Skip Navigation

Federal Communications Commission

English Display Options

Commission Document

Media Ownership Study 3-Submitted Study

Download Options

Released: June 15, 2011
Report to the FCC:
How the Ownership Structure of Media Markets affects
Civic Engagement and Political Knowledge, 2006-2008
Lynn Vavreck,1 Simon Jackman,2 and Jeffrey B. Lewis3
Saturday 23rd April, 20114
1Associate Professor, Department of Political Science, University of California, Los Angeles.
e-mail: lvavreck@ucla.edu
2Professor, Department of Political Science, Stanford University. e-mail: jackman@stanford.edu
3Associate Professor, Department of Political Science, University of California, Los Angeles.
email: jblewis@ucla.edu
4DMA is a registered trademark of, and DMA region boundaries and names are the proprietary
information of The Nielsen Company, used under license.

Contents
I
Introduction
1
1
Data
2
1.1
The Cooperative Election Studies . . . . . . . . . . . . . . . . . . . . . . . .
2
1.2
Validated Turnout Data from U.S. Secretaries of States . . . . . . . . . . . .
10
1.3
Nielsen Gross Ratings Points Data for Political Advertising . . . . . . . . . .
13
2
Civic Engagement and Political Information
14
II
Analyses: Political Interest, Knowledge, Uncertainty, and
Participation (Project A)
17
3
Analyses of Variation
17
3.1
Civic Engagement: Dependent Variables . . . . . . . . . . . . . . . . . . . .
18
3.2
DMA region Fixed Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . .
21
4
Bayesian Analysis of Hierarchical and Multilevel Models
36
4.1
Interest in Politics
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
39
4.1.1
Multilevel model . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
40
4.2
Overall Political Knowledge/Information . . . . . . . . . . . . . . . . . . . .
43
4.2.1
Multilevel model . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
43
4.3
Ability/Willingness to Place Obama
. . . . . . . . . . . . . . . . . . . . . .
46
4.3.1
Multi-Level Model
. . . . . . . . . . . . . . . . . . . . . . . . . . . .
47
4.4
Validated Voter Turnout . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
4.4.1
Multilevel Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
4.5
Conclusion: Project A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
51
III
Analyses: Identification of Local Candidates -- A Study
Representative of DMA regions (Project B)
53
5
The Pattern across DMA regions and its Explanations
54
5.1
How many people can recognize the candidates? . . . . . . . . . . . . . . . .
55
5.2
Is There a Pattern to the Variation at the DMA region Level? . . . . . . . .
60
5.3
Individual-Level Attributes . . . . . . . . . . . . . . . . . . . . . . . . . . . .
62
5.4
Explaining the DMA region-level Fixed Effects . . . . . . . . . . . . . . . . .
66
IV
Report Conclusion
78
References
79

List of Figures
1
Map highlighting the nine Midwest designated market areas (DMA regions)
analyzed in this study. From the Northwest to the Southeast, these DMA
regions are: Minneapolis--St. Paul (MN), Madison (WI), Milwaukee (WI),
Chicago (IL), Champaign&SprngfldDecatur (IL), Lansing (MI), Detroit (MI),
Cleveland-Akron (Canton) (OH), and Columbus, OH (OH). The black bound-
aries on the maps outline Congressional districts in the 109th Congress. Black
dots highlight Congressional districts that intersect the 9 studied DMA regions.

3
2
Both maps show the representation ratio of the CCAP sample within each of
the 210 Designated Market Areas (DMA regions) in 2007. Reds reflects DMA
regions that are overrepresented in the CCAP sample. Blues reflect DMA
regions that are underrepresented in the CCAP sample. Darker colors reflect
larger degrees of over- or under-representation. Panel (b) is a cartogram that
deforms the US map such that each DMA region's area is proportional to the
size of its adult (18+) population. Alaska's two DMA region's are shown to
the Southwest of Texas. The Hawaii DMA region is shown the Southeast of
Texas.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
3
Both maps show the weighted representation ratio of the CCAP sample within
each of the 211 Designated Market Areas (DMA regions) in 2007. In these plot
the CCAP data are weighted to correct for the oversample of battle ground
states and for other incidental demographic imbalances. The color scale is the
same as in Figure 2.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
9
4
Histograms of representation ratios for the CCAP sample across DMA regions.
In Panel (a) the ratios are based on the unweighted CCAP observations. In
Panel (b) the ratios are based on weighted CCAP sample. The weights adjust
for CCAP's battleground-state oversample and for demographic imbalances.

10
5
Registration Status of CCAP respondents in voter files, by state . . . . . . .
13
6
Scatterplot Matrix of Fixed Effects . . . . . . . . . . . . . . . . . . . . . . .
23
7
Number of Independently Owned TV Stations by Fixed Effects
. . . . . . .
26
8
Number of Independent Radio Stations by Fixed Effects
. . . . . . . . . . .
27
9
Number of Multiple-TV Station Parents by Fixed Effects . . . . . . . . . . .
29
10
Number of Parents Owning at least one Television and one Radio Station by
Fixed Effects

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
30
11
Number of Parents Owning TV and Radio Stations and a Newspaper by Fixed
Effects
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
31
12
Fraction of Households that Subscribe to 200 KBS Internet Service by Fixed
Effects
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
32
13
Log of Total Presidential Campaign Advertising GRPs by Fixed Effects, 2008
33
14
Number of Radio Parents with News/Talk Format by Fixed Effects . . . . .
34
15
Log of Population over 18 by Fixed Effects . . . . . . . . . . . . . . . . . . .
35
16
Scatterplot Matrix of Level Two Variables . . . . . . . . . . . . . . . . . . .
37
1

17
Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level ordinal logistic regression model for respondent self-reported levels of
interest in politics. The estimates have been sorted from low to high and by
construction have zero mean.
. . . . . . . . . . . . . . . . . . . . . . . . . . .
41
18
Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level regression model for political information scores. The estimates have
been sorted from low to high and by construction have zero mean.
. . . . . .
45
19
Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level logistic regression model for respondent ability/willingness to place Obama
on health policy. The estimates have been sorted from low to high and by
construction have zero mean.
. . . . . . . . . . . . . . . . . . . . . . . . . . .
49
20
Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level logistic regression model for validated voter turnout. The estimates have
been sorted from low to high and by construction have zero mean.
. . . . . .
52
21
Differences in Ability to Identify Incumbents by DMA regions . . . . . . . .
58
22
Ability to Identify Candidates by Congressional District
. . . . . . . . . . .
59
23
Scatterplot Matrix of DMA region Fixed Effects from Challenger Identification
Model and Ownership Variables
. . . . . . . . . . . . . . . . . . . . . . . . .
68
24
DMA region-Level Fixed Effects and Number of Independent TV Voices . . .
69
25
DMA region-Level Fixed Effects and Number of Independent Radio Voices .
70
26
DMA region-Level Fixed Effects and Number of Multi-Ownership Parents . .
71
27
DMA region-Level Fixed Effects and Number of Cross-Ownership Parents . .
72
28
DMA region-Level Fixed Effects and Gross Ratings Points . . . . . . . . . .
74
29
Explaining DMA region-Level Fixed Effects with Incumbents' Ad Buys . . .
75
30
Explaining DMA region-Level Fixed Effects with Challengers' Ad Buys . . .
76
List of Tables
1
Number of CCES respondents by DMA region . . . . . . . . . . . . . . . . .
5
2
Zip code matches, CCAP and Vote Validation File from Third-Party Firm. .
11
3
Least squares regression analysis of Political Information (mean 0, sd 1) and
Political Interest (1, 2, 3), October wave. All models include unreported fixed
effects for income levels.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
22
4
Least squares regression analysis of Ability/willingness to locate Obama on
health care issue, September wave of CCAP (0,1) and Validated Turnout (0,1).
All models include unreported fixed effects for income levels.
. . . . . . . . .
24
5
Estimates of Multilevel Ordinal Logistics Regression Model of Respondent
Self-Reported Levels of Political Interest. Entries above the line are estimates
of micro-level parameters, ; entries below the line are estimates of DMA-level
parameters, . are threshold parameters in the ordinal logistic regression
model. is the standard deviation of the error component at the DMA level
of the model.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
42
2

6
Estimates of Multilevel Regression Model of Political Information Scores En-
tries above the line are estimates of micro-level parameters, ; entries below
the line are estimates of DMA-level parameters, . is the standard devia-
tion of the error component at the micro, individual level of the model; is
the standard deviation of the error component at the the DMA level of the
model.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
44
7
Estimates of Multilevel Model of Ability/Willingness to Place Obama on
Health Care. Entries above the line are estimates of micro-level parameters,
; entries below the line are estimates of DMA-level parameters, .
. . . . .
48
8
Estimates of Multilevel Model of Validated Voter Turnout. Entries above
the line are estimates of micro-level parameters, ; entries below the line are
estimates of DMA-level parameters, .
. . . . . . . . . . . . . . . . . . . . .
50
9
Percent Correctly Identifying Congressional Candidates, 2006 . . . . . . . . .
56
10
Ability to Identify Incumbent Images as Function of DMA region Fixed Effects 60
11
Ability to Identify Challenger Images as Function of DMA region Fixed Effects 61
12
Ability to Identify Incumbent Images as Function of Demographics
. . . . .
63
13
Ability to Identify Challenger Images as Function of Demographics
. . . . .
63
14
Reliance on Local News for Information about Elections as a Function of
Employment
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
65
15
Ability to Identify Challenger Images as Function of Demographics and DMA
region Fixed Effects
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
65
16
Correctly Identifying the Challenger in a District from His/Her Image, DMA
region-Level
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
77
3

Abstract
This report investigates whether the structure of media ownership in television markets
affects the levels of civic/political engagement and political information of people living
within those markets. The FCC provided data on the structure of market ownership to
which we appended three types of data:
Proprietary survey data, collected in 2006 (CCES) and 2008 (CCAP)
Validated turnout information collected from Secretaries of States' offices
Political advertising data in the form of gross ratings points for each television media
market purchased from Nielsen, Inc.
The FCC's localism and diversity goals are partially designed to provide citizens with
enough local information and a diversity of viewpoints such that media consumption increases
the likelihood of participating in politics and people's levels of information about politics.
In this study, engagement, participation, and knowledge are indicators of the success at
reaching the localism and diversity goals at the market-level.
Localism We measure the effects of localism using 3,000 respondents from a portion of a
larger survey fielded during the 2006 midterm elections (Cooperative Congressional Election
Study, CCES). This module of the CCES targets 9 Midwestern TV markets covering 63
congressional districts. The focus on midterm elections lets us determine whether media
ownership structure is directly related to knowledge about local Congress members and the
people running against them. Essentially we ask whether the local media provide citizens
with information that helps them learn about candidates for Congress and whether media
market structure has anything to do with the answer. We find that there is variation across
media markets with respect to people's knowledge about local candidates, but that these
differences are due mainly to features of the congressional district (how much advertising the
candidates do in any particular race) and not due in any appreciable way to the ownership
characteristics of the market.
Diversity We measure the effects of diversity using a 20,000 person, 6-wave nationwide
survey fielded during 2007-8 (Cooperative Campaign Analysis Project, CCAP). We use cam-
paign interest, knowledge of candidates' positions on issues, general levels of political infor-
mation, and turnout in the election to examine whether media markets are in any way related
to citizens' levels of civic or political engagement and whether the ownership structure in
the markets are related to these differences. We find that there is variation on political
engagement across media markets, but that these differences cannot be explained by the
structural conditions related to ownership.

Part I
Introduction
The FCC has commissioned this investigation into the way that media ownership struc-
ture affects political and civic engagement. By ownership structure, we refer here to the
number of independent television or radio voices in a given market,1 how many parent cor-
porations own multiple broadcast stations in the market,2 whether any of the broadcast
outlets are also owned by a parent company that owns a newspaper in the market,3 the
number of unique radio stations with a news or talk format,4 and the fraction of households
in the DMA region that subscribe to Internet service of at least 200KBS.5 These measures
are meant to convey where the markets align, along an imaginary continuum, in terms of the
amount of unique political information broadcast outlets in the market provide to viewers.
Hypothetically speaking, on one extreme is a market with one television or radio station.
The people living in this DMA region do not receive very much political information via
broadcast media at all. On the other extreme is a market with 20 television or radio stations,
all uniquely owned and all providing some unique content about politics. People in this
DMA region may be exposed to a lot of political information via the media available to
them. Somewhere in the middle is the market with 20 stations, but also a set of four parent
companies that own five stations each. If the coverage across the five stations owned by each
conglomerate is the same, then the people living in this market have more opportunities
to hear political news, but the chances that they hear unique points of view are diminished
relative to people living in the 20-station market described above. Our goal was to exploit this
structure in the analyses that follow, however, our inability to relate any of these ownership
dimensions to political engagement limits the need to bring this level of nuance to the
analyses.
The FCC provided data on the structure of media ownership for television stations, radio
stations, and the ownership of newspapers within television markets.6 We use these data to
construct our characterization of the market ownership environment using the variables listed
above. To these data we appended three additional sources of information. We describe the
data in detail in the next section. These data cover the fall of 2006, just before the midterm
1As measured in the FCC provided dataset named TVMarkets.dta, using the variables called TVVOICES
and RADIOVOICES
2Variables MULTICOMTVPARENTS and COMRADIOCOMTVPARENTS
3Variable NEWSPAPERTVPARENTS
4Variable RADIONTPARENTS
5Variable BROADBAND200PCT
6These data are contained in the file TVMarkets.dta delivered originally to us on October 22, 2010 and
updated by the FCC on December 22, 2010.
1

elections to Congress; and the year leading up to the 2008 presidential election. We use the
FCC data on media ownership from 2005 and 2007 to investigate the effects of ownership
structure on political engagement in the 2006 and 2008 studies, respectively. Our general
approach is to determine whether there is any variation in political engagement or knowledge
across markets and to investigate whether we can characterize the variation in terms of the
ownership structure of markets.
In no case do we find that the ownership structure of the local media market affects levels
of civic or political engagement or knowledge. We conclude that there is significant within-
market variation that we can explain a portion of with individual-level demographics, but
that the across-market variation is not explained by the ownership structure in the market.
Although there is between-market variation in civic interests, it appears to be driven by two
things -- the local political context in the DMA region and the level of Internet penetration
in the DMA region. Specifically, we measure the local political context at the DMA region
level with the total advertising Gross Ratings Points that were purchased at the market level
for the election in reference. There are market-level factors that drive political engagement
and participation, but they are not tied to the ownership structure in the market to any
appreciable degree.
1
Data
1.1
The Cooperative Election Studies
The Cooperative Congressional Election Study (CCES) and the Cooperative Campaign
Analysis Project (CCAP) are in the family of election studies known as the Cooperative
Election Studies. These projects bring together Political Scientists, Economists, Commu-
nication scholars, Sociologists, and Psychologists from around the world in order to pool
financial resources for the purposes of running an election study with a large amount of
statistical power (through the acquisition of tens of thousands of cases). Each research team
buys in to the project in exchange for the opportunity to field original content to a portion
of the study's respondents.
Vavreck served as the study director for the first cooperative project in 2006 (the CCES)
(Vavreck and Rivers 2008) and, along with Jackman, served as principal investigator of the
CCAP in 2008 (Jackman and Vavreck 2009). We use these two datasets in our analysis of
media ownership and political engagement.
The data were gathered by the survey research firm YouGov/Polimetrix, Inc. of Palo
Alto, California. Interviews were conducted via the Internet. The data are representative
2

Figure 1:
Map highlighting the nine Midwest designated market areas (DMA re-
gions) analyzed in this study.
From the Northwest to the Southeast, these DMA re-
gions are: Minneapolis--St. Paul (MN), Madison (WI), Milwaukee (WI), Chicago (IL),
Champaign&SprngfldDecatur (IL), Lansing (MI), Detroit (MI), Cleveland-Akron (Canton)
(OH), and Columbus, OH (OH). The black boundaries on the maps outline Congressional
districts in the 109th Congress. Black dots highlight Congressional districts that intersect
the 9 studied DMA regions.
of target populations as described below. Details on the process by which panelists are
recruited and samples are made to be representative can be found in Vavreck and Rivers
(2008) and Jackman and Vavreck (2010).7
CCES
The subset of the CCES data with which we work was commissioned by teams from
University of California Los Angeles (UCLA) and the University of Wisconsin. These data
cover nine Midwestern Designated Market Areas (DMA regions) and include a total of 3,000
7Studies using data from each of these projects have been published across a wide variety of peer reviewed
journals in Political Science, including the profession's most prominent journal, the American Political Science
Review.
3

respondents.8 The UCLA/Wisconsin project was specifically designed to track media effec-
tiveness. The teams asked a number of questions about people's media habits but also asked
about things citizens are likely to learn from local media. The University of Wisconsin team
was particularly interested in the effects of local television news on attitudes and information
while the UCLA team was most interested in the effects of local campaigns. The project,
therefore, contains unique questions asking respondents to identify pictures of their local
candidates for U.S. House and their sitting Members of Congress. Tying these measures to
structural factors within the market provide a rare opportunity to assess whether geography
-- in terms of media markets -- affects local political knowledge at all, and if it does, to
what extent the ownership environment structures this effect.
The nine DMA regions and the Congressional districts that they include are shown in
Figure 1. The data were constructed to be representative of the general population with
respect to a number of demographic and political variables and were gathered in October of
2006 in the weeks leading up to the 2006 midterm Congressional elections (see Vavreck and
Rivers (2008) for a complete description of the project).
The nine Midwestern DMA regions range from large markets such as Chicago (3rd
largest in adult population) and Detroit (11th largest), to mid-sized markets, Columbus
(33rd largest) and Milwaukee (34th largest), to smaller markets, Champaign (82nd largest)
and Lansing (111th largest).9 The DMA regions cover in whole or in part 65 Congressional
districts. While most of these 65 districts intersect with only one of the nine DMA regions,
eight districts interest two of the DMA regions.10
A breakdown of the number of CCES respondents by DMA region is given in Table 1.
Table 1 also shows the size of the adult population in each of the nine DMA regions and
enumerates the Congressional districts that intersect each DMA region. The final column
of the table shows the representation ratio of the CCES sample with respect to each DMA
region. If the CCES sample sizes for each DMA region had been in perfect proportion
to the size of the DMA region, each of these ratios would be one.
Ratios larger than
one are associated with DMA regions in which the CCES sample over-represents the DMA
region, while ratios less than one are associated with DMA regions that are underrepresented
in the CCES. The deviations from proportionality in the DMA region-level sample sizes
are not cause for serious concern because we will be inferring differences among the DMA
8The nine Nielsen markets are: Champaign&Sprngfld-Decatur, Chicago, Cleveland-Akron (Canton),
Columbus, OH, Detroit, Lansing, Madison, Milwaukee, Minneapolis/St. Paul
9The markets are ranked by their 2005 eighteen and older population as provided by the variable
POP18PLUS in the file tvmarkets.dta furnished by the FCC.
10These districts are Michigan's 7th and 8th, Ohio's 4th, 5th, and 18th, and Wisconsin's 1st, 2nd, and
3rd.
4

regions rather than aggregate features of these nine DMA regions as a whole. However, it is
reassuring that most of the representation ratios fall between 0.75 and 1.25. Conveniently,
the most over-represented DMA regions are those that are the smallest in population. Thus,
we have many more respondents to leverage in the study of the two DMA regions in which
we have the smallest number of observations (Madison and ChampaignSpringfieldDecatur)
than would have been generated by a sample that was perfectly representative of the DMA
regions.
Designated Market Area
Observations
Adult
Representation
(Congressional districts)
(Weighted)
Population
Ratio
Champaign&Sprngfld-Decatur, IL
164
736,831
1.70
(IL 15,17,18,19)
(166.93)
Chicago, IL
685
7,079,787
0.72
(IL 111, 1316; IN 1,2)
(680.51)
Cleveland-Akron (Canton), OH
491
2,966,440
1.22
(OH 4,5,9,1018)
(485.38)
Columbus, OH
271
1,701,532
1.20
(OH 4,5,7,12,15,18)
(272.82)
Detroit, MI
479
3,766,926
0.95
(MI 715)
(476.31)
Lansing, MI
77
512,131
1.12
(MI 7,8)
(76.53)
Madison, WI
234
699,538
2.52
(WI 13)
(236.00)
Milwaukee, WI
260
1,682,118
1.15
(WI 1,2,46)
(258.00)
Minneapolis--St. Paul, MN
336
3,224,155
0.78
(MN 18; WI 3,7)
(336.73)
Table 1: Number of CCES respondents by DMA region
CCAP
The CCAP data consist of 20,000 impaneled observations representative of registered
voters with respect to a number of demographic and political variables.11 Registered voters
from 15 states in which the 2008 Presidential election was expected to be closely contested
and in which extensive campaign activity was likely to take place were over-sampled.12 This
11See Appendix A for a complete description of the sampling attributes of CCAP.
12These "battleground" states were selected based upon the likely competitiveness within each state and
5

