Using Item Response Theory for the Development of a New Short Form of the Eysenck Personality Questionnaire-Revised

Causal relationship between neuroticism and frailty: A bidirectional Mendelian randomization study

BACKGROUND

Observational studies have shown that neuroticism is associated with frailty, but the causal relationship between them remains unclear.

METHODS

A two-sample Mendelian randomization (MR) study was conducted to explore the bidirectional causal relationship between neuroticism (n = 380,506 for the primary analysis, n = 79,004 for the validation) and frailty (n = 175,226) using publicly available genome-wide association study data. The inverse variance weighted (IVW), weighted median, and MR-Egger were used to obtain the causal estimates. Findings were verified through extensive sensitivity analyses and validated using another dataset. Multivariable MR (MVMR) analysis was performed to estimate the direct causal effects with adjustment of potential confounders. Two-step MR technique was then conducted to explore the mediators in the causal effects of neuroticism on frailty.

RESULTS

Genetically-predicted higher neuroticism score was significantly correlated with higher frailty index (IVW beta: 0.53, 95%CI: 0.48 to 0.59, P = 9.3E-83), and genetically-determined higher frailty index was significantly associated with higher neuroticism score (IVW beta: 0.28, 95%CI: 0.21 to 0.35, P = 1.3E-16). These results remained robust across sensitivity analyses and were reproducible using another dataset. The MVMR analysis indicated that the causal relationships remained significant after adjusting for the potential confounding factors. Mediation analysis revealed that depression, years of schooling, and smoking were significantly mediated the causal effects of neuroticism on frailty.

CONCLUSIONS

A bidirectional causal relationship existed between neuroticism and frailty. Our findings suggested that early intervention and behavioral changes might be helpful to reduce the neuroticism levels and prevent the development of frailty.

Dissecting the causal relationship between neuroticism and osteoarthritis: a univariable and multivariable Mendelian randomization study

Background

Mental health has been found to be associated with risk of osteoarthritis (OA), but the causal relationship was not fully clarified.

Methods

Two-sample Mendelian randomization (MR) study was conducted to investigate the causal relationship between neuroticism (n = 329,821) and the two most frequently affected parts of osteoarthritis (OA) (knee OA: case/control =24,955/378,169; hip OA: case/control = 15,704/378,169) using large scale summary genome-wide association study (GWAS) data. Inverse variance weighted (IVW), weighted median, and MR-Egger were used to estimate the causal effects. Multiple sensitivity analyses were conducted to examine the robustness of the causal estimates. Multivariable MR analysis was used to estimate the direct effects of neuroticism on OA after accounting for the other OA risk factors. Two-step MR approach was employed to explore the potential mediators of the causal relationship.

Results

Univariable MR analysis indicated that 1-SD increase in genetically predicted neuroticism score was associated with an increased risk of knee OA (IVW: OR, 1.17; 95% CI, 1.087-1.26; p = 2.72E-05) but not with hip OA. The causal effects remained significant after accounting for the effects of BMI, alcohol drinking, and vigorous physical activity but were attenuated with adjustment of smoking. Further mediation analysis revealed that smoking initiation mediated a significant proportion of the causal effects of neuroticism on knee OA (proportion of mediation effects in total effects: 22.3%; 95% CI, 5.9%-38.6%; p = 7.60E-03).

Conclusions

Neuroticism has significant causal effects on knee OA risk. Smoking might partly mediate the causal relationship. Further studies were warranted to explore the underlying mechanisms and potential use of neuroticism management for OA treatment.

Artificial Neural Networks for Short-Form Development of Psychometric Tests: A Study on Synthetic Populations Using Autoencoders

Short-form development is an important topic in psychometric research, which requires researchers to face methodological choices at different steps. The statistical techniques traditionally used for shortening tests, which belong to the so-called exploratory model, make assumptions not always verified in psychological data. This article proposes a machine learning-based autonomous procedure for short-form development that combines explanatory and predictive techniques in an integrative approach. The study investigates the item-selection performance of two autoencoders: a particular type of artificial neural network that is comparable to principal component analysis. The procedure is tested on artificial data simulated from a factor-based population and is compared with existent computational approaches to develop short forms. Autoencoders require mild assumptions on data characteristics and provide a method to predict long-form items' responses from the short form. Indeed, results show that they can help the researcher to develop a short form by automatically selecting a subset of items that better reconstruct the original item's responses and that preserve the internal structure of the long-form.

