Open-ended questions should be used sparingly.
The rule of thumb in designing questions is to only ask what is important to know rather than what you would like to know ‘while you are there’. This applies equally to open-ended and forced choice. With this extra proviso for open-ended: can this question be better handled as a forced-choice?
- Don't opt for open-ended unless forced choice cannot elicit the free range of answers you require.
- Don't opt for open-ended for the sake of providing 'variety' in your survey.
- Don't opt for open-ended as a convenient way out to elicit choices you have not thought out properly yourself.
Processing open-ended answers will require allocation of human resources to read and sort answers into meaningful groupings.
Having made the decision...
You can use the E-valuate-IT coding tool to assist. You are able to create category names by which to classify open-ended responses. You can annotate each category with comments to guide others in applying and interpreting the categories consistently. You can inspect and print all open-ended responses before and after coding. When you assign codes to each answer, you can report aggregated results and grouped answers.