CKChris van Klaveren

Chris van Klaveren

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Chris van Klaveren has a total of 108 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Long-run impacts of school-based mentoring for at-risk youth: evidence from a randomized field experiment

    by Nienke Ruijs, Ilja Cornelisz, Chris van Klaveren, Marc van der Steeg, Dinand Webbink

  • Open Access

    A regularized approach to the synthetic control method: an empirical evaluation

    by Tom Stolp, Ilja Cornelisz, Chris van Klaveren

    The synthetic control method (SCM) estimates the treatment effect for a single treated unit by constructing a counterfactual outcome as a weighted average of the outcomes from untreated donor units. SCM requires that the estimated weights be greater than zero and sum to one. However, these constraints can be suboptimal if the treated unit's time series exhibits consistently higher or lower values or variance, or opposite trends relative to the donor units. To address this limitation, we systematically evaluate a regularized version of SCM (RSCM), allowing weights to take on any value. Simulation results show that SCM and RSCM perform equally well when outcomes of the treated unit are located inside the outcome set of donor units. However, when treated unit outcomes are at the boundary, RSCM outperforms SCM. Finally, placebo tests in two empirical applications confirm that RSCM accurately predicts employment levels for different labour market sectors.

  • Open Access

    Augmented-Reality Cognitive Behavior Therapy for Contamination-Related Obsessive-Compulsive Symptoms: A Protocol for a Multicenter Randomized Controlled Trial (Preprint)

    by Andrej Skoko, Jón Ingi Hlynsson, Marketa Ciharova, Thomas Berger, Per Carlbring, Tobias Krieger, Jan Bergström, Tom Van Daele, Chris van Klaveren, Ilja Cornelisz, et al.

  • Open Access

    Augmented-Reality Cognitive Behavior Therapy for Fear of Contamination in Obsessive-Compulsive Disorder: A Protocol for a Multicenter Randomized Controlled Trial (Preprint)

    by Andrej Skoko, Jón Ingi Hlynsson, Marketa Ciharova, Thomas Berger, Per Carlbring, Tobias Krieger, Jan Bergstörm, Tom Van Daele, Chris van Klaveren, Ilja Cornelisz, et al.

    BACKGROUND Obsessive-compulsive disorder (OCD) affects a substantial proportion of society and is associated with high economic burden. Despite the availability of effective treatments, many individuals do not receive evidence-based treatments due to barriers to accessibility, affordability, and scalability. Augmented Reality Exposure Therapy (ARET) has emerged as a promising alternative to conventional face-to-face therapy. OBJECTIVE We aim to evaluate an AR-based cognitive-behavioral self-help intervention for contamination-related fear, ZeroOCD. Specifically, we will assess whether ZeroOCD reduces the severity of contamination fear at post-intervention and at the 6-month follow-up, as well as its cost-effectiveness and user-friendliness compared with cognitive behavior therapy (CBT) delivered via videoconferencing and a waitlist control condition. METHODS A three-arm randomized controlled trial (RCT) will be conducted, comparing: (1) ZeroOCD with minimal therapist guidance, (2) CBT delivered via videoconferencing, and (3) a waitlist control condition. A total of 189 participants experiencing fear of contamination will be recruited from the Dutch (n = 63), German (n = 63), and Swedish-speaking (n = 63) general population. RESULTS Recruitment and data collection began in September 2025 and are expected to continue through 2026. CONCLUSIONS This study aims to improve access to evidence-based treatment for OCD by evaluating a scalable, AR-based self-help intervention with minimal therapist involvement. CLINICALTRIAL ClinicalTrials.gov NCT07194473; https://clinicaltrials.gov/study/NCT07194473

