Pergamon zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Person. indiuid. Off Vol. 21, No. 5, 719-725, 1996 pp. Copyright 0 1996 Else&r Science Ltd Printed in Great Britain. All rights reserved zyxwvutsrqpo S019143869(96)00118-3 0191-8869/96 %15.00+0.00 THE BENEFITS OF OPTIMISM: A META-ANALYTIC REVIEW OF THE LIFE ORIENTATION TEST Gerhard Andersson Department of Clinical Psychology, Uppsala University and Department of Audiology, Uppsala University Hospital, Box 1225, S-751 42 Uppsala, Sweden (Received I6 January 1996) Summary-In 1985 Scheier and Carver set out to begin to explore the possibility that optimism, construed as a stable personality characteristic, has implications for the manner by which people regulate their actions. They developed a scale to measure optimism, the Life Orientation Test (LOT), that has been zyxwvutsrqponmlkjih use d in several studies over the years. In this meta-analysis 56 studies were included adding up to a total number of 98 effect size estimates. Three combined estimates were obtained for measures of coping (+0.27), symptom reporting (-0.23) and negative affect (-0.43). The results from combined estimates of sig- nificance were highly significant as well as homogeneous. High fail-safe n’s were also found. Combined estimates of effect sizes were, however, heterogeneous in all three estimates. Analyses were therefore repeated using the Ways of Coping Checklist for coping (+0.20) and the Beck Depression Inventory for negative affect (- 0.46). The effect size estimate for coping did, however, remain significantly heterogeneous. Although optimism is highly significantly associated with measures of coping, symptom reporting and negative affect, it is concluded that the most reliable association is between optimism and negative affect. Copyright 0 1996 Elsevier Science Ltd. INTRODUCTION On the basis of their theory of behavioural self-regulation, Scheier and Carver (1985) developed the Life Orientation Test (LOT). Their aim was to measure dispositional optimism as a personal construct. More specifically, dispositional optimism was intended to be a global generalized tendency to believe that one will generally experience good vs bad outcomes in life (Scheier & Carver, 1985, 1992). Moreover, it was believed that optimism would serve as a protective factor when facing difficulties in life such as illness. Since the publication of the LOT in 1985, it has been used in several studies in health and personality psychology and has been linked to both psychological and physical well-being (Scheier & Carver, 1992). More recently, however, its psychometric properties have been debated (Chang, D’Zurilla, & Maydeu-Olivares, 1994; Marshall, Wortman, Kusulas, Hervig & Vickers, 1992; Mroc- zek, Spiro III, Aldwin, Ozer & BOSSY, 1993; Smith, Pope, Rhodewalt & Poulton, 1989). In a couple of these studies, an optimism and a pessimism subscale of the LOT have been separated and found to correlate differently with criterion variables. For example, Marshall et zyxwvutsrqponmlkjihgfedcbaZY al. (1992) found LOT pessimism to be associated with neuroticism and negative affect, whereas LOT optimism was associated with extraversion and positive affect. Even the constructors of the LOT found that it consisted of two factors, namely optimism and pessimism, with factor loadings corresponding to the negatively and the positively worded items (Scheier & Carver, 1985). They did, however, consider the LOT to be unidimensional for pragmatic reasons, even though they acknowledged the possibility of examining optimism and pessimism scores separately. More recently, in response to the criticism, the constructors of the scale have suggested a modification of the LOT, but they concluded that the original LOT still is a viable instrument for assessing people’s generalized sense of optimism (Scheier, Carver & Bridges, 1994). Given the continued interest in the LOT and in optimism as a predictor variable of health and coping it would be valuable to summarize the data that has been presented so far on the relationship between optimism as assessed by the LOT and various measures ofcoping and of somatic complaints. Also worth considering is its relationship with measures of negative affect or distress (Watson & Clark, 1984), that may be an important factor in itself, but that also may be viewed as an alternative interpretation of research results found (Smith et al., 1989). One way to summarize large sets of data collected in many studies is by meta-analysis that has 719