Well this subject line grabbed my attention as my dad worked his entire career for Nielsen Media Research -- the TV ratings company. He was led the sampling division which is essentially the whole process. If the sample isn't created correctly, the ratings are crap. I was very influenced by my father which led me to get a master's degree in statistics and work my entire career in market research.
I worked for them in the summers during college.
I could discuss this topic for hours but don't want to bore you all.
A few fun things though:
1/ Art Nielsen, who was the founder of the company, was a genius. And a VERY GOOD MAN on top of it.
2/ My dad had to travel to NYC a lot when I was a kid to meet w/the network executives. He would also complain that the networks would whine, complain, and scrutinize the ratings when their ratings went down ("Something is wrong!") . But when their ratings went up, he never heard a peep. I, too, work in market research and this same thing holds true with my own clients. None of my clients question the results of a study that says their product is fabulous.
3/ Do you all remember "sweeps week"? This is when the local affiliates would host super attention-grabbing programming. Like special reports on your local news. They did this to jack their local ratings. One clever local news affiliate decided to have a special report on "what it means to be a Nielsen family". Well, if you WERE a Nielsen family, this made you want to tune in b/c it was about YOU. Generated completely artificial high ratings. So Nielsen essentially voided that affiliate's ratings for that time period and the affiliate sued Nielsen. My dad had to testify in court explaining the statistical methodologies behind sampling. I don't remember the outcome.
Back in the old days there were both paper diaries and set-top electronic devices.
This is a story about ghosts. About how the technological networks of the twentieth century made us more distant, and yet more connected…
medium.com