Yvonne Chen

Modeling the Bullying Prevention Program Preferences of Educators: A Discrete Choice Conjoint Experiment

We used discrete choice conjoint analysis to model the bullying prevention program preferences of educators. Using themes from computerized decision support lab focus groups (n = 45 educators), we composed 20 three-level bullying prevention program design attributes. Each of 1,176 educators completed 25 choice tasks presenting experimentally varied combinations of the study’s attribute levels. Latent class analysis yielded three segments with different preferences. Decision Sensitive educators (31%) preferred that individual schools select bullying prevention programs. In contrast, Support Sensitive educators (51%) preferred that local school boards chose bullying prevention programs. This segment preferred more logistical and social support at every stage of the adoption, training, implementation, and long term maintenance processes. Cost Sensitive educators (16%) showed a stronger preference for programs minimizing costs, training, and implementation time demands. They felt prevention programs were less effective and that the time and space in the curriculum for bullying prevention was less adequate. They were less likely to believe that bullying prevention was their responsibility and more likely to agree that prevention was the responsibility of parents. All segments preferred programs supported by the anecdotal reports of colleagues from other schools rather than those based on scientific evidence. To ensure that the bullying prevention options available reflect the complex combination of attributes influencing real world adoption decisions, program developers need to accommodate the differing views of the Decision, Support, and Cost Sensitive segments while maximizing the support of parents and students.

Providing Information to Parents of Children with Mental Health Problems: A Discrete Choice Conjoint Analysis of Professional Preferences

We used discrete choice conjoint analysis to model the ways 645 children’s mental health (CMH) professionals preferred to provide information to parents seeking CMH services. Participants completed 20 choice tasks presenting experimentally varied combinations of the study’s 14 4-level CMH information transfer attributes. Latent class analysis revealed three segments. Open Access professionals (32.2%) preferred that intake workers automatically provide all parents with CMH information. They preferred information prepared by professional organizations and located at accessible settings such as public schools. They responded favorably to the internet as a source of information for parents. Controlled Access professionals (22.2%) preferred information that was approved and recommended by a child’s therapist, prepared by an experienced clinician, and located at hospitals and CMH clinics. Process Sensitive professionals (45.6%) showed a stronger preference for active learning materials with parenting groups and therapist “coaching” calls supporting the knowledge transfer process. Simulations suggested that realizing the benefits of CMH information requires the development of knowledge transfer strategies that align the preferences of professionals with those of the families they serve.

Adaptive Choice-Based Conjoint Analysis

Conjoint analysis (CA) has emerged as an important approach to the assessment of health service preferences. This article examines Adaptive Choice-Based Conjoint Analysis (ACBC) and reviews available evidence comparing ACBC with conventional approaches to CA. ACBC surveys more closely approximate the decision-making processes that influence real-world choices. Informants begin ACBC surveys by completing a build-your-own (BYO) task identifying the level of each attribute that they prefer. The ACBC software composes a series of attribute combinations clustering around each participant’s BYO choices. During the Screener section, informants decide whether each of these concepts is a possibility or not. Probe questions determine whether attribute levels consistently included in or excluded from each informant’s Screener section choices reflect ‘Unacceptable’ or ‘Must Have’ simplifying heuristics. Finally, concepts identified as possibilities during the Screener section are carried forward to a Choice Tournament. The winning concept in each Choice Tournament set advances to the next choice set until a winner is determined.A review of randomized trials and cross-over studies suggests that, although ACBC surveys require more time than conventional approaches to CA, informants find ACBC surveys more engaging. In most studies, ACBC surveys yield lower standard errors, improved prediction of hold-out task choices, and better estimates of real-world product decisions than conventional choice-based CA surveys.

