Norbert Schwarz

Introduction

Norbert Schwarz (born March 28, 1953) is a German psychologist whose work supplies the standard account of how context, feeling and question wording shape judgment. His central claim is that people do not hold stable, coherent attitudes that self-report can simply read off; opinions are constructed on the spot, and small features of the immediate situation — the weather, a scale's numeric range, the effort of recalling an example — exert a disproportionate influence on what people say. A bibliometric analysis of 1997–2017 placed him among the 0.1% most-cited scientists across all fields in the Scopus database.

He is best known for the feelings-as-information account, developed with Gerald Clore, which treats moods, emotions and metacognitive experiences such as ease of retrieval as inputs that people consult — and sometimes misattribute — when making judgments. He is equally known for applying that logic to the survey interview, treating a questionnaire as a conversation governed by implicit rules. He holds a Provost Professorship at the University of Southern California in both psychology and marketing, and co-directs the USC Dornsife Mind and Society Center.

Early Life and Education

Schwarz was born in Annweiler am Trifels, Germany. His training was unusually cross-disciplinary by later standards: he completed a doctorate in sociology at the University of Mannheim in 1980 — not a psychology Ph.D., as is sometimes stated — and then a habilitation in psychology at the University of Heidelberg in 1986. The sociology foundation shows throughout his work: the survey, the interview and the measurement instrument have always been objects of study in their own right rather than neutral windows onto the mind.

He taught at Heidelberg from 1981 to 1992 while completing the habilitation and building a research programme on judgment and communication. His early collaborations with Fritz Strack on the role of mood in autobiographical recall date from this period.

Academic and Professional Career

From 1987 to 1992 Schwarz was scientific director of ZUMA (Zentrum für Umfragen, Methoden und Analysen), the German social-science survey infrastructure centre now known as GESIS. The position gave him direct access to large-scale survey operations at the moment when cognitive psychology was beginning to be applied to questionnaire design, and it framed the research agenda he pursued for the next three decades.

In 1993 he moved to the United States and joined the University of Michigan, where he spent twenty years and held four simultaneous appointments: Charles Horton Cooley Collegiate Professor of Psychology in the social psychology programme, Professor of Marketing at the Ross School of Business, Research Professor in the Program in Survey Methodology, and Research Professor at the Institute for Social Research. He was a Fellow at the Center for Advanced Study in the Behavioral Sciences in 2000–01 and 2009–10 and held visiting positions in Europe and Asia, including Würzburg and the Hong Kong University of Science and Technology.

He moved to the University of Southern California in 2013 as Provost Professor in the Department of Psychology and the Marshall School of Business. He is an elected member of the American Academy of Arts and Sciences, the German National Academy of Sciences Leopoldina and the Academia Europaea, and a fellow of the American Psychological Association, the Association for Psychological Science, the Psychonomic Society, the Society for Consumer Psychology, the Society of Experimental Social Psychology and the Society for Personality and Social Psychology.

His awards include the Wilhelm Wundt Medal of the German Society for Psychology (awarded jointly with Strack), the Oswald Külpe Prize, the Wilhelm Wundt–William James Award of the American Psychological Association and the European Federation of Psychologists' Associations, the Donald T. Campbell Award of SPSP, the Distinguished Scientist Award of SESP, the Distinguished Scientific Contribution Award of the Society for Consumer Psychology, the Thomas M. Ostrom Award, Michigan's Rackham Distinguished Graduate Mentor Award, and an honorary doctorate from the University of Basel.

Major Contributions to Psychology

Feelings as information

The feelings-as-information hypothesis holds that when people evaluate a target they ask themselves, often implicitly, how they feel about it, and treat that feeling as evidence. This usually works, because feelings are usually about something relevant. It fails when the source of the feeling is misattributed.