"battleground" oversample can be corrected for by applying sampling weights and is not a
cause for concern in this study because we are seeking to make comparisons among media
markets and only 47 of the 210 media markets include territory from both battleground and
non-battleground states.13 The data collection began in December of 2007 and continued
across 5 additional panel waves in January, March, September, October, and November of
2008. The CCAP data contain questions on general civic responsibilities such as turnout
in elections, knowledge of candidates' positions on issues, overall levels of general political
information, and interest in public affairs. Once we merge these data with the FCC data on
media markets, we are able to place 19,159 respondents in 210 unique Nielsen media markets
(DMA regions) and 279 unique radio metro areas.14
CCAP respondents hailed from all 210 DMA regions that existed in 2007. As we would
expect, small DMA regions such as Glendive (MT) and Harrisonburg (VA) have small num-
bers of respondents (1 and 4 respectively), but all of the DMA regions have at least one
respondent. The largest number of respondents are found in the New York, Los Angeles,
Tampa--St. Petersburg--Sarasota, Philadelphia, and Chicago DMA regions with 691, 680,
573, 569, and 526 respondents respectively. The median DMA region had 40 CCAP re-
spondents and 50 percent of DMA regions have between 18 and 110 CCAP respondents.
Also as expected, the Tampa--St. Petersburg--Sarasota DMA region has more respondents
than larger markets such as Boston, Dallas--Forth Worth, Washington (DC), and Atlanta,
because of the CCAP battleground oversample that led to disproportionately more Florida
respondents and fewer respondents from non-battleground states such as Texas, Georgia,
and Washington (DC).
Figure 2 shows representation ratios for the CCAP sample for the 210 DMA regions.
DMA regions shaded in blue are under-represented in CCAP whereas DMA regions shaded
in red are over-represented. Darker shades of red or blue reflect larger degrees of over- or
under-representation respectively. DMA regions from Alaska and HI appear to the Southwest
and Southeast of Texas, respectively. Some of Alaska is not shown because those areas of
Alaska were not assigned to any DMA region in the data provide to us by the FCC. The
representation ratios ranged from an under-representation of 1 to 8.25 in the St. Joseph (MO)
DMA region to an over-representation of 2.7 to 1 in the Marquette, MI DMA region. This
are: CO, FL, IA, ME, MI, MN, NC, NH, NM, NV, OH, OR, PA, WI, and WV.
13In these 47 DMA regions, residents are not usually evenly split among the battleground and non-
battleground states. Rather, large majorities of these DMA regions' populations are generally either battle-
ground or non-battleground residents.
14We drop 841 cases from the analysis due to an inability to place the respondent in a single DMA region
or radio metro area. Our conversations with FCC representatives Tracy Waldon and Jonathan Levy on
December 22, 2010 about acquiring shape-files to place the remaining respondents in DMA regions led to
the conclusion that they were not worth using or purchasing for these purposes.
6

means that the one Marquette respondent represents 8 times as many adults in the DMA
region as the average CCAP respondent. Conversely, each of the 41 Marquette DMA region
CCAP respondents represent almost 3 times fewer residents of their DMA region than did
the average CCAP respondent. Ninety percent of DMA regions have representation ratios
that fall between 1 to 2.4 and 1.8 to 1. Because of the oversample of battleground states the
median DMA region representation ratio is 1 to 1.09.
The effect of the battleground oversample can be seen in Figure 2. Most of the electorally
competitive Midwestern states along with Florida, Colorado, New Mexico, Nevada, and
Oregon are shaded in red (over-represented). The states that were electorally uncompetitive
in the Presidential election, such as those of the deep South, New York, and California) are
shaded in blue (under-represented). The overall sense of the distribution of representation
ratios given in Panel (a) of Figure 2 is somewhat misleading because there are large differences
in population size across DMA regions and those differences are only weakly related to the
geographic extent (area) of the DMA region ( = 0.2). Thus, Panel (a) over-emphasizes
the smaller DMA regions in which the representation ratios are inherently less stable (DMA
regions in which only a few respondents are to be expected).
The map presented in Panel (b) of Figure 2 deforms the DMA region map such that each
DMA region remains in the same relative location (maintains all of its neighbors), but the
area of each DMA region is adjusted to be proportional to the population of the DMA region.
Displayed in this way, DMA regions like Los Angeles and New York become much larger
while the DMA regions of the mountain West are compressed. Once the larger DMA regions
are emphasized, we see much less variation in the degree of over- or under-representation
across DMA regions. Two clear remaining outliers are New York which was underrepresented
and Palm Springs which is over-represented. Once variation due to small DMA regions is
de-emphasized, the role of the battleground oversample becomes even more clear.
Figure 3 presents representation ratios based on the number of weighted CCAP respon-
dents. These weights correct for the oversample of voters from battleground states and
for some demographic imbalances. Once weighted, the CCAP sample becomes more rep-
resentative of the DMA regions. Ninety percent of the weighted representation ratios fall
between 1:1.7 and 1.6:1. Weighting to adjust for the battleground-state oversample, the
median representation ratio across DMA regions is 1.01:1. The regional pattern of over and
under-sampling becomes much less pronounced and the frequency which the darkest shades
of red and blue appear on the DMA region map is greatly reduced. Interesting, even after
weighting, the battleground-state DMA regions appear to be somewhat overrepresented in
the sample relative to non-battleground DMA regions.
7

(a) Designated Market Areas (DMA regions)
(b) DMA regions distorted so that their areas are proportional to their adult populations
More than 2:1
10:9 to 4:3
9:10 to 1:1
1:2 to 3:4
4:3 to 2:1
1:1 to 10:9
3:4 to 9:10
Less than 1:2
Figure 2: Both maps show the representation ratio of the CCAP sample within each of the
210 Designated Market Areas (DMA regions) in 2007. Reds reflects DMA regions that are
overrepresented in the CCAP sample. Blues reflect DMA regions that are underrepresented
in the CCAP sample. Darker colors reflect larger degrees of over- or under-representation.
Panel (b) is a cartogram that deforms the US map such that each DMA region's area is
proportional to the size of its adult (18+) population. Alaska's two DMA region's are shown
to the Southwest of Texas. The Hawaii DMA region is shown the Southeast of Texas.
8

(a) Designated Market Areas (DMA regions)
(b) DMA regions distorted so that their areas are proportional to their adult populations
More than 2:1
10:9 to 4:3
9:10 to 1:1
1:2 to 3:4
4:3 to 2:1
1:1 to 10:9
3:4 to 9:10
Less than 1:2
Figure 3: Both maps show the weighted representation ratio of the CCAP sample within each
of the 211 Designated Market Areas (DMA regions) in 2007. In these plot the CCAP data
are weighted to correct for the oversample of battle ground states and for other incidental
demographic imbalances. The color scale is the same as in Figure 2.
9

50
25
40
20
30
15
20
10
5
10
0
0
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0
1
2
3
(a) Representation ratios
(b) Weighted representation ratios
Figure 4: Histograms of representation ratios for the CCAP sample across DMA regions. In
Panel (a) the ratios are based on the unweighted CCAP observations. In Panel (b) the ratios
are based on weighted CCAP sample. The weights adjust for CCAP's battleground-state
oversample and for demographic imbalances.
The effect that applying the weights has on the DMA region representation ratios can be
seen more clearly in two histograms presented in Figure 4. In the left panel, we see a bimodal
distribution of ratios reflecting the separate populations of over-sampled battleground-state
DMA regions and under-sampled non-battleground state DMA regions. One the weights are
applied correcting for the oversample (among other things), the distribution of representation
ratios becomes unimodal and centered at 1. A few large outliers remain (and a few outliers
are exaggerated by the weighting), however these are largely confined to small population
DMA regions (as can be verified by examining the lower panel of Figure 3.)
1.2
Validated Turnout Data from U.S. Secretaries of States
To the 2008 survey data described above we append an indicator for whether the Secretary
of State in the respondent's state of residence validated, through state-based records of
participation, that the respondent was registered to vote and cast a ballot in the respective
election. These data are publicly available in most states. States vary in terms of the costs
associated with obtaining these data. Our validated turnout data was gathered by the survey
research firm YouGov/Polimetrix, Inc.
The process of actually doing the matching and validating is itself interesting and war-
10

rants some brief elaboration. Vavreck and Jackman were the PIs of CCAP, and in effect,
clients of YG/PMX. As such, information sufficient to identify the respondents on a voter
file is not delivered to the PIs and is not their intellectual property. Like almost any sur-
vey research firm, YG/PMX makes quite stringent guarantees of privacy to their panelists.
Aware of these circumstances, we did not seek (nor we were offered) identifying information
such as respondent names and addresses from YG/PMX.
Accordingly, the PIs were "firewalled" from the vote validation process. YG/PMX sent a
file of identifying information (e.g., names, addresses, date of birth) to a third-party firm that
specializes in vote validation.15 YG/PMX received a file with turnout history for respondents
for whom the contractor was able to match in its extensive data bases.16
Ninety-seven out of 20,000 CCAP respondents appear twice in the voter file data returned
to us by YG/PMX. In these instances we choose the voter file record with an entry for the
2008 general election; in every instance this is unambiguous.
For another three CCAP
respondents we have no corresponding records in the vote validation file and we presume
that no match could be found for these respondents.
n
%
On File
Unmatched
Unregistered
5 digit zip match
15,127
75.6
15,127
0
0
No zip on validation file
3,236
16.2
35
2,603
598
Zips do not match
1,599
8.0
1,599
0
0
Bad zip from YG/PMX,
31
0.2
31
0
0
good zip from validation file
Bad zip from YG/PMX,
7
0.0
0
5
2
no zip from validation file
Table 2: Zip code matches, CCAP and Vote Validation File from Third-Party Firm.
YG/PMX did provide the respondent ip codes (at least as reported by the respondents
to YG/PMX); YG/PMX also forwarded the ip code found by the third-party vote validation
firm. Table 2 summaries the congruence between the ip codes provided by the YG/PMX
15This is consistent with YG/PMX's relationship with its panelists, in that YG/PMX was not selling the
data, nor disclosing anything about the panelists other than that they might possibly be YG/PMX panelists.
In fact, the passage of data between YG/PMX and its contractors is subject to a non-disclosure agreement;
under the terms of this agreement the contractor returns the registration and vote history data and destroys
the data set of identifying information received from YG/PMX. On the respondent side, YG/PMX's agree-
ment with its panelists allows them to perform matching with data from third parties who agree to abide
with YG/PMX's privacy policies.
16Beyond the voter files, vote-matching firms also uses consumer data bases to help ascertain residency and
registration if someone is not in the voter files; e.g., if a person with a certain set of identifying characteristics
can be found in the consumer data bases, but not on the voter file, then this is strong evidence that this
person is not a registered voter (the alternative hypothesis is that the respondent is "real", but the voter file
records are incomplete or contain errors sufficient to make a match impossible).
11

respondents and the ip codes found by the vote validation. For three quarters of our (osten-
sibly) registered voter sample, the 5 digit ip supplied by the respondent corresponds with
the 5 digit match found by the validation contractor. For 16.2% of our respondents, the firm
could not find a matching record or a ip code in the voter files; for about one-fifth of these
cases (n = 598), the firm had enough information to identify the individual and conclude
that this person was not on a voter file and hence designated as "unregistered". For a small
number of individuals (n = 35), they did find the individual in the voter files, but did not
return the ip code from the voter file (presumably because the particular voter file did not
have that information). For 1,600 respondents (8%), the ip code provided to YG/PMX by
the respondent does not match the ip code on the matching record on the voter file found by
the validating firm. Intriguingly, there are 31 cases where the respondent provided a "bad"
ip code (e.g., less than five digits in a state without a "leading zero" in their ip codes);
nonetheless, in each of these cases the firm was able to find a matching record on the voter
files with a "good" ip code. Finally, there are another 7 cases with "bad" ip codes from the
respondent; 5 of these respondents could not be found in the voter files; for 2 of these 7 cases
enough information was found to conclude that this person was in fact not registered.
In 600 out of 20,000 cases (3%), we conclude that contrary to the respondent's asser-
tion, the respondent was in fact not a registered voter. In another 2,603 cases (13%), the
respondent could not be found on a state voter file. Of course, this is not to say that the
respondent is in fact not a registered voter; but given that a professional data-matching firm
with one of the best maintained collection of voter files in the country utilized a good deal of
identifying information from YG/PMX (at least name, address, gender, date of birth) and
utilized information in other sources and could not find a matching entry on any state voter
file for these respondents leads us to conclude that these respondents are not registered to
vote.
Rates of matching reveal that YG/PMX is able to track respondents across state lines;
these are almost all cases where (a) the respondent moved across state lines between com-
pleting a YG/PMX profile survey at time of initial recruitment, but updated their voter
registration records, or (b) the respondent gave erroneous state of residence information to
YG/PMX. The voter file data appear to exclude Nevada; the "state of registration" variable
returned to us has zero cases for Nevada, and on the YG/PMX side, the only matches for
the 213 respondents thought to reside in Nevada are from 16 respondents matched on voter
files in other states, or identified on consumer files as residing elsewhere.
Nevada aside, rates of matching our ostensibly registered voter sample on voter files
display considerable variation across states. In Mississippi, just 67 of 100 CCAP panelists
were found on voter files from that state. Similarly low rates come from Wyoming (21 out
12

70
75
80
85
90
95
70
75
80
85
90
95
q All
MT q
q VA
ND q
q SC
SD q
q OR
NH q
q NC
OH q
q TX
ME q
KS q
q KY
ID q
q NY
CT q
q AZ
MD q
q OK
IN q
q NJ
WI q
q WV
MN q
q VT
MI q
q HI
NM q
q GA
AR q
q LA
MA q
q TN
DE q
q CO
PA q
q CA
WA q
q UT
FL q
q AL
RI q
q AK
IA q
q DC
IL q
q WY
NE q
q MS
MO q
70
75
80
85
90
95
70
75
80
85
90
95
Figure 5: Registration Status of CCAP respondents in voter files, by state
of 31, or 67.7%), the District of Columbia (29 out of 40, 72.5%) and Alaska (38 out of
52, or 73.1%). States with high rates of matching include the Dakotas (SD: 46 out of 51
respondents, or 90.2%; ND, 40 out of 43 respondents, 93%) and Montana (53 out of 55
respondents, 96.4%). Across the entire set of 20,000 respondents, 84.0% were found on state
voter files, and another 600 respondents (3%) were determined to be unregistered. Figure 5
provides a graphical display of the data, dropping the problematic case of Nevada. Most
of the state-specific rates of verified voter registration for CCAP respondents are clustered
around the average of 84%.
1.3
Nielsen Gross Ratings Points Data for Political Advertising
In addition to the validated vote information from the voter files described above, we
contracted with the Nielsen Corporation to provide data on the advertising gross ratings
points purchased and aired by political candidates or parties running for any office in each
13

DMA region for 2006 and running for president in 2008. A gross rating point (GRP) is a
measure of the penetration of an advertisement into its targeted market. For example, one
GRP indicates that one percent of the targeted population saw the ad in question one time.17
The GRP data on political ad buys are an important control in a study of civic or
political engagement that aims to explain the effect of characteristics of media markets
unrelated to politics (ownership structure). The ads that candidates run in their DMA
regions are meant to increase political participation, engagement, and knowledge among the
population. If variation in GRPs is correlated with variation in ownership structure at the
DMA region level, without controls for political advertising, we run the risk of concluding
that market structure affects civic engagement when in fact, it was the political advertising
that was driving political interest, participation, and knowledge, not the other attributes of
the markets.
Nielsen, Inc. provided these data beginning August 1 of each election year under in-
vestigation and running through election day. The data are broken out by candidate, day,
creative, and outlet. The data we use here are limited to broadcast television outlets in the
market. We collapse the data to the level of candidates in 2006 because we are interested
in controlling for the effects of advertising on knowledge about each Congressional candi-
date over the course of the campaign. In 2008, we collapse the ad data for all candidate or
party-sponsored political advertising for the presidency in each market in order to control for
the amount of candidate-provided political information that was being broadcast to voters
during the 2008 presidential campaign in any given market.
2
Civic Engagement and Political Information
Political scientists have long been interested in questions of civic engagement, political
participation, and political knowledge. Figuring out how citizens acquire the information
needed to discharge the duties of citizenship is a large part of the investigations of many
prominent political scientists (Rosenstone, Hansen and Reeves 2003; Wolfinger and Rosen-
stone 1980; Delli Carpini and Keeter 1997; Lupia and McCubbins 1998; Gerber and Green
2000; Verba, Schlozman and Brady 1995).
The media have not been overlooked in these investigations. But, while the media play
a role in shaping the kinds of thing people think about (Iyengar and Kinder 1988) and the
criteria on which voters evaluate candidates (Iyengar and Kinder 1988), the main drivers of
17Technically, GRPs are defined as an advertisement's "REACH" multiplied by its "FREQUENCY". Thus
there are different ways to explain 10 GRPS, for example. One possibility is that 10% of the population saw
the ad a single time. Another possibility is that one percent of the population saw the ad 10 times.
14

participation are well-established as individual-level covariates such as age, race, gender, ed-
ucation level, income, marital status, employment status, and home ownership (Rosenstone,
Hansen and Reeves 2003; Wolfinger and Rosenstone 1980; Verba, Schlozman and Brady
1995).
Recently, a productive line of field experimentation has demonstrated the positive and
robust effects of canvassing and leafleting to remind voters that an election is coming up
(Gerber and Green 2000), shaming voters into participating by publishing the names of
those who do not vote in newspapers or newsletters (Gerber, Green and Larimer 2008), and
of non-partisan cable TV advertising, such as that done by groups like Rock the Vote (Green
and Vavreck 2008). These field tests have shown effects that range from 3 points for the
cable TV study to more than 10 points for the shaming study.
The extant literature on participation suggests that both individual-level attributes and
campaign-efforts at stimulation will be important drivers of political engagement.
With respect to media and market structures, what can we say about their role in fostering
civic engagement? Communication studies scholars have shown that despite the increases
in available political information (through cable TV news and increased outlets) political
knowledge and participation have not increased dramatically (Sunstein 2002; Prior 2007,
2005). Sunstien proposes that people's increasing ability to customize their political infor-
mation will have an unexpected effect on democracy because people are simply less likely to
encounter political news as a byproduct of turning on the television. This is not unlike Sam
Popkin's conception of the byproduct theory of information (Popkin 1991).
Popkin and Sunstien suggest that even those who are uninterested in politics may come
across some snippets of political information simply because they are watching television or
listening to the radio. Take for example a typical household in the 1960s. If the television
was on in the evening, chances are some form of news was part of the programming --
across all three broadcast networks. People could not escape the news in this setting. In
contrast, today it is easy to escape all forms of news when watching television -- in fact,
people's ability to expose themselves to only the type of television programming they like is
impressive and has lead to an entire industry devoted to understanding market segmentation
and micro-targeting. Narrowcasting has replaced broadcasting.
The implication of the Sunstien argument is that greater choice allows politically inter-
ested people to access more information and increase their already impressive amounts of
political knowledge, while people who prefer pure entertainment or sports to politics can
avoid being exposed to political information all together. Markus Prior (2005) demonstrates
that this pattern exists and that the gaps in knowledge and participation between those who
prefer news on television and those who prefer entertainment shows is widening over time.
15

