To Be or Not to Be: Exploring the Nature of Positively and Negatively Keyed Personality Items in High-Stakes Testing

Including equal numbers of positively and negatively keyed items is common in Five-Factor Model (FFM) personality measures. Much literature has demonstrated the presence of positive and negative keying factors in low-stakes testing situations, but there is a dearth of research investigating these factors in high-stakes testing. To address this gap, we investigated whether an FFM measure used in high-stakes testing was influenced by positive and negative keying factors. We also examined the overlap of the positive and negative keying factors with social desirability, rule-consciousness, acquiescence, and cognitive ability. Confirmatory factor analysis supported the inclusion of distinct factors associated with positively and negatively keyed items and suggested that the keying factors accounted for a substantial portion of variation in responses to FFM items. Social desirability and rule-consciousness were found to have significant relations with both keying factors, whereas acquiescence was only related to the negative keying factor. Implications for the construct validity of FFM measures used in high-stakes testing and directions for future research are discussed.     

Structured personality test item characteristics and validity

Using the structured personality test item as the unit of analysis, the purpose of this research was to evaluate the relationship between validity and a variety of other test item parameters. Of particular interest was the relationship of test item criterion validity to negative keying and to negative wording. By drawing a distinction between negative keying and negative wording it was demonstrated that the use of balanced scales to control acquiescence need not result in a reduction in item criterion validity. Whereas the use of negative wording has in the past reduced validity, data demonstrated that positively worded, negatively keyed items did not. In addition, results indicated that clear, moderately short, relevant test items tended to be the most empirically valid.     

Implications of Item Keying and Item Valence for the Investigation of Construct Dimensionality

Factor analysis and nomological network analysis are commonly used as complementary procedures in the investigation of the dimensionality of constructs (e.g., self-esteem, job satisfaction). Although it has been demonstrated that factor analyses are often biased toward a two-dimensional solution for measures including regular- and reverse-keyed items, less attention has been paid to the implications for nomological network analyses. We propose, and demonstrate empirically in two studies, that item keying is confounded with item valence (i.e., favorability of item content), and that item valence can bias the results of both factor analysis and nomological network analysis toward a two-dimensional interpretation. We also demonstrate that the valence effect is related to, but distinguishable from, social desirability response bias. We caution that the practice of excluding reverse-keyed items to achieve unidimensionality can lead to distortion in correlations among constructs, and we offer alternative remedies to the valence problem.     

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。     

The role of item extremity in the emergence of keying-related factors: an exploration with the life orientation test

Keying-related factors in psychological scales are variously interpreted substantively or as products of violations of the assumptions underlying item keying. The present study investigated whether the extremity of the wording of items may contribute to the emergence of item-keying factors in a commonly used psychological scale. Respondents (N = 277) completed the Life Orientation Test (M. F. Scheier & C. S. Carver, 1985) in either its original or modified, more moderately worded form. Results indicate that the interaction of item extremity and item keying significantly affected subscale means and, more important, that the more moderately worded scale was substantially more unidimensional. Results are explained partially through the association of lesser and greater extremity with the tendency for some respondents to agree or disagree with items irrespective of keying direction. These results, although demonstrated in only 1 scale, have potential relevance to any scale comprising positive and negative items.     

Creators and Presses: The Person–Situation Interaction in Negative Creativity

Given the recent emphasis on exploring valence in creative behavior, this study examines negative creativity via a person–situation interactionist perspective. By manipulating goal valence (uses or misuses) and object valence (positive or negative), four conditions of an adapted Divergent Thinking task were used to predict positive and negative creativity. Participants (N = 178, 103 females, M_age = 23.82, SD = 4.03, range: 18–38) responded to a single condition along with the Big Five and Dark Triad personality scales in a between-groups design. Hierarchical multiple regressions revealed that goal and object valence significantly explained variance in valences of creative responses, beyond individual differences of personality. Furthermore, the congruence between goal and object valence predicted valenced creativity; that is, the positive objects and goals condition yielded more positive-original ideas, and the negative objects and goals condition yielded more negative-original ones. Object valence alone (material presses) did not contribute significantly to explaining valenced creativity. Negative-original responses were inversely related to conscientiousness, and directly to intellect/imagination and secondary psychopathy. Thus, negative creativity was attributed relatively more to person as compared to situation variables. Results are discussed from a valence-based and interactionist perspective.     

