A: You are somewhat correct in that zero (0) is not a meaningful number as every number used in the LPI scale (1 – 10) indicates an action (or non-action). In addition, my co-author Jim Kouzes and I fundamentally believe that it is actually impossible to never, ever, ever do something. As a number in statistical scales, zero also tends to skew a normal distribution off from "normal" or bell-shaped with more weight given to the bottom of the scale. This makes statistical tests more problematic but also makes the "curve" look off-balance. Even in school, a student who receives an F on an exam gets more than zero points.
To the question of using 1, 2, or 3 to indicate the LPI statement is not applicable…again, it is our belief that it is impossible to never, ever do something. Therefore, we chose to use the number three (3) as a surrogate for that response ("seldom"), leaving the respondent the option of giving a meaningful one (1) or two (2) as a response which would be an affirmative statement that this behavior is almost never or rarely observed. We have found empirically that the number three was the least used response in the ten-point scale.
