What specific question did he ask to determine dgu?
Presumably he asked "Did you use a gun in self-defence in the past year?" and I know he asked "Have you actually seen the attacker, rather than merely having heard a suspicious noise?". And if the respondents say "yes" to those two questions, then he asked "Was the gun actually fired?" (to which around 20% of respondents said yes) and "Do you think you would have died if you didn't have a gun?" (to which around 16% of respondents said yes).
Why telescope up but not down?
Because that's how studies show telescoping works: people, on average, tend to remember events which occurred 14 months ago as if they occurred less than a year ago.
How reliable is a data pool of 5,000 and was that from a gun county or from a non gun county?
I am not sure what you mean. By dialing a random number on a telephone, you get people from all over the US.
what was asked?
you would presume.
and you would be wrong.
kleck pg13
what is a
throat clearing question? - priming the bias of the responder and how they feel about cops and tough on crime (table 4).
and then followed up with
did you [feel] you used your gun for self protection? - well obvious if they feel it was for protection then they feel they were protecting themselves.
kleckpg14: story relevance:
1 involved a human, not animal and the respondent was not in security/ service.
2 an actual person, not just a noise.
3 defender could state a specific crime.
4 gun was actually used in some way.
so without knowing the flow of the conversation, i assume the above was guided by table 3 guideline:
and we note B home and C rob-theft-burg-trespass-assault.
if i feel threatened i will assume they're going to rob me or assault me to rob me.
well similarly that is a valid feeling, but is in no way a valid claim.
plenty of people shot last few years for knocking on the wrong door.
and suspiciously E
no threat nearly half the time.
i would give you 50% error bars.
like i said a few years ago here, a more realistic survey would be to rate convenience store robberies and successes of gun-no gun and magnitude-population.
telescoping, location and sample?
kleck pg13
we also oversampled within contacted households for males, who are more likely to own guns and to be victims of crimes in which victims might use guns defensivley. Data were later weighted for oversamplingso the calls weren't completely random...
so they've been "randomly" selected to be from, poorer neighbourhoods and male (cuts the population in half right there - did kleck include the females in his telescoping?).
so back table 4 64%respondants are <44,999$anual.
if the bottom 50% make up 50% of the population, we cut that 50 male/female by another 50 poor/mid-upper.
survey was also done by florida.
i would wager the interviewers were they biased.
kleck paper:
chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://scholarlycommons.law.northwestern.edu/cgi/viewcontent.cgi?article=6853&context=jclcthe title says it all:
extreme overestimateschrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/https://scholarlycommons.law.northwestern.edu/cgi/viewcontent.cgi?article=6936&context=jclc
