To Trust a Survey or Not
Data provided from a survey can be so convincing. Seeing the percentages posted can make you feel as if it is trustworthy. They are backing up their point with numbers. Does that mean it’s true? There are factors that you should consider before blindly believing what is presented to you. Look at the survey and the full report before deciding.
Sample Size
How many people did they have to take the survey? Look for the sample size. Anything can look convincing if the sample size is small. If they have fifty people out of a hundred, agree with them. That is fifty percent. Now imagine it was fifty out of a thousand. That is only five percent. Any subgroup can seem significant if the sample size is small. It can seem like a huge effect if it is taken out of context. Before taking the results of a survey seriously, make sure that as many people took it as possible. The larger the sample size, the more accurate the results will be. And the less chance of having a small group overwhelm the results.
Demographic
The age, race, beliefs of every person cannot be the same. Unless a survey is specific to a particular group. If it is looking to see the health choice of millennials. Then all those surveyed need to be in that age range. Looking to test the beliefs of the U.S., those users need to be as different as possible. Not only would the sample size need to be large, but they would also need people from every region. If it is not, you cannot trust the results.
People Lie
It is human nature. People lie. What is stopping them from lying on a survey? Any controversial topic will have more people lying. Either to avoid embarrassment or ridicule. In the report, it will tell you how the study was presented. More lying will happen when it is not done anonymously. Having someone asking and recording the answers will get less of the truth. Questionaries’ that are completed without knowing whose answers belong to who will get more honest answers. Use your judgment on if it was presented in a way that would be best for the circumstance. If you would be shy giving the answers, that is a good indication that the statistics may not be an accurate reflection.
Bad Questions
The quality of the questions asked can affect the results. How they are worded can make people react differently. There is a whole science to creating a good survey. Read the questions and evaluate them yourself. How do your results compare to the percentages that they are reporting? Look for biased or leading questions. Any of these should be a red flag that the results may not be the best. Only trust surveys that have adequate questions.
Do your research. Do not blindly believe what you are told. Put the tests to the test before you trust them.