Call for Proposals: NBER-Nielsen Media Projects
May 18, 2017
Nielsen and the National Bureau of Economic Research (NBER) have entered into an agreement to make Nielsen’s media-related data available for academic research. On the NBER side, the initiative is led by Andrew Sweeting (University of Maryland and NBER), Matthew Gentzkow (Stanford University and NBER) and Jesse Shapiro (Brown University and NBER). This initiative has received generous financial support from the Sloan Foundation, and the NBER hopes that this agreement may become a model for how proprietary datasets can be made available to academic researchers.|
Under the agreement, the NBER will solicit and select research proposals that will make use of Nielsen's data on media. The amount of data that can be supplied will depend on the number of Nielsen analyst hours required to extract the data. The Sloan Foundation's funding will support data extraction costs in the first year of this initiative. The data can only be used for academic research, and both working and published papers will be subject to review by Nielsen to make sure that commercially-sensitive data are not disclosed and that their data is being described appropriately. The supplied data will be stored and used on the NBER's computing infrastructure.
The Nielsen data available cover the media usage and advertising that Nielsen monitors and the consumer surveys that it conducts in the United States. Examples of the types of data available include:
For more information on the data, researchers should review the following materials prepared by Nielsen.
- Nielsen's basic TV ratings data (e.g., network, program, daypart, episode ratings, by age, gender and DMA), from 1991 to the present
- Nielsen's advanced TV ratings data, which include minute-by-minute ratings data for programs and commercials; these ratings can be broken down by both demographics and other household information, such as whether the household subscribes to services such as Hulu or Netflix, and they can be combined to look at the extent to which households watch combinations of programming. Much of this data will be available back to 2002
- topline data on website traffic back to 2011, with data available by demographic group for the last 24 months
- data on browsing and app usage on mobile devices (smartphones and tablets)
- national and local radio ratings, together with information on station formats, back to 2010 (this data would be similar to Arbitron data from earlier years)
- Nielsen's Advertising Intelligence data, tracking brand-level advertising across 22 media types, including dollar expenditures and estimates of exposures, back to 1993
- information from a range of Nielsen's syndicated surveys, including Scarborough that track consumer behavior. Requests for adding additional questions to these surveys will also be considered.
Nielsen's scanner/point of sale and Homescan datasets are not included in this agreement, and respondent/household-level media data will also not be available.
We welcome proposals on a wide range of topics, including the way people consume media in the digital age; the effects of media on individual behavior, social behavior and political attitudes; the industrial organization of media markets; and, the effectiveness of advertising and its role in the broader economy.
In order to be considered for access to the data individual researchers or research teams should submit written proposals of no more than four single-spaced pages, including references, tables and figures. Proposals should be submitted via
http://papers.nber.org/confsubmit/backend/cfp?id=NMPs17 by June 28, 2017. These proposals should
The NBER organizers will review these proposals and work with Nielsen to identify a set of proposals for which it will be possible to supply data. There may be some negotiation with teams about their exact data requirements so that the NBER can support as many projects as possible. Successful proposals will receive the full documentation that Nielsen provides to its commercial clients, and some on-going support from Nielsen, to answer questions and correct any problems with the data supplied, will be available.
- identify research questions and explain why they are novel, interesting and important;
- describe the Nielsen media data that will be required and any other data that will be used in conjunction with the Nielsen data;
- list any financial resources (such as grant or university support) that the researchers may be able to provide to support their purchase of data; this support may be important where large amounts of data, or data from multiple Nielsen databases, is being requested;
- list the set of researchers involved in the proposal; all research teams must include at least one NBER affiliated researcher.
Please contact Andrew Sweeting at firstname.lastname@example.org or Alterra Milone at email@example.com with any questions.