Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

General Discussion

Mountainguy

(2,145 posts)

11. It's called small sample sizes

Reply to Eliot Rosewater (Original post)

Mon May 26, 2025, 09:07 PM

May 26

Subgroups in polls have much smaller samples than the overall population of the poll. That means their margins of error are much higher than the full poll. A poll with a MOE of 3 might have an MOE of ~10 with a given subgroup. So a number in a subgroup going from 38 to 44 between two polls can easily be explained by normal variation.

Edit history

Please sign in to view edit histories.

Recommendations

0 members have recommended this reply (displayed in chronological order):

14 replies

= new reply since forum marked as read

Highlight:

Help me understand the Polling allegedly [View all] Eliot Rosewater May 26 OP

Trump has claimed that his tariffs are working. Media (including social) has repeated that. Stock market and economic Doodley May 26 #1

DJIA Is Down... ProfessorGAC May 26 #3

As I wrote, Trump has been claiming the tariffs have been working, and that has been repeated, and the stock market Doodley May 27 #13

Oh, please. Those were election polls. A damned lot has happened SINCE then... hlthe2b May 26 #2

I am purposely not providing links Eliot Rosewater May 26 #5

He isn't picking up young *Democrats* Fiendish Thingy May 26 #4

Did the supreme court with the traitors on it reverse Eliot Rosewater May 26 #8

What's that got to do with your OP? Nt Fiendish Thingy May 26 #9

Nothing, I just didn't know the answer and I didn't wanna start a whole thread about it Eliot Rosewater May 26 #12

Made me look. Believe it or not, here's one article. I'm sure we'll bash the findings and source. Silent Type May 26 #6

That's the one I saw, but I purposely didn't link it Eliot Rosewater May 26 #7

Article is garbage. They quote biased polls and cherry pick their data even then. Wiz Imp May 26 #10

It's called small sample sizes Mountainguy May 26 #11

OLD POLLING from the 2024 election results shows that... Jack Valentino May 27 #14