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 General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

sabrina 1

(62,325 posts)
Mon Nov 16, 2015, 02:41 PM Nov 2015

STUDY: More Tweets More Votes! Social Media as a Quantitative Indicator of Political Behavior!

For some people it is difficult to move away from the past even when progess and new technology demand it. Due to this phenomenon, people continue to place undue emphasis and trust in the old Political Polling Methodologies, despite pollsters themselves acknowledging that they have NOT kept up with new technology and its effects on their old Methodologies, which have been increasingly failing for a number of reasons.

It was inevitable therefore that studies would be forthcoming regarding the currently DISMISSED impact of online activity on various Social Media sites because not EVERYONE rejects NEW SCIENCE thankfully, when it comes along.

This is one study and I'm sure there will be more to determine what if any effect Social Media has on elections, among other things:

More Tweets More Votes! Social Media as a Quantitative Indicator of Political Behavior!

Joseph DiGrazia,1? Karissa McKelvey,2 Johan Bollen,2 Fabio Rojas 1
1Department of Sociology
2School of Informatics and Computing

Indiana University Bloomington, IN 47408, USA

?To whom correspondence should be addressed; E-mail: jdigrazi@indiana.edu.


An increasingly important question is whether social media activity can be used to assess off-line political behavior. Online social networking environments present a tremendous scientific opportunity: they generate large scale data about the communication patterns and preferences of hundreds of millions of individuals

...........

Here we show a statistically significant relationship between tweets and electoral outcomesthat persists after accounting for these potentially confounding variables. We compiled two large-scale datasets. First, we collected 2010 election outcomes and sociodemographic variables from all 435 U.S. House districts (18). Second, we retrieved a random sample of 537,231,508 tweets posted from August 1 and November 1, 2010. Then, we extracted 113,985 tweets that contained the name of the Republican or Democratic candidate for Congress.



First, the data do not include any information about the meaning or context ofa name mention (e.g., “I love Nancy Pelosi” vs. “Nancy Pelosi should be impeached”). The relative share of attention compared to the opponent is all that is needed.

This is evidence for the conventional wisdom that “all publicity is good publicity.”

Second, the models show that social media matters even when controlling for traditional television media, such as CNN, which many scholars have argued is important because it shapes political reality via agenda setting (27, 28), but does not seem to have a significant effect in our models.

Finally, this study adds to the mounting evidence that online social networks are not ephemeral, spam-infested
sources of information. Rather, social media may very well provide a valid indicator of the American electorate.



Acknowledgements We would like to thank Emily Winters and Matt Stephens for data collection as well as Clem Brooks, Elizabeth Pisares, and the Politics, Economy, and Culture Workshop at Indiana University for helpful discussions and contributions.

We gratefully acknowledge support from the National Science Foundation (grants SBE 0914939, CCF 1101743), the
Andrew W. Mellon Foundation, and the McDonnell Foundation

References and Notes

1. W. Bainbridge, Science 317, 4726 (2007).

2. D. Lazer, et al., Science 323, 7213 (2009).

3. A. Vespignani, Science 325, 4258 (2009).

4. M. Naaman, J. Boase, C.-H. Lai, Proceedings of the 2010 ACM conference on Computer
supported cooperative work, CSCW ’10 (ACM, New York, NY, USA, 2010), pp. 189–192.

5. A. Java, X. Song, T. Finin, B. Tseng, Proceedings of the 9th WebKDD and 1st SNA-KDD
2007 workshop on Web mining and social network analysis (ACM, 2007), pp. 56–65.

6. A. Mislove, S. Lehmann, Y.-Y. Ahn, J.-P. Onnela, J. N. Rosenquist, ICWSM ’11: 5th International
AAAI Conference on Weblogs and Social Media (Barcelona, Spain, 2011), pp.
554–557.

7. M. D. Conover, B. Gonc, A. Flammini, F. Menczer, EPJ Data Science 1, 1 (2012).

8. E. Hargittai, Journal of Computer-Mediated Communication 13, 276 (2007).

9. T. Correa, A. W. Hinsley, H. G. d. Ziga, Computers in Human Behavior 26 (2010).

10. S. Stephens-Davidowitz, Quarterly Journal of Economics (2011).

11. S. Asur, B. A. Huberman, Proceedings of the 2010 IEEE/WIC/ACM International Conference
on Web Intelligence and Intelligent Agent Technology - Volume 01, WI-IAT ’10 (IEEE
Computer Society, Washington, DC, USA, 2010), pp. 492–499.

12. J. Bollen, H. Mao, X. Zeng, Journal of Computational Science 2, 1 (2011).

13. S. Golder, M. Macy, Science 333, 187881 (2011).

14. P. Dodds, K. Harris, I. Kloumann, C. Bliss, C. Danforth, PloS one 6, e26752 (2010).

15. A. Tumasjan, T. O. Sprenger, P. G. Sandner, I. M. Welpe, Word Journal Of The International
Linguistic Association 280, 178 (2010).

