Abstract: Personal meaning in subjective experience is a key element in the treatment of persons with mental disorders. Open-response speech samples would appear to be suitable for studying this type of subjective experience, but there are still important challenges in using language as data. Scientific principles involved in sample size calculation, validity, and reliability may be applicable, by analogy, to data collected in the form of words. We describe a rationale for including computer-assisted techniques as one step of a qualitative analysis procedure that includes manual reading. Clarification of a framework for including language as data in psychiatric research may allow us to more effectively bridge biological and psychometric research with clinical practice, a setting where the patient’s clinical “data” are, in large part, conveyed in words.