APA provides the following definition:
a personality trait marked by a calculating attitude toward human relationships and a belief that ends justify means, however ruthless. A Machiavellian is one who views other people more or less as objects to be manipulated in pursuit of his or her goals, if necessary through deliberate deception.
This definition, while establishes that a Machiavellian person believes manipulation is an efficient method of achieving one's goals, tells nothing about an actual willingness to manipulate others. Similarly, a person can believe that having a sexual affair with the boss is an efficient way to get a raise or a promotion. Yet, I expect, the amount of people actually willing to do that is many times lower.
The first standard test created to measure Machiavellianism was MACH-IV. The questions there are depersonalized. For example:
Anyone who completely trusts anyone else is asking for trouble.
While a personalized variant would be:
I don't completely trust anyone.
It's possible that the amount of people responding positively to the first statement would differ from the amount of people responding positively to the second statement. And these two questions therefore would measure different things.
The other issue is that some of the questions have little to do with proneness to manipulate people. Consider the following item:
People suffering from incurable diseases should have the choice of being put painlessly to death.
This question might be measuring negative empathy, because people who want to put themselves to death painlessly likely suffer. Meanwhile, manipulative people are considered to lack empathy. Which, although, also is questionable, because deferred empathy, for example, might help to keep people you have offended. After all, there is a reason that out of 71 original items this one has made it to the selected 20.
Because MACH-IV has problems, another scale, MPS, was created. Unlike in MACH-IV, in MPS most items are personalized. An average score in MPS was significantly lower than an average score in MACH-IV, which is consistent with my prediction.
However, this scale might have its own problems. It's not actually clear if people who are more likely to manipulate others have a significantly higher desire for power or control. The reasons for manipulation can be different, including conflict avoidance, for example, by making up excuses.
Scores of another scale to measure Machiavellianism, FFMI, are relatively weakly correlated with MACH-IV and MPS scores.
One of the facets of this scale is ambitiousness. It's unclear why it was chosen as a facet of Machiavellianism, as defined by APA. While it is true that people in high-ranking positions, such as CEOs, score higher in both MACH-IV, MPS and FFMI, it can't be yet concluded that ambition plays a significant role in it. There can be a lot of ambitious people unwilling to manipulate, but since they are unwilling to manipulate, they might have problems actually satisfying their ambitions.
At the same time, the amount of manipulative people likely is significantly greater than the amount of ambitious people. There likely are more people who overreport the amount of work they have done than there are people who flatter their boss.
A few other scales that measure Dark Triad traits as a whole exist. Research using it shows that Machiavellianism subscale scores are correlated with drug consumption. This might mean that pleasure seeking is a more important trait associated with willingness to manipulate than ambitiousness.
However, yet, since so many different measures, all significantly differing from each others, exist, which ones would be considered more accurate in measuring Machiavellianism, as defined by APA. Would a new scale be required to accurately measure Machiavellianism? How would one approach this? Would the research have to directly ask participants about their willingness and frequency of manipulation? Ask participants what they'd do in a particular situation?