The psychometric property of a short-form of the Social Axioms Survey (SAS II)

BACKGROUND

Social Axioms are generalized beliefs and broad assumptions about the world, guiding behaviors across various social situations. Social Axioms are usually assessed by Social Axioms Survey II (SAS II). Nevertheless, the length of the scale may limit its usefulness in studies with strict time constraint. The present study aimed at developing a shorter version.

METHODS

A survey was conducted among 455 college students. First, we performed psychometric evaluation on the full item version of SAS II to identify items with superior psychometric properties for a brief version of SAS II. Second, we validated the psychometric properties of the brief version of SAS II.

RESULTS

A 20-item version of SAS II (SAS II-20) was developed, and it demonstrated adequate reliability and validity. The correlations between SAS II-20 and personality variables, cognitive flexibility, interpersonal trust, locus of control, and paranormal beliefs were consistent with past studies.

CONCLUSIONS

SAS II-20 is psychometrically acceptable and provides a time-efficient measurement tool for investigating social beliefs.

Enhancing Computerized Adaptive Testing with Batteries of Unidimensional Tests

The article presents a new computerized adaptive testing (CAT) procedure for use with batteries of unidimensional tests. At each step of testing, the estimate of a certain ability is updated on the basis of the response to the latest administered item and the current estimates of all other abilities measured by the battery. The information deriving from these abilities is incorporated into an empirical prior that is updated each time that new estimates of the abilities are computed. In two simulation studies, the performance of the proposed procedure is compared with that of a standard procedure for CAT with batteries of unidimensional tests. The proposed procedure yields more accurate ability estimates in fixed-length CATs, and a reduction of test length in variable-length CATs. These gains in accuracy and efficiency increase with the correlation between the abilities measured by the batteries.

Development of a scale for capturing psychological aspects of physical-digital integration: relationships with psychosocial functioning and facial emotion recognition

The present work aims at developing a scale for the assessment of a construct that we called "physical-digital integration", which refers to the tendency of some individuals not to perceive a clear differentiation between feelings and perceptions that pertain to the physical or digital environment. The construct is articulated in four facets: identity, social relationships, time-space perception, and sensory perception. Data from a sample of 369 participants were collected to evaluate factor structure (unidimensional model, bifactor model, correlated four-factor model), internal consistency (Cronbach's α, McDonald's ω), and correlations of the physical-digital integration scale with other measures. Results showed that the scale is valid and internally consistent, and that both the total score and the scores at its four subscales are worthy of consideration. The physical-digital integration scores were found to be differently associated with digital and non-digital behaviors, individuals' ability to read emotions in the facial expressions of others, and indicators of psychosocial functioning (anxiety, depression, and satisfaction with social relationships). The paper proposes a new measure whose scores are associated with several variables that may have relevant consequences at both individual and social levels.

Relationship of neurocognitive ability, perspective taking, and psychoticism with hostile attribution bias in non-clinical participants: Theory of mind as a mediator

Objectives

Hostile attribution bias is reportedly common from non-clinical population to those with serious mental illness, such as schizophrenia, and is known to be closely related to theory of mind (ToM). This study aimed to investigate whether ToM skills mediate the relationship among neurocognitive ability, personality traits, and attribution bias.

Methods

A total of 198 (101 females) non-clinical youths were recruited. To assess their neurocognitive ability and ToM skills, the participants were asked to complete Raven’s Standard Progressive Matrices (SPM) and the Korean version of the Reading the Mind in Eyes Test (K-RMET). To determine their personality traits, the Eysenck Personality Questionnaire (psychoticism) and interpersonal reactivity index (perspective taking) were used. To evaluate hostile attribution bias, the Ambiguous Intentions Hostility Questionnaire was administered. Path analysis and bias-corrected percentile bootstrap methods were used to estimate model fit and the parameters of the mediating effects.

Results

Based on model comparison, the best model characterized (1) two direct pathways from psychoticism and the K-RMET to hostility attribution bias and (2) three indirect pathways, wherein SPM, perspective taking, and psychoticism influenced hostile attribution bias through K-RMET. The final model fit indices were good [x²/df = 1.126; comparative fit index = 0.996; root mean square error of approximation = 0.026; standard root mean square residual = 0.026 and Akaike information criterion = 28.251] and the K-RMET fully mediated the association between SPM, perspective taking, psychoticism, and hostile attribution bias.