16. B. OConnor, R. Balasubramanyan, B. R. Routledge, N. A. Smith, Proceedings of the International
AAAI Conference on Weblogs and Social Media (AAAI Press, 2010), vol. 5, p.
122129.

17. D. Gayo-avello, Arxiv preprint arXiv12046441 pp. 1–13 (2012).

18. U. S. FEC, Federal Elections 2010: Election Results for the U.S. Senate and the U.S. House
of Representatives (2010), pp. 39–150.

19. C. Klarner, PS: Political Science & Politics 41, 723728 (2008).

20. A. I. Abramowitz, The Western Political Quarterly 28 (1975).

21. H. Brady, S. Verba, K. Schlozman, American Political Science Review pp. 271–294 (1995).

22. K. Schlozman, N. Burns, S. Verba, Journal of Politics 56, 963 (1994).

23. S. Verba, K. Schlozman, H. Brady, N. Nie, British Journal of Political Science 23, 453
(1993).

24. J. Boucher, C. E. Osgood, Journal of Verbal Learning and Verbal Behavior 8 (1969).

25. D. Garcia, A. Garas, F. Schweitzer, EPJ Data Science 1, 1 (2012).

26. P. Rozin, L. Berman, E. Royzman, Cognition & Emotion 24 (2010).

27. M. E. McCombs, D. L. Shaw, Public Opinion Quarterly 36, 176 (1972).

28. M. S. Roberts, Journalism and Mass Communication Quarterly 69, 878 (1992)


Every time there is progress in human history, it is initially resisted by people who are more comfortable with the old ways. That is very natural because change can be scary.

Fortunately there are always, also, those who do not resist change but welcome it. And then there are those who want to know whether or not progress/change is good or bad for humanity.

Some things though simply cannot be denied, no matter how much resistance there may be.



