If you have participants in your data who don't match the demographics you ask for, please first confirm that the prescreening data provided by participants does not match the screeners you applied. You can do that by downloading the demographic data for your study.
If the participants' data from the export matches the screener but they have given you contradictory answers in your survey, please get in touch with them to ask which response is correct and if necessary, update their information within Prolific. It’s possible that there was a genuine misunderstanding arising from any differences in how the screening questions are worded. If this is the case, we recommend you ask these participants to return their submission so that they’re not penalised and you don’t have to send them the reward.
If it is clear that there was no misunderstanding and the answers they have provided are simply not truthful, this is valid grounds to reject the submission. Please note that rejections can only be made on the basis of inconsistent screening responses if the questions within your survey are presented exactly as they appear on Prolific. This is to avoid rejecting participants because of subjective misinterpretations, arising out of any differences in wording between your screening questions.
If your survey software supports it, you are allowed to redirect participants who provide inconsistent screening responses to a separate end of survey message, which asks them to return their submissions. Please see here for further guidance on this.