Conclusion

The main findings suggested that ToM skills, such as the RMET, play an important role in explaining the relationship among neurocognitive ability, personality traits, and hostile attribution bias. ToM skills and a remediation strategy may need to be developed to balance the enhanced hostility bias that underlies the paranoia.

Peabody Developmental Motor Scales-2: The Use of Rasch Analysis to Examine the Model Unidimensionality, Motor Function, and Item Difficulty

The Peabody Developmental Motor Scales-Second Edition (PDMS-2) is a valid and reliable instrument used in several countries, including Brazil, to assess gross and fine motor skills and identify motor deficits and eligibility for intervention for children with and without disabilities. However, the analysis of PDMS-2 items regarding the unidimensionality of the model, order of item difficulty, and whether the items portray the children's developmental trajectories still lacks investigation. Therefore, this study aims to: (1) analyze the unidimensionality of PDMS-2, (2) verify the model's capacity to explain the variance in the motor function responses, and (3) identify the level of difficulty of the items for Brazilian children. Children (n = 637; 51% girls) newborn to 71 months (M age = 21.7, SD = 18.6) were assessed using the PDMS-2. The Rasch analysis was conducted; the indexes of infit and outfit, and the point-biserial correlations coefficient were analyzed. The model unidimensionality was investigated using percentages of variance in the Rasch model (40% of variance). Results indicated that (1) for reflexes subscale, 62.5% of the items had correlations with the factor above 0.60, and two items had unadjusted infit and outfit; (2) for stationary subscale, 83.3% of the correlations of the items with the factor were above 0.50, and one item had unadjusted infit and outfit; (3) for locomotion subscale, 80.0% of the correlation of the items with the factor were above 0.50; all items had adequate infit and outfit; (4) for object manipulation subscale, 79.9% of the correlation of the items with the factor were above 0.50, and one item had unadjusted infit and outfit; (5) for grasping subscale, 92.3% of the correlation of the items with the factor were above 0.50, and one item had unadjusted infit and outfit; and (6) for the visual-motor integration subscale, 73.6% of the correlation of the items with the factor were above 0.50, and six items had unadjusted infit and outfit. The items with unadjusted fit were removed for further analysis. No changes in reliability and separation of items and people scores were observed without the unadjusted items; therefore, all items were maintained. A unidimensional model was found, and the reliability and discriminant capability of the items were adequate, and all items should be used to assess children. The PDMS-2 is appropriate for assessing Brazilian children.

The Caregiving System Scale

Abstract. According to attachment theory, the propensity of human beings to care for others is governed by an inborn caregiving behavioral system that aims to promote welfare and reduce the distress of other people through effective provision of care. However, some individuals may develop non-optimal caregiving strategies, such as anxious hyperactivation and avoidant deactivation. These two non-optimal caregiving strategies can be evaluated in adults using the Caregiving System Scale (CSS). Recent findings suggested that the factor structure of the instrument may be more complex than was intended. The present work examines in-depth the factor structure of the CSS to provide a clearer understanding of the underlying dimensions. Gender invariance and the contribution of attachment orientations to CSS scores are also examined. Findings reveal that, whereas the CSS-deactivation subscale is unidimensional, the CSS-hyperactivation subscale is better represented by two distinct yet related constructs – caregiving-related worries/doubts and intrusive/coercive caregiving. Partial strict gender invariance is supported. The contribution of attachment orientations to non-optimal caregiving strategies is consistent with theoretical expectations. Results and future research directions are discussed in the final section.

An Item Response Theory-Based Scoring of the South Oaks Gambling Screen-Revised Adolescents

The South Oaks Gambling Screen-Revised Adolescent (SOGS-RA) is one of the most widely used screening tools for problem gambling among adolescents. In this study, item response theory was used for computing measures of problem gambling severity that took into account how much information the endorsed items provided about the presence of problem gambling. A zero-inflated mixture two-parameter logistic model was estimated on the responses of 4,404 adolescents to the South Oaks Gambling Screen-Revised Adolescent to compute the difficulty and discrimination of each item, and the problem gambling severity level (θ score) of each respondent. Receiver operating characteristic curve analysis was used to identify the cutoff on the θ scores that best distinguished daily and nondaily gamblers. This cutoff outperformed the common cutoff defined on the sum scores in identifying daily gamblers but fell behind it in identifying nondaily gamblers. When screening adolescents to be subjected to further investigations, the cutoff on the θ scores must be preferred to that on the sum scores.

