PhD Researchers & Doctoral Candidates
A well-designed instrument and sampling plan for a thesis—so the data can bear the analysis, and the measures survive review.
Research Methods
Survey & Primary Research
The best analysis in the world can't repair a broken instrument or an unrepresentative sample. We get the foundations right—questions that measure what you mean, a sample that reflects your population, and weights that make the numbers generalizable—so your primary data can carry the conclusions you draw from it.
Questionnaire & sampling design
Weighting & complex-survey analysis
R · Stata · Qualtrics
Reproducible, journal-ready
Sample output
strata
weighted sample
Overview
Data quality is built in, not analyzed in
Survey research has a hard truth at its center: the quality of your conclusions is capped by the quality of your data collection, and no analysis recovers what a bad instrument or an unrepresentative sample failed to capture. A leading question, an ambiguous scale, a sampling frame that misses part of the population—these are decided before a single response arrives, and they set a ceiling on everything after.
So we work upstream. We design questionnaires where each item cleanly operationalizes a construct, pilot them to catch problems early, and build a sampling scheme—stratified, clustered, or multistage—matched to your population and budget. We compute the sample size from a power analysis rather than a rule of thumb, and plan the weighting before collection so the sample can be generalized back to the population.
Then we analyze it correctly: reliability and validity of the measures, complex-survey estimation that respects the design, non-response and weighting adjustments, and the SEM or regression models your questions call for. The deliverable is primary data you can defend—and reproducible code you keep.
Who We Work With
For research that collects its own data
If you're running a survey or gathering primary data, this is the right desk to write to.
Management & Marketing Researchers
Questionnaire design, scale validation, and SEM-ready survey data for organizational and consumer studies.
Social Scientists & Public-Opinion Researchers
Representative sampling, weighting, and complex-survey analysis for population-level conclusions.
Policy Organizations & Agencies
Survey design and analysis where results have to generalize and stand up to external scrutiny.
Research Institutes & Data Teams
Large or ongoing survey programs needing rigorous design, weighting, and non-response handling.
Corporates & Market Researchers
Primary research on customers, employees, or markets, designed to yield defensible, actionable data.
Capabilities
The full survey toolkit, design to analysis
Organized across instrument design, measurement, and complex-survey analysis. If your study needs a method not listed here, ask—this is the core, not the boundary.
Getting the data right
The upstream decisions—instrument, sample, and size—that set the ceiling on everything after.
- Questionnaire design
- Survey design
- Sampling design
- Sample-size calculation
- Power analysis
- Data collection design
Validating & using constructs
Establishing that the measures hold, then modeling the relationships between them.
- Reliability analysis
- Validity analysis
- CFA
- SEM
- PLS-SEM
- Mediation
- Moderation
- Multigroup analysis
Generalizing to the population
The design-aware methods that turn a sample into valid population-level estimates.
- Survey weighting
- Complex survey analysis
- Non-response analysis
How the Analysis Works
Six steps from question to representative data
A transparent sequence weighted toward design—because in survey research, the data's quality is fixed before it's collected. Nothing is a black box.
Steps are adapted to your study: the population, the mode of collection, and whether you need population estimates or model relationships. We settle the design with you before fielding.
-
1
Define
Clarify the constructs, the target population, and the sampling frame the study will rest on.
Inputs: constructs · population · frame
-
2
Design instrument
Build the questionnaire—item wording, scales, ordering—and pilot it to catch problems early.
Steps: item design · scales · pilot · refine
-
3
Plan sample
Choose the sampling scheme, compute the sample size via power analysis, and plan the weighting.
Methods: stratified/cluster · power · weighting plan
-
4
Collect
Field the survey with monitoring for response rates, data quality, and coverage of the frame.
Checks: response rate · data quality · coverage
-
5
Validate & weight
Assess reliability and validity, apply survey and non-response weights, and estimate with the correct design.
Methods: CFA · reliability · weighting · complex-survey
-
6
Report
Deliver validated estimates and models, figures, methodology, and reproducible code you keep.
Output: estimates · figures · methods · R/Stata code
Rigor by default
The checks that make survey data trustworthy
Representativeness and measurement quality are what reviewers probe first. Establishing them is standard on every engagement.
Included on every project
- Power analysis and sample-size justification
- Reliability and validity of measurement scales
- Design-appropriate weighting and estimation
- Non-response and coverage assessment
- Reproducible instrument and analysis code you keep
What You Receive
Every engagement, delivered in full
Not a black-box result and a number, but a complete, documented package you can submit, defend, and reproduce.
- Designed and piloted questionnaire
- Sampling plan and sample-size justification
- Clean, weighted dataset
- Reliability and validity evidence
- Population-level or model-based estimates
- Non-response and weighting documentation
- Interpretation of results and their limits
- Reproducible R (survey) or Stata code
- Journal-ready methodology and results sections
Where this fits
Part of a larger arc
Survey research is strongest when the design ahead of it is deliberate and the measurement modeling after it is rigorous—each handled with the same care.
Research Design & Planning
Identification strategy, power, and specification decided before estimation begins.
Explore methodsStatistical & Methodological Audit
An independent check of assumptions, specification, and reproducibility before submission.
Explore auditPublication & Research Support
Methods and results reporting, journal selection, and reviewer-response support.
Explore support
FAQ
Common questions
Answers to what most researchers and project leads ask before we begin a survey or primary-research engagement.
Yes, and it is where survey quality is won or lost. Question wording, response scales, ordering, and the operationalization of each construct determine whether the data measures what you intend. We design the instrument, pilot it, and refine it before full collection begins.
Through a power analysis tied to the effects or model you plan to estimate, the expected effect size, and your precision targets—adjusted for the design (stratification, clustering) and for anticipated non-response. The goal is a sample large enough to answer the question and no larger.
It depends on your population and goals. Simple random sampling is rarely feasible at scale; stratified, cluster, and multistage designs trade off cost against precision, and each implies a different analysis. We design the sampling scheme and, crucially, carry its structure through to the analysis with appropriate weights.
Usually, yes, if you want population-level estimates. Design weights correct for unequal selection probabilities, and post-stratification or raking weights align the sample to known population margins. Ignoring weights on a complex sample produces biased estimates and wrong standard errors—we handle both correctly.
We assess whether non-response is likely to bias results, compare respondents to the target population and to known frame characteristics, and apply non-response weighting or, where appropriate, principled imputation—and we report the assumptions those corrections rest on.
Yes. We assess reliability (Cronbach's alpha, composite reliability), and convergent and discriminant validity through confirmatory factor analysis, before any construct is used in a structural model—the measurement work that reviewers increasingly require.
R (survey, lavaan, and related packages), Stata (with its complex-survey commands), and Mplus or SmartPLS for SEM, plus survey platforms such as Qualtrics for collection. You receive the instrument, versioned analysis code, and a methods section written to journal standards.
Yes, and it is the ideal time. The questionnaire, the sampling frame, the sample size, and the weighting plan all need to be settled before collection—fixing them in advance prevents a dataset that can't support the analysis it was gathered for.
Related Expertise
Often paired with econometrics
Planning a survey?
Tell us your population and what you need to measure—we'll design the instrument and sample, and make sure the data generalizes.