The canonical demonstration, by Schwarz and Clore, is the weather study. People interviewed on a sunny day report higher life satisfaction than people interviewed on a rainy day. But when the interviewer first mentions the weather, the effect disappears: respondents now attribute their mood to the weather rather than to their lives, and the mood stops functioning as information about life satisfaction. The same logic explains why induced moods shift judgments of unrelated targets, and why the shift is cancelled when the true source of the mood is made salient.

Metacognitive experiences and ease of retrieval

Schwarz extended the argument from moods to the felt difficulty of thinking. People asked to recall twelve examples of their own assertive behaviour subsequently rate themselves as less assertive than people asked to recall six — because twelve is hard, and the difficulty is read as evidence that few such examples exist. This reversed the standard interpretation of the availability heuristic: it is not the number of instances retrieved that carries the information but the subjective ease of retrieving them.

The same mechanism was shown to govern judgments of truth, frequency, risk and beauty. Easy-to-process stimuli are judged truer, more frequent, less risky and more beautiful. Because processing ease can be manipulated by figure-ground contrast, font legibility or muscle movement, judgments that ought to depend on content turn out to depend on incidental features of presentation — including findings that hard-to-pronounce names are judged riskier, and that repeated statements feel truer.

The survey interview as conversation

Schwarz's second research programme treats the survey as a conversation governed by cooperative norms of the kind described by Paul Grice: respondents try to be informative, truthful, relevant and clear. This generates testable predictions. Respondents avoid repeating information they have already given to an earlier question, so asking about marital satisfaction before general life satisfaction changes the answer to the second question. Respondents infer the intended meaning of a question from the response alternatives offered, so the numeric range of a scale changes what the labels mean: when a success scale ran from 0 to 10, 34% of respondents endorsed values between 0 and 5; when the formally equivalent scale ran from −5 to +5, only 13% endorsed the same range. Frequency questions about behaviours, questions about "how often" versus "how much," and the framing of time reference periods all shift reported rates by large margins.

Assimilation and contrast

With Herbert Bless he proposed the inclusion/exclusion model, which explains why a contextual prime sometimes pulls judgment toward it and sometimes away. If the prime is included in the mental representation of the target, assimilation results; if it is excluded and used as a comparison standard, contrast results. Whether a celebrity's name is brought to mind as an example of "politicians" or as a standard against which politicians are judged determines the direction of the effect.

Major Publications

  • Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45, 513–523.
  • Schwarz, N., Bless, H., Strack, F., Klumpp, G., Rittenauer-Schatka, H., & Simons, A. (1991). Ease of retrieval as information: Another look at the availability heuristic. Journal of Personality and Social Psychology, 61, 195–202.
  • Schwarz, N., Knäuper, B., Hippler, H.-J., Noelle-Neumann, E., & Clark, L. F. (1991). Rating scales: Numeric values may change the meaning of scale labels. Public Opinion Quarterly, 55, 570–582.
  • Schwarz, N., & Sudman, S. (Eds.). (1994). Autobiographical Memory and the Validity of Retrospective Reports. Springer.
  • Schwarz, N. (1996). Cognition and Communication: Judgmental Biases, Research Methods, and the Logic of Conversation. Erlbaum.
  • Reber, R., & Schwarz, N. (1999). Effects of perceptual fluency on judgments of truth. Consciousness and Cognition, 8, 338–342.
  • Kahneman, D., Diener, E., & Schwarz, N. (Eds.). (1999). Well-Being: The Foundations of Hedonic Psychology. Russell Sage Foundation.

Impact and Legacy

Schwarz changed the practice of survey research more than any other individual in his generation. Cognitive pretesting, the routine use of question-comprehension and response-formulation steps in questionnaire design, and the treatment of question order and scale construction as substantive methodological decisions all derive from work he did or directed. The same ideas entered marketing, where his appointments in business schools reflect the direct commercial relevance of knowing that a scale's numbers or a question's position can move an answer by tens of percentage points, and into behavioural economics, where Daniel Kahneman's constructed-preference arguments rest partly on his findings.