Personality effects and sex differences on the International Affective Picture System (IAPS): A Spanish and Swiss study

The present study analyses the relationship between Anxiety and Impulsivity personality factors and emotions, by controlling for country and sex effects in a sample of Spanish and Swiss university students. Emotions were assessed through the International Affective Picture System (IAPS) of pictures (valence/arousal) using the Self-Assessment Manikin (SAM) procedure. The mixed valence/arousal groups’ pictures were formed according to Tok, Koyuncu, Dural and Catikkas procedure (2010). Results showed that females scored significantly higher in Anxiety factor and men in Impulsivity factor in both countries. The effect of sex was highly significant for Anxiety (ŋ²: 0.12), but there was no significant effect of the country. Also, females obtained higher scores in the four valence/arousal pictures groups. The sex effect was particularly robust for negative valence–high arousal (ŋ²: 0.13). Impulsivity correlated with high ratings of positive valence–high arousal while Anxiety correlated with high ratings of negative valence–high arousal and with high ratings of negative valence–low arousal in both sexes, although scores were higher for females. Structural Equation Modelling confirmed these relationships. Nevertheless, Anxiety and Impulsivity explained only a small amount of the accounted variance of the self-reported valence and arousal of the pictures.     

Hostility-Related Prejudice Against Turks in Adolescents: Masked Affective Priming Allows for a Differentiation of Automatic Prejudice

The masked affective priming task was used as an unobtrusive measure of intergroup prejudices in a sample of German adolescents (aged 13–15). Pictures of Turks and Germans were used as masked primes that preceded positive and negative target adjectives conveying either other-relevant valence (e.g., honest, evil) or possessor-relevant valence (e.g., talented, dull). Affective priming indices (denoting relative negativity of Turkish primes) were positively correlated with the open expression of prejudices towards Turks and foreigners in general in questionnaires as well as with discriminative interaction behavior in a virtual ball-tossing game. As expected, these correlations were found only for priming indices based on other-relevant targets, thereby emphasizing the differentiation of automatic prejudice into (imputed) hostility and depreciation.     

Bandwidth and fidelity on the NEO-Five Factor Inventory: replicability and reliability of Saucier's (1998) item cluster subcomponents

Many users of the NEO-Five Factor Inventory (NEO-FFI; Costa & McCrae, 1992) are unaware that Saucier (1998) developed item cluster subcomponents for each broad domain of the instrument similar to the facets of the Revised NEO Personality Inventory (Costa & McCrae, 1992). In this study, I examined the following: the replicability of the subcomponents in young adult university and middle-aged community samples; whether item keying accounted for additional covariance among items; subcomponent correlations with a measure of socially desirable responding; subcomponent reliabilities; and subcomponent discriminant validity with respect to age-relevant criterion items expected to reflect varying associations with broad and narrow traits. Confirmatory factor analyses revealed that all subcomponents were recoverable across samples and that the addition of method factors representing positive and negative item keying improved model fit. The subcomponents correlated no more with a measure of socially desirable responding than their parent domains and showed good average reliability. Correlations with criterion items suggested that subcomponents may prove useful in specifying which elements of NEO-FFI domains are more or less related to variables of interest. I discuss their use for enhancing the precision of findings obtained with NEO-FFI domain scores.     

Beyond the Score: Clinical Evaluation Through Formal Psychological Assessment

Questionnaires generate numerical scores, but endorsing different sets of items could produce the same score despite reflecting qualitatively different configurations of clinical features. Formal psychological assessment (FPA) attempts to overcome this by identifying the clinical features entailed by observed response patterns. This study illustrates an application of FPA to the cleaning subscale of a questionnaire assessing obsessive–compulsive symptoms and DSM–IV–TR diagnostic criteria for obsessive–compulsive disorder. A deterministic model of the items–criteria relationships was constructed by mapping each item to each diagnostic criterion. The resulting model was tested on a large community sample (N = 4,412). Results indicate that the theoretical model has adequate fit; item error rates and probabilities for each of the criteria are examined. Clinically relevant examples of the items–criteria relationships are discussed. Possible applications of FPA to personality assessment are also discussed, including long multidimensional questionnaires and questionnaires that use subtle item content.     

