Consulting Lab

Who we engage with

All Kansas State University faculty, staff, graduate students and undergraduates actively engaged in research may seek statistical support through the Statistical Consulting Laboratory. Faculty with Kansas State Research & Extension (KSRE) appointment and their research teams will have priority if time or resources are constrained.
In order to enable a productive dialogue, understanding of basic statistics concepts will be expected of researchers seeking assistance from the Statistical Consulting Laboratory.

How we engage with research

We highly encourage researchers to engage the Statistical Consulting Laboratory prior to conduction of a research study.

“To consult the statistician after an experiment is finished is often merely to ask her to conduct a post mortem examination. She can perhaps say what the experiment died of.”

Sir Ronald Fisher (1890 - 1962)

Contributions if engaged during planning of a study (HIGHLY preferred): Fine-tuning of research question, assistance with efficient design of experiments, computation of statistical power and determination of required sample size. In addition, if engaged during planning stages, the Statistical Consulting Laboratory will be better positioned for data analysis.

Contributions if engaged after data collection: Assistance with choice of proper statistical methods, model specification and fitting, software implementation (mostly SAS and R), interpretation of results, advice on preparation of figures for presentation and publication, preparation of research manuscripts for publication.

Direct engagement of the Statistical Consulting Laboratory with research projects is mediated primarily by Statistics graduate students with supervision from Statistics faculty.

On a case-by-case basis, Statistics faculty may engage in collaborative research with a wide variety of Kansas State University researchers to apply established statistical methods to novel situations or to develop new statistical methods motivated by specific problems.

What will be expected

Any time that a member of the Statistical Consulting Laboratory is involved in any substantial way with a research study through experimental design, data analysis, interpretation of results and/or manuscript preparation, it will be expected that such contribution will be recognized in the form of co-authorship in any peer-reviewed publications and/or in conference abstracts or proceedings, as relevant for each discipline.
Similarly, any time that a member of the Statistical Consulting Laboratory is asked to be involved in any substantial way with preparation of a grant proposal (i.e. experimental design, sample size calculation, power analyses, etc.) or to provide a letter of statistical support for experiments on the grant, it will be expected that a corresponding budget line will be incorporated into the grant proposal to ensure proper coverage of expenses related to grant work. The specifics of the budget line will depend on the scope of the proposed work. Researchers should discussed budgetary issues in advance and directly with the Director of the Statistical Consultling Laboratory.

The Statistical Consulting Laboratory does not provide data management service. But we are happy to review data entry form(s) and offer suggestions on database construction. We encourage the submission of historical datasets or data entry templates as additional resources to aid in the better understanding of the presented problem. Before sending data for statistical analyses, please perform data quality control. In particular, variables (columns) need to be consistently formatted; missing values should be clearly labelled; please do not merge cells or color-code the data values.

The Statistical Consulting Laboratory unfortunately cannot help with inquiries about SAS or R programming. Neither do we develop SAS macros or R packages for broad implementation. All of our coding applies specifically to each submitted consulting request associated with a given study. We do not guarantee its feasibility under general settings. Programming code will be provided to clients upon request. Please do not circulate them outside the KSU campus.

Resources that might be helpful:

We are not engaged in teaching statistics concepts, such as interpretation of p-values, calculation of confidence interval, estimation of random effects, mathematical reasoning behind experimental design, etc. The K-State courses listed below could be of use to students interested in learning more about statistics

STAT 325: Introduction to Statistics
STAT 705: Regression and Analysis of Variance
STAT 720: Design of Experiments
STAT 870: Analysis of Messy Data

Statistical Consulting Lab Policies

Who we engage with

How we engage with research

What will be expected

How we function

What we are NOT