Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling

R. M. Baron and D. A. Kenny (1986; see record 1987-13085-001) provided clarion conceptual and methodological guidelines for testing mediational models with cross-sectional data. Graduating from cross-sectional to longitudinal designs enables researchers to make more rigorous inferences about the causal relations implied by such models. In this transition, misconceptions and erroneous assumptions are the norm. First, we describe some of the questions that arise (and misconceptions that sometimes emerge) in longitudinal tests of mediational models. We also provide a collection of tips for structural equation modeling (SEM) of mediational processes. Finally, we suggest a series of 5 steps when using SEM to test mediational processes in longitudinal designs: testing the measurement model, testing for added components, testing for omitted paths, testing the stationarity assumption, and estimating the mediational effects.     

Picture interpretation and Jungian typology

This study examined a Jungian picture interpretation schema, which utilizes a specific quadrant method. This proposed schema, which is used in training at the C. G. Jung Institute in Zürich, attaches significance to specific areas of a drawing. The upper left of a drawing is associated with the father; the lower left with the unconscious; the lower right with the mother; and the upper right with the future. Three questions served as the foci of this study: Is the schema valid? Do certain types within Jung's personality theory respond better than others to the schema? Does the schema work best when it elicits a strong affective response from the person who created the picture? Subjects took one of two tests: a space domain test, or a colour associative test. The results revealed only two significant effects: a positive association (in the space domain test) between unconscious material with the unconscious quadrant and a negative association (in the colour associative test) of mother material with the mother quadrant. These results suggest that the Jungian quadrant schema used to interpret drawings is invalid and of minimal practical utility. Other methods of picture interpretation are discussed as well as suggestions for future research that would be helpful to this field.     

Investigating the influence of organizational role on perceptions of risk in deep coal mines

The authors replicated O. Ostberg's (1980) study of forestry workers using a range of personnel groups in coal mines. Findings demonstrate a high level of consensus between a range of personnel groups for a set of pictorially depicted behavioral risks, at the level of rank order, but significant differences in terms of relative magnitudes. Magnitudes of perceived risk were found to be greater for those closest to the point of hazard than for members of more geographically and experientially distal groups. Findings are interpreted with reference to experiential influences associated with organizational role. Revealed differences between personnel groups are explained in terms of cognitive availability, specifically effects associated with habituation and familiarity with risk, these sources of perceptual bias appearing to interact with organizational role.     

Maternal correlates of toddler insecure and dependent behavior

The present study was designed to examine the relationship between characteristics of mothers and their toddler's insecure and dependent behavior. The authors studied 254 2-year-old toddlers and their mothers via a structured questionnaire administered to the mothers in their homes. The extent to which insecure and dependent behavior is related to the domains of maternal child rearing, maternal personality traits, parental marital relations, and maternal drug use was assessed. Using Pearson correlations and hierarchical multiple regression analyses, the authors found that the maternal child-rearing and maternal personality domains have a direct effect on the toddlers' insecure and dependent behavior. The maternal child-rearing domain also served as a mediator for the domains of the maternal personality attributes, parents' marital relations, and maternal drug use. There also was evidence suggesting an indirect effect of maternal personality attributes on the toddlers' insecure and dependent behavior, which is mediated by the domain of maternal child-rearing practices. Implications for the prevention of insecure and dependent behavior in toddlers are discussed.     

Event-related potential and behavioral measures of attention in 5-, 7-, and 9-year-olds

In 2 experiments, the development of auditory selective attention in children was assessed. The participants, aged 5, 7, and 9 years, responded to target stimuli in the left ear for 1 series (ignoring the standard stimuli) and in the right ear for the other series. In Experiment 1, event-related potentials (ERPs) evoked by the auditory stimuli were recorded from frontal, central, and parietal sites. The 9-year-olds showed a greater processing negativity (Nd) to the attended channel compared with the 5-year-olds. Both 7- and 9-year-olds showed significantly larger amplitudes for the P3 component of ERPs in the attended vs. ignored condition. Behaviorally, the 5-year-olds made fewer hits and more false alarms than did older children, and the 9-year-olds made significantly more false alarms than did the 7-year-olds. The results of Experiment 2 showed that the detriment in the performance of the 9-year-olds was a result of task parameters. The inability of 5-year-olds to attend selectively appears to involve problems with the inhibition of the processing of irrelevant information and with selection of the correct response.     

