Nice LE Use of Force training survey

quote: Dr. Bill Lewinski often notes in his public presentations that the average high school football player gets more training in his sport in his brief career than the average peace officer receives in use-of-force instruction across his or her entire working life. Now a first-of-its-kind survey by Calibre Press has confirmed that dismal truth.


Recently, Calibre editors invited readers of its popular "Street Survival" newsletter to complete an anonymous Survey Monkey poll regarding their departmental training policies. Nearly 900 officers from small agencies to large participated, with these results: (added: https://www.surveymonkey.com/results/SM-QFX9QXRR/)


Range time. Nearly two-thirds of officers said they are required by policy to shoot on the range with their sidearm only once (23.66%) or twice (37.66%) a year. Only about 8% have to shoot as often as monthly.


Qualification. Monthly official qualification with their sidearm is required for only 1.37%, while roughly 84% need to qualify only annually (46.81%) or semi-annually (37.24%).


Scenarios. The monthly requirement shrinks even more (to 0.91%) when it comes to "dynamic 'shoot/don't shoot' scenario-type training." One-quarter never have to experience such training, and over half (56.26%) do so only once a year or less often.


DT training. Close to 15% of officers said they are never required by policy to do "defensive/control tactics-type training." For two-thirds (63.82%), such training is mandated only once a year (42.32%) or less (21.5%). Fewer than 2% must train hands-on monthly.


Less-lethal. Monthly requirement virtually fades off the chart (at 0.57%) when it comes to training with "less-than-lethal weapons" such as TASERS, batons, and OC spray. Annual training predominates at over 55%. One in five officers trains less frequently than that, and over 10% never have to engage in this type of training. unquote Force Science institute news #312.

i do not understand how or why the initiators of the survey are so bloody proud of quote nearly 900 unquote (what on earth does nearly represent?? 8001?) respondents to their query when, quote: In 2008, state and local law enforcement agencies employed more than 1.1 million persons on a full-time basis, including about 765,000 sworn personnel (defined as those with general arrest powers). Agencies also employed approximately 100,000 part-time employees, including 44,000 sworn officers. unquote. https://en.wikipedia.org/wiki/Law_enforcement_in_the_United_States

this equates to .08% eligible 1.1M and it doesn't advise if the respondent was still active.

more misinformation being disseminated.

ipse

I think that the US Army also only requires range time once or twice a year too. (see Soldier's Manual Of Common Tasks, 2009, Warrior Skills Level 1).

Even the skill of zeroing a M16 is only needed to be performed 2x/yr....and its a basic thing.

I see nothing inherently wrong with requiring only 1 or 2x a year range time ... the Army finds it OK....and from my experience, its fine.

So how many hours of training should a LEO have in the various aspects of the job.

Firearms Iinstuctor said:

So how many hours of training should a LEO have in the various aspects of the job.

Click to expand...

Zero. We train monkeys, not people. And you want someone to list down every job duty ? Give someone a year to fully respond.

Firearms Iinstuctor said:

So how many hours of training should a LEO have in the various aspects of the job.

Click to expand...

query, which training are you wanting to know about less lethal or deadly force training or talking the suspect down before we blow them away type of training?

my thought would be less lethal and talk training instead of immediately reaching for the bloody firearm first. please understand Firearms, just like citizens who get into trouble because they failed to recognize and understand their SA, the same consideration applies to the nice LEs who perhaps need to tread lightly before coming into the situation like a bear in a china shop ~ i shall refrain from providing examples of nice LEs failing to assess their SA and getting in over their head and getting stuck into a 'i feared for my life' situation and blew the suspect away and then told it was a righteous shoot by the nice judicial system, and oh BTW glad you are all right mentality.

ipse

davidmcbeth said:

I think that the US Army also only requires range time once or twice a year too. (see Soldier's Manual Of Common Tasks, 2009, Warrior Skills Level 1).

Even the skill of zeroing a M16 is only needed to be performed 2x/yr....and its a basic thing.

I see nothing inherently wrong with requiring only 1 or 2x a year range time ... the Army finds it OK....and from my experience, its fine.

Click to expand...

You are correct in that the required qualification with the individual weapon is normally not more than twice per year. However, that is for the M16/M4 weapons family, or other rifles. Firing, and zeroing, a rifle is a far more durable skill than is firing a handgun. One of the men who taught me pistol marksmanship back in the day recommended a minimum of every other month shooting at least fifty rounds. He also strongly recommended adding the practice of firing with the non-dominant hand.

To me, a good compromise would be to require law enforcement officers to qualify a minimum of once per quarter with the aforementioned fifty rounds. There should be a requirement to fire at least ten rounds with the non-dominant hand.

solus said:

i do not understand how or why the initiators of the survey are so bloody proud of quote nearly 900 unquote (what on earth does nearly represent?? 8001?) respondents to their query when, quote: In 2008, state and local law enforcement agencies employed more than 1.1 million persons on a full-time basis, including about 765,000 sworn personnel (defined as those with general arrest powers). Agencies also employed approximately 100,000 part-time employees, including 44,000 sworn officers. unquote. https://en.wikipedia.org/wiki/Law_enforcement_in_the_United_States

this equates to .08% eligible 1.1M and it doesn't advise if the respondent was still active.

Click to expand...

