Log in or create a free Rosenverse account to watch this video.
Log in Create free account100s of community videos are available to free members. Conference talks are generally available to Gold members.
A Mixed Method Approach to Validity to Help Build Trust
Summary
Quantitative instruments are frequently sought because 1) they can be quickly fielded to lots and lots of people, and 2) when carefully sampled, they can be generalizable to the population of users/customers. However, because many times the focus is on speed to launch because decision-makers need results quickly, there is not much depth given to their development, nor an investigation of the validity evidence. In the session, I will share a framework that centers validity and is necessarily a mixed methods approach to research. I will also share ideas on how to scale the research over time so that findings and insights are able to be iteratively delivered to stakeholders, while also iteratively informing one another in a qual-quant research dance that brings more trustworthy, user-centered evidence to decision-makers. Finally, I will share ideas for a course I am developing for supporting qualitative researchers to become more mixed in their approach.
Key Insights
-
•
Validity involves five evidence sources: content, response processes, internal structure, relations to other variables, and consequences of testing.
-
•
Qualitative methods, especially cognitive interviews, are crucial for understanding how respondents interpret survey items, supporting validity.
-
•
Surveys should be treated as products that need ongoing iteration and testing, not one-off tools.
-
•
Ethical considerations extend beyond data privacy to how survey results affect user experience and product decisions.
-
•
Mixed methods approaches leverage both qualitative insights and quantitative analyses to build a stronger validity argument.
-
•
Breaking down survey validation work across multiple teams and 'bite-sized' efforts makes the process manageable.
-
•
Revising surveys over time to improve validity complicates measuring change longitudinally but increases trustworthiness.
-
•
Stakeholder buy-in improves when validity processes are communicated as phased insights offering tangible results quickly.
-
•
Analyses such as factor analysis and Rasch modeling reveal survey internal structure and help identify item bias across subpopulations.
-
•
It is important to revisit validity considerations continuously, especially after product or user base changes.
Notable Quotes
"Validity is the degree to which evidence and theory support the interpretations of test scores for proposed uses."
"Qualitative research in a mixed methods setting needs to think about validity to support the bigger validity argument."
"What would happen if you do think about validity? Would it change your process or research plans?"
"Surveys are products too, so they need to be iteratively tested."
"You don’t know what you don’t know. Let’s write surveys that cover those blind spots."
"If the survey wording changes, measuring change over time becomes difficult, but improving the survey builds trust."
"Ethical considerations should go beyond typical privacy reviews and think deeply about user impact throughout the process."
"It is better to partner with someone who has quantitative expertise to interpret internal structure analyses."
"Management and stakeholders often care more about getting usable information quickly than understanding the full validity process."
"We can get quick insights in phases to keep teams engaged and slowly build a complete validity argument."
Or choose a question:
More Videos
"This conference is a polished snapshot of conversations already happening at scale in the community."
Bud Caddell Kristin Skinner Alana WashingtonDesignOps Community Sensing Session
May 13, 2021
"Getting weird in Miro lets people play and connect remotely without pressure to prepare or present."
Shipra Kayan Tess DixonHow Tess Dixon Facilitates Team Engagement and Collaboration at Condé Nast Using Miro
October 1, 2021
"Sponsor sessions are not sales pitches; they offer content similar in quality to the main program."
Bria AlexanderOpening Remarks
September 29, 2021
"AI is automating inequality; if we don’t include inclusive thinking in AI design, the biases will become worse."
John Maeda Alison RandAbout Design Organizations
May 13, 2019
"A lot of people will try a prompt and not get what they were hoping to get and maybe miss out on all the benefits of AI."
Kate KalcevichDesigning inclusively with AI
June 5, 2024
"We connect design decisions back to business performance and risk metrics to answer the concerns of our partners."
Frank DuranPartnership Playbook: Lessons Learned in Effective Partnership
January 8, 2024
"If we avoid how we work together, we create disjointed experiences and cause confusion and frustration for riders."
Kristin TaylorBuilding Bridges Across Organizational Silos
November 18, 2022
"The DEO looks at all business problems as design problems, solvable through creativity, imagination, and metrics."
Maria GiudiceEmpowering change: Reigniting purpose, passion and impact in research
March 13, 2025
"Be present, put devices down, turn off Slack notifications, and really engage."
Shelby SwitzerMaking Space for Community Knowledge-sharing in a Distributed World
December 10, 2021
