What are the mathematical bounds for coefficient α?

Coefficient α, although ubiquitous in the research literature, is frequently criticized for being a poor estimate of test reliability. In this note, we consider the range of α and prove that it has no lower bound (i.e., α ∈ ( - ∞, 1]). While outlining our proofs, we present algorithms for generating data sets that will yield any fixed value of α in its range. We also prove that for some data sets-even those with appreciable item correlations-α is undefined. Although α is a putative estimate of the correlation between parallel forms, it is not a correlation as α can assume any value below-1 (and α values below 0 are nonsensical reliability estimates). In the online supplemental materials, we provide R code for replicating our empirical findings and for generating data sets with user-defined α values. We hope that researchers will use this code to better understand the limitations of α as an index of scale reliability. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Thinking thrice about sum scores, and then some more about measurement and analysis.

Measurement is fundamental to all research in psychology and should be accorded greater scrutiny than typically occurs. Among other claims, McNeish and Wolf (Thinking twice about sum scores. Behavior Research Methods, 52, 2287-2305) argued that use of sum scores (a) implies that a highly constrained latent variable model underlies items comprising a scale, and (b) may misrepresent or bias relations with other criteria. The central claim by McNeish and Wolf that use of sum scores requires the assumption that a parallel test model underlies item responses is incorrect and without psychometric merit. Instead, if a set of items is unidimensional, estimators of reliability are available even if the factor model underlying the set of items does not have a highly constrained form. Thus, dimensionality of a set of items is the key issue, and whether strict constraints on parameter estimates do or do not hold dictate the appropriate way to estimate reliability. McNeish and Wolf also claimed that more precise forms of scoring, such as estimating factor scores, would be preferable to sum scores. We provide analytic bases for reliability estimation and then provide several demonstrations of reliability estimation and the relative advantages of sum scores and factor scores. We contend that several claims by McNeish and Wolf are questionable and that, as a result, multiple recommendations they made and conclusions they drew are incorrect. The upshot is that, once the dimensional structure of a set of items is verified, sum scores often have a solid psychometric basis and therefore are frequently quite adequate for psychological research.

Understanding the Magnitude of Psychological Differences Between Women and Men Requires Seeing the Forest and the Trees.

Whether women and men are psychologically very similar or quite different is a contentious issue in psychological science. This article clarifies this issue by demonstrating that larger and smaller sex/gender differences can reflect differing ways of organizing the same data. For single psychological constructs, larger differences emerge from averaging multiple indicators that differ by sex/gender to produce scales of a construct's overall typicality for women versus men. For example, averaging self-ratings on personality traits more typical of women or men yields much larger sex/gender differences on measures of the femininity and masculinity of personality. Sex/gender differences on such broad-gauge, thematic variables are large relative to differences on their component indicators. This increased effect magnitude for aggregated scales reflects gains in both their reliability and validity as indicators of sex/gender. In addition, in psychological domains such as vocational interests that are composed of many variables, at least some of which differ by sex/gender, the multivariate distance between women and men is typically larger than the differences on the component variables. These analyses encourage recognition of the interdependence of sex/gender similarity and difference in psychological data.

Between- and within-person structures of value traits and value states: Four different structures, four different interpretations.

OBJECTIVE: The circular structure of values has been verified mostly at a between-person level and on measures of general value preferences. In this manuscript, we argue that it is a simplification that neglected significant aspects of the value structures and distinguish four different types of structures: (a) between-person structure of value traits, (b) within-person structure of value traits, (c) between-person structure of value states, and (d) within-person structure of value states. We argue that the within-person structure of value states addresses the circular structure of values most accurately. METHOD: To compare all four structures, we collected three partially dependent samples (N1  = 449, N2  = 293, N3  = 218) of adults (age 17-57, M = 24). At three time points, separated by 5-7 weeks, respondents completed a questionnaire measure (Portrait Values Questionnaire-Revised [PVQ-RR]) of value preferences (value traits) and reported the importance of values in their everyday actions (value states) for 1 week in an experience sampling study. RESULTS: The four types of value structures were stable over time. All four were also consistent with Schwartz's value model to some degree, but at the same time, there were some deviations. CONCLUSIONS: It is important to distinguish four types of value structures and be aware of their different interpretations that we outlined in this paper.

