Testing Differential Item Functioning in Small Samples.

Differential item functioning (DIF) is a pernicious statistical issue that can mask true group differences on a target latent construct. A considerable amount of research has focused on evaluating methods for testing DIF, such as using likelihood ratio tests in item response theory (IRT). Most of this research has focused on the asymptotic properties of DIF testing, in part because many latent variable methods require large samples to obtain stable parameter estimates. Much less research has evaluated these methods in small sample sizes despite the fact that many social and behavioral scientists frequently encounter small samples in practice. In this article, we examine the extent to which model complexity-the number of model parameters estimated simultaneously-affects the recovery of DIF in small samples. We compare three models that vary in complexity: logistic regression with sum scores, the 1-parameter logistic IRT model, and the 2-parameter logistic IRT model. We expected that logistic regression with sum scores and the 1-parameter logistic IRT model would more accurately estimate DIF because these models yielded more stable estimates despite being misspecified. Indeed, a simulation study and empirical example of adolescent substance use show that, even when data are generated from / assumed to be a 2-parameter logistic IRT, using parsimonious models in small samples leads to more powerful tests of DIF while adequately controlling for Type I error. We also provide evidence for minimum sample sizes needed to detect DIF, and we evaluate whether applying corrections for multiple testing is advisable. Finally, we provide recommendations for applied researchers who conduct DIF analyses in small samples.

Has the "M" word been framed? Marijuana, cannabis, and public opinion.

Over the past two decades, a growing cadre of US states has legalized the drug commonly known as "marijuana." But even as more states legalize the drug, proponents of reform have begun to shun the term "marijuana" in favor of the term "cannabis." Arguing that the "M" word has been tainted and may thus dampen public support for legalization, policy advocates have championed "cannabis" as an alternative and more neutral name for the drug. Importantly, however, no one has tested whether calling the drug "cannabis" as opposed to "marijuana" actually has any effect on public opinion. Using an original survey experiment, we examine whether framing the drug as "marijuana" as opposed to "cannabis" shapes public attitudes across a range of related topics: support for legalization of the drug, moral acceptance of its use, tolerance of activities involving the drug, perceptions of the drug's harms, and stereotypes of its users. Throughout each of our tests, we find no evidence to suggest that the public distinguishes between the terms "marijuana" and "cannabis." We conclude with implications of our findings for debates over marijuana/cannabis policy and for framing in policy discourse more generally.

Statistical Analysis for Subjective and Objective Evaluations of Dental Drill Sounds.

The sound produced by a dental air turbine handpiece (dental drill) can markedly influence the sound environment in a dental clinic. Indeed, many patients report that the sound of a dental drill elicits an unpleasant feeling. Although several manufacturers have attempted to reduce the sound pressure levels produced by dental drills during idling based on ISO 14457, the sound emitted by such drills under active drilling conditions may negatively influence the dental clinic sound environment. The physical metrics related to the unpleasant impressions associated with dental drill sounds have not been determined. In the present study, psychological measurements of dental drill sounds were conducted with the aim of facilitating improvement of the sound environment at dental clinics. Specifically, we examined the impressions elicited by the sounds of 12 types of dental drills in idling and drilling conditions using a semantic differential. The analysis revealed that the impressions of dental drill sounds varied considerably between idling and drilling conditions and among the examined drills. This finding suggests that measuring the sound of a dental drill in idling conditions alone may be insufficient for evaluating the effects of the sound. We related the results of the psychological evaluations to those of measurements of the physical metrics of equivalent continuous A-weighted sound pressure levels (LAeq) and sharpness. Factor analysis indicated that impressions of the dental drill sounds consisted of two factors: "metallic and unpleasant" and "powerful". LAeq had a strong relationship with "powerful impression", calculated sharpness was positively related to "metallic impression", and "unpleasant impression" was predicted by the combination of both LAeq and calculated sharpness. The present analyses indicate that, in addition to a reduction in sound pressure level, refining the frequency components of dental drill sounds is important for creating a comfortable sound environment in dental clinics.

A pilot study measuring changes in student impressions before and after clinical training using a questionnaire based on the semantic differential technique.