He assumes this is due to the increasing choice over programming in the media environment.
For our purposes, if choice breeds a widening gap between types of viewers in terms of
participation and knowledge, we may be able to see the pattern in markets in which there
are many broadcast outlets compared to those in which there are fewer. Of course, cable and
satellite television provide ample numbers of choices for respondents across media markets,
so perhaps the pattern will be difficult to detect given our focus on the effects of broadcast
stations alone, however, we note that cable programming is typically not local in orientation.
We present the evidence on whether the structure of ownership affects civic engagement
in the following manner:
Structure -- We operationalize structure as it relates to unique information providers
in the market by investigating the role of the number of independently owned tele-
vision stations in the market, the number of radio stations in the market, the number
of parent entities that own more than one television station in the market, the num-
ber of parent entities that own at least one television station and a radio station in
the market, the number of parent entities that own at least one television station, a
radio station, and a newspaper in the market, the number of Internet service providers
providing service at 200KBS, and the number of unique radio stations with a news or
talk format in the market.18
Engagement -- We operationalize civic engagement in two forms, participation and
knowledge. In terms of participation, we use the information from the vote validation
in 2008 to investigate the drivers of turnout in the presidential election. In terms of
knowledge, using the 2008 data we investigate people's overall levels of general political
information, their abilities to place the candidates on important political issues relevant
to the campaign, and their level of interest in public affairs generally. In 2006, we
analyze people's abilities to identify the candidates running for local office (Congress).
We begin with the 2008 study, which we refer to as "Project A" and from there move to
the 2006 evidence, "Project B." In each study, we proceed in the following manner:
1. Start with simple models of engagement using only the individual-level attributes that
have been shown to drive participation and knowledge.
2. Add fixed effects for the DMA regions.
3. Assess whether any of the variation in engagement is explained by DMA region indi-
cators. If so, we attempt to explain that structure.
18The exact names of the variables and the datasets from which they were drawn are listed on page X.
16

4. Recover the appropriate regression/logistic parameters describing the fixed effects for
the markets and treat those as the "biggest possible net effects" due to market-level
attributes.
5. Individually demonstrate the relationship between the structural characteristics of mar-
kets (number of voices, cross-ownership features, etc . . . ) and the fixed-effect
parameters for each measure of engagement.
6. Model engagement hierarchically using all the market-level indicators to assess the role
of structural factors across the measures of engagement when all indicators are allowed
to work simultaneously.
We turn now to Project A, which focuses on people's levels of interest, knowledge, par-
ticipation, and uncertainty regarding politics and candidates.
Part II
Analyses: Political Interest,
Knowledge, Uncertainty, and
Participation (Project A)
3
Analyses of Variation
In project A we examine four general indicators of civic engagement: Levels of interest in
politics and current affairs, overall levels of political knowledge, people's willingness to place
the candidates on important campaign issues (their level of uncertainty about the candidates'
positions), and turnout in the 2008 presidential election. We use the 2008 CCAP data, which
covers 210 DMA regions as described earlier.
For each of the four dependent variables, we begin the analyses with the approach de-
scribed above: we use the fixed effects from a model with basic demographics and DMA
region indicators and we attempt to model the structure of those fixed effects with the
variables detailing the ownership characteristics of markets. From there, we move into a
full-hierarchical model allowing the effects of covariates to change based on characteristics
of the markets.
17

3.1
Civic Engagement: Dependent Variables
We use the following four measures as dependent variables tapping in to civic engagment:
Interest We begin with a general measure of interest in politics. We asked people, "How
interested are you in politics and current affairs?" People could answer in three decreasingly
interested categories (very much, somewhat, not that much).
Willingness to Place Candidates on Issues We also asked people to place the candi-
dates on important political issues that were being discussed during the 2008 presidential
campaign. One of those issues was health care reform. We asked people:
Which comes closest to Barack Obama's view about providing health care in the
United States? (Choose one of the following)
The Government should provide everyone with health care and pay for it
with tax dollars
Companies should be required to provide health insurance for their em-
ployees and the government should provide subsidies for those who are not
working or retired.
Health insurance should be voluntary. Individuals should either buy insur-
ance or obtain it through their employers as they do currently. The elderly
and the very poor should be covered by Medicare and Medicaid as they are
currently.
I'm not sure, I haven't thought much about this
We take the health care question and we dichotomize it into two categories, giving people
a one if they will not place the Barack Obama on health care and a zero if they will place
him at a position. This "unwillingness" to place the candidate on an important campaign
issue is best thought of as a measure of uncertainty about the candidate's position on the
issue. These kinds of uncertainty measures are known to be robust predictors of vote choice
and favorability (Bartels 1986; Vavreck 2009), and they correlate in predictable ways with
demographic attributes such as education, age, gender, and income.19
19People who were not asked this question are dropped from the analysis as they are not considered
unwilling or too uncertain to place the candidate.
18

Political Knowledge In order to characterize people's overall level of political sophistica-
tion or knowledge more generally, we asked a series of 12 questions that we use together in
a scale. Ten of these questions take the same form, they ask people to place a prominent
politician or business person in one of three possible jobs -- a Member of the House of Rep-
resentatives, a Senator, or neither of those things. People's responses are coded in binary
fashion as either correct or incorrect. We asked about the following ten people: John Dingell,
Nancy Pelosi, Bill Gates, John Boehner, Susan Collins, Henry Waxman, Jon Kyl, Dennis
Kucinich, Patrick Leahy, and Ted Kennedy. Additionally, we asked people to choose what
job Condoleeza Rice held (from four choices) and whether people could correctly identify
why Guant
anamo Bay had been in the news lately (from four choices).
There is some interesting variation in the item difficulties with the item about Senator
Susan Collins being the hardest item (only 39% of respondents got that one right in a setup
where 33% could get it right by guessing). The item about Bill Gates, CEO of Microsoft
Corporation is the easiest item -- 97% of respondents got that right.
To form the scale we use an Item Response Theory model (IRT). We pass this set of
binary coded items to the following item-response theory model, a model widely used in the
social-sciences to recover a latent measure from a set of binary items:
ij = Pr(yij = 1|i, j, j) = F (ij - j)
(1)
where
yij {0, 1} is the i-th subject's answer to the j-th item (e.g., yij = 1 if correct, yij = 0if
incorrect), where i = 1, . . . , n indexes respondents and j = 1, . . . , m indexes items;
i R is an unobserved attribute of subject i (typically considered ability in the
test-taking context, or ideology in the analysis of legislative data)
j is an unknown parameter, tapping the item discrimination of the j-th item, the
extent to which the probability of a correct answer responds to change in the latent
trait i
j is an unknown item difficulty parameter, tapping the probability of a correct answer
irrespective of levels of political information
F () is a monotone function mapping from the real line to the unit probability interval,
typically the logistic or normal CDF.
A one parameter version of the model results from setting j = 1, j; i.e., items vary in dif-
ficulty, but not in terms of their discrimination, and is often called a Rasch model. Connec-
19

tions between IRT models for binary indicators and the factor analysis model for continuous
indicators have been noted by Takane and de Leeuw (1987) and Reckase (1997).
The statistical problem here is inference for = (1, . . . , n) , = (1, . . . , m) and =
(1, . . . , m) . We form a likelihood for the binary data by assuming that given i, j
and j, the binary responses are conditionally independent across subjects and items; this
assumption is called "local independence" in the argot of IRT. That is,
n
m
L
y
=
ij (1 -
ij
ij )1-yij
(2)
i=1 j=1
where ij is defined in equation 1.
The model parameters are unidentified. For instance, any linear transformation of the
i can be offset by appropriate linear transformations for the j and j; an obvious case is
scale invariance, in which ij = F (ij - j) indistinguishable from the model with =
ij
F ( -
= c
=
i
j
j ) where
i
i and
j
j /c, c = 0. A special type of rotational invariance
arises with c = -1.
Any two linearly independent restrictions on the latent traits are
sufficient for at least local identification, in the sense of Rothenberg (1971); a typical example
is a mean-ero, unit variance restriction on the i, while setting at least one pair of (j, j) item
parameters to fixed values is one way of obtaining global identification. Here we impose the
identifying restriction that the latent traits (the i) have mean zero and standard deviation
one across respondents.
We use a Gibbs sampler to generate a Monte Carlo based exploration of the posterior
density of the model parameters = (, , ) . By Bayes Rule, this posterior density is
p(|y) p(y|)p() where p(y|) is the likelihood defined above (equation 2) and p() is
a set of prior densities. We assume a priori independence across all elements of , specifically
i N (0, 1), j N (0, 52) and j N (0, 52), each iid across i and j, respectively. Note
also that post-estimation, we impose the identifying restriction that the Bayes estimates of
the i (the means of their respective marginal posterior density) have mean ero and standard
deviation one across respondents; this normalization is functionally equivalent to imposing
some additional prior structure on the parameters. The Gibbs sampler is implemented in the
ideal function in the pscl package in R, as shown below. Further details appear in Jackman
(2009, section 9.3).
We use the resulting scores from this item response model as a dependent variable as-
sessing people's underlying level of general political knowledge or sophistication.
Turnout Finally, as described in Subsection 1.2 above, we acquired validated turnout infor-
mation from Secretaries of States offices. We append these data to our survey data and use
20

them as our final measure of civic engagement.
3.2
DMA region Fixed Effects
We begin the analysis with a purely exploratory investigation of the relationship between
DMA regions and engagement. We are interested in decomposing the variation in engage-
ment into within-DMA region and between-DMA region components. The DMA region
fixed effects give us an indication of how much of the variation in each of the engagement
dependent variables comes from between-DMA region variation. The individual-level demo-
graphics explain the within-DMA region variation.
It is worth noting at this point that we do not view these analyses as structural models
for interpreting the effects of market-level features. We are, at this point, merely interested
in discovering whether accounting for DMA region-level variation increases the overall level
of explained variation in each of the engagement variables. Because of this, we estimate the
models for the variance decomposition using a simple least squares linear probability model
(LPM) across all of the dependent variables, even the dichotomous and ordered ones. For
political interest, willingness to place Obama on healthcare, and turnout, the LPM will not
be appropriate when we move to a structural model of the market ownership frameworks.
Logistic regression will replace the LPM at that point.
In other words, for this exploratory analysis, we disregard levels of measurement, fitting
linear regressions irrespective of whether we have a continuous, binary or ordinal dependent
variable. Our goal at this stage is merely to understand the structure of the data. Specifically,
all we want to know at this stage is how much variation in a given y can be accounted for
with a set of "fixed effects" for DMA regions, versus, say, other plausible sources of variation
in the data (e.g., fixed effects for state, fixed for birth year, etc).
We define a function that we use repeatedly in this exploratory mode of analysis: a func-
tion that regresses a given y on various fixed effects, and reports the resulting r2. We ask the
following questions: How much of the variation in each of the engagement dependent vari-
ables is explained by DMA region-level fixed effects alone? And what do the demographics
add? We present these results in Table 3 and Table 4.
There appears to be modest structure to the civic engagement markers based on DMA
region indicators. In other words, we conclude there is modest between-DMA region variation
to explain in these data. But most of the variation-explained comes from within-DMA
region indicators that take the form of individual-level demographics including: gender,
race (black, white, hispanic, other), education (college or not), age, and income (5-levels).
These demographics explain 19% of the variation in overall levels of political information.
21

Table 3: Least squares regression analysis of Political Information (mean 0, sd 1) and Political
Interest (1, 2, 3), October wave. All models include unreported fixed effects for income levels.
Info 1
Info 2
Interest 1
Interest 2
Female
-0.41
-0.41
-0.18
-0.18
(0.01)
(0.01)
(0.01)
(0.01)
Black
-0.23
-0.26
-0.05
-0.07
(0.02)
(0.02)
(0.02)
(0.02)
Hispanic
-0.20
-0.26
-0.03
-0.09
(0.03)
(0.03)
(0.02)
(0.02)
Other Race
-0.05
-0.07
0.01
-0.02
(0.03)
(0.03)
(0.02)
(0.02)
College
0.46
0.44
0.18
0.18
(0.02)
(0.02)
(0.01)
(0.01)
Age (Years/10)
0.09
0.09
0.06
0.06
(0.00)
(0.00)
(0.00)
(0.00)
DMA Fixed Effects
No
Yes
No
Yes
r2
0.19
0.22
0.12
0.16
N
19,650
19,650
16,404
16,404
Standard errors in parentheses
Adding DMA region indicators increases the percent of variation explained to 22%. The
between-DMA region indicators increase explanatory power by 3 points. Similar trends
exist for political interest (12% explained by demographics alone and 16% with DMA region
indicators) and the respondent's willingness to place Obama on healthcare (8% to 13% with
the markets accounted for).
For turnout, however, we see an opposite trend. Only 4% of the variation in validated
turnout is explained by the individual-level characteristics of respondents. This is increased
to 10% when the DMA region indicators are added. Here, the DMA region fixed effects
increase the percent of variation in turnout that is explained by the model by 6-points.
While the size of this effect is not that different from the size of the increases due to DMA
region fixed effects from the other dependent variables, the level of this effect -- being
larger than the percent of explained variation from the demographics -- is what draws our
attention.
Despite the modest levels of between-DMA region variation in engagement, we can ascer-
tain whether the fixed effects are related to one another across the four dependent variables
measuring engagement. If the fixed effect estimates are related across the dependent vari-
ables, it opens the possibility that a common factor is driving the variation across all the
engagement measures. We present these comparisons in Figure 6.
22

1.0
1.5
2.0
2.5
0.2
0.4
0.6
0.8
0.0
-0.5
Info FE
0.39
0.35
0.25
-1.0
-1.5
2.5
q
q
q
q
q
q
q q
q
q
q
q q
q
q
q
qq q
q
q
q
q
q
q
q
qq q
q
q
q q
q
q q qq q
q
q q
q
q q q
q
q
q
q
2.0
qq
q
q qq
q
q q
q q
q
q
q
qq
qq q
q
q qq
q
q
q
qqqq
q q
q
q
q
q q
q
q q
q
q
q
q
qq
q q
q
q q
q q
q
q q
q q
q
q
q
q
q q
q
q q
q
q
q q
q
q
q q
q q
q
q
q
q
q
q
q
q
q q
q q
q
q q
q q
q
qq q
q
qq
q q
q
q
Interest FE
0.43
0.12
qq
q
q q
q
q
q
q q
q
qq
qqq q
q
q q
q
q
q
q
q
q q
q
1.5
q
q
q
q
q
q
q
q
q
1.0
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.8
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q q
q
q q
q
q q
q q q
qq
q
q
q
q q q
q
q
q
q
q
q
q qq q
qq
q
q
q
q
q
q
q
q
q
q
qq
q
q q
q
q q
q
q q
q qq
qq
q q
q
qq
q
q q
q q
q
q
q q
qq q q
q q
q
q q
q
q
q q
q q
q q q
q qq q
q q
q
qq q
q
q q
q q
q
q q
q q
q
q qq
q
q qq q
qq
q q q
q qq
q
q qq
q
q
q
q
q
q q
q q
q q
q
0.6
q
q
qq
q
q
qqq
q q
q
q
q
q
q q
q q
q
q q qq q q q
q q
q
q q
q
q
q q q
q
q
q
q
q
q q
q q
q q
q
q
q
q
qqq q qq
q
q
q
q
q
q qq q q
q
q
q
q q
q
q
q
q
q q
q
q q qq
q q
q qqqqq
q
q q q
q
q q
q
q
qqq
q
q
q q
q
qq q
q
qqq q
Obama Health Care
q
q qq
q q
q
q
q q
q
q
q
q
q q
qq
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
0.17
q
q
q q
q
q
q
q
q
qq
0.4
q
q
q
q
q
Ability to Place FE
q
q
q
q
q
q
q q
q
q
q q
q
qq
q
q
q
q
q
qq
q
q
q
q
0.2
0.0
q
q
q
q
q
q
q q
0.8
q
q
q
q
q
q
q
q
q
q q
q
qq q
q
q
q
qq
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q q
q
q q q
q
q
q
q
q
q
q q
q
q
q q
q
q
q q
q
q
q
q
q
q
q
q qq q
q
q q q
q
q
qq
q
q
q q
q
q
q
q
q
q q
q
q
q qq
q
q
q
q
q
q q
q
q
q
q
q q
q
q
q
q
q
q q
q
q
q
q
q q
q
q
q
q q
q
q q q
q
q q
q
q
qq
q q
qq
q
q q
q q
q
q q
q q
q
q
q
q
q
q
q
q
q q
q
q q
q
q
qq
q
qq
q
qqq
q
q
q
q
q
q q q
qq
q q
q
qq q
q
qq
qq
q q
q
q qqq
q
q
q qq q
q
q q
q q
q
q
qq
q
q q
q q
q
q
q q
q
q q q qq
q
q q
q
q
qq q qq
qq q
q
qqq
0.6
q q
q q q
q
qq qq
q
q
q
q
q q
q
q q
q
q q
q
q
q
qq q
qqq
q qq
q
qq
q q
qq
q
q
q
q
q
q
q
q
q q
qqq
q q
q
q
q
q q q
q q
q
q
q
q
q
q
q q qq q
q
q
q
q
q q qq
q
q q qq
q q
q
q
q
q
q q
q
q
q
q
q qq
q
q
q
q q q
q q q
q
q
q qqq
q q q qq
q
q
qq q
qq
q qq q
q q
q
q
qq
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q qqq q
q
q
q q
q
q
q
q
q
q
q
q
qq qq
qq
q
q
q
q q
q
qq q qq q
q
q
q
q
q
q
q
qq
q
q
q q
q
q
q
q
q qq
q
Validated
q
q
q
q
q
q
q
q q q
q
q
q
q q q
q q
q
q
q
q q q
q
q qq qq
q
q q q
q
q
q
q
q
q
q
0.4
q
q
qq
q
q q
Turnout FE
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.2
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
-1.5
-1.0
-0.5
0.0
0.0
0.2
0.4
0.6
0.8
Figure 6: Scatterplot Matrix of Fixed Effects
23

