Harris Biostatistics Consulting
Provided by the team at statswithr.com
At Harris Biostatistics Consulting, we offer the services you need to go from study conception to data archiving. This can start from initial study design questions before data collection begins and finish with analysis justification for reviewers. Our team of graduate-level statisticians and applied scientists have years of experience in the manipulation and proper modeling of data. This includes the familiar data cleaning steps that may require several weeks, and can be accelerated using statistical software (R, SAS, Python). In the data cleaning space, our statistical programming support saves your team time and energy, allowing you to focus on data collection and conceptual concerns. Our primary statistician, Michael Harris, MS, MAS is a well-recognized R programming expert with years of experience in using R for statistical analysis, data manipulation, and application development. In the study design and analysis realm, which is described here as statistical consulting, we have tools to help you as well. We have published papers using a variety of endpoints to answer the question you have in the best way we know. A non-exhaustive list of the outcome variable types we have modeled includes: continuous, binary, count variables, time-to-event data, and longitudinal data. Additionally, we are familiar with other advanced methodologies like exploratory factor analysis, missing data imputation methods, and causal mediational modeling. As a third tier of service, we also offer statistical data science consulting if there is a strong match between project requirements and current capacity.
Services We Offer
Tier 1 - Statistical Programming
Clear and fully traceable data cleaning
Well-documented calculated columns and scales
Automated and standardized data reporting
Automated experimental data backup (if your institution allows this)
Data archiving to adhere to data sharing agreements
Tier 2 - Statistical Consulting
Guidance on data collection procedures based on the research question
Model selection based on the study outcomes and data structure
Manuscript preparation, including all analysis related sections
Peer-reviewer model justification or reanalysis (if appropriate)
Simulation studies that support manuscript findings
Tier 3 - Statistical Data Science
Builds interrelated machine learning infrastructure to address a scientific need
Scientific application development that requires complex back-end support
Big data aggregation from multiple sources in real-time (data engineering)
Includes methods like natural language processing and deep learning
Tier 3 is reserved for projects requiring unusually high technical complexity, custom infrastructure, or advanced statistical computing beyond standard consulting support.
Deliverables You Can Expect
Reproducible R code and output
Cleaned and clearly documented datasets
Statistical analysis plans
Tables and figures for manuscripts
Methods and results text support
Guidance on reviewer-response strategy
Written recommendations for next steps
Who We Work Best With
We work best with researchers, faculty, research staff, and graduate-level teams in the broadly defined social sciences. This includes work in psychology, public health, education, sociology, communication, political science, economics, human development, social work, and related interdisciplinary fields.
Our services are especially well suited for clients who need help with things like:
study design and analytic planning
survey and observational data
longitudinal and repeated measures data
regression modeling and generalized linear models
missing data strategies
manuscript-ready analyses and reporting
reviewer-response support
reproducible workflows in R
We are often a strong fit for investigators and teams who want more than simple software assistance. Many of our clients are looking for a collaborator who can help think through the statistical side of the project from a research perspective, explain methodological choices clearly, and produce work that can stand up to peer review.
At this time, we are not focused on highly regulated FDA clinical trial work or projects that require the specialized compliance infrastructure typically associated with that environment. Our strengths are better aligned with academic, behavioral, observational, survey-based, and other non-FDA-regulated research settings where rigorous statistical reasoning and clear communication are essential.
Pricing
| Client Type | Typical Services | Rate (USD/hour) |
|---|---|---|
| Tier 1 - Statistical Programming | Data cleaning, manipulation, and reporting | $125/hr |
| Tier 2 - Statistical Consulting | Study design, analysis, and manuscript preparation | $200/hr |
| Tier 3 - Statistical Data Science | Application development, complex machine learning, and data engineering | $250/hr |
View our consultation and payment policies.
Why Choose Us?
Graduate-level statistical expertise
Support from study design through publication
Strong fit for social science and academic research
Clear, reproducible workflows in R
Statistical reasoning that can stand up to peer review
Book an Initial Consultation
This up to one-hour consultation is designed for researchers who want focused statistical guidance on a project, analysis question, or manuscript issue. During the session, we can discuss study design, data structure, analytic options, interpretation, reviewer comments, or next-step recommendations. The goal is to help you leave with a clearer direction and practical guidance for moving forward.
Please provide the following when signing up:
Name, institution, and email
Project title and short description
Main question you want to discuss
Current project stage
Whether data have already been collected
Approximate sample size
Type of data involved
Any relevant materials, such as a manuscript draft, reviewer comments, codebook, output, or code
Data Privacy Policy
We take data privacy seriously and request de-identified data whenever possible. For most statistical consulting projects, a de-identified analytic dataset provides everything needed for analysis while reducing confidentiality risk. Before transfer, clients should remove direct identifiers whenever they are not necessary to the project, including items such as names, contact information, medical record numbers, social security numbers, and other personally identifying details. When appropriate, study IDs or internal record codes should be used in place of direct identifiers, and only the minimum data necessary for the engagement should be shared. Clients remain responsible for confirming that any necessary IRB, HIPAA, institutional, contractual, or other approval requirements have been addressed before data transfer. If a project requires identifiable or especially sensitive data, additional documentation or review may be required before the engagement can proceed.
Need More Information?
We hope to expand in the future, but we are currently only available to clients associated with US academic institutions. If you are affiliated with one of these institutions and are interested in our services, the most direct way to contact us is through the form below.