A cautionary note on using internal cross validation to select the number of clusters

A highly popular method for examining the stability of a data clustering is to split the data into two parts, cluster the observations in Part A, assign the objects in Part B to their nearest centroid in Part A, and then independently cluster the Part B objects. One then examines how close the two partitions are (say, by the Rand measure). Another proposal is to split the data into k parts, and see how their centroids cluster. By means of synthetic data analyses, we demonstrate that these approaches fail to identify the appropriate number of clusters, particularly as sample size becomes large and the variables exhibit higher correlations.The authors express their thanks to the Sol C. Snider Entrepreneurial Center, Wharton School, for support of this project.     

A cluster randomised controlled trial of an intervention based on the Health Action Process Approach for increasing fruit and vegetable consumption in Iranian adolescents

Objective: To evaluate an intervention programme based on the Health Action Process Approach and designed to increase the intake of fruit and vegetables (F&V) among Iranian adolescents aged 13 to 18.

Design: A randomised controlled trial with three arms examined the short- (1 month) and long-term (6 months) effects of the intervention. There were two intervention groups (one included adolescents only [A group; n = 510]; the second included mothers and adolescents [M + A group; n = 462]) and a control group (n = 483). All participants were recruited from schools.

Main outcome measures: Social cognitions, self-regulatory processes and F&V intake.

Results: The intervention led to an increase in F&V intake for adolescents in the short and long terms. Adolescents in the M + A group increased their F& V intake more than adolescents in the A group. Outcome expectancies, self-monitoring, intentions, action and coping planning, perceived social support and behavioural automaticity mediated the effect of the intervention on F&V intake.

Conclusion: The theory-based intervention led to an increase in F&V intake and promoted more positive social cognitions and self-regulatory processes among Iranian adolescents. The findings also provide evidence that involving mothers in an intervention can confer additional benefit.     

Be a Mom,a Web-Based Intervention to Prevent Postpartum Depression: Results From a Pilot Randomized Controlled Trial

Be a Mom is a self-guided web-based intervention, grounded in cognitive behavioral therapy, delivered to postpartum women to prevent persistent postpartum depression [PPD] symptoms. We aimed to evaluate Be a Mom in terms of its preliminary efficacy, feasibility, and acceptability. A pilot randomized, two-arm controlled trial was conducted. Eligible women (presenting PPD risk-factors and/or early-onset PPD symptoms) were enrolled in the study and were randomly assigned to the intervention (Be a Mom) or to the waiting-list control group. Participants in both groups completed baseline (T1) and postintervention (T2) assessments. The 194 women presenting risk factors/early-onset PPD symptoms were allocated to the intervention (n = 98) or to the control (n = 96) group. A significant Time × Group interaction effect was found for both depressive and anxiety symptoms, with women in the intervention group presenting a larger decrease in symptoms from T1 to T2 (p < .05). Less than half of the women (41.8%) completed Be a Mom. Most women (71.4%) would use Be a Mom again if needed. Results provide preliminary evidence of the Be a Mom’s efficacy, acceptability and feasibility, although further research is needed to establish Be a Mom as a selective/indicative preventive intervention for persistent PPD.     

Evaluating discrete trial teaching with instructive feedback delivered in a dyad arrangement via telehealth

Telehealth research in the field of applied behavior analysis (ABA) has focused on the training and supervision of others to implement various behavior analytic procedures. Current guidelines for practicing telehealth suggest that direct telehealth services may be appropriate for some individuals diagnosed with autism spectrum disorder, but more research is needed. This study evaluated the effects of discrete trial teaching with instructive feedback in a dyad arrangement delivered directly via telehealth to teach tact relations to 6 children diagnosed with autism spectrum disorder. All participants and the experimenter were located in different physical locations. All participants learned their primary and secondary targets, and 5 of the participants acquired the observational primary and secondary targets without direct teaching. Areas of future research and clinical implications are discussed in the context of telehealth service delivery.     