Table 4: Least squares regression analysis of Ability/willingness to locate Obama on health
care issue, September wave of CCAP (0,1) and Validated Turnout (0,1). All models include
unreported fixed effects for income levels.
Placement 1
Placement 2
Turnout 1
Turnout 2
Female
-0.10
-0.10
0.00
0.01
(0.01)
(0.01)
(0.01)
(0.01)
Black
0.01
-0.01
0.00
0.02
(0.01)
(0.01)
(0.01)
(0.01)
Hispanic
-0.06
-0.09
-0.04
-0.05
(0.01)
(0.01)
(0.01)
(0.01)
Other Race
-0.01
-0.02
-0.01
-0.02
(0.02)
(0.02)
(0.01)
(0.01)
College
0.14
0.13
0.06
0.06
(0.01)
(0.01)
(0.01)
(0.01)
Age (Years/10)
0.02
0.02
0.03
0.03
(0.00)
(0.00)
(0.00)
(0.00)
DMA Fixed Effects
No
Yes
No
Yes
r2
0.08
0.13
0.04
0.10
N
15,509
15,509
16,929
16,929
Standard errors in parentheses
Figure 6 is a scatterplot matrix showing the way the coefficients on the DMA region
fixed effects for each measure of engagement are related to one another. These coefficients
can be thought of as representing the exceptionalism of each DMA region on the dependent
variable under analysis, net of individual-level predictors. In other words, these off-sets can
be thought of as the "remainders" in each DMA region mapping directly onto the dependent
variable and due specifically to something occurring in the physical space of that market.
The first column shows the relationships between fixed effects in the information model,
the second column is for interest in politics, the third is willingness to place Obama on
healthcare and the last column in turnout. The rows are ordered similarly. The bottom
triangle plots the off-sets against one another. As such, there are roughly 210 datapoints
in each square. The top triangle reports the pearson correlation between the two sets of
estimates.
The purpose of this analysis is to search for patterns of any functional form between the
DMA region off-sets on different measures of engagement. The method does not require us
to have theoretical expectations about whether and how the DMA region-level off-sets might
be related to one another. We are looking for patterns.
What might some of these patterns look like? If the estimates were perfectly correlated
we would expect to see all of the datapoints on a line. This type of relationship would sug-
24

gest that something at the DMA region-level was driving the observations of both dependent
variables in the same manner. To the extent that there is no relationship between the esti-
mates, we would expect to see datapoints scattered in the squares. This type of relationship
would suggest no structure to the DMA region-level off-sets across the dependent variables
-- for example, the markets may very well differ from one another on one measure, but
the way that they differ bears no relationship to the way they differ on other measures, net
of individual-level predictors. This lack of patterning would suggest no common driver of
political engagement at the DMA region-level; it does not, however, rule out the possibility
that different DMA region-level factors affect different measures of engagement.
To assist readers in visualizing these relationships, we fit a loess smoother in each square
in red.20
The loess curves show slight positive associations between the market off-sets
across measures of engagement. The strongest relationship is between political interest and
willingness to place Obama on healthcare. This association indicates that there is something
ordering the markets from high to low levels of political interest that that orders them
similarly from willing to unwilling to place Obama. The off-sets for interest are also highly
correlated with the off-sets for political information more generally. The off-sets for turnout,
in contrast, seem to have little relationship to any of the other market-level off-sets (with
the exception of political information). Net of individual-level predictors, the relationships
among these off-sets for market suggest an underlying structure at the market level that is
worth further exploration.
We attempt to define this underlying structure by taking each dependent variable and
uncovering whether the ownership structure of the DMA region has any relation to these
off-sets and their patterns of correlation.
We do this by presenting similar matrices of
scatterplots for each measure of engagement and the market-level offsets. We do this for
each of our measures of market ownership. Once more, we evaluate the non-parametric loess
curves searching for any indication of a pattern in the data.
In Figure 7 and Figure 8 we present the relationships between the number of independent
commercially owned television and radio stations and the fixed effects, separately. In both
cases, the relationships across all four dependent variables are flat. The data span most of
the X-axis, but there is no clear pattern to the effects of increasing market voices, whether
television or radio, and political interest, knowledge, uncertainty, or participation.
20Loess, sometimes called locally weighted scatterplot smoothing, combines much of the simplicity of linear
least squares regression with the flexibility of nonlinear regression. It does this by fitting simple models to
localized subsets of the data to build up a function that describes the deterministic part of the variation in
the data, point by point. One of the chief attractions of this method is that the data analyst is not required
to specify a global function of any form to fit a model to the data, only to fit segments of the data. In
25

5
10
15
20
Can Place Obama, Health Care
Validated Turnout
q
q
q
q
q
q
q
q q
q
q
0.8
q
q
0.8
q
q q
q
q
q
q
q
q
q q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q q
q q q
q
q
q
q
q
q
q
q
q
q
q q q
q q q
q
q
q q
q q
q
q q q
q
q
q
q q q
q q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
0.6
q
q
q q q q
q
q
q
q
q
q
q
q
q
q q
q
q
q q q
q
q
q
q q
q
q
q q
q
q
q q
q
q q q q
q
q
0.6
q
q q
q
q
q
q
q
q
q
q
q q q
q
q q q
q
q q q
q q
q q
q q
q
q
q
q q
q
q
q
q q q q
q
q
q
q
q q
q q q
q
q
q
q
q
q
q
q
q
q q q
q q
q
q
q q
q
q
q
q
q
q
q
q
q q q
q
q
q
q q
q q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q q
q
q
q q q
q q
q
q
q
q
0.4
q q q
q
q
q
q
q
q
q q
q q
q
q
q
q q
0.4
q
q
q
q
q
q
q q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
0.2
q
q
q
q
0.2
q
q
q
q
0.0
q
q
q q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q
q
q
q q q
q
q
q
q q
q q q
q
q
q q
ed Eff
q
-0.5
q q
q
q q
q
q q
q
q
q
q
q
q q q q
q
q
q
q
q
q
q
q q q q q q
q
q q
q
q q
q
q q q q
q
q
q
q
Fix
q
q
q
q
q q
q
q
q
q
q q
q q
q q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q q q q
2.0
q
q
q q q
q
q
q
q
q q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q q
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q q q
q
q
q
q
q q
q
q
q q
q q q q q
q
q
q
q q q
q q q
q
q
q
q
q q
q
q q
q
q
q
q
q
q q q
q
q
q
q q
q q
q
q
q q q
q
q
q
q q q
q
q
q
q
q
q q q q
q
q q
q
q
q
q q q
q q
-1.0
q
q q q
q q
q
q q
q
q
q
q q
q
q q q q
q
q
q
q
q q
q
q
q q q q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q q
q
q q
q q
q
q
q
q
q
q
q q q q q
q q
q
q
q q q q q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q q
q
q
1.5
q
q
q
q
q q
q
q
q
q
q
q
-1.5
q q
q
q
q
q
q
q
q
q
q
q
q
1.0
5
10
15
20
Number of Independent TV Stations (TVVOICES)
Figure 7: Number of Independently Owned TV Stations by Fixed Effects
26

0
20
40
60
80
100
120
Can Place Obama, Health Care
Validated Turnout
q
q
q
q
q
q
q
q
qq
q
0.8
q
q
0.8
q q
q q
q
q qq
q
q
q
q
q
q
q
q q
q
q
qq
q
qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q q
qq q
q
q
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q q
q
q q q
q
q q
q
q q
q
q q
q
q
q q
q q
q
q
q
q
q
q q
q
q
q
q q q
q
q
q
q
q q
q
q
q q
q
q
q q
q
q
q
q
q q
q
q q q q
q
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q
q
q
q
q
q
q
0.6
q
q
q
q
q
q q
q
q q
q
q q
q
q
q
q
qq q
q q
q
q q
q
q q
q
q
q
q
q
q
q
q
q
q
q q qq
q q
q
q
q
q
0.6
q
q
q
q
q
q
q
q
q
q
q
q
q qq
qqq
q q
qq
q
q q
q
q
q
q
q
q q q
q q q
q qq q
q q q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q q
q
q
qq
q q
q q
q
q q
q
q
q
q
q
q
q
q q
q q qq qq
q
q
q
q
q q
q q
q
q
q q
q
q
q
q
q
q
q
q
q q
qq
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
0.4
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.4
q
q
q
q
q
qq q
q
q
qq
q
q
q
qq
qq
qq
q
q
0.2
q
q
q
q
0.2
q q
q
q
0.0
q
q
q
q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q
q
q
q
qq q q
q
q
q
q
q
q
q
q
q
q
ed Eff
q
-0.5
q
q
q
qq q
q
q q
q
q
q q
q
q
qq
q q q
q q
q
q
q
q
q
q
q
q q
q
q q q
q
q
q
q
q
q
q
q q q
q q
qq
q q
q
q
q
q
Fix
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q
q q
q
q
q
q q
q
q
q
q
q
q q
q
q
q
q
2.0
q
q
q
q
q
q
q
q
q
q
q
qq
q q q
q q
q
q
q
q
q q q q
q
q
q
q q qq
q q
q
q
qq
q
q q
q q
q
q
q q
q
q
q
q
q
q
qq
q
q
q
q q q
q
q
q
q
q
q q q
q q
q
q q
q
qq q
q
q
q qq
q
q
q
q q
q
q
q
q q
q
qq
q
q
q
q q
q
q
q
q
q
q
q
q q q
q
q
q
q
q
q
q
q q
q
q
q q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q q
-1.0
q
q
qq
q
q
q q
q
q
q
q
qq q
q
q
q
q q
q
q q
q
q
q
q
q
q
q
q
q
q q
q q
q q q
q
q q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q qq
q
q
q
q
q
q
q q
q
q
q
q
q q
q q
q
q
q
q
q
1.5
q
q
q
q q
q
q
q
q
q
q
q
-1.5
q q q q
q
q
q
q
q
q
q
q
q
q
q
1.0
0
20
40
60
80
100
120
Number of Independent Radio Stations (RADIOVOICES)
Figure 8: Number of Independent Radio Stations by Fixed Effects
27

The only slightly interesting movement is in the relationship between radio stations and
political interest at the low end of the scale, where increasing numbers of stations are leading
to decreasing levels of interest in politics (remember that interest is coded in reverse), but
this trend is not clear enough, even with this large amount of data, from which to make any
inferences. If ownership is related to the structure of these DMA region-level effects, it will
have to come from multiple- and cross-ownership and not from the sheer numbers of voices.
Looking at Figure 9, Figure 10, and Figure 11, however, we see that these trends are flat
as well. Whether we examine multiple television station ownership, the television-radio cross-
ownership, or this cross plus a newspaper -- there is simply no pattern to the relationships.
The FCC provided data on the penetration, at the market level, of Internet usage. In
Figure 11 we present the data on the percentage of the households with Internet access of
greater than 200 bauds per second by the DMA region level fixed effects.
While these trends show more shape than the ownership structure patterns, they are
essentially presenting equally null relationships. There may be a slight association between
Internet penetration rates and the fixed effects for validated turnout, but the movement is
slight and only for the lower half of the penetration scale.
Finally, using Nielsen data on the total number of gross ratings points purchased by
the presidential candidates in each DMA region, we look for patterns between these ad-buy
levels and the DMA region off-sets. The measures of GRP in each DMA region are simply
the sum of all the political advertising that was purchased in a given DMA region by any
presidential candidate or party on behalf of a presidential candidate from August 1, 2008 to
Election Day. These data are presented in Figure 13.
The patterns are not striking, but each of the loess lines slopes in a downward direction,
suggesting that areas with a lot of presidential advertising are areas with low levels of interest,
turnout, information, and knowledge. This is no surprising when one considers the fact that
the design of this test is not experimental -- meaning we cannot conclude that increased
advertising leads to declines in participation, for example, since candidates are deciding
where to place these ads strategically, and they may be placing ads in exactly those places
where people are less engaged and less likely to participate.
this case, we fit a weighted quadratic least squares regression over the span of values on the y-axis using a
bandwidth setting of .8.
28

Can Place Obama, Health Care
Validated Turnout
0.8
0.8
q
q
0.6
q
q
q
q
0.6
q
q
q
q
0.4
0.4
q
q
q
0.2
q
q
0.2
q
q
q
0.0
q
q
ects
q
q
Political Information
Political Interest
ed Eff
0.0
q
2.5
Fix
q
q
-0.5
q
2.0
q
q
q
q
q
q
q
q
q
-1.0
q
1.5
-1.5
q
q
q
q
q
q
q
1.0
0
1
2
3
4
0
1
2
3
4
Number of Parents owning more than 1 TV Station (MULTICOMTVPARENTS)
Figure 9: Number of Multiple-TV Station Parents by Fixed Effects
29

Can Place Obama, Health Care
Validated Turnout
0.8
0.8
q
q
q
0.6
q
q
0.6
q
q
q
q
q
q
q
q
q
0.4
0.4
q
q
0.2
q
0.2
q
q
q
q
0.0
q
q
ects
q
q
Political Information
Political Interest
ed Eff
0.0
q
2.5
Fix
q
-0.5
q
2.0
q
q
q
q
q
q
q
q
q
q
q
q
-1.0
1.5
-1.5
q
q
q
q
q
q
q
q
1.0
0
1
2
3
4
5
0
1
2
3
4
5
Number of Parents owning at least 1 TV & Radio Station
(COMRADIOCOMTVPARENTS)
Figure 10: Number of Parents Owning at least one Television and one Radio Station by
Fixed Effects
30

Can Place Obama, Health Care
Validated Turnout
0.8
0.8
0.6
q
0.6
q
q
q
q
q
0.4
0.4
q
q
q
0.2
q
0.2
q
q
q
0.0
q
q
ects
q
q
Political Information
Political Interest
ed Eff
0.0
q
2.5
q
Fix
q
q
-0.5
2.0
q
q
q
q
q
q
-1.0
1.5
-1.5
q
q
q
q
q
q
q
q
q
1.0
0
1
2
0
1
2
Number of Parents owning at least 1 TV & Radio Station
& a Newspaper (NEWSPAPERTVPARENTS)
Figure 11: Number of Parents Owning TV and Radio Stations and a Newspaper by Fixed
Effects
31

0.2
0.4
0.6
0.8
1.0
Can Place Obama, Health Care
Validated Turnout
q
q
q
q
q
q
q
q
q
q q
0.8
q
q
0.8
q
q
q
q
q
q
q
q
q
q
q
qq
qq
q
q
q
q
qq q
q
q
q q
q
q
q
q
q
q
q
q
q qq qq
q
q
q
q
q
q
qq qq q
q
q
q
q
q
qq
q q
q q q q qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
qq q qq
qq
qq q q q
q
q
q
q
q
q q
q
q
q
q q
q q
qq q
q
q
q q
q
q q
qq q
q
q
q
q
q
q
0.6
q q
q
q q
q
q q q
q
q
q
q
q q q
q q qq
q
q
qq
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
qq
q
q
q q
q q q
q q
q
0.6
q
q q
q
q
q
q
q
q
qqqq
qq
q
q q
q
q
q
qq qq
q
q
q
q
q
q
q qq
q q
q
q
q
q
q
q q
q
q
q
q
q
q
q q q
q
q q
q
qq
q
q q
q
q
q
qq
q q
q
q q
q
q
q
q
qq q
q q
q
q q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q q q q
q
q
q
q
q
qq
q q
q
q
q
q
qq
q
0.4
q
q
q
q
q
qq
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
0.4
q
q q
q
q
q
q
q
q
q q
q
q
q q
qq
q
q
q
0.2
q
q
q
q
0.2
q
q
q
q
0.0
q
q
q
q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
ed Eff
q q q
q
q
q
q q
q
q qqq q
q
q
q
q
qqq
q
q
-0.5
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q
q
q
qq
q qq
q
q
q
q
qq q q
q
qq
q
q
q
Fix
q
q
q
q q
q qq
q q
q
q q q
q
q q
q
q
q
q
q
2.0
q
q
q
q
q
q
q
q
q
qq
q
q
q
q q
q
q
q
q q
q
q
q q q q
q
qq
q
q q
q
q q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q q
q
q
q
q
qqqq
q
q
q
q q
qq
q
q
q q q q q
q
q
q
q
q
q
q
q
q
q
q q q
q q q q
q
q
q
q
q q q q
q
q
q q
q
qqq
q
q
q
q q
qq
q
q qq
q
q
q
q q
q q
q
q
q q q
qq
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q q
q
q q
q
q
q q
q
q
q q
q
qq
q
q q
q q q
q q
q
q q
q
q q
q
q
qqq
q
q
q
q
q
q
q
q
q
q
q
q
q qq q
q
q
-1.0
q
q
qq
q
q
q
q q q
q q
q q
qq q
q
qq
q
q q q
q
q
q
q
q
q
q
q
q
q q q q
q
q
1.5
q
q
q
q
q
q
q
q
q
q
q
q
q
q qq
q
q
q
q
q
q
q
-1.5
q
q
q
q
q
1.0
0.2
0.4
0.6
0.8
1.0
Fraction of Households that Subscribe to 200Kbs Internet Service
(BROADBAND200PCT)
Figure 12: Fraction of Households that Subscribe to 200 KBS Internet Service by Fixed
Effects
32

4
6
8
10
12
Can Place Obama, Health Care
Validated Turnout
q
q
q
q
q
q
q
q
q
q
0.8
q
q
0.8
q
q q q
q
q
q
q
q
q
q
q
q
q
q
q qqq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q q
q
q q q
q qqq
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q q q
q q
q
q
qq
q
q
q
q q q
qq qqqq
q
q
q
q
q
q
q
q
q q
q
q q qq
q
qq
q q
q
q q
q
qq q q
q
qq q
q
q q
q
q
q
q q
q
q
q
q
q
q
q q qq
q
q
q
q q q
0.6
q q
q
q
q
q
q q q
qq
qq q
q
q
q
q
q q q q
q
q
q qq
q
q q q
q q
q
qq qq
q
0.6
q q q
qq q
q q
q
q
q q q
qq q
q
q
q
q
q
qq
q
q
q q
q q q
q
q
q q
q
q q
q
q
qq
q
q q q
q
q
q q
q
q q
q
q
q q
q
q
q
q q q
q
q qq
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q q q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q q
q q
qq
q
q
q
q q
q
q
q
q
q
q q
q q q
q
q
q q
q
q
q
q
q
q
q
q q
0.4
q
q
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
0.4
q
q
q q
q
q
q
q
q
q
q
q qq
q
q
q
q
q
q
0.2
q
q
q
q
0.2
q
q
q
q
0.0
q
q
q
q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q q
q
q q
ed Eff
-0.5
q
q q q
q q
q
q
q
q
qq
q
q
q
qqqq
q
q
q
q
q
q q
qq
q
q qq
q
q
q q
q
q
q
q
q
q
q
q q qq
q q
q q
q
q
q
q
q
q
q
q q
q
q q q
q
q
Fix
q
q
qq q
q
qq
q
q
qq
qq q
q
q q
q q
q
q
q
2.0
q
q
q q
q
q
q
q
q
q
q q q
q
q
q q
q
q q
q
q qqq
q
qqq q
q q
q
q
q
q
q
q
q
q
q
q qq
q
qqq
qq q
q
q
q
qqq q q
qq
q
q
q
q
qq
q
q
q
q
q
q
q
q q q q
q
q
q
q
q
q
q q
q
q
q
q
qq q
q
q q
qq
qq q q
q
q
q
q q
q q q
q
q
q
qq
q
q
q q
q q
qqqq
q
q
q
q q
q q q
q q
qq
q q qq
q
-1.0
q
qq
q
q
q
q
q
q
q
q
q
q
qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
qq
qq
q
q
q
q
q
q q
qq
q
q
q
q
q
q
q
q
q qq
q
q
q q
qqq
q
q
q
qq q
q
q
q
q
q
qq
q q q
q
q
q
q
q
qq
q
q
q
q
qqqq
q
q
q qq
1.5
q
q
q
q
q
q
q q
q
q
q
q
q
q
-1.5
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
1.0
4
6
8
10
12
Total Presidential Advertising GRPs
Purchased from Labor Day to Election Day (log scale)
Figure 13: Log of Total Presidential Campaign Advertising GRPs by Fixed Effects, 2008
33

