DHHS Director Intro Memo

Popular Press Assignment
Claims about the mind everywhere
Tension
Imagine you are a research scientist
You’ve spent years on a project
You carefully selected every word
Your claims are qualified and nuanced
Then some journalist writes an article that focuses on one small
part of your work and gives it a misleading, sensationalized
title.

Tension
Imagine you are a journalist
You’ve only got 1,000 words
You need to make the article catchy
You’ve got competition
Example:
Prize fight
Video
Thoughts
Was it engaging?
Did you learn something new?
What did you like about it?
What could have been improved?

Was it engaging?
Did you learn something new?
What did you like about it?
What could have been improved?
6
Inoue & Matsuzawa, 2007
Assignment
Part I – 10% of grade – Due on March 2nd at 11 AM
Read the Time magazine article entitled, “Watching TV Steers
Children Toward Eating Junk”
Answer corresponding questions on Worksheet 1
Read the research study entitled, “Associations of Television
Viewing With Eating Behaviors in the 2009 Health Behaviour in
School-aged Children Study”
Answer remaining questions of Worksheet 1
Part II – 10% of grade – Due on April 11th at 11 AM
Read “Priming Effects of Television Food Advertising on
Eating Behavior”
Write 750-1000 word popular press article about the study
Make it engaging, not a dry summary
Have fun and be creative
Pt 2 Expectations
Absolutely no plagiarism.

Two-quotation maximum.
Keep it clear and concise.
Important content. You will, of course, want to describe the (a)
motivation for the research study, (b) aspects of the method
used, and the (c) results. But it may also be important to discuss
(d) the broader implications of the research and (e) possible
limitations or criticisms of the research.
Be engaging.
Don’t forget a title!
More details
Things to keep in mind:
What are the 2 or 3 main points that you want your readers to
take away from your article?
Make sure those points are very clear
What is the research question? What is the motivation for this
question?
How did the researchers answer the question?
What did the researchers find? Broadly speaking, what were the
results?
What are the implications? Why should people care? What
questions remain?
If you thought the research wasn’t solid, why? What alternative
explanation do you think should be considered?
General Rubric
50 Points
Writing: 20 points
clear and easy to read, logical organization

follows guidelines (e.g., only two quotes)
no spelling or grammatical errors
Engaging (but not overly sensationalized)
Don’t give a dry description of what the research was – help
your reader understand the motivation and logic behind the
work
Academic citations not needed (e.g., APA style), but quotes
should have citations (e.g., Smith and colleagues state,
“……….”)
General Rubric
50 Points
Content: 30 points
Required content (e.g., title, description of research)
The author clearly understands the original research
Accurate description of research question, method, findings
Appropriate level of detail (just detailed enough to get the main
ideas across)
If multiple studies, you may not need to explain each one in
detail.
Relevant information
Focus on the research- don’t add lots of unrelated filler
Appropriate length (750-1000 words)
Suggestions
Have a friend or parent read it
Do not wait until the last minute
Write it and then put it out of your mind for a few days
Read it with fresh eyes. Does it still make sense? Is it still
clear?

Avoid writing on the same article as your friends
This is not a group project
Articles that are too similar will raise red flags
Health Policy and Management Introduction Memorandum
Guidelines and Examples
Remember that memos are a way to bring attention to an issue,
solve a problem, or officially acknowledge something that needs
to be recorded. Memos are often used to discuss procedure and
policy changes, outcomes of a meeting, or simply act as
management notices. Memos are NOT MEANT for sensitive
material. Memos are most effective when they are SHORT,
PURPOSEFUL, AND INTERESTING.
Formatting:
· No longer than one page
· Text *Arial 10pt, 1.5 line spacing, 1 inch margins
· Creation of fictional information:
o When discussing your experience, background, and health
policy goals, this is your time to think like a manager. Consider
what you would want to change if you were in this position of
authority and lay out the goals you think would be achievable if
you were in the position of the DHHS Director.
o In the closing paragraph, create fictional information for
contacts or just specify a department or organization.
· Ensure that all formatting and content specifications are
covered based on the example below. See the rubric, available
after the example or in the assignment submission space in
Blackboard, for more detailed information about how you will
be
graded._______________________________________________
_____________
To: Who should this go to?
From: Your Name, Director, Department of Health and Human

