What ever happened to the 2015 deer harvest survey???

[fullthrottlehunter]

Ridge, give me some time and I'll be posting up an example of why and how bias effects survey reports. More specifically...Why the DNR strives to achieve a higher return rate, and how a lower return rate can introduce non-response bias to the results. This non-response bias can skew not only the actual results, but also effect predicted/long-term estimated results by quite a bit. The DNR realized this and made adjustments based on other historical years where data was collected but return rate varied greatly.

If return rates continue to fall, I fear their long term data collection results may appear to vary wildly from year to year. The higher the return rate, the less chance there will be for variability and a higher chance that non-response bias will not skew the actual results.

[Dale Malusi]

Jul 19, 2016 9:42:48 GMT -5 fullthrottlehunter said:

Jul 19, 2016 6:26:19 GMT -5 Dale Malusi said:

Simple Definition of estimate

: to give or form a general idea about the value, size, or cost of (something) : to make an estimate of (something).

Why try to further dilute a guess?

That's where your logic or understanding of this subject falls short. Their estimates aren't just a random guess. They're based on many factors compiled through years of statistical analysis.


From the report......

"Statistical tests are used routinely to determine the likelihood that the differences among estimates are larger than expected by chance alone. The overlap of 95% confidence intervals was used to determine whether estimates differed. Non-overlapping 95% confidence intervals were equivalent to stating that the difference between the means was larger than would be expected 995 out of 1,000 times, if the study had been repeated (Payton et al. 2003)."

Sorry Dale, they're not a simple guess.

I am not short on anything.
An estimate is a guess, plain and simple

[fullthrottlehunter]

For those interested, I will honestly try to explain to you how non-response bias effects survey results. There is no conspiracy or data manipulation going on by the DNR guys.

I'm going to take a sample size of 10 to make things easier to follow along with. In the real world there are far more than 10 samples, so the effects of non-response bias may be quite less than with a sample size of 10. However, the point is to show what non-response bias can do to a sample. I will also be using one of the data points that the DNR collects on, Hunter Effort (days). Hunter Effort being the total number of days hunted divided by the number of hunters. This number has been around 14 days per hunter for the last two harvest reports, and still around 14 days looking back to 2006-2007. Keep in mind that this is for all seasons combined. Hunter Effort (days) was higher during archery season. Also keep in mind that the Hunter Effort (days) was adjusted for the 2014 Deer Harvest Survey. The adjustment was done to take into account the effects of non-response bias and was decreased 4.7%, which equals 0.6 (days) per hunter.

Nonresponse Bias:

Sometimes, in survey sampling, individuals chosen for the sample are unwilling or unable to participate in the survey. Nonresponse bias is the bias that results when respondents differ in meaningful ways from nonrespondents. Nonresponse is often problem with mail surveys, where the response rate can be very low.

Let's get started....

Hunters - (days hunted).

1. 22

2. 5

3. 10

4. 18

5. 25

6. 17

7. 10

8. 13

9. 12

10. 8

Example #1 - 30% return rate

Hunter 1, 3, and 6 returned surveys. It's impossible to ascertain why only those three returned them. You now are calculating hunter effort (days).

22 + 10 + 17= 49

Divide 49 by 3 to get average hunter effort (days). This gives you 16.3 days of hunter effort average.

Example #2 - 50% return rate

Hunter 1, 2, 3, 6, and 9 returned surveys. It's impossible to ascertain why only those five returned them. You now are calculating hunter effort (days).

22 + 5 + 10 + 17 + 12 = 66

Divide 66 by 5 to get average hunter effort (days). This gives you 13.2 days of hunter effort average.

Example #3 - 70% return rate

Hunter 1, 2, 3, 5, 6, 9, and 10 return surveys. It's impossible to ascertain why only those seven returned them. You now are calculating hunter effort (days).

22 + 5 + 10 + 25 + 17 + 12 + 8= 99

Divide 99 by 7 to get average hunter effort (days). This gives you 14.1 days of hunter effort average.

Example #4 - 100% return rate

All hunters return surveys.

22 + 5 + 10 + 18 + 25 + 17 + 10 + 13 + 12 + 8= 140

Divide 140 by 10 to get average hunter effort (days). This gives you 14 days of hunter effort average.

As you can see, Example 1 and 2 results vary widely. It's also noted that the results are either well above or well below the actual result when all are hunters returned their survey. It's not until you reach the 70% threshold that things start getting closer to the actual result when all hunters return their survey. You don't know their reasoning for returning them or not, but the much larger return rate decreases the chances of non-response bias. I hope this helps some understand more about the statistics and math behind these surveys instead of just assuming that they are being manipulated for no reason. They're not.

