What it does and how I built it

Firstly I started off by writing Python code that will load in the information from the Excel file to a Python array, in a neat and useful format.

I then collected all the scores from each of the questions for each person. By separating out the introvert questions (i.e. the questions which introverts are more likely to agree with) from the extrovert questions (same thing but for extroverts), I converted the scores to the introvert questions by reversing the response, e.g. 5 (Agree) becomes 1 (Disagree) and vice versa, 4 (Slightly Agree) becomes 2 (Slightly Disagree) and vice versa, and 3 (Neutral) remains as 3. This is similar to adding a 'not' in the introvert questions, hence turning them into extrovert questions, and then reversing the responses so that the original response matches with the negated questions. This is done so that we can take an average of the scores from 1 to 5 for all the questions answered by each person, and to do this we convert these scores to a scale of increasing likelihood that an extrovert will agree with the question from 1 to 5 (with 5 being the most likely an extrovert would agree and 1 being the least likely).

For example, Q1 is an introvert question, as an introvert is more likely to 'never audition to be on a game show'. Hence an extreme introvert would pick '5' - Agree. So, to convert it to an extrovert question (which an extrovert would agree with), the '5' must be converted to a '1' - Disagree, as the extrovert is least likely to agree with this question.

As I mentioned above, I took an average of these scores for each person, so that we could determine the average score of this person on the extrovert scale. I then plotted out the graphs for 10 people to more clearly visualise the distribution of the scores. Since the scores were from 1 to 5, I determined any score above 3 to indicate an extrovert (since they, on average, agreed more with the extrovert questions than disagreed) and any score below 3 to indicate an introvert (since they, on average, agreed less with the extrovert questions than disagreed), and any score equalling 3 would be neither, since they, on average, agreed as much on the extrovert questions as they disagreed. Using this criteria, I was able to separate out the extroverts, introverts and the neither-verts (and how many of each there were) for all 7188 participants.

Since we were also given the data on whether each person identified as an introvert, extrovert or neither, I compared these results to the results I determined for each person being an extrovert, introvert or neither. Through this comparison, I was able to find which people were lying and which people were telling the truth, and the number of liers and truth-tellers.