2014

BACK TO BASICS

Data Collection Guidelines: The People Element

by Scott A. Laman

Collecting high quality data is essential to the success of any project, process improvement or new product development. People, especially operations employees familiar with an actual situation, often play key roles in data collection.

The following guidelines were developed specifically to help obtain data with operations personnel.

The People

To begin, employees involved in the collection process must remember the process—not individual employees—is being evaluated. Employees don’t need to look for hidden management motives. All are encouraged to contribute ideas to the planning and execution of the data collection.

Employees also must be aware that honest data collection helps the company. Fabricating numbers when data are missing doesn’t help. It’s better to leave a space blank on a form than to guess the number. Every employee must understand what he or she is being asked to do and shouldn’t hesitate to ask for clarification if something doesn’t make sense.

If two employees do things differently, these differences might affect the data. There is nothing wrong with that. However, if the different methods affect process, it should be noted and perhaps a best practice suggested. If necessary, bring the subject up with the supervisor or engineer or during a team meeting. This isn’t an issue of right and wrong but of reducing variation so the data are valid.

If something happens that might affect data collection, write it down, preferably on the form provided during the data collection process. Make sure the tools being used for measurements are calibrated and in good condition. Remember the document control requirements of ISO 9001 and your company’s quality system. Make sure the data collection form is compliant to whatever regulations apply to your business.

Other Considerations

Further considerations in planning data collection activities include:

  • Identify the problem. Put a clear statement in writing. Gather data to specifically answer the problem statement. In other words, ask yourself: Where is the problem occurring? Where was it first noticed? How often does it occur? What products and sizes are affected? Are certain process conditions perceived to be the cause of the problem?
  • Find any historical data that might be helpful.
  • Define exactly what is to be measured. List all important characteristics.
  • Select the measurement technique. Make sure it is as precise and accurate as needed.
  • Design an uncomplicated data form. Have traceability to data collection details such as what time the data were collected, what tool was used and who did the measuring. Allow space for comments. Unusual circumstances during data collection must be recorded.
  • Decide who will collect the data. Define any training needs or consistency checks.
  • Choose an appropriate sampling method.
  • Do a measurement error study.
  • Decide who will analyze, interpret and report the results.
  • Run a short pilot to ensure people understand how to collect and record data. Pilot the analysis to test the adequacy of the information.

To prevent misusing the data and drawing incorrect conclusions, be aware of several potential problem areas:

Done but incomplete. Experiments can end but might be incomplete if you don’t know what you want. Good data are not really good if they do not answer the key question.

Human error. Mistakes can happen when recording information. Proper training and a well designed data collection form can address this potential problem.

Calibration. Tools and machines also must be in top condition. Make sure they are calibrated or verified.

Data analysis. Know what analysis you are doing, even before the first data point is collected. Misusing historical data doesn’t help. Make sure it is relevant.

Misuse of a gut feeling. It is OK to have a theory; just make sure it is supported by data before taking it too far.

Not enough data. Collect data for a long enough time period to introduce all sources of variation.

Data collection is just too important to take lightly. The time spent planning and communicating with all parties involved is well worth the effort. Use these tips on data collection as starting points to gather excellent data. 


SCOTT A. LAMAN is a manager of quality engineering for Arrow International Inc. in Reading, PA. He earned a master’s degree in chemical engineering from Syracuse University. He is a senior member of ASQ and is certified as a quality engineer, reliability engineer, quality manager, Six Sigma Black Belt and quality auditor. He is a member of the ASQ Certification Board and is the past chair for the certified reliability engineer exam and the present chair for new exam development.


Average Rating

Rating

Out of 0 Ratings
Rate this article

Add Comments

View comments
Comments FAQ


Featured advertisers