Leaderboard

@llawgoch And you have no understanding of the normalization factors commonly used in statistics to take out the widely varying exceptions. You see, I do. I use this type of statistical correction method with weather normalization that is embedded within our software platform, so you're mistaken when you say that I know nothing about statistics. Maybe you can give me your background in statistics and how you utilize in your profession.

Alex, as we've repeatedly tried to tell you, feel free to run the numbers any way you want but don't expect your results to be held in high regard. Garbage in - garbage out, it's really that simple. Your computer normalization program works so well on weather because you are only asking it to predict an overall average for temperature or rainfall for example and you have a huge sample size obtained over more than a hundred years with calibrated instruments which results in only a few outlying results that can be disregarded. Even with this ideal setup we are more than 30 degrees below our expected average temp today, have received snow and are expecting a killing frost overnight. I'm sure 30 degrees below normal falls well within the expected range or probability density of temps for today. Statistics are great for possible ranges based on lots of good data. They are also good for determining probability densities when even small samples are closely matched to limit noisy data and are precisely monitored - that's why drug trials use similar aged patients, with similar conditions, and base dosages on weight. The point being that there is lots of verifiable data and/or lots of controls to ensure good data. We don't have any of that with the bigfoot reports you plan on using. There is no control over who submits, no control over data accuracy and no way to clean up the data by normalizing because we have no standard like we would in a rainfall study with calibrated gauges. I believe you mean to use no criteria for inclusion but rather rely on throwing out some percentage of the reports? Are you using sightings only or including encounters based on sounds or possible footprints? SWAG1. What if 90% or more of whatever reports you choose were mistaken or hoaxed as some proponents believe? What if it's 50%? Because you are not using calibrated instruments but will instead rely on humans who are prone to error your study will obviously be less precise but by how much? Imagine basing your rainfall data on people's quick perceptions gathered in the amount of time a sighting takes place rather a calibrated gauge. How close would you expect the data to track now? Introduce SWAG2. What percentage will you ascribe to outright lying? Introduce SWAG3. How many are sightings of the same creature at different times or places? SWAG4. How many SWAGs will there be? Do you realize that depending on how you set this up potential errors in the SWAGs can compound? The difficulty in doing statistics is not in applying simple math to a sample but in choosing the sample and correctly determining what factors to apply. It's impossible to do if you if you can't trust the data itself due to unreliable reporting. All I can say is good luck!

Enough! dmaker should soon be out of plusses for the day

I was calling your comment pathetic; surely that was clear from the context. If you don't think saying "You are making less and less sense as this goes on" is insulting then I don't know what you think is. The fact is I was making sense, and your failure to understand is your problem, not mine, so if you want to back out of the conversation just do so without trying to have a dig at the other person. Tell me how the normalization factors apply to this situation then, and how my lack of understanding has caused any issues. I've explained exactly how your lack of understanding of how the Bigfoot sighting sample is obtained leads to a problem. Incidentally my knowledge of statistics is very limited and I do not use it at all in my profession, but that doesn't stop me giving you clear examples of why you are making an error here. So argue with the points I make instead of trying to argue from a claimed authority.

How did I insult you? I'm sorry if I hurt your feelings in any way. I thought I was leaving the conversation by saying I was bored. Now, who are you calling pathetic?

It's not "all people". It's "all people reporting Bigfoot". (And "mistaken or liars" rather than "stupid or liars") Seriously, if you can't see the difference this makes, you have no understanding at all of statistics and sampling.

Then move on without the insults and attempts to belittle me. Instead of telling me I am making no sense, simply admit you can't counter the argument, or leave it without making your snide remarks. Admin or no. If you wanted to move on all you had to do was stop misstating my position in your replies. No, pathetic. if you want to stop having a conversation on the internet, all you have to do is stop contributing to it.

Ok, try this--I'm bored with the conversation and wish to move on

You might be surprised. You'd hafta tread the Tar Pit to see my true thoughts, my friend. Another 23 postings, and you're eligible, should you wish to join the Premium Access section.

I don't think the wildlife survey paradigm works where awareness of self and deliberate planning / avoidance rather than operating on instinct are involved. That means I think the underlying assumptions are flawed. As Thom Powell once wrote, think Guerilla rather than Gorilla. If you fail, I already told you so. But I might be wrong. Good luck. MIB