Search

In this commentary, we discuss the proposed criteria in Gärtner et al. (2022) for hiring or promoting quantitative methods researchers. We argue that the criteria do not reflect aspects that are relevant to methods researchers and typical publications they produce. We introduce a new set of criteria that can be used to evaluate the performance of methods researchers in a more valid fashion. We discuss the necessity to balance scientific expertise and open science commitment in such ranking schemes.

Rationale Recently, experimental paradigms have been developed to strengthen automatic avoidance or inhibitory responses for smoking cues. However, these procedures have not yet been directly compared regarding their effectiveness and mechanisms of action. Objective This study compared the effects of avoidance vs. inhibitory training as an add-on to a brief smoking cessation intervention. The standard Approach-Avoidance-Task (AAT) was adapted for both training types and control conditions. Methods One hundred twenty-four smokers attended behavioral counseling for smoking cessation and were thereafter randomized to one of four training conditions: avoidance-AAT, sham-avoidance-AAT, inhibition-AAT, sham-inhibition-AAT. During a 2-week training period including five training sessions, smokers in the avoidance-AAT trained to implicitly avoid all smoking-related cues, while smokers in the inhibition-AAT trained to implicitly inhibit behavioral response to smoking cues. During sham training, no such contingencies appeared. Self-report and behavioral data were assessed before and after training. Cigarette smoking and nicotine dependence were also assessed at 4- and 12-week follow-ups. Results At posttest, avoidance training was more effective in reducing daily smoking than inhibition training. However, this difference was no longer evident in follow-up assessments. All training conditions improved other smoking- and health-related outcomes. Neither training changed smoking-related approach biases or associations, but approach biases for smoking-unrelated pictures increased and Stroop interference decreased in all conditions. Smoking devaluation was also comparable in all groups. Conclusions Avoidance training might be slightly more effective in reducing smoking than inhibitory training. Overall, however, all four training types yielded equivalent therapy and training effects. Hence, a clear preference for one type of training remains premature.

Conclusions about the individual development of delinquent behaviours during the life-course are often made by repeatedly interviewing the same respondents (i.e. panel data). Missing data, especially unit nonresponse and panel attrition are often a problem for the analysis of panel data, as they pose a threat to the validity of statistical inferences. Multiple imputation (MI) is a standard state-of-the-art technique to address these problems. However, until very recently, MI methods to impute highly skewed, zero-inflated and repeatedly measured count data such as the count of delinquent behaviours per year, were not available. Solutions that were often applied included data transformations and rounding, so that available MI methods, usually based on the multivariate normal model, could be applied. This approach has also been used by Reinecke and Weins (Qual Quant 47(6):3319–3334, 2013), who analysed delinquency data from an adolescents’ four-wave panel. Recent missing data research, however, suggests that these “normalizing” practices could be problematic and that imputation models with implausible distributional assumptions should generally be avoided when the empirical data depart too heavily from these assumptions. In the present paper, we re-analyse the data from Reinecke and Weins (2013) using MI models, where parametric assumptions of the imputation model are compatible to the subsequent analysis model (a growth curve model for zero-inflated count data). Results show that the chance of reporting zero punishable offences decreases over time, with a turning point at around the age of 15. Likewise, the versatility of delinquent behaviours increases early on in adolescence, and decreases later (reflecting the typical age-crime curve). Boys and students from the bottom-level branch of the German educational system exhibit a higher versatility in delinquent activity. A comparison of present results with the original ones from Reinecke and Weins (2013) corroborates recommendations e.g. by Yu et al. (Stat Methods Med Res 16(3):243–258, 2007) to opt for missing data methods with fitting distributional assumptions.

Cognitive bias research draws upon the notion that altered information processing is key for understanding psychological functioning and well-being. However, little attention has been paid to the question of whether the frequently used experimental paradigms hold adequate psychometric properties. The present study examined the psychometric properties of three widely used cognitive bias tasks: the Approach-Avoidance Task (AAT), the visual dot-probe-task, and the Implicit Association Test (IAT). Approach, attentional, and association biases towards valenced stimuli were repeatedly measured at five different time points in a sample of 79 healthy young adults. Two different devices were used for assessment: a personal computer (PC) and a touchscreen-based tablet. Reliability estimates included internal consistency and temporal stability. Validity was inferred from convergence across different behavioral tasks and correlations between bias scores and self-reported psychological traits. Reliability ranged widely amongst tasks, assessment devices, and measurement time points. While the dot-probe-task appeared to be completely unreliable, bias scores obtained from the PC-based version of the AAT and both (PC and touchscreen) versions of the IAT showed moderate reliability. Almost no associations were found across information processing tasks or between implicit and explicit measures. Cognitive bias research should adopt a standard practice to routinely estimate and report psychometric properties of experimental paradigms, investigate feasible ways to develop more reliable tools, and use tasks that are suitable to answer the precise research question asked.