Services
Date: Due Date
Re: Insert a Subject
Introduction Paragraph. State your purpose in issuing the
Memo. Keep in mind who are you addressing. State what this
memo is regarding, and what a reader will know by the end.
Acknowledge your audience.
2nd Paragraph. This is the body of your memo. Make sure to use
smooth transition sentences. Many people use bullet points,
which is fine if the information can be concisely presented in
list form.
Closing paragraph. The reader has all of the information now;
you want to close with a courteous ending that states what
action you want your reader to take, if any. Make sure to
ALWAYS thank them and ALWAYS invite questions or
concerns. If you cannot answer them yourself, make sure to
direct to the right people who can.
Closing,
Signature (Or Typed Name)
Cc: All Dept. Staff
_____________________________________________________
_______________________________
Further Memo Guidelines and Examples
· https://owl.english.purdue.edu/owl/owlprint/590/
·
https://www.fsb.muohio.edu/heitgedl/Memo%20writing%20tips
%20ACC333%20SP06.pdf
· http://www.law.cuny.edu/legal-
writing/students/memorandum/memorandum-3.html
Health Policy and Management
Introduction Memo Rubric

Good to Excellent
100-90%
Fair to Good
89-70%
Poor to Fair
69-0%
Political
Neutrality 35%
Memo maintains neutrality and references health policy goals
Memo maintains neutrality but does not have clear health policy
goals
Memo is biased and opinion based
Professionalism
35%
Voice of memo maintains objective tone with concise
explanations
Voice of memo maintains objective tone but does not provide
concise explanations
Voice of memo is casual with verbose explanations
Format 20%
Follows example format and addresses all sections provided in
the prompt
Follows example format but does not address all sections
provided in the prompt
Does not follow example format provided and fails to address
all of the sections provided in the prompt
Writing
Mechanics 10%
Memo contains no grammatical errors
There are grammatical errors, but they do not
affect the readability of the memo
Grammatical errors affect the readability of the memo

Document A704-23-ZL5K-9
FROM THE DESK OF THE GOVERNOR
Document A704-23-ZL5K-9
DATE: 01/01/2017
TO: Director, DHHS
RE: Appointment
FROM: Chief of Staff, Governor’s Office
To the DHHS Director:
Now that the State Legislature has approved your appointment,
the Governor’s Office is pleased to
officially offer you the position of Director of the Department
of Health and Human Services. You are
now the highest official for the Department and will be directly

below the Governor in the chain of
command. The Governor is very excited about the potential you
and your experience bring to the
table. We expect that you will start your work immediately.
As you know, the Department of Health and Human Services in
our State has over 3000 employees,
including another 1000 plus per diem and part-time workers.
Although your appointment was in the
newspapers and on television, most of the Department is
unfamiliar with you and your work.
May I suggest that for your first action as Director, you draft an
Introduction Memorandum introducing
yourself to your Department and the other Directors in the
Administration? You should discuss your
experience, background, and health policy goals for the
remainder of your appointment. Since
Directors as well as employees in the Department are extremely
busy, I would limit your Memo to
one page. Finally, please keep in mind that this Memorandum,
as well as with any other
Memorandum you issue, could be released to the public or press
through a FOIA request or internal
leak, so keep it professional and do not share information you
do not think should be public.
I would add, since the Governor was just elected, and the
Administration is still working out official
stances on many policy issues. We request that you refrain from
taking any public political positions
at this time. The Governor would like you to remain politically
neutral until further notice.
As we have discussed, you will be working out of our Capital
Office and will report directly to the

