We suggest that to capitalize on future methodological advances in text analysis, researchers should supplement closed-ended questions with more open-ended questions in high-quality, representative, longitudinal studies.

We assess the value of an open-ended question about well-being (“What do you do to make life go well?”) in three samples of the Midlife in the United States study. We show that closed- and open-ended measures of well-being similarly predict objective health indicators (e.g., mortality). But while closed-ended measures of well-being are strongly correlated with hundreds of other closed-ended self-ratings, the open-ended measure is distinct.