I could not find any study related to the exact question, but hacking a single website to spread a political message is very unlikely to be effective:
- time to detect web site changes (defacement) can be very small, as very efficient defacement detection algorithms can be implemented
- once the defacement is detected, the web site is closed, so it can be restored
On the other hand, using several methods (using legitimate web sites for politics news, fake news on various social networks etc.) can be effective.
This article explains how the political cyberhacker Andrés Sepúlveda managed to use digital techniques to manipulate the outcome of the elections in Mexico.
According to the cited article, these techniques are not very effective overall, but may have important influence on the political outcome:
Many of Sepúlveda’s efforts were unsuccessful, but he has enough wins
that he might be able to claim as much influence over the political
direction of modern Latin America as anyone in the 21st century.
One "success story" depicts how the hacking was performed:
With a budget of $600,000, the Peña Nieto job was by far his most
complex. He led a team of hackers that stole campaign strategies,
manipulated social media to create false waves of enthusiasm and
derision, and installed spyware in opposition offices, all to help
Peña Nieto, a right-of-center candidate, eke out a victory.
As a side note, this article from Symantec shows a list of security vulnerabilities that could be exploited by hackers:
[...] we discovered that there’s an opportunity for a hacker to modify
the code put on a voter’s chip card [...] We found a card reader that fits neatly into the palm of our hand and used it to reset our fake voter chip cards two different ways.
The lack of full disk encryption on the internal hard drive (as well
as the external cartridges) presents opportunities for hackers to
reprogram and alter ballots.
Also, the methods used by Andrés Sepúlveda are mentioned:
By propagating misinformation, a hacktivist or attacker could cause
voter distrust of election results.
In our simulated election, we broadcast our results “live” on YouTube.
We found that it’s plausible for hackers to hijack means of
communication and spread false results on YouTube, broadcast media,
social media and other channels. If voters were to follow the poll
leader, they might not choose to go through the trouble of voting in
an election if it looked like they were in for a landslide victory.
This is hardly an exhaustive answer. However, I think it is plausible that concerted hacking actions is effective in some contexts.