Taurus Firearm Forum banner

1 - 18 of 18 Posts

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #1 · (Edited)
Ironic, look at the town the sponsoring agency is located in. News Releases

They do not report their methodology and its an online survey. I would be very interested to know how they drew their sample. Not knowing this its hard to know whether or not this is a valid representative sample and thus have confidence in the findings. Nevertheless, here it is...a one page summary...an easy read.


More coverage of this press release...a little more info

http://www.guns.com/2013/06/25/nssf-releases-results-of-first-time-gun-buyer-study/
 

·
Registered
Joined
·
2,554 Posts
I realize this was an online survey but for surveys in general, where do they get their stats?
You see a new data almost daily stating XX% of Americans are for this, or against that.. Who are they polling?
I have never been contacted concerning any polls and neither has anyone I know? Have any of you ever participated in the polls we hear so much about.
 
  • Like
Reactions: whichfinger

·
Registered
Joined
·
300 Posts
When I took Stats in college... I learned one thing... you can manipulate the data to make it say what ever you want it to. So you have to really read and know where the data comes from.

Scott
 

·
Registered
Joined
·
7,741 Posts
I realize this was an online survey but for surveys in general, where do they get their stats?
You see a new data almost daily stating XX% of Americans are for this, or against that.. Who are they polling?
I have never been contacted concerning any polls and neither has anyone I know? Have any of you ever participated in the polls we hear so much about.
I don't know if I've ever been the recipient of a phone poll. We have caller ID and I decide who I want to talk to. If a name or number pops up that I don't recognize, it doesn't get answered. If caller Id is blocked, it doesn't get answered. If it's a toll-free number it doesn't get answered. The older I get, the more curmudgeonly I become. AFAIC, the telephone is there for my convenience. Harumph.
 

·
Registered
Joined
·
1,131 Posts
This was an online based survey. The NSSF is a great organization that just happens to be located in the ill fated Newtown. They are public enemy #2 (NRA is #1) of our gun grabbing Governor. I think most of their research base was hosted by a 3rd party but if I remember correctly, the survey appeared on a large local progun website www.CCDL.us. NSSF is a national organization that just happens to be located in the belly of the beast and was there long before the incident.

FWIW, although it may be local results they seem accurate with why those I talk to and those taking the NRA class here in CT are getting guns.
 
  • Like
Reactions: aphdmansoc

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #6 · (Edited)
I realize this was an online survey but for surveys in general, where do they get their stats?
You see a new data almost daily stating XX% of Americans are for this, or against that.. Who are they polling?
I have never been contacted concerning any polls and neither has anyone I know? Have any of you ever participated in the polls we hear so much about.
Data/stats are derived from the sample. The sample is a subset of a larger population/universe that one wants to generalize back to from the sample. How the sample is drawn is key in determining whether the "stats" are valid representations of the population. Their is a margin of error and a confidence level that is associated with the size of the sample that is used. For example, one might come up with (made up for illustration purposes) a sample stat that 18% of TA members are 60 years old or older. Assuming the sample was drawn correctly (some form of random sample) with a large enough sample, let's say 400, then the margin of error would be + or - 5% from the sample statistic of 18%. That means the "true" percentage of elderly would be (if we could take a full census of TA members) between 13 to 23% (18% being the midpoint of our estimate). If we were to do this study over and over again we would wind up these results, i.e, the sample statistic would be between 13 and 23% 95% of the time...this is called the confidence interval...with 18% being the best single point estimate of the true population percentage of older TA members.

If you want to know how you get a random sample...google it or pay me a consultation fee, or sign up for a college course on research methods! There are a lot of good sites on the web that will go into greater detail if you are really interested. How you phrase a question is another matter...but a basic course will cover that too. To do a valid study is both art and science. I used to do this for a living...teaching this stuff and doing polling. :)

The fact that you have never been contacted doesn't mean much. The odds of anyone individual being selected is very very small. How large a sample you need is dependent on several things...the most important are the size of the population you want to generalize to and the expected frequency of the occurrence you are looking for. A small expected occurrence will require a larger sample than one in which the characteristic is split near 50/50, for example, are you a man or a women. The other important variable is how large a margin of error are you willing to have.

