The traditional smear test is more reliable than a newer screening technique, researchers suggest.
Smear tests show cancerous cells
Some countries, including the US and Switzerland, have introduced liquid based cytology screening.
But others, including the UK, continue to use the traditional smear tests to spot cervical cancer.
The French research team recommend using the conventional system because liquid based cytology (monolayer testing) is more likely to give incorrect readings.
They also looked at the effectiveness of testing for Human papillomavirus, which causes most cases of cervical cancer.
They studied 828 women referred to hospital because abnormalities had been detected on previous smears and 1,757 women attending for routine smears.
Each woman was given a conventional smear test.
The evidence for change is not there at the moment
Dr Richard Sullivan, Cancer Research UK
In addition, half the women had a liquid cytology test and the rest the HPV test.
Conventional smear tests were found to be 91% reliable, compared to 87% reliability for the liquid cytology tests.
HPV testing was no more effective than the conventional tests, they found.
They said HPV tests needed to be researched further before being introduced.
Writing in the British Medical Journal, the Paris based researchers led by Dr Joel Coste, said: "Monolayer testing is less reliable and more expensive and should not replace conventional smear tests for cervical cancer screening.
"While human papillomavirus testing should be further evaluated.
"Our results emphasise the need to improve the "hard evidence" in studies of new technologies for cervical screening by using adequate methodological standards."
Dr Richard Sullivan, the head of clinical programmes for Cancer Research UK, told BBC News Online: "We know that the screening programme has been an excellent success.
"It saves around 1,300 lives a year."
He said any new screening method must be properly evaluated before the existing one was replaced.
"Liquid cytology is a possible replacement. The question is, is there enough evidence for that now, or are we going to say the evidence for change is not there at the moment and we need to do more studies.
"I think, at the moment, it's the latter."