Rigidity in children's drawings and its relation with representational change

Four studies tested the application of to the drawing domain. In particular, we tested her claim that young children are inhibited in their attempts at changing their graphic representations (representational change) due to being constrained by the order in which the elements of the representation are drawn (procedural rigidity). The first study required 60 children (4- to 6-year-olds and an older comparison group of 8-year-olds) to make three drawings of a familiar and novel topic. From these drawings each child was measured for procedural rigidity. In a further drawing the child was asked to modify their usual representation of each topic. Regression analyses revealed procedural rigidity levels were not predictive of manipulation performance. A second study, testing 75 4- to 6-year-olds and a third study, testing 30 3- to 4-year-olds, revealed that when young children were specifically asked to manipulate rigid sub-procedures on a familiar topic they were indeed able to do so. Finally, a fourth study (testing 40 5-year-olds and 40 8-year-olds) removed the notational trace in drawing (a possible aid for procedural interruption) but this still produced no evidence of a relation between procedural rigidity and representational change. We suggest how the concept of procedural rigidity might be re-interpreted for the drawing domain so that the RR model can remain as a domain-general theory of cognitive development. We also suggest the development of information processing may be crucial for flexibility in drawing.     

Power in stereotypically masculine domains: a Social Influence Strategy X Stereotype Match model

This work examines the hypothesis that stereotypes of groups to which low-power people belong should influence the perceptions and behavior of powerful people only when those stereotypes are both contextually relevant (e.g., women in masculine domains) and provide information of relevance given powerful people's beliefs about the relation between subordinates and goal attainment. Findings across two studies supported predictions. In a masculine domain, when high-power men were attentive to subordinate weaknesses that may produce thwarts to goal attainment, stereotypes of women defined the contextually relevant shortcomings of women, and stereotype-consistent high-power behaviors ensued. In contrast, when powerful men were attentive to subordinate strengths that may enhance goal strivings, stereotypes of women were uninformative (i.e., did not contain information about relevant strengths); female and male employees were responded to and, in turn, performed and reacted similarly. The implications of these findings for theorizing on the relation between power and stereotyping are discussed.     

Starting at the beginning: an introduction to coefficient alpha and internal consistency

Cronbach's a is the most widely used index of the reliability of a scale. However, its use and interpretation can be subject to a number of errors. This article discusses the historical development of a from other indexes of internal consistency (split-half reliability and Kuder-Richardson 20) and discusses four myths associated with a: (a) that it is a fixed property of the scale, (b) that it measures only the internal consistency of the scale, (c) that higher values are always preferred over lower ones, and (d) that it is restricted to the range of 0 to 1. It provides some recommendations for acceptable values of a in different situations.     

Multisensory temporal order judgments: when two locations are better than one

In Experiment 1, participants were presented with pairs of stimuli (one visual and the other tactile) from the left and/or right of fixation at varying stimulus onset asynchronies and were required to make unspeeded temporal order judgments (TOJs) regarding which modality was presented first. When the participants adopted an uncrossed-hands posture, just noticeable differences (JNDs) were lower (i.e., multisensory TOJs were more precise) when stimuli were presented from different positions, rather than from the same position. This spatial redundancy benefit was reduced when the participants adopted a crossed-hands posture, suggesting a failure to remap visuotactile space appropriately. In Experiment 2, JNDs were also lower when pairs of auditory and visual stimuli were presented from different positions, rather than from the same position. Taken together, these results demonstrate that people can use redundant spatial cues to facilitate their performance on multisensory TOJ tasks and suggest that previous studies may have systematically overestimated the precision with which people can make such judgments. These results highlight the intimate link between spatial and temporal factors in determining our perception of the multimodal objects and events in the world around us.     

Is there a “strength effect” in automatic semantic priming?

According to spreading activation models of automatic priming, highly related associates should yield stronger priming effects than weakly related associates. The strength of relations is usually based on word association norms. However, this strength effect has been found in some studies but not in others. The present study suggests that one factor that might be responsible for this inconsistency is whether the weak associate is the primary or nonprimary response in the word association norms This possibility was explored by comparing priming effects of weak nonprimary and primary associates with those of strong primary associates. Comparable priming effects were found for the strong and weak primary associates whereas the weak nonprimary associates did not yield any priming effects. These results were obtained both in paired (Experiment 1) and single (Experiment 2) presentation priming procedures. Thus, the rank of the associate is an important factor in predicting the magnitude of the priming effect.