An Ofsted survey into teacher wellbeing has been criticised as being unreliable and unrepresentative, with claims that anyone who was sent a link to the poll could answer the questions.

The inspectorate announced initial findings of its survey last week, showing that three-quarters of teachers said the job was having a negative impact on their mental health.

However, Ofsted has now faced questions about whether the survey was secure, and how the 1,000 schools it selected to take part were chosen.

Posting on Twitter, Ed Finch said: “Are you aware that the link to the survey allows absolutely anyone to take it? It asks you to select a school from a drop-down menu but it’s in all ways absolutely insecure. I don’t see that you can put any trust in the data it has collected. A shame as it’s a worthwhile study.”

There were also several posts on social media saying that people had taken part in the survey by pretending to be from another school.

Who's answered the ofsted survey as another school because their own one didn't get the invite?@Ofstednews so many people will do this... Making your survey pointless...

Tip: junk this survey. Put it out again to all. Do it online and analyse the data with no effort.

— ClaymanSLT (@ClaymanPE) December 2, 2018

The inspectorate has also been criticised for restricting the number of schools that could take part.

Ofsted had posted a reminder that the survey was due to finish over the weekend and urged schools to participate.

However, some school leaders then reacted with dismay when they discovered that their school was not listed to take part.

In response to criticism, Ofsted posted: “We’re sorry the original tweet suggested the survey was open to all.

“It isn’t, for the reason that we wouldn’t be able to process all the responses if it were. Our survey of 1,000 schools - contacted at random to minimise bias - will give us a representative sample.”

However, this statement also attracted criticism from some Twitter users, who questioned how a random survey could be representative.

Randomisation does not mean it would minimise bias. It leaves it completely up to chance. You could theoretically randomly select 1,000 'outstanding' or 'good' schools and not get a fair representation. And how did you make this random selection? Random number generator online?

— Hayden Reynolds (@hr_teacher) December 2, 2018

Ofsted also came under fire from Mary Bousted, joint general secretary of the NEU teaching union.

In a message about the inspectorate’s tweet promoting the survey, she wrote: “Incredible. Ofsted measuring the damage it has itself caused to teacher wellbeing through excessive workload generated by the bureaucracy its inspections generate around lesson planning, assessment and data. All compounded by the uneven quality of its inspectors.”

Ofsted has been approached for comment.

Findings from the survey so far show that three-quarters of respondents believe their job impacts negatively on their mental health, while 60 per cent claim it is affecting their physical health.

The main reasons cited for teaching having a negative impact were the behaviour of pupils and inconsistent management, as well as workload and marking.

Ofsted also highlighted a lack of funding as being one of the issues impacting on teachers’ wellbeing.