Math 216: Statistics for Engineering

See Jeopardy and The Weakest Link especially 

There's a slightly higher-quality version of this talk on YouTube: http://www.youtube.com/watch?v=hVimVzgtD6w.

"anyqs" I'm very interested in how exactly the poll defines a "Republican-leaning independent". Was it self-reported? Were the subjects asked about their views on certain issues, and, if so, how was a "Republican" viewpoint determined? In addition, I'm curious as to what the sample size is for this data, as well as what a rough population number would be. For registered Republicans, I feel like this would be easy to calculate, but it would be rather difficult to determine the population size for "Republican-leaning independents" depending on what method was used.

This is a sister website of the very popular www.speedtest.net, which does many millions of internet download/upload tests every month. Aggregating that data across the world provides some pretty robust statistics on regional and national connectivity speed. You probably never knew South Dakota had the second highest average download rate in the country. http://www.netindex.com/download/2,1/United-States/ I'm guessing that's related to their cheap power and ample space in encouraging data centers to move in.

This is a branch of Computer Science, springboarded by my Computer Engineering major.

I think these maps are much nicer than the color maps that I found in my high school history classes. The 3D aspect of them is much nicer than the heat maps I have become accustomed to. They really allow you to see the how many more people live in some regions than in others. anyqs

The graph of the population distribution of the US is a very interesting statistical representation. There is one outlying area that I noticed. In southern Florida there seems to be a very sparsely populated area and this map brings up the question of why this would be the case. Most of the coast surrounding this area is densely populated, so is there a geographic barrier there that prevents dense human settlement? anyqs

Nuclear Explosions since 1945

anyqs It is interesting to note the relationship between the locations of the nuclear explosions and the population densities around those regions. Furthermore, it seems that there is also a relationship between the frequency of nuclear detonations and how long ago they were detonated (i.e. there were several nuclear detonations 50 years ago, but just 2 in the last 10 years). Lastly, it is interesting to see that the UK and France conducted the majority of their nuclear explosions in their colonies during the 1960s and 1970s, away from their population.

every instance of the phrase "is the new" encountered from various sources in 2005

The question that really jumps out at me is in the section with the greyscale plane diagrams - Do the planes with the most fatalities also have more cumulative flights? The ratio of planes in service to accidents gives some scaling data, but it's entirely possible that the seemingly dangerous Boeing 737 family has made many more cumulative flights than its counterparts. "anyqs?"

In the display of the number of crashes by month, does the thickness of the circle represent anything in the data because some circles are thicker than others, or are the numbers being compared just based on the diameter of the circles?

anyqs In the second display, I'm confused with in ratio of the number of the planes in service to fatal accidents. The ratio is interpreted as an indicator of the safety factor of the plane. However, in my opinion, the larger plane is taking a higher risk of suffering more casualties rather than comparing their in service ratios.

75% #probability in next 4 years of magnitude 7+ quake --------------------------------- Says based on data gathered, so presumably run through some sort of simulation and determined that way. Even though a priori, I think closest to an empirical treatment of probability.

Compares the voted top 100 companies via jobs applications, current number of employees, and amount of jobs created/lost. Overall it is a fairly simple graphic but is interesting to see how these companies compare

anyqs This is a beautiful relative comparison graph. However, it would still be better to see all the companies losing a percentage of their workforce instead of just the number. This would show a better representation of decline or prosperity of a company. for example, is microsoft facing the recession more than anyone else, or did they simply have too many employees before

anyqs This is a nice graph, especially in that it compares multiple variables. One question I have though is who voted on these best companies? What was their definition of best? Also, this is pretty technical, but is the center of the circle or the top of the circle even with the total number of employees?

McCreeley seems to invoke the axiomatic notion of probability, but her view is likely much more subjective.

anyqs This is one of the most creative visualizations I've ever seen. I would be very interested to know more about the point at which Earth simply can't handle more people (as depicted by the nearly overflowing vials). Asia's vial, for example, is nearly filled to the brim by the year 2015, indicating that it will start to overflow soon. Does this mean that Asia's population growth will have to slow down before that of other continents? All the vials are the same size in this representation, but I would be curious to see which vial fills up first if the size of each vial was adjusted to represent a "theoretical maximum capacity" for each continent/country (based on available land, available resources, average consumption rates, etc.). From an execution standpoint, I think it would be useful to see a table of values with population growth rates for each continent/country, particularly because the vials mainly indicate relative rates. anyqs

anyqs This was pretty cool the first time around. I like the use of the glasses, but the way they have China, India and the rest of Asia in separate cups. How does the way they break up the world into regions affect the data? The separations make sense when trade/communication were minimal. But in the last 200 years, the most dramatic time, the importance of location in the data diminishes. Like Irene said, I also thought the "max capacity" of the cups was misleading. I like the part at the end where the world cup it about to overflow but I don't think it works with the individual cups.

from 82games.com, a great site for finding advanced basketball metrics

anyqs: The article says "more women are entering college, which in turn is changing the relative popularity of disciplines," but doesn't say exactly what percentage of degrees awarded in each discipline go to women. I'm wondering what that percentage is. It would have been nice to have seen the number of degrees awarded to women and men in each discipline visualized with the number of degrees awarded.

anyqs: It is true that more students are earning science degrees than they did twenty years ago, but it is also true that more students are earning degrees, period. Note the growth in arts, music, and business degrees awarded. It would be good to know the percentage of the total increase in degrees awarded for each discipline.

anyqs When was the originating date for the number of total persons born data? Is it from the "beginning of time"? Also, how are abortions measured? Are these just registered, legal abortions, or an approximation of self-inflicted abortions?

Click on the link to the online database just below the title of the site and the spinning baseball

Some interesting data on mean salaries for different universitites and different majors

The actual dataset can be found by scrolling down to "Getting the Data" (Section 5).