A group of do-gooder college students believes that herbal tea has remarkable powers. To test this belief, they make weekly visits to a local nursing home, visiting with the residents and serving them herb tea. The nursing home staff reports that after several months many of the residents are more cheerful and healthy. A dour professor scoffs at the idea that herb tea helped the residents. It's all confounding, says the professor.
(a) Identify the treatment and outcome/response variables in this study.
(b) What sort of confounding might the professor have in mind?
A scholar is interested in analyzing the effect of spirituality and spiritual practice on personal attitudes about health. She sends out a questionnaire asking individuals to report (i) whether they have a chronic illness (possible answers are "yes" or "no"), (ii) the importance of spirituality to themselves (possible answers are "not important" or "important or very important"), (iii) their attitude about their own health (possible answers are "positive or very positive" or "not positive").
(a) State the treatment the scholar is interested in studying, and state the outcome on which she wishes to ascertain its effect.
(b) Is this an experiment?
(c) Suppose the scholar finds a negative relationship between the treatment and the outcome you identified in the first sub-part. Can you explain this relationship purely as a selection effect, i.e. assuming that the treatment effect is 0?
(d) Suppose the scholar finds a positive relationship between the treatment and the outcome you identified in the first sub-part. Can you explain this relationship purely as a selection effect, i.e. assuming that the treatment effect is 0?
(e) Suppose instead of using the questionnaire described above, the scholar invites some randomly chosen individuals to spiritual counseling sessions provided at her expense. Then, after all counseling sessions are finished, she asks all individuals in the study (those invited to counseling and those not invited) to report their attitudes about their own health. Is this an experiment, and can you critique the possible relationships she might find on grounds of selection bias?