Students with a positive impression of their studies can become more motivated. This study measured the learning impact of clinical training by comparing student impressions before and after clinical training. The study included 32 students of radiological technology in their final year with the Division of Radiological Science and Technology, Department of Health Sciences, School of Medicine, Hokkaido University. To measure student impressions of x-ray examination training, we developed a questionnaire using the semantic differential technique. The resulting factor analysis identified 2 factors that accounted for 44.9% of the 10 bipolar adjective scales. Factor 1 represented a "resistance" impression of x-ray examination training, and factor 2 represented a "responsibility" impression. The differences in factor scores before and after the clinical training suggest that student impressions are affected by clinical training.

Why don't girls choose technological studies? Adolescents' stereotypes and attitudes towards studies related to medicine or engineering.

Gender differences in choice of studies emerge already in adolescence. Two studies with adolescents are presented, the goal of which is to explore the influence of gender by assessing males and females who choose studies related to Medicine or Engineering. Study 1, correlational (N = 330, mean age 15.9, 56.7% girls), shows that girls who choose technology are more poorly appraised than girls who choose other studies. Study 2 (N = 130; mean age 16.77, 56.2% girls), experimental, measures implicit attitudes (using the IAT) towards males and females from Medicine and Engineering. Implicit attitudes are more favorable towards women if they are studying Medicine and towards men if they study Engineering. The results are analyzed with relation to the percentages of boys and girls in the different fields of study.

The interplay between semantic and referential aspects of anaphoric noun phrase resolution: Evidence from ERPs.

In this event-related brain potential (ERP) study, we examined how semantic and referential aspects of anaphoric noun phrase resolution interact during discourse comprehension. We used a full factorial design that crossed referential ambiguity with semantic incoherence. Ambiguous anaphors elicited a sustained negative shift (Nref effect), and incoherent anaphors elicited an N400 effect. Simultaneously ambiguous and incoherent anaphors elicited an ERP pattern resembling that of the incoherent anaphors. These results suggest that semantic incoherence can preclude readers from engaging in anaphoric inferencing. Furthermore, approximately half of our participants unexpectedly showed common late positive effects to the three types of problematic anaphors. We relate the latter finding to recent accounts of what the P600 might reflect, and to the role of individual differences therein.

Parents' representations of their children: an exploratory study using the Osgood Semantic Differential Scales.

The results of an exploratory national study carried out in Italy using Osgood's Semantic Differential Scales (Osgood, Suci, & Tannenbaum, 1957) with parents of 100 normal 6- to 11-yr.-old children in mainstream classrooms are reported. The aim was to devise a simple tool to be used to explore parents' perceptions of their children in this age group. Parents were asked to rate "my child" using some of Osgood's Semantic Differential Scales especially adapted for this study. Participants were 97 fathers and 100 mothers. Fathers were 32 to 57 years of age (M = 43.2, SD = 5.4), mothers were 29 to 49 years of age (M = 39.7, SD =4.4). In factor analysis four factors were identified: Activity, Evaluation, Emotions evoked by the child, Personality/Physical contact. Average factor scores were significantly different. There were no significant differences between fathers and mothers on the four factor scores. Children were more positively evaluated the younger they were.

Contrast of frequency of positive and negative feeling state changes associated with standard and reduced cardiovascular exercise.

A field investigation was conducted to assess whether cardiovascular exercise assigned at a level to maximize improvements in fitness, i.e., a standard exercise amount or 30 min. at approximately 80% maximal heart rate, differed significantly in frequency of positive exercise-induced feeling changes when contrasted with exercise that was reduced to 20 min. at approximately 65% maximal heart rate. A significantly more frequent occurrence of positive changes in feelings just after exercise (chi1(2), = 4.95, p < .05, phi(c) = .26) was noted for adults initiating programs at the reduced (n = 36; 86% positive change) vs the standard (n = 36; 64% positive change) amount of exercise. After replication, implications for adjusting exercise to increase adherence were discussed.

Entropy and environmental mystery.

