Symptom-positive and symptom-negative items in the state-trait anxiety inventory: A comparison and replication

Substantially higher mean scores on symptom-negatively versus symptom-positively worded items have consistently been reported in the literature for the balanced Stait-Trait Anxiety Inventory. In this study we aimed at replicating and comparing these findings for the Dutch adaptation of the inventory. Results indicated significantly higher mean subscale scores for symptom-negative as opposed to symptompositive items of both the Trait and State version, across sexes and age groups as well as across different levels of distress in students, policemen and psychiatric out-patients. Positive-negative mean difference scores were consistently larger for the State than for the Trait version. State and Trait mean difference scores were found to be mainly confined to symptom-positive subscales. Scale intercorrelations were lowest between symptom-positive and symptom-negative subscales both within and across measures. Factor analyzing the combined versions identified separate symptom-negative and symptom-positive factors, tentatively labeled “absence of positive affect” and “presence of negative affect”. Several explanations of the findings among which item-intensity specificity, the response style of social desirability and the model of positive and negative affectivity are discussed.     

人格测验中题目正反向陈述的效应

研究中将EPQ和NEO-FFI中的一半题目进行语义反转（肯定陈述的改为否定陈述,否定陈述的改为肯定陈述）,而另一半的题目陈述方向不变,形成的量表用于初测。重测时则把初测问卷中的题目全部进行语义反转。测试后得到363名大学生的正反EPQ和412名大学生的正反NEO-FFI数据。对比发现两套量表在使用正向题时信度略高,而使用反向陈述题目时效度略好;验证性因素分析发现正反向题能测量同一特质。正向题可能更易受反应偏差影响     

项目表述效应对人格测验信度和效标效度的影响——以核心自我评价量表(CSES)为例

项目表述效应是指由项目表述方式的差异引起的与测量内容无关的系统变异,项目表述效应模型的统计本质是一种双因子模型。本研究以核心自我评价量表(CSES)为例,探讨项目表述效应对人格测验信效度的影响。采用核心自我评价量表、生活满意度量表和积极情感消极情感量表对340名“蚁族”进行测查。结果表明,CSES在核心自我评价特质以外,还存在一个反向题项目表述效应因子;忽视项目表述效应对CSES的同质性信度和效标关联效度有重要影响：高估CSES的同质性信度,低估核心自我评价与生活满意度、积极情感的正相关,高估核心自我评价与消极情感的负相关。     

Simultaneous administration of the Rosenberg Self-Esteem Scale in 53 nations: exploring the universal and culture-specific features of global self-esteem

The Rosenberg Self-Esteem Scale (RSES) was translated into 28 languages and administered to 16,998 participants across 53 nations. The RSES factor structure was largely invariant across nations. RSES scores correlated with neuroticism, extraversion, and romantic attachment styles within nearly all nations, providing additional support for cross-cultural equivalence of the RSES. All nations scored above the theoretical midpoint of the RSES, indicating generally positive self-evaluation may be culturally universal. Individual differences in self-esteem were variable across cultures, with a neutral response bias prevalent in more collectivist cultures. Self-competence and self-liking subscales of the RSES varied with cultural individualism. Although positively and negatively worded items of the RSES were correlated within cultures and were uniformly related to external personality variables, differences between aggregates of positive and negative items were smaller in developed nations. Because negatively worded items were interpreted differently across nations, direct cross-cultural comparisons using the RSES may have limited value.     

Method effects: the problem with negatively versus positively keyed items

Using confirmatory factor analyses, we examined method effects on Rosenberg's Self-Esteem Scale (RSES; Rosenberg, 1965) in a sample of older European adults. Nine hundred forty nine community-dwelling adults 60 years of age or older from 5 European countries completed the RSES as well as measures of depression and life satisfaction. The 2 models that had an acceptable fit with the data included method effects. The method effects were associated with both positively and negatively worded items. Method effects models were invariant across gender and age, but not across countries. Both depression and life satisfaction predicted method effects. Individuals with higher depression scores and lower life satisfaction scores were more likely to endorse negatively phrased items.     

Comparing the predictive utility of trait affect and average daily affect for the prospective prediction of health outcomes

In preregistered secondary data analyses, we compared the predictive utility of trait affect and average daily affect for predicting three health outcomes across nine years (N = 1,376). Trait positive and negative affect were assessed using a 25-item dispositional questionnaire. Average daily affect was assessed as the mean of eight daily diary reports of the same items. Trait affect and average daily affect both had medium associations with self-reported general health and chronic health conditions. Moreover, both types of affect predicted mortality when adjusting for baseline health and demographics. Effect sizes were comparable for trait compared to daily affect. These findings demonstrate convergent predictive validity of trait and daily affect measures.     