While that seems to make sense to most people, that is simply not the way statistics works. Provided the sample size is a good statistical representation of the population i.e. an appropriately randomized sample, the size of the population has little bearing on the determination of the appropriate sample size.

For example, given your population size of 1,100,000 law enforcement personnel, and a common confidence interval (aka margin of error) of +/- 3, to achieve a common confidence interval of 95%, you only need 1,066 respondents. In English, this means "You can be 95% sure that the true metric of the population you're trying to measure is between +/- 3 points of the measured sample statistic."

Now, let's increase the size of your population by ten-fold, to 11 million. You might think you would have to increase your sample size by ten-fold, as well, to 10,860, right? Nope. In fact, you only need to increase your sample size by one, to 1,067.

The reason this works is due to the "law of large numbers." "In probability theory, the law of large numbers (LLN) is a theorem that describes the result of performing the same experiment a large number of times. According to the law, the average of the results obtained from a large number of trials should be close to the expected value."

Look at it this way: If you flip a coin 50 times, you'll get a fairly decent idea that it will wind up heads roughly half the time. Flip it 500 times, and your level of confidence (the confidence interval) will improve, but it won't be ten times greater. Rather, it might inch up a bit, say, a few more percent. But if you flip the coin 5,000 times, you'll only be very slightly more certain that the odds of it coming up heads is 50% than when you flipped it 500 times.

Put another way, increasing the number of trials provides diminishing additional information:




Thus, the reason you said, "i do not understand how or why the initiators of the survey are so bloody proud of quote nearly 900 unquote" is because you do not understand statistics. But that's OK, as most people don't understand statistics.

However, the reason "why the initiators of the survey are so bloody proud of quote nearly 900 unquote" is because with a population of 1,100,000 law enforcement personnel, a sample size of 900 respondents, they can be 95% sure that the true metric of the population they're trying to measure is between +/- 1.11 points of the measured sample statistic.

That's pretty darn accurate, solus. Absolutely nothing to be ashamed of. At all.

ETA: I am a statistician. However, if anyone would like to understand basic statistics, the best source to date is Kahn Academy's Statistics playlist, available on YouTube. In just 68 courses averaging about 12 minutes each, you can gain a good appreciation of all things statistic beyond the simple task of computing a mean.

Lets say you want to say its more likely than not .. with 1.1E6 in the population ... and say about a 51% confidence level then that number of required persons to be surveyed goes way down LOL

With a 55% confidence level , allowing for 3% of a confidence interval .... then its 159

Fun with stats !

I abuse stats for political gain LOL

davidmcbeth said:

Zero. We train monkeys, not people.

Click to expand...

Wrong. Whether you're talking about law enforcement or military personnel of all ranks and branches, it's called "training."

And you want someone to list down every job duty ? Give someone a year to fully respond.

Click to expand...

Wrong again. The training pipeline for my duty in the Air Force was two years long, and that was after the mandatory four years of college required for selection. Additional and recurrent training for various specialties pushed that to about three years in-service training, total. Even so, I could easily recount i.e. "list down" every job duty I ever had -- and I had quite a few during my twenty years -- in several hours.

davidmcbeth said:

Lets say you want to say its more likely than not .. with 1.1E6 in the population ... and say about a 51% confidence level then that number of required persons to be surveyed goes way down LOL

With a 55% confidence level , allowing for 3% of a confidence interval .... then its 159

Click to expand...

A 55% C.I. is meaningless, only slightly better than "half wrong," unless, all you had was 159 respondents. Then, it would tell you that your sample statistic is only slightly more informative than "no clue."

since9 said:

A 55% C.I. is meaningless, only slightly better than "half wrong," unless, all you had was 159 respondents. Then, it would tell you that your sample statistic is only slightly more informative than "no clue."

Click to expand...

You can claim "its more likely than not" ... sufficient for political rhetoric...

davidmcbeth said:

You can claim "its more likely than not" ... sufficient for political rhetoric...

Click to expand...

But in what way and more importantly, to what extent? Politicians always claim they're better than their opponents. That's like saying Tom Brady is better than Joe Montana. How much better? 5%? Is that even statistically significant?

No.

since9 said:

But in what way and more importantly, to what extent? Politicians always claim they're better than their opponents. That's like saying Tom Brady is better than Joe Montana. How much better? 5%? Is that even statistically significant?

No.

Click to expand...

Its not 5% ... its more likely than not. This can cost you $$ in a civil case, and for political rhetoric, its fine.

Is it statistically significant? Its sure is. To the parameters set by me and defined by math.

So lets say you want to push for some legislation ... your district is 20000 folks. You only need about 60 people to survey to go to your representative and say that most people in his district support such legislation.

Politicians don't understand math and you don't have to present him with your data ... you gotta play the game and know how to play the game.

davidmcbeth said:

Its not 5% ... its more likely than not.

Click to expand...

Dave, even 1% better is "more likely than not."

This can cost you $$ in a civil case.

Click to expand...

Surely, you jest! Civil cases have been won by those with less than 1% advantage over defendants with a 99% advantage. Our system of justice is "usually" fair, but it remains all too often unfair.

So lets say you want to push for some legislation ... your district is 20000 folks. You only need about 60 people to survey to go to your representative and say that most people in his district support such legislation.

Click to expand...

Politicians don't understand math and you don't have to present him with your data ... you gotta play the game and know how to play the game.[/QUOTE]

Sadly, that's all too often true.