[daappleknocker]

FTH, your last post made me laugh. For several years I have been saying that the sample size and the response bias in the MAR's surveys have led to  manipulated results. I have been called a conspiracy theorist who wears a tinfoil hat. Now, when it comes to the ADHSR, it suddenly becomes acceptable? The funny part is that they do not differentiate between the actual harvest data, finite and not affected by response bias, and opinion data, affected by response bias. Harvest data and opinion data are two very different animals. As in the MAR's surveys, the DNR has proven that those who advocate for MAR's are much more likely to return their surveys, and are encouraged to do so through organization.

[daappleknocker]

FTH, the only thing that your example? shows us is that hunter effort from archery hunters and firearms hunters should be separated. Also, do not pretend that you do not know why hunters do or do not return surveys. You do know and it is designed (manipulated) into the process. How many red M&Ms do you have?

[fullthrottlehunter]

Appreciate the personal attack for simply trying to explain this to the readers here. Not surprised. I give up. Continue to bury your head in the sand, spread nonsense, and preach conspiracy theories. Don't care.

[daappleknocker]

Personal attacks? Because someone refutes your posts, does not make it a personal attack. so what you are saying is that 51% return rate is acceptable but a 49% return rate is not? I totally understand your example above. But, you are saying that in the ADHSR, 28,000 hunters are not enough to get a reliable result? But, in the MAR's surveys they accept the results with much fewer hunters responding.

[fullthrottlehunter]

Jul 22, 2016 7:13:57 GMT -5 daappleknocker said:

Personal attacks? Because someone refutes your posts, does not make it a personal attack. so what you are saying is that 51% return rate is acceptable but a 49% return rate is not? I totally understand your example above. But, you are saying that in the ADHSR, 28,000 hunters are not enough to get a reliable result? But, in the MAR's surveys they accept the results with much fewer hunters responding.

No, I said nothing of the sort. I did show that if response rates drop, non-response bias may be present. To what extent may never be determined. In regards to the MAR survey....you know full well that is a completely different type of survey. Not only a different type of survey, but a smaller population of hunters are part of the sample. Statistically speaking, as long as response rates remain high (which they were extremely high in the MAR survey), the results of the MAR survey are very solid. Whether that be a pass or fail.

[daappleknocker]

FTH, will you at least agree that all of the MAR's surveys are invalid because they are far from random and unbiased? I say this because the prerequisite of receiving one is that you had to return your Annual Deer Harvest Survey. The MDNR-WD proved that hunters who belong to hunting organizations are far more apt to return surveys and support MAR's regulations when they do. (2001 Hunter attitudes on QDM by Peter Bull and Dr. Ben Peyton.)

[daappleknocker]

For what its worth, in the above survey, every member (439) of the QDMA Chapter in Michigan was sent a survey. The results? An 82% response rate! Compare this to the general hunting public where it took three survey mailings and a reduction in questions to achieve a response rate of over 50%. Now, explain to me why there is not an obvious bias in the MAR's survey process. I'm waiting. 

[fullthrottlehunter]

Curtis, this has been explained time and time again over the years. I completely disagree with your assertion that the MAR survey was biased or faulty. There simply isn't any proof of such. I'm not going to pontificate what if's with you. The fact remains, whether or not an individual is more predisposed to returning a survey in no way predicts how they will likely vote. Any statement claiming such is pure conjecture.

[daappleknocker]

That's funny because even Peter Bull and Dr. Ben Peyton admit that members of deer hunting organizations are more apt to return surveys AND more apt to support QDM. MAR's IS a QDM agenda. Read this important report, it might open your eyes. Plus they admit that the QDM philosophy will create a smaller deer herd, fewer hunters and lower success rates. Read my other post in this wild west section, the statistics prove it. If you can't understand or refuse to see the facts, move on. I can't state it any clearer.

[fullthrottlehunter]

Show the readers of this forum the report, and maybe we could then discuss it. Let's see it.


Peyton, R.B. and P. Bull. 2001. An assessment of possible antler restrictions and quality deer
management by Michigan deer hunters.

[daappleknocker]

Don't know how. Not that computer savvy. Was given to me as a participant in the 2008-2009 Michigan Deer Advisory Team. It is not a report that they would leave on their site as it does not support their agenda of QDM. If anyone can find it, please put up the link.

