• Hi All

    Please note that at the Chandoo.org Forums there is Zero Tolerance to Spam

    Post Spam and you Will Be Deleted as a User

    Hui...

  • When starting a new post, to receive a quicker and more targeted answer, Please include a sample file in the initial post.

Statistical Analysis Project

T

Tom Nice

New Member
Hey Chandooers,

I'm pretty new to the scope of what excel can do and have been assigned with a project to analyse the movement of keyword positions in Google. Basically, I need to assess the difference in keyword positions between having google settings on 10 results per page or 100 results per page. The study is looking at the first 30 results (so 10 results per page x 3 against 30 out of 100 results on the one page) and see if there is any difference in keyword positions between the two settings.

I am having trouble finding a method to analyse the difference between the datasets i.e. which statistical test/s could be used. I have, so far, put the data 60k+ and 250k+ rows respectively into pivot tables that show the count of the keyword rankings one for a domain address for one date and ideally would like to be able to compare them.

Hopefully someone here might understand what I'm going on about and maybe have some tips as how to help effectively compare these two sets of data. Here is the link to the workbook that I am using if you would be so kind as to have a look at it and see what ideas come to mind. The data is scraped from google for online dating websites.

https://www.dropbox.com/s/2odx1yd529tsm5z/Chandoo - Data and Pivots.xlsx

Many thanks,

Tom
 
Hi Tom

Excel can only do what it is made to do. Unless you explain what you are after, it is difficult to help you. What sort of analysis you are after?
 
I'd like to be able to get data (in last 2 tabs of workbook) into the form where I can carry out analysis on whether unique keywords differ in position between data sets. I have tried to create pivot tables that show domains that are reached for keywords searched and the count of the ranking they receive. It would be great to analyse by what degree the unique keywords change in position depending on the setting. I was originally thinking a mann-whitney u test but just wanted to find a way to present the data to get to the stage of analysis. Hope this makes sense.

Sathish - Unfortunately the file is too large to upload here but if you have another suggestion for downloading it, I'd be happy to do that.

Thanks,

Tom
 
Hey again,

I've uploaded the file to wetransfer (as binary didn't reduce the size much) and here is the link.

http://we.tl/u6n38XhLRN

The workbook contains two data sets, as mentioned before, one that is for google search results of 10 per page x 3 and one that is 100 results in one page. There are tabs for pivot tables which show one search date and counts the number of instances a keyword has been clicked on to reach the listed domain name. I want to be able to compare these datasets by calculating any difference in the accuracy or volatility of where these keywords land the user. I welcome any suggestions you guys might have.

Thanks,

Tom
 
You must log in or register to reply here.
Back
Top