VALIDITY THREATS

Collecting data is important to analyzing and evaluating processes, products, and people. It is imperative, therefore, to ensure that whatever tool is used to measure processes, services, products, or people measures what it is supposed to. Otherwise, inferences from the data could be erroneous.

A number of threats can occur that can skew or bias inferences. These include the time that the evaluation occurred, the tool or technique used, sequence of collection actions, flawed criteria, tendency to generalize from a narrow band of test, and institutional pressures.

One benefit of knowing about the possible threats to validity is that you can quantify the results of your testing or study. Another benefit is that you can take compensatory actions to offset the influences of specific validity threats.

for Dealing with Validity Threats

• Define the goals and objectives of the study or data collection.
• Identify all the potential threats to validity.
• Prioritize those threats, and estimate their probability of occurrence.
• Determine how to account for the influences of those threats.