Examining Parents’ Preferences for Group and Individual Parent Training for Children with ADHD Symptoms

Parent training (PT) programs have been found to reduce some behavioral impairment associated with childrens attention deficit hyperactivity disorder (ADHD) as well as improve parenting competence, but poor uptake and participation by parents are formidable barriers that affect service effectiveness. We used a discrete-choice experiment (DCE) to examine how parent preferences for treatment format (i.e., group vs. individual) might influence their participation in PT. Participants were 445 parents seeking mental health services for children with elevated symptoms of ADHD in Ontario, Canada. Parents completed a DCE composed of 30 choice tasks used to gauge PT format preference. Results showed that 58.7% of parents preferred individual PT; these parents were most interested in interventions that would make them feel more informed about their childs problems and in understanding—as opposed to solving—their childs problems. A minority of parents (19.4%) preferred group PT; these parents were most interested in active, skill-building services that would help them solve their childs problems. About one fifth of parents (21.9%) preferred the Minimal Information alternative (i.e., receiving neither individual or group PT); these parents reported the highest levels of depression and the most severe mental health problems in their child. Results highlight the importance of considering parent preferences for format and suggest that alternative formats to standard PT should be considered for multiply stressed families.

Modeling Mental Health Information Preferences During the Early Adult Years: A Discrete Choice Conjoint Experiment

Although most young adults with mood and anxiety disorders do not seek treatment, those who are better informed about mental health problems are more likely to use services. The authors used conjoint analysis to model strategies for providing information about anxiety and depression to young adults. Participants (N = 1,035) completed 17 choice tasks presenting combinations of 15 four-level attributes of a mental health information strategy. Latent class analysis yielded 3 segments. The virtual segment (28.7%) preferred working independently on the Internet to obtain information recommended by young adults who had experienced anxiety or depression. Self-assessment options and links to service providers were more important to this segment. Conventional participants (30.1%) preferred books or pamphlets recommended by a doctor, endorsed by mental health professionals, and used with a doctors support. They would devote more time to information acquisition but were less likely to use Internet social networking options. Brief sources of information were more important to the low interest segment (41.2%). All segments preferred information about alternative ways to reduce anxiety or depression rather than psychological approaches or medication. Maximizing the use of information requires active and passive approaches delivered through old-media (e.g., books) and new-media (e.g., Internet) channels.

Modeling the bullying prevention program design recommendations of students from grades five to eight: a discrete choice conjoint experiment

We used a discrete choice conjoint experiment to model the bullying prevention recommendations of 845 students from grades 5 to 8 (aged 9-14). Students made choices between experimentally varied combinations of 14 four-level prevention program attributes. Latent class analysis yielded three segments. The high impact segment (27.1%) recommended uniforms, mandatory recess activities, four playground supervisors, surveillance cameras, and 4-day suspensions when students bully. The moderate impact segment (49.5%) recommended discretionary uniforms and recess activities, four playground supervisors, and 3-day suspensions. Involvement as a bully or bully-victim was associated with membership in a low impact segment (23.4%) that rejected uniforms and surveillance cameras. They recommended fewer anti-bullying activities, discretionary recess activities, fewer playground supervisors, and the 2-day suspensions. Simulations predicted most students would recommend a program maximizing student involvement combining prevention with moderate consequences. The simulated introduction of mandatory uniforms, surveillance cameras, and long suspensions reduced overall support for a comprehensive program, particularly among students involved as bullies or bully-victims.

Modeling improvements in booster seat use: a discrete choice conjoint experiment.

Automobile crashes are the leading cause of death in children aged 1-14 years. Many children, however, are not properly restrained in safety seats that reduce serious injury and death. This study used a discrete choice conjoint experiment to study factors influencing the decision to use booster seats. Parents of 1714 children aged 4-9 years from nine Canadian provinces completed choice tasks presenting experimentally varied combinations of 15 4-level booster seat promotion attributes. Latent class analysis yielded three segments of parents. The choices of the Benefit Sensitive segment (50%) were most sensitive to the injury prevention benefits of booster seats. The choices of parents in the Context Sensitive segment (33.5%) were more likely to be influenced by installation complexity, oppositional behavior, and the prospect that their child may be teased for riding in booster seats. Parents in the High Risk segment (16.5%) were younger, less educated, and less knowledgeable about vehicle safety legislation. They anticipated fewer benefits, expected more barriers and were less likely to use booster seats. Simulations suggest that consistent enforcement coupled with advertising focusing on injury prevention and the use of booster seats by other parents would increase adoption.

