A new US study has claimed that research into LGBT young people could be based on faulty data either by “confused teens” or “jokesters” providing inconsistent responses about their sexual orientations.

According to MCT Information Services, the report focuses on the National Longitudinal Study of Adolescent Health, which is a study following a group of tens of thousands of teenagers into adulthood.

Ritch Savin-Williams, a professor at Cornell University, said that more than 70 percent of the teens who had ever experienced a same-sex “romantic attraction” in the study later told researchers they were straight, calling into questions the validity of the collected data.

He said the trend was unusual as teenagers usually come out of the closet, and not the other way round.

In his analysis published last month in the Archives of Sexual Behaviour, Mr Savin Williams and another researcher concluded that either teenagers were not understanding the question, or that “jokesters” were having fun with the survey.

The researchers found that you people who were “inconsistent” on their sexual attractions scored lower in intelligence and got lower grades.

Mr Savin-Williams added others may have answered falsely for fun, as earlier research on the Health survey had shown signs of dishonesty by teenagers falsely reporting they had an artificial hand, arm, leg, or foot.

He said the findings raise questions about a whole host of research on LGBT youth that is based on the same survey, with many studies showing reports such as worse physical and mental health for teens attracted to the same sex.

“It’s not that we saw something that no one else had seen,” Mr Savin-Williams added. “But they kept using the data…People should have said, ‘Hold on here. Who are these kids?'”

However, other experts have claimed that the idea that LGBT teenagers are higher at risk of suffering worse physical and mental health is entirely consistent with many other studies from different sources of data.

Ilan Meyer, a researcher on sexual orientation and gender identity law and public policy at the University of California, said that though the new study points to some “interesting issues,” it is not clear “what the results would look like if these inconsistent youth were not included.”

He added though the study raises the question, it fails to actually answer it.

Other experts say that “inconsistent” teens are not necessarily just “jokesters” or confused.

University of Arizona sociologist Stephen Russell, who has used the Health data in his own research on sexual attraction and health risks, said it made perfect sense that young people might have “romantic feelings” for the same sex, yet later call themselves straight.

He added: “Why is it unlikely they might be questioning who they are — and grow up to identify as straight?”

Previously, data from the National Longitudinal Study of Adolescent Health revealed that gay and bisexual youth are 40 per cent more likely to be punished at school or by the police and courts than their straight peers.