Skip to main content

Posts

Showing posts from February, 2014

Pew's "games"

One of my favorite things to do with students is take a break and take Pew's quizzes as a class-- they usually have a good time see how they do relative to all the people who take the quiz and I get them to look closely at numbers and statistics!

Pew Research Center Quizzes
Pew Research Center Interactives

Remember that you can refine results by projects such as US Politics, Media & News, Religion, Internet & Tech, Global, Social Trends, and Hispanic.

Enjoy!

places to access the GSS

There are several places to access the GSS. The main site of the GSS is here: http://www.norc.org/GSS+Website/
On this site, do the following to quickly access some basic information about GSS variables
Browse VariablesSubject index -> A Choose abortion Spousal consent Notes Section—click “here” to see associated publications with variable. To look at the questionnaire, click Publications and in the drop down menu choose GSS questionnaires.
V1-V6: Sections A-K- Core, Ballots 1-6—look at the Manual Map Another popular place to access the GSS is Berkeley's Archive: http://sda.berkeley.edu/archive.htm
Choose GSS 1972-2006 – Quick TablesClick on Politics and Voting by Background Variables Explore the column variables (background variable) Explore the row variables (political id.) Choose Bar Chart You may also want to try the following:
go back to GSS 1972-2006 – Cumulative DatafileRow (choose a variable you’re interested in)—notice that the names are very specific-- "abany"…

Study Finds Gender Gap in Doctors' Pay - NYTimes.com

What a troubling report- even in such "meritocratic" careers such as medicine, the gender gap persists: http://well.blogs.nytimes.com/2012/06/28/among-doctors-too-women-are-paid-less/. It may be counter-intuitive, but many studies have actually found that higher levels of education equates with a larger gender gap in earnings.
As in some earlier studies, the researchers found a difference in income, with a male doctor's annual salary averaging just over $200,000 and a female's averaging about $168,000. And like previous researchers, they found that the female doctors tended to be in lower-paying specialties, have fewer publications, work fewer hours and hold fewer administrative leadership positions.  But when these researchers ran the numbers again, this time adjusting for differences in specialty, publications, academic rank, hours worked and leadership positions, they found that the expected average salary for women still fell behind that of their male colleagues…

reading a methodology section from Gallup's survey on Confidence in U.S. Medical System

This comes from this Gallup poll: http://www.gallup.com/poll/155381/Confidence-Medical-System-Health-Law-Passed.aspx

Pay particular attention to the methodology section (reproduced below). Do you recognize the statistical terms?
Survey Methods
Results for this Gallup poll are based on telephone interviews conducted June 7-10, 2012, with a random sample of 1,004 adults, aged 18 and older, living in all 50 U.S. states and the District of Columbia.  For results based on the total sample of national adults, one can say with 95% confidence that the maximum margin of sampling error is ±4 percentage points.  Interviews are conducted with respondents on landline telephones and cellular phones, with interviews conducted in Spanish for respondents who are primarily Spanish-speaking. Each sample includes a minimum quota of 400 cell phone respondents and 600 landline respondents per 1,000 national adults, with additional minimum quotas among landline respondents by region. Landline telephone numbe…

r, r^2, and R^2

R's can get confusing. There is r, then there is r^2, and finally, there is R^2. I haven't been able to find one place that concisely clarifies these correlation statistics so here it is.


income inequality as seen from space

A very interetsing post, think stats and GIS.
Income inequality, as seen from space
by Tim De Chant, persquaremile.comMay 24th 2012 Last week, I wrote about how urban trees—or the lack thereof—can reveal income inequality. After writing that article, I was curious, could I actually see income inequality from space? It turned out to be easier than I expected.
Below are satellite images from Google Earth that show two neighborhoods from a selection of cities around the world. In case it isn't obvious, the first image is the less well-off neighborhood, the second the wealthier one.

are you better off than your parents?

Here's a link to the Bureau of Labor Statistics CPI calculator. You may also want to consider this article and the accompanying graphic: A Rough Ten Years for the Middle Class


So are you better off than your parents?

Worldmapper: The world as you've never seen it before

What a great site! Have a look at these maps redrawn based on specified criteria, such as land mass, poverty, population, etc. Their website describes the site in the following way:

Worldmapper is a collection of world maps, where territories are re-sized on each map according to the subject of interest.

There are now nearly 700 maps. Maps 1-366 are also available as PDF posters. Use the menu above to find a map of interest. To learn a little more about this and other map projections read this: Worldmapper and map projections.

They recently added a new series of maps of the world's population using a new mapping technique: Worldmapper Population Atlas

Here is an interactive map of the world's population living on less than $1/day to those earning over $200/day.

© Copyright 2006 SASI Group (University of Sheffield) and Mark Newman (University of Michigan)

Visit WorldMapper for other great maps!

stats on Coursera

Coursera is just one of many fast-growing MOOC (Massive Open Online Course) sites. There's a fantastic stats class available here (https://www.coursera.org/course/stats1) taught by Professor Andrew Conway, a senior lecturer at Princeton University. Have a look and keep in mind that this is just one of the many stats resources available (often for free) online.