Assessing the impact of a family nurse-led intervention on young mothers’ references to internal states

A mother's propensity to refer to internal states during mother–child interactions is important for her child's developing social understanding. However, adolescent mothers are less likely to reference internal states when interacting with their children. We investigated whether young mothers’ references to internal states are promoted by the Family Nurse Partnership (FNP) intervention, an intensive home-visiting programme designed to support adolescent mothers in England. We also investigated family, maternal, and child factors associated with young mothers’ references to inner states during interactions with their children. Adolescent mothers (n = 483, aged ≤ 19 years when recruited in pregnancy) and their children participated in an observational substudy of a randomized controlled trial investigating the impact of FNP compared to usual care. Mother–child dyads were video-recorded during free play, and mothers’ speech was coded for use of internal state language (references to cognitions, desires, emotions, intentions, preferences, physiology, and perception). We found no differences in mothers’ use of internal state language between the FNP and usual care groups. A sample-wide investigation identified that other features of mothers’ language and relationship status with the child's father were associated with internal state language use. Findings are discussed with reference to targeted interventions and implications for future research.     

Effects of training and display content on Level 2 driving automation interface usability

IntroductionLevel 2 driving automation features, such as adaptive cruise control (ACC) combined with lane centering, primarily communicate their operating statuses through the instrument cluster. It remains unclear how interface-specific training and display content influence the ability to understand Level 2 activity in production vehicles.MethodsEighty participants viewed videos recorded from the driver’s point of view in a variety of driving scenarios with Level 2 activity displayed in the instrument cluster of a 2017 Mercedes-Benz E-Class. Half viewed one of two instrument cluster layouts (simple or complex), and half received an orientation to the interface before the experiment. After each video, they were asked about the scenario they had just seen. We then examined what information in the displays participants used to identify Level 2 activity and the usability of the displays.ResultsTraining improved recognition accuracy of when lane centering was temporarily inactive and understanding of why it was inactive. Neither training nor display content affected the ability to recognize when ACC had adjusted the vehicle’s speed or detected a vehicle ahead, or when ACC initially did not detect a lead vehicle and understanding of why it had not detected it. Both factors influenced which sources of information participants used to determine Level 2 activity. Recognition accuracy of system activity improved when participants used valid sources of information. Training, but not display content, improved lane centering usability, but not ACC usability.DiscussionBasic training improves detection of notifications that potentially require further driver action, but not of those that display persistent status information. Training does not result in full understanding of all system notifications or functional limitations, which reinforces the need for intuitive, salient communication about system behavior and its limitations.     

Distance stability analysis in multidimensional scaling using the jackknife method

Stability or sensitivity analysis is an important topic in data analysis that has received little attention in the application of multidimensional scaling (MDS), for which the only available approaches are given in terms of a coordinate‐based analytical jackknife methodology. Although in MDS the prime interest is in assessing the stability of the points in the configuration, this methodology may be influenced by imprecisions resulting from the inherently necessary Procrustes method. This paper proposes an analytical distance‐based jackknife procedure to study stability and cross‐validation in MDS in terms of the jackknife distances, which is not influenced by the Procrustes method. For each object, the corresponding jackknife estimated points are considered as naturally clustered points, and stability and cross‐validation are analysed in terms of the MDS distances arising from the jackknife procedure, on the basis of a weighted cluster‐MDS algorithm. A jackknife‐relevant configuration is also proposed for cross‐validation in terms of coordinates, in a cluster‐MDS framework.     

Comorbidity and Internet-Delivered Transdiagnostic Cognitive Behavioural Therapy for Anxiety Disorders

Internet-delivered transdiagnostic anxiety interventions aim to reduce symptoms across several anxiety disorders using one treatment protocol. However, it is unclear whether comorbidity affects outcomes of such treatment. This study re-examined data from a recent randomised controlled trial (N = 129) that evaluated the efficacy of an Internet-delivered transdiagnostic cognitive behavioural therapy (iCBT) intervention for participants with principal diagnoses of generalised anxiety disorder (GAD), social phobia (SP) panic disorder and agoraphobia (PDA), of whom 72% met criteria for a comorbid anxiety disorder or depression. Participants were divided into two groups based on whether or not they had a comorbid disorder before treatment. Participants with comorbid conditions reported higher symptom levels at pre-treatment, post-treatment, and follow-up across a range of measures. Both groups showed significant reductions in symptoms over treatment; however, participants with comorbid disorders showed greater reductions in measures of GAD, PDA, SP, depression, and neuroticism. In addition, treatment significantly reduced the number of comorbid diagnoses at follow-up. These results indicate transdiagnostic iCBT protocols have the potential to reduce comorbidity.