I recently came across a reference to a phenomenon that got my interest. It is termed the Nocebo Effect. The reference was in a story of a patient who was erroneously informed that he had terminal liver cancer. The patient died, and the subsequent autopsy revealed that there was no cancer. (link to the story omitted due to a "subscribe to read the whole story")
I think that a big reason for my interest lies in my recent reflection on the power of the mind in programming behavior. The fact that the Matrix was on TV tonight didn't hurt much either. I have heard a lot of talk in the past few months about the value of affirmations. I have seen in my own practice how daily reading of a desired set of behaviors can have a positive effect.
Years ago I read about a study of patients' reactions to being informed of terminal illness. I recall the study grouped the patients into three broad groups. One group basically gave up, one group accepted their fate, and the third group basically dug in their heels and fought. I don't recall all of the statistics that were mentioned, but I believe the third group had around a 75% survival rate. The other two groups didn't fare anywhere near as well. The mind is a powerful thing.
Saturday, April 17, 2010
Monday, April 05, 2010
Statistics
Statistics are a favorite of mine. I like looking at sports statistics. I like delving into the meaning behind the numbers and what they mean. I like to look at if the stats are relevant. There are plenty of irrelevant statistics in sports, but they can still be fun. Some of them are just too absurd to really pay attention to, like wins in January, or points scored on Tuesdays.
I really like when statistics are meaningful and paint a picture of who won and why. In volleyball a team with a lot of serve receive errors is going to have bad passing, and they will likely have poor hitting efficiency as a result. Conversely the team with little to no serve receive errors and good hitting efficiency is going to have really good passing, and they will roll over a lot of teams. To really get something from stats you have to know what they measure and what they don't measure. It helps even more to know enough about what is being measured to know if it is important. Also you need to know if there are going to be lurking variables explaining things.
I just came across a story about Nintendo that had some really cool stats. Unfortunately they were all garbage. The stats were showing the top three most popular Nintendo franchises and the number of games sold. Number one was Mario with a whopping 240 million games sold. As a point of contrast Halo was included with a measly 27 million. The guys looking at the page and not thinking (hopefully not the majority, but I think not) will visualize 240 million copies of whatever Mario title is available on the Wii. They will think that there were 240 million Wii consoles with guys playing Mario compared to 27 million Xbox consoles with guys playing Halo.
The little nagging detail that is not included in that lovely little slice of the internet is the fact that there is an unbelievable number of Mario titles. Even only counting the Wii and DS there are about 30 Mario titles. The Halo titles ring up to a handful (5), but only 3 are Xbox 360 titles. Sure the Mario titles sell very well, but the stats are misleading. They may as well post how a certain percentage of best sellers are Mario titles and suggest that if it has Mario in it it will automatically sell. The Mario/Halp discrepancy can be further explained by looking at culture. Xbox is a non-entity in Japan. That is a big market for video games. Nintendo sits at the other end of the spectrum, and it really plays to the video game playing demographic in Japan. That combined with the sheer number of Mario titles explains the sales discrepancy far better than any implied notion of which games are "better". Fun statistics, but they don't really say much that is meaningful.
I really like when statistics are meaningful and paint a picture of who won and why. In volleyball a team with a lot of serve receive errors is going to have bad passing, and they will likely have poor hitting efficiency as a result. Conversely the team with little to no serve receive errors and good hitting efficiency is going to have really good passing, and they will roll over a lot of teams. To really get something from stats you have to know what they measure and what they don't measure. It helps even more to know enough about what is being measured to know if it is important. Also you need to know if there are going to be lurking variables explaining things.
I just came across a story about Nintendo that had some really cool stats. Unfortunately they were all garbage. The stats were showing the top three most popular Nintendo franchises and the number of games sold. Number one was Mario with a whopping 240 million games sold. As a point of contrast Halo was included with a measly 27 million. The guys looking at the page and not thinking (hopefully not the majority, but I think not) will visualize 240 million copies of whatever Mario title is available on the Wii. They will think that there were 240 million Wii consoles with guys playing Mario compared to 27 million Xbox consoles with guys playing Halo.
The little nagging detail that is not included in that lovely little slice of the internet is the fact that there is an unbelievable number of Mario titles. Even only counting the Wii and DS there are about 30 Mario titles. The Halo titles ring up to a handful (5), but only 3 are Xbox 360 titles. Sure the Mario titles sell very well, but the stats are misleading. They may as well post how a certain percentage of best sellers are Mario titles and suggest that if it has Mario in it it will automatically sell. The Mario/Halp discrepancy can be further explained by looking at culture. Xbox is a non-entity in Japan. That is a big market for video games. Nintendo sits at the other end of the spectrum, and it really plays to the video game playing demographic in Japan. That combined with the sheer number of Mario titles explains the sales discrepancy far better than any implied notion of which games are "better". Fun statistics, but they don't really say much that is meaningful.
Subscribe to:
Posts (Atom)