رابطه بین خوش بینی جهت مند، بدبینی جهت مند، مقابله سرکوبگر و اضطراب خصلتی

Abstract The purpose of this study was to investigate the relationship between dispositional optimism, dispositional pessimism, repressive coping and trait anxiety. The Marlowe-Crowne scale (MC) and the Bendig version of the Taylor Manifest Anxiety Scale (MAS) were used to select repressor and control groups (N=143, aged between 18 and 47 years), who subsequently completed a measure of dispositional optimism, the Life Orientation Test (LOT), which consists of two sets of items: optimism and pessimism. Repressors (high MC, low MAS) reported significantly lower pessimism than all other participants. When the pessimism scale was divided into high and low categories, there were significantly more repressors in the low pessimism group than the other low trait anxiety group, low anxious (low MC, low MAS). However, for the optimism scale the opposite pattern of results was found, with significantly more low anxious in the high optimism group than repressors. These results support the assertion that the pessimism and optimism scales of the LOT are not the same construct and should be measured separately.

1. Introduction It has been suggested that dispositional optimism, as defined in terms of general expectancies, is beneficial to both psychological and physical well-being (see Scheier, Carver, & Bridges, 1994). Dispositional optimism is usually measured by an eight item questionnaire, the Life Orientation Test (LOT, Scheier & Carver, 1985). The LOT consists of 4 positively worded items and 4 negatively worded items. The evidence linking high LOT scores (i.e. high dispositional optimism) to good psychological health outcomes is strong. For example, high dispositional optimism has been associated with less negative mood in bone marrow transplant patients (Curbow, Somerfield, Baker, Wingard, & Legro, 1993). Similarly, Chamberlain, Petrie, and Azariah (1992) found that both pre- and post-surgery, optimism was positively correlated with life satisfaction and positive well-being, but negatively correlated with psychological distress. Carver and Gaines (1987) reported that, in pregnant women, both before and after birth, dispositional optimism was negatively correlated with depression. In a study by Litt, Tennen, Affleck, and Klock (1992), optimistic women compared to pessimistic women responded better to an unsuccessful in vitro fertilisation procedure. In contrast, the link between dispositional optimism and physical health outcomes is less clear. Scheier et al. (1989) reported that after coronary artery bypass surgery, optimistic men made a more rapid physical recovery than pessimistic men, when compared on physiological measures. Staff rated optimists as progressing faster than pessimists, but this difference was only marginal (p<0.10). In the Chamberlain et al. (1992) study optimism was not related to self-reported health. Two other studies found similar findings, both Robbins, Spence, and Clarke, 1991 and Smith et al., 1989 report that the relationship with self-reported health and optimism became negligible when other personality variables were taken into account. However, Lyons and Chamberlain (1994), in a longitudinal study, reported a significant relationship between optimism and self-reported health, but this relationship was present at time 1 but not at time 2 in their study. In a study by Scheier and Carver (1985), optimism was negatively associated with scores on a physical symptom check list both at time 1 and time 2. In summary, it is not clear whether optimism has an association with physical health. One of the reasons for such inconsistent results may be due to the nature of dispositional optimism measurement. The studies discussed assume that individuals can be divided into optimists and pessimists using the LOT and this is a measure of dispositional optimism. However, there are potential problems with this view of dispositional optimism measurement. Firstly, is the LOT unidimensional/bipolar or do optimism and pessimism scales measure separate personality dimensions? Scheier and Carver (1985) identified two factors on the LOT: the positively-worded and the negatively-worded items. However, they preferred to use the LOT as a unidimensional measure. Later studies have suggested that the two sets of items, the optimism and pessimism scales, do not measure the same constructs. For example, Dember, Martin, Hummer and Melton (1989) found correlations of between −0.52 and −0.57 for the optimism and pessimism in two different samples. They argued that these correlations were not high enough to consider optimism and pessimism as a single construct. Similarly, Marshall, Wortman, Kusulas, Hervig, and Vickers (1992) found correlations of between −0.54 and −0.47 in two samples of young men. Optimism was correlated with extraversion and positive affect and pessimism was correlated with neuroticism and negative affect. Therefore, they posited that the LOT consisted of two separate factors which correlated differently with different measures. In addition, three studies suggest that optimism and pessimism may be independent in older participants. Mroczek, Spiro, Aldwin, Ozer, and Bosse (1993) in a study of older men report correlations of −0.28 between pessimism and optimism. They also found that pessimism and optimism