项目表述效应的统计控制：以中文版生活定向测验为例

本研究以中文版生活定向测验（LOT-R）为例,使用CTCM和CTCU方法进行建模,旨在探讨中文表述的量表中是否存在项目表述效应以及影响项目表述效应的特质因素。采用中文版生活定向测验等量表组成的问卷对334名大学生进行测试。结果表明,中文版LOT-R存在项目表述效应,是一种反向题的项目表述效应,也可称为语言标签效应;在分离了反向题项目表述效应后,发现中文版LOT-R只存在乐观因子,而不再有悲观因子。这表明乐观-悲观同属于一种人格特质,而非两种不同的人格特质;反向题项目表述效应是一种结构性误差,会导致量表结构的扭曲。研究还发现担心错误和父母期望与反向题项目表述效应存在显著的负相关,表明越是担心错误或父母期望水平越高的被试,他们对反向题的项目表述效应（语言标签效应）越小。社会赞许性对反向题项目表述效应没有影响。     

Method Effects: The Problem With Negatively Versus Positively Keyed Items

Using confirmatory factor analyses, we examined method effects on Rosenberg's Self-Esteem Scale (RSES; Rosenberg, 1965) in a sample of older European adults. Nine hundred forty nine community-dwelling adults 60 years of age or older from 5 European countries completed the RSES as well as measures of depression and life satisfaction. The 2 models that had an acceptable fit with the data included method effects. The method effects were associated with both positively and negatively worded items. Method effects models were invariant across gender and age, but not across countries. Both depression and life satisfaction predicted method effects. Individuals with higher depression scores and lower life satisfaction scores were more likely to endorse negatively phrased items.     

Assessing clergy work-related psychological health: reliability and validity of the Francis Burnout Inventory

ABSTRACT

Drawing on the classic model of balanced affect, the Francis Burnout Inventory (FBI) conceptualised good work-related psychological health among clergy in terms of negative affect being balanced by positive affect. In the FBI negative affect is assessed by the Scale of Emotional Exhaustion in Ministry (SEEM) and positive affect is assessed by the Satisfaction in Ministry Scale (SIMS). In support of the idea of balanced affect, previous work had shown a significant interaction between the effects of SEEM and SIMS scores, showing that the mitigating effects of positive affect on burnout increased with increasing levels of negative affect. In this paper a convenience sample of 155 priests serving with the Roman Catholic Church in Italy have been assessed on the Purpose in Life Scale (PILS) as an independent measure of well-being and concurrently on the two scales, SEEM and SIMS. Crucially for confirming the idea of balanced affect, there was a significant interaction between the effects of SEEM and SIMS scores on scores recorded on the PILS, confirming that the mitigating effects of satisfaction in ministry on purpose increased with increasing levels of emotional exhaustion.     

Work-Related Psychological Health Among Clergy Serving in The Presbyterian Church (USA): Testing the Idea of Balanced Affect

Drawing on the classic model of balanced affect, the Francis Burnout Inventory (FBI) conceptualises good work-related psychological health among clergy in terms of negative affect being balanced by positive affect. In a random sample of 744 clergy (539 clergymen and 205 clergywomen) serving in The Presbyterian Church (USA), negative affect was assessed by the Scale of Emotional Exhaustion in Ministry (SEEM) and positive affect was assessed by the Satisfaction in Ministry Scale (SIMS). At the same time, burnout was independently assessed using self-report measures of overall health and burnout, and by the extraversion and neuroticism scales of Eysenck’s dimensional model of personality. These independent measures of burnout indicated higher burnout among those who were emotionally exhausted and lower burnout among those who had high levels of satisfaction with their ministry. Crucially for proving the idea of balanced affect, there was a significant interaction between the effects of SEEM and SIMS scores on these independent measures of burnout, showing that the mitigating effects of positive affect on burnout increased with increasing levels of negative affect.     

