June 21, 2013

The challenges of conducting surveys of youth

FT_13.06.21_Troublepollingteens_640x300

Why don’t you do more research on youth?

I get this question all the time because I do a lot of research with online teenagers. The short answer is: because surveying children under 18 is expensive and complicated. But for those interested in the slightly longer answer, keep reading.

Research with minors presents some unique challenges. When we survey young people ages 12 to 17, we’re surveying minors, who legally fall into a protected class of people. By law, minors cannot consent; their parents must give consent for them (in this case to participate in the research project). The consent requirement is a legacy of the belief that people under the age of 16, 17 or 18 (to complicate matters the age of consent varies by state) are not fully equipped to make good decisions in their best interest and that youth may be unduly susceptible to coercion.

So, all this means we must obtain parental consent to interview minors younger than 16, and because of state differences, we typically seek parental consent for all youth under the age of 18. This means that we must speak or interact with two individuals in each household (in a specific order with the parent first) rather than one, as in traditional surveys of adults. Interviewing two people increases the complexity of the project, and requires more phone calls or messages to reach eligible respondents in the proper order.

Further, families with children between the ages of 12 to 17 make up about 14% of U.S. households. This means that in a random digit dial survey, only one in every seven households will be eligible for the survey.

Also, the increasing use of the cell phones over the past decade means that in order to reliably reach a representative sample of teens and families we must conduct research on mobile phones. While it is now possible to survey a representative sample of cell phone users, the cell phone presents its own challenges, for youth research in particular.

Landline phones are attached to households; so as researchers we can call the same landline number to reach two different people (parent and child) and have a reasonable expectation that both might be available to answer the phone.

In contrast, cell phones are typically attached to individuals (even though they are sometimes shared within a household). With a mobile phone, we’re calling the individual parent, who must then either give us another phone number (often a cell number) for the teen, or must be physically in the same space as the child to hand the phone over to them. Completing two interviews with two individuals where the parent is reached on their cell phone often takes more time and effort and requires additional calls back to try to reach the teen, thereby increasing the already substantial cost of telephone surveys with youth. Even in online survey administrations (where respondents take the survey on a website), making the jump from the parent consent to youth completion of the survey can prove challenging.

All of these complications make surveying children and families very expensive. While a typical telephone survey with adults costs between $50,000 and $140,000 depending on survey and sampling variables, a single telephone-based survey of teens and parents costs between $150,000 and $300,000.

Despite the challenges in surveying teens and their parents, we’ve surveyed teens and parents eight times since 2000, and are looking forward to fielding our ninth survey in early 2014.

Resources on ethical research with children:

http://www.naeyc.org/resources/research/ethical

http://www.srcd.org/about-us/ethical-standards-research

 

Topics: Research Methodology

  1. Photo of Amanda Lenhart

    is Director of Teens and Technology Research at the Pew Research Center’s Internet & American Life Project.

Leave a Comment

Or

All comments must follow the Pew Research comment policy and will be moderated before posting.

1 Comment

  1. Michael E Jones10 months ago

    It seems that you are striving hard to preserve the rights of the “protected” class “minors” which is good.

    I now ponder:
    Hmmmmm…….can’t have age requirement for morning after pill…..hmm, no danger there,………survey must be dangerous….. class-rooms are getting to point of obtaining physiological data from students real time…..hmmm.

    The current conversations in the NSA scandal, may be the perfect time to review data collection in this group.

    Reply