Meta S. Brown

When Less (Data) Is More (Information)

NO RATINGS
1 saves
View Comments: Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
louisw900
User Rank
Blogger
(Super) Sample Size
louisw900   8/13/2013 10:31:48 PM
NO RATINGS
Thanks Meta for the great tips for quieting the "big talker" when it comes  issues of sample size by asking how we account for it.  Anyone who can explain this in a cogent way has my vote for the position ! ; ) 

One of the most fundamental issues that a analyst must honestly confront IMO.

chris.had
User Rank
Prospector
Re: Truly Random?
chris.had   8/10/2013 11:34:16 AM
NO RATINGS
@Meta: Indeed, training is an investment which should be done to get the maximum out of the people you have hired and identified as the future leaders of that particular organization. You need people who are willing to learn things from the scratch if its needed. Not ones who think they can do anything and does not participate in any training session at all.

Meta S. Brown
User Rank
Blogger
Re: Truly Random?
Meta S. Brown   8/9/2013 3:16:42 PM
NO RATINGS
First, you have to care!

If you can get people who have relevant training, great, but rather than looking for perfect people, I suggest hiring good, consicentious people and investing in training to fill gaps.

SethBreedlove
User Rank
Data Doctor
Re: Truly Random?
SethBreedlove   8/9/2013 1:05:56 AM
NO RATINGS
One of my recent favorites was this helmet that used lasers to help men gain their hair back, that just got FDA approval.  The effectivness and saftey study was done on 5 men, yes that is FIVE, count on one hand - 1, 2, 3 4 just 5.  For a $675 dollar price tag, I hope they will be able to afford a bigger study.

True, you don't have to test everyone, but you do need a group large enough to have a decent confidence level.   

 

Alexis
User Rank
Data Doctor
Re: Truly Random?
Alexis   8/8/2013 12:06:31 PM
NO RATINGS
Meta, from your experience what are the most effective strategies for ensuring accurate sampling? Just hiring the right people?

Meta S. Brown
User Rank
Blogger
Re: Truly Random?
Meta S. Brown   8/7/2013 2:14:23 PM
NO RATINGS
That's true.


Still, we need to make a reasonable effort at selecting appropriate samples for our analyses, or the results will be of no value.

CandidoNick
User Rank
Data Doctor
Truly Random?
CandidoNick   8/7/2013 2:01:38 PM
NO RATINGS
A true random sampling is a rare occurance. If you decide to set up shop at a grocery store, and record how long it takes men vs women to collect their items and purchase them, you are still being selective by not moving to another store across the country. The people in the first store are all fro the area, and thus making the data you collect not random. A true random sampling would be a nigh impossible feat.

Meta S. Brown
User Rank
Blogger
Re: Majority
Meta S. Brown   8/7/2013 1:55:03 PM
NO RATINGS
Yet large size, in itself, is rarely an important feature of a sample. We may need large samples to detect subtle efffects, for example, although subtle effects are sometimes mere curiousities as far as practical applications go. We certainly need large samples when we have many subgroups to examine - or to put it another way, we need many little samples that are collected in a coordinated, efficient manner.

Misrepresentation certainly occurs online. But I don't hold it against respondents. If we put more thought into how we recruit respondents and what we demand of them, we could improve the quality of our data.

Phoenix
User Rank
Data Doctor
Re: Majority
Phoenix   8/7/2013 12:39:39 PM
NO RATINGS
Big data gives us the opportunity to collect larger samples than ever before. However, with online hacking and misrepresentations I wonder whether the data we collect is authentic.

Meta S. Brown
User Rank
Blogger
Re: Majority
Meta S. Brown   8/7/2013 12:12:03 PM
NO RATINGS
Oh, they took a sample. They probably did not do it well, but they did it.

Page 1 / 2   >   >>
Information Resources
More Blogs from Meta S. Brown
Today's data analyst needs to understand the factors that have an impact on their research, and the best way to gain that understanding is the good old library.
Data alone won't make an analyst's work memorable or actionable in the eyes of a business executive. A story puts it into perspective.
When it comes to acquiring the data that will feed your analytics initiative, "free" isn't always the best approach.
Data presents an opportunity for enterprises to utilize personalization. Do it right and you can win. Do it wrong and you turn off those customers.
Pundits and analysts feel the need to expand the original designation of volume, velocity, and variety.
Radio Show
A2 Conversations
UPCOMING
Jessica Davis
Data Analysts in Training: Meeting Tomorrow's Demand


3/8/2017   REGISTER   0
ARCHIVE
Jessica Davis
Our Bodies, Our Data: Medical Records For Sale


2/21/2017  LISTEN   62
ARCHIVE
Jessica Davis
Energy Analytics: Using Data to Find Savings


2/14/2017  LISTEN   44
ARCHIVE
Jessica Davis
Sharpen Your Analytics & Data Management Strategy


2/8/2017  LISTEN   74
ARCHIVE
Jessica Davis
Analytics: Make the Most of Data's Potential in 2017


1/19/2017  LISTEN   19
ARCHIVE
Jessica Davis
A2 Radio: Can You Trust Your Data?


12/20/2016  LISTEN   70
ARCHIVE
James M. Connolly
Retail Analytics: See Where Style Meets Statistics


12/6/2016  LISTEN   53
ARCHIVE
James M. Connolly
Why the IoT Matters to Your Business


11/29/2016  LISTEN   45
ARCHIVE
James M. Connolly
Will Data and Humans Become Friends in 2017?


11/22/2016  LISTEN   40
ARCHIVE
James M. Connolly
We Can Build Smarter Cities


10/20/2016  LISTEN   31
ARCHIVE
James M. Connolly
Visualization: Let Your Data Speak


10/13/2016  LISTEN   70
Information Resources
Quick Poll
Quick Poll
About Us  |  Contact Us  |  Help  |  Register  |  Twitter  |  Facebook  |  RSS