106 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
STUDY: More Tweets More Votes! Social Media as a Quantitative Indicator of Political Behavior! (Original Post) sabrina 1 Nov 2015 OP
OMG!!! OMG!!! OMG!!! zappaman Nov 2015 #1
Stalkers are creepy! sabrina 1 Nov 2015 #2
Sorry a member replied to your hilarious OP, but that's how it works. zappaman Nov 2015 #11
Like I said, Stalkers are Creepy. sabrina 1 Nov 2015 #15
Like I said, you post, people reply. zappaman Nov 2015 #52
If you are going to post a lot of OPs treestar Nov 2015 #59
Clearly those are not the people i am referring to. But thanks for weighing in anyhow ... sabrina 1 Nov 2015 #60
It would have been nice if you could have come up with your own unique term. Sheepshank Nov 2015 #17
Thanks for your support. Stalkers are creepy, my words, and documenting them is sabrina 1 Nov 2015 #43
plagarized...but I'm flattered. n/t Sheepshank Nov 2015 #70
You may flatter yourself if you wish. Nothing stopping you from doing so. Just don't mistake sabrina 1 Nov 2015 #72
Watch out! zappaman Nov 2015 #74
I think I'll write some notes too..... Sheepshank Nov 2015 #77
I dunno. zappaman Nov 2015 #78
Sounds like stalking... sheshe2 Nov 2015 #89
Hmmmm zappaman Nov 2015 #90
;) sheshe2 Nov 2015 #92
They are writing a list and checking it twice. sheshe2 Nov 2015 #95
Why is it that anytime someone responds to you with something you don't agree with... George II Nov 2015 #31
Why is it you make stuff up about other DUers even when it is obvious that sabrina 1 Nov 2015 #33
Okay, what am I "doing"? George II Nov 2015 #49
Apparently, if you respond to the OP, you are "stalking" zappaman Nov 2015 #53
why are you responding to your "Stalkers". Sheepshank Nov 2015 #69
That's hilarious. polly7 Nov 2015 #42
Lol, that was my reaction also! What can you do but laugh? sabrina 1 Nov 2015 #44
I know right? polly7 Nov 2015 #46
Love that graphic! Lol. sabrina 1 Nov 2015 #48
You were alerted for that Oilwellian Nov 2015 #81
You were alerted on Oilwellian Nov 2015 #84
It's always alerted on. Documenting stalkers is necessary, glad to see juries understand sabrina 1 Nov 2015 #86
#4 got it right! zappaman Nov 2015 #91
LMAO! MaggieD Nov 2015 #13
Nailed it right out of the box. NCTraveler Nov 2015 #18
Is that your OMG this is science face? TM99 Nov 2015 #19
I guess they don't like Science. I thought that's what they wanted, but I guess sabrina 1 Nov 2015 #35
It was never about the science. TM99 Nov 2015 #37
Yes, that has been obvious. But now all the cries for Science have been answered and we have sabrina 1 Nov 2015 #45
More memes from Down The Rabbit Hole, Alice's Adventures in Wonderland: freshwest Nov 2015 #63
You got that right! n/t sabrina 1 Nov 2015 #94
If that's the case, then...Donald Trump will be our next president. JaneyVee Nov 2015 #3
Actually no. Bernie beats Trump in online formats as well as traditional polls. sabrina 1 Nov 2015 #4
LOL! Segami Nov 2015 #6
It's always been my philosophy to never ask a question you don't know the answer to sabrina 1 Nov 2015 #23
Meh,... Segami Nov 2015 #28
The anger over the online polls is unprecedented for some reason. I'm thinking sabrina 1 Nov 2015 #38
This reads and feels more like anti vax science. Untested, wishful thinking. Sheepshank Nov 2015 #71
Well we have the 2008 election to go by. And now another study apparently. Resisting progress sabrina 1 Nov 2015 #73
Bernie ain't Obama sheshe2 Nov 2015 #97
Lol! I think we all know that! But hey, thanks for kicking my thread. sabrina 1 Nov 2015 #98
Trust me. sheshe2 Nov 2015 #99
Lol, yes, some truly excellent comments. I too am glad they are being read. sabrina 1 Nov 2015 #100
Kim Kardashian!!!!! JoePhilly Nov 2015 #5
You said it whatchamacallit Nov 2015 #7
Thanks for helping prove my point. JoePhilly Nov 2015 #8
I believe it's "trees" whatchamacallit Nov 2015 #10
I went smaller than trees, because that's what you guys are doing. JoePhilly Nov 2015 #14
They are but another source of data whatchamacallit Nov 2015 #22
Sometimes, you throw out a source of data because all it adds to ... JoePhilly Nov 2015 #25
Didn't read the study, did you? sabrina 1 Nov 2015 #47
I did, but then again, I actually understand how to read such studies. JoePhilly Nov 2015 #54
Well, we women are not capable of reading and understanding complex scientific studies of sabrina 1 Nov 2015 #56
"Well, we women are not capable of reading and understanding complex scientific studies" zappaman Nov 2015 #57
did someone say that in this thread? Sheepshank Nov 2015 #75
Your inability to understand the extent to which you can extrapolate ... JoePhilly Nov 2015 #80
LOL! Segami Nov 2015 #20
You fell through the same door he fell through. JoePhilly Nov 2015 #26
Ya,..sure... Segami Nov 2015 #29
Hey, keep thinking tweets predict votes. JoePhilly Nov 2015 #30
Oh now,..come on... Segami Nov 2015 #32
Ah, but see ... JoePhilly Nov 2015 #34
Didn't read the study, did you? Lol! sabrina 1 Nov 2015 #39
I don't think WYMCI missed anything! Beautiful response to another comment with sabrina 1 Nov 2015 #12
Lol, you ran right through that door that was opened wide, with perfect timing! sabrina 1 Nov 2015 #9
The door he ran through opened over a cliff. JoePhilly Nov 2015 #16
What do you have to say about the actual topic of this OP? You're in the thread, I assume you've sabrina 1 Nov 2015 #21
The idea that tweets predict votes in any meaningful way is nonsense. JoePhilly Nov 2015 #24
Well that's now what the study says. But hey, it's there for you to read or not. sabrina 1 Nov 2015 #40
The WIU Mock Election thread contained a link to other predictive models ... 1StrongBlackMan Nov 2015 #51
I did a more complete review of the full article down below. JoePhilly Nov 2015 #55
Will win in 2016! It's all over! I give up! n/t freshwest Nov 2015 #58
President Ron Paul will not be defeated in his reelection bid tishaLA Nov 2015 #27
Ron Paul talking point already used and dismissed. Go to Obama, 2008! sabrina 1 Nov 2015 #41
talking point? WTF makes it a talking point? tishaLA Nov 2015 #50
Or whoever gets a thousand clicks by ten people each... freshwest Nov 2015 #64
In that case, Corrupt the Record will be "winning" the next debate for Hillary. winter is coming Nov 2015 #36
“I love Nancy Pelosi” vs. “Nancy Pelosi should be impeached” - bad news for Bernie? ieoeja Nov 2015 #61
I don't tweet about Hillary unless it's during a debate or to forward a link someone has posted sabrina 1 Nov 2015 #65
K&R. JDPriestly Nov 2015 #62
When the votes actually start getting counted, how long will it take the Bernie crew to start scream tritsofme Nov 2015 #66
How about we follow the example of the 'Benrie crew' and just deal with FACTS. Your sabrina 1 Nov 2015 #67
There's going to be plenty of screaming no mattter which way the votes go Fumesucker Nov 2015 #68
so this backs up Sanders winning in all 50 states JI7 Nov 2015 #76
indubitably n/t Sheepshank Nov 2015 #79
No, it doesn't. Why do you ask, did you see something IN the study that prompted that question sabrina 1 Nov 2015 #82
do you disagree with the op that you posted ? JI7 Nov 2015 #83
You didn't answer my questions. But then i never really expect actual discussion of issues from sabrina 1 Nov 2015 #87
off topic ? how is it of topic ? both are about how a candidate JI7 Nov 2015 #88
If you want to join the non contributing factors here, that's your choice. The rest of us actually sabrina 1 Nov 2015 #96
this OP supports your earlier OP. why are you upset about that ? JI7 Nov 2015 #101
Thanks again for kicking the thread. I tend to be different form most DUers who use the ignore sabrina 1 Nov 2015 #102
what's the FP ? JI7 Nov 2015 #103
Thanks, I couldn't do it without my small fan club! sabrina 1 Nov 2015 #104
i guess it's a great accomplishment JI7 Nov 2015 #105
Take a bow. No matter how small, it is an accomplishment. Funny isn't it, how sometimes things sabrina 1 Nov 2015 #106
You were alerted on Oilwellian Nov 2015 #85
#4 for the win! zappaman Nov 2015 #93

