13 Comments
User's avatar
Eric Rasmusen's avatar

This makes me think of a paper we never got published- poli sci journals hated it-- with a lot of methodology you'd fine useful, since the math problem is exactly the same as finding the best short IQ test for measuring g. In our case, we wanted to find the best N questions for measuring conservative-liberal, for N= 1,...10. At minimum, you should discover the best ONE-ITEM IQ test, which would be a very fun result.

Voter Ideology: Regression Measurement of Position on the

Left-Right Spectrum

June 28, 2016

J. Mark Ramseyer and Eric B. Rasmusen

Abstract

For scholars who need a measure of political preferences, a person’s position on the ideological

spectrum provides a good start. Typically, scholars identify that position through factor analysis on

survey questions. In effect, they assume that the calculated synthetic variable marks the person’s location

on the liberal-conservative spectrum. They then use that ideology variable either as the focus of a study

on ideology, or as a control variable in other regressions. The leading attitudinal surveys— the GSS, the

CCES, and the ANES— include a variable giving a respondent’s self-identified ideology. Factor analysis

assigns this variable no special prominence. To treat this self-identification appropriately, we urge

scholars to instead measure ideology using the fitted value from a regression of self-identified ideology on

other survey responses. In contrast to factor analysis, the regression approach assigns proper priority to

self-identification; it lets us test whether voters identify their own ideology through identity-group

variables; it avoids the bias introduced in choosing the issue variables to include in the factor analysis;

and it identifies the issues that the average voter thinks best define “liberal” and “conservative.” http://rasmusen.org/papers/spectrum-ramseyer-rasmusen.pdf.

Expand full comment
Emil O. W. Kirkegaard's avatar

Interesting approach. The main issue with this idea, maybe, is that the regression models would not be consistent across time. Issues will change their importance in determining self-perceptions. Trying to apply the model over decades, then, probably results in non-invariance and strange results. Something you could look into by fitting your model in the different decades and cross-validating across.

Expand full comment
Michael Watts's avatar

> At minimum, you should discover the best ONE-ITEM IQ test, which would be a very fun result.

There can't be a best one-item IQ test. In the absence of any measurement error of any kind (which wouldn't occur even under many idealizing assumptions), the most you can do with a one-item test is classify people into two groups, above a threshold and below a threshold.

That might be enough for you, but where you want the threshold to be will vary depending on what you're trying to accomplish.

Expand full comment
Wire's avatar

Black people are now regretting that they invented testing.

Expand full comment
Truth's avatar

American University's researched this highly at the turn of the 20th century and nothing has changed.

Blacks are substantially less intelligent...... we all know this it's a law of science essentially now.

Expand full comment
mike's avatar

Could somebody do #9 for me?

Expand full comment
mike's avatar

"Dividing the profits equally means each person gets 1/3 of them...$1600.

Expand full comment
John Duckitt's avatar

I wonder what the r is/would be with longer IQ tests of more established validity?

Expand full comment
Emil O. W. Kirkegaard's avatar

You can just adjust for the reliability and do the same for the others, and the validity should be roughly the same.

Expand full comment
Truth's avatar

Nothing would change.

Expand full comment
Realist's avatar

"There is another aspect to measuring intelligence, which is that to measure intelligence well, you need to use multiple different kinds of items. This is because every item (or test of items) measures intelligence in addition to one or more other abilities."

Yes, some finer aspects of intelligence are missed when an intelligence test is shortened.

"Universities are these days obsessed with keeping everybody enrolled (🤑), this might make the correlation weaker."

True, as with most aspects of our society, avariciousness reigns supreme.

Thanks for the analysis.

Expand full comment
Truth's avatar

No matter how you slice and dice it blacks are substantially less intelligent than whites or Asians..... you can rearrange the deck chairs on the Titanic as many times as you want it's not going to change the outcome.

Expand full comment
User's avatar
Comment deleted
Feb 17, 2024
Comment deleted
Expand full comment
mike's avatar

An there races 2.

Expand full comment