Broadband.gov
Federal Communications Commission



Denying Bill Shock by Distorting the Facts

July 15th, 2010 by Admin User

CTIA CLAIM #1: The FCC interviewed teenagers. According to the CTIA, “only 30% of the respondents [in the FCC study] said they were over 18.”

FACT: All the respondents in our study were over 18.

Let’s begin with how CTIA describes the survey sample:

“The FCC commissioned a survey in mid-April to ask a range of questions -- 3,005 people participated. From what we can tell, only 902 of the respondents, when asked directly if they were 18 years or older, said “yes.” Let me repeat -- only 30% of the respondents said that they were over 18. Despite the survey instructions directing the termination of the interview and recording as ineligible if the respondent was under 18, it appears the survey went on to ask questions and record answers about wireless bills. How many people under the age of 18 actually have ever seen a wireless bill? But wait, there’s more…”

These statements are incorrect. No one under age 18 was interviewed for the survey. A close look at the data will show why. The statement that “only 902 respondents” said they were over the age of 18 comes from the response to question labeled “S3” in the raw data file; all 902 of those who received this question stated they were over 18, with system missing coded for 2,103 other respondents in the total sample that were not asked the question. What of those not asked the question? They turn up in a preceding question (S1), which asked “May I please speak to the person age 18 or older in the household who had the most recent birthday?” Now, it seems like survey interviewers first asked people (in S1) if they were over 18 and then, very shortly thereafter at S3, asked again if they were over the age of 18.1  It would be odd (and probably annoying to respondents) if the survey had unfolded that way. Did it?

The answer is “no.” The survey design did not set out to annoy respondents and it did not fail to capture adults in the sample. In fact, the questions were addressed to two different sampling frames. This is survey lingo for saying that some set of respondents were given question S1, another set question S3. Those getting S1 were people who responded to the survey on landline phones at home; those getting S3 responded on their cell phones (perhaps at home, perhaps not). As everyone in survey business knows, given that so many people have given up their landline phones to go “cell only,” it only makes sense to include cell phone numbers in a national random digit dial telephone survey. This approach yields a reliable sample – and of course, one wants to be sure that a cell phone respondent willing to take a survey is over 18. This is not only good practice, but it also prevents survey firms from hassling under age customers of … the companies CTIA represents.

All this is by way of saying that the FCC’s survey of 3,005 respondents contained 902 people who were reached on their cell phones and 2,103 who were reached on landlines at home. The survey methodology ensured that respondents were 18 years of age or older. The survey also asked respondents to tell us how old they were (standard practice in surveys). All respondents – except for 3% who refused to give an age – said they were age 18 or older.

Granted, to the untrained eye, these technical aspects of inspecting the raw data might invite confusion. One response to seeing such apparently anomalous results (that is, a survey of adults in which only 30% of respondents seemed willing to own up to being of voting age) would be to go to the source to try to resolve the confusion. That is not what CTIA did. No one from CTIA contacted anyone at the FCC about this.

CTIA CLAIM #2: The FCC study shows only 13 percent of Americans have experienced these unexpected increases. CTIA bases this on responses from 391 out of 3005 people interviewed.

FACT: The true number is 17 percent. CTIA based its calculation on raw numbers from the survey. Standard practice for all national surveys is to weight the responses to match a representative sample of the population. The appropriate weighted number is 17 percent of the U.S. population.

Again, let’s start with the CTIA blog post:

Question 52 of the survey asked respondents if their “cell phone bill ever increased suddenly, from one month to the next, even if [they] did not change the calling or texting plan on [their] phone.” 

Of the 3,005 people interviewed, and knowing that a significant percentage were likely under the age of 18:

  • 1862 responded “no” 
  • 8 refused to answer
  • 202 didn’t know
  • 542 were not valid 
  • 391 said “yes” 


CTIA here is reporting the raw, unweighted numbers from the dataset. For Q52, as is clear in the questionnaire, this question was asked of those respondents who have a personal cell phone. Those respondents who have no personal cell phone (a total of 542 respondents) were not asked the question and are listed in the dataset as System Missing. The statement above also repeats the erroneous claim that some respondents were under the age of 18.

Using unweighted numbers to report results from this question is not a good practice and can be misleading. Weighting raw data is a standard practice in the survey industry, and it corrects for known systematic biases that result from telephone interviewing. In most situations, the survey results should be examined using the weighted percentages. All the FCC reports on the survey have correctly used the weighted results. Here are the correct, weighted results for Q52, as reported by the FCC:

Q52.    Has your cell phone bill ever increased suddenly, from one month to the next, even if you did not change the calling or texting plan for your phone?

Based on those who have a personal cell phone (n=2463)
    %
17    Yes
75    No
7    (DO NOT READ) Don’t know
*    (DO NOT READ) Refused

1 Question S2 asked landline respondents who were not over the age of 18 if the survey interviewer could call back another time to attempt to administer the survey to an adult in the household.



Capture The Phone Numbers Using Your Camera Phone

If you have a camera and a 2D matrix code reader on your mobile phone, you can capture the FCC Phone numbers right to your phone by following these three easy steps:
Step 1: Take a photograph of one of the codes below using the camera on your mobile phone.
Step 2: Use your phone's Datamatrix or QR Code reader to decode the information on the photograph. Please note, these code readers are device specific and are available to download on the internet.
Step 3: Store the decoded address information to your phone's address book and use it with your Maps or GPS application.

Datamatrix and QR FCC Phones