A map of desire: multidimensional scaling of men's sexual interest in male and female children and adults.

BACKGROUND: Men sexually interested in children of a specific combination of maturity and sex tend to show some lesser interest in other categories of persons. Patterns of men's sexual interest across erotic targets' categories of maturity and sex have both clinical and basic scientific implications. METHOD: We examined the structure of men's sexual interest in adult, pubescent, and prepubescent males and females using multidimensional scaling (MDS) across four datasets, using three large samples and three indicators of sexual interest: phallometric response to erotic stimuli, sexual offense history, and self-reported sexual attraction. The samples were highly enriched for men sexually interested in children and men accused of sexual offenses. RESULTS: Results supported a two-dimensional MDS solution, with one dimension representing erotic targets' biological sex and the other dimension representing their sexual maturity. The dimension of sexual maturity placed adults and prepubescent children on opposite ends, and pubescent children intermediate. Differences between men's sexual interest in adults and prepubescent children of the same sex were similar in magnitude to the differences between their sexual interest in adult men and women. Sexual interest in adult men was no more associated with sexual interest in boys than sexual interest in adult women was associated with sexual interest in girls. CONCLUSIONS: Erotic targets' sexual maturity and biological sex play important roles in men's preferences, which are predictive of sexual offending. The magnitude of men's preferences for prepubescent children v. adults of their preferred sex is large.

Alterations in Emotional Diversity Correspond With Increased Severity of Attenuated Positive and Negative Symptoms in the Clinical High-Risk Syndrome.

Background: Alterations in emotional functioning are a key feature of psychosis and are present in individuals with a clinical high-risk (CHR) syndrome. However, little is known about alterations in emotional diversity (i.e., the variety and relative abundance of emotions that humans experience) and clinical correlates in this population. Methods: Individuals meeting criteria for a CHR syndrome (N = 47) and matched healthy controls (HC) (N = 58) completed the modified Differential Emotions Scale (used to derive scores of total, positive, and negative emotional diversity) and clinical interviews (i.e., Structured Interview for Psychosis-Risk Syndromes). Results: Findings showed that the CHR group experienced lower levels of positive emotional diversity compared to HCs. Among the CHR individuals, lower levels of positive and higher levels of negative emotional diversity were associated with more severe attenuated positive and negative symptoms. Analyses controlled for mean levels of emotion and current antipsychotic medication use. Discussion: Results demonstrate that altered emotional diversity (in particular lower levels of positive and higher levels of negative emotional diversity) is a clinically relevant marker in CHR individuals, above and beyond alterations in mean levels of emotional experiences. Future studies may probe sources, downstream consequences, and potential modifiability of decreased emotional diversity in individuals at CHR.

Reliability from α to ω: A tutorial.

Reliability is a fundamental problem for measurement in all of science. Although defined in multiple ways, and estimated in even more ways, the basic concepts seem straightforward and need to be understood by practitioners as well as methodologists. Reliability theory is not just for the psychometrician estimating latent variables, it is for everyone who wants to make inferences from measures of individuals or of groups. For the case of a single test administration, we consider multiple measures of reliability, ranging from the worst (ß) to average (α, λ3) to best (λ4) split half reliabilities, and consider why model-based estimates (ωh, ωt) should be reported. We also address the utility of test-retest and alternate form reliabilities. The advantages of immediate versus delayed retests to decompose observed score variance into specific, state, and trait scores are discussed. But reliability is not just for test scores, it is also important when evaluating the use of ratings. Estimates that may be applied to continuous data include a set of intraclass correlations while discrete categorical data needs to take advantage of the family of κ statistics. Examples of these various reliability estimates are given using state and trait measures of anxiety given with different delays and under different conditions. An online supplemental materials is provided with more detail and elaboration. The online supplemental materials is also used to demonstrate applications of open source software to examples of real data, and comparisons are made between the many types of reliability. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

The Big Five, Everyday Contexts and Activities, and Affective Experience.