As a matter of practice and economics sample sizes in the range of 400 will give you a margin of error of 5%. A sample of 1000-1,500 will give you a margin of error of 3%. Considering each interview might cost (illustration only) the sponsor of the survey $20 you can see that costs could range from $8,000 to as high as $30,000. So how accurate do you need to be? You pay for that accuracy. Thus, the idea of the value of information. If you are running a political campaign or evaluating the effectiveness of an advertisement and people's careers are on the line then the value of the information is very important and must be balanced out with other considerations...like how much money do you have...is it better to do a couple of smaller surveys or one large survey? A lot of practical questions go into determining the value of information. This idea is not always appreciated by the people who are paying for the study.

I had a funny experience of being called up for an interview by a competitor...who I was on friendly terms. Gave me a good insight on how good a job he did with his questionnaire and how well trained his interviewers were.
 

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #7 · (Edited)
I don't know if I've ever been the recipient of a phone poll. We have caller ID and I decide who I want to talk to. If a name or number pops up that I don't recognize, it doesn't get answered. If caller Id is blocked, it doesn't get answered. If it's a toll-free number it doesn't get answered. The older I get, the more curmudgeonly I become. AFAIC, the telephone is there for my convenience. Harumph.
I wonder why you don't get asked questions by polling organizations. :D
 
  • Like
Reactions: whichfinger

·
Registered
Joined
·
5,151 Posts
I dont trust poles or surveies but the figures are believable
and could be close

jhp
 

·
Registered
Joined
·
2,554 Posts
Thanks for the quick lesson.

Which was kinda my point, they said 90% of all Americans were in favor of more gun control but who did they ask?? The people of Newtown, when the wounds were fresh?

Like someone else said. I don't believe in any of them it's too easy to skew the #'s to get the results you want.
 

·
Registered
Joined
·
3,046 Posts
Wow aphdmansoc that's a lot to learn and well laid out in only a few paragraphs. I guess though I'll leave all of that up to others to do and worry about... that's one topic/education I don't see at the top of my bucket list.. :) But well said, seems you have lots of experience in it.
 
  • Like
Reactions: aphdmansoc

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #11 · (Edited)
This was an online based survey. The NSSF is a great organization that just happens to be located in the ill fated Newtown. They are public enemy #2 (NRA is #1) of our gun grabbing Governor. I think most of their research base was hosted by a 3rd party but if I remember correctly, the survey appeared on a large local progun website www.CCDL.us. NSSF is a national organization that just happens to be located in the belly of the beast and was there long before the incident.

FWIW, although it may be local results they seem accurate with why those I talk to and those taking the NRA class here in CT are getting guns.
Since NSSF is a industry association I would expect that they would have paid for a national study. The reason I don't think this would be a small town study is ...not to be too glib...but who cares what first time gun buyers in a small l community think or do? I'm sure NSSF contracted this out...it generally would be cheaper to do so if polling is not there primary business.

Universities with smart leadership generally see the advantages of having polling capabilities. They already are paying salaries for the "brain power" all they need is the equipment - computers, phones, software - and a few technical people and they are ready to go.
 

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #13 ·

·
Registered
Joined
·
1,131 Posts
The reason I said I believe it was on a large local website is I am pretty sure I remember seeing it there as I am a member of that group.

I also worked at the Quinnipiac Polling Institute for 3 days many years ago (1996) and my roomate worked there all semester (he was a little weird and thought badgering people was fun). Not all polls are created equally. I didn't even pick up my paycheck for those 3 days work, I felt sleezy just going in the building. I was a student at QU (then QC) and it was a work-study. The questions were biased and pollsters were told to hang up on people who disagreed to much. Nowadays, it seems the auto-dialer sure likes to call "friendly numbers" and since my wife took one poll about 6 years ago, we get repeat calls relatively often, maybe one a month. We are in the same town as QU, go figure. Even "national" polls conducted by "national instsitutes" may not be as "national" as some would believe.
 

·
Registered
Joined
·
27,647 Posts
I don't know if I've ever been the recipient of a phone poll. We have caller ID and I decide who I want to talk to. If a name or number pops up that I don't recognize, it doesn't get answered. If caller Id is blocked, it doesn't get answered. If it's a toll-free number it doesn't get answered. The older I get, the more curmudgeonly I become. AFAIC, the telephone is there for my convenience. Harumph.
Ha! What a grouch!
You don't know the entertainment and life enhancement you are missing.
In a typical day I can get:
a roofing company
  • 3 duct cleaning companies
  • a carpet cleaning service
  • The Indian Children's Relief Fund
  • the FOP
  • the Sheriff's Association
  • "my" Congressman
  • Buy Gold Now
  • 2 subscription offers
  • Chem-Gro
  • "my" alumni association
  • Work-At-Home
  • Red Cross
  • Purple Heart
  • 4 miscellaneous charities
  • AND a telephone survey that won't take my answers because I don't fit their predetermined demographic
And you, poor fellow, are missing out on the fun.
 