zappaman

(20,606 posts)
11. Sorry a member replied to your hilarious OP, but that's how it works.
Mon Nov 16, 2015, 03:04 PM
Nov 2015

You post.
People reply.
Would it make you happy if I posted in your other OP?

treestar

(82,383 posts)
59. If you are going to post a lot of OPs
Mon Nov 16, 2015, 05:16 PM
Nov 2015

A lot of the same people will respond to them. It's a message board.

 

Sheepshank

(12,504 posts)
17. It would have been nice if you could have come up with your own unique term.
Mon Nov 16, 2015, 03:09 PM
Nov 2015

I coined that term several months ago against a Bernie Supporter who had been following me around. I suppose it was magnificent and said it all. But I used it against a persistent stalker, not just against someone who would simply respond to occassional posts.

sabrina 1

(62,325 posts)
43. Thanks for your support. Stalkers are creepy, my words, and documenting them is
Mon Nov 16, 2015, 03:49 PM
Nov 2015

a good way to keep track of them. They never fail to oblige.

sabrina 1

(62,325 posts)
72. You may flatter yourself if you wish. Nothing stopping you from doing so. Just don't mistake
Mon Nov 16, 2015, 06:49 PM
Nov 2015

YOUR flattery of yourself to anything I have said. That would only embarrass you.

The documentation of stalkers which I am conducting, is my documentation.


Since I don't pay much attention to what you are doing here unless I run into you on a thread likes this, I am completely unaware of what you have said or didn't say unless it was directed to ME.

Sorry about that, it seems so important to you. But facts are facts!

zappaman

(20,606 posts)
78. I dunno.
Mon Nov 16, 2015, 07:09 PM
Nov 2015

I think keeping documentation on other DUers is kinda...creepy.
I wouldn't do that...would you?

sheshe2

(83,785 posts)
89. Sounds like stalking...
Tue Nov 17, 2015, 12:06 AM
Nov 2015

Lists~

Hmmm someone the other night said they were book marking threads so they would be handy for reference. Wasn't meant in a nice way.

George II

(67,782 posts)
31. Why is it that anytime someone responds to you with something you don't agree with...
Mon Nov 16, 2015, 03:35 PM
Nov 2015

...you call them a "stalker"?

 

Sheepshank

(12,504 posts)
69. why are you responding to your "Stalkers".
Mon Nov 16, 2015, 06:41 PM
Nov 2015

...it comes across as baiting and the complaining when they bite? OR, they could just be responding because you are asking questions or making statements that the responders feel is inaccurate and need refuting.

It really does appears you don't really believe they are stalkers until they say something you don't like then *BAM* to affix the label. It's not a particularly clever tactic to bandy around so much.

polly7

(20,582 posts)
46. I know right?
Mon Nov 16, 2015, 03:52 PM
Nov 2015

[URL=.html][IMG][/IMG][/URL]

and ......... great study and thread, Sabrina. You knew the kind of response you were going to get though, right? Never fails!!

[URL=.html][IMG][/IMG][/URL]

sabrina 1

(62,325 posts)
48. Love that graphic! Lol.
Mon Nov 16, 2015, 03:58 PM
Nov 2015

Well, the OP kind of takes away their argument that there is no 'science' regarding online formats and their relationship to election results.

So far I have received not one single reference to the findings in the study. It's interesting!

Oilwellian

(12,647 posts)
81. You were alerted for that
Mon Nov 16, 2015, 07:52 PM
Nov 2015

Until the alert, I couldn't see most of the responses in your thread due to an anti-flyswatter hiding binge I went on recently. I had to laugh at what you were responding to.

