Biostatistics & Study Design

Parkland’s Office of Research Administration promotes a research model in which the statistician is a partner in the research effort. As such, we can be of best service when we are involved early in the research effort. Resources exist to provide research and education support by offering services and collaboration in:

• Hypothesis generation
• Study design, sample size and power calculations, analytic plan
• Statistical analysis of preliminary data
• Other data management issues (e.g. validation) – data entry services not available
• Statistical analysis of data
• Organizing your data for capture into a database
• Advice on survey design and data collection methods
• Advice on data sources
• Advice on preparing manuscripts and responding to journal reviews

Grant Support

Investigators submitting requests for statistical assistance with research grant preparation are expected to include a budget for statistical effort in the grant. Grant support is typically identified as a % FTE.

Authorship

The authorship list on a scientific publication identifies those who are responsible for the integrity of the results as well as those who should be credited for the findings. Thus, co-authorship is appropriate when the statistician has made a substantial intellectual contribution to the design, analysis, or interpretation of the results. The assignment of authorship is a matter of scientific integrity and is independent of funding. In general, the biostatistics group cannot exchange for authorship in lieu of payment for services, and payment for services does not preclude authorship credit.

Request a Consultation

Contact information

Health System Research
T: 214-590-4642 or 214-590-1074
E: Lonnie.Roy@phhs.org or Larry.Brown@phhs.org
Chase Bank Bldg, Ste 240
6300 Harry Hines Blvd.
Dallas, TX 75235

What to bring (if available)

• Define the single most important question to be answered by the study.
• Get as much information as possible about what you expect to find. Bring expected and hypothesized parameters for your variables. (means, standard deviations, percentages)
• Define your effect size: difference in proportions, change in means etc.
• Decide on a (p-value for rejecting the null hypothesis), the power of the study, and what your maximum feasible sample size would be.
• Consider important subgroups of the study population.
• Consider whether or not there are multiple important comparisons that will need to be performed during data analysis
• Allow ample time for discussions with the statistician during proposal development and for data analysis.
• If possible, send a draft proposal to the statistician prior to your meeting

For more information regarding data for researchers, visit Conducting Your Study: Data Services or email ResearchData@phhs.org.