[Dale Malusi]

Jul 22, 2016 9:56:40 GMT -5 fullthrottlehunter said:

Curtis, this has been explained time and time again over the years. I completely disagree with your assertion that the MAR survey was biased or faulty. There simply isn't any proof of such. I'm not going to pontificate what if's with you. The fact remains, whether or not an individual is more predisposed to returning a survey in no way predicts how they will likely vote. Any statement claiming such is pure conjecture.

So what about Maprs protecting fawns and such, isn't that wording bias? Or was it just faulty management ? After all the DNR admitted it wasn't kosher and should not have happened. What about that.
Predisposition or not, you should not have to "adjust" anything.
Now if we want to talk about what this surveys real purpose is............

[ridge]

The only place that I can find it is at the end of a large group of reports.  The QDMA eliminated it from its original position.  I will try to copy and paste here: (This is not the original report but a DNR Summary which is much different than the one I linked in the last reply to this report.
www.learningace.com/doc/594966/cb626b41389d69b1a4fbb5fe3066f6df/perm-annual-report-98-03, Pages 51-52)
The QDMA and the DNR have omitted and blocked the original report.  Perhaps someone better than I can find the original one.  It is probably in the MSU Library but I do not have time at the present to search those files.)  Please see my last reply for the best summary of this report.)


An Assessment of Possible Antler

Restrictions and Quality Deer

Management by Michigan Deer Hunters


Management Problem


As defined by the Michigan Department of

Natural Resources (MDNR), “Quality Deer

Management (QDM)” is an approach to deer

management that requires restrictive buck

harvests and sustained antlerless harvests to

produce a more balanced sex ratio and

populations in balance with available habitat”

(MDNR 1999). The QDM Association’s

definition is more holistic, but both definitions

imply that QDM is more than trophy buck

management. Michigan has been

experiencing a growth in interest among

some hunters to have special regulations in

their hunting areas, which protect yearling

bucks. The growing interest presents policy

questions and increased issue activity. This

study was undertaken to sort through the

perceptions of Michigan hunters to provide a

baseline understanding of hunters’ support

for QDM-related restrictions as well as

assess their understanding of the range of

goals and strategies involved in QDM.


Study Description


A questionnaire was mailed to a random

sample of 9,500 deer license holders in the

spring of 2001. The response rate adjusted

for non-deliverables was only 60%. A short

non-response follow-up survey and

comparisons with age and residence data in

the original sample database suggested that

non-response bias would not dramatically

influence the results. In addition to the

random sample of hunters, surveys were

mailed to all members of the Michigan

Chapter of the QDM Association. These

responses were compared with responses

from the larger random survey. Response

rate to the QDMA survey was 82%.


Findings


Over a third of our respondents had not heard

of QDM prior to our survey. Only a fifth knew

someone who practiced QDM on land they

owned or leased. Areas where QDM-related

regulations have been established to protect

yearling bucks were not known to most of the

respondents. The largest group of informed

respondents were the 80% of the QDM

Association practitioners who knew of one

example. As a group, respondents reported

they discussed QDM very little, although they

did discuss QDM-related topics such as

protecting spike bucks. These and other

results suggest that issues associated with

QDM practices are known to most deer

hunters, but QDM is not recognized as a

formal, well-defined management approach

such as that defined by the QDM Association.

Most QDM goals and strategies were

recognized by more than 65% of

respondents. However, many erroneously

thought supplemental feeding was a QDM

strategy in Michigan, including over a fourth

of QDM Association members.


A majority (55%) of our respondents agreed

that the agency should manage for an older

age structure among bucks in Michigan. We

posed four regulations that would contribute

to older age classes of bucks. Respondent’s

supported (59%) the idea of an antler

restriction to protect yearling bucks in their

hunting area. Support was essentially

unrelated to type of land access or age of

respondent.


Of the four regulations proposed, the one

buck rule received the most approval. The

earn-a-buck proposal was least liked (39%

supported, 43% opposed). Although more

publication

analysis

data collection

design

2000 2000.5 2001 2001.5 2002
52

respondents supported the use of a buck tag

for a button buck (54% vs 28% opposed), and

a majority of those who supported this option

strongly supported it, two-thirds of those

opposed were strongly opposed. All three

options would be highly contentious if

proposed in the state in spite of some

majority support.


The Natural Resources Commission process

for evaluating proposed QDM-related antler

restrictions in a DMU includes a survey of

hunters in the area in which key questions

present an “I don’t care” as well as

“undecided” response option. This would

allow “don’t care” responses to be dropped

when calculating the approval rate among

respondents. In this study, we found that 4%

did not care and 13% were undecided about

antler restrictions to protect yearling bucks.