Reliving emotional personal memories: affective biases linked to personality and sex-related differences

Although available evidence suggests that the emotional valence and recollective properties of autobiographical memories (AMs) may be influenced by personality- and sex-related differences, overall these relationships remain poorly understood. The present study investigated these issues by comparing the effect of general personality traits (extraversion and neuroticism) and specific traits linked to emotion regulation (ER) strategies (reappraisal and suppression) on the retrieval of emotional AMs and on the associated postretrieval emotional states, in men and women. First, extraversion predicted recollection of positive AMs in both men and women, whereas neuroticism predicted the proportion of negative AMs in men and the frequency of rehearsing negative AMs in women. Second, reappraisal predicted positive AMs in men, and suppression predicted negative AMs in women. Third, while reliving of positive memories had an overall indirect effect on postretrieval positive mood through extraversion, reliving of negative AMs had a direct effect on postretrieval negative mood, which was linked to inefficient engagement of suppression in women. Our findings suggest that personality traits associated with positive affect predict recollection of positive AMs and maintenance of a positive mood, whereas personality traits associated with negative affect, along with differential engagement of habitual ER strategies in men and women, predict sex-related differences in the recollection and experiencing of negative AMs. These findings provide insight into the factors that influence affective biases in reliving AMs, and into their possible link to sex-related differences in the susceptibility to affective disorders.     

Valence, comparison focus and self-positivity biases: does it matter whether people judge positive or negative traits?

The present research questions whether mere valence affects self-other comparisons in the domain of trait characteristics. While some previous studies have reported greater positivity bias for the self when traits were positive than when traits were negative, we suggest that this is an ambiguous finding, because valence and content were confounded. When we unconfounded content and valence, valence had no effect on the magnitude of self-positivity bias displayed. We also replicate several findings for our unconfounded set of traits. Firstly, comparing others to the self, rather than comparing the self to others, lowered self-positivity for positive and negative traits (focus effect). Secondly, extremely positive and negative traits triggered greater positivity bias than did more moderately evaluated ones. Finally, we suggest that comparative self-positivity biases may be based on a general positivity bias.     

When bad gets worse: Negative wording amplifies negative recall in persons with the borderline personality trait

The emotional valence of memory recall strongly influences the extreme, and often self‐destructive behavioral response. Overall, 4427 people from the community filled out our screening questionnaire, of which 674 (mean age: 32.9 years, 64.7% women) passed the threshold for the likely presence of borderline personality disorder (BPD). Participants watched one of 9 possible video clips differing in valence (positive, negative or neutral). Results indicate support (ß = .151, p < .05) for a negative cognitive tendency in the evaluations proportionate to the severity of BPD but only when the impression is assessed with negatively‐worded items. This effect is detectable regardless of other factors such as negative mood, extremity of the answers or general impression of the character. These effects are unaffected by acute trauma. To conclude, different factors influence the recall of emotional stimuli, depending on the valence of the stimuli and the valence of the rating scale in persons with strong borderline tendencies.     

Independent effects of word concreteness and word valence on immediate serial recall

The present study examined the roles of word concreteness and word valence in the immediate serial recall task. Emotion words (e.g. happy) were used to investigate these effects. Participants completed study-test trials with 7-item study lists consisting of positive or negative words with either high or low concreteness (Experiments 1 and 2) and neutral (i.e. non-emotion) words with either high or low concreteness (Experiment 2). In serial recall performance, we replicated the typical item concreteness effect (concrete words are better recalled than abstract words) and obtained an item valence effect (positive/neutral words are better recalled than negative words). However, there was no concreteness × valence interaction. We conclude that both word valence and word concreteness independently contribute to the serial order retention of emotion words in the immediate serial recall task.     

