Login

Welcome, Guest. Please login or register.

May 24, 2024, 04:45:49 pm

Author Topic: 1st year stats question - sampling  (Read 10852 times)  Share 

0 Members and 1 Guest are viewing this topic.

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #15 on: April 14, 2020, 11:05:50 pm »
0
2.2 What’s the design? Read the brief description related to a statistical study below, and
 a) explain why it is an experiment rather than an observational study.
Runners recovering from lower leg fractures were randomly allocated to three
different training programs. Half of each group also followed a high calcium diet.
Bone density measurements were taken for all subjects.
Also:
b) identify the factor(s) in the experiment and the number of levels for each.
c) give the total number of treatments.
d) state the response variable(s) measured.
e) explain whether it was blind, or double-blind

2.3 Some studies suggest that having a pet can provide positive mental and physical health benefits for the owner.
A particular study involving elderly people who lived alone reported that dog owners had lower blood pressure on
average than the non dog-owners, with the difference in average blood pressure between the groups statistically
significant.
a) In this context, what does “statistically significant” mean?
b) Explain why this study would be an observational study rather than an experiment.
c) Does this result mean that if an elderly person buys a dog that their blood pressure will decrease? Explain.
d) Suggest some possible associated confounding variables for this study.

Can I please have help with these questions??
I'm not sure how to do any of them
Thanks

????
I'm not sure how to do any of them, which is why I haven't shown any working out

pahm

  • Forum Regular
  • **
  • Posts: 74
  • Respect: +22
Re: 1st year stats question - sampling
« Reply #16 on: April 15, 2020, 10:31:42 pm »
+2

I don't think most of these questions are even relevant to the study design. can anyone confirm?
❝ Everything is possible for one who believes ❞
Mark 9:23
.
Economics [45] • Legal [47] • English • Methods • Specialist

Sine

  • Werewolf
  • National Moderator
  • Great Wonder of ATAR Notes
  • *****
  • Posts: 5135
  • Respect: +2103
Re: 1st year stats question - sampling
« Reply #17 on: April 15, 2020, 10:36:37 pm »
+2
I don't think most of these questions are even relevant to the study design. can anyone confirm?
Yeah, they aren't. I think they are uni assignment questions.

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #18 on: April 15, 2020, 10:53:36 pm »
0
For the following statement about a statistical survey, identify the following items (if possible). If you can’t tell,
then say so – this often happens when we hear or read about a survey.
A recent online poll conducted by an Australian newspaper of 1083 registered voters found
 that 43% of the respondents believe it is likely that bushfires are linked to climate change

a) The possible population of interest
Registered voters
b) The population parameter
proportion of registered voters who believe it's likely that bushfires are linked to climate change
c) The sampling frame
1083 registered voters
d) The sample
43% of 1083
e) The sampling method, including whether or not randomisation was employed
voluntary response survey, randomization is NOT used
f) Who (if anyone) was left out of the study
people who didn't vote and people who didn't participate in the survey
g) Any potential sources of bias you can detect and any problems you see in generalising to the population of
interest
people are more likely to participate in the survey if they have a strong opinion towards either side
Can you please tell me if my answers are correct? Be as harsh as you can
And yes, this is 1st year uni maths, and I don't have a textbook why is why i'm having trouble with these questions
Thanks

Bri MT

  • VIC MVP - 2018
  • Administrator
  • ATAR Notes Legend
  • *****
  • Posts: 4719
  • invest in wellbeing so it can invest in you
  • Respect: +3677
Re: 1st year stats question - sampling
« Reply #19 on: April 16, 2020, 09:23:57 am »
+1
There are some parts of your answer which you could improve by using information you have already been given so I would suggest that you reread the advice you have already received and incorporate that before asking for further help & feedback.

In particular, you have been explicitly told what the sample is for this question and been given more sources of bias.


For stats the internet has a lot of resources and information so not having a textbook shouldn't be as big a hinderance as you might think

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #20 on: April 16, 2020, 11:37:29 am »
0
There are some parts of your answer which you could improve by using information you have already been given so I would suggest that you reread the advice you have already received and incorporate that before asking for further help & feedback.

In particular, you have been explicitly told what the sample is for this question and been given more sources of bias.


For stats the internet has a lot of resources and information so not having a textbook shouldn't be as big a hinderance as you might think
But for the above questions I've answered all parts of the question, and all I'm asking is for someone to verify if they're correct
Surely this is ok?


Guys I know I've asked this question before but I just want to make sure my answers are correct so I know i've understood it the correct way
For the following statement about a statistical survey, identify the following items (if possible). If you can’t tell,
then say so – this often happens when we hear or read about a survey.
A recent online poll conducted by an Australian newspaper of 1083 registered voters found
 that 43% of the respondents believe it is likely that bushfires are linked to climate change

a) The possible population of interest
Registered voters
b) The population parameter
proportion of registered voters who believe it's likely that bushfires are linked to climate change
c) The sampling frame
1083 registered voters
d) The sample
43% of 1083
e) The sampling method, including whether or not randomisation was employed
voluntary response survey, randomization is NOT used
f) Who (if anyone) was left out of the study
people who didn't vote and people who didn't participate in the survey
g) Any potential sources of bias you can detect and any problems you see in generalising to the population of
interest
people are more likely to participate in the survey if they have a strong opinion towards either side
Can you please tell me if my answers are correct? Be as harsh as you can
Thanks


