Statistics HW pls help

[frankies]

here's the problems:

The Business Week 1993 U.S. subcriber Study collected data from a sample of 1597 subcribers. 66% of the respondents indicated that their annual income was $50,000 or more and 51% reported having an American Express credit card.

a) What is the population of interest in this study?
b) What is the sample in this study?
c) Is the ownership of an American Express card a categorical or quantitative variable? Why?

[felixdd]

here's the problems:

The Business Week 1993 U.S. subcriber Study collected data from a sample of 1597 subcribers. 66% of the respondents indicated that their annual income was $50,000 or more and 51% reported having an American Express credit card.

a) What is the population of interest in this study?
b) What is the sample in this study?
c) Is the ownership of an American Express card a categorical or quantitative variable? Why?

a) the world.
b) subscribers to the magazine only.
c) categorical

This is very basic stuff...maybe you should get some extra help or listen in class or something.

[Icedawn]

a) the world.
b) subscribers to the magazine only.
c) categorical

This is very basic stuff...maybe you should get some extra help or listen in class or something.

umm... personally...

a) population? all the subcribers of Business Week
b) sample? the 1597 people that were sampled...
c) categorical...

hate to say this, but maybe YOU should go back and listen in class ;)

[felixdd]

umm... personally...

a) population? all the subcribers of Business Week
b) sample? the 1597 people that were sampled...
c) categorical...

hate to say this, but maybe YOU should go back and listen in class ;)

Depends. You're talking about sample population. I'm talking about target population.

you're right on (b). I should be more specific that it's the 1597 of the subscribers, not all of them. But it's all semantics now.

[Icedawn]

Depends. You're talking about sample population. I'm talking about target population.

you're right on (b). I should be more specific that it's the 1597 of the subscribers, not all of them. But it's all semantics now.


Well, I'm talking about the answer in the context of the answer. Sure, technically, for every survey regarding people, a correct answer could be "the world". However, in this case, you have to look at the objective. 1597 subscribers were sampled out of ... who know how many actual subscribers, lets just say 10000 for example. so, these 1597 were asked questions in order to gain perspective upon the population of ALL the subscribers, not the population of the whole world.

[enforcerviper]

umm... personally...

a) population? all the subcribers of Business Week
b) sample? the 1597 people that were sampled...
c) categorical...


I'm with you.

[felixdd]

Well, I'm talking about the answer in the context of the answer. Sure, technically, for every survey regarding people, a correct answer could be "the world". However, in this case, you have to look at the objective. 1597 subscribers were sampled out of ... who know how many actual subscribers, lets just say 10000 for example. so, these 1597 were asked questions in order to gain perspective upon the population of ALL the subscribers, not the population of the whole world.

I see your point and it's a valid one. You're probably hearing my statistics prof speaking through me on this issue, as he emphasized repeatedly the flaws of a self-selected, non-random population. He'll keep saying that such a study is trying to be bigger and better than what it actually is.

In this case, since many magazine subscribers use a credit card to pay their subscription fees, wouldn't the resulting figures be artificially elevated and subject to more bias than, if done on a randomly selected and perhaps even stratified sample population?

Sure, the magazine can always phrase the hypothesis to their study such that their target population falls only within their subscribers. But the astute reader would immediately pick up on this little flaw in the study. As well, the more you constrict your target population, the less externalizing would be possible with the resulting data. How applicable would a study be with the hypothesis, "I want to know how many people go on RFD at exactly 6:00:00 AM EST as a sample of average useres online?". But I guess the issue isn't as "simple" as I once said it was, and I retract my statement about "listening in class" :o

[Icedawn]

I see your point and it's a valid one. You're probably hearing my statistics prof speaking through me on this issue, as he emphasized repeatedly the flaws of a self-selected, non-random population. He'll keep saying that such a study is trying to be bigger and better than what it actually is.

In this case, since many magazine subscribers use a credit card to pay their subscription fees, wouldn't the resulting figures be artificially elevated and subject to more bias than, if done on a randomly selected and perhaps even stratified sample population?

Sure, the magazine can always phrase the hypothesis to their study such that their target population falls only within their subscribers. But the astute reader would immediately pick up on this little flaw in the study. As well, the more you constrict your target population, the less externalizing would be possible with the resulting data. How applicable would a study be with the hypothesis, "I want to know how many people go on RFD at exactly 6:00:00 AM EST as a sample of average useres online?". But I guess the issue isn't as "simple" as I once said it was, and I retract my statement about "listening in class" :o


Ahhh... I love how although I'm a math major, I've never actually taken a stats course, (every stats course was really a probability course).

Hm, upon reading your response, it seems like you know what you're talking about as well. Now I'm curious why we're differing in our answers.

Tell me if I'm right

a) You interpreted that the objective of the study is to the gain perspective onto the whole world, because based on this assumption, then yes, I would agree with you.

b) I interpreted that the objective of the study is to gain perspective onto JUST the subscribers of the magazine. For example, if the magazine wanted to give to advertisers some data regarding the demographic reading it.

Is that why we are concluding different things?