Prior research shows that personality traits predict time spent with different people and frequency of engagement in different activities. Further, personality traits, company, and activity are related to the experience of affect. However, little research has examined personality, context, and affect together in the same study. In the current study, 78 people described their Big Five traits and took part in a 1-week experience sampling study using mobile phones as a means for data collection. Participants indicated their current company, activity, and momentary affect along the dimensions of energetic arousal (EA), tense arousal (TA), and hedonic tone (HT). Poisson regressions revealed that traits predicted higher frequencies of trait-consistent contexts: for example, extraversion was related to more frequently being with various types of company. Results predicting contexts from multilevel logistic regressions were sparser. Multilevel models revealed that traits and contexts had main effects on affect, yet there were relatively few interactions of traits X contexts predicting affect. We discuss more specific implications of these findings.

Factor Analysis of Negative Symptom Items in the Structured Interview for Prodromal Syndromes.

BACKGROUND: Negative symptoms occur early in the clinical high risk (CHR) state and indicate increased risk of conversion to psychotic disorder and poor functional outcome. However, while the negative symptom domain has shown to be parsimoniously explained by a 2-factor construct in schizophrenia, there has yet to be an established factor structure of negative symptoms in CHR. METHODS: 214 individuals meeting the Structured Interview for Psychosis-Risk Syndromes (SIPS) criteria for CHR were recruited through 3 active research programs in the United States. Exploratory Factor Analysis was conducted on the 6 negative symptom items of the SIPS, and factors were evaluated with respect to functional outcome and depression. RESULTS: Factor analysis indicated a 2-factor hierarchical model with 2 negative symptom dimensions reflecting volition (Occupational Functioning and Avolition) and emotion (Expression of Emotion, Experience of Emotion and Social Anhedonia). Linear Regression showed that the emotion factor was associated with poor social function, and the volition factor was associated with poor role function and depression. CONCLUSIONS: Similar to factor solutions identified in adults diagnosed with psychotic disorders, results indicated that the SIPS negative symptom subscale is not a unidimensional construct. Rather, the SIPS negative subscale has 2 distinct factors that have different associations with clinical outcome and should be interpreted independently. Results have significant relevance for informing the valid assessment and conceptual interpretation of early clinical phenomenology in the psychosis prodrome.

The personality of U.S. states: Stability from 1999 to 2015.

Researchers have shown an interest in the aggregated Big Five personality of U.S. states, but typically they have relied on scores from a single sample (Rentfrow, Gosling, & Potter, 2008). We examine the replicability of U.S. state personality scores from two studies (Rentfrow et al., 2008; Rentfrow, Gosling, Jokela, & Stillwell, 2013) across a total of seven samples, two of them new. Same-trait correlations across samples are, on average, positive for all five traits, indicating score agreement. Additionally, three traits (Conscientiousness, Neuroticism, and Openness) show strongly consistent patterns of correlations with sociodemographic variables across samples. We find rank order stability in state personality scores for a 16-year period (1999-2015).

A robust data-driven approach identifies four personality types across four large data sets.

Understanding human personality has been a focus for philosophers and scientists for millennia1. It is now widely accepted that there are about five major personality domains that describe the personality profile of an individual2,3. In contrast to personality traits, the existence of personality types remains extremely controversial4. Despite the various purported personality types described in the literature, small sample sizes and the lack of reproducibility across data sets and methods have led to inconclusive results about personality types5,6. Here we develop an alternative approach to the identification of personality types, which we apply to four large data sets comprising more than 1.5 million participants. We find robust evidence for at least four distinct personality types, extending and refining previously suggested typologies. We show that these types appear as a small subset of a much more numerous set of spurious solutions in typical clustering approaches, highlighting principal limitations in the blind application of unsupervised machine learning methods to the analysis of big data.

