Biases in multilevel analyses caused by cluster-specific fixed-effects imputation

When datasets are affected by nonresponse, imputation of the missing values is a viable solution. However, most imputation routines implemented in commonly used statistical software packages do not accommodate multilevel models that are popular in education research and other settings involving clustering of units. A common strategy to take the hierarchical structure of the data into account is to include cluster-specific fixed effects in the imputation model. Still, this ad hoc approach has never been compared analytically to the congenial multilevel imputation in a random slopes setting. In this paper, we evaluate the impact of the cluster-specific fixed-effects imputation model on multilevel inference. We show analytically that the cluster-specific fixed-effects imputation strategy will generally bias inferences obtained from random coefficient models. The bias of random-effects variances and global fixed-effects confidence intervals depends on the cluster size, the relation of within- and between-cluster variance, and the missing data mechanism. We illustrate the negative implications of cluster-specific fixed-effects imputation using simulation studies and an application based on data from the National Educational Panel Study (NEPS) in Germany.     

CFS MATLAB toolbox: An experiment builder for continuous flash suppression (CFS) task

CFS toolbox is an open-source collection of MATLAB functions that utilizes PsychToolbox-3 (PTB-3). It is designed to allow a researcher to create and run continuous flash suppression experiments using a variety of experimental parameters (i.e., stimulus types and locations, noise characteristics, and experiment window settings). In a CFS experiment, one of the eyes at a time is presented with a dynamically changing noise pattern, while the other eye is concurrently presented with a static target stimulus, such as a Gabor patch. Due to the strong interocular suppression created by the dominant noise pattern mask, the target stimulus is rendered invisible for an extended duration. Very little knowledge of MATLAB is required for using the toolbox; experiments are generated by modifying csv files with the required parameters, and result data are output to text files for further analysis. The open-source code is available on the project page under a Creative Commons License (http://www.mikkonuutinen.arkku.net/CFS_toolbox/ and https://bitbucket.org/mikkonuutinen/cfs_toolbox).     

Examining the Cross-National Measurement Invariance of the Extended Satisfaction with Life Scale in the United States and Hungary

The measurement invariance properties of four domains from the Extended Satisfaction with Life Scale (ESWLS) were examined in samples of young adults from the United States and Hungary. The four domains (scales) were general life satisfaction, social life satisfaction, family life satisfaction, and satisfaction with one’s self. For both countries, confirmatory factor analyses supported a correlated four-factor model, with each ESWLS domain constituting a single independent factor. Multigroup invariance testing provided support for both configural invariance and metric invariance, suggesting that across the U.S. and Hungarian samples the four ESWLS scales evidenced the same pattern of factor loadings as well as largely equivalent regression weights. These results suggest that respondents in both countries attributed the same meaning to the ESWLS constructs and that associations between the four ESWLS scales and other variables can be meaningfully compared across the two nations. Latent factor analyses indicated that with the exception of the correlation between general- and self-satisfaction, which was higher for the U.S., the magnitudes of the correlations between ESWLS domains were similar for the U.S. and Hungary. Possible explanations for the lack of measurement invariance across all of the nested models (i.e., item intercepts through residuals) are discussed.     

Remembering successes and failures: rehearsal characteristics influence recollection and distancing

We investigated the relationship between components of rumination, brooding and reflection, and autobiographical remembering by testing whether voluntary and involuntary rehearsal mediated rumination-related variation in the sensory-affective and metacognitive features of memory experience. We focused on achievement and failure memories as both are goal-related events, yet they represent distinct experiences in terms of valence and functionality. For failure memories, brooding was associated with intense recollection and reduced psychological distance. Brooding was related to enhanced distance of achievements, indicating the disruptive effects of brooding on remembering. Although reflection attenuated the recollective experience for both achievement and failure memories, it brought achievement memories to a subjective closer past. Structural equation modelling demonstrated the mediating role of involuntary remembering on the pattern of remembering experience.     

Sympathetic reactivity in late pregnancy is related to labour onset in women

Stress regulation during pregnancy is considered to be connected to the timing of labour initiation. Although increasing knowledge is emerging on the regulation of parturition, there is currently no way to predict the start of spontaneous labour in women. The main aim of this study was to assess pain threshold and the sympathetic nervous system response to cold pain in relation to the onset of labour in healthy pregnant women. Ninety-three pregnant women were recruited and assessed for skin conductance (SC) activity during a cold pressor test in gestational week 38. Pain threshold and cold endurance were also measured and the results were compared with data obtained from hospital records. Seventy-four women had a spontaneous labour onset and a valid SC measurement. SC activity during the cold pressor test decreased significantly with the number of days left to spontaneous parturition. This may indicate a gradual decrease in sympathetic autonomic nervous system reactivity even during the last weeks of pregnancy. Measuring SC activity during mild stress provocation is a rapid and non-invasive means to study variation in sympathetic reactivity during pregnancy, and may be useful in research on stress regulation in pregnancy and its relation to labour initiation.     

Stress, glucocorticoids and liquorice in human pregnancy: programmers of the offspring brain

A suboptimal prenatal environment may induce permanent changes in cells, organs and physiology that alter social, emotional and cognitive functioning, and increase the risk of cardiometabolic and mental disorders in subsequent life ("developmental programming"). Although animal studies have provided a wealth of data on programming and its mechanisms, including on the role of stress and its glucocorticoid mediators, empirical evidence of these mechanisms in humans is still scanty. We review the existing human evidence on the effects of prenatal maternal stress, anxiety and depression, glucocorticoids and intake of liquorice (which inhibits the placental barrier to maternal glucocorticoids) on offspring developmental outcomes including, for instance, alterations in psychophysiological and neurocognitive functioning and mental health. This work lays the foundations for biomarker discovery and affords opportunities for prevention and interventions to ameliorate adverse outcomes in humans.     

Negative Social Cognitions in Socially Anxious Youth: Distorted Reality or a Kernel of Truth?

We review studies that investigate negative social cognitions of socially anxious youth in relation to two specific domains: interpretation of ambiguous social situations and self-evaluation of social performance, including social skills and nervous behaviors. In this review, we address the question whether socially anxious youth’s negative perceptions are distortions of reality or reflect a kernel of truth as compared to other sources of information including independent adult observers and age peers. Studies key to this question are those that investigate not only the social perceptions themselves but also the social behavior of socially anxious youth. Hence the selection of studies for the review was based on this criterion. From the relevant literature it is, as yet, unclear whether the negative interpretations of ambiguous social situations shown by socially anxious youth are distorted or a reflection of reality. Socially anxious youth’s self-evaluations of social skills appear partly distorted and partly true, depending on the person judging the social skills. In contrast, self-evaluations of nervous behaviors appear distorted. The studies reviewed indicate that research would benefit from including a wider range of perceptions from persons relevant to the socially anxious youth’s daily social environment, not only parents and teachers but also their age peers.