Discuss the characteristics of valid and reliable data.

Answers will vary. Validity refers to the degree to which the data measure what you intend to measure. It generally results from careful planning of the questionnaire or interview questions or items. Cautious wording, preliminary testing of items to detect misunderstandings, and some statistical techniques are helpful in determining whether the responses to the items are valid. A pilot test of the instrument is often conducted prior to the full-scale survey so that a smaller number of participants can test the instrument, which can then be revised prior to wide-scale administration.
Reliability refers to the level of consistency or stability over time or over independent samples; that is, reliable data are reasonably accurate or repeatable. Reliability results from asking a large enough sample of people so that the researcher is reasonably confident the results would be the same even if more people were asked to respond or if a different sample were chosen from the same population. For example, if you were to ask 10 people to react to a questionnaire item, the results might vary considerably. If you were to add 90 more people to the sample, the results might tend to reach a point of stability where more responses would not change the results. Reliability would then be reasonably established.