The programme has also attracted sustained criticism, on four main grounds.

First, robustness — and here the record can be stated precisely rather than gestured at. The founding demonstration of feelings-as-information was reported from roughly fourteen respondents per condition. Lucas and Lawless (2013, Journal of Personality and Social Psychology, 104(5), 872–884) matched daily weather to the life-satisfaction reports of more than a million Americans assessed over five years, across a wide range of temperature, precipitation, cloud cover, pressure, wind and humidity variables, and found essentially nothing: the few significant associations were of the order of 0.02 on the satisfaction scale and were detectable only because of the sample size. Feddersen and colleagues (2016), working from panel data with comparable power, found a weather effect roughly one hundredth the size of the original. Yap and colleagues (2017) then ran nine new experiments deliberately modelled on the classic studies but with far larger samples: mood inductions moved life satisfaction significantly, weather did not, and the effect sizes were d = .09 and .11.

The same pattern appears in the survey work. Strack, Martin and Schwarz (1988) reported that asking about dating before asking about life satisfaction produced an item-order effect of d = 1.83 (a figure restated, for example, in Diener, Lucas and Oishi's 2018 review of subjective well-being research); Schimmack and Oishi's meta-analysis (2005) found that the typical item-order effect across later studies was an order of magnitude smaller.

Schwarz and Clore's reply is on record and is not dismissive. They argue that the original study sampled deliberately extreme days — the first sunny days of spring against a return of winter — and that it buried the life-satisfaction item behind dozens of earlier questions, so an instrument using ordinary weather variation and a different question order is not testing the same claim. Simonsohn's counter is that the replication data were in fact less noisy than the original rather than more, and that restricting the analysis to days with large weather swings, which is the direct test of the extreme-days account, still produced no effect. Defenders also point out that context effects are by construction sensitive to features of the testing situation, so a failed replication is ambiguous evidence; critics reply that a theory able to absorb any failure is hard to falsify. The dispute is unresolved, and it is worth noting that the conclusion the large-sample work supports is also Schwarz's own: the biases are demonstrable in the laboratory and small enough in practice that self-report measures are not, on this evidence, invalid.

Second, mechanism. The conversational-inference account is not the only explanation of response effects. Jon Krosnick's satisficing theory derives many of the same phenomena from effort minimisation rather than from cooperative inference, and predicts different patterns across survey modes — notably that satisficing should dominate in self-administered questionnaires where the conversational cue of an interviewer is absent. Which account fits better continues to be debated. Similarly, mood effects on judgment can be modelled by associative network and affect-infusion accounts that require no misattribution step, so the distinctive claim of feelings-as-information — that the source attribution is doing the work — is not always uniquely supported.

A related revision comes from inside the tradition. Huntsinger, Isbell and Clore (2014) argue that affect does not reliably shift people toward either a global or a detail-focused processing style, but confers value on whichever style is already dominant — so a positive mood promotes attention to detail when detail focus is what the situation has activated. On that reading the direction of the effect is contextual rather than fixed, which narrows what the original formulation claimed and moves the theory closer to the associative accounts it was written to replace.

Third, interpretation. Schwarz's own position is that his findings argue for better instruments, not for the conclusion that self-report is worthless; he has argued that attitudes are constructed but not arbitrary, and that context effects are systematic and therefore fixable. That nuance is frequently lost when the findings are cited to dismiss survey data in general.

Fourth, Schwarz has been an outspoken critic of the reform movement in psychology, characterising those who emphasise the replication crisis and propose methodological reforms as conducting a "witch hunt." Supporters of the reforms regard this as untenable given how many of the affected findings come from his own subfield; his own argument is closer to the claim that uniform remedies applied without attention to measurement and context become ritual rather than improvement. The disagreement is unresolved and, given his stature, consequential.

Learn More

References

  • Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal of Personality and Social Psychology, 45(3), 513-523.

Last updated: 2026-10-20

Menu
Loading