A randomised controlled trial to determine the effect on response of including a lottery incentive in health surveys [ISRCTN32203485]

...since 2) conclusion is not restricted to this popn & 3) what is really important is randomisation between treatment & control https://t.co/1A93FTG6XV

Thanks. Great example, 1).but the fact that those targetted are a simple random sample of some unrepresentative population is unimportant https://t.co/1A93FTG6XV