Department of Physical Medicine and Rehabilitation, Ohio State University, Columbus (Dr Corrigan) and Santa Clara Valley Medical Center, Rehabilitation Research Center, San Jose, California (Drs Kolakowsky-Hayner and Wright, Ms Bellon, and Mr Carufel).
Checking for direct PDF access through Ovid
THE SATISFACTION WITH LIFE SCALE (SWLS) is a measure of life satisfaction developed by Diener et al.1,2 Life satisfaction is one factor in the more general construct of subjective well-being, which has at least 3 components: positive affective appraisal, negative affective appraisal, and life satisfaction.3 Life satisfaction is distinguished from affective appraisal in that it is more cognitively than emotionally driven.SCALE STRUCTUREThe SWLS consists of 5-items that require a rating on a 7-point Likert scale. Administration is rarely more than a minute or 2 and can be completed by interview (including phone) or paper and pencil response. The instrument should not be completed by a proxy answering for the person. Items of the SWLS are summed to create a total score that can range from 5 to 35.The SWLS is in the public domain. Permission is not needed to use it. Further information regarding the use and interpretation of the SWLS can be found at the author's Web site http://internal.psychology.illinois.edu/∼ediener/SWLS.html. The Web site also includes links to translations of the scale into 27 languages.PSYCHOMETRIC PROPERTIESResearch on the SWLS has shown that current mood tends to have a small effect on life satisfaction4 but personality traits (ie, extraversion, neuroticism, agreeableness, and conscientiousness5) have a modest “top down” effect. A large twin study from the Netherlands found 38% of variance in the SWLS was attributable to heritability, broadly defined, including the shared personality characteristics of twins.6 The SWLS does not measure satisfaction with specific domains of life (eg, family, employment, and income); however, one's overall satisfaction is significantly associated with those specific domains that a person considers important.5RELIABILITYA meta-analysis of 60 studies that assessed the reliability of the SWLS found a mean Cronbach alpha of 0.78 with 95% confidence intervals ranging from 0.766 to 0.807.7 In the original validation study, Diener et al found a 2-month test-retest correlation coefficient of 0.82. Studies since have reported 0.808 and 0.849 for 1-month intervals; 0.54 for a 4-year interval10; and 0.51 for 5-year averages with a 7-year interval between.11 Lower test-retest reliability as time passes is consistent with expectations for variability in life circumstances and thus life satisfaction.In the initial validation study, Diener et al concluded that a single factor best represented the results and multiple subsequent studies have replicated this finding.12–15 The first item is consistently the most associated with the total score and latent factor, while the fifth item is frequently the least associated. Some studies have concluded that a 2-factor solution best fits the data, primarily due to item 5 being less correlated.16–20 The fifth item inquires about past versus current circumstances. Pavot and Diener3 maintain that all 5 items can be used confidently, but if the investigator is strictly interested in current satisfaction, item 5 can be dropped.ValidityThe original validation studies correlated the SWLS with 10 other measures of subjective well-being. Most measures correlated at r = 0.50 or higher for each of the 2 samples from the original work. Numerous subsequent studies have found comparable or higher correlations with other populations when interviewer ratings, informant reports, or other objective measures are used. See Pavot and Diener3 for a more detailed summary of these studies.Evidence for the construct validity of the SWLS can be drawn from a plethora of investigations—again, see Pavot and Diener3 for a summary of these studies.