On Mon Nov 16, 2015, 06:02 PM an alert was sent on the following post:

Stalkers are creepy!
http://www.democraticunderground.com/?com=view_post&forum=1251&pid=817391

REASON FOR ALERT

This post is disruptive, hurtful, rude, insensitive, over-the-top, or otherwise inappropriate.

ALERTER'S COMMENTS

Stop calling other DUers stalkers.

You served on a randomly-selected Jury of DU members which reviewed this post. The review was completed at Mon Nov 16, 2015, 06:41 PM, and the Jury voted 2-5 to LEAVE IT.

Juror #1 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #2 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #3 voted to HIDE IT
Explanation: Seems the how thread is flamebait.
Juror #4 voted to LEAVE IT ALONE
Explanation: Hilarious. Cat was cute.
Juror #5 voted to LEAVE IT ALONE
Explanation: Another bogus alert.
Juror #6 voted to HIDE IT
Explanation: Oh for Pete's sake. If you post OPs/post responses in a thread, people are going to respond. Some are going to find your posts tiresome and they will express it. If you can't handle it don't post
Juror #7 voted to LEAVE IT ALONE
Explanation: I agree, stalkers are creepy.

Thank you very much for participating in our Jury system, and we hope you will be able to participate again in the future.

Oilwellian

(12,647 posts)
84. You were alerted on
Mon Nov 16, 2015, 07:57 PM
Nov 2015

Had to laugh at the irony.

On Mon Nov 16, 2015, 06:02 PM an alert was sent on the following post:

Stalkers are creepy!
http://www.democraticunderground.com/?com=view_post&forum=1251&pid=817391

REASON FOR ALERT

This post is disruptive, hurtful, rude, insensitive, over-the-top, or otherwise inappropriate.

ALERTER'S COMMENTS

Stop calling other DUers stalkers.

You served on a randomly-selected Jury of DU members which reviewed this post. The review was completed at Mon Nov 16, 2015, 06:41 PM, and the Jury voted 2-5 to LEAVE IT.

Juror #1 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #2 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #3 voted to HIDE IT
Explanation: Seems the how thread is flamebait.
Juror #4 voted to LEAVE IT ALONE
Explanation: Hilarious. Cat was cute.
Juror #5 voted to LEAVE IT ALONE
Explanation: Another bogus alert.
Juror #6 voted to HIDE IT
Explanation: Oh for Pete's sake. If you post OPs/post responses in a thread, people are going to respond. Some are going to find your posts tiresome and they will express it. If you can't handle it don't post
Juror #7 voted to LEAVE IT ALONE
Explanation: I agree, stalkers are creepy.

Thank you very much for participating in our Jury system, and we hope you will be able to participate again in the future.

sabrina 1

(62,325 posts)
86. It's always alerted on. Documenting stalkers is necessary, glad to see juries understand
Mon Nov 16, 2015, 08:40 PM
Nov 2015

how creepy stalking is. But it's good to have this kind of thing documented which I will continue to do making it much easier to show the pattern.

Thanks for the link!

sabrina 1

(62,325 posts)
35. I guess they don't like Science. I thought that's what they wanted, but I guess
Mon Nov 16, 2015, 03:40 PM
Nov 2015

I was wrong. I notice not a single refutation of the Science in the OP. Which can only mean, they have no response.

 

TM99

(8,352 posts)
37. It was never about the science.
Mon Nov 16, 2015, 03:44 PM
Nov 2015

If Clinton was winning on the online polls, they would be just fine.

It is just the standard talking points, memes, deflections, and frankly, bullshit that is spouted as loudly and with great fury.

It really doesn't mean a damned thing.

Thanks for the excellent article. I am not surprised.

sabrina 1

(62,325 posts)
45. Yes, that has been obvious. But now all the cries for Science have been answered and we have
Mon Nov 16, 2015, 03:52 PM
Nov 2015

'scientific' proof that the anger over these polls was never about anything other than Bernie is winning them all by huge margins.

Like you, I have suggested that if Hillary were winning, and I don't know she is not considering all candidates have equal opportunity to do so, we would hear how 'scientific' they are. Lol!

freshwest

(53,661 posts)
63. More memes from Down The Rabbit Hole, Alice's Adventures in Wonderland:
Mon Nov 16, 2015, 05:56 PM
Nov 2015


Oh, my paws and whiskers! I'll be late!

sabrina 1

(62,325 posts)
4. Actually no. Bernie beats Trump in online formats as well as traditional polls.
Mon Nov 16, 2015, 02:48 PM
Nov 2015

Thanks for reminding me, though that was not part of THIS study, it's interesting to see that it is consistent with the study's findings.

sabrina 1

(62,325 posts)
23. It's always been my philosophy to never ask a question you don't know the answer to
Mon Nov 16, 2015, 03:16 PM
Nov 2015

or go into a thread WITHOUT reading the OP and attempt to make an intelligent comment. I'm wondering why all of a sudden Hillary supporters have lost interest in 'science'! I was hoping to accomodate them assuming they were serious. But, well, so far we have 'leaves and trees' and for some totally inexplicable reason 'Kim Kardishian'.

 

Segami

(14,923 posts)
28. Meh,...
Mon Nov 16, 2015, 03:25 PM
Nov 2015

Whomever is handing out their talking-points orders today seems to be a little mid-day doozy. Definitely, their morning call envelopes have been mixed.





sabrina 1

(62,325 posts)
38. The anger over the online polls is unprecedented for some reason. I'm thinking
Mon Nov 16, 2015, 03:45 PM
Nov 2015

the campaign knows what happened last time and is attempting to 'mock' them away, to no avail apparently. You're right that the talking points don't seem to be working.

I would think by now they would have fired the 'experts' they've been using, and hired some more qualified personnel. But I'm not complaining, all this is helping Sanders so as far as I am concerned, they are doing just fine, for US! Loll!

 

Sheepshank

(12,504 posts)
71. This reads and feels more like anti vax science. Untested, wishful thinking.
Mon Nov 16, 2015, 06:49 PM
Nov 2015

a one off article you are trying to pass off as well tested and faultless. haa haaa haaaaaaaa.

lets see what happens in the real Primaries and we can test your new found scientific knowledge against reality

bookmarking

sabrina 1

(62,325 posts)
73. Well we have the 2008 election to go by. And now another study apparently. Resisting progress
Mon Nov 16, 2015, 06:51 PM
Nov 2015

as I said, is nothing new, so feel free to do so.

sheshe2

(83,785 posts)
99. Trust me.
Tue Nov 17, 2015, 01:43 AM
Nov 2015

I am thrilled your thread is going to the top.

Excellent comments here that need to be read!

JoePhilly

(27,787 posts)
14. I went smaller than trees, because that's what you guys are doing.
Mon Nov 16, 2015, 03:07 PM
Nov 2015

Tweets predict votes.

Its too funny.

whatchamacallit

(15,558 posts)
22. They are but another source of data
Mon Nov 16, 2015, 03:14 PM
Nov 2015

"Legit polls" are not without issues too. Demos, landlines, question crafting... Totally dismissing internet and social media trends is as foolish as totally relying on them.

JoePhilly

(27,787 posts)
25. Sometimes, you throw out a source of data because all it adds to ...
Mon Nov 16, 2015, 03:20 PM
Nov 2015

... your predictive ability is meaningless "noise" ... also known as unexplained statistical variation.

The idea that tweets predict votes is rather silly, especially when the vote in question is more than 5 months away.

sabrina 1

(62,325 posts)
47. Didn't read the study, did you?
Mon Nov 16, 2015, 03:53 PM
Nov 2015

Sometime you throw out comments in a thread and prove you have no basis for the comments even though you COULD have since the material is right there for you to read. But to each their own.

JoePhilly

(27,787 posts)
54. I did, but then again, I actually understand how to read such studies.
Mon Nov 16, 2015, 04:18 PM
Nov 2015

And because I enjoy reading this kind of study ... my assessment.

1) The first thing you miss is that the results only predicted elections where Republicans won. You assume that the results would transfer to Democrats or to primary voters. On what basis do you make that assumption? Big question unaddressed, why does their data only work for Republicans?

2) The data they collected was for the 3 months IMMEDIATELY prior to the election. The current data you refer to is more than 5 months out from the relevant election.

3) The authors do not claim nearly as much as what you claim. They suggest this is an interesting area for future study. Here is their strongest conclusion.

Finally, this study adds to the mounting evidence that online social networks are not ephemeral, spam-infested sources of information. Rather, social media may very well provide a valid indicator of the American electorate.


4) Your authors are not very good. Read this quote from the article.

Although the effect size is reduced to 154.7 in the full model, the effect remains highly significant.

There is no such thing as a "highly significant" effect in behavioral research. The effect either reaches statistical significance or it does not. Your authors make this mistake multiple times, which suggest that they are not terribly familiar with the methodology they are using. Computers make running these statistics much easier than it was when I was learning this stuff. But learning to do it by hand forces you to understand what it actually means.

5) I enjoyed this sentence.

Interestingly, the percentage of whites also has a highly significant positive effect on vote margin in both models.

Do you have any theories on why? Or how about the importance of POC to Democratic elections? They do not go into it, but perhaps, POC are not tweeting as much as whites are. But this might explain why their data correlated well with Republican tweeters, who we can assume, are primarily white

6) Here's an interesting observation from the article.

We find that districts where the model under-performs tend to be relatively
uncompetitive. If there is little doubt about who the winner will be, there may be little reason to
talk about the election.

I wonder if Hillary being up by 25% makes things uncompetitive?