0
5
10
15
20
25
Can Place Obama, Health Care
Validated Turnout
q
q q
q
q q
q
q
q
q q
0.8
q q
0.8
q
q
q q
q
q q
q
q q
q
q q
q q q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q
q
q
q
q q
q q q
q
q
q
q
q
q q
q
q
q
q
q q
q
q q
q q q
q q
q
q q q
q q
q q
q
q
q
q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q q
q q
q q
q
q
q q
q
q
q
q
q
q q
q
q
q
q q q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q q
0.6
q
q
q
q q
q
q
q
q q q
q
q
q
q
q
q
q
q
q q
q
q q
q q q
q
q q q q
q q
q
q q
q
q q
q
0.6
q
q q
q
q
q
q
q
q
q q
q q q q q
q
q
q q
q q
q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q
q
q q
q
q
q q
q q q
q
q q
q
q
q
q q
q
q
q q
q q
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q
q
q
q
q q
q
q
q q q
q
q
q
q
q
q
q
q
q
q q
q q
q q
q q q q
q
q q
q
q
q
q
q
q
0.4
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.4
q
q
q
q
q
q
q q
q
q
q q
q
q
q
q
q
q
q
q
q q
0.2
q
q
q
q
0.2
q
q
q
0.0
q
q
q
q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q q
q
q
q q q q
q
q
q
q
q q
q
q
q q
q
ed Eff
q
-0.5
q q q
q q q
q
q q
q
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q q q
q
q q
q
q
q q q q
q
q
q
Fix
q
q
q
q
q
q q q q q
q
q
q
q q
q q
q
q
q q
q
q q
q
q
q
q
q
q q q
q
q
2.0
q
q q
q q
q q q
q q q
q
q
q
q q q
q
q
q
q
q
q q
q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q q q
q
q
q
q
q
q
q
q q
q
q
q
q q q q
q
q
q
q
q q
q
q
q
q
q q q
q
q
q q q
q
q q
q
q
q
q q
q q
q
q
q q
q
q
q
q
q q
q
q
q
q
q
q q
q
q
q
q q q q q q q
q q
q
q q
q
q q
q
q
q
q
q
q q
q q
q
-1.0
q
q q
q
q
q
q q q q q
q q
q
q
q
q
q
q q
q
q
q
q q
q q
q
q
q
q
q q
q q
q
q q
q q q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q
q
q
q
q
q q
q
q
q
q
q
q q
q
q
q
q q q
q
q
q
q
1.5
q
q
q
q
q
q q
q
q
q
q
q
q
-1.5
q q
q
q
q
q q
q
q
q
q
q
q
q
1.0
0
5
10
15
20
25
Number of Parents owning a News/Talk Radio Station (RADIONTPARENTS)
Figure 14: Number of Radio Parents with News/Talk Format by Fixed Effects
34

10
12
14
16
Can Place Obama, Health Care
Validated Turnout
q
q q
q
q
q
q
qq
q
q
0.8
q
q
0.8
q q
q qq
q
q
q qq
q
q
q
q
q
q q q q q q
q
q
q
q
q q
q
q
q
q
q q
q
q
q
q q q
q
q
q
q
q q
q
q
q q
q
q q q q
q
q
q
q
q
q q q
q
q q
q
qq
q
q qq
q
q q
q
q q
q qqq
q
q
q
q
q
q q
q
q
q
q
q
q
q
qq
q
q
q
q q q q
q
q
q
q
q
q q
q
q
q q
q q
q
q
q
q
q
q q q
q
q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.6
q
q
q
q
q q
q
q q
q q q
q q
q q
q q
q
q qq
q q
qq
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q
q
q
q q
q
0.6
q
q
q q
q
q
q
q
q
q q
q
q
q
q
q q
q q
qq
q
q
q
q q q qq
q
q
q
q
q
q
q
q
q q
q q
q
qq
q
q
q
q
q
q
q
q
q q
q
q q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q q q
q
q q
q
q q
q
q
q
q
q
q
q
q
q
q
qqq
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q qq
q
q
q
qqq q q q
q
q
q
q
qq
q
qq
qq
q
q
q
q
0.4
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.4
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
0.2
q
q
q
q
0.2
q
q
q
q
0.0
q
q
q q
Political Information
Political Interest
0.0
q
2.5
q
ects from Regression
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q
q
qq
q
ed Eff
q
-0.5
q
q
q
q
q qq
q q
q
q
q
q qq
q qq
qqqq
q q
q
q
q
q
q
q
qq
q
q
q q q
q
q
q
q q
q
q
q q
q
q q
q
q
q
q
q
q
Fix
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
qq
qq q
q
q
q
qq
q
q
q
q q
qq
qq
q
2.0
q
q
q
q
q q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q q q
q
q
q
q qq q q q
q
q
q q
q q
q
q
q
q
qq
qq
q qq
q
q
q
qq q q q
q q
q
q
q
q
q
q
q
q
q
q
q q qq
q qq q q
q
q
q
q
q q q
qq
q
q
q
q
q
qq
q q q q
q q
q
q
q
q
q q q
q
q
q
q
q q
q
q
q
q q
q
q
qqq
q
q q
q
q
q q q
q q
q
q
q
q
q
q
q
q
q
q
q
q q
q
-1.0
q
q
q
q
qq q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q q
q
q
q
q q q
q
q
q
q
q
q
q
q q q
q
q
q q
q
q
q
q
q
q
q
q
q
q q q
q q q
q q
q q
q
q q
q
qq
q
q
q
q
q
q
q
q
q
qq
q
q
q q
q
q
q
q q
q
q
1.5
q
q
q
q
q
q
q
q
q q
q
q
q
-1.5
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
1.0
10
12
14
16
Population over the age of 18 (POP18PLUS, log scale)
Figure 15: Log of Population over 18 by Fixed Effects
35

We also examined the relationship between the number of parents owning radio stations
with a news or talk format and the DMA region offsets as well as the size of the adult
population and the offsets. We see the familiar flat line for the news/talk radio relationship,
but a slight hint of a downward trending relationship for adult population size.
Thus the non-ownership measures (Internet penetration, advertising, and adult popula-
tion) gave us the strongest pattern, although all the relationships were weak in general.
These results led us to analyze the relationship among all of these DMA region level
factors. We suspected that many of the market ownership indicators were highly correlated
with size of adult population and in fact, by definition, some of these measures must be highly
correlated with independent voices because they are transformations of the voices count
based on FCC rules and regulations. We present the relationships among these indicators in
Figure 16.
You can see the correlations in Figure 16 in two ways: by examining the pearson correla-
tions in the upper triangle or by focusing on the loess lines in the lower triangle. Either way,
the dependencies are obvious. Adult population, in particular, is correlated with news/talk
radio parents at .76, with radio and TV voices at .84, respectively. In contrast, Internet pen-
etration and presidential advertising do not seem to be highly correlated with the market
ownership structure. This correlation among ownership factors is likely to make it difficult,
even with the heightened power of our 20,000 respondent survey, to find strong effects.
We turn our attention now to a set of multilevel and hierarchical models that estimate
the overall effect of ownership structure on civic engagement, even though we did not find
much of the variation across DMA regions was explained by these indicators.
4
Bayesian Analysis of Hierarchical and Multilevel Mod-
els
Descriptions of the specific multilevel models we fit appear below.
Generally, we will be fitting models of the sort
E(yi|xi) = xi + j(i)
(3)
V (yi|xi) = 2,
(4)
where i indexes observations and j indexes DMA regions; i.e., respondent i is located in DMA
region j(i). Regression, logistic regression and ordinal logistic regression are the particular
cases we will encounter below.
36

0
40
80
120
0
1
2
3
4
5
0.4
0.8
4
6
8
10
20
15
TV Voices
0.83
0.67
0.65
0.30
0.24
0.75
0.30
0.84
10
5
120
q
q
100
q
q
q qq
q
q
80
q
q q q
60
q
q
q
q q q
q
q
q
Radio Voices
q
q
qq
0.67
0.59
0.31
0.22
0.86
0.36
0.84
q
q
q q
qq
q q
q
q
q
q
q q q
q
40
q q
q
q
q q
q
q
q
q
q
q
q
q
qq
q q
q q
qq
qq q
q
q
q q
q
q
q
q q
q
q q
qq
qq
q
q
q
20
q
q q
q q
q
qq q q
q
q q
q
q
q
qq
q
q q q
q q
q
qqq
qq
q
q q
qq
q
q
q
qq
qq
q
q
q q
qq
q
q q
q
q
q
q q
qq
q
q
0
q
q
q q
q
q
q
q
q
q
q
q
q
4
q q
q
q
q
q q q
q
q
q
q q
qq q q
3
MULTICOM-
q q
q q
q q
q
q
q q q
q
qqq
q q
q
q
q q
qq q
q q q q
0.55
0.25
0.25
0.61
0.26
2
0.67
TVPARENTS
q qq
q
q
q
q
q
q
q
q q
q
qq
q
q
q
q
q q
q
qq
q
q q
q
q q
q
q q
q q
q
q q
q q
q
q qq
q
1
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
qq
qq q
q
q
q
q q
q
q q
qq
qq q
q
q
q
q
q
q q
q q
q q
q q
0
5
q
q
q
4
q
q
q
q q q
q
q
3
qq
q
q q
q
q
q
q
COMRADIO-
0.40
0.22
0.53
0.09
0.55
2
q q
q
q qq
q q
q q
q q q
q
qq q q
q q
q q
qq q
q
q
q
q
COMTVPARENTS
1
q q
q
q
q q
q q
q q
q q
q q
q
qq q
q q
qq
q q
q
q
q
q
qq q
q q
qq q
q
q
q
q
q
0
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q q
q
q
q
q q
q
q q
qq
qq
qq
q
q q
q
q q
q
q q
q
qq
qq
qqq
qq
q q
q
q q q
q
q
q
q
q
q
q
q
q
q
2.0
1.5
NEWSPAPER-
q q
q q
q
q
q q
q
q q
q q
q
q
q q
q q
q
q q
q q
q
q q
q q q q
q q
q
q
q
q
q
q
q
q
q
q
0.098
0.26
0.069
1.0
0.32
TVPARENTS
0.5
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q q
q
q
qq
qq q
q
q
q
q q
q
q q
qq
qq q
q
q
q
q
q
q q
q
qq
qq
q
q q
q
q q
qq
q q
q
q q
q q
q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
0.0
q
q
q
q
q
1.0
0.8
q q q
q
q
q
q
qq
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
BROADBAND-
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q q
q
q
q
q q q
qq
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q
q
q
q
q
q
q
q
q
q q q
q
qqq
q qq
q
qq
q q q
q
q
q
q
q
q
q
q
q
q
q
0.27
0.11
0.33
0.6
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
qq qq
q
q
q
q q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q
q
q
q
q
q
q
q
qq
qq
qq qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q
qq qq
q
q
q
q
q
q
q
q
q
q
q
200PCT
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q q
q
q
q
q q
q q
q
q
q
q
q
q
q
qq
q
q
q
q
qq
q q
q
q
q q
q
q
q q
q
qq q
q qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q qq
q
q
q
q q
q
q
q
q
q
q
q
q
q
q q qq
q q
q q q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
qq
q
q q
q
q
q
q
q
q
q
q q
q
q q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q qqq
q
q
q
q q
q
q
qq q
q q q
q qq
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
0.4
q
q
q
q
q
qq
q q
qq q
q
q
q
q
q
q
qq
qq q
q
qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
25
q
q
q
q
q
q
qq
q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
q q q
20
q q
q
q
q
q
q
q
q
q
q
q
qq
qq
q
q
q
q
q
RADIONT-
q
q
qq
q
q
q
q
q
q
q
q q
q
15
q
q
q
q
q
q
0.33
0.76
q
q
q
q qq
q
q
q
q
q
q
q
q
qq
q qq
q q q
q qq
q
qqq
qq
q q q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q q
q q
q
q
PARENTS
q
qq q
q
qq
q
q qq
q q
q
q
q
q
q
q
q
q
q
q
qq
q
q
10
q
q q
q
q
q
q
q
q qq
q q
qq
q
q qq
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q q
q
q
q q
qq q qq
q
q
q
q q
qq q
q
q
qq
qq q
q
q
q
q
q
q
q
q
q
q
q
qq qq
q q
q qq
q
q
q
q
q
q
q
qq
q
q
q
q q
q
q
q
q q
q q
qq
q
q q
q
q
q
q
q
q
q q
qq
qq
q q
q
q
q q
q qq
qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q
q q
q qq q
q q
q q
qq
q
q q
q
q
q
q
q
5
q q
q
q
q
qq
q q
q
q q
q
q
qq
q
q
qqq
q q
q
qqq
q q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q q
qq
q q
q
q
q
q
q q
q
q
q q
q q
q q
q
q
q q
q
q q
qq
q
q
q q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q qq
qq q
q
q qq
q
q q
q q
q q
q q
q
q
q q
q
q
q
qq
q
q
q
q
q
q q
q
qq
q
qq
q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q qq q
q
qqq
q
q
q
q
q q q
q q q
q
q
q
q
q q
q q
q q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q q
qqqqqq
q
q
q
q
0
12
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q q
q
q
q
qq
q
q q
q
q
q
q
q
qq
q
q
q q
q
q
q
q
q
qqq
q
q q
qq
q
q
q
q
q
q
q
q
q
q
q
q q
q q
qqq
qq q q
q
q q
q
qq q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
qq qq q
q q q
q q
q q q
q
q
q
q
q q
q qq q
q
q q q
q
q q
qq q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q q
q
q
q qqq q
qq q
q
q q
q
q q q
q q
qq
qq q
q
q q
q
q
q
q
q
q
qq
q
q qq
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q q
q qq
q
q
q q q
q qq
q
q
q q
q
q q
q
q q
q
q qq
q q
q q
q q q
q q
qqq
q
q
q
q
q
q
q
q q
q
q
q q
q q qq
q
q
q
q q qq qq q
qq
q
q
q
q qq
q q
q q
q q q
qq
q
q
q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q qq
q q
q
q qq
q
q
qq
q q q
q
q
q
q qq
q q
q q
q
qq q
q
q
q
q
q
qq
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q
q q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q q q
q q
qq q
q
q
q
q
q
q
qq q
q
q q
q q
q
q q q
q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q q
q q qq
qq
q
q q
q
qq q
q q
q
q q
q q
q q
q
q
q
q
q
q
q
10
q
q qq
q
q
q
q
q
q
q
q
q q
q q q
q q
q
q q
q
q
q
qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qqqq q q
q
q q
q
q
qq
q q
q q q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
qq
q
q
q qq
qq
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q qq
q
q
q qq q
qq
q
q
q
q
q
q
q
qq
q qq
q
q
q
q
q
q
q
qq
q
q q q
q
q q
q q
q
q q
q
q
q
q
q
qq q
q q
q q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q qq
q
q
qq
q
q qq
qq
q q
q
q
q q
q
q
q
q
q
q
q
q
q qq
q
q
q qqq q
q
q q q
q
q
q
q q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q q
q q q
q
q q q
q
Total
q
q q
q
q
q
q q q
q
q
q
q q
qq
q
q
q q
q qq
q
q q
q q
q
q
q
q
q
q
q
q
q
qq q
q qq q
8
qq q
q
qq
q
q
q
q
q q
q
q q
q
q q
q
qq
q
q
q
qq
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
Presidential
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
0.54
q
q
q
q
q
q
q
q
qq
GRPs
6
q q
q q
q
q
q
q
q
q
q q
(logged)
4
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
16
qq
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q qq
q
q
q
q
q
q
q
q
qq q
q
q
q
q
q
q
q
q
q
q
q qq q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q q
q
q
q q q
q
q
q
q
q
qqqq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q q
q
qq
q
q
q qq
q q
q
q
q
q
q
q
q
q
q
q
q q
q q
q q
q
q
q
q
q q
q q
q
q
q q qq
q
q q
q qq
q
q
q
q q
q
q
q
q
q
q
qq qqq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q q
q q q
q
q
qq
q
q
q
q q
q q
q q q
q
q
q
q q
q
q q
q
q q
q q
q q
q qq
q q
q
q
q
q
q
q
q
q
q
q
q
q q
q qq q
qq q
q q q
q
qq
q q
q
q q q
q q
q q
q
q
q
q q
q q
q
q
q
qq
q
q
q
q q
qq q
q q
q
qqqq q
q
q
q
q
q
q
q
q q q
q
q
q
q q
q q q
q qq qq
Popln
14
q q
q q
q
qq
q
q
q
q
q qq q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq q
q q q
q q
q q
q
qqq q
q q
q
q qq
q
q q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
q
qq
qq
q q
q
q
q
q q
q
q
q q
q
qq
q
q
q q
q q q q
q
qq q
q q q
q
q
q q
q
q
q q q
q q
q
q
q
q
q
q
q q
q
q
q q
q q
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q
q
qq q
q qq
q q q
q q q
q
q q
q qq q
q
q
q q
q q
q
q
q
q
q
q
q
q
q q
q
q
q q
q
q
q q q
q
q
q q
q
q q
q q
q
q
q
q
q
q
q q
q
q q
q
q
qq
q
q
q
q q
q q
q q
q q
qq
q
q
q q
q
q
q q
q
q q
q
q q
qq
q
q q
q
q
q
q q
q qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
qq q q
q
q
q
q
qq q
q
q
qq
q q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
qq
q
q q q
q
q
q q
q
q q
q q q
q
q
q
q
q
q
q q q
qq q
q
q
q q
qq
q
q
q
qq
q
q
q q
q
q
q
q
q
q
q
q q
q qq
q q
q q
q
q
q
q qq
q
q
q q
q q
q
q q
q
q
q
qq
q q
q q
qq
q q
q
q
q
q
q
q
q
q
q
q
qq q
qq q
q
q q
q
q
q
q q
q
q
q q
q
q q
q q
q
q
18+
q
q
q
q
qq
q
q
q
q
q
q
q
q
q q
q
q
q
q
q
q qq
q
q
qq
qqq
qq q
q
q
q
q
q
q
q
q
qq
q q
qq
q
qq
q
q
q q
q q
q q
q
q
q
q
q
q q
q q
q
qqq
q
q
q
q
q
q
q
q
q
q
q qq q
qq q
q q
q
q q
q q
qq q
q
q q
q
q
q
q
q
q
q
q q
qq q
q
q q
q
qq
q qq
q
q
q q
q
q q
q
q
q
q
q q
q
q
q q
q q
q
q
q
q
q
q
q
q
q
q q
q q
q
q
q
q
q
q
q q
q
q
q
q
q
q
12
q
q
q
q
q
q
q
q
qq q
q
q q
q
q
q
q
q
q
q q
q
q q
q
qq
q
q
q
q
q
q q
q
q
q
q
q
q
q
q
q
q q
q
q
q
q q
q
qq
q
q q
q
q q
q
q
q
q
q qq
(logged)
q
q
q
q
q
q
q
q
q
q q
q
q
qq
q
q
qq
q
q
q
q
q
q
q
q
q
q
qq
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
10
q
q
q
q
q
q
q
q
5
10
15
20
0
1
2
3
4
0.0
1.0
2.0
0
5
15
25
10
12
14
16
Figure 16: Scatterplot Matrix of Level Two Variables
37

