Link love

racism is terrible

Want more diversity in your romance novels?

“The kitten, at press time, was very cute. ”  Here’s another kitty.

bookcase and reading nook

ack!  ha ha ha look at that poor grad student

My mom was just telling me this

argh people are so dumb!  (probably even me)

here’s a linky link:  why you hate work


I think Elmo actually knows, he’s just pausing for effect.

this is pretty good advice

For those with kids who want ideas about how to teach financial skills to their teens

More places to clickbait your stuffs.

This is one of the sweetest cutest animes ever made (there’s also a manga if you prefer it that way):

5 Responses to “Link love”

  1. Chelsea Says:

    Totally off topic, but I saw your response about the 2012 ATUS data on Laura’s blog, and I wanted to ask you a question about analyzing the data. I do a lot of data analysis but not with survey data. All I really remember from grad school (and from a friend who is a sociologist) about survey data is a warning that you have to be careful when analyzing it because you have to account for the way the survey sampling was done. So with the ATUS data, do you need to account for the sampling scheme somehow or can you use the data as is and assume the results are generalizable to the US population because of the way the sampling was done? Or do you only have to account for the sampling scheme if you want to generalize to the US population but not if you just want to make statements about the specific study sample? Or something else? Thanks for the response if you have the time!

    • nicoleandmaggie Says:

      I didn’t check to see if ATUS has sampling weights (and I don’t have the data at home with me right now). I didn’t account for weighting in my quick regressions and ttests, though you are right that I should have if I wanted to say something generalizable about the US population. If I were using it for research I would have to look at what the survey documentation says about weighting and then decide whether or not to use weights. (I do this in my actual work work, I promise. Though we don’t cover weighting in my intro to stats course, which is why it wasn’t in my exam answers.)

      The ATUS is built off the CPS and the CPS is supposed to be representative, though it also has various weights that you can use depending on if you’re trying to say something representative about people or about households (so there are person weights and household weights). Usually with the CPS weighting doesn’t make a huge difference, but sometimes it can mean the difference between significance and not. (The HRS, on the other hand, it’s important to use the weights because for various reasons the sampling frame doesn’t match the US population.) The biggest thing that weighting generally affects is the representation of young black males.

      The ATUS also imputes missing values for the major variables of interest, which adds an additional layer. Sometimes these surveys incorporate the weights already and they put in a variable that allows you to undo that, but I don’t know what this survey did.

      Here’s the place I downloaded the ATUS: . Chapter 7 of the documentation is on how you’re supposed to weight, why, and what weights to use. And I didn’t, so my estimates in LV’s comments section will be off. I suppose I should fix that in her ask the grumpies too. Good thing I made a log file when I did that.

      So great question.

  2. chacha1 Says:

    the bookcase reading nook is a win. :-)

  3. Debbie M Says:

    I think you guys might enjoy this: . It’s one of those educational role-reversal things fighting rape culture.

    • Debbie M Says:

      That was a link. Let’s try again. Start with the h and the t and the t and the p and the colon and the two slashes. Then thesaltcollective dot org then /modesty-whensuitsbecomestumblingblock/.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: