| Superman |
08-26-2023 05:51 AM |
Quote:
Originally Posted by fintstone
(Post 12074389)
The fact that they did not use violent crime rate tells you everything you need to know.
|
Not sure what it tells you, but it tells me that this study deliberately did not seek to estimate actual crime rates or actual probabilities of violence per capita. Other studies already do that. Instead, this study sought to measure attitudes and perceptions.
Quote:
Originally Posted by rwest
(Post 12074478)
My question about all these surveys is: aren’t you just getting opinions from the sort of people who are willing to answer their phone and spend 10-20 minutes talking to someone they don’t know?
That has to be a very specific demographic and therefore not representative of the public at whole. Pretty much every phone survey is just the opinions of one type of person- probably a very lonely person!
|
Yes, phone surveys do not capture data from people like me, who decline to participate. Then, as you discerned, the question becomes whether the non-respondents have the same perception of cities' safety as the respondents, or whether those perceptions are different between the two groups. In other words, would the results be different if the survey only included non-respondents? I have a guess about that and it might not be what most of you think. What would be interesting now is to compare these results with studies showing actual crime rates. And there could be even more interesting findings if we knew some demographic information about the respondents so that we could correlate their perceptions with, for example, their information sources. And stuff.
http://forums.pelicanparts.com/uploa...1693057758.jpg
And there are other problems with surveys like this. A note in the published results mentioned this. Did all respondents get the same exact question? What about voice inflection? Were other questions also included? Did the respondent ask questions and did the respondent and pollster have casual conversation? What were the attitudes and perceptions of the pollsters? Humans are VERY perceptive, and if the attitudes of the pollsters were different, then the data gathered could be different too....and this can be analyzed in the study. Studies can often identify their own bias.
Here is a particularly amusing meme about statistics, simply because it may be accurate. Probably is:
http://forums.pelicanparts.com/uploa...1693057875.jpg
|