The models we fit here have a hierarchical component sometimes taking the simple form
j N (0, 2 )
(5)

but of greater interest are multilevel models of the sort
j N (i, 2).
(6)

The densities used in the hierarchical component need not be normal; here we focus on
normal densities since the j R and the normal is a convenient choice for when modeling
continuous-valued quantities without restrictions on their support; note that it is straight-
forward to introduce covariates into the model for the mean of the normal via the familiar
linear, additive regression model.
We adopt a Bayesian approach to inference for these models; Bayesian analysis of hi-
erarchical and multilevel models is now well-established as one of the preferred methods
for dealing with what classical statistics and econometrics would refer to as "random coef-
ficients" or "varying coefficients" model. Other branches of the social sciences sometimes
refer to these models of this sort as "mixed" models. Discussion and further details appear
in Gelman and Hill (2007) or Jackman (2009, Ch 7).
In the Bayesian approach to inference, we seek to characterize the posterior density of
the model parameters
m
R . By Bayes Rule, the posterior density of is
p(|data) p(data|)p()
where p(data|) is the likelihood function for the data and p() is the prior density of
. Note immediately that with a "vague" or approximately locally-constant prior density,
the prior is absorbed into the constant of proportionality in Bayes Rule, and the posterior
density has the same shape as the likelihood function. Moreover, in large samples, posterior
densities for many parameters are approximately normal; the symmetry of the normal in turn
ensures that the mean of a (marginal) posterior density will often be close to the maximum
likelihood estimate. For this reason, many practitioners have adopted Bayesian approaches
out of convenience -- especially in light of the computational details discussed in the next
paragraph -- as a means of generating MLEs "by any other means".
Hierarchical models give rise to lots of parameters; e.g., in this case we have over 200
j DMA region-specific parameters. Modern approaches to characterizing high-dimensional
posterior densities make use of the Gibbs sampler, an algorithm that sample successively
from the lower dimensional conditional densities that together constitute some joint density
38

of interest (e.g., a high-dimensional posterior density). Details of the mechanics of the Gibbs
sampler appear in many places: see Robert and Casella (2004) or Jackman (2009, Chapter
5).
Of great practical significance is that Gibbs sampling for hierarchical models (and other
Bayesian models) is easily implemented using freely available software; e.g., JAGS (Plummer
2010) or OpenBUGS/WinBUGS (Spiegelhalter et al. 2003). In fact, it is no exaggeration to
observe that the Gibbs sampler and its implementation in user-customizable programs lie
behind the surge of interest and utilization of hierarchical models in the last decade or two.
4.1
Interest in Politics
We model this variable with an ordinal logistic regression model, utilziing the micro-
level covariates introduced earlier. Let the ordinal self-report of level of political interest be
yi {1, 2, 3} and let xi be a vector of covariates, i = 1, . . . , n. The ordindal response model
presumes that the observed responses are an interval-censored version of a latent continuous
variable: i.e.,
yi = 1

y
i
1
yi = 2

1 < y
i
2
yi = 3

y >
i
2
where y = x
i
i + j(i) + i is a regression model relating the covariates to the latent response,
with a vector of unknown parameters and i Logistic giving an ordinal logit model. The
term j(i) is unobserved term specific to the j-th DMA region; the notation j(i) simply
denotes the index of the DMA region in which respondent i is located.
The thresholds 1 < 2 are also unobserved parameters. A common parameterization
(which we adopt here) is to omit an intercept term from xi, which means both thresholds
can be estimated.
With these definitions, we can define the likelihood for the model as follows. The prob-
abilities of each of the three outcomes are:
Pr[yi = 1] = Pr[y
i
1]
= Pr[xi + j(i) + i] = Pr[ i 1 - xi - j(i)] = F [1 - xi - j(i)]
Pr[yi = 2] = Pr[1 < y
i
2]
= F (2 - xi - j(i)) - F (1 - xi - j(i))
Pr[yi = 3] = 1 - F (2 - xi - j(i))
39

The likelihood contribution of each observation is just the probability of the particular
outcome we observe. We assume conditional independence across survey respondents, such
that the likelihood is just the product of the particular respondent-specific probabilities.
In our Bayesian approach to inference, proper priors complete the model specification.
Specifically, we assume vague, independent, normal priors over the elements of ; i.e., k
N (0, 102), with k indexing the elements of . A simple hierarchical model for the j is
j N (0, 2), with a relatively innocuous prior on , e.g., Unif(0, 5); at the upper
bound of 5 on the prior for a 95% credible interval for any one of the j is [-10, 10], which
spans a huge range of values on the logit scale of the latent y. The multilevel version of
the model posits that the j are a linear function of DMA region-level covariates, j plus
random error: i.e., j N ( , 2), with a proper prior on the unknown hyperparameters
j
completing the specification.
The threshold parameters are assigned vague normal priors subject to the ordering con-
straint, i.e., k N (0, 102), k = 1, 2 subject to the constraint 1 < 2.
4.1.1
Multilevel model
The multilevel model utilizes DMA region-level predictors, j in the hierarchical component
of the model j N ( , 2). We fit a multilevel version of an ordinal logistic model. We
i
present the point estimates of the parameters (means of marginal posterior densities, as
estimated from the Gibbs sampler output) and standard deviations in Table 5.
The pattern that stands out in Table 5 is the dominance of the micro-level attributes
in explaining political interest levels relative to the DMA region-level attributes. Greater
levels of interest are observed from older, wealthier, white men with college degrees, for
example. Each of the individual-level indicators is estimated precisely and in the expected
magnitude and direction. The DMA region-level indicators do not show the same predictive
qualities. Only one of the DMA region-level estimates is twice the size (or even bigger than)
its standard deviation and that is the measure of advertising GRPs. DMA regions with
more advertising, on average, have lower average levels of political interest, all else equal.
This is not necessarily surprising, once we remember that the design of this project is not
a randomized experiment assigning advertising GRPS. It is possible that candidates and
parties advertise in places where political interest and engagement are low in order to boost
turnout and interest in the election and specifically in their candidacy.
Examining the coefficients demonstrates differences in the magnitudes of the effects be-
tween the micro-level and DMA region-level measures. The gender, age, and income effects
40

-1.0
-0.5
0.0
0.5
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
-1.0
-0.5
0.0
0.5

Figure 17: Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level ordinal logistic regression model for respondent self-reported levels of interest in politics.
The estimates have been sorted from low to high and by construction have zero mean.
41

Est
StdDev
Female
-0.705
0.031
Hispanic
-0.219
0.071
Black
-0.205
0.061
College
0.496
0.038
Age (years/10)
0.196
0.011
Income (1-14)
0.085
0.005
1
-1.212
0.078
2
1.018
0.078
# TV Voices
0.004
0.017
# Radio Voices
0.001
0.003
# Parents with 2+ TV Stations
0.021
0.037
# Parents with TV and Radio
-0.020
0.045
# Parents with TV, Radio and Newspaper
0.042
0.083
Broadband 200bps (%)
0.104
0.319
# Parents with News/Talk Radio Station
-0.012
0.011
Total Presidential Ad GRPs (logged)
-0.060
0.034
Population 18+ (logged)
0.067
0.064

0.280
0.031
Table 5: Estimates of Multilevel Ordinal Logistics Regression Model of Respondent Self-
Reported Levels of Political Interest. Entries above the line are estimates of micro-level
parameters, ; entries below the line are estimates of DMA-level parameters, . are
threshold parameters in the ordinal logistic regression model. is the standard deviation of
the error component at the DMA level of the model.
are quite large compared to the DMA region-level effects, such as TV voices. The gender
effect is .71 (increase in interest for men) while the TV voices effect, albeit imprecisely es-
timated, is .004. There would have to be an enormous number of television stations in the
market in order to have the same kind of effect that we observe on political interest driven
by gender.
In terms of political interest, characteristics of the ownership structure of the DMA region
do not appear to affect people in an appreciable way. In order to visualize the variation we are
trying to explain, we take the model's output and generate a plot of the DMA region-specific
terms, the off-sets (j). We present these data in Figure 17.
By construction, the estimates of the DMA region-specific off-sets have mean zero. The
vertical line down the center of the graph represents this point (zero). Each of the dots on
the graph are one DMA region, and the lines extending in each direction represent the 95%
credible intervals around the estimate. Some of the intervals are smaller than others. This is
a function of the amount of data that we have in any particular DMA region. As discussed
42

earlier, CCAP was not designed to be representative at the DMA region level, thus we have
respondents in each DMA region that are roughly proportional to the size of the DMA region
in the population as a whole.
As the figure makes clear, despite the variation in DMA regions in terms of political
interest, net the other factors in the model, most of the estimates of (j) have credible
intervals that cross zero meaning there may actually be no difference in the DMA regions
in terms of political interest. This makes it very hard to "explain" the differences with DMA
region-level factors. When the differences are not clear, there is not much to be explained.
The range of these off-sets, from roughly -.5 to .5 suggests that from the DMA region with
the least average political interest to the DMA region with the highest average level of
interest, the net effect is about the same as it is for gender (at the micro-level). Of course,
the important point to recognize from these analyses is that none of the DMA region-level
ownership variables explain the pattern of variation we see in Figure 17.
4.2
Overall Political Knowledge/Information
Respondent levels of overall political knowledge or information is estimated using the
IRT analysis described earlier in Section 3.1.21 The resulting measure is continuous (nor-
malized to have mean zero and standard deviation one); we use a normal regression model
for this variable at level 1, greatly simplifying the computation of the posterior density of
the parameters in the multi-level model. For the multi-level model, we model the DMA
region-specific terms as j N ( , 2), with vague-yet-proper priors on . In this specifica-
j
tion 2 is an unknown "within" DMA region error variance, presumed to be constant across
DMA regions. In the multi-level model, once DMA region-level covariates are introduced,
2 becomes an error variance, tapping how much variation in the j is unaccounted for after
we introduce the DMA region-level regressors j.
4.2.1
Multilevel model
As with political interest, we present the results of the multilevel model first in a table
and then in visual form showing the DMA region level offsets we are trying to explain. The
results for overall political knowledge are presented in Table 6.
The pattern that emerges from the investigation of overall political knowledge has simi-
larities but also differences when compared to the model of political interest levels above. For
example, once again, the micro-level predictors are robust and strong predictors of overall
21Further documentation on the construction of the knowledge scale is also available in the replication file
for this report, available from the authors or from the FCC.
43

Est
StdDev
Intercept
-0.674
0.030
Female
-0.424
0.013
Hispanic
-0.187
0.029
Black
-0.271
0.025
College
0.375
0.014
Age (years/10)
0.086
0.004
Income (1-14)
0.045
0.002

0.908
0.005
# TV Voices
0.012
0.009
# Radio Voices
0.002
0.002
# Parents with 2+ TV Stations
-0.008
0.019
# Parents with TV and Radio
-0.010
0.020
# Parents with TV, Radio and Newspaper
-0.070
0.038
Broadband 200bps (%)
0.238
0.120
# Parents with News/Talk Radio Station
-0.004
0.006
Total Presidential Ad GRPs (logged)
0.017
0.013
Population 18+ (logged)
-0.041
0.036

0.122
0.012
Table 6: Estimates of Multilevel Regression Model of Political Information Scores Entries
above the line are estimates of micro-level parameters, ; entries below the line are estimates
of DMA-level parameters, . is the standard deviation of the error component at the micro,
individual level of the model; is the standard deviation of the error component at the the
DMA level of the model.
political knowledge -- and, they work in the same directions that they did for political in-
terest. These individual-level characteristics drive many different types of political behavior
and we are not surprised to see them working so heartily in these analyses.
Where Table 6 differs from the previous estimation for interest, however, is in the way
the DMA region-level covariates affect the dependent variable. To be clear, none of these
effects are particularly precise, and certainly most are small, but there are hints in these data
that factors such as the number of independent TV voices, the number of parents with a
TV station, radio station, and newspaper, advertising GRPs, and Internet penetration may
be slightly related to overall levels of political knowledge, all else equal. We stress, however,
that these results are not particularly precise (with the exception of Internet penetration the
t - statistics hover around 1.3). Both the number of voices and the ability to have a parent
entity own across the three types of outlets (TV, radio, and newspaper) are correlated with
size of the DMA region and population. This model controls for population, so these effects
are net of the role of population.
44

-0.4
-0.2
0.0
0.2
0.4
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
-0.4
-0.2
0.0
0.2
0.4

Figure 18: Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level regression model for political information scores. The estimates have been sorted from
low to high and by construction have zero mean.
45

Of particular interest in Table 6 is the estimate on Internet penetration. This effect is
large relative to any other DMA region-level factor we have seen thus far. In fact, the effect
of Internet penetration -- moving from no penetration to full penetration -- is nearly the
size of the effect for having a college education. The mean level of broadband household
penetration in these data is 63%, thus the average effect of broadband is about the same as
the effect of being a non-hispanic white.
We present the DMA region-level offsets in the same fashion as above. Once again we
note that the 95% credible intervals mainly cover zero -- indicating that we cannot be sure
there is any variation at the DMA region level on overall political knowledge. In terms of
the spread of the off-sets, the range covers nearly one-half a unit on the information scale
-- this scale ranges from -2.3 to 1.5 by construction. The distribution of the off-sets covers
roughly 10% of the information scale's range. These two facts, that the estimates' confidence
intervals cover zero and that there is not much spread to the variation, make it very difficult
for us to uncover much in the way of DMA region-level explanation for these off-sets.
4.3
Ability/Willingness to Place Obama
We consider respondent's ability/willingness to place Obama on the health care issue. We
re-create this variable and compute its marginal distribution: n.b., yi = 1 if the respondent
could locate Obama on the health care policy dimension, yi = 0 if the respondent could not
locate Obama, and we also have a reasonable amount of missing data.
We implement a logistic regression model. More formally, the model is:
yi Bernoulli(i)

ln
i
= xi + j(i)
1 - i
j N (, 2)
N (0, 10 I6)
N (0, 42)
Unif(0, 3)
where i indexes the respondents, j indexes the DMA regions such that respondent i resides
in DMA region j(i), j is an unobserved effect specific to each DMA region, is a vector
of unobserved micro-level parameters, is a "grand mean" or intercept in the mode, and
is the standard deviation of the DMA region-specific terms j. The multilevel version of
this model posits the j as linear function of DMA region-level covariates, j, plus random
variation: i.e., j N ( , 2), with a proper prior on the unknown hyperparameters
j
46

completing the specification.
In our Bayesian approach to inference for this model, the parameters , and are
assigned proper priors. The prior for is a multivariate normal density with large standard

deviations of
10 in each dimension. The intercept parameter is also given a vague normal
prior with mean zero and standard deviation 4 (where "vague" here is defined with respect
to the logit scale); similarly, the prior for the standard deviation of the DMA region effects
is uniform over the interval zero to 3, again allowing the j to span a large portion of the
logit scale, at least a priori. In short, we use priors that are not especially restrictive with
respect to the logistic scale that we use in this model for the binary outcomes.
4.3.1
Multi-Level Model
We add structure to the model by introducing DMA region-level ("level two") covariates,
writing the model as
yi N (xi + j(i), 2)
j N (j, 2)
where j are "level two" or DMA region-level covariates and is a set of unknown parameters.
In the simple model considered below, j is a single covariate, the DMA.tvvoices variable. In
this case the level-two regression model becomes 1 + 2 .
j
We present the result from this multilevel estimation in Table 7. Once again, the demo-
graphics are strong predictors of the ability to place Obama on the health care issue, with
one exception. The indicator for "Black" is not significant in this model, while it has been
in the other models of engagement and is well-known to be a strong predictor of knowledge
and engagement in politics. This, of course, is because of Obama's strong ties to the Black
community and the unprecedented levels of support he received from the Black community
as a candidate.
The lack of effect for "Black" indicator, however, sheds light on exactly the kind of
processes we are trying to explain in this report. We could measure DMA region ownership
factors in as great a detail as possible and gather much more data than we have here --
but at the end of the day, what drives whether Black respondents will place Obama on an
issue scale (which presumably measures their knowledge of his issue position)? Their race,
not any factor of their media environment. Women and Hispanics remain less willing to
place Obama on health care than men and non-Hispanic whites, all else equal, but Black
respondents demonstrate the opposite trend -- being as willing to place Obama as non-Black
47

Est
StdDev
Intercept
0.408
0.096
Female
-0.766
0.047
Hispanic
-0.368
0.093
Black
-0.091
0.086
College
0.794
0.054
Age (years/10)
0.119
0.014
Income (1-14)
0.084
0.007
# TV Voices
0.011
0.023
# Radio Voices
0.005
0.005
# Parents with 2+ TV Stations
-0.027
0.055
# Parents with TV and Radio
-0.035
0.058
# Parents with TV, Radio and Newspaper
-0.112
0.106
Broadband 200bps (%)
0.426
0.340
# Parents with News/Talk Radio Station
0.000
0.016
Total Presidential Ad GRPs (logged)
0.011
0.036
Population 18+ (logged)
-0.025
0.069

0.307
0.035
Table 7: Estimates of Multilevel Model of Ability/Willingness to Place Obama on Health
Care. Entries above the line are estimates of micro-level parameters, ; entries below the
line are estimates of DMA-level parameters, .
respondents. We, of course, cannot be sure, but the best guess as to what may be driving
this effect is Obama's race.
In fact, not a single DMA region-level estimate is notably bigger than its standard de-
viation, with the exception again of Internet penetration in the DMA region, which is still
imprecisely estimated, but has a t - statistic of roughly 1.25 -- and again is large in size.
The effect of Internet penetration in the DMA region is larger for issue-placement than it is
for knowledge, in this case, as it moves across its range, it exceeds the effect of non-Hispanic
ethnicity and delivers roughly half the effect of having a college education.22
Figure 19 presents the familiar lack of variation across DMA regions that we have seen
for both interest and knowledge. Even with the amount of data that we have in CCAP, we
22It is worth pointing out at this point that CCAP was a survey conducted entirely on the Internet.
Respondents opted in to the PollingPoint panel, run by YouGov/Polimetrix, Inc. and all had broadband
access prior to being invited to complete the CCAP survey. Having said that, however, the broadband
variable that is affecting political knowledge and issue placement is not a CCAP survey measure, but
instead a measure from the FCC furnished datasets, specifically, TVMARKETS.DTA. If there are biases that
come from regular Internet users (the CCAP respondents) "relying more on the Internet" to uncover political
information, they would result in an attenuated role for average broadband penetration in DMA regions,
not in the relationship we observe here.
48

-1.0
-0.5
0.0
0.5
1.0
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
-1.0
-0.5
0.0
0.5
1.0

Figure 19: Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level logistic regression model for respondent ability/willingness to place Obama on health
policy. The estimates have been sorted from low to high and by construction have zero mean.
49

cannot precisely estimate these differences, and even if we could, it does not seem that the
DMA region-level factors that we have measured are driving this pattern of off-sets, with
perhaps the exception of Internet penetration. As with the other engagement indicators, the
level of variation in the off-sets is low. The off-sets are tightly clustered around their mean
of zero, with very little spread.
4.4
Validated Voter Turnout
Validated voter turnout is also a binary outcome, and so the model is identical to that
used for willingness/ability to place Obama on health care (section 4.3).
4.4.1
Multilevel Model
Est
StdDev
Intercept
0.552
0.109
Female
-0.132
0.052
Hispanic
-0.311
0.089
Black
-0.103
0.080
College
0.348
0.054
Age (years/10)
0.182
0.016
Income (1-14)
0.069
0.008
# TV Voices
-0.043
0.031
# Radio Voices
0.007
0.007
# Parents with 2+ TV Stations
-0.045
0.073
# Parents with TV and Radio
-0.020
0.086
# Parents with TV, Radio and Newspaper
-0.078
0.154
Broadband 200bps (%)
-0.088
0.501
# Parents with News/Talk Radio Station
0.017
0.022
Total Presidential Ad GRPs (logged)
-0.071
0.056
Population 18+ (logged)
0.042
0.101
Table 8: Estimates of Multilevel Model of Validated Voter Turnout. Entries above the line
are estimates of micro-level parameters, ; entries below the line are estimates of DMA-level
parameters, .
Recall that the multilevel model estimates the regression model j N ( , 2) for the
j
prior j N (, 2).
In Table 8 we present the results of the multilevel model of turnout. It bears a striking
similarity to the model of issue placement above, particularly for the micro-level indicators,
which once more have strong effects in the anticipated direction except for the indicator for
50

"Black" racial identification. The micro-level characteristics explain the within DMA region
variation, but an examination of the bottom of the table suggests that the DMA region-
level measures do not match up. Only advertising GRPs come close to a precise estimate
(t - statistic of 1.4). None of the other point estimates are bigger than their standard
deviations.
Internet penetration does not emerge as a driver of turnout in the 2008 presidential
election at the DMA region-level, unlike its role in shaping overall political knowledge and
ability to place Obama on health care. Recall also that it was not related to general political
interest either. It seems that the Internet is used to gather political information, but does
not affect a person's political behavior or interest, all else equal.
Presidential advertising GRPs are related to turnout in much the same manner that they
were related to interest. It seems as if DMA regions with higher average levels of GRPs have
lower average levels of turnout, but again, we caution that this result should not be viewed
as causal. The design suffers from the endogeneity of candidate advertising strategies, as it
is not a randomized trial assigning advertising volume to DMA regions. If candidates use
their advertising to mobilie low-propensity voters, the effect of advertising may be negative,
as it is here. In terms of its magnitude, the size of the advertising effect appears to be on
par with the effect of income.
Figure 20 presents the graphical representation of the DMA region off-sets for turnout
in the 2008 general election among our registered voter sample. Perhaps at the low-end of
the scale there is more differentiation than we seen by DMA region, but this is not matched
at the high-end. For the turnout off-sets, at the low-end, the alphajs extend up to 2 units
away from the mean of zero, but at the high-end, they extend only the typical half-unit or
so.
4.5
Conclusion: Project A
What explains registered voters' levels of political interest, knowledge, information, and
participation? Here we have demonstrated what political scientists have long known, individual-
level demographics are strongly related to political engagement and behavior. But we have
also explored whether factors related to the media ownership structure within markets affect
these things.
For the most part, we find no substantively important or in fact measurable effect of
DMA region-level ownership indicators. While there are DMA region-level measures that
affect civic engagement, they are not related to the ownership structure of the local mar-
ket. Instead, the things that sometimes affect political behaviors and information are the
51

-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
q
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5

Figure 20: Estimates and 95% Credible Intervals, DMA region-specific offsets (j), multi-
level logistic regression model for validated voter turnout. The estimates have been sorted
from low to high and by construction have zero mean.
52

amount of presidential advertising aired within a DMA region (during the 2008 election)
and the penetration of the Internet in the market (at the level of 200kbs). The occasional
impact of these two factors raises important considerations about how people gather political
information -- or how political information finds them.
Political behaviors, like interest and turnout, are affected by the supply of information
available to people, but our analyses suggest that it is the supply of information provided
by candidates, not by local media outlets, that sometimes affect civic engagement. And
similarly, demand for information matters, too. the Internet provides citizens interested in
politics with 24 hour, easy access to any information they want. In the 21st century, the
provision of political information is likely to become more demand-driven and less supply-
driven, whether that supply is provided by candidates or by existing media in the market.
The results of Project A provide little indication that market ownership structure affects
civic engagement, but we must reiterate an important caveat. CCAP is one of the largest
studies of presidential elections ever fielded, but despite it's 120,000 interviews (20,000 re-
spondents interviewed 6 times), it was not designed to have the power to detect the kinds
of effects that this inquiry necessitates. Although we have coverage in most of the nation's
DMA regions, our coverage is not representative of the population within those DMA re-
gions -- and in some cases of small DMA regions, we have only a single person in the DMA
region. In order to completely understand the effect of DMA region ownership structure, a
study representative of the DMA regions should be done with enough people in each DMA
region to get precise estimates, but designed to maximize the number of DMA regions in
the project. This study would cost hundreds of thousands of dollars and thus is unlikely
to be done by an independent researcher, a foundation, or the federal government; but an
appropriately designed project that stratified on DMA region would at least have the chance
of returning the kids of effects in which the FCC is interested.
We turn now to a small example of one such study. The 2006 CCES Midwestern Sample,
a project fielded by researchers at UCLA and the University of Wisconsin during the 2006
Congressional elections. This project is representative of 9 DMA regions in the Midwest and
ties characteristics of each DMA region directly to people's levels of knowledge about local
political candidates.
53

Part III
Analyses: Identification of Local
Candidates -- A Study
Representative of DMA regions
(Project B)
5
The Pattern across DMA regions and its Explana-
tions
In this project we leverage a survey designed to be representative at the DMA region level
to demonstrate the importance of local political context. We link the market environment
and the political context to people's ability to identify the Congressional candidates running
in their districts from their photos alone. Admittedly, this is a rather tough test. If, however,
something about the media market structure results in more local, political television news
programming then this test is well-suited to demonstrate this relationship.
In CCES, we showed each respondent a photograph of their Member of Congress and
another of the person running against their member from the other major party, if their was
a challenger in the race.23 This totaled 127 candidates for the House of Representatives over
65 congressional districts. Of these 65 districts, candidates ran advertising in 10 of them
between July 1 and election day, November 3.24
We use people's abilities to identify candidates running for Congress as a measure of their
level of local political knowledge -- particularly as gained from watching local television
-- and will examine the differences in these abilities across media markets to understand
whether the structure of media ownership affects people's abilities to recognize local political
candidates. While this test of local political knowledge may be tougher than any measure
of engagement we used in Project A, the design of this study is also imperfectly suited
to test the effects of market-level factors since it covers only 9 unique DMA regions. In
essence, we have 3,002 observations at the individual-level clustered into 9 media markets.
In terms of power, we face the possibility of being underpowered at level 2 -- the DMA region
23In open seats we showed the two major party candidates running for the office.
24The 10 districts with candidate advertising in 2006 are: IL10, IL15, IL6 IL8, MI9, MN1, MN2, MN6,
OH12, and WI2.
54

level. The small number of level 2 units also limits our ability to run multivariate models
as we lose degrees of freedom very quickly in this design. In terms of benefits, however,
the direct relevance of the dependent variable to the provision and quality of local political
news cannot be denied. If there are DMA region-level differences in people's abilities to
recognize Congressional candidates from their photos it is quite likely that those off-sets can
be explained by some structural condition in the DMA region.
5.1
How many people can recognize the candidates?
We begin with a simple question. What percentage of Americans can recognize their
sitting member of Congress, a candidate running for Congress, or both? The week after the
2006 congressional elections, we asked 3,002 respondents the following question:
Some people remember faces well and others do not. If you happen to know the
name of the individual pictured here, please type it in to the text-box. Do not
worry about spelling. If you do not know, please leave the box blank and proceed
to the next question.
After this stem, we showed the respondent a head-shot of one of the two major party
candidates who ran for Congress in his district. We followed the first image by asking them
to take a look at one more picture, and then we showed them the other candidate's image.
It is worth noting that we did not give the respondents any clue that the images were of
Congressional candidates, or for that matter, candidates at all.
Respondents were free to type up to 55 characters in the text box. We coded the verbatim
answers that people gave such that spellings and names that came close to being correct were
deemed correct. We were lenient in this regard. For example, Bobby Rush was an incumbent
Democrat running for re-election to Congress in 2006. If, upon seeing his image, one of his
constituents typed "Bobby-guy who is in politics in DC" we count that as correct. Similarly,
"Bobby R.", "Rush", and "our Congressman" are counted as correct.
In Table 10 we present summary statistics for the entire data set broken out by incum-
bency and party. The data indicate that incumbents who run for re-election are easier for
people to identify than are challengers running for the same seat. Fifty-four percent of the
respondents in CCES could correctly identify their sitting member of Congress if he or she
was running for the office again in 2006. Only 21% could correctly identify the challengers.25
That is, more than twice as many people could visually identify their incumbent Member
25This difference in means has a t-statistic of 29.5 with 2,650 degrees of freedom and is significantly different
from zero at a level of p .0001.
55

Table 9:
Percent Correctly Identifying Congressional
Candidates, 2006
Identifying Candidates From Images
Incumbent
Incorrect
54.2
Correct
45.8
Challenger
Incorrect
79.1
Correct
20.9
Republican
Incorrect
69.5
Correct
30.5
Democrat
Incorrect
63.8
Correct
36.2
Correct Responses
None
56.7
One
23.0
Two
20.3
Source: CCES
N=3002
56

of Congress than could identify the person running against the incumbent at the end of the
2006 campaign season.
The differences between parties is also highly significant, but not substantively as impres-
sive as the difference between incumbents and challengers. Examining each party separately,
roughly one-third of CCES respondents could correctly identify Congressional candidates,
ignoring incumbency status.
More than half the respondents, nearly 57%, were unable to come up with the name of
either candidate. Almost a quarter could name at least one candidate running for Congress,
but only 20% could name both of the major party candidates.
The identification of the candidates for Congress in 2006 make up the dependent variable
for this exercise. We are interested in explaining the variation in abilities to correctly identify
candidates with a particular interest in whether the structure of the respondent's media
market affects knowledge of the candidates. Does living in a DMA region with no multi- or
cross-ownership of television, radio, and newspapers lead to a greater or lesser probability of
being able to correctly identify candidates for local offices? What about the size of the DMA
region in terms of television households, but also in terms of television and radio stations?
Is there variation in people's abilities across DMA regions and how much of it is attributable
to the way the DMA region is structured?
We begin with a basic question: How much variation is there in the ability to identify
congressional candidates by media market? We present these data in Figure 21. Keep in mind
that some media markets have as many as 18 congressional districts within the boundary
(Chicago, for example), thus while we are interested in variation at the DMA region level,
there may also be interesting variation within the DMA region due to characteristics of the
districts themselves. We explore these possibilities in turn.
Over the nine DMA regions covered in this analysis, the span of average values for
correctly identifying the incumbent candidate ranges from a low of 41% in the Detroit DMA
region to a high of 58% in the Lansing DMA region, right next door. Across all districts
represented in these data, the spread of average correct identifications covers a larger range,
from 0% in Michigan's 15th to 68% in Ohio's 11th. We present these data in Figure 22.26
Clearly there are district-level factors that influence respondents' abilities to identify the
candidates for Congress. These factors may have something to do with the structure of
26We have very little data in many of the districts that register 0% correct identifications, but in Michigan's
15th and 13th districts we have 85 and 30 respondents, respectively. Similarly, we have one district where
the sole respondent was able to identify the incumbent, Wisconsin's 7th, but we cannot generalize from
this estimate to anything meaningful about the district. Keep in mind that these data were drawn to be
representative of the DMA region not of the Congressional district.
57

Figure 21: Differences in Ability to Identify Incumbents by DMA regions
58

Correctly Identifying Incumbent Candidates
WI 7
OH 11
WI 4
OH 10
WI 2
IL 4
OH 15
IL 2
OH 9
OH 12
IL 15
MI 8
IL 5
MN 1
MI 14
MN 3
OH 14
MN 2
MI 9
IL 10
IL 8
MI 10
IL 18
IL 1
IL 7
MI 12
IL 9
WI 5
OH 16
WI 1
OH 17
MN 4
IN 1
IL 3
MN 8
IN 2
IL 14
MI 11
IL 11
IL 13
MN 7
WI 3
OH 7
IL 19
IL 16
WI 6
OH 5
OH 3
MI 5
MI 3
MI 15
MI 13
IN 3
0
.2
.4
.6
.8
1
Percent
CCES 2006
Figure 22: Ability to Identify Candidates by Congressional District
59

the media market, but more than likely they do not. For example, it is possible that a
media conglomerate "has it out" for a certain member of Congress and focuses a lot of its
local political coverage in a DMA region that covers many Congressional districts on this
one member. More than likely, however, is the opposite scenario. A member of Congress
who sits on a powerful and popular committee garners a lot of coverage from the local news
because of what he or she is doing in Washington, even though viewers from many other
congressional districts live in his or her DMA region.
5.2
Is There a Pattern to the Variation at the DMA region Level?
In the previous section we demonstrated the variation in local knowledge of candidates
across DMA regions. But there is also plenty of variation in levels of knowledge about
candidates within DMA regions. One example of this is the variation across Congressional
districts, some of which reside in the same DMA region, but imagine other possibilities such
as the variation in local political knowledge within DMA regions associated with people's
interest levels in politics generally, or their level of education, their household income, or
even their marital status. In this section, we attempt to demonstrate how much of the
variation in local political knowledge is happening across DMA regions compared to the
amount of variation that we observe within DMA regions. This will give us some indication
of the upper-bound for how much of the variation in levels of knowledge can be affected by
DMA region-level attributes like ownership structure. If very little or none of the variation
is happening across DMA regions and most of it is happening within DMA regions, there
will be little reason to believe the structure of ownership can affect levels of local political
engagement and knowledge. We begin this assessment just as we did in Project A, by an
exploratory analysis of the variation explained. As before, we use simple ordinary least
squares regressions (OLS) despite the form of the dependent variable, since we are not so
much interested in interpreting the coefficients as we are in examining the percent of variation
explained by the DMA region fixed effects and by other individual level factors.
Table 10: Ability to Identify Incumbent Images as Func-
tion of DMA region Fixed Effects
Variable
Coefficient
Std. Err.
Detroit
-0.034
0.100
Minneapolis
0.052
0.042
ClevelandAkron (Canton)
0.143
0.054
Continued on next page...
60

... table 10 continued
Variable
Coefficient
Std. Err.
Milwaukee
0.004
0.061
Columbus, OH
0.185
0.060
Champaign&SprngfldDecatur
0.149
0.085
Madison
0.208
0.057
Lansing
0.118
0.030
Chicago
0.450
0.024
N
2650
R2
.031
Significance levels :
: 10%
: 5%
: 1%
Table 10 shows that only 3.1% of the variation in incumbent recognition is explained by
variation across the DMA regions.27 In terms of knowledge of local politics about incumbent
Members of Congress, it does not appear that the across-DMA region variation explains much
of the pattern. This makes intuitive sense, since Congress members have other methods of
proselytization at their disposal -- through the franking privilege and in-person visits to
districts.
In many ways, a better test of whether the structure of the DMA region is likely to have
any effect on local political knowledge is whether a DMA region-level fixed effect can explain
much of the variation in respondents' abilities to identify the challengers in Congressional
races. These candidates have no other "free" (or federally sponsored) media at their disposal,
thus any exposure they receive is likely coming from television, radio, and newspapers.28
Table 11: Ability to Identify Challenger Images as Func-
tion of DMA region Fixed Effects
Variable
Coefficient
Std. Err.
Detroit
0.001
0.082
Continued on next page...
27For the analyses that follow, the dependent variable is a person's ability to recognize the incumbent
Member or the challenger. Because of this, respondents who reside in districts with open seats (no incumbent
in the race) are eliminated from the analyses as it would not be appropriate to classify the candidates in
these districts as either incumbents or challengers. This reduces the total N from 3,002 to 2,650.
28Complete details of the estimation for all subsequent models can be found in the work documentation
for Project B in the Appendix of this report.
61

... table 11 continued
Variable
Coefficient
Std. Err.
Minneapolis
0.203
0.067
ClevelandAkron (Canton)
-0.052
0.049
Milwaukee
-0.028
0.066
Columbus, OH
0.432
0.074
Champaign&SprngfldDecatur
0.214
0.120
Madison
0.332
0.063
Lansing
0.023
0.036
Chicago
0.159
0.035
N
2650
R2
.139
In Table 11 we present the same results for challengers as we did for incumbents earlier.
The data show that much more of the variation in challenger familiarity is explained by the
simple DMA region-level fixed effect. In fact, more than four times the variation is explained
in the challenger model than was in the incumbent model. Almost 14% of the variation in
ability to identify the challenger from his or her image is a function of living in a particular
DMA region. If the ownership structure of DMA regions affects local political knowledge it
is likely to do so via knowledge of challengers not incumbents.
We turn now to an analysis of the variation explained depending on individual-level
factors.
5.3
Individual-Level Attributes
Earlier we introduced the idea that the main drivers of political engagement were likely to
be personal characteristics such as age, education, income level, gender, race, marital status,
and employment. In contrast to the DMA region-level fixed effect model we just presented,
how much of the variation in candidate knowledge do these individual-level characteristics
explain? We present these results for identification of the incumbent in Table 12 and for the
challenger in Table 13.
The data indicate that the individual-level demographics explain roughly 6% of the vari-
ation in ability to identify the incumbent candidate for Congress and 3% of the variation in
identifying the challenger. This pattern is reversed from the one we observe in the analysis of
DMA region fixed effects. The DMA region characteristics explain four-times the variation
62

in knowledge about the challenger than they do about the incumbent, but individual-level
demographics explain twice as much variation in incumbent familiarity than they do for
challengers.
Thus is seems that features of respondents are more important in identifying sitting
members of Congress than DMA region-level factors, but some characteristic of the DMA
regions is more important in recognizing the challenger in the race. In a healthy democracy
with competitive elections, this seems, normatively, like a good pattern. The local media
environment may be helping people to become familiar with challengers for Congressional
seats.
Table 12: Ability to Identify Incumbent Images as Func-
tion of Demographics
Variable
Coefficient
Std. Err.
Post Graduate Degree
0.176
0.046
College Degree
0.103
0.030
White
0.042
0.039
Male
0.075
0.018
Income Group
0.001
0.003
Age
0.006
0.001
Married
0.062
0.021
Employed
-0.023
0.022
Intercept
0.048
0.071
N
2594
R2
.06
Table 13: Ability to Identify Challenger Images as Func-
tion of Demographics
Variable
Coefficient
Std. Err.
Post Graduate Degree
0.144
0.048
College Degree
0.085
0.025
White
0.073
0.031
Continued on next page...
63

