The Accelerate book uses surveys as one of their primary means of gathering information. They were one of the first in broad knowledge within the industry to do so. Many of their questions use the Likert scale.
They run their survey through analytic constructs. “construct” in this case is a grouping of multiple individual statistical tests and putting them under a single name.
Discriminant validity : making sure that items that are not supposed to be related are actually unrelated (e.g., that items that we believe are not capturing organizational culture are not, in fact, related to it).
Convergent validity : making sure that items that are supposed to be related are actually related (e.g., if measures are supposed to measure organizational culture, they do measure it).
Reliability : making sure the items are read and interpreted similarly by those who take the survey. This is also referred to as internal consistency.