solo diving survey results

[skip]

Many of you took the solo diving survey on "surveymonkey" at my request posted here on the forum awhile back. Preliminary results are in with responses from 298 divers (cave, wreck, reef, rec and tec). All numbers are 1-10 with 10 being most safe and 1 being least safe.

Is scuba diving safe? Mean is 8.02 with sd of 1.65.
Is buddy diving safe? Mean is 7.66 with sd of 1.79
Is group diving safe? Mean is 6.82 with sd of 2.10
Is solo diving safe? Mean is 7.00 with sd of 2.2
Can solo training make it safe? Mean is 7.83 with sd of 2.95

with such a large n, statistically significant results are a bit easy to find. even small differences become "significant." But here goes:

buddy diving considered safer than solo diving (7.66 vs 7.00).
buddy diving is safer than group diving (7.66 vs 6.82).
solo diving with training can be as safe as buddy diving and safer than group diving, and safer than solo diving without training (7.83 vs 7.66 and 6.82 and 7.00).

these differences are pretty small so I would not argue if someone held that personal opinions on safety all come out pretty much the same whether we are considering diving with or without a buddy or group.

I will be breaking it down by type of diver and gender, maybe age too...and report back then.

Thanks to all who took the time to help out.

-skip

[Luxrok]

I am assuming sd mean Standard Deviation. If so, I couldn't remember what it meant so I looked it up on Wikipedia. Here is what I found:

In probability and statistics, the standard deviation is a measure of the dispersion of a collection of values. It can apply to a probability distribution, a random variable, a population or a data set. The standard deviation is usually denoted with the letter σ (lowercase sigma). It is defined as the root-mean-square (RMS) deviation of the values from their mean, or as the square root of the variance. Formulated by Galton in the late 1860s,[1] the standard deviation remains the most common measure of statistical dispersion, measuring how widely spread the values in a data set are. If many data points are close to the mean, then the standard deviation is small; if many data points are far from the mean, then the standard deviation is large. If all data values are equal, then the standard deviation is zero. A useful property of standard deviation is that, unlike variance, it is expressed in the same units as the data.

So the quick translation for me is that most of the responses came within the sd on either side of the mean. Correct me if I am wrong.

You can find that definition for Standard Deviation on Wikipedia.

No onto my question, what do you mean by such a large N?

Hope that is as helpful to you as it is me.
Hans

[scblade27]

N is normally what is used for the sample size. In this case it is the 298 divers that responded. The bigger the N/people that respond, the more significant the results. The more significant the results, the more you can generalize about the total population.

i.e. If you asked 10 Americans who they were voting for president, your results would not be very significant. If you asked 200 million Americans, you would have much more reliable data.

[Luxrok]

N is normally what is used for the sample size. In this case it is the 298 divers that responded. The bigger the N/people that respond, the more significant the results. The more significant the results, the more you can generalize about the total population.

i.e. If you asked 10 Americans who they were voting for president, your results would not be very significant. If you asked 200 million Americans, you would have much more reliable data.

Thanks for the explanation. I hope I wasn't the only fool wondering.

[akcaver]

Within 1 standard deviation of the mean value, you should have 95% of all the answers. That means in the first question, Is Scuba Diving Safe? 95% of all the answers were between 6.37 and 9.67, and 98% of all the answers were between 4.72 and 11.32 (or 10, but in general stats 98% of all values/answers come within 2 standard deviations of the mean). Hope that helps also.

[MORGAN]

I still have bad memories of my statistics and probability courses after many years. I was forced to take them - I was a wildlife biology major; I wanted to be in the woods counting deer turds!

I still have a question about this standard deviation thing: How can anything that's standard be considered a deviation? Ain't that one o' them oxygenmaroons?

Mike

PS - I enjoy your research projects, Skip. I'm eagerly anticipating your study on the personality disord- I mean characteristics of cave divers. The analyses you sent Tracy and I were almost embarrassingly accurate.

[skip]

Yes, sd is standard deviation and yes, deviants can be standard and nonstandard!

99% of all scores are within + and - 3 sd's.

68% are within + and - 1 sd, so + and - 1 is the "normal" range of scores, e.g., if your IQ is within + and - 1 sd you are considered of normal intelligence (which is 85-115, mean is 100, sd is 15). if you are above 115, but below 130 you are bright, and if above 130 you are MENZA material. 85-70 is dull, and below 70 is retarded (excuse my lack of pc, what is it called these days...special, mentally handicapped?).

But sd's really only matter to statisticians, although comparing different means requires knowing the sd's associated with each mean. I did not present the statistical significance tests (anova and post hoc), but just summarized the meaning in those short comparative statements.

N is sample size, 298, and although things are never more or less significant (they are or are not sig), small differences are easier to detect with larger sample sizes (like more power on a micoscope lets you see smaller critters).

Sometimes this is not a good thing. If you don't like your results with N=10 or N=50, just ask more people until N = 1,000 and you will most likely get a "significant" finding! Unless there is no significant finding to find in the first place, then you won't find it no matter how many people you ask!

There is another statistic - effect size - that takes a look at "statistically significant" findings to see if the difference between two means is "meaningful." I have not done effect size calculations yet, but seeing how small the mean differences are I suspect the effect size will show that the difference is not very meaningful. Most researchers don't calculate effect sizes cause it's embarassing to say you spent all that time finding significant effects only to discover they weren't very meaningful! Indeed, psychology researchers never include effect size in their published findings (partly why psych is such a mess if you ask me). If everyone did publish effect size along with "significant differences," very few would have their grants renewed!

However, even if the significant differences found here are not meaningful differences, it's still rather interesting that the opinion of divers seems to be that solo diving is about as safe as diving with a buddy or group and certainly as safe with solo training.

In light of the recent updated accident analysis proposing that solo diving be added to the list of why divers die, we seem to have a community of divers who view solo diving as relatively safe. In other words, we have empirical evidence that seems to suggest that diving solo has become a major factor in diver deaths at the same time that divers perceive solo diving as safe. So should we take another look at the empirical data and see if the interpretation is justified (is the increase in deaths of solo divers real?) or should we adjust our attitudes?

-skip

[Sludge]

what is it called these days...special, mentally handicapped?

IQ 70 - 79: Borderline mentally handicapped
IQ 50 - 69: Mildly intellectually handicapped
IQ 35 - 49: Moderately intellectually handicapped
IQ 20 - 34: Severely intellectually disabled
IQ below 20: Profoundly intellectually disabled

I work with the top two on a regular basis, and the third one occasionally, in the public high schools.

[Major Restriction]

and if above 130 you are MENZA material.
-skip

Mensa.

[EGIB]

IQ 70 - 79: Borderline mentally handicapped
IQ 50 - 69: Mildly intellectually handicapped
IQ 35 - 49: Moderately intellectually handicapped
IQ 20 - 34: Severely intellectually disabled
IQ below 20: Profoundly intellectually disabled

I work with the top two on a regular basis, and the third one occasionally, in the public high schools.

What kind of teacher are you?