Velocity Explains the Links between Personality States and Affect.

The present research examined whether perceived rate of progress toward a goal (velocity) mediated the relationships between personality states and affective states. Drawing from control theories of self-regulation, we hypothesized (i) that increased velocity would mediate the association between state extraversion and state positive affect, and (ii) that decreased velocity would mediate the association between state neuroticism and state negative affect. We tested these hypotheses in 2 experience sampling methodology studies that each spanned 2 weeks. Multilevel modeling analyses showed support for each of the bivariate links in our model, and multilevel path analyses supported our mediation hypotheses. We discuss implications for understanding the relations between personality states and affective states, control theories of self-regulation, and goal striving.

Within-subject covariation between depression- and anxiety-related affect.

Studies find a strong positive relationship between the affective components of anxiety and depression. However, most research thus far has examined the between-person correlations among these constructs, while ignoring how changes in these two types of affect covary over time within a person. Within-person correlations could differ meaningfully from how anxiety- and depression-related affect relate across individuals. Further, individuals may differ in terms of how highly these constructs covary over time. The current study aimed to (1) compare between- and within-person correlations between anxious and depressive affect, (2) examine lagged effects between anxious and depressive affect over time, (3) test whether individuals differ in their within-person correlations between these two types of affect, and (4) examine whether the mean level of affective intensity moderated these individual differences. These questions were explored using college undergraduates (N = 50) who rated their depression- and anxiety- related affect six times a day for two weeks. A higher average correlation was observed between anxious and depressive affect in between-person compared to within-person analyses. Significant bidirectional lagged effects were observed between these constructs. Individuals with higher average levels of anxious affect experienced stronger within-person correlations between anxious and depressive affect.

Finding a Life Worth Living: Meaning in Life and Graduation from College.

Graduation from college is an important milestone for young adults, marked by mixed emotions and poignancy, and therefore is an especially salient context for studying meaning in life. The present research used experience-sampling methodology to examine the antecedents and consequences of students' experience of meaning in life over the course of graduation. Participants were 74 graduating students who provided a total of 538 reports over the span of three days, including commencement day. Increased levels of state meaning in life during the days around commencement were linked to spending time with people in general and with family in particular, as well as thinking about one's years in college. Thinking about one's years in college mediated the effects of present company on state meaning in life. Graduates who experienced higher levels of state meaning in life during the days around their commencement ceremony had higher trait levels of meaning in life one week following commencement. We discuss how making meaning of a poignant experience has implications for healthy psychological development.

The Data Box and Within-Subject Analyses: A Comment on Nesselroade and Molenaar (2016).

Nesselroade and Molenaar suggest that it is the relationships between latent variables within subjects that are invariant across subjects and thus the appropriate unit of analysis. We disagree and take the view that between-factor correlations may differ systematically across subjects. Further, individual differences in these correlations may be an important source of information about each unique individual. Following from this premise, analyses of consistencies and differences between subjects of the within-subject pattern of interfactor correlations is a step toward an integrative science of behavior. We give several examples demonstrating how this approach has the potential to yield novel insights into personality functioning.

Measuring Positive Emotion With the Mood and Anxiety Symptom Questionnaire: Psychometric Properties of the Anhedonic Depression Scale.

Low positive emotion distinguishes depression from most types of anxiety. Formative work in this area employed the Anhedonic Depression scale from the Mood and Anxiety Symptom Questionnaire (MASQ-AD), and the MASQ-AD has since become a popular measure of positive emotion, often used independently of the full MASQ. However, two key assumptions about the MASQ-AD-that it should be represented by a total scale score, and that it measures time-variant experiences-have not been adequately tested. The present study factor analyzed MASQ-AD data collected annually over 3 years (n = 618, mean age = 17 years at baseline), and then decomposed its stable and unstable components. The results suggested the data were best represented by a hierarchical structure, and that less than one quarter of the variance in the general factor fluctuated over time. The implications for interpreting past findings from the MASQ-AD, and for conducting future research with the scale, are discussed.