contributed unique variance to three different measures: hassles, psychological symptoms and illness severity. Similarly, Robinson-Whelan, Kim, Maccullam, and Kiecolt-Glaser (1997) in a study of older adults reported that the LOT consisted of two separate optimism and pessimism factors. Plomin et al. (1992) found that optimism and pessimism were uncorrelated in a sample of older men (r=−0.02). Therefore, the evidence strongly suggests that optimism and pessimism are two different constructs. Secondly, it has been argued that the LOT might be indistinguishable from neuroticism (e.g. Smith et al., 1989). However, results from other studies challenge this view and suggest that neuroticism measures do not account for optimism findings Scheier et al., 1994 and Chang, 1998. Thirdly, it has been hypothesised that there may be more than one type of optimist. Wallston (1994) makes a distinction between `cautious optimists' and `cockeyed optimists'. He suggests that cautious optimists are in touch with reality and are fairly certain of favourable outcomes but still do everything in their power to obtain desired outcomes. On the other hand, cockeyed optimists live in a world of illusion and believe that everything will be fine without doing anything to gain their desired outcome. It is the cautious optimists which appear to fit with the Scheier and Carver (1985) notion that optimism will have important behavioural consequences and that people with positive expectancy will work towards their goals. Wallston (1994) suggests that rather than dividing participants into optimists or pessimists by a median split on the LOT, cautious and cockeyed optimists may be identified through a 3 or 4 way split on the LOT. The aim of the current study was to further explore these three issues by comparing low and high trait anxious individuals, especially focusing on individuals who possess a repressive coping style (repressors). The construct of repressive coping is associated with being positive (e.g. Weinberger, 1990 and Myers & Vetere, 1997). More specifically, previous research suggests that repressors may fail to report negative affect rather than reporting more positive affect. For example, repressors compared to nonrepressors recalled fewer negative autobiographical memories, but there was no difference for positive memories (Myers, Brewin, & Power, 1992), they recalled less negative information but not positive material in an intentional recall task (Myers & Brewin, 1995) and an incidental recall task (Myers, Brewin, & Power, 1998). Repressors have been found to answer many self-report instruments in a less negative fashion than nonrepressors. For example, compared to nonrepressors, repressors rated themselves as significantly less likely than the average person to experience negative events and rated negative words as significantly less descriptive of self than others. However, there were no differences for positive events and for positive words (Myers & Brewin, 1996). On a positively-phrased measure of secure adult attachment type, repressors rated themselves as no more securely attached than nonrepressors (Myers & Vetere, 1996). Repressors have been operationally defined by their high defensiveness scores, usually measured with the Marlowe-Crowne Social Desirability Scale (MC; Crowne & Marlowe, 1964) and low scores on a self-report measure of trait anxiety (e.g. Weinberger, Schwartz, & Davidson, 1979). Typically, they are compared with 3 control groups, another low anxiety group, low anxious (low anxiety-low defensiveness), and two high anxiety groups: high anxious (high anxiety-low defensiveness) and defensive high anxious (high anxiety-high defensiveness) individuals. The current study compared repressors with the three control groups on their responses to the LOT. Due to the propensity of repressors to fail to report negative affect rather than reporting more positive affect, it was hypothesised that if the LOT is not unidimensional and in fact measures two different personality dimensions, repressors would score higher on the pessimism scale of the LOT than the three control groups, i.e. they would be less pessimistic, but would not score higher than the other low trait anxious group, the low anxious on the optimism scale of the LOT, i.e. they would not be more optimistic. The current study used the Bendig short form of the Taylor Manifest Anxiety Scale (MAS; Bendig, 1956) as a measure of trait anxiety. This has been considered to be a valid measure of Neuroticism (Watson & Clark, 1984). It was hypothesised that if the LOT is just a measure of neuroticism, there will be no group differences between the two low anxiety groups; repressors and low anxious. However, if the LOT is more than a measure of neuroticism there will be group differences between repressors and low anxious. To test the Wallston (1994) cockeyed/cautious theory, the LOT was divided initially by median splits into high and low, and secondly, using quartile splits, to see whether results would be clearer using quartile splits rather than median splits. As previous research indicates that repressors are overly optimistic for negative future events Myers & Brewin, 1996 and Myers & Reynolds, 1998 it was hypothesised that repressors may fall into the Wallston (1994) category of cockeyed optimists, especially for the pessimism scale.