Mod edit: Merged double post
« Last Edit: April 17, 2020, 11:19:27 am by Bri MT »

Bri MT

  • VIC MVP - 2018
  • Administrator
  • ATAR Notes Legend
  • *****
  • Posts: 4719
  • invest in wellbeing so it can invest in you
  • Respect: +3677
Re: 1st year stats question - sampling
« Reply #21 on: April 17, 2020, 11:24:36 am »
+3


I'm not trying to have a go or say you can't ask for help, what I'm saying is that you have missed where you have directly been told the answer to some of your questions. E.g. You have been directly told that the sample is the 1083 respondents.  (not a percentage of them)

For your benefit, and because given that people who are helping you are volunteering to do so, I recommend you reread the replies to your post so you see that and can fix your answers accordingly. It's better for both your sake and for people volunteering their time to help you that you do this before receiving further feedback.

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #22 on: April 17, 2020, 01:43:00 pm »
0
I'm not trying to have a go or say you can't ask for help, what I'm saying is that you have missed where you have directly been told the answer to some of your questions. E.g. You have been directly told that the sample is the 1083 respondents.  (not a percentage of them)

For your benefit, and because given that people who are helping you are volunteering to do so, I recommend you reread the replies to your post so you see that and can fix your answers accordingly. It's better for both your sake and for people volunteering their time to help you that you do this before receiving further feedback.
Hi Bri MT.
In one of the earlier posts, you said the sample is a subset of the sampling frame, whilst colline said the sample is the 1083 registered voters, which is why i'm confused

Bri MT

  • VIC MVP - 2018
  • Administrator
  • ATAR Notes Legend
  • *****
  • Posts: 4719
  • invest in wellbeing so it can invest in you
  • Respect: +3677
Re: 1st year stats question - sampling
« Reply #23 on: April 17, 2020, 02:30:34 pm »
+1
Hi Bri MT.
In one of the earlier posts, you said the sample is a subset of the sampling frame, whilst colline said the sample is the 1083 registered voters, which is why i'm confused


The sample is 100% definitely the 1083 registered voters.

In general, a sampling frame is where you get your samples from (hence the sample being a subset of the sampling frame). When I said this, I didn't mean to imply that it was impossible for the samples to cover the entire frame (this is the case in a census) but I can definitely understand why you might have read it that way.  For example, if the newspaper only contacted people who subscribe to their newspaper to take part in the survey and this is how they collected their data, then the sampling frame would be their subscribers list.

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #24 on: April 17, 2020, 03:23:17 pm »
0
The sample is 100% definitely the 1083 registered voters.

In general, a sampling frame is where you get your samples from (hence the sample being a subset of the sampling frame). When I said this, I didn't mean to imply that it was impossible for the samples to cover the entire frame (this is the case in a census) but I can definitely understand why you might have read it that way.  For example, if the newspaper only contacted people who subscribe to their newspaper to take part in the survey and this is how they collected their data, then the sampling frame would be their subscribers list.
A recent online poll conducted by an Australian newspaper of 1083 registered voters found
 that 43% of the respondents believe it is likely that bushfires are linked to climate change
tbh in this question I don't think a sampling frame is even mentioned?

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #25 on: April 18, 2020, 12:49:18 pm »
0
Guy I know i've asked this question quite a few times and I apologise for it, but can someone please tell me if my answers are correct?
A recent online poll conducted by an Australian newspaper of 1083 registered voters found
 that 43% of the respondents believe it is likely that bushfires are linked to climate change

The population of interest is registered voters
The population parameter is proportion of registered voters who believe it's likely that bushfires are linked to climate change
The sample is the 1083 registered voters
The sampling frame is not stated in the question
The sampling method is a voluntary response survey, randomization is not used
People who didn't participate in the survey were left out of the study
There is bias involved as people who have a strong opinion on either side are more likely to participate in the survey
Thanks

statistics

  • Adventurer
  • *
  • Posts: 21
  • Respect: 0
Re: 1st year stats question - sampling
« Reply #26 on: April 21, 2020, 01:03:28 pm »
0
Can someone just please tell my if my answers are correct or not?
I'm not able to find any resources online, which is why i'm having to post my questions on atarnotes

Be as harsh as you can
For the following statement about a statistical survey, identify the following items (if possible). If you can’t tell,
then say so – this often happens when we hear or read about a survey.
A recent online poll conducted by an Australian newspaper of 1083 registered voters found
that 43% of the respondents believe it is likely that bushfires are linked to climate change

a) The possible population of interest
Registered voters
b) The population parameter
Proportion of registered voters who believe it's likely that bushfires are linked to climate change
c) The sampling frame
Not stated in the question
d) The sample
1083 registered voters
e) The sampling method, including whether or not randomisation was employed
voluntary response sample, randomization is NOT used
f) Who (if anyone) was left out of the study
People who didn't participate in the study
g) Any potential sources of bias you can detect and any problems you see in generalising to the population of
interest
People are more likely to participate in the survey if they have a strong opinion towards either side. What source of bias would this be?