  • Open Access

    Predicting Adolescent Risk-Taking Behavior: A Comparative Study of Theory-Based vs. Data-Driven Variable Selection for Predictive Analysis

    by Tatvan Todor, Ilja Cornelisz, Chris Van Klaveren, Barbara Braams

    Adolescence is a developmental period characterized by an increase in risk-taking behavior. To understand the complex interactions between biological, cognitive, and social factors underlying risk-taking, research has largely focused on explanatory approaches and cross-sectional designs. While this approach is powerful for testing hypotheses it may overlook other, non-hypothesized, determinants of risk-taking. Predictive modelling offers a complementary, discovery-oriented approach. By optimising out-of-sample prediction over a wide range of factors, it can reveal influential predictors that may otherwise be overlooked.Here longitudinal data was used from 218 adolescents collected across three measurement waves. We used a Random Forest (RF) approach to predict the self-reported likelihood of risk-taking behavior two and four years after the initial assessment. To provide a baseline benchmark for evaluating predictive performance, we also estimated a traditional Ordinary Least Squares (OLS) regression model.Results showed superior predictive accuracy for the RF model compared to the OLS model. RF highlighted biological and hormonal factors (e.g., testosterone, estradiol, DHEA, and pubertal development) as the strongest predictors. An exploratory clustering step within the RF model, using a single decision tree, further illustrated how these predictors combine into distinct profiles, pointing toward potential mechanisms of adolescent risk-taking.These findings underscore the added value of data-driven methods in identifying predictors of adolescent risk-taking and demonstrate the potential of RF to reveal influential factors beyond those suggested by theory.

  • Open Access

    Moderators of Parenting Programme Effects on Child Behavioural Difficulties: An Individual Participant Data Meta-Analysis Using Machine Learning

    by Constantina Psyllou, Chris Van Klaveren, Marjolein Luman, Patty Leijten, Tycho J. Dekkers, Saskia Van der Oord, Barbara J. van Den Hoofdakker, Annabeth P. Groenman

    Background: Behavioural parenting programmes are effective for many children with attention-deficit/hyperactivity disorder (ADHD) and oppositional defiance disorder (ODD) or conduct disorder (CD) problems, but treatment effects vary widely. We aimed to identify key moderators of intervention effects on child behaviour difficulties using individual participant data from randomised controlled trials (RCTs). Methods: We pooled individual data from 48 RCTs (5729 families) of behavioural parenting programmes for children with (sub)clinical characteristics of ADHD and/or ODD/CD. Fifteen parent-reported child, parent, and family characteristics at baseline were examined as potential moderators. The primary outcome was parent-reported child behavioural difficulties (e.g., hyperactivity-impulsivity, defiance, irritability, aggression, rule-violation) at the first post-intervention assessment, harmonised across instruments and trials. We used a machine learning method, the Least Absolute Shrinkage and Selection Operator (LASSO) regularization to select robust moderators and then tested selected moderators in a standard regression model. Findings: Two significant moderators were identified: Baseline child internalising problems (e.g., anxiety, depression, withdrawal) and baseline non-constructive parenting (e.g., harsh or inconsistent discipline). Moderating effects were driven by increases in child behavioural difficulties among families in control condition with higher baseline levels of child internalizing problems and non-constructive parenting, rather than larger improvement among intervention recipients. For non-constructive parenting, this increase was specific to waitlist conditions. Interpretation: Our findings suggest that children with higher levels of internalising problems and/or families reporting higher non-constructive parenting are at greatest risk of deterioration in behavioural difficulties without support. Results may inform triage: to identify children with co-occurring internalizing problems who may benefit from timely access to behavioural parenting programs and minimize waiting times for families with higher non-constructive parenting.

