[Previous] Applying Philosophy to Politics | Home | [Next] Bad Scholarship by Ari Armstrong

Alex Epstein Scholarship Problem

Alex Epstein wrote:
In the US, 30 years ago the average household had 3 electronic devices—today it has 25, overwhelmingly thanks to fossil fuels.
For a source, he linked this youtube video from "Alphanr" (Alpha Natural Resources). It's titled "Did You Know" and has no description. It's 89 seconds long and packed full of factual claims including the one Epstein asserted, but lacks sources. It ends with a link to their website. Searching for "Did You Know" on the site to try to find extra details results in an error. Their site says, "Alpha is a leading global coal company and the world’s third largest metallurgical coal supplier". I doubt they have expertise at surveying household electronic device usage.

Sourcing your assertion to someone else's unsourced assertion isn't scholarship. It's how lies spread.

I've investigated this topic. Several studies have been done about electronic device usage in U.S. households. However, none of them would be acceptable sources. All the ones I found have a large methodological error. For example:
Of the 37 CE [Consumer Electronic] devices surveyed, the average U.S. household owns 24, the same number as last year, and spent $961 on consumer electronics over the past 12 months, down more than $200 from last year. The average adult individually reports spending $552 on CE in the past 12 months, down $100.
They are surveying how many devices U.S. households have from a list of 37 devices. That list is incomplete. The number of devices from the list that a household has is different than the number of consumer electronic devices the household has.

There is an additional problem here. When the surveys use a specific list of devices, they change it over time. Today we would expect smartphones to be on the list. Thirty years ago, they would not have been. Changing the list of which devices count means the surveys from different years are not comparable because they measure different lists.

When you pretend what's being measured is "(consumer) electronic devices", then it would make sense to compare studies from different years, because they appear to measure the same thing. But really one survey is about list A and another survey from another year is about list B, and calling the two lists by the same name doesn't make them the same thing. Epstein's comparison between the present and thirty years ago, using two different surveys of two different things, is a mistake.

I informed Epstein of these scholarship errors and he did not fix them. He should not repeat unsourced claims from Youtube that are presumably based on misusing the readily available invalid research. The truth matters.

Elliot Temple on June 12, 2014


What do you think?

(This is a free speech zone!)