Civilian PTSD scales: relationships with trait characteristics and everyday distress

Strong associations between civilian posttraumatic stress disorder (PTSD) scales and measures of general psychological distress suggest that the scales are nonspecific to PTSD. Three common PTSD scales were administered to 122 undergraduates who had experienced an emotionally salient, nontraumatic event: a college examination. Results indicated that normal levels of anxiety associated with the examination were positively correlated with scores on the Impact of Event Scale-Revised, Revised Civilian Mississippi Scale, and PTSD Checklist-Civilian Version. Multiple regression analyses indicated that substantial variance in the PTSD scales was accounted for by trait characteristics such as negative and positive affectivity and affect intensity (R2 .29 to .53). Negative affectivity correlated as highly with PTSD measures (r = .46 to .71) as those measures correlated with each other (r = .48 to .65). A high proportion of participants exceeded clinical cutoffs on these measures. The findings suggest that these PTSD scales may be overly sensitive to nontraumatic stressors such as everyday distress and trait characteristics.     

An examination of the wording effect in the Rosenberg Self-Esteem Scale among culturally Chinese people

Previous psychometric studies of the Rosenberg Self-Esteem Scale (RSES; 1965) have shown that items with positive and negative words tend to form 2 factors instead of a single factor for global self-esteem. Recent studies using confirmatory factor analysis have indicated that there is an additional method effect behind negatively worded items. However, researchers conducted these studies using Western participants. Because J. L. Farh and B. S. Cheng (1997) suggested that culturally Chinese people tend to exhibit a modesty bias in self-evaluation, especially on positively worded items, researchers may infer that a wording effect of positively worded items would be evident for culturally Chinese people. The author examined the wording effect in the RSES for culturally Chinese people by comparing different confirmatory factor models. The author analyzed data from 2 independent samples of students at the National Taiwan University (ns = 393, 441) and a national sample of juniors recruited from 140 universities and colleges in Taiwan in 2004 (n = 28,862). Results showed that in addition to a global factor for self-esteem, method effects of positively and negatively worded items should also be specified for a model fitting culturally Chinese people.     

First and Third Person Pronouns in Sex Role Questionnaires

First and third person wording of items on a sex role attitude questionnaire were examined. No correlation was found between women's scores on two versions of the Wellesley Role Orientation Scale (WROS) which were identical except for the pronouns. These results suggest that despite similar content, items worded in the first person are not equivalent to third person items as measures of sex roles. Third person scores of women were much higher than first person scores and were similar to third person scores of men. This was interpreted as indicating that normative expectations regarding women's roles are far more traditional than stated preferences of college women in the sample studied.     

Turkish validation of the Emotional Approach Coping Scale

The Emotional Approach Coping Scale is frequently used to assess coping, which consists of emotional processing and emotional expression. The present aim was to examine the psychometric properties of this scale by utilizing two independent samples: university students (n = 481) and community members (n = 284). Based on goodness-of-fit indices in confirmatory factor analysis, a two-factor model yielded significant findings in these samples. The results of multi-group analysis revealed that the theoretical structure of the dispositional Emotional Approach Coping Scale was the same for men and women. In addition to sufficient internal consistency and test-retest reliability, the relationships between the Emotional Approach Coping Scale and five conceptually related measures (coping styles, positive affect, negative affect, depression, and trait anxiety) demonstrated concurrent validity. Furthermore, the present study provides a map of emotional approach coping styles in a non-Western culture.     

Perceived Perfectionism from God Scale: Development and Initial Evidence

In this study, the Perceived Perfectionism from God Scale (PPGS) was developed with Latter-day Saints (Mormons) across two samples. Sample 1 (N = 421) was used for EFA to select items for the Perceived Standards from God (5 items) and the Perceived Discrepancy from God (5 items) subscales. Sample 2 (N = 420) was used for CFA and cross-validated the 2-factor oblique model as well as a bifactor model. Perceived Standards from God scores had Cronbach alphas ranging from .73 to .78, and Perceived Discrepancy from God scores had Cronbach alphas ranging from .82 to .84. Standards from God scores were positively correlated with positive affect, whereas Discrepancy from God scores was positively correlated with negative affect, shame and guilt. Moreover, these two PPGS subscale scores added significant incremental variances in predicting associated variables over and above corresponding personal perfectionism scores.     

Investigating the Affective Signature of Forgivingness across the Adult Years

Though dispositional forgiveness has been associated with affective well-being, it remains unclear whether these associations are consistent throughout adulthood. The current study investigated whether forgivingness interacted with chronological age or future time perspective to predict affect at the trait and daily level. Participants (n = 332, mean age: 45.5 years) completed baseline measures of forgivingness, positive and negative affect, and future time perspective, along with daily assessments of positive and negative affect for up to ten days. Results suggest that the associations between forgivingness and affective well-being differ somewhat based on age and future time perspective, and level of analysis.     

