Winter 2025 Issue

Discrete Choice Experiments: Guiding Social Work Policies and Practices With Patient Goals
By J.E. Whilldin
Social Work Today
Vol. 25 No. 1 P. 30

Social work research requires collaboration and engagement with different stakeholders to understand their diverse preferences and needs. Informed research incorporates stakeholder views with the goal of improving research questions and generating more relevant research.1 However, currently there is no standard methodology for measuring stakeholder preferences in social work, says Alan Ellis, PhD, MSW, an associate professor at the School of Social Work at North Carolina State University. “So, researchers and practitioners typically develop a set of survey or interview questions and make inferences about stakeholder preferences based on the responses,” he explains.

Ellis and his colleagues recently published an article proposing the use of a discrete choice experiment (DCE) methodology for social work research. The article provides an introduction to DCE methodology and an overview of designing and conducting a DCE for social work applications. DCE methodology includes the typical components of a research study: the research question, study design, measuring instrument design, data collection, analysis and interpretation, and reporting. However, a DCE is used to analyze and rank participant preferences.1 Ellis explains, “Although traditional survey methods can be used to evaluate stakeholder perspectives, the DCE is one of several methodologies specifically designed to assess the degree to which people prioritize one thing over another. We propose that social work researchers adopt DCEs as a robust tool for capturing stakeholder preferences on any number of issues.”

Stakeholder’s Choice
A quantitative research methodology, a DCE uses a survey-based experimental study design in which study participants (relevant stakeholders) are presented with a series of hypothetical scenarios called choice tasks. In each choice task, the stakeholder chooses from two or more alternatives (for example, products or services). The premise of DCE methodology is that when stakeholders engage in decision making, they will choose the alternative associated with the greatest benefit to them. Each alternative in a choice task has a set of two or more attributes assigned so as to require a trade-off by the study participant.2 For example, in a marketing DCE about cell phone service providers, survey alternatives would be different wireless plans with attributes for different monthly service costs, unlimited/limited data, internet, streaming service channels offered as a bonus, and other benefits. The DCE is designed such that the attribute trade-off involves the study participant giving up some amount of an attribute to gain more of another attribute (eg, giving up unlimited data for a lower monthly cell service cost).2

DCE methodology is not new—it has been used for decades in marketing and has also become increasingly applied in health care research and policy development. Health care-focused DCEs commonly evaluate patient experience and trade-offs between factors related to patient experiences and health outcomes.1,2

“DCEs are often used to guide health care and health policy decisions, and the health care field has a lot of overlap with social work,” Ellis notes.

The use of DCEs and other preference studies has been previously proposed for social work approximately 10 years ago. However, the proposals were general and did not provide specific direction for implementation in social work research. In addition, published research on DCEs in social work journals is very limited. Ellis and his colleagues provide an orientation to DCE methodology geared toward social work researchers.

“Typically, DCEs are used to understand population preferences and how those preferences vary across individuals rather than to measure the preferences of a specific individual or small group in a clinical setting,” Ellis explains. “We aimed to introduce DCEs to researchers rather than practitioners [in this publication]. However, DCEs are practical in that they could be used by an individual agency to guide the development or selection of social work programs or interventions,” he adds.

Evaluating study participant choices in the controlled DCE design allows researchers to understand trade-offs among attributes by analyzing participants’ choices among alternatives and quantifying the relative strengths of participant preferences.

“Because DCEs were designed specifically to assess what choices people would make when confronted with realistic choice scenarios, DCEs can improve the use of stakeholder input in several ways,” Ellis says.

In social work, for example, Ellis explains that a DCE can help researchers with the following:

• understanding the trade-offs stakeholders think about when considering social work interventions or policies;

• assigning relative importance of different features of social work interventions;

• estimating how much stakeholders would pay or what they would give up for a certain improvement in services; and

• predicting what percentage of a population would choose a given social work intervention or policy.

“So, DCEs can provide helpful information that applies directly to social work policy and practice recommendations,” Ellis notes. In their published article on DCEs, Ellis and his colleagues add, “Further, DCEs make it possible to measure and understand variation in preferences across diverse stakeholder groups, consistent with the value that our profession places on social justice.”

Advancing Social Work
DCEs not only align closely with social work values but also could benefit the profession in specific ways.”1 Applying a DCE methodology to social work research could have the following benefits1:

• facilitating the formulation of social work research questions;
• choosing relevant social work research study outcomes;
• developing social work programs, policies, and interventions;
• assisting organizations with resource allocation; and
• helping organizations incorporate stakeholder preferences into decisions about social work efforts.

“Preferences that a DCE study design might evaluate include, for example, priorities of parents when selecting mental health treatment for their children. A DCE can explore this question by presenting scenarios that include different types of mental health care providers, treatment methods, costs, locations, and so on,” Ellis says. Trade-offs of attributes might include out-of-pocket treatment costs vs decreased travel time or access to highly rated therapists.1 “Caregivers’ stated choices in these scenarios can provide a lot of information about their priorities,” Ellis emphasizes.

In designing a DCE, social work researchers may engage relevant stakeholders to assist in the selection of alternatives and attributes. This engagement often includes qualitative data collection, such as interviews or focus groups, which can also be used to obtain feedback on a pilot DCE survey.1

In addition to evaluating parent preferences for mental health treatment, specific applications for a DCE methodology in social work research might also include1:

• understanding parent/caregiver preferences for behavioral treatment (eg, medications vs cognitive behavioral therapy vs combination treatment);

• determining social work educational preferences (eg, online vs in-person education programs);

• assessing how therapists select social work interventions for clients;

• prioritizing funding allocations by identifying the social work programs valued by relevant stakeholders for a certain target population;

• evaluating new social work interventions for specific populations (eg, individuals with homelessness, substance use); and

• motivating stakeholder engagement in community-based social work research.

“DCEs align closely with social work values and with community-engaged research principles. So, by adopting this methodology, we could do a better job of grounding social work research, policy, and practice in stakeholder values,” Ellis says. In addition, using a DCE methodology can strengthen the link between social work research and practice. “Having a better understanding of our clients’ preferences and goals will make us better social workers,” Ellis adds.

Although DCE methodology is currently being recommended for the social work research setting, wide adoption of DCEs in social work could eventually lead to clinical practice applications. A DCE could be applied to enhance shared decision making between a social worker and their individual clients.1

— J.E. Whilldin is a medical research analyst and writer from the Reading, Pennsylvania, area.

References
1. Ellis A, Cryer-Coupet QR, Weller BE, Howard K, Raghunandan R, Thomas KC. How to use discrete choice experiments to capture stakeholder preferences in social work research. J Soc Social Work Res. 2024;15(20).

2. Szinay D, Cameron R, Naughton F, Whitty JA, Brown J, Jones A. Understanding uptake of digital health products: methodology tutorial for a discrete choice experiment using the Bayesian efficient design. J Med Internet Res. 2021;11;23(10):e32365.