The philosophy of 仕方がない

I read a book review of Kimi Cunningham's Silver lake dust: One family’s story of America’s Japanese Internment a couple weeks ago and requested the book from the local library. I got the book last night. In spite of my busy schedule, I started reading it as soon as I returned from the library. I finished the book just one evening!

It is a book that you just cannot put it down. It is a memoir of the author's grandmother who lived in Heart Mountain Japanese Internment Camp during the WWII. As she promised her grandmother, the author did not unveil her grandmother's name, simply addressed her as Obaasan (grandma in Japanese). The conversation between the author, Kimi and her grandmother just keeps holding my interests.

It is a personal memoir, yet it is also a book that offers a better understanding of the Japanese shikata ga nai 仕方がない philosophy.

Grading on a curve?

The practice of grading student's progress on a curve is derived from the concept of statistical distributions of scores (bell curve). This common practice in education is a method of assigning grades designed to yield a desired distribution (K12 Academics, 2012). There are pros and cons among educators and students. Some educators consider the curve grading system a way to motivate “lagging students” (Richert, 2012), while some teachers use this method to divide students into three categories, above average, average, and below average. The frequency distribution of the number of students in these categories can be determined learning improvement and advancement. Based on a conversation with a teacher who is in favor of curve grading, the bell curve grading in fact assist him how to write recommendation letters. In general, hard-working students would not welcome bell curve grading and consider it as an unfair method or even a "punishment for overachievers." According to Miller (2009), the main problem with this style, is grading students who are between two grades.

Here are some of the examples that I think bell curve grading does not reflect students’ performance in the class.

Scenario 1: A legal studies instructor addressed his class that he uses the bell curve grading system. The top 40% of the class received an A, 50% would have a B, and the remaining 10% would get a C. At the end of the semester, Lisa had a final average of 94.5%. She had no doubt that she would get an A. However, when the grades were posted, Lisa was shocked to discover that she had received a B. She inquired about her grade, and learned that the cut off score was 97%. This revealed that 40% of her classmates scored 97% or above for their overall assignment, which places Lisa's, score of 94.5% in the 50% (B) group.

Scenario 2: In an Algebra class, 80% of the students scored 50% an exam. The highest score was 60%. With the bell curve grading system, the instructor would inevitably give the students an A even though they only scored 60%.   

Can Twitter predict political election results?

I have compared the 2010 Belgian general election, 2011 Singapore general election, and 2011 Irish general election results with Twitter messages. It appears that Twitter users and resulting tweets might in fact play a role in the political elections; although, I will argue that Twitter cannot predict political election results for the following reasons:

• ·         The number of Twitter users is only a small fraction of the voting population, according to the sampling percentages; Belgium 0.11%, Germany 0.24%, and Singapore 4.45%, this prevents an adequate representative sampling.
• ·         When using tweets to conduct scientific research, data miners need to identify the variables of the sample such as the various strata of users based on age, gender, income, political affiliation, technical knowledge, etc. The profiles of Twitter users should play a significant role in sampling research.
• ·         The validity of tweets is questionable due to the fact users can use a pseudo name; therefore, it is possible that tweets come from the same person.
• ·         Even though Twitter users post their viewpoint in favor of one political party, they do not necessarily have a negative sentiment against other parties. How can researchers distinguish or classify the negative and positive roles that tweets play on election results.
• ·         If retweets were counted as separate data for the study, this data probably would skew the analysis.
• ·         A political party might be the center of the discussion among Twitter users, thus the party largely received the number of positive and negative tweets. However, it is not necessary an indicator for the election result; such as the People's Action Party in Singapore.
• ·         We have to bear in mind that popularity is not necessary equal to prediction. The prediction of future events needs comprehensive and in-depth scientific research, not just numbers or contents of the tweets received.
• ·         This research dominantly used quantitative analysis and statistical inference methods. Utilizing combined qualitative methods for these studies might reveal different results.
•  

In my view, the study of tweets and the current events is an interesting research area. From these included studies, we can see the relevance of tweets in the political arena. One cannot draw a conclusion that tweets enable prediction of future events, such as election results; that could require detailed studies.