The PANAS in one page: 20 mood words, two largely independent scales, how time instructions change what it measures, published reliability and how to cite it.
Twenty mood words that measure two relatively independent dimensions of affect, positive and negative, with a brief questionnaire that is easy to repeat. The time instruction decides whether you capture state or trait affect. With a Spanish validation.
By Víctor Ciudad-Fernández, PhD · Updated: October 6, 2026 · Leer esta ficha en español · Instruments
The PANAS builds on a central finding of research on the structure of mood: when people rate how they feel on a wide set of mood words, their answers do not line up on a single good-to-bad continuum but on two fairly independent dimensions. Positive affect (PA) reflects how active, enthusiastic and alert a person feels; negative affect (NA) reflects subjective distress and unpleasant engagement, not merely the absence of PA.
That is why a person can score high on both scales at once, or low on both: they are not opposite poles of one variable, and should be interpreted separately. Watson, Clark and Tellegen built the two 10-item scales because many existing mood measures had low reliability or poor convergent and discriminant validity, and they wanted scales that were also brief and easy to administer.
The time instruction changes what you are measuring: “right now” captures state affect, useful as a manipulation check after a mood induction, while “in general” captures trait affect, a more stable affective style used as a personality variable or a covariate. In the original study the general ratings were stable enough over eight weeks for the authors to suggest using them as trait measures.
Respondents rate the extent to which they have felt each of the 20 moods during the specified time frame on a 5-point scale (very slightly or not at all, a little, moderately, quite a bit, very much), scored 1 to 5. Each scale is the sum of its ten words, from 10 to 50. There are no reverse-keyed items.
The original paper tested seven time frames: right now, today, the past few days, the past week, the past few weeks, the past year and in general. Mean scores rise as the time frame lengthens, so compare only scores obtained with the same instruction.
Factor analyses of the 20 words recover two dimensions, with the PA words loading on one and the NA words on the other. In the student samples, the correlation between the scales ranged from −.12 to −.23 across time frames, so they shared only 1% to 5% of their variance.
In a broadly representative UK adult sample (N = 1,003), Crawford and Henry (2004) found the best fit for two correlated factors, allowing correlated errors between words from the same mood subcategories, with measurement invariance across demographic groups. Their conclusion: the PANAS is reliable and valid, but the hypothesis of complete independence between PA and NA must be rejected.
Sandín, Chorot, Lostao, Joiner, Santed and Valiente (1999) validated the Spanish PANAS with 712 undergraduates in Madrid. Exploratory and confirmatory factor analyses showed a robust, stable two-dimensional structure (positive and negative affect), with support for its construct validity, internal consistency and cross-cultural convergence with the original US data.
There is also a version for children and adolescents aged 12 to 17, the PANASN, by the same research group. If you work with Latin American university students, an invariance study across Colombia and Ecuador is a useful reference.
License: Not sold by a commercial publisher: the authors make the manual of the expanded form (PANAS-X), which contains the original 20-item PANAS, openly available through the University of Iowa repository and ask to be contacted with specific questions about reuse.
The original paper describes the scales and their development. The Spanish validation is open access in Psicothema.
Cite the original paper and, if you use it, the Spanish validation:
@article{watson1988panas,
author = {Watson, David and Clark, Lee Anna and Tellegen, Auke},
title = {Development and validation of brief measures of positive and negative affect: The {PANAS} scales},
journal = {Journal of Personality and Social Psychology},
year = {1988},
volume = {54},
number = {6},
pages = {1063--1070},
doi = {10.1037/0022-3514.54.6.1063}
}If your design gives the PANAS twice —to check the stability of trait affect, or because two observers rate the same session— similar means are not enough: you need an agreement index between the two measurements. The ICC calculator computes the intraclass correlation coefficient with its confidence interval, ready for your results section.
No. PA and NA are two separate dimensions, and the instrument is built to be scored and interpreted as two scales. Subtracting one from the other throws away the information the PANAS was designed to provide.
It depends on your question: “right now” for state affect (for example, a manipulation check), “in general” for trait affect. Report the instruction you used, because scores obtained with different time frames are not comparable.
This fact sheet is an academic reference for information purposes: it does not reproduce any questionnaire item. Using the instrument in your research or practice may require citing its authors or asking for permission: see “How to get the instrument”.