Prospective associations of low positive emotionality with first onsets of depressive and anxiety disorders: Results from a 10-wave latent trait-state modeling study.

Unipolar depressive disorders and anxiety disorders co-occur at high rates and can be difficult to distinguish from one another. Cross-sectional evidence has demonstrated that whereas all these disorders are characterized by high negative emotion, low positive emotion shows specificity in its associations with depressive disorders, social anxiety disorder, and possibly generalized anxiety disorder. However, it remains unknown whether low positive emotionality, a personality trait characterized by the tendency to experience low positive emotion over time, prospectively marks risk for the initial development of these disorders. We aimed to help address this gap. Each year for up to 10 waves, participants (n = 627, mean age = 17 years at baseline) completed self-report measures of mood and personality and a structured clinical interview. A latent trait-state decomposition technique was used to model positive emotionality and related personality traits over the first 3 years of the study. Survival analyses were used to test the prospective associations of low positive emotionality with first onsets of disorders over the subsequent 6-year follow-up among participants with no relevant disorder history. The results showed that low positive emotionality was a risk marker for depressive disorders, social anxiety disorder, and generalized anxiety disorder, although evidence for its specificity to these disorders versus the remaining anxiety disorders was inconclusive. Additional analyses revealed that the risk effects were largely accounted for by the overlap of low positive emotionality with neuroticism. The implications for understanding the role of positive emotionality in depressive disorders and anxiety disorders are discussed.

Affect, Behavior, Cognition, and Desire in the Big Five: An Analysis of Item Content and Structure.

Personality psychology is concerned with affect (A), behavior (B), cognition (C) and desire (D), and personality traits have been defined conceptually as abstractions used to either explain or summarize coherent ABC (and sometimes D) patterns over time and space. However, this conceptual definition of traits has not been reflected in their operationalization, possibly resulting in theoretical and practical limitations to current trait inventories. Thus, the goal of this project was to determine the affective, behavioral, cognitive and desire (ABCD) components of Big-Five personality traits. The first study assessed the ABCD content of items measuring Big-Five traits in order to determine the ABCD composition of traits and identify items measuring relatively high amounts of only one ABCD content. The second study examined the correlational structure of scales constructed from items assessing ABCD content via a large, web-based study. An assessment of Big-Five traits that delineates ABCD components of each trait is presented, and the discussion focuses on how this assessment builds upon current approaches of assessing personality.

A relational structure of voluntary visual-attention abilities.

Many studies have examined attention mechanisms involved in specific behavioral tasks (e.g., search, tracking, distractor inhibition). However, relatively little is known about the relationships among those attention mechanisms. Is there a fundamental attention faculty that makes a person superior or inferior at most types of attention tasks, or do relatively independent processes mediate different attention skills? We focused on individual differences in voluntary visual-attention abilities using a battery of 11 representative tasks. An application of parallel analysis, hierarchical-cluster analysis, and multidimensional scaling to the intertask correlation matrix revealed 4 functional clusters, representing spatiotemporal attention, global attention, transient attention, and sustained attention, organized along 2 dimensions, one contrasting spatiotemporal and global attention and the other contrasting transient and sustained attention. Comparison with the neuroscience literature suggests that the spatiotemporal-global dimension corresponds to the dorsal frontoparietal circuit and the transient-sustained dimension corresponds to the ventral frontoparietal circuit, with distinct subregions mediating the separate clusters within each dimension. We also obtained highly specific patterns of gender difference and of deficits for college students with elevated attention-deficit/hyperactivity disorder traits. These group differences suggest that different mechanisms of voluntary visual attention can be selectively strengthened or weakened based on genetic, experiential, and/or pathological factors.