  • Like
Reactions: whichfinger

·
Registered
Joined
·
5,072 Posts
Discussion Starter · #16 ·
The reason I said I believe it was on a large local website is I am pretty sure I remember seeing it there as I am a member of that group.

I also worked at the Quinnipiac Polling Institute for 3 days many years ago (1996) and my roomate worked there all semester (he was a little weird and thought badgering people was fun). Not all polls are created equally. I didn't even pick up my paycheck for those 3 days work, I felt sleezy just going in the building. I was a student at QU (then QC) and it was a work-study. The questions were biased and pollsters were told to hang up on people who disagreed to much. Nowadays, it seems the auto-dialer sure likes to call "friendly numbers" and since my wife took one poll about 6 years ago, we get repeat calls relatively often, maybe one a month. We are in the same town as QU, go figure. Even "national" polls conducted by "national instsitutes" may not be as "national" as some would believe.
I certainty do agree that polling quality varies. I had the luxury of being my own boss. I had the intellectual commitment to doing quality and honest research. I had complete control over the development of every major aspect of the surveys - sampling, questionnaire construction and data analysis. Most polling organizations use hired help for many of these important aspects of the study. Sometimes the "main guys" are the outreach people drumming up the business and they may leave too much of the "grunt" work to others - who have varying commitments to the project. And projects vary in their importance to a company and that influences the quality of the personnel they assign to a particular project.

Biased polling does the client no good. They are paying for useful information to plan and make decisions. In the case of obvious bias (which most people would not have access to the decisions made) that leads to propaganda press releases - more likely in the case of some political campaign for public relations the company doing that would, in the long run, suffer by losing credibility.

I remember doing a "private poll" for a politician. The results contradicted a Mason-Dixon poll that was getting a lot of publicity. We (his inner circle of which I was a part) went to one of the local watering holes for dinner. I was hoping we would run into some of the local crowd...knowing my results and the PR that the Mason-Dixon generated, I wanted to bet someone a $1,000 on the results of the election. Alas, none of the gambling crowd was at the place when we were there. My man won by a good margin. The only way to explain why the large discrepancy between the two polls was they either used a less experienced team and didn't put the resources into doing a good poll...they cut corners to increase profits or they were inept...maybe a little bit of both.
 
  • Like
Reactions: JoshNix9

·
Registered
Joined
·
7,741 Posts
Ha! What a grouch!
You don't know the entertainment and life enhancement you are missing.
In a typical day I can get:
a roofing company
  • 3 duct cleaning companies
  • a carpet cleaning service
  • The Indian Children's Relief Fund
  • the FOP
  • the Sheriff's Association
  • "my" Congressman
  • Buy Gold Now
  • 2 subscription offers
  • Chem-Gro
  • "my" alumni association
  • Work-At-Home
  • Red Cross
  • Purple Heart
  • 4 miscellaneous charities
  • AND a telephone survey that won't take my answers because I don't fit their predetermined demographic
And you, poor fellow, are missing out on the fun.
And loving it.
 

·
Registered
Joined
·
1,131 Posts
I had the intellectual commitment to doing quality and honest research.
This is the heart of the matter. It seems to me that more often than not, those ordering the "poll" are looking for a desired result and make it known to the polling institute. A major poll done in the interest of pure research and scientific inquiry is like a unicorn in my opinion. Please do not take offense, I believe someone who is obviously well versed is capable of doing an accurate unbiased demographical study, I just think that is more often the exception not the rule. I may be soured by my experience on both ends (albeit short and long ago on the polling end). As the recipient of poll calls rather frequently, I can assure you the majority of the ones I get are definitely worded to get a specific result and opposing viewpoints are often hung up on. I would be more than happy to participate in some actual unbiased and well constructed polls, as I believe they can have some real value. YMMV
 
1 - 18 of 18 Posts
Top