€˜€˜Are natural languages neutrally, positively, or negatively biased?€ That is the question a recent study set out to answer. They found the top 5000 words used in English in each of four different places: Twitter, books on the Google Book Project, The New York Times, and music lyrics. The resulting complete list was 10,222 unique words in the English language. Each word was then evaluated independently by 50 different people, each giving a rating on how the word made them feel on a 1 to 9 scale where 1 = least happy, 5 = neutral, and 9 = most happy. (The highest rated word was €˜€˜laughter€ while the lowest was €˜€˜terrorist.€) The distributions of the ratings for all 10,222 words for each of the four media sources were surprisingly similar, and all had approximately the shape shown in Figure 2.14.   

(a) Describe the shape of the distribution.
(b) Which of the following values is closest to the median of the distribution:

(c) Will the mean be smaller or larger than the value you gave for the median in part (b)?

Transcribed Image Text:

1 5 Posltivity Score 3. 3.5 5 6.5 7 7.5