Two studies are reported regarding the effects of entropy, lighting, and occlusion on impressions of mystery in physical environments. The theoretical context of this study was the "informational theory" of environmental preference, which, among other claims, holds that mystery can be measured by the extent to which people perceive a promise of more information if they move deeper into an environment. Entropy, in the context of this article, is visual diversity as measured using information theory. Mystery was measured by a semantic differential scale. The definition of mystery was left up to each individual participant. Entropy of occluded objects was used to obtain an objective, experimentally manipulatable and operational definition of "promise of more information." Exp. 1 had 12 stimuli and 15 participants. Exp. 2 had 12 stimuli and 16 participants. Entropy of occluded objects ranged from 0 to 6 bits. Entropy of occluded objects was used to measure the promise that there would be more information if one moved deeper into an environment. Overall, amount of light had the strongest effect on responses of mystery (r = -.63, darker was more mysterious), followed by occlusion (r = .26, occluding objects made a scene seem more mysterious), and by the promise of more information if one moved about in the scene (r = .13), the more entropy in occluded objects, the greater the impression of mystery). The theoretical contribution of this work is that a relationship between subjective impressions of mystery and an objective measure of "promise of more information" was found.

Significado de la obesidad para los adolescentes de Temuco-Chile / The meaning of obesity for adolescents of Temuco-Chile

RESUMEN: La obesidad en la población adolescente ha tenido un alarmante aumento, siendo la dieta uno de los factores que más ha influido en estos cambios. La dieta se caracteriza por una alta ingesta de grasas, azúcares, alimentos procesados, todo esto estimulado por atractivos mensajes publicitarios que aumentan su consumo.Con el propósito de analizar el significado que tiene la obesidad para los adolescente de los colegios de Temuco, Chile, realizamos este trabajo, utilizando la técnica de redes semánticas naturales. Entrevistamos a 372 adolescentes, alumnos de primero y segundo de Enseñanza Media de colegios de la ciudad deTemuco, Chile. Para los adolescentes, las palabras que estaban presentes con alto valor semántico del concepto "obesidad" fueron: gordura, grasa, sobrepeso, discriminación, depresión, enfermedad, ansiedad, problemas, baja autoestima, chatarra y sedentarismo, entre otras. Esta representación demuestra que los alumnos manejan información de las alteraciones que conlleva la obesidad, que se debe estimular la actividad física, la buena alimentación de los adolescentes y que debe haber un manejo integrado de este problema, como también la mantención del peso corporal o provocar la pérdida de peso, dependiendo de la situación de cada uno de ellos.

School children's responses on a semantic differential over a 10-year span.

A semantic differential scale was administered to 208 school children when they were in the second, fourth, sixth, eighth, tenth, and twelfth grades. Their perceptions towards two concepts were measured, Education (going to school) and Work (having a job). Each semantic differential scale had 15 adjective pairs and reflected the three underlying factors of Evaluative, Potency, and Activity. Because the study was conducted for 10 years (ages seven to 18), the changing cognitive developmental stages of the children were expected to influence factor analytic and reliability results. Confirmatory factor analysis, which forced the data into three factors, did not clearly identify the expected three factors, although more items loaded on the three factors with age. An exploratory factor analysis identified a trend across grades from six to four factors over time. Reliability also improved across age groups. Caution should be exercised when using the semantic differential with young children in investigations of abstract concepts.

Analyzing the aesthetic impressions of alexithymic Japanese students.

We investigated the association between alexithymic tendencies as measured by the 20-item Toronto Alexithymia Scale and the characteristics of aesthetic impressions for words and drawings. Impressions were measured using the semantic differential technique in three groups of Japanese students (High scores n = 31, Moderate scores n = 40, and a Control group n = 182) who were presented only words (Word condition), only drawings (Drawing condition), or pairs comprising one word plus one drawing that represented the same emotional categories (Drawing/Word condition). In a factor analysis on the data from participants, three factors (Evaluation, Activity, and Potency) were extracted. Based on the factor scores, the distances among the three conditions were calculated for each of the groups. For Potency, significant group differences were found between the Drawing/Word and other conditions. In the High-scoring alexithymia group, in particular, the Potency impressions based on tactile sensation, e.g., soft-hard, blunt-sharp, were amplified regardless of stimulus condition. These results are discussed in the context of somatosensory amplification associated with alexithymia and the difficulty of distinguishing between emotion and somatic sensations.

Distributed neuronal networks for encoding category-specific semantic information: the mismatch negativity to action words.

