Research & QI Center

When analyzing data is is important to be aware of and avoid bias.

• Catalog of Bias: A collaborative project mapping all the biases that affect health care evidence

How to Avoid Bias in Scientific Test - Stile Education, Dec. 7, 2020.
Recommend: turn cc on if you can't use sound.

Surveys are susceptible to data collection errors and bias as well.

Proper data collection begins at the project design phase. Good project design can help avoid common data errors in sample, selection, size, response rate and other factors that will ultimately affect the quality of the research and results you produce.

Common data collection errors (bias):

• Population specification - not selecting the proper population for the intended study
• Sampling and sample frame error - incorrect subpopulation selection
• Selection - using a non-random sampling method
• Non-responsive - low response rate can greatly affect results
• Measurement - survey does not measure what you intended or misses the population

From: 5 Common Errors in the Research process

Selection of proper data collection and analysis tools can also aid in effective outcomes. Some of the resources available for HonorHealth researchers (contact your program director for more information):

Survey Tools:

• Qualtrics - advanced survey research, requires authorization
• Survey Monkey - basic survey research

Statistical tools:

• SPSS - advanced statistical resource, has a learning curve, requires license
• Curt Bay - consultant for HonorHealth (contact information to come)

 Use Zotero to keep your research & citations in one place.  Zotero also saves you time with automatic citation formatting.

Get Zotero (Harvard University Library guide)

Start Using Zotero (Harvard University Library guide)

Analytic Services from HonorHealth

The Analytic Services team provides reporting, visualization, management, and governance of data that supports, but is not limited to, operations, clinical care, patient experience, quality improvement, and patient safety activities.  The team also extracts and manages data from a variety of sources but primarily the Epic Clarity database, Epic Caboodle data warehouse, and Midas database.