First, I took statistics courses, so don’t start this with the arguing point that this is a sample size. I know that, but in statistics for your survey to actually carry real weight this is what would have to be done if we made a guess at the average amount of teens at 3.5 million per age from 13-19:
- Find the total number of teens in the US. According to census reports there are 25 to 28 million teens in the U.S. Piper Jaffray did a survey size of “6,100 teens with an average age of 16.0 years.” If the average amount of teens at 16 was at 3.5 million the number of teens surveyed needs to be considered. 6100 is what Piper Jaffray did and every news outlet and analyst retweeted the study without doing any homework.
- Once you have the total number of teens you are analyzing then you have to consider your confidence level. Since we are looking at 25 million teens and averaging the number of 16 year olds at 3.5 million, I’d like a 99% confidence level.
- You would then need a margin of error. Let’s say that’s at 1.
Basically to have a more accurate reflection in a survey, Piper Jaffray would need to survey 16,563 teens at 3.5 million. The fewer people surveyed the less accurate the survey.
Population Size = N | Margin of error = e | z-score = z
e is percentage, put into decimal form (for example, 3% = 0.03). (via surveymonkey)
I have to admit simply looking at the numbers 6100 is a small sample size. I mean there are high schools in San Diego, where I used to teach, in the Division 1 classification that have 2000 – 3000 high school students. This means that Piper Jaffray basically asked two high schools out of 100 schools in San Diego what they thought was cool.
This micro to macro comparison leaves a lot to be considered when every newspaper in the country is utilizing this information to state that “teens no longer find Nike cool”. In a clickbait society if you want to get people to visit your site, then you simply need to create a survey and then release it to the public. It seems that no one considers the numbers involved in a survey. I’ve seen some data analyst tell people they shouldn’t argue the numbers they present. No data company is infallible. All data should be checked.
Nike is still cool.
If I were to consider the most popular social media outlets among teens it would be Instagram. A quick look at brand pages on IG and the numbers speak to the cool factor of the brand.
Nike 74.7 million followers
adidas 16.8 million followers
Under Armour 3.8 million
If you look at the last post placed on IG for all brands the interaction looks like this (each post is a video):
Nike 573K plays 330 comments
adidas 153K plays 168 comments
52.1K plays 67 comments
This in no way disqualifies the Piper Jaffray survey. It does place into question why so many online sites, analysts and reporters utilize the information so readily without taking at least a beat to consider the accuracy. In the words of Jay-Z, when it comes to teens no longer finding Nike cool, “I don’t believe you. You need more people.”