Liar’s Index

Goal: To come up with a blueprint for measuring one’s capacity to give alternative facts.

In light of our honorable new leader’s honorable policy on honesty, let’s see if we can’t think through a bigly way to measure it. Firstly, some base cases:
• If you ask someone who’s 100% honest if they will tell the truth 100% of the time, they will say “Yes” 100% of the time.
• If you ask someone who’s 100% dishonest if they will tell the truth 100% of the time, they will still say “Yes” 100% of the time.
• If you ask someone who’s 50% honest if they will tell the truth 100% of the time, they will say “Yes” 50% of the time, and “No” 50% of the time.
• If you ask someone who’s 50% honest if they will tell the truth 50% of the time, they will say “Yes” 50% of the time, and “No” 50% of the time.
• If you ask someone who’s 75% honest if they will tell the truth 75% of the time, they will say “Yes” 75% of the time, and “No” 25% of the time.

Based on the above we can come up with a more generic formula:
• If you ask someone who’s X% honest if they will tell the truth Y% of the time, they will say “Yes” X% of the time.

Variable Y above has no bearing on the results, so let’s remove it:
• If you ask someone who’s X% honest if they tell the truth, they will say “Yes” X% of the time.

Next step is to factor in sample size and the law of averages (i.e. asking several questions will yield better results than asking one question):
• If you ask someone who’s X% honest once if they will tell the truth, they will either say “Yes” once or “No” once
• If you ask someone who’s X% honest 10x if they will tell the truth, they will either say “Yes” (X*10/100) times or “No” 10-(X*10/100) times
• If you ask someone who’s X% honest, 100x if they tell the truth, they will either say “Yes” (X*100/100) times or “No” 100-(X*100/100) times

For the sake of this exercise, we’ll pick a sample size of 100 questions.
Unfortunately, asking someone straight out if they’re honest or dishonest is not verifiable. Therefore we need to replace the question with something that could be verifiable with evidence, something like:
• If you ask someone who’s X% honest, 100x if they’ve ever made fun of a disabled reporter, they will either say “Yes” (X*100/100) times or “No” 100-(X*100/100) times

However, each question in the sample size must be unique, so as to avoid bias towards one particular subject or line of questioning:
• Q1: If you ask someone who’s X% honest if they’ve ever made fun of a disabled reporter, they will either say “Yes” or “No” once
• Q2: If you ask someone who’s X% honest if they’ve ever made fun of a woman, they will either say “Yes” or “No” once
• Q3: If you ask someone who’s X% honest if they’ve ever made fun of an immigrant, they will either say “Yes” or “No” once
• Q4: If you ask someone who’s X% honest if they’ve ever made fun of a child, they will either say “Yes” or “No” once
• …up to 100 questions total
Liar’s Index: Total # of “No” answers/Total # of questions

The next step, perhaps the most difficult, is to factor in the person’s personal motivation behind answering each question dishonestly. What does the person have to gain or lose by lying on a particular question?
For simplicity let’s normalize all questions so that “YES” denotes honesty and “NO” denotes dishonesty, then use the following motivational scale (abbreviated MO):
MO 0 – Either answering honestly, or no impact for lying. (e.g. “Did you drink a glass of orange juice this morning? If not then oh well.”)
MO 1-4 – Minimal impact (e.g. “Did you eat fast food today? If so then people may question your commitment to a healthy lifestyle, a little bit.”)
MO 5 – Medium impact (e.g. “Did you make fun of a disabled reporter? If so then you may be seen as an insensitive human being, which could jeopardize your chances for re-election.”)
MO 6-9 – High impact (e.g. “Does a foreign entity have incriminating material on you? If so then you may have your brand tarnished, destroy your chances of remaining president or getting re-elected, and possibly bring criminal charges to you or your family sometime in the future.”)
MO 10 – Critical impact (e.g. “Did you strangle that adorable puppy for peeing on your leg? In front of your entire staff? On national television?”)
Liar’s Index: (Q1’s MO+ Q2’s MO+Q3’s MO+…+Q100’s MO)/1000

One additional factor that could be counted separately from the motivational index is magnitude. The magnitude of a person’s lie may/may not be different from that person’s motivation behind that lie. For example, something like an administration’s intentions towards a foreign country’s trade or immigration policy may have severe implications for multiple countries and millions of individuals, but have relatively little bearing on that individual’s personal motivations.

So factoring in a magnitude scale (abbreviated MA) of 0-10, with 10 being highest,
Final Formula for Liar’s Index: (Q1’s MI+ Q2’s MI+Q3’s MI+…+Q100’s MI)/1000+(Q1’s MA+ Q2’s MA+Q3’s MA+…+Q100’s MA)/1000

Next step is to plug some test subjects into this formula to see what gets spit out on the other end. Wonder what President Lincoln would have gotten?

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: