Statistics Homework Question: Do specific postcodes experience more burglaries?

M

mintcake32

New member
Joined
Jan 3, 2021
Messages
3
Hi,
I have a homework question that I am struggling to answer and would be grateful if the forum can help. I am a statistics newbie so please be gentle! Do specific postcodes experience more burglaries?

The question asks us to visualise if there is any correlation between postcode / zip and the number of burglaries per household. I have the data which contains a list of house address and postcode and a list of houses address and burglaries to date. I have a subset of data so only 500 postcodes. We do not need to worry about population density etc.

I am guessing I need to show some sort of central tendency and certain postcodes that are outside x percentile? We can only use standard charts, e.g. bar, scatter etc.

Thanks,
Paul
 
mintcake32 said:
Hi,
I have a homework question that I am struggling to answer and would be grateful if the forum can help. I am a statistics newbie so please be gentle! Do specific postcodes experience more burglaries?

The question asks us to visualise if there is any correlation between postcode / zip and the number of burglaries per household. I have the data which contains a list of house address and postcode and a list of houses address and burglaries to date. I have a subset of data so only 500 postcodes. We do not need to worry about population density etc.

I am guessing I need to show some sort of central tendency and certain postcodes that are outside x percentile? We can only use standard charts, e.g. bar, scatter etc.

Thanks,
Paul
Click to expand...
Please show us what you have tried and exactly where you are stuck.

Please follow the rules of posting in this forum, as enunciated at:


Please share your work/thoughts about this problem.
 
Thank you for your reply.

I am stuck as postcode is categorical data so I am not sure how to correlate this with number of burglaries.

I initially tried plotting the total number of houses in each postcode on the x-axis and the mean burglaries per postcode on the y-axis and showing a linear regression together with pearsonr but I don’t think this answers the question.

Thanks,
Paul
 
mintcake32 said:
I am stuck as postcode is categorical data so I am not sure how to correlate this with number of burglaries.

I initially tried plotting the total number of houses in each postcode on the x-axis and the mean burglaries per postcode on the y-axis and showing a linear regression together with pearsonr but I don’t think this answers the question.
Click to expand...
I haven't attempted to answer because I can't tell what you have learned that could be used, and therefore what might be expected of you. From the OP, I wouldn't have guessed that you have any linear regression tools available. I think this is an open-ended question, perhaps without any one specific correct answer; but that depends on my assumption that you haven't been taught anything specific enough.

Certainly you are right that the "correlation" intended in the problem is not the r you might get from a linear regression. That, together with the phrase "visualise if there is any correlation", is why my first impression was that "correlation" is not being used as a technical term, and therefore a formal answer may not be expected.

Can you list the topics that have been covered?
 
Thanks Dr. Peterson.

The topics we have covered are correlation and covariant which is why I am confused.

I have given this more thought and think I could show the coviance between postcodes on the x and y as a heat map matrix? The thing that worries me is there are lots of labels for postcodes so might need to group postcodes, so SW19 5AE grouped into SW19?

Does this make sense?

Thanks,
Paul
 
You must log in or register to reply here.
Top