Any other questions?

sabrina 1

(62,325 posts)
56. Well, we women are not capable of reading and understanding complex scientific studies of
Mon Nov 16, 2015, 05:07 PM
Nov 2015

data, I'm sure. But we TRY and yes, I did manage to read AND understand the findings even though I am both a woman and a Bernie supporter.

You don't like the fact that now finally they ARE beginning to acknowledge, which only makes logical sense, the power of Social Media.

It took a while before many people acknowledged the power of TV also, Radio before that, the Printing Press before THAT.

How 'scientific' are polls that admit now they are far beyond regarding new technology, making it impossible for them to accurately predict the results of elections. I'll take their word for it, especially considering their failures over the past number of years.

zappaman

(20,606 posts)
57. "Well, we women are not capable of reading and understanding complex scientific studies"
Mon Nov 16, 2015, 05:12 PM
Nov 2015

Why would you say such a silly thing?

 

Sheepshank

(12,504 posts)
75. did someone say that in this thread?
Mon Nov 16, 2015, 07:00 PM
Nov 2015

I haven't made my way through the whole thread yet, but i will be calling out anyone that said such a thing. Or are you making this up?

JoePhilly

(27,787 posts)
80. Your inability to understand the extent to which you can extrapolate ...
Mon Nov 16, 2015, 07:49 PM
Nov 2015

... the results of that study have nothing to do with your gender, or your support for Bernie ... although the latter might cause one to lose their objectivity.

And apparently, you are unaware that the behavior sciences (psychology and sociology in particular) tend to be dominated by women. You could learn from them.

As for the polls you whine about ... I could help you understand why some are more predictive than others ... but I would not want to suggest that you need a man to explain such things to you.

Instead, you should seek out some female behavioral scientists, of which there are many, and I suspect they'd be happy to provide you with some of the fundamentals of behavioral science.

JoePhilly

(27,787 posts)
30. Hey, keep thinking tweets predict votes.
Mon Nov 16, 2015, 03:29 PM
Nov 2015

You should write a tweet program to auto tweet about Bernie ... that way he can't lose!!!

sabrina 1

(62,325 posts)
12. I don't think WYMCI missed anything! Beautiful response to another comment with
Mon Nov 16, 2015, 03:06 PM
Nov 2015

zero content, unrelated to the topic, and the kind of comment that contributes nothing to the discussion on this forum.

JoePhilly

(27,787 posts)
16. The door he ran through opened over a cliff.
Mon Nov 16, 2015, 03:09 PM
Nov 2015

He did not get that I used leaves instead of trees specifically because they are even smaller.

Tweets predict votes.

Its hilarious.

sabrina 1

(62,325 posts)
21. What do you have to say about the actual topic of this OP? You're in the thread, I assume you've
Mon Nov 16, 2015, 03:13 PM
Nov 2015

read the entire study. So far, you're only contribution to this study was incredibly, 'Kim Kardashian' Could you explain for those of us who simply didn't get the connection between a completely moronic reality show personality, good friend of Hillary's or not, and a study on the impact of the New Media on elections? I would try to figure it out myself, but I just don't have that much time ....

JoePhilly

(27,787 posts)
24. The idea that tweets predict votes in any meaningful way is nonsense.
Mon Nov 16, 2015, 03:17 PM
Nov 2015

And I say that as one who has spent 3 decades working in behavioral science research.

And the idea that tweets more than 5 months before anyone votes predict anything is even more ridiculous.

Having said that, I am getting a huge kick out of your OPs suggesting that online activities, in which the participants self select, somehow predict anything.

Its a rather humorous form of wishful thinking. About as likely to be predictive as the calls for a primary challenge to Obama.

 

1StrongBlackMan

(31,849 posts)
51. The WIU Mock Election thread contained a link to other predictive models ...
Mon Nov 16, 2015, 04:10 PM
Nov 2015

showing equally strong positive correlations ... like whether the Redskins win their last game before the election and the ending of the previous years Oscar winning movie.

Gotta love science for the numerically illiterate!

freshwest

(53,661 posts)
64. Or whoever gets a thousand clicks by ten people each...
Mon Nov 16, 2015, 05:59 PM
Nov 2015

But don't you think that the Don is too 'muy feo' to win?

winter is coming

(11,785 posts)
36. In that case, Corrupt the Record will be "winning" the next debate for Hillary.
Mon Nov 16, 2015, 03:40 PM
Nov 2015

Which will be less obvious than footing the bill for dubious polling.

 

ieoeja

(9,748 posts)
61. “I love Nancy Pelosi” vs. “Nancy Pelosi should be impeached” - bad news for Bernie?
Mon Nov 16, 2015, 05:35 PM
Nov 2015

I know Bernie's supporters are bigger on social media. But do they post more positives about Bernie, or negatives about Hillary? According to this study if tweets about Hillary, good and bad, outnumber tweets about Bernie that would indicate a win for Hillary.



sabrina 1

(62,325 posts)
65. I don't tweet about Hillary unless it's during a debate or to forward a link someone has posted
Mon Nov 16, 2015, 06:08 PM
Nov 2015

that has to do with policies. I follow hundreds of Bernie supporters and most of them post positive things about Bernie, not so much about Hillary at all. Unless there is something relevant in the news.