  • Open Access

    Sensitivity of the PHQ-9: Does Item Ordering Effect Questionnaire Outcomes and Item Responses (Preprint)

    by Zahraa Abdulhussein, Darragh Glavin, Pepijn van de Ven, Chris van Klaveren, Sascha Struijs

    BACKGROUND The PHQ-9 is a standard self-report questionnaire used to assess the severity of depression in individuals. It consists of nine items, typically presented in a fixed order. Although shorter versions of the PHQ-9 exist, they use the same items in the same sequence. Since different individuals may benefit from different question orders, it is important to determine whether item order influences responses or overall scores. In this study, we investigated whether altering the order of these questions would affect the overall depression score or the responses to individual items. We conducted a survey involving seven groups of university students in 9 universities across the Netherlands, with each group completing the PHQ-9 in a different question order. One group received the standard order (Original), while the remaining six groups received different orders. To evaluate potential differences, we performed hypothesis testing using t-test, Welch’s t-test, ANOVA, the Mann–Whitney U test, and permutation tests. We compared both the total scores and the item-level responses between the standard order and the other orderings. The results showed no significant differences in mean total scores, whereas individual item scores (phq1, phq6 and phq8) showed statistically significant differences, but these are unlikely to be of practical importance. These findings suggest that reordering PHQ-9 items does not impact the overall depression assessment or individual item responses. OBJECTIVE The aim of this project is to examine whether the PHQ-9 is sensitive to the order in which its items are presented. This study therefore addresses a broader research question concerning the robustness of such measures to item-ordering effects. The investigation is particularly relevant for validating ongoing research in which we are experimenting with a personalized PHQ-9 using machine learning, where both the order and number of items presented may vary across individuals. In such a dynamic version of the test, item selection could be adaptively tailored based on participants’ previous responses. Furthermore , within the context of the Caring Universities project, this investigation is also important for evaluating the administration of potential new ultra-brief questionnaires designed for the student population. If these new versions are to be implemented, it is essential to assess how item ordering influences questionnaire outcomes and individual response patterns, particularly when certain items may be skipped. These considerations raise important questions about the validity and reliability of ultra-brief instruments. Addressing these research questions in advance is therefore a critical step before deploying such questionnaires in practice. METHODS Hypothesis testing including t-test, Welch’s t-test, ANOVA, the Mann–Whitney U test, and permutation tests. RESULTS The results showed no significant differences in mean total scores, whereas individual item scores (phq1, phq6 and phq8) showed statistically significant differences, but these are unlikely to be of practical importance. These findings suggest that reordering PHQ-9 items does not impact the overall depression assessment or individual item responses. CONCLUSIONS In this study, we tested whether changing the order of PHQ-9 items influences depression severity scores or individual item responses. Analyses of 9,758 participants across seven order types revealed no significant differences in overall PHQ-9 total scores, as confirmed by both independent-samples t-tests and ANOVA. Between-group variance was negligible compared to within-group variance, indicating that differences in scores were primarily driven by individual variation rather than item order. At the item level, Mann–Whitney U tests identified a small number of statistically significant effects for phq1, phq6, and phq8; however, these effects were inconsistent across order types and of very small magnitude. Permutation tests further confirmed these patterns, showing minor but statistically detectable differences, with most effects close to zero and one (phq8 in the cognitive–somatic order) non-significant. Taken together, these results demonstrate that reordering PHQ-9 items does not meaningfully affect either the overall depression score or responses to individual items. While isolated fluctuations emerged at the item level, they lacked consistency and practical significance. Furthermore, the findings of this study validate the use of ultra-brief questionnaires for the student population within the Caring Universities project and support our current work on developing a personalized PHQ-9. Such developments have the potential to improve efficiency and enhance the respondent experience, representing an important step toward more individualized approaches to psychological assessment.

