Skip to main content

Exploring the Pew Research Center data-set on EReaders and Gadgets

Pew Research Center hosts variety of open datasets spanning across US Politics, Media & News, Social Trends, Internet & Tech, Global, Religion. The data-set that got my attention is survey of about 1007 respondents on the use of  electronic gadgets for E-reading.



Distribution of number of books read in the past 12 months (broken by gender for each bin)



Mean (including those who answered “none”) – 12 books
Median – 5 books

Based on those who read any books in the past 12 months N=780

a. Printed books – 91%
b. Audiobooks – 19%
c. E-books – 37%

Now for those who answered Yes for an Ebook, lets start to see how many books they tend to read on average based on electronic gadgets that they own. The left most segment which only owns an EReader like Kindle/Nook is the most voracious of all. The group that stands out lowest is coincidentally the exact inverse  ie  they own everything but the Ebook reader. 

The average number of books shown for each category can also include paper books and audio books. The survey does not provide a breakup by the device type or mode (paper,audio,ebook). 




For each of the device categories Yes/No question, some participants left NULL's or refused to answer the question. I have omitted them from the above analysis.

Lets now look at the max level of education attained and how that affects number of books read in past 12 months.  

In terms of total number of books read, here are top 3

1 ) High school grads 
2) Some college - No Degree
3) Postgrad or Professional degree


For average number of books in last 12 months, here are top 3

1) Some Postgrad or Professional schooling
2) Postgrad or Professional degree
3) Some College- No Degree

































The next thing we look at his how Gender and Race play a part in the amount of time spent reading.

We will look at both average number of books and median of books read in last 12 months

 Averages :  White females top the list with an average  of 19 books in last 12 months. That's a lot of book reading and overall numbers seem to be pretty high.








Median:   The leader spot changes from above and is now shared by  Male spread across 2 different categories   Asian / Pacific Highlander and Mixed



The survey also included  'Other' race with a very small composition and has been included from the Race/Gender analysis. Here is a breakdown of Other category

Race                                        Number of Records
Hispanic                                     13
Latino                                          3
Mexican                                       2
Mexican American                      2
Puerto Rican                                2


Part 2 of the analysis to follow soon..

Ciao
                                         


Comments

Popular posts from this blog

Deep Dreams with Keras & Tensorflow

Made some modifications to the DeepDreamcode courtesy François Chollet and added few extra layers, changed the loss function settings a bit and viola !!!

Used the pre-trained model based on the VGG16 network architecture.

I will post the github link shortly. 



Iteration 1


                                                                            Iteration 2


                                                                          Iteration 3


Iteration 4

Iteration 5


Random !