Search

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.

There are three fundamental problems in Sijtsma (Psychometrika, 2008): (1) contrary to the name, the glb is not the greatest lower bound of reliability but rather is systematically less than ωt (McDonald, Test theory: A unified treatment, Erlbaum, Hillsdale, 1999), (2) we agree with Sijtsma that when considering how well a test measures one concept, α is not appropriate, but recommend ωt rather than the glb, and (3) the end user needs procedures that are readily available in open source software.

Within the debate on the structure of affect, a consensus began emerging in the last decade regarding the bipolarity of happiness–sadness. We argue that this consensus is premature. Focusing on the psychometrics of momentary affect, particularly happiness and sadness, and using a simulation study, a large-scale data set, and 2 experiments manipulating affect, we plot a map of affective space that departs from the consensus. One key departure is the finding that happiness and sadness are not bipolar opposites. Another is that nonuniform skewness plays a major role in studies of affective structure, but can be addressed with appropriate analyses.

We make theoretical comparisons among five coefficients—Cronbach’s α, Revelle’s β, McDonald’s ω h , and two alternative conceptualizations of reliability. Though many end users and psychometricians alike may not distinguish among these five coefficients, we demonstrate formally their nonequivalence. Specifically, whereas there are conditions under which α, β, and ω h are equivalent to each other and to one of the two conceptualizations of reliability considered here, we show that equality with this conceptualization of reliability and between α and ω h holds only under a highly restrictive set of conditions and that the conditions under which β equals ω h are only somewhat more general. The nonequivalence of α, β, and ω h suggests that important information about the psychometric properties of a scale may be missing when scale developers and users only report α as is almost always the case

This 2-year quasi-experiment evaluated the effect of peer-led workshop groups on performance of minority and majority undergraduate biology students. The workshop intervention used was modeled after a program pioneered by Treisman (1992). Majority volunteers randomly assigned to workshops (nD 61) performed significantly better than those assigned to the control group (nD 60, p< 0:05) without spending more time studying. Workshop minority students (nD 25) showed a pattern of increasing exam performance in comparison to historic control minority students (nD 21), who showed a decreasing pattern ( p< 0:05). Volunteers (nD 121) initially reported that biology was more interesting and more important to their futures than to nonvolunteers’ (nD 435, p< 0:05). Volunteers also reported higher levels of anxiety related to class performance ( p< 0:05). The relationship of anxiety to performance was moderated by volunteer status. Performance of volunteers was negatively associated with self-reported anxiety (rDi 0:41, p< 0:01). Performance of nonvolunteers was unrelated to self-reported anxiety (rDi 0:02). Results suggest elevated anxiety related to class performance may increase willingness to participate in activities such as workshop interventions. In addition, students who volunteer for interventions such as workshops may be at increased risk of performance decrements associated with anxiety. Even so, workshop programs appear to be an effective way to promote excellence among both majority and minority students who volunteer to participate, despite the increased risk of underperformance associated with higher levels of anxiety.

Research using the emotion Stroop task has established that individuals with various emotional disorders exhibit increased interference for stimuli specifically related to their disorder due to the concern relevance and negativity of these stimuli. Our study expands this research to normal populations. We examined the influence of emotionality, mood congruence, and concern relevance following experimental mood-induction procedures (MIPs) on emotion Stroop performance of college students. Participants completed a computerized emotion Stroop task following positive and negative MIPs. Results supported the mood congruence and concern relevance but not the emotionality hypotheses. The implication of these findings for theories of emotional breakdown and the importance of studying the idiographic aspects of affective experience are discussed.

Research has generally supported the existence of familial–genetic factors for male sexual orientation, but has not shed much light on the specific nature of those influences. Gay men with gay brothers provide the opportunity to examine several hypotheses. Sixty-six men, representing 37 gay male sibling pairs, completed questionnaires assessing behavior on various measures including childhood and adult gender nonconformity, timing of awareness of homosexual feelings, self-acceptance, and the quality of family relationships. Consistent with prior findings using twins, gay brothers were similar in their degree of childhood gender nonconformity, suggesting that this variable may distinguish etiologically (e.g., genetically) heterogeneous subtypes. The large majority of gay men with brothers knew about their own homosexual feelings before they learned about their brothers' homosexual feelings, suggesting that discovery of brothers' homosexuality is not an important cause of male homosexuality.

Two explanations for the Yerkes-Dodson Effect (1908) were tested. Easterbrook (1959) proposed that arousal reduces the range of cue utilization. Thus, arousal should interfere with the capacity for simultaneous (dual) memory scans. In contrast, Humphreys and Revelle (1984) proposed that arousal facilitates sustained information transfer but interferes with short-term memory. Arousal should thus reduce the time needed to prepare to respond but increase the time needed to scan memory. Either caffeine or placebo was given to 78 subjects who differed in impulsivity. They completed three versions of a memory-scanning task: two single-task versions (physical and category matches) and one dual-task version (either type of match). As predicted by Humphreys and Revelle, relative to placebo, caffeine lowered the intercept (p

The Multifactor Racial Attitude Inventory (MRAI), a major 10-factor measurement of Whites’ attitude toward Blacks, was reanalyzed by using a new hierarchical cluster technique to locate factors with greater unidimensionality. 4 clusters emerged (Public, Private, Black Superiority, and Gradualism) with reliabilities between .90 and .66. The mean intercorrelation between the 4 new clusters was .23, contrasted to .54 for the 10 MRAI factors. All 4 of the new clusters met a 39% or less overlap criterion, whereas only 3 of the original 10 factors met this criterion. The 4-scale solution produced subscales with markedly greater independence than existed for the subscales in the 10-factor solution.

Several recent studies have attempted to identify theinformation-processing components involved in analo­gical reasoning. Some have focused specifically on geo­metric analogies, in which terms of the analogy aregeometric shapes that can be related to one another bytheir degrees of rotation, the number of shapes in eachterm, the number of sides forming each shape, and otherproperties of the figures. Under current information­processing models of analogical reasoning, each anal­ogy term is seen as having defining characteristics orattribute dimensions, and each takes on specific val­ues along these dimensions. For example, in the ver­bal analogy FATHER:SON: :MOTHER..., FATHERcan be represented as a set of attribute values