Preferences for evidence-based practice dissemination in addiction agencies serving women: a discrete-choice conjoint experiment.

AIM   To model variables influencing the dissemination of evidence-based practices to addiction service providers and administrators. DESIGN   A discrete-choice conjoint experiment. We systematically varied combinations of 16 dissemination variables that might influence the adoption of evidence-based practices. Participants chose between sets of variables. SETTING   Canadian agencies (n = 333) providing addiction services to women. PARTICIPANTS   Service providers and administrators (n = 1379). MEASUREMENTS   We estimated the relative importance and optimal level of each dissemination variable. We used latent class analysis to identify subsets of participants with different preferences and simulated the conditions under which participants would use more demanding professional development options. FINDINGS   Three subsets of participants were identified: outcome-sensitive (52%), process-sensitive (29.6%) and demand-sensitive (18.2%). Across all participants, the number of clients who were expected to benefit from an evidence-based practice exerted the most influence on dissemination choices. If a practice was seen as feasible, co-worker and administrative support influenced decisions. Client benefits were most important to outcome-sensitive participants; type of dissemination process (e.g. active versus passive learning) was more important to process-sensitive participants. Brief options with little follow-up were preferred by demand-sensitive participants. Simulations predicted that initiatives selected and endorsed by government funders would reduce participation. CONCLUSIONS   Clinicians and administrators are more likely to adopt evidence-based addiction practices if the practice is seen as helpful to clients, and if it is supported by co-workers and program administration.

Modeling Parenting Programs as an Interim Service for Families Waiting for Children's Mental Health Treatment

Using a discrete choice conjoint experiment, we explored the design of parenting programs as an interim strategy for families waiting for childrens mental health treatment. Latent class analysis yielded 4 segments with different design preferences. Simulations predicted the Fast-Paced Personal Contact segment, 22.1% of the sample, would prefer weekly therapist-led parenting groups. The Moderate-Paced Personal Contact segment (24.7%) preferred twice-monthly therapist-led parenting groups with twice-monthly lessons. The Moderate-Paced E–Contact segment (36.3%), preferred weekly to twice-monthly contacts, e-mail networking, and a program combining therapist-led sessions with the support of a computerized telephone e-coach. The Slow-Paced E–Contact segment (16.9%) preferred an approach combining monthly therapist-led sessions, e-coaching, and e-mail networking with other parents. Simulations predicted 45.3% of parents would utilize an option combining 5 therapist coaching calls with 5 e-coaching calls, a model that could reduce costs and extend the availability of interim services. Although 41.0% preferred weekly pacing, 58% were predicted to choose an interim parenting service conducted at a twice-monthly to monthly pace. The results of this study suggest that developing interim services reflecting parental preferences requires a choice of formats that includes parenting groups, telephone-coached distance programs, and e-coaching options conducted at a flexible pace.

Modeling the anti-cyberbullying preferences of university students: Adaptive choice-based conjoint analysis

Adaptive choice-based conjoint analysis was used to study the anti-cyberbullying program preferences of 1,004 university students. More than 60% reported involvement in cyberbullying as witnesses (45.7%), victims (5.7%), perpetrator-victims (4.9%), or perpetrators (4.5%). Men were more likely to report involvement as perpetrators and perpetrator-victims than were women. Students recommended advertisements featuring famous people who emphasized the impact of cyberbullying on victims. They preferred a comprehensive approach teaching skills to prevent cyberbullying, encouraging students to report incidents, enabling anonymous online reporting, and terminating the internet privileges of students involved as perpetrators. Those who cyberbully were least likely, and victims of cyberbullying were most likely, to support an approach combining prevention and consequences. Simulations introducing mandatory reporting, suspensions, or police charges predicted a substantial reduction in the support of uninvolved students, witnesses, victims, and perpetrators.