a warning

Nov. 23rd, 2016 06:26 pm
archersangel: ("normal")
[personal profile] archersangel
from [personal profile] twistedchick 
Reposted from my Facebook entry:

There is a supposed survey with Trump's and Pence's logo at the top that is alleged to be an official survey of attitudes toward their plans. Don't take it. Don't fill it out. Don't put your name on it.

It is deceptive to the max. It gives you the titles of bills they plan to submit to Congress, and programs they plan to do -- but these are doublespeak. It does not give you enough information to correctly interpret what you are being asked. The 'questions' are misleading, biased, incomplete, inadequate. For example, the 'infrastructure' plan to rebuilt highways (and that's all you're told -- do you like it or not?) is actually about privatizing them, selling them to individuals and making them all toll roads. But they don't tell you that. It's the same with everything else on this so-called survey -- all of it is 1984-style doublespeak, lies. There is no place to say that you are against anything, just 'no comment' -- so from the start the entire thing is biased, incorrect, not valid for true statistical measurement.

A valid survey would give you options similar to these: Like very much, like a little, neutral, dislike a little, dislike very much. This survey's choices are: Highly important, somewhat important, not important, no opinion. Do you see the difference? There is no way to say you oppose anything, only that you have no opinion -- which is a lie, if you are motivated enough to take the survey in the first place.

Let me also point out that posting it on Facebook is not a statistically correct way of obtaining survey results. Those who reply will be self-selected; there is no way to evaluate the respondents in any way to make sure there is a representative sampling of people from all viewpoints and geographic locations. In a real survey, respondents are carefully selected to make sure they meet the criteria set up by the surveyors -- for example, geographic location, income level, race, political views, age. There is no way to do that online. Respondents are likely to be either rampant Trump/Pence supporters or people who are deeply opposed -- but since the survey is slanted toward Trump/Pence support, anything else won't be measured properly.

I have written surveys. I have done surveying for elections and other purposes. This is not how it is done. None of this is professionally done. None of it is properly organized. There is no way to use this survey to arrive at a statistically correct result of any kind.

I suspect that what this is for is to collect the names of US citizens. I do not know why. But this thing is so fake, so purposely skewed to favor the incoming administration, that it cannot have a neutral or good purpose.

Don't fill it out. Don't answer it. Do please feel free to repost what I have written here.

February 2017

S M T W T F S
   1234
567891011
12131415161718
192021 22232425
262728    

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Apr. 28th, 2017 06:28 am
Powered by Dreamwidth Studios