3. Results Firstly, to examine the relationship between the two sets of items, the pessimism and optimism scales were correlated. The correlations between the pessimism scale and optimism scale was r=0.50 (p<0.001). Secondly, it was investigated whether trait anxiety was related to the different scales of the LOT. Because the trait anxiety data were not continuous, rather than using correlations, participants were divided into low anxiety (R and LA groups) and high anxiety (HA and DHA groups) and independent sample t-tests were performed to establish whether there was an effect for trait anxiety. Results indicated that the low anxiety group scored higher on all three scales than the high anxiety group (bipolar optimism, low anxiety mean=23.05 (S.D. 6.22), high anxiety mean=15.05 (S.D. 6.44), t(141)=7.53, p<0.001; pessimism, low anxiety mean=12.37 (S.D. 3.47), high anxiety mean=7.81 (S.D. 2.43), t(141)=7.84, p<0.001; optimism, low anxiety mean=10.97 (S.D. 3.84), high anxiety mean=7.23 (S.D. 3.66), t(141)=5.90, p<0.001). Therefore, trait anxiety was related to high bipolar optimism, low pessimism and high optimism. For all further analyses, participants were divided into the four groups, R, LA, HA and DHA. One way analysis of variance (ANOVA) indicated that there were no gender differences for bipolar optimism, F(1, 142)=024, ns, pessimism F(1, 142)=0.62, ns and optimism, F(1, 142)=0.04, ns. Therefore, gender was disregarded in further analyses. Group differences in the means were tested for bipolar optimism, pessimism and optimism using ANOVA. Newman–Keuls tests were used for all post hoc comparisons, with significance levels set at p<0.05. Means can be found in Table 2. Table 2. Mean scores and standard deviations for bipolar optimism, pessimism and optimism for the repressor (R), low anxious (LA), high anxious (HA) and defensive high anxious (HA) groups. Different subscripts indicate group differences (p<0.05) R LA HA DHA F(3, 139) Bipolar optimism 23.78acd (6.19) 21.71abc (6.15) 14.53bd (5.88) 15.81bd (7.25) 19.84*** Pessimism 13.45a (2.76) 10.39bc (3.80) 7.74bd (3.43) 7.92bd (3.50) 27.66*** Optimism 10.55a (3.82) 11.75abc (3.84) 7.05bd (3.60) 7.50bd (3.81) 12.33*** ∗∗∗ p<0.001 Table options There was a significant main effect for bipolar optimism, with repressors scoring significantly higher than the two high anxiety groups, HA and DHA, but not the other low anxiety group (LA). In addition, the LA group scored significantly higher than the HA and DHA groups. A similar pattern was found for optimism with repressors scoring significantly higher than the two high anxiety groups, HA and DHA, but not LA. In addition, the LA group scored significantly higher than the HA and DHA groups. However, there was a different pattern for pessimism, with repressors rating themselves as significantly less pessimistic than all the groups, LA, HA and DHA. In addition, the LA group rated themselves as significantly less pessimistic than the HA and DHA groups. In further analyses, bipolar optimism, pessimism and optimism were divided into categorical variables, firstly by median splits and then by quartile splits. To test differences between the two low anxiety groups, LA and R, these two groups were compared using chi square statistic. 3.1. Median splits See Fig. 1 for percentage high scores for bipolar optimism, optimism and pessimism. Image for figure Fig. 1 Fig. 1. Figure options 3.1.1. Bipolar optimism For the repressor group, 13 were classified as low bipolar optimism and 38 were classified as high bipolar optimism (74.5%). For LA, 10 were classified as low bipolar optimism and 18 were classified as high bipolar optimism (64%). For HA, 30 were classified as low bipolar optimism and 8 were classified as high bipolar optimism (21.1%). For DHA, 19 were classified as low bipolar optimism and 7 were classified as high bipolar optimism (25.0%). 3.1.2. Pessimism For the repressor group, 4 were classified as low (high pessimism) and 47 were classified as high (low pessimism) (92.2%). For LA, 11 were classified as low (high pessimism) and 17 were classified as high (low pessimism) (60.8%). For HA, 29 were classified as low (high pessimism) and 9 were classified as high (low pessimism) (23.7%). For DHA, 16 were classified as low (high pessimism) and 10 were classified as high (low pessimism) (38.4%). 