Quantitative Psychology · Measurement Science · Applied Research

I study how complex human experience can be measured and understood over time.

My work sits at the intersection of quantitative psychology, measurement science, and behavioral and health research. I develop and apply quantitative methods for answering difficult questions about human behavior, health, and development, particularly when the phenomena of interest cannot be directly observed, when measurement may function differently across people or contexts, or when the most important information lies in how individuals change over time.

Across academic, clinical, healthcare, and industry research settings, I work with multidisciplinary teams to translate complex data into rigorous and interpretable evidence.

Psychometrics Measurement Science Longitudinal Research Structural Equation Modeling Behavioral Science Quantitative Methods Study Design Research Leadership

Research & Methods

Methodological interests


Across these areas, my work is guided by a common question: how can we make credible claims about complex human phenomena when measurement is imperfect, individuals differ, and observations unfold over time?


Applications: Behavioral health · Human development · Patient-reported outcomes · Engagement · Clinical research · User experience

01

Measurement & psychometrics

Construct validity · Measurement equivalence · Latent variable modeling · Instrument development · Reliability · Sensitivity to change

How do we determine whether a measure is capturing the construct we intend to study and whether that measurement retains its meaning across people, groups, and time?

02

Modeling change

Longitudinal analysis · Structural equation modeling · Multilevel models · Growth trajectories · Individual differences in change

Many of the phenomena we care about most are dynamic. A single observation provides a snapshot; understanding trajectories of change reveals a much richer picture of development, behavior, and health.

03

Complex data & research questions

Mediation and moderation · Missing data · Predictive modeling · Behavioral data · Research design

I use quantitative methods to investigate complex relationships, identify meaningful heterogeneity, and design studies that can give credible answers to difficult scientific questions.

Research portfolio

Selected work

01

When does change mean the same thing for everyone?

Measurement equivalence and developmental growth

Measurement validation · Growth modeling · Longitudinal analysis · SEM · Construct validity

When we observe differences in developmental or behavioral scores, how do we know they represent real differences in the underlying construct rather than differences in how the measurement performs?

The challenge

Longitudinal research often assumes a score means the same thing across people and across time. Without that assumption, an apparent difference in a growth trajectory could reflect a change in the measurement rather than genuine development.

My work

One project examined the measurement and developmental properties of the Early Communication Indicator, a standardized assessment of early communication development.

With a multidisciplinary team, I led analyses using parallel-process and piecewise latent growth models to examine developmental trajectories, relationships among communication domains, and transitions between communication strategies.

Why it matters — a measurement system should do more than assign a score. It should give confidence that observed change reflects the phenomenon we intend to understand: not only where someone is, but how they are changing.

02

Looking beyond the average trajectory

Understanding heterogeneity in longitudinal change

Longitudinal modeling · Multilevel models · Growth trajectories · SEM

What do we lose when we summarize an entire population with a single average trajectory?

The challenge

Population averages obscure meaningful heterogeneity. Two individuals may share an identical baseline while following entirely different paths of improvement, deterioration, recovery, or stability.

What I do

Longitudinal and multilevel models that distinguish individual patterns of change from measurement noise and population-level trends across developmental, behavioral, psychological, and clinical outcomes.

  • Who changes?
  • How and when does meaningful change occur?
  • Why do individuals follow different trajectories?
  • Under what conditions do those trajectories differ?

Why it matters — a single measurement is only a snapshot. Whether the subject is well-being, engagement, development, health, or user experience, the trajectory tells the more meaningful story.

03

Measuring what we cannot directly observe

Latent variables and structural equation modeling

SEM · Factor analysis · Latent variables · Psychometrics · Measurement invariance

Well-being, engagement, symptoms, functioning, motivation — none can be measured directly. How do we know our observed measures represent the construct at all?

The challenge

Observed variables are imperfect indicators. A survey response, behavioral observation, or clinical assessment carries both meaningful information and measurement error, and scientific conclusions depend on telling them apart.

My approach

Latent variable methods — factor analysis and structural equation modeling — used to define constructs, examine validity, test theoretical models, and separate meaningful variation from error over time.

I am most interested in questions where the measurement model and the substantive model cannot be treated as separate problems.

Why it matters — good measurement begins with a clear understanding of the construct, but rigorous measurement means asking whether the data actually support the claims we want to make.

04

Finding the signal in complex data

Missing data and scalable quantitative methods

Missing data · Dimension reduction · Principal component analysis · Computational statistics

How can researchers preserve useful information when real-world datasets are incomplete?

The challenge

Missing data are rarely a technical inconvenience: how missing observations are handled can change a study’s conclusions. Large datasets hold thousands of potentially useful variables, but including them directly is often computationally impractical.

My approach

My methodological research examines how auxiliary information can be brought into missing-data analyses, including dimension-reduction techniques for finding the useful signal inside large datasets. The goal is methods that are both statistically rigorous and practically scalable.

Why it matters — the most sophisticated method is not necessarily the most useful one. I care about approaches that improve inference while staying practical enough for researchers working with real data.

Research philosophy

How I approach research

01

Start with the construct

Before selecting a model, ask what we are actually trying to understand or measure.

02

Treat measurement as part of the problem

Validity, equivalence, reliability, and sensitivity decide what conclusions the data can support.

03

Study change, not just states

Behavior, health, and experience are dynamic. Look past isolated observations to trajectories.

04

Match the method to the question

The best approach aligns with the question, the design, the measurement properties, and the data.

05

Make complexity useful

Advanced methods earn their place only when the results improve understanding and decisions.

Leadership

Research leadership

I work across the full research lifecycle — from formulating the scientific question through design, analysis, interpretation, dissemination, and methodological development.

As a Principal Biostatistician and research methods leader, I collaborate with investigators across clinical, behavioral, and translational research: building statistical strategy for grant applications, leading primary analyses, and advancing methodological approaches.

A cornerstone of my work is building quantitative capacity beyond individual projects. I mentor statisticians, analysts, faculty, programmers, and research staff, and develop methods, workflows, and infrastructure that can be reused across scientific questions.

Start with the problem, understand the context, adapt the approach, and build systems that help people do their best work.

Background

About

I am a quantitative psychologist and research leader with experience spanning academic research, healthcare, clinical research, and industry.

My background includes quantitative psychology, psychometrics, structural equation modeling, longitudinal analysis, multilevel modeling, missing-data methodology, and predictive modeling.

Throughout my career I have collaborated with clinicians, behavioral scientists, epidemiologists, data scientists, engineers, and organizational leaders on complex research questions.

Questions that motivate my work
  • How do we know that we are measuring what we think we are measuring?
  • Does a measure function equivalently across different people and contexts?
  • How can we distinguish meaningful change from measurement error?
  • Why do individuals with similar starting points follow different trajectories?
  • How can complex quantitative evidence inform real-world decisions?

Let’s measure something difficult.

Open to collaborations involving measurement science, quantitative research, behavioral data, longitudinal analysis, psychometrics, research methodology, and product and experience measurement.

Get in touch

Quantitative Research & Measurement Scientist