Automatic approach biases toward smoking-related cues have been implicated in the development and maintenance of addictive behaviors. Studies aiming at modifying such biases have shown promise in changing maladaptive approach tendencies for smoking cues and reducing smoking behavior. However, training effects tend to be small and partly inconsistent. The present randomized-controlled trial incorporated virtual reality (VR) technology into Approach Bias Modification (ABM) to improve efficacy. One-hundred-eight smokers attended behavioral counseling for smoking cessation and were thereafter randomized to receive VR-ABM or VR-control training. During VR-ABM, participants trained to implicitly avoid smoking-related objects and to approach alternative objects, while no such contingency existed in the VR-control condition. Trainings were administered in six sessions within a two-week period. Assessments were conducted at baseline, post-intervention (three weeks after baseline), and at follow-up (seven weeks after baseline). VR-ABM did not change approach biases, nor other cognitive biases, but it was superior in reducing daily smoking. However, this effect was limited to the two-week training period. Both groups improved in other smoking- and health-related variables across time. Future work should continue to investigate working mechanisms of ABM, in particular crucial training ingredients. VR could prove valuable for public health as the potential of VR-based treatments is large and not fully explored.

In this chapter the concept of a missing data mechanism and the classification of missing data as being missing completely at random (MCAR), missing at random (MAR) or missing not at random (MNAR) is introduced. Consequences of different possible missing data mechanisms are considered and illustrated with simple examples. In addition, conditions are given for analysing data sets without the need to explicitly model the missing data mechanism (“ignorability”). We also review diagnostic tools for incomplete data sets, both descriptive and based on a statistical test.

In this chapter, we provide an overview over the theory of multiple imputation. Topics covered are justification of multiple imputation strategies based on monotone and non-monotone missing patterns, the approach based on joint modeling of all variables with missing values and the fully conditional modeling approach, where only univariate marginal models are used to generate imputations. Additional topics are rounding, how to deal with restrictions and how to treat interaction or higher polynomial terms.

In this chapter missing data procedures and techniques are reviewed and discussed. Among them are both, ad-hoc methods but also more sophisticated techniques including maximum likelihood estimation, weighting and imputation. We discuss pros and cons of the different approaches and techniques, and give practical advice which procedure might be suited best in a given scenario because valid inferences in applied research can only be expected based on informed decisions. A conclusion of this chapter will be that there is not the one method or technique that works best under every possible scenario.

The purpose of this chapter is to characterize and describe missing data situations, to introduce terminology used in this field, e.g., the terms item nonresponse, unit nonresponse and attrition, and review the most common statistical approaches to inference from samples to populations. Missing data patterns (univariate, monotone, arbitrary) are described as far as they are relevant for choosing an appropriate technique to compensate for missing data. Since the scope of the method of multiple imputation is rather general, those aspects of the most common statistical approaches relevant to understanding when and how multiple imputation works are reviewed. The different statistical approaches and the potential usefulness of methods to compensate for missing data are demonstrated through simple examples and illustrations.

In this chapter, we discuss new developments in the field of multiple imputation. Our focus will be on non-standard missing data problems, i.e., we focus on situations, where distributional assumptions of standard MI approaches (for example, multivariate normality or homoscedasticity) are violated, and outline solutions to these problems.

In this chapter, we discuss the most important and most commonly used multiple imputation tools in R (Table 5.1 gives an overview of the download frequencies of various MI packages in R.) for both multivariate and clustered data sets, including packages mice, norm2, Amelia, mi, pan, as well as function aregImpute( ) from package Hmisc, and show practical applications. We give hands-on step by step tutorials regarding how to carry out MI in practice.