Genotype-Environment Correlation and Its Relation to Personality - A Twin and Family Study

The aim of the study was to examine the Family and School Psychosocial Environment (FSPE) questionnaire in relation to a possible genotype-environment correlation and genetic mediation between the FSPE variables and personality variables, assessed by the Junior Eysenck Personality Questionnaire. A sample of 506 Swedish children aged 10-20 years from 253 families were recruited via the Swedish state population and address register and SchoolList.Eu. The children were divided into 253 pairs: 46 monozygotic twin pairs, 42 dizygotic twin pairs, 140 pairs of full siblings and 25 pairs of half-siblings. The behavioral genetic analysis showed that both FSPE factors, Warmth and Conflicts, may be partly influenced by genetic factors (suggesting genotype-environment correlation) and that nonadditive genetic factors may mediate the relationship between FSPE factors and psychoticism/antisocial personality (P). An indication of a special shared monozygotic twin environment was found for P and Lie/social desirability, but based on prior research findings this factor may have a minor influence on P and L. P and L were negatively correlated, and the relationship seems to be partly mediated by nonadditive genetic factors. Nonshared environment and measurement errors seem to be the most influential mediating factors, but none of the cross-twin cross-dimension correlations suggest a common shared environmental mediating factor.

A Comparison of Classical and Modern Measures of Internal Consistency

Three measures of internal consistency – Kuder-Richardson Formula 20 (KR20), Cronbach’s alpha (α), and person separation reliability (R) – are considered. KR20 and α are common measures in classical test theory, whereas R is developed in modern test theory and, more precisely, in Rasch measurement. These three measures specify the observed variance as the sum of true variance and error variance. However, they differ for the way in which these quantities are obtained. KR20 uses the error variance of an “average” respondent from the sample, which overestimates the error variance of respondents with high or low scores. Conversely, R uses the actual average error variance of the sample. KR20 and α use respondents’ test scores in calculating the observed variance. This is potentially misleading because test scores are not linear representations of the underlying variable, whereas calculation of variance requires linearity. Contrariwise, if the data fit the Rasch model, the measures estimated for each respondent are on a linear scale, thus being numerically suitable for calculating the observed variance. Given these differences, R is expected to be a better index of internal consistency than KR20 and α. The present work compares the three measures on simulated data sets with dichotomous and polytomous items. It is shown that all the estimates of internal consistency decrease with the increasing of the skewness of the score distribution, with R decreasing to a larger extent. Thus, R is more conservative than KR20 and α, and prevents test users from believing a test has better measurement characteristics than it actually has. In addition, it is shown that Rasch-based infit and outfit person statistics can be used for handling data sets with random responses. Two options are described. The first one implies computing a more conservative estimate of internal consistency. The second one implies detecting individuals with random responses. When there are a few individuals with a consistent number of random responses, infit and outfit allow for correctly detecting almost all of them. Once these individuals are removed, a “cleaned” data set is obtained that can be used for computing a less biased estimate of internal consistency.

Development and psychometric properties of a short form of the Illness Denial Questionnaire

Background

Coping with chronic illness can be overwhelming for patients and caregivers, and may be inhibited by the denial mechanism, and therefore, denial represents a critical issue for health professionals. Assessing illness denial is far from easy, and brief tools suitable for medical settings are lacking. In this paper, the development of a short form of the Illness Denial Questionnaire (IDQ) for patients and caregivers is presented.

Methods

In study 1, the IDQ was administered to 118 patients and 83 caregivers to examine the internal structure of denial; then the properties of the items (DIF, fit, and difficulty) were evaluated according to the Rasch model in order to select the best items for the Illness Denial Questionnaire-Short Form (IDQ-SF). Study 2 included 202 participants (113 patients and 89 caregivers). The internal structure of the IDQ-SF was tested via confirmatory factor analysis (CFA). Reliability and concurrent validity were also studied using the Anxiety and Depression Questionnaire-Reduced Form (AD-R).

Results

The CFA showed a two-factor structure encompassing "Denial of negative emotions" and "Resistance to change". Results of the Rasch analyses led to the selection of 4 items for each dimension. The resulting IDQ-SF (8 items) showed a two-factor structure as well as good reliability and concurrent validity with AD-R.

Conclusion

The IDQ-SF represents a valid tool for quickly evaluating the core of illness denial in patients and caregivers. This brief and easily administrable questionnaire allows health professionals to outline the presence and severity of illness denial in order to set individually tailored interventions.