@joeyd999
After seeing that ad, I don't care about the doctors and surveys - I need to learn more about this T-zone story in the lower right corner.

 

Sorry, I thought this was in 'Off-Topic' and attempted to comply.

I hope you don't get banned again. I really thought THAT earlier post for which you got banned was funny. Perhaps it was too close to fact to be appreciated for what it was.

 

View attachment 226306

The top page headings only show if you're at the top of the page. This is not necessarily the case if you've previously viewed the thread and then later clicked the link for "new posts".

I hope you don't get banned again. I really thought THAT earlier post for which you got banned was funny. Perhaps it was too close to fact to be appreciated for what it was.

I don't know to which post you are referring. But, yes, my humor (and more frequently these days -- my annoyance with this site) is wearing thin on many here.

Getting banned would not be so horrible. I made a New Year's wish last year. Alas, my stars didn't align (then again, it was 2020; who's did?).

 

@joeyd999
After seeing that ad, I don't care about the doctors and surveys - I need to learn more about this T-zone story in the lower right corner.

First, you'll need to walk a mile for a Camel.

 

First, you'll need to walk a mile for a Camel.

Heck, my driveaway is 2000 feet to the mailbox. Close. I have one small pipe "spiff" (not MJ ) per day. That's where I get my tobacco.

That banning seems to have been in relation to the Coronavirus thread when you gave a list of acceptable, vetted sources such as MSNBC, ABC, CBS, ESPN, and NPR. Frankly, I thought it was very well done and funny, but I was just a little too late to give a like before it vanished.

 

... acceptable, vetted sources such as MSNBC, ABC, CBS, ESPN, and NPR.

That looks a lot like the list of folks accepting goodies from the Chinese Communist Party.
https://thefederalist.com/2020/05/04/has-china-compromised-every-major-mainstream-media-entity/

 

That banning seems to have been in relation to the Coronavirus thread when you gave a list of acceptable, vetted sources such as MSNBC, ABC, CBS, ESPN, and NPR. Frankly, I thought it was very well done and funny, but I was just a little too late to give a like before it vanished.

Ahhh...I remember!

I've come to suspect that there is a cadre of "usual suspects" that routinely report posts of either a particular provenance or of certain topics (or of one side of those topics, actually). The powers-that-be take the path of least resistance, AFAIKT.

FYI: I've never reported a post or a thread that was not spam*. I just ignore members from whom I don't wish to hear (only 2 on the list, currently).

*Edit: except for grievous violations like plagiarism -- the world should know that "original" contributions aren't!

 

Hey people, we are way off topic. I think Andrew had another, legitimate plan for this thread. I was tempted to give the Camel quote early on, but respected his perspective (post #5 ). We all know, or should know, that market share for a very large number of products depends entirely on the number of virtually identical brands of a commodity that a manufacturer makes, Liggett and Myers developed that strategy a long time ago.

 

Thanks guys

I know its new year to the western world
have a happy one, and in a years time, those still here, remember today.

And if anyone has a hard number answer to the question, I would like to know,

 

Hey people, we are way off topic. I think Andrew had another, legitimate plan for this thread. I was tempted to give the Camel quote early on, but respected his perspective (post #5 ). We all know, or should know, that market share for a very large number of products depends entirely on the number of virtually identical brands of a commodity that a manufacturer makes, Liggett and Myers developed that strategy a long time ago.

Thank you @jpanhalt

 

As n increases, the binomial distribution looks increasingly like the normal distribution with SD = sqrt(n/4). Here n = 67 and so SD = sqrt(67/4) = 4.1. Allegedly 43 "heads" were observed. That's 9.5, or 2.3 SD, more than the expected result.

For most published work, a finding more than two SD away from the null hypothesis is considered statistically significant. That alone is falling out of favor as "proof" of anything.

The normal distribution gives the probability for x heads in n flips as:



I get 1.345% for the values in this case.

I think an important point is being missed in all this. Customers rate a shampoo on numerous factors including smell (before and after, in the hair), cleaning potency, foaminess, and so on. The ad under discussion mentions none of these. "Hair shininess" is not a feature I would have thought of. I'm guessing the marketing folks did their research – focus group testing and polling – and found out that their product had a standout property of "shininess", which has a value in the market.

When your product offers a unique property, that's what you advertise. You hope to connect to the niche in the market that most value that unique benefit of your product. Their words are wasted on luddites like me that attach no value to "shininess".

 

When your product offers a unique property, that's what you advertise. You hope to connect to the niche in the market that most value that unique benefit of your product. Their words are wasted on luddites like me that attach no value to "shininess".

I've got some carwash soap in the garage that contains carnauba wax as one of its ingredients. I'm gonna substitute this for my wife's shampoo to see what happens.

 

I've got some carwash soap in the garage that contains carnauba wax as one of its ingredients. I'm gonna substitute this for my wife's shampoo to see what happens.

I if you look into it, you'll find hair shampoos claiming words like shines, gloss, bright will contain 1-3% carnuba wax or, PALM wax (another commercial name for the same product). The art of cleaning is the art of selling. Not much difference in any grease cutting, dirt dispersing, adhesion breaking formulations. Marketing people found that using the name, CARnuba was great for automotive shampoos. I learned so much working for P&G for a few years - more about marketing than electrical engineering and process automation. Summary, your wife may appreciate the gloss.

 

I've got some carwash soap in the garage that contains carnauba wax as one of its ingredients. I'm gonna substitute this for my wife's shampoo to see what happens.

I don't see what this has to do with understanding the statistics,

 

I if you look into it, you'll find hair shampoos claiming words like shines, gloss, bright will contain 1-3% carnuba wax or, PALM wax (another commercial name for the same product). The art of cleaning is the art of selling. Not much difference in any grease cutting, dirt dispersing, adhesion breaking formulations. Marketing people found that using the name, CARnuba was great for automotive shampoos. I learned so much working for P&G for a few years - more about marketing than electrical engineering and process automation. Summary, your wife may appreciate the gloss.

I don't see what this has to do with understanding the statistics,

 

There are three kinds of lies: lies, damned lies, and statistics.

 

All surveys are inaccurate because of the observer effect.

Ask any question: "How much alcohol beverages do you consume in a week?"

The moment you ask the question, the interviewee is already biased one way or another in their thinking and response.

 

There is another confounder that may not be considered by everyone. That is, how was the survey conducted?

I doubt whether it was a simple yes/no question, like is your hair brighter? More likely it was some version of the "5-star " system, at least in the United States that is what I have seen. Look at the Amazon system, is 3-star good or bad?

Here's an artificial example of a client satisfaction survey:

• Extremely satisfied
• Somewhat satisfied
• Satisfied
• Not satisfied
• Extremely dissatisfied

I obviously have manipulated those responses; paid consultants are much more subtle. Now, for such an ordinal list, is an average valid? Or, more likely, the department head will report to the manager group, "60% of our clients were satisfied to extremely satisfied." Don't laugh. I have seen hospital satisfaction surveys completed while patients were still in recovery, and while conscious, they were very "high." It even happened to me after a colonoscopy with analgesia many years ago. My response, "You have to be kidding."

 

To answer the TS's question, if you toss a penny 100 times and record your head/tail ratio.
You'll have to do about 1000 sets of 100 tosses to see that 1 to 2 times out of 1000 sets will get a 67/33 distribution or more. Some random sets will have zero and some will have 3 or more in a lognormal distribution - but most groups of a 1000 sets (of 100 flips per set) will have 1 or 2 sets of 66 or more HEADS. If you want to count any wide distribution, 67 or more heads OR, 67 or more tails, then double the probability.

So, to answer your question,

If lets say, the odds of a person agreeing is 50:50
How many would we have to ask, to have a 50:50 chance of finding a run of 67 where 43 agree ?

It is not a question of how many coin flips would it take but, "how many groups of a 100 would you have to hit before getting that ratio?" If you have a fair sample size.
For a reasonable sample size of 100, the answer is about 1 or 2 per 1000 samples of 100.

Finally, your odds go up considerably if you only sample 3 women to get two (66%) in that case, half of the samples of three people will have a 66% or more skew of like vs dislike. Or, specifically, 37.5% of two women will like it and 12.5% that 3 of 3 women will like it.

The greater your sample size, the greater chance your results will be close to 50:50

 

@MrSalts
And your source for how the survey was conducted is? Please fill us in on the details? Moreover, the group was 67, not 100.

Doesn't that number seem a bit odd?