Mismatch negativity (MMN), an index of experience-dependent memory traces, was used to investigate the processing of action-related words in the human brain. Responses to auditorily presented movement-related English words were recorded in a non-attend odd-ball protocol using a high-density electroencephalographic (EEG) set-up. MMN was calculated using responses to the same words presented as standard and deviant stimuli in different sessions to avoid contamination from phonetic-acoustic differences. The topography of the mismatch negativity to action words revealed an unusual centro-posterior distribution of the responses, suggesting that activity was at least in part generated posterior to usually observed frontal MMNs. Moreover, responses to hand-related word stimulus (pick) had a more widespread lateral distribution, whereas leg-related stimulus (kick) elicited a more focal dorsal negativity. These differences, remarkably reminiscent of sensorimotor cortex topography, were further assessed using distributed source analysis of the EEG signal (L2 minimum-norm current estimates). The source analysis also confirmed differentially distributed activation for the two stimuli. We suggest that these results indicate activation of distributed neuronal assemblies that function as category-specific memory traces for words and may involve sensorimotor cortical structures for encoding action words.

Processing of semantic categorical and associative relations: an ERP mapping study.

Brain imaging studies have repeatedly shown a dominant involvement of the left hemisphere (LH) in the processing of semantic information. However, some behavioural studies have suggested that the right hemisphere (RH) also processes lexico-semantic information although in a qualitatively different way. More specifically, it has been proposed that the LH shows a greater sensitivity to categorical (CR, or intraconceptual) relationships, while the RH appears more sensitive to associative (AR, or interconceptual) ones. Using semantically unrelated and related word pairs comprised of CR and AR, our aim in this study was to verify the involvement of the RH in the processing of associative relationships. Strength-independent topographical analysis of ERP map series using temporal segmentation procedures showed the same sequence of eight stable map segments in CR and AR conditions. Statistical analysis revealed no difference between conditions either in terms of duration or in terms of time of occurrence of the segmentation maps. However, field strength comparisons between the two conditions using the global field power measure at each time point revealed higher field strength in CR than in AR condition between 300 and 330 ms. Distributed source localisation showed a similar pattern of activation during this time period, involving predominantly left frontal regions. Nevertheless, the strength of the source found in the left medial frontal gyrus differed between CR than AR conditions, thus explaining the differences observed in field strength. Our data suggest that the processing of CR and AR words depend on the same neural network in the LH.

Differences between noun and verb processing in a minimal phrase context: a semantic priming study using event-related brain potentials.

The aim of the present study was to enforce the priming of either nouns or verbs in order to evoke word-category-specific N400 effects. In two experiments two primes which were either a verb-noun or a noun-noun pair were followed by a semantically related or unrelated target which was a noun or verb, respectively. This target always completed the word sequence to a minimal phrase comprising verb, subject, and object (VNN or NNV triplets). In experiment I subjects judged the semantic relatedness of the target to the two primes, in experiment II subjects first generated an appropriate target of the required word category and then judged the semantic relatedness between self-generated word and target. ERPs were recorded from 124 scalp electrodes. In both experiments verbs and nouns as such evoked reliably distinct ERP topographies between 300 and 800 ms. With verbs in relation to nouns the amplitudes were most often found to be more positive over central to frontal or parietal areas and more negative over occipital and temporo-parietal areas. In contrast, N400 effects proved as topographically invariant for noun and verb targets in both experiments. The results suggest that access to noun and verb representations involves topographically distinct cell assemblies while the N400 effect seems to reflect semantic evaluation and integration processes which are more abstract and independent from a particular word category.

Matching faces for semantic information and names: an event-related brain potentials study.

In face identification, it has been controversial whether or not access to biographical information and to a person's name are mediated by qualitatively different loci. We recorded ERPs while participants saw two successive faces and performed a matching task that either required retrieval of semantic information ("same or different profession?"), or retrieval of the person's name ("same or different number of forename syllables?"). For both tasks, slow ERP activity between the first and the second face was characterized by a prominent right posterior negativity, with the asymmetry being larger for the name than the semantic matching task. ERPs to the second face showed a difference between congruent (matching) and incongruent (mismatching) trials, with more negative ERPs for incongruent trials. In the semantic matching task, these differences were significant between 450 and 550 ms, and resembled an N400, with a maximum negativity over the vertex. In the name matching task, the topography of this congruency effect was qualitatively different from that seen in semantic matching. These findings suggest that different brain substrates mediate the access to semantic and name information.

Medial temporal lobe activation during semantic language processing: fMRI findings in healthy left- and right-handers.