Even a 4% proportion makes the measure

worthwhile since it reduces the number of

respondents who must approve the proposal

to achieve the 66% approval rate.


Another contentious aspect of the QDM

evaluation procedure has been the

requirement that the mail survey achieve a

66% approval for the proposed regulation

change. We asked respondents what

percent of hunters should approve a

proposed antler restriction change before it

was put in place. Statewide, the respondents

recommended an average 61% approval rate

which is acceptably close to the 66%

approval requirement when “I don’t care”

responses are considered.


Communication is a useful tool to address

QDM-related issues and our findings

provided some implications for

communication efforts by both the agency

and private interest groups. Some

recommendations were broad, generic

approaches to target messages statewide.

Additional findings suggest using

segmentation to target messages and

strategies to specific groups. One major

misconception, which needs to be addressed,

involves supplemental feeding. Over a fourth

of the respondents believed winter feeding to

supplement habitat was a strategy of QDM.

There is a need to inform hunters--and QDM

Association members--that supplemental

feeding is not an acceptable strategy in

Michigan. Findings suggest not all

information sources will be equally effective

for all hunter segments. Sources vary in the

credibility assigned them and the frequency

of use by hunter segments relating to region,

age, and preference of hunting method.


Current Status


Study completed.


Investigators

Wildlife Division

Brian Frawley


Michigan State University

Ben Peyton, PI

Peter Bull


Reports


Bull, P and Peyton B. 2001. An assessment

of possible antler restrictions and Quality

Deer Management by Michigan deer hunters.


Bull, P and Peyton B. 2001. Report to

Michigan wildlife managers: The perceptions

of Michigan deer hunters regarding Quality

Deer Management (QDM) and related issues.


Bull, P and Peyton B. 2001. The perceptions

of Michigan deer hunters regarding Quality

Deer Management (QDM) and related issues.

[ridge]

Here is another interesting submission by the same authors:  The Perceptions of Michigan Deer Hunters Regarding Quality Deer Management (QDM) And Related Issues


citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.524.7482&rep=rep1&type=pdf

[ridge]

Here is an Extremely interesting report:  Are Michigan Deer Hunters Satisfied Stewards or Coerced Conservationists? (also by Peyton and Bull)


michigansaf.org/Tours/05Deer/07-Peyton.pdf

[ridge]

Here is a FASCINATING SUMMARY OF THE PEYTON AND BULL REPORTS:  (AN ABSOLUTE MUST READ!)

IT WILL NOT COPY AND PASTE.  PLEASE GO TO PAGES 105-106 of this PERM Annual Report.  It is the best of all of the above.

www.learningace.com/doc/594966/cb626b41389d69b1a4fbb5fe3066f6df/perm-annual-report-98-03

[daappleknocker]

Ridge, the one I liked above is the one I was talking about. Thanks for your hard work and search. It proves that FTH is wrong and the MDNR knows the truth about hunter acceptance of QDM and MAR's.

[ridge]

According to a librarian that I just spoke to, this report may be over 100 pages in length.  I have an email in to the MSU library for clarification.

[fullthrottlehunter]

Jul 22, 2016 12:04:32 GMT -5 daappleknocker said:

Don't know how. Not that computer savvy. Was given to me as a participant in the 2008-2009 Michigan Deer Advisory Team. It is not a report that they would leave on their site as it does not support their agenda of QDM. If anyone can find it, please put up the link.

You want people to read a report that you don't have or cannot find access to, but pawn off as some type of proof? Let's see the report, look at it, and then discuss it like adults.

[fullthrottlehunter]

Jul 22, 2016 17:17:57 GMT -5 daappleknocker said:

Ridge, the one I liked above is the one I was talking about. Thanks for your hard work and search. It proves that FTH is wrong and the MDNR knows the truth about hunter acceptance of QDM and MAR's.

Im wrong? Let's see the report so we can discuss it like adults. I'm sorry, I don't just take someone's word for anything when discussing this topic. That's not meant to be a slam directed at you or anyone. I look at information, facts, from both sides first.

[fullthrottlehunter]

From a post above by Ridge....



A majority (55%) of our respondents agreed that the agency should manage for an older age structure among bucks in Michigan. We posed four regulations that would contribute to older age classes of bucks. Respondent’s supported (59%) the idea of an antler restriction to protect yearling bucks in their hunting area. Support was essentially unrelated to type of land access or age of respondent.

Let's look at that real quick.

9423 surveys were mailed to hunters and 439 surveys were mailed to QDM members. That's 9862 total surveys. 59% of respondents supported the idea of an antler restriction to protect yearling bucks.