... table 13 continued
Variable
Coefficient
Std. Err.
Male
0.054
0.019
Income Group
0.002
0.004
Age
0.000
0.001
Married
0.081
0.020
Employed
-0.033
0.017
Intercept
0.033
0.041
N
2594
R2
.033
Although we do not want to interpret these coefficients in any structural sense (since
we simply ran OLS for this exploratory stage of the analysis), it is also worth noting that
the indicators behave generally as anticipated. That is, respondents with higher levels of
education, who are white, male, and married have a higher probability of recognizing these
candidates from their images alone. In the case of the challenger, it is interesting to note that
being employed full time actually decreases a person's chance of knowing who the challenger
is in a Congressional race, all else equal. We assume this is due to the fact that people who
work full time are less likely to encounter the local news on a routine basis compared to
those who do not work full time. We confirm this pattern below in Table 14. We present
marginal breakdowns by employment status (full time work versus not) on a question asking
respondents how often they rely on their local television news as a source for information
about elections. Our hunch about employment is confirmed, as people who work full time
are 8-points less likely to say they rely mostly on local news for information about elections.
This pattern is interesting since we typically think of employment as increasing a per-
son's interest, knowledge, and engagement in politics. Here, however, we see employment
being a drain on a person's ability to recognize the challenger in a Congressional race and
simultaneously we find that working people are less likely to rely on local news as a source
of election news. These two findings, even taken together, are not enough to conclude that
local news is a good source of information about local elections, but the trend does seem to
indicate that there may be something worth investigating in terms of the content of local
news and local elections.
Of course we cannot identify whether it is the content of the local media environment
that is driving the recognition of challengers or whether it is some kind of campaign effort
64

by the challenger that is tied to the competitiveness of the race (which is correlated with
the DMA region since districts reside mainly within single DMA regions). What we can rule
in or out is whether the ownership structure of the DMA region explains the knowledge of
challengers relative to incumbents as driven by something about the DMA region.
Table 14: Reliance on Local News for Information about
Elections as a Function of Employment
Reliance Level
Unemployed
Employed Full Time
Total
Rely solely on Local News
10.3
8.7
9.4
Rely Mostly on Local News
25.2
17.3
20.7
Rely Somewhat on Local News
26.0
28.8
27.6
Rarely Rely on Local News
11.3
12.6
12.0
Do Not Rely on Local News
22.7
29.4
26.6
Don't Know
4.6
3.1
3.7
N
2946
X2 = 25.8,
p .000
Finally, as we did in Project A, we generate a set of DMA region-level off-sets for each
media market, net of micro-level factors, and use those as dependent variables that we
attempt to explain with the structural features of the markets. The combined DMA region
and individual-level model explains nearly 17% of the variation in abilities to recognize the
challenger which is not unlike the explanatory power of the joint models from Project A.
We present these coefficients are presented in Table 15.
Table 15: Ability to Identify Challenger Images as Func-
tion of Demographics and DMA region Fixed Effects
Variable
Coefficient
Std. Err.
Post Graduate Degree
0.126
0.035
College Degree
0.074
0.007
White
0.044
0.022
Male
0.059
0.014
Continued on next page...
65

... table 15 continued
Variable
Coefficient
Std. Err.
Income Group
0.004
0.005
Age
0.001
0.001
Married
0.063
0.023
Employed
-0.035
0.018
Chicago
-0.051
0.072
Detroit
-0.073
0.079
Minneapolis/St. Paul
0.151
0.077
ClevelandAkron (Canton)
-0.107
0.076
Milwaukee
-0.071
0.076
Columbus, OH
0.381
0.074
Champaign&SprngfldDecatur
0.123
0.076
Madison
0.259
0.077
Lansing
-0.066
0.076
N
2594
R2
0.169
5.4
Explaining the DMA region-level Fixed Effects
What is driving the differences in knowledge across these DMA regions, particularly
for recognition of the challenger? We are interested in the structure of these data, as in
Project A, and use these exploratory analyses to give us some sense of how much of the
variation can be explained by DMA region-level structural factors. The nine off-sets become
our dependent variables in a new level-two analyses of market structure. Essentially, we
are asking whether the pattern represented in these nine coefficients can be explained by
properties of the markets that relate to ownership structure (particularly the number of
television and radio stations, the number of parent corporations owning more than a single
television station, or a television and radio station, or a television and radio station along
with a newspaper).
We begin with simple correlations between the fixed effects and the DMA region-level
ownership variables in Figure 23. Reading down the first column of the figure shows the
relationship between the ownership variables and the DMA region-level fixed effects from
the challenger-identification model. The largest correlation in this column is between the
66

fixed effects and the number of television stations in the market -- -.5803. This suggests that
increasing numbers of stations in a market, on average, corresponds to the people living in
that market being less able to identify the challenger in a Congressional race, all else equal.
Perhaps as Sunstien (2001) suggests, more choices mean an increased ability to opt out of
politics. Number of radio stations correlate with the fixed effects less well, -.09. The number
of multiple- and cross-owning parent entities correlates slightly with the fixed effects, at -.27
and -.2 respectively.29 As we saw in Project A, the market ownership variables are highly
correlated with one another by definition this must be true for many of them as they are
transformations of one another. The multiple-ownership and cross-ownership variables are
correlated with the voices variables at levels higher than .85. RADIOVOICES is correlated
with MULTICOMTVPARENTS at .95 in these data. These correlations make it difficult to
determine the unique effects of any single factor in a multivariate framework.
In Figures 24 through Figure 27 we examine these relationships more closely. Although
the correlation between the number of television stations and the DMA region fixed effects
is high, a regression line fit through these data demonstrates the noise in the data. We
cannot be sure that the line is not actually flat. With more data, however, we might have
been able to isolate this effect more precisely. Figure 24 (and all the ones that follow) show
the scatterplot between the fixed effects and the ownership variable, a regression line fit to
these data, and the 95% confidence interval around the fitted line. The space between the
gray interval lines is suggestive of the uncertainty associated with this relationship. Number
of television stations explains 34% of the variation in the DMA region off-sets from the
challenger identification model.30
In contrast to Figure 24, examine the mainly flat line in Figure 25. None of the variation
in the fixed effects is explained by the number of radio stations in the market alone. The
cross- and multiple-ownership relationships are shown in Figure 26 and Figure 27. Both of
the lines are downward sloping, but characterized by a large amount of uncertainty. It is not
so clear in these cases that even with more data the relationships would amount to anything
substantively important. Each of these measures explains less than 4% of the variation in
the fixed effects.
29Incidentally, it is worth noting that these correlations are higher than similar correlations done with the
fixed effects for the model identifying the incumbent candidate in the district.
30Complete results from the OLS regressions represented in these figures can be found in the work docu-
mentation in the Appendix of this report.
67

Figure 23: Scatterplot Matrix of DMA region Fixed Effects from Challenger Identification
Model and Ownership Variables
68

Figure 24: DMA region-Level Fixed Effects and Number of Independent TV Voices
69

Figure 25: DMA region-Level Fixed Effects and Number of Independent Radio Voices
70

Figure 26: DMA region-Level Fixed Effects and Number of Multi-Ownership Parents
71

Figure 27: DMA region-Level Fixed Effects and Number of Cross-Ownership Parents
72

While there may be some hint of a relationship between the number of independent
television stations and the fixed effects, these DMA region level ownership variables for the
most part bear no relationship to what differentiates the DMA regions in their respondents'
abilities to recognize local candidates. We mentioned earlier that an important element of
DMA region-level differentiation in terms of political engagement was the level of campaign
effort or activity in the political units within the DMA regions. In other words, some DMA
regions may have competitive Congressional districts within the border and others may have
relatively uncompetitive races in the districts that constitute the market. In Project A, we
did not see much of a relationship between the presidential advertising and the market off-
sets, although there was a trend. In this case, being able to recognize a previously unknown
candidate's face maybe more directly tied to the amount of advertising (or overall campaign
effort) in a media market or district.
In order to test whether the thing that is driving the difference across DMA regions in
terms of political knowledge is political -- not structural (in terms of ownership of media
outlets), we bought data from the Nielsen Corporation on the number of gross ratings points
purchased by each candidate running for Congress in these nine DMA regions. Many of the
candidates, in fact most, ran no advertising. This is a well-known limitation of Congressional
campaigning -- it is inefficient for most candidates for Congress to advertise on television
because their DMA region usually includes many people who live in different districts. In
essence, they are paying for eyeballs of people who cannot vote for them. Still, many can-
didates for Congress do advertise on TV and we use these data to look for the effects of
political media in the DMA region.
We present the scatterplots of the GRP data and the fixed effects in Figure 28. In contrast
to Figure 23, it is easy to see the strong relationship between the number of GRPs purchased
in the DMA regions by the candidates for Congress and the DMA region off-sets. Figures 29
and 30 show the precision of the estimate of these relationships, especially in contrast to the
relationships with ownership structure. The advertising GRP data explain 85% and 74% of
the variation in the DMA region-level fixed effects. Of the ownership structure data, only
the number of television voices came close to this level of association, with 33%. The others
were in single digits or non-existent.
Even among people who do not rely on local news for information about politics or public
affairs, the advertising GRPs explain as much as 80% of the variation in the fixed effects
from a model using only these non-political news viewers. In contrast, among people who
report they do rely on the local news for information about politics and public affairs, the
73

Figure 28: DMA region-Level Fixed Effects and Gross Ratings Points
74

Figure 29: Explaining DMA region-Level Fixed Effects with Incumbents' Ad Buys
75

Figure 30: Explaining DMA region-Level Fixed Effects with Challengers' Ad Buys
76

advertising data explains an impressive 95% of the variation in the fixed effects.
The 2006 data illustrate the kind of effect that can exist at the market level, but that
we fail to find in 2006 or in 2008 in terms of ownership structure. Nearly all of the DMA
region-level variation in respondents' abilities, on average, to identify visually the candidates
running for Congress comes from variation in the amount of candidate advertising in the
DMA region. No structural ownership variable comes close to having the explanatory power
of the paid-advertising effects.
In terms of explaining correctly identifying the candidates, we are somewhat limited by
having only 9 DMA regions with which to work in this design, however, we present results
of two parsimonious specifications in Table 16. These models are simply logit estimations
performed at the DMA region-level with 9 observations. All probability calculations are
done holding variables at their means.
Table 16: Correctly Identifying the Challenger in a Dis-
trict from His/Her Image, DMA region-Level
Model 1
Variable
Coefficient
Std. Err.
Challengers GRPs (in hundreds)
0.009
0.003
Independent TV Voices
-0.020
0.016
Constant
.371
.162
R2 = .77
N=9
Model 2
Variable
Coefficient
Std. Err.
Challengers GRPs (in hundreds)
0.008
0.003
Independent TV Voices
-0.060
0.035
Parents with > 1 Station
.11
.084
Parents with TV, Radio, News
-.022
.078
Constant
.612
.248
R2 = 0.84
N=9
Note: Cell entries are OLS regression estimates on 9 DMA region observations.
Significance levels :
: 10%
: 5%
: 1%
77

In terms of being able to recognize local candidates, specifically the challenger in a Con-
gressional election, the only significant driver of DMA region-level factors is that candidate's
advertising penetration as measured by Neilsen gross ratings points.31 Moving from no ad-
vertising in a market to a typical 1000 GRP ad-buy (100% of the target audience sees the
ad 20 times) boosts the probability of being able to recognize the candidate from his or her
image by 20-points, from .19 to .39.32
Even with controls for the number of independent television voices, the number of parent
companies owning more than one TV station, and the number of parent companies owning
TV, radio, and newspaper outlets, the effect of advertising remains strong, in fact, virtually
unchanged. These findings are consistent with other work in communication and political
science that demonstrates the effectiveness of political advertisements relative to news cover-
age in terms of educating voters (Vavreck 2009; Gilens, Vavreck and Cohen 2007). When it
comes to providing the information necessary to dispense the duties of citizenship, it appears
that television is doing its part -- through advertising, however, not through the ownership
structure of the news environment.
Part IV
Report Conclusion
We conclude that while there is a pattern to the civic and political engagement in media
markets across the country we are unable to explain this pattern with market-ownership
indicators like TV and radio voices, and multi- and cross-ownership.
In terms of political and civic engagement, there is some within DMA region variation that
can be explained at the individual-level and in addition to that, what drives participation,
learning, knowledge, and interest in politics and public affairs seems to be the media context
as cultivated by the candidates in the market through paid advertising (or other efforts
with which paid advertising is correlated, like a ground-campaign or direct mail effort); and
sometimes, the level of Internet penetration in the market.
The advertising effect is complicated by the fact that any political messaging is likely to
have as its goal electioneering, which means that one side is trying to persuade voters to
do one thing and the other is trying to persuade them to do the opposite. To the extent
that we can show any effects of political advertising effort at all in these analyses without
accounting for the content of the ads, it is likely that we are underestimating the effects of
31In these models, GRPS are denominated in units of 100.
32For Incumbents, the movements is a much smaller 6-points, from .52 to .58.
78

the ads, since in most cases the competing messages are likely to cancel each other out.
The clear and direct effect of challenger advertising on respondents' abilities to recognize
images of the candidate demonstrate the power of television to affect political knowledge or
engagement. That we cannot show this kind of effect for television news -- as measured by
number of voices or conglomerate cross-ownership -- suggests that the intensity of political
television advertising is so great that its effects will drown out any effects of multi- or cross-
ownership.
References
Bartels, L.M. 1986. "Issue voting under uncertainty: An empirical test." American Journal
of Political Science pp. 709728.
Delli Carpini, M.X. and S. Keeter. 1997. What Americans know about politics and why it
matters. Yale University Press.
Gelman, Andrew and Jennifer Hill. 2007.
Data Analysis Using Regression and Multi-
level/Hierarchical Models. New York: Cambridge University Press.
Gerber, Alan S. and Donald C. Green. 2000. "The Effects of Canvassing, Telephone Calls,
and Direct Mail on Voter Turnout: A Field Experiment." American Political Science
Review 94(3):653663.
Gerber, A.S., D.P. Green and C.W. Larimer. 2008. "Social pressure and voter turnout: Evi-
dence from a large-scale field experiment." American Political Science Review 102(01):33
48.
Gilens, M., L. Vavreck and M. Cohen. 2007. "The Mass Media and the Public's Assessments
of Presidential Candidates, 19522000." Journal of Politics 69(4):11601175.
Green, D.P. and L. Vavreck. 2008. "Analysis of cluster-randomized experiments: A compar-
ison of alternative estimation approaches." Political Analysis 16(2):138.
Iyengar, S. and D.R. Kinder. 1988. News that matters: Television and American opinion.
University of Chicago Press.
Jackman, S. and L. Vavreck. 2010. "Primary politics: race, gender, and age in the 2008
Democratic primary." Journal of Elections, Public Opinion & Parties 20(2):153186.
Jackman, Simon. 2009. Bayesian Analysis for the Social Sciences. Hoboken, New Jersey:
Wiley.
79

Jackman, Simon and Lynn Vavreck. 2009. "The 2007-8 Cooperative Campaign Analysis
Project, Release 2.0.".
Lupia, A. and M.D. McCubbins. 1998. The democratic dilemma: Can citizens learn what
they need to know? Cambridge Univ Pr.
Plummer, Martyn. 2010. JAGS Version 2.20 manual.
URL: http://mcmc-jags.sourceforge.net">http://mcmc-jags.sourceforge.net
Popkin, S.L. 1991. The reasoning voter: Communication and persuasion in presidential
campaigns. University of Chicago Press.
Prior, M. 2005. "News vs. entertainment: How increasing media choice widens gaps in
political knowledge and turnout." American Journal of Political Science 49(3):577592.
Prior, M. 2007. Post-broadcast democracy: How media choice increases inequality in political
involvement and polarizes elections. Cambridge Univ Pr.
Reckase, Mark D. 1997. "The Past and Future of Multidimensional Item Response Theory."
Applied Psychological Measurement 21(1):2536.
Robert, Christian P. and George Casella. 2004. Monte Carlo Statistical Methods. Second ed.
New York: Springer.
Rosenstone, S.J., J.M. Hansen and K. Reeves. 2003. Mobilization, participation, and democ-
racy in America. Longman.
Rothenberg, Thomas J. 1971.
"Identification in Parametric Models." Econometrica
39(3):577591.
Spiegelhalter, David J., Andrew Thomas, Nicky G. Best and Dave Lunn. 2003. WinBUGS
User Manual Version 1.4. Cambridge, UK: MRC Biostatistics Unit.
Sunstein, C.R. 2002. Republic. com. Princeton Univ Pr.
Takane, Yoshio and Jan de Leeuw. 1987. "On the relationship between item response theory
and factor analysis of discretized variables." Psychometrika 52:393408.
Vavreck, L. 2009. The message matters: the economy and presidential campaigns. Princeton
Univ Pr.
Vavreck, Lynn. and Douglas Rivers. 2008. "The 2006 Cooperative Congressional Election
Study." Journal of Elections, Public Opinion & Parties 18(4):355366.
80

Verba, Sidney, Kay Lehman Schlozman and Henry Brady. 1995. Voice and Equality: Civic
Voluntarism in American Politics. Cambridge, MA: Harvard University Press.
Wolfinger, R.E. and S.J. Rosenstone. 1980. Who votes? Yale Univ Pr.
81

Document Outline

  • I Introduction
    • Data
      • The Cooperative Election Studies
      • Validated Turnout Data from U.S. Secretaries of States
      • Nielsen Gross Ratings Points Data for Political Advertising
    • Civic Engagement and Political Information
  • II Analyses: Political Interest, Knowledge, Uncertainty, and Participation (Project A)
    • Analyses of Variation
      • Civic Engagement: Dependent Variables
      • DMA region Fixed Effects
    • Bayesian Analysis of Hierarchical and Multilevel Models
      • Interest in Politics
        • Multilevel model
      • Overall Political Knowledge/Information
        • Multilevel model
      • Ability/Willingness to Place Obama
        • Multi-Level Model
      • Validated Voter Turnout
        • Multilevel Model
      • Conclusion: Project A
  • III Analyses: Identification of Local Candidates A Study Representative of DMA regions (Project B)
    • The Pattern across DMA regions and its Explanations
      • How many people can recognize the candidates?
      • Is There a Pattern to the Variation at the DMA region Level?
      • Individual-Level Attributes
      • Explaining the DMA region-level Fixed Effects
  • IV Report Conclusion
    • References

Note: We are currently transitioning our documents into web compatible formats for easier reading. We have done our best to supply this content to you in a presentable form, but there may be some formatting issues while we improve the technology. The original version of the document is available as a PDF, Word Document, or as plain text.

close
FCC

You are leaving the FCC website

You are about to leave the FCC website and visit a third-party, non-governmental website that the FCC does not maintain or control. The FCC does not endorse any product or service, and is not responsible for, nor can it guarantee the validity or timeliness of the content on the page you are about to visit. Additionally, the privacy policies of this third-party page may differ from those of the FCC.