Sample surveys and opinion polls can be manipulated in several ways.
For example, you can ask the same question in two different tones.
1. "Do you want to punish 16-year olds with jail time similar to the case with adult offenders, if they commit serious offenses? Or should they be sent to special juvenile institutions instead?"
and
2. "Do you think that rapists deserve jail-time, even if they are slightly under the age of 18?"
I am quite sure that the percentage of people agreeing with #2 will be more than #1.
That just means your methodology sucks and that you have no clues about real scientific studies.
Well he's just explaining how a study can be flawed...
He's not saying that all studies are useless and stupid because numbers can be manipulated, he's just saying that each study is influenced by the way it was planned and then by the people behind it paying the scientists conducting the study.
Rather true IMHO
I understood that.
And for me it shows either bad methodology and/or the abstinence of objectivity.
Both is a the A and O for a real and good scientific study:
Not biased and be able to reproduced.
Oh ok. I agree then. It's the real point that should be checked before considering a study. Not who made it but has scientific method been respected.