MPs received more than 130,000 toxic messages in just six weeks as scale of online abuse revealed

• Special report by ITV Granada Reports' journalist Emma Sweeney

An MP says the level of abuse he receives for his job is making him consider not standing in the next general election.

Scott Benton, the MP for Blackpool South, says he has suffered abuse both on and offline, including death threats and sustained attacks.

It comes as a study reveals the scale of online abuse directed at politicians.

Analysis of almost three million tweets mentioning MPs across a six week period found over 130,000 were considered 'toxic'.

Male MPs were more likely to be called 'liars', while female MPs were often targeted by sexualised comments, or called 'thick' or 'ignorant'.

The analysis showed Mr Benton was mentioned in 13,509 tweets during this time and 342 of those tweets (2.5%) were considered as toxic.

The Conservative representative said: "I've been called 'moronic', an 'evil toad' and those are the more pleasant ones without expletives attached to them.

"There's a civilised society and there's a free society and we have to ask whether or not people have a right to say those things.

"I would ultimately say they do, so I don't necessarily have a problem with some of those tweets, unpleasant as they are, in my mind it's the more serious tweets when we talk about potential death threats online, when we talk about being called words such as a Nazi."

"To give an example," the Conservative MP added, "The local police force, Lancashire, are investigating two different incidents for me at the moment.

"One was a death threat and one was continued abuse via email where a gentlemen sent me over 60 abusive emails.

"That's not even including the online day-to-day Twitter and Facebook discourse and some of the abuse."

MPs from all three main political parties were subject to online abuse in the North West during the six week period.

Shadow Chancellor Angela Rayner was mentioned in 83,797 tweets - 3,390 (4%) of which were considered as toxic.

What classifies as a toxic message?

• The researchers used a Perspective API, a machine-learning tool built to identify harmful conversations online. It defines toxic comment as a rude, disrespectful or unreasonable comment that is likely to want to make someone want to leave the conversation.

The research, led by the BBC Shared Data Unit, also found that often the abuse spiked after MPs spoke in parliament, or in the news, about issues they were passionate about.

Rosalynd Southern, a Senior Lecturer at the University of Liverpool, said: "Quite commonly we'll see people trying to make a valid criticism or trying to hold an MP to account and they almost veer into incivility or abuse almost by-the-by.

"They almost do drive-by misogyny while they're trying to hold them to account for policy."

Lucy Powell, Labour MP for Manchester Central, received almost 170 toxic tweets during the six week period, which accounted for 3.4% of the overall tweets she received.

As the shadow Culture Secretary, she believes the government needs to act fast - starting with the much-delayed Online Safety Bill.

Ms Powell said: "It's the same kind of hate we see in society, but online it gets amplified in a way that it wouldn't do in a work place or a pub or a hime situation."

She added: "What's most important about this bill is it will give the regulator, Ofcom, powers to really scrutinise the business models, the algorithms, the engagement tools that so reward sensationalised disinformation, hate speech on these platforms.

"They are actively rewarded on these platforms because they foster engagement and they monetise hate and they monetise misery."

Elected MPs say they fear the toxic level of debate online is creating an environment where they are seen as fair game for abuse and violence.

There is growing pressure on the government to take the Bill - which has been hit by a series of delays - through parliament before Christmas otherwise it could be scrapped altogether.

The latest delay is because the government is said to be re-examining a part of the bill that would tackle content that's actually legal but still potentially harmful.

Emma Sweeney joins Gamal and Ann in the studio

Twitter has been contacted for a response but the firm was unable to reply.

Previously it has said it is committed to opposing abuse as outlined in its Hateful Conduct Policy.

On its website, it says: "We are committed to combating abuse motivated by hatred, prejudice or intolerance, particularly abuse that seeks to silence the voices of those who have been historically marginalized.

"For this reason, we prohibit behaviour that targets individuals or groups with abuse based on their perceived membership in a protected category."

Want a quick and expert briefing on the biggest news stories? Listen to our latest podcasts to find out What You Need To Know...