Background: The latest version of the International Classification of Diseases (ICD-11) proposes a posttraumatic stress disorder (PTSD) diagnosis reduced to its core symptoms within the symptom clusters re-experiencing, avoidance and hyperarousal. Since children and adolescents often show a variety of internalizing and externalizing symptoms in the aftermath of traumatic events, the question arises whether such a conceptualization of the PTSD diagnosis is supported in children and adolescents. Furthermore, although dysfunctional posttraumatic cognitions (PTCs) appear to play an important role in the development and persistence of PTSD in children and adolescents, their function within diagnostic frameworks requires clarification. Methods: We compiled a large international data set of 2,313 children and adolescents aged 6 to 18 years exposed to trauma and calculated a network model including dysfunctional PTCs, PTSD core symptoms and depression symptoms. Central items and relations between constructs were investigated. Results: The PTSD re-experiencing symptoms strong or overwhelming emotions and strong physical sensations and the depression symptom difficulty concentrating emerged as most central. Items from the same construct were more strongly connected with each other than with items from the other constructs. Dysfunctional PTCs were not more strongly connected to core PTSD symptoms than to depression symptoms. Conclusions: Our findings provide support that a PTSD diagnosis reduced to its core symptoms could help to disentangle PTSD, depression and dysfunctional PTCs. Using longitudinal data and complementing between-subject with within-subject analyses might provide further insight into the relationship between dysfunctional PTCs, PTSD and depression.

Mindfulness meditation yields beneficial effects on the processing of emotions. However, it is still unclear whether the focus of attention during meditation influences these effects. In the present study we aimed at comparing the effects of breathing meditation and emotion-focused meditation on the immediate and delayed processing of negative and positive emotions. The study included 65 adult novice meditators who were exposed to positively and negatively valenced film clips. Participants were randomly assigned to three conditions. While watching the films at t1, they were asked to mindfully focus on their breath (condition 1), on emotions (condition 2), or on nothing in particular (condition 3). Ten minutes later at t2, comparable film clips were shown but all participants watched them without taking up a mindful attitude. Dependent measures were emotional states at t1 and t2. Participants of both meditation conditions particularly showed a more preferable delayed emotional reaction to negative stimuli than participants of the control condition. Breathing meditation and emotion-focused meditation may constitute effective emotion regulation strategies to deal with negatively valenced emotional states.

Abstract. Predictive mean matching (PMM) is a state-of-the-art hot deck multiple imputation (MI) procedure. The quality of its results depends, inter alia, on the availability of suitable donor cases. Applying PMM in small sample scenarios often found in psychological or medical research could be problematic, as there might not be many (or any) suitable donor cases in the data set. So far, there has not been any systematic research that examined the performance of PMM, when sample size is small. The present study evaluated PMM in various multiple regression scenarios, where sample size, missing data percentages, the size of the regression coefficients, and PMM’s donor selection strategy were systematically varied. Results show that PMM could be used in most scenarios, however results depended on the donor selection strategy: overall, PMM using either automatic distance-aided selection of donors ( Gaffert, Meinfelder, & Bosch, 2016 ) or using the nearest neighbor produced the best results.

Empirical count data are often zero-inflated and overdispersed. Currently, there is no software package that allows adequate imputation of these data. We present multiple-imputation routines for these kinds of count data based on a Bayesian regression approach or alternatively based on a bootstrap approach that work as add-ons for the popular multiple imputation by chained equations (mice) software in R (van Buuren and Groothuis-Oudshoorn, Journal of Statistical Software, vol. 45, 2011, p. 1). We demonstrate in a Monte Carlo simulation that our procedures are superior to currently available count data procedures. It is emphasized that thorough modeling is essential to obtain plausible imputations and that model mis-specifications can bias parameter estimates and standard errors quite noticeably. Finally, the strengths and limitations of our procedures are discussed, and fruitful avenues for future theory and software development are outlined.

Predictive mean matching (PMM) is a standard technique for the imputation of incomplete continuous data. PMM imputes an actual observed value, whose predicted value is among a set of k ≥ 1 values (the so-called donor pool), which are closest to the one predicted for the missing case. PMM is usually better able to preserve the original distribution of the empirical data than fully parametric multiple imputation (MI) approaches, when empirical data deviate from their distributional assumptions. Use of PMM is therefore especially worthwhile in situations where model assumptions of fully parametric MI procedures are violated and where fully parametric procedures would yield highly implausible estimates. Unfortunately, today there are only a handful of studies that systematically tested the robustness of PMM and it is still widely unknown where exactly the limits of this procedure lie. I examined the performance of PMM in situations where data were skewed to varying degrees, under different sample sizes, missing data percentages, and using different settings of the PMM approach. It was found that small donor pools overall yielded better results than large donor pools and that PMM generally worked well, unless data were highly skewed and more than about 20% to 30% of the data had to be imputed. Also, PMM generally performed better when sample size was sufficiently large.

We find that existing multiple imputation procedures that are currently implemented in major statistical packages and that are available to the wide majority of data analysts are limited with regard to handling incomplete panel data. We review various missing data methods that we deem useful for the analysis of incomplete panel data and discuss, how some of the shortcomings of existing procedures can be overcome. In a simulation study based on real panel data, we illustrate these procedures’ quality and outline fruitful avenues of future research.