I should point out that the course uses R as the software for statistical analysis. R is a powerful package, but can be intimidating for some. I should further add that although I have not actually taken the Coursera course, I did glance at it and my initial impressions are that it's pretty good (I also glanced at Udacity's Stat's class and liked it less largely because I felt the same way as the AngryMath person but understand that as this post makes clear, the goals of Udacity and AngryMath person may be different and my goals align more with the AngryMath person).

descriptive comments in SPSS syntax file

Once you become accustomed to working with a syntax in SPSS, you may prefer to work with the syntax file rather than the familiar point-and-click method. To help you make sense of your syntax, you may wish to add descriptive comments to the syntax-- these are called "Comments" in SPSS. A Comment is descriptive text that doesn't tell SPSS to do anything. But you have to write such comments in a particular way to differentiate them from SPSS commands. If you don't do this properly, SPSS will interpret your text as a command that it doesn't understand and you'll receive an error code. Here are 3 ways you may write Comments to differentiate them from commands in SPSS:
Comment method- Write the word COMMENT (in caps) at the beginning, write your comment, and end it with a period:COMMENT This is a comment and will not be executed.Asterisk method-Use an asterisk at the beginning, write your comment, and end with a period: * This is a comment and will continue to be a…

confidence intervals, margin of error, alphas, and p-values

There's often a lot of confusion surrounding the terms confidence interval, margin of error, alphas, and p-values.

Let's begin with a confidence interval. The confidence interval formula gives you an interval that is likely to include your population value. A margin of error is really a part of the equation of confidence intervals in statistics. Here's the formula for confidence intervals:

Notice that the margin of error calculation actually includes a Z score. Moreover, the Z score is determined by the alpha (╬▒) selected by the researcher. So while we're used to seeing the margin of error reported, it would be more accurate to also report the alpha selected by the researcher. Remember, the ╬▒determines the risk of being right/wrong and is selected by the researcher. You could still obtain a margin of error with a risk of being wrong 98% of the time! That's one of the reasons why it's important to report the ╬▒. When we see margin of errors reported in popular med…

Khan Academy on the Central Limit Theorem and Law of Large Numbers

One of the most exciting changes in education in recent years is the explosion of interest in making learning and lessons freely available on the Internet. Khan Academy is a non-profit organization offering many opportunities for learning almost anything you would like to learn.

Here's a video from Khan Academy about the Central Limit Theorem:

The full lesson with comments is available here.

If you would like to compare the Central Limit Theorem to the Law of Large Numbers, watch this video:

Again, the full lesson comes from here on Khan Academy.

free alternative to SPSS - PSPP

[updated 7/31/2012]
This is a GNU alternative to SPSS. I have not played with it some can't say much about it, but it's suppose to be very similar to SPSS so you may want to try it out. At some point, I'll try to test it and see how it compares. I've tried out the software and it's easy to install and works quite well. It resembles SPSS and while there may be some bells and whistles missing, it should be more than enough for the requirements of a basic undergrad class.

The software's main page is here: http://www.gnu.org/software/pspp/ This site contains guides, etc. and is pretty informative.

If you just want to download and install the software, visit this page for the Windows version: http://pspp.awardspace.com/. Choose the appropriate package for your operating system (32bit or 64bit) and download the latest version. This package includes an installer package so it should be just like installing any other kind of software.

I don't use Mac so I can not s…

options for accessing SPSS

SPSS is a very expensive software package and that makes student access to it a recurring problem. SPSS is accessible to student on nearly all computers on campus.

But if you are one of those students who must have it on your home computer, here are the options available to you:

First off, there are several SPSS packages with many add-on modules. The idea is that starting from a 'base,' consumers can expand the software to fit their needs. SPSS also offers Academic editions such as a 'Grad Pack' or a 'Faculty Pack' that vary in price. These packs also come in Base, Standard, or Premium versions. For most undergraduate introductory statistics courses the Base pack is all you'll need.

Both the Grad Pack and Faculty Pack (all editions) are significantly discounted from the commercial edition and may be rented from various places including E-Academy's OntheHub eStore (6-month for $40 or 12-mths for $70)StudentDiscounts.com (12-month for $100)Studica.com (12-…

Fox News and charts

A great graphic of how important a visual can be.
Fox News tried to show the change in the top tax rate if the Bush tax cuts expire, so they showed the rate now and what'd it be in 2013. Wow, it'll be around five times higher.

Wait. No.

The value axis starts at 34 percent instead of zero, which you don't do with bar charts, because length is the visual cue. That is to say, when you look at this chart, you compare how high each bar is. Fox News might as well have started the vertical axis at 34.9 percent. That would've been more dramatic.

Here's what the bar chart is supposed to look like:

With a difference of 4.6 percentage points, the change doesn't look so crazy.

[via Effective Graphs] http://flowingdata.com/2012/08/06/fox-news-continues-charting-excellence/

a normal distribution and a paranormal distribution...

This is one of my favorite social media discoveries...see, stats can be fun!

switching between string and numeric views in the SPSS data editor

Usually, the default view in SPSS's data editor window is something like this:

If you prefer to view string information instead of numbers, it's easy to switch the view. In SPSS 21, just click this icon in the menu bar.
If you're using an earlier version of SPSS, use Edit -> Options and on the General tab, Under Variable Lists, select "Display labels" for numeric view and "Display Names" for the string view.