大小判断任务中正负号及其异同对SNARC效应的影响

SNARC效应(Spatial-Numerical Association of Response Codes)是指被试对数字做按键反应时,对于较小的数字,按左键的速度快于按右键;对于较大的数字,按右键的速度快于按左键。本研究以ERP作为测量手段,采用修正的大小判断任务,旨在探究数字正负号及其异同对SNARC效应的影响。行为结果发现,在反应时上,当目标数字与基线数字正负号相同且基线数字为+5时,一致条件显著快于不一致条件。ERP结果发现,当目标数字与基线数字正负号相同时,无论基线数字为+5还是–5,在反应选择阶段,不一致都比一致条件更负且均诱发了P3。当目标数字与基线数字正负号相异时,若基线数字为+5,一致比不一致条件在刺激呈现阶段诱发了波幅显著更小的N300;若基线数字为–5,一致比不一致条件在反应执行阶段诱发了更正的LPP。无论目标数字与基线数字正负号相同还是相异,在反应选择阶段,不一致都比一致条件更负且均诱发了P3,表明出现了SNARC效应。同时,SNARC效应的出现激活了额叶头皮位置,负数加工伴随左额叶的激活,而正数加工伴随右额叶的激活,溯源分析结果进一步表明SNARC效应定位于额叶与顶叶。这些结果说明负数按实际大小表征在心理数字线上,支持了负数空间表征的个体发展论假说;表明符号捷径机制会改变SNARC效应的发生时间;同时证明了负数与正数的空间表征具有不同的优势半球。     

Looking behaviour and preference for artworks: The role of emotional valence and location

The position of an item influences its evaluation, with research consistently finding that items occupying central locations are preferred and have a higher subjective value. The current study investigated whether this centre-stage effect (CSE) is a result of bottom-up gaze allocation to the central item, and whether it is affected by item valence. Participants (n = 50) were presented with three images of artistic paintings in a row and asked to choose the image they preferred. Eye movements were recorded for a subset of participants (n = 22). On each trial the three artworks were either similar but different, or were identical and with positive valence, or were identical and with negative valence. The results showed a centre-stage effect, with artworks in the centre of the row preferred, but only when they were identical and of positive valence. Significantly greater gaze allocation to the central and left artwork was not mirrored by equivalent increases in preference choices. Regression analyses showed that when the artworks were positive and identical the participants' last fixation predicted preference for the central art-work, whereas the fixation duration predicted preference if the images were different. Overall the result showed that item valence, rather than level of gaze allocation, influences the CSE, which is incompatible with the bottom-up gaze explanation. We propose that the centre stage heuristic, which specifies that the best items are in the middle, is able to explain these findings and the centre-stage effect.     

“Authoritarianism” and Agreement with Things Authoritative

The use of personality and attitudinal questionnaires in private executive and military samples is briefly reviewed. The need for evidence regarding the factor similarity of such questionnaires in comparative studies is noted. Such a comparative factorial study of an abbreviated version of the Shure-Meeker (11) Personality/Attitude Schedule (PAS) is reported. In terms of the general scales examined, three of the four scales emerge in both samples. The order of emergence, as well as the number and item content of the factors, differs slightly between the two samples. Caution is suggested in assuming factor similarity across samples in comparative studies utilizing the PAS and other personality and attitudinal instruments.     

The word concreteness effect occurs for positive,but not negative,emotion words in immediate serial recall

The present study examined the roles of word concreteness and word valence in the immediate serial recall task. Emotion words (e.g. happy) were used to investigate these effects. Participants completed study‐test trials with seven‐item study lists consisting of positive or negative words with either high or low concreteness (Experiments 1 and 2) and neutral (i.e. non‐emotion) words with either high or low concreteness (Experiment 2). For neutral words, the typical word concreteness effect (concrete words are better recalled than abstract words) was replicated. For emotion words, the effect occurred for positive words, but not for negative words. While the word concreteness effect was stronger for neutral words than for negative words, it was not different for the neutral words and the positive words. We conclude that both word valence and word concreteness simultaneously contribute to the item and order retention of emotion words and discuss how Hulme et al.'s (1997) item redintegration account can be modified to explain these findings.     

The effect of availability heuristics in online consumer reviews

This research investigates principles of judgmental heuristics and dual processing systems in the online purchasing environment. It examines the effects of availability cues in restaurant reviews on dining intentions and menu item choice. Two experiments are reported in which consumers make dining out and food choice decisions using simulated online review sites. The first experiment evaluates primacy–recency effects of positive and negative reviews along with different review types. The findings indicate that text plus ratings produce higher visit intentions and expectations compared with either cue by itself. The second experiment examines the effects of rating format, visual cues, and review valence on menu item choice. Pictures increase likelihood to choose a positively reviewed menu item but do not influence likelihood to choose a negatively reviewed item. This finding supports the negativity bias, whereby consumers place more weight on negative versus positive information. Consumers are more likely to choose a menu item with pictures when ratings are in numerical versus star rating format. The findings can be interpreted in terms of System 1 (heuristic) and System 2 (systematic) processing.