Twitter "I voted" mentions as a measure of enthusiasm

Jim Barksdale is quoted as saying:

If we have data, let’s look at data. If all we have are opinions, let's go with mine

Previously I've looked at clearly imperfect but interesting proxies for measuring grassroots support and enthusiasm in the Republican runoff.

I counted Twitter for mentions of having voted for either "Dewhurst" (anywhere in tweet) or "Ted Cruz" or "tedcruz."

Today, between 12:01am and 8pm during Tuesday early voting, the results were:

Voted for David Dewhurst: 4
Voted for Ted Cruz: 159

Yesterday in the morning I also counted (between 7am and 1pm on Monday morning) and the count was:
Dewhurst: 3
Cruz: 42

Most interestingly, there were more people who said they voted against Dewhurst without ever mentioning Cruz (7) than actual people mentioning they'd voted for Dewhurst (4). That would support the notion that Dewhurst's attacks are driving up Dewhurst's negatives more than they are convincing people not to vote for Dewhurst.

A note on methodology: searching for Dewhurst should pick up all his support, while if a tweet mentioned only "Cruz," it wouldn't have been counted. As such, if anything my count undercounted Cruz.

Comments

Honestly reminds of the Democrats arguing Bill White was ok, because he had more Twitter followers than Perry. Using Twitter as a measure of anything is fraught with complications.

Posted by Toby Belch @ 07/24/12 09:57 PM

How about bumper stickers in the court house parking lot vs number of people voting. I saw 6 Cruz bumper stickers, no Dewhurst bumper stickers, and there were 6 people voting. The excitement over Cruz may just give him a win, walking away. After all, it's hard to get too excited about a guy like Dewhurst, who puts the audience to sleep with his prepared debate closing statement.

Posted by John Gaver @ 07/24/12 10:21 PM

Bill White didn't have more twitter followers than Rick Perry. He briefly had more Facebook fans after spending more than 100K on Facebook ads for that specific aim. Perry did not spend on Facebook ads to gain fans.

Perry had far more twitter followers than Bill White. FAR more. Same goes for Perry versus Hutchison. On any and all organic online/social metrics, Perry destroyed both Hutchison and White. On one paid metric, White beat Perry because he paid and Perry didn't.

Posted by In the know @ 07/25/12 07:00 AM

I would note that I didn't draw strong conclusions from the data. Obviously many people tweet about voting for Cruz because he thanks most of the people who do.

However imperfect the data point might be, I prefer imperfect data to no data. That is especially true when other independent data tells the same story. As I said, I think it is more worrisome for Dewhurst that more people (who I don't recognize from #txsen) tweeted about voting against Dewhurst than people who tweeted about voting for Dew.

Posted by Rick Perry vs World @ 07/25/12 03:16 PM

Add Comments

No flames or impolite behavior. HTML will be stripped. URLs will be transformed into hyperlinks.