Personality dimensions across cultures

In order to generalize the dimensional structure of personality-relatively independent groups of covarying traits-across languages and cultures, a large number of cultures must be studied. Until recently only a few worldwide personality datasets have been available. The first large-scale studies indicate that the pattern of covariation between personality traits is universal and is relatively easily generalizable across languages and cultures. In contrast to the structure of personality, the comparison of the mean trait scores across cultures is much more problematic because cross-cultural differences turned out to be very small in their magnitude, about one-third of the magnitude of individual differences within culture. More integral (e.g., the similarity between personality profiles) or subtle (e.g., the disparity between positively and negatively worded items) measures can reveal more systematic relationships with relevant socioeconomic and geographic variables than the mean scores themselves. Relatively modest sizes of cross-cultural differences in the mean values may imply that a reasonable scalar equivalence can be achieved, and all individuals, irrespective of their language and culture, can be represented in a common metric.     

Development of a random response scale for the multiple affect adjective check list-revised

A scale was constructed to identify random responses on the Multiple Affect Adjective Check List-Revised. Items chosen were the 14 least frequently checked items and 14 most frequently checked items, plus the seven most frequently checked negative items and the seven least frequently checked positive items (total=42). The Random Response Scale successfully differentiated random protocols from those produced by 420 college students, and scores on the scale were significantly higher for the college students than for the random sample. In addition, correlations between scores on the Random Response Scale and the Communality Scale (Adjective Check List) and the NEO-FFI Conscientiousness Scale suggest its usefulness as a measure of "conscientiousness" or "dependableness."     

Normative and idiosyncratic measures of positive and negative affect in sport

Objectives: The purposes of the study were to: (a) ascertain the degree of similarity amongst normative and idiosyncratic measures of affect; (b) test the notion of reversal effects on the functional impact and the hedonic tone of emotions; (c) analyse the differences on the intensity of facilitating-positive, facilitating-negative, inhibiting-positive, and inhibiting-negative performance emotion content categories, and (d) test whether competitive trait anxiety intensity could predict pre-performance normative or idiosyncratic negative affect intensity.Design: A cross-sectional study design was employed using normative and idiosyncratic measures of affect.Methods: Experienced male soccer and volleyball players (N =124) were individually assessed. Normative scales were the Positive and Negative Affect Schedule (PANAS) used as a sport-specific trait measure of competitive affect, and the Sport Anxiety Scale (SAS) administered for the assessment of competitive trait anxiety. Idiosyncratic affect occurring prior to or during optimal and poor competitions was identified in the conceptual framework of the Hanin's Individual Zones of Optimal Functioning (IZOF) model (Hanin, Y.L. (1997). Emotions and athletic performance: Individual zones of optimal functioning model. European Yearbook of Sport Psychology, 1, 29–72).Results: Findings revealed low overlap amongst the PANAS items and the idiosyncratic items, reversal effects on the functional impact and the hedonic tone of emotions, higher level of positive affect associated with optimal performance when compared with all other affect categories, and lower levels of facilitating-negative category when compared with the facilitating-positive. Moreover, sport trait anxiety intensity was found to predict negative affect intensity as assessed by the PANAS normative scales and by the idiosyncratic items. For idiosyncratic affect, however, significant differences emerged only when comparing individuals with very high or very low anxiety symptoms.Conclusions: Idiosyncratic affect scales together with normative scales are recommended for research and applied purposes. The functional impact as well as the hedonic tone of emotion items need to be examined according to individual perception.     

Acquiescent responding in partially balanced multidimensional scales

We propose a method for controlling acquiescent response in which acquiescence response variance is isolated in an independent factor. This kind of procedure is available for perfectly balanced scales (i.e. half of the items are worded in the opposite direction to the other half with respect to a general trait). However, few questionnaires are designed so that exactly half of the items are worded in this way. If this is not the case, the available methods are useless. We propose to adapt the rotation method of Lorenzo‐Seva and Rodríguez‐Fornells to handle partially balanced scales (i.e. only a few items in the scale are worded in the opposite direction). The most important characteristic of our method is that it removes the variance due to acquiescent response from all the items in the questionnaire (i.e. the balanced subset of items, but also the unbalanced subsets of items). The usefulness of the method is illustrated in a numerical example.