For chairs & faculty
What your student gets, and what I will not do.
If you are considering referring a candidate, you are entitled to know exactly what outside statistical support looks like here. This page is written for you rather than for them.
Who I am
I am Brian Harward, PhD, a statistics educator and researcher in the United States, working alone. I received the Distinguished Faculty Award for teaching statistics, served as Director of Research at San Diego University for Integrative Studies, and have been a Research Scientist at Ethical Systems, NYU Stern School of Business. I have worked with more than 200 doctoral students through analysis and defense.
There is no team behind me, no offshore analysts, and no subcontracting. Every decision on a student's project is mine. More background
The model
Coaching, not ghost-writing
The student stays in the work
Every analysis is accompanied by coaching in which the student learns the rationale: why this design, why this test, what the assumptions require, and what the results can and cannot claim. If a candidate cannot explain a result, the engagement is not finished.
The student remains the author
I do not write dissertation prose. I support the methods and results (structure, tables, reporting language, and accuracy of claims), and the writing is the candidate's.
Your guidance takes precedence
Where my read of the methods differs from the committee's, the committee's direction governs. I explain my reasoning to the student, and the student decides what to do with it. I am not a second chair, and I do not put candidates in the position of choosing between us.
Defense readiness is the measure
Before a defense I run a mock question session on the candidate's own results. The test I apply is whether the student can answer your committee's questions unaided.
Ethics & disclosure
Ethical, and the student's call
Universities generally permit doctoral candidates to work with a statistical consultant, in the same category as editors, writing centers, and transcription services. Policies differ by institution, and checking them is the student's responsibility, not something I verify. What I control is the kind of support I provide: coaching on their own work, never work passed off as theirs.
Whether a student tells anyone they worked with me is their decision. I do not require disclosure and I do not report engagements to anyone.
AI policy
Human-led analysis
Client data is never uploaded to an AI service, and no analysis decision or piece of advice comes from an AI system. Statistical work here is done by a person who can be questioned about it.
Confidentiality
How student data is handled
- Datasets are used only for the candidate's project and are not shared, resold, or used as examples without written permission.
- De-identified data is requested wherever the design allows it.
- Where a study is under IRB oversight, I work within the approved protocol and its data-handling terms.
- Files are removed at the end of an engagement on request.
Referring a student
How referring works
Send the candidate to the intake form, or send me a note yourself and I will follow up with them. I reply with scope and a flat price in writing, so a referral costs your student nothing and commits them to nothing.
There is no referral fee, in either direction. I would rather you refer students because the work holds up.
If it helps your candidate
- An honest assessment up front, including telling a student they do not need to pay for help.
Questions before you refer anyone
Faculty enquiries are welcome by email, before any student is involved.