Medial temporal lobe (MTL) areas are well known to serve episodic memory functions; their contribution to semantic memory has been occasionally noticed but not studied in detail. In the present fMRI study, 35 right-handed and 35 left-handed healthy subjects performed a semantic decision paradigm during which subjects heard spoken concrete nouns designating objects and had to decide on whether these objects were available in the supermarket and cost lest then a certain amount of money. The control paradigm consisted of sequences of low and high tones where subjects had to decide whether a sequence contained two high tones. The resulting contrast activation of semantic decision versus tone decision involved neocortical temporal, parietal, and prefrontal areas. Additional significant, bilateral activations in the MTL, the hippocampal formation, and adjacent areas were found. The exact incidence and location of activation was studied in a single-subject analysis for all 70 subjects. At the chosen threshold of P<0.001, 94% of subjects showed activations in the MTL and inferior temporal lobe (ITL). Activations were found along the longitudinal axis of the MTL, including the hippocampal formation and the parahippocampal gyrus. In the ITL, parts of the fusiform and lingual gyri were activated. Activations were similar in right- and left-handers. We conclude from this study that the MTL and parts of the ITL can be added to the areas activated by semantic verbal memory processing.

Category-specific medial temporal lobe activation and the consolidation of semantic memory: evidence from fMRI.

Semantic memory consolidation was studied by comparing medial temporal lobe (MTL) fMRI activation to ANIMAL, IMPLEMENT and ABSTRACT nouns in healthy seniors to that of young adults. Relative to healthy seniors, young adults were predicted to show greater MTL activation for IMPLEMENTS, but not ANIMALS, because the ANIMALS category consists of highly intercorrelated and overlapping features that should require less MTL-mediated binding than IMPLEMENTS over a shorter period of time during concept consolidation. ABSTRACT meanings are context-dependent and do not consist of fixed feature sets. Thus it was predicted that ABSTRACT words would not involve age-related feature binding mediated by the MTL. These predictions were confirmed by the results. Our observations are consistent with the hypothesis that the structure of a category influences the consolidation of knowledge in semantic memory.

The construction of the SEDAS: a new suicide-attitude questionnaire.

OBJECTIVE: Only two scales exist for measuring attitudes towards suicidal behaviour. Because of theoretical and methodological limitations of both scales, the SEDAS was constructed to offer an alternative instrument. METHOD: The semantic differential scale, from which the SEDAS was devised, consisted of 36 items (pairs of adjectives) constructed to measure attitudes towards suicidal behaviour of different actors in various situations. It was administered to 142 Dutch SOS-telephone volunteers for seven different actor/situation sets. RESULTS: Item analysis reduced the number of suitable items to 15. Simultaneous component analysis revealed two dimensions on which the actor/situation sets were scored: health/illness and acceptance/rejection, jointly accounting for 47% of variance. Psychometrics of the instrument in terms of internal consistency (values between 0.70 and 0.86), item-rest correlation and test-retest correlation (between 0.63 and 0.87) are acceptable to good. CONCLUSION: The SEDAS may be of value in population surveys, and facilitate effect evaluation of staff training and suicide prevention programmes.

Perceptions of schizophrenic patients and their therapists: application of the semantic differential technique to evaluate the treatment relationship.

This study is an attempt to evaluate the treatment relationship with schizophrenic patients by examining the patients' and their therapists' perceptions of themselves and each other, which are hypothesized to reflect features of the relationship. One hundred fifty-eight schizophrenic patients and 11 psychiatrists who maintained a supportive relationship with the patients as a therapist estimated their perceptions using the semantic differential (SD) technique with 17 adjective pairs. Eight composite scales with sufficient internal consistency were constructed from the estimations. The interrelationship among the perceptual elements, which was represented by correlation analysis of the composite scale scores, seemed consistent with our clinical experience. A factor-analytic study of the scales yielded 3 orthogonal factors that could be assumed to characterize the treatment relationship. The patient-therapist cooperation factor indicated the degree of trust between the two participants, supposedly the affective or relational aspect of the therapeutic alliance. The therapist passivity factor reflects the therapist's passive role-taking and the clinical stability of the patient. The patient strength factor was related to the condition-related and characterological strength of the patient. It is demonstrated that the estimations performed by patients and therapists are valid and useful for evaluation of the treatment relationship in the current status.