  • Open Access

    Truncated History Framework for Synthetic Control Approaches

    by Peter Spoelstra, Tom Stolp, Bart H.H. Golsteyn, Ilja Cornelisz, Chris van Klaveren


  • Open Access

    Feasibility of a Guided Web-Based Procrastination Intervention for College Students: Open Trial

    by Sevin Ozmen, Arpana Amarnath, Leonore de Wit, Chris van Klaveren, Pim Cuijpers, Annemieke van Straten, Sascha Struijs, Caring Universities Consortium

    Background: College students commonly struggle with procrastination, which is linked to mental health complaints and poor academic performance. Interventions based on cognitive behavioral therapy can be effective in reducing procrastination. Traditional face-to-face therapy and online interventions have shown promising outcomes, with the latter overcoming help-seeking barriers such as lengthy referral processes and waiting lists. Objective: This study aims to examine the feasibility and acceptability of a new eHealth intervention targeting procrastination for college students (“GetStarted”) with guidance by student e-coaches. This cognitive behavioral therapy–based intervention was designed specifically for and together with the target demographic of students studying in the Netherlands. Guidance was offered by trained clinical psychology students in the form of written motivational, supportive messages. Methods: We conducted a single-arm study. The primary outcomes were satisfaction (8-item Client Satisfaction Questionnaire [CSQ-8]), usability (10-item System Usability Scale [SUS-10]), and adherence (completion rate). The secondary outcomes were changes to procrastination (Irrational Procrastination Scale [IPS]), depression (9-item Patient Health Questionnaire [PHQ-9]), stress (10-item Perceived Stress Scale [PSS-10]), quality of life (Mental Health Quality of Life Questionnaire [MHQoL]), and e-coaching satisfaction (Working Alliance Inventory for Guided Internet Interventions [WAI-I]). Results: Of 734 participants who started the intervention, 335 (45.6%) completed the posttest. Students reported being satisfied with the intervention (CSQ-8: mean 23.48, SD 3.23) and found it very usable (SUS-10: mean 34.39, SD 4.52). Regarding adherence, participants completed 68.95% of the intervention on average, while 36.65% (n=269) of participants completed the full intervention. Participants showed a significant decrease in procrastination (IPS: mean decrease 35.39-32.56, Cohen d=0.63), depression (PHQ-9: mean decrease 9.27-7.73, Cohen d=0.35), and stress (PSS-10: mean decrease 20.79-19.02, Cohen d=0.31) as well as an increase in quality of life (MHQoL: mean increase 12.81-13.65, Cohen d=0.37) from baseline to posttest to follow-up. Participants reported a moderate-to-strong alliance with their e-coach (WAI-I: mean 45.26, SD 7.72). Conclusions: The internet-based, student-guided intervention “GetStarted” targeting procrastination appears to be acceptable and feasible for college students in the Netherlands. However, high attrition rates and the lack of a control group mean that results must be interpreted with caution. To further examine intervention effectiveness, a randomized controlled trial needs to be conducted.


  • Open Access

    Labor market entry prospects of people with visual impairments: evidence from administrative register data in the Netherlands

    by Katharina Diehl, Eline Heppe, Melvin Vooren, Ilja Cornelisz, Chris van Klaveren

    The UN’s Convention on the Rights of Persons with Disabilities includes a right to equal work opportunities. People with visual impairments (VI) are targeted by this convention and experience low employment rates. The initial transition from education to labor is a predictor for later labor market participation. Exploiting administrative register data, we identified a graduated subpopulation with VI and matched them to a control group. This paper evaluates (1) group differences in various socio-economic outcomes following graduation, and (2) following initial employment. Following graduation there are fewer transitions into (self-)employment for people with VI and more transitions to disability insurance or social benefits, but similar patterns for unemployment insurance and further education. Scrutinizing subsequent dynamics for those that secured initial employment revealed a greater share of those with VI maintained employment as dominant source of income given there was a change in employment status. However, there are more dynamics recorded for individuals with VI, which may reflect difficulties retaining a job. This group remained overrepresented in disability insurance as well as social benefits. Differences with regards to self-employment, unemployment insurance, further education and the other no income status did not reach significance. In conclusion, this study did not find equal labor market prospects for declarants with VI. The finding that VI declarants face less favorable labor market prospects than their matched counterparts, despite having similar cognitive ability, underscores the need for policies that not only improve their labor market outcomes but also promote greater equality in opportunities to reach the UN convention’s ideals.