Let's just assume that all 439 QDM members wanted them and count them as 100% supporting.

How about some math....

9862 total surveys x 59%= 5818

5818 - the 439 from QDM= 5379

5379 ÷ 9862= 54.5%

What does this tell us?
That even when all QDM members supported the antler point restriction, 54.5% of the remaining respondents supported them.

54.5% - That's a majority gentlemen.

[QDM MARS]

Jul 22, 2016 18:26:17 GMT -5 ridge said:

According to a librarian that I just spoke to, this report may be over 100 pages in length.  I have an email in to the MSU library for clarification.

Thanks for taking all the time Ridge.

[fullthrottlehunter]

My above example is based on if everyone returned their survey. They didn't, so let's look at how the numbers actually came in.

60.4% of 9423 returned the survey which equals 5691. 82% of the 439 QDM members returned theirs which equals 360.

This equals 6051 returned surveys. 59% of that is 3570. 3570 minus the 360 equals 3210. 3210 ÷ 6051= 53%

Of those that returned surveys, 53% of non QDM members supported an antler point restriction to protect yearling bucks if we assume all QDM respondents supported them. Do you think 100% of QDM members support them?

[ridge]

Jul 22, 2016 19:26:22 GMT -5 fullthrottlehunter said:

From a post above by Ridge....



A majority (55%) of our respondents agreed that the agency should manage for an older age structure among bucks in Michigan. We posed four regulations that would contribute to older age classes of bucks. Respondent’s supported (59%) the idea of an antler restriction to protect yearling bucks in their hunting area. Support was essentially unrelated to type of land access or age of respondent.

Let's look at that real quick.

9423 surveys were mailed to hunters and 439 surveys were mailed to QDM members. That's 9862 total surveys. 59% of respondents supported the idea of an antler restriction to protect yearling bucks.

Let's just assume that all 439 QDM members wanted them and count them as 100% supporting.

How about some math....

9862 total surveys x 59%= 5818

5818 - the 439 from QDM= 5379

5379 ÷ 9862= 54.5%

What does this tell us?
That even when all QDM members supported the antler point restriction, 54.5% of the remaining respondents supported them.

54.5% - That's a majority gentlemen.

Page 3 of the citeseerx report disagrees with this post.  It states, " 'QDM' is not particularly salient for the general hunting population."  and "QDM is not recognized as a formal, well-defined management approach." and "It provides further evidence that 'quality deer management' has not yet become a well-defined and standardized movement."  "No doubt the NRC will continue to face regulation-specific issues primarily involving production of more mature bucks, but is less likely that strong public demands will be heard in  the near future . . ." 

[hartman756]

54.5% - That's a majority gentlemen.

I make no claims to the accuracy of your figures . My point has to do with when was it  we made this change to setting game regs based on a popularity contest. And why is the QDMA the only special interest group that gets the privilege to have the DNR/NRC hold such a popularity contest on their behalf ? If that is the way we are going to do things now, why do we even need to spend tax dollars on the DNR/NRC ? And lets make it fair and do it for any and all special interest. Maybe the DNR/NRC  can even sponsor  a vote on behalf of the ANTI's  on  the gam regs they want in place

[ridge]

My problem with the report is that in the DNR's own words, "It is very difficult to measure these biases." (page 6)  See Pinefarm's thread on the survey report just out for which he put a link on M-S.  They admit this is true for low response bias. How can they be sure that their estimate based on historic behavior of data is correct. There are varied factors that could be involved in a change of data behavior. Assuming that the rate is the same or similar is common but it can corrupt the possible true results of a survey. I do not know whether or not that has happened with the survey report. It may be impossible to ever know for sure, which is why, not publishing the original data can cast a shadow on the results that are given.

It should not be a chore to publish the original data or results as they require it to make their mathematical estimates or adjustments.  Showing consistency in trends may be important but it is not the only vital service of a survey report.  Those trends only have value if their basis are made on true continuing rates.  That value is completely compromised if some factor created a change in those trending ratios which is why having the original data is equally important.

[fullthrottlehunter]

We have already covered what was changed.

"Estimates of harvest, hunting effort, and hunter numbers were reduced by 6.4%, 4.7%, and 1.9%, respectively...."

Please, explain why this matters. I've gone one step further and showed what a 4.7% reduction means in regards to hunter effort (days). With nearly 2 decades of statistical sampling to compare to, I'm confident they understand why and how the change needed to be made.

The problem here is a few guys that don't understand statistics cry fowl and proclaim manipulation. It's a perception problem, not a manipulation one.