But it is what it is, eg, are Hillary supporters posting more NEGATIVE things about Bernie, than positive things about Hillary?

tritsofme

(17,379 posts)
66. When the votes actually start getting counted, how long will it take the Bernie crew to start scream
Mon Nov 16, 2015, 06:17 PM
Nov 2015

that Hillary "stole" the election? My guess is that the transition will be seamless.

sabrina 1

(62,325 posts)
67. How about we follow the example of the 'Benrie crew' and just deal with FACTS. Your
Mon Nov 16, 2015, 06:25 PM
Nov 2015

CT speculative nonsense adds nothing to the discussion.

sabrina 1

(62,325 posts)
82. No, it doesn't. Why do you ask, did you see something IN the study that prompted that question
Mon Nov 16, 2015, 07:55 PM
Nov 2015

Or did you get the false memo that I might be intimidated by people kicking an old thread or something?

I can assure you that I am always flattered when one of my FP OPs are remembered. So whosever Idea that was,it is about as effective as all the other troll like efforts to try to slow down Sanders incredible momentum. Which is 'zero'.

But let me ask you something, don't you feel embarrassed, just a teeny bit, to become part of a small group of massively ineffective swarmers when you could come up with something a lot better, because let's face it, that was one of the lamest attempts to stalk any Duer I've seen yet?? I know I would be!

sabrina 1

(62,325 posts)
87. You didn't answer my questions. But then i never really expect actual discussion of issues from
Mon Nov 16, 2015, 09:37 PM
Nov 2015

people who jump into threads making off topic comments, following the lead of childish pranksters who are generally ignored by most posters here. But I am always optimistic that I might be wrong for once.

sabrina 1

(62,325 posts)
96. If you want to join the non contributing factors here, that's your choice. The rest of us actually
Tue Nov 17, 2015, 01:13 AM
Nov 2015

have families, children, elderly relatives we care about.

Elections for us ordinary people are about issues. They are not about anonymous stalkers on the internet. But we all have to make choices, mine is to choose the candidate that has a record of fighting for ordinary Americans.

Btw, thanks for kicking this thread, I am getting a lot of excellent feedback from it thanks to those who kept it going.

sabrina 1

(62,325 posts)
102. Thanks again for kicking the thread. I tend to be different form most DUers who use the ignore
Tue Nov 17, 2015, 02:01 AM
Nov 2015

feature to disappear non constructive commentary. I believe in using everything, including non constructive drive bys to kick important threads. Seems to work for me.

Mmm, I just checked the FP. Looks like you have helped me to get two OPs on the FP. Wow, thanks, AGAIN!

sabrina 1

(62,325 posts)
106. Take a bow. No matter how small, it is an accomplishment. Funny isn't it, how sometimes things
Tue Nov 17, 2015, 03:10 AM
Nov 2015

just backfire?

But even backfires have their silver linings, not necessarily for those who inadvertently cause them. This one is for Bernie. Thanks again!

Oilwellian

(12,647 posts)
85. You were alerted on
Mon Nov 16, 2015, 08:00 PM
Nov 2015

Had to laugh at the irony.

On Mon Nov 16, 2015, 06:02 PM an alert was sent on the following post:

Stalkers are creepy!
http://www.democraticunderground.com/?com=view_post&forum=1251&pid=817391

REASON FOR ALERT

This post is disruptive, hurtful, rude, insensitive, over-the-top, or otherwise inappropriate.

ALERTER'S COMMENTS

Stop calling other DUers stalkers.

You served on a randomly-selected Jury of DU members which reviewed this post. The review was completed at Mon Nov 16, 2015, 06:41 PM, and the Jury voted 2-5 to LEAVE IT.

Juror #1 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #2 voted to LEAVE IT ALONE
Explanation: No explanation given
Juror #3 voted to HIDE IT
Explanation: Seems the how thread is flamebait.
Juror #4 voted to LEAVE IT ALONE
Explanation: Hilarious. Cat was cute.
Juror #5 voted to LEAVE IT ALONE
Explanation: Another bogus alert.
Juror #6 voted to HIDE IT
Explanation: Oh for Pete's sake. If you post OPs/post responses in a thread, people are going to respond. Some are going to find your posts tiresome and they will express it. If you can't handle it don't post
Juror #7 voted to LEAVE IT ALONE
Explanation: I agree, stalkers are creepy.

Thank you very much for participating in our Jury system, and we hope you will be able to participate again in the future.

Latest Discussions»Retired Forums»2016 Postmortem»STUDY: More Tweets More ...