Featured Tesla to expand charging station program with Hilton hotels to up to 200 chargers

We all have our biases, conscious and unconscious.

The question is can we put them aside and let the facts drive our expressed opinions. I think that there are some who can. I think CR does its best to do so. But I have always been suspect of the self selecting bias of its readership and their ability to respond to a survey in a meaningful way.

I have owned 50 cars, 10 brands, 5 types. You have bought one, rented a few. Should our opinions have the same weight?

You bought the top of the line, loaded. Me the bottom of the line stripped. Same weight?

You use and IPhone. I a Droid. You use the connectivity features. I don't. Same weight?

You read the manual, and again. I didn't. Same weight?

You live in Topeka, I in LA. Same weight on HVAC performance?

I have a great dealer and a wonderful relationship. Yours sucks. Same weight? How does this affect our opinions of the cars?

I live 1 mile from the dealer. You live 50 miles? How does this affect our opinions of the cars?

You bought from your local dealer, I bought 4 states away? How does this affect our opinions of the cars?

I bought in the first month of availability, one recall. You in the 12th, none. How does this affect our opinions of the cars?

My dealer checked the alignment before it was delivered to me. Yours didn't and your tires wore out prematurely. Same weight?

You got brand M tires. I got brand H. Same weight?

I really hate the car but want its resale value high as I'm trying to sell it. Do I report honestly?

You are pleased with your 29 MPG compared to your previous cars. I'm only getting 35 and am disappointed. Same weight?

What I'm trying to say here is there are many permutations and combinations of what influences the broad perceptions of the brands. Many definitions of what constitutes quality.

 

The idea of a statistically valid survey is that you have enough replies so that the extremes you correctly point out tend to blend in to the averages...somewhat equally for all the car makers and models.

There is an effect where people with a lot to complain about (valid complaints or not) will be more likely to respond to surveys.
The same goes for enthusiasts for a new car or technology. Good surveying techniques will minimize this effect.

Because the internet makes it very easy to do a survey there are lots of poor surveys.

Mike

 

Errors in a survey is very different than the purpouslfull manipulation you speak of.
The surveys are far better than your personal, impression. Especially when you make accusations with zero evidence of wrong doing.
It is no better than screaming “fake news” when you simply don’t like what the news has to say.

 

almost sounds like and admission that your marketing claims were highly spurious.

but yea, marketing @1 time - me too ... the thing is - it's most likely there would be no issue at all, if it was OUR top 10 picks - being identical as CR's respondents. It's a rare bird who will THEN still claim a survey is suspect.
So funny how that works.

.

 

So what is the quantity that makes the outliers meaningless and the norm meaningful? When you are surveying maybe 30 brands with a total of 100 models and you have N responses. You have cars in the same brand for $16k and another for $60k. They have to have widely deviating expectations from their buyers.

 

We all know about the tobacco companies, McDonald's and 60 Minutes.

 

Surveys of preference are not research and do not measure quality. To me, all surveys are suspect because I know how easy it is manipulate findings and/or how easy it is to make mistakes. BTW, I brush with baking soda.

 

Walk into the store and compare the active ingredients in brand A's several price points toothpastes. Startlingly the same. Then go to Brand B's offerings.

 

I've taken enough statistics to say that if I was a professional in that field I'd be able to tell you.

Standard deviation - Wikipedia

Mike