3.1.3. Optimism For the repressor group, 29 were classified as low optimism and 22 were classified as high optimism (43.1%). For the LA group, 6 were classified as low optimism and 22 were classified as high optimism (78.6%). For HA, 28 were classified as low optimism and 10 were classified as high optimism (26.3%). For DHA, 19 were classified as low optimism and 7 were classified as high optimism (25.0%). 3.1.4. Comparison of R and LA Chi square statistic indicated that for bipolar optimism there was no group differences (χ2=0.91, df=1, ns). However, for pessimism there were significantly more R than LA in the low pessimism group (χ2=11.62, df=1, p<0.001). There was the opposite pattern for optimism, with significantly more LA in the optimistic group than R (χ2=9.20, df=1. p<0.01). 3.2. Quartile splits The four cells were labelled low, medium, high or very high. 3.2.1. Bipolar optimism For the repressor group, 3 (5.9%) were classified as low optimism, 8 (15.7%) were classified as medium optimism, 18 (35.3%) were classified as high optimism and 22 (43.1%) were classified as very high optimism. For the LA group, 4 (14.3%) were classified as low optimism, 6 (21.4%) were classified as medium optimism, 8 (28.6%) were classified as high optimism and 10 (35.7%) were classified as very high optimism. For HA, 18 (47.4%) were classified as low optimism, 12 (31.6%) were classified as medium optimism, 7 (18.4%) were classified as high optimism and 1 (2.6%) was classified as very high optimism. For DHA, 11 (42.3%) were classified as low optimism, 8 (30.8%) were classified as medium optimism, 5 (19.2%) were classified as high optimism and 2 (7.7%) were classified as very high optimism (see Fig. 2). Image for figure Fig. 2 Fig. 2. Figure options 3.2.2. Pessimism For the repressor group, 1 (2.0%) was classified as low (very high pessimism), 4 (7.8%) were classified as medium (high pessimism), 13 (25.5%) were classified as high (medium pessimism) and 33 (64.7%) were classified as very high (low pessimism). For LA, 7 (25.0%) were classified as low (very high pessimism), 6 (21.4%) were classified as medium (high pessimism), 5 (17.9%) were classified as high (medium pessimism) and 10 (35.7%) were classified as very high (low pessimism). For HA, 16 (42.1%) were classified as low (very high pessimism) 14 (36.8%) were classified as medium (high pessimism), 5 (13.2%) were classified as high (medium pessimism) and 3 (7.9%) were classified as very high (low pessimism). For DHA, 11 (42.3%) were classified as low (very high pessimism), 7 (26.9%) were classified as medium (high pessimism), 6 (23.1%) were classified as high (medium pessimism) and 2 (7.7%) were classified as very high (low pessimism) (see Fig. 3). Image for figure Fig. 3 Fig. 3. Figure options 3.2.3. Optimism For the repressor group, 6 (11.8%) were classified as low optimism, 23 (45.1%) were classified as medium optimism, 7 (13.7%) were classified as high optimism and 15 (29.4%) were classified as very high optimism. For LA, 6 (21 4%) were classified as low optimism, zero were classified as medium optimism, 10 (35.7%) were classified as high optimism and 12 (42.9%) were classified as very high optimism. For HA, 18 (47.4%) were classified as low optimism, 10 (26.3%) were classified as medium optimism, 10 (26.3%) were classified as high optimism and zero were classified as very high optimism. For DHA, 11 (42.3%) were classified as low optimism, 8 (30.8%) were classified as medium optimism, 6 (23.1%) were classified as high optimism and 1 (3.8%) was classified as very high optimism (see Fig. 4). Image for figure Fig. 4 Fig. 4. Figure options 3.2.4. Comparison of R and LA Chi square statistic indicated that for overall bipolar optimism there were no group differences (χ2=2.27, df=3, ns). For pessimism there was a significant group difference (χ2=15.36, df=3, p<0.01) and for optimism, there was a significant group difference (χ2=18.76, df=3, p<0.001.)