the same code word every time. Before sending the email with the data away, please don't forget to add your comments to questions that you may find within the email body.
WebQ runs best with version 4.x (or better) of either Netscape Navigator (NN) or MS Internet Explorer (MSIE). The older (3.x) versions of both browsers unfortunately do not support the transparent mode of inserting data into the email message. Also, some respondents may have set up a separate email-client program which the browser may not be able to access, other people may have browser-access only and use a web-based service for email. Therefore the exact copy of the email containing data and the comment-prompt(s) is also displayed in the lower (ranking) frame, to allow for preparing the email via cut&paste.
This page and the WebQ
software were developed by Peter
Schmolck. |