Governor. Since this is the highest position in the Department,
there will be no orientation, no
instruction, and you will begin your job duties straightaway.
Since the Governor is extremely busy,
you are unlikely to have much face-to-face time either. I will
forward correspondence for the
Governor. Please expect to find our correspondence via your
“Modules”.
There will be a variety of additional projects that will need your
attention very soon. I look forward to
your Introduction Memo for the Department.
Respectfully,
Chief of Staff
Office of the Governor
ARTICLE
Associations of Television Viewing With Eating
Behaviors in the 2009 Health Behaviour
in School-aged Children Study
Leah M. Lipsky, PhD, MHS; Ronald J. Iannotti, PhD
Objective: To examine associations of television view-
ing with eating behaviors in a representative sample of

US adolescents.
Design: Cross-sectional survey.
Setting: Public and private schools in the United States
during the 2009-2010 school year.
Participants: A total of 12 642 students in grades 5 to 10
(mean [SD] age, 13.4 [0.09] years; 86.5% participation).
Main Exposures: Television viewing (hours per day)
and snacking while watching television (days per week).
Main Outcome Measures: Eating (�1 instance per
day) fruit, vegetables, sweets, and sugary soft drinks; eat-
ing at a fast food restaurant (�1 d/wk); and skipping
breakfast (�1 d/wk).
Results: Television viewing was inversely related to in-
take of fruit (adjusted odds ratio, 0.92; 95% CI, 0.88-
0.96) and vegetables (0.95; 0.91-1.00) and positively re-
lated to intake of candy (1.18; 1.14-1.23) and fast food
(1.14; 1.09-1.19) and skipping breakfast (1.06; 1.02-
1.10) after adjustment for socioeconomic factors, com-
puter use, and physical activity. Television snacking was
related to increased intake of fruit (adjusted odds ratio,
1.06; 95% CI, 1.02-1.10), candy (1.20; 1.16-1.24), soda
(1.15; 1.11-1.18), and fast food (1.09; 1.06-1.13), inde-
pendent of television viewing. The relationships of tele-
vision viewing with fruit and vegetable intake and with
skipping breakfast were essentially unchanged after ad-
justment for television snacking; the relationships with
intake of candy, soda, and fast food were moderately at-
tenuated. Age and race/ethnicity modified relationships
of television viewing with soda and fast food intake and

with skipping breakfast.
Conclusion: Television viewing was associated with a
cluster of unhealthy eating behaviors in US adolescents
after adjustment for socioeconomic and behavioral
covariates.
Arch Pediatr Adolesc Med. 2012;166(5):465-472
D
IETARY INTAKES OF US
youth fall short of recom-
mendations for whole
fruit, whole grains, le-
gumes, and dark green
and orange vegetables and exceed recom-
mendations for fat, sodium, and added sug-
ars,1-3 increasing the risk of obesity and
chronic disease throughout the life-
span.4-12 Further understanding of the fac-
tors contributing to youth eating behav-
iors is necessary to improve dietary intakes
and associated health outcomes.
Television viewing (TVV) in youth has
been associated with unhealthy dietary in-
take and food preferences that may track
into early adulthood.13 Positive associa-
tions have been found with intakes of fast
food,14-18 soda,16,17,19-27 refined grains,28 and
energy-dense foods,17,20,25,27,29,30 as well as
with energy intake.6,16,25 In addition, TVV
has been inversely associated with fruit and
vegetable intake.16,19-21,25,27,31

A primary explanation for these find-
ings is the impact of exposure to food ad-
vertisements, which highlight primarily
energy-dense, nutrient-poor products and
influence food preferences and intake in
a variety of youth populations.32-34 Eating
while watching TV is another hypoth-
esized mechanism for observed relation-
ships between TVV and diet,17,29,35,36 al-
though, to our knowledge, this has not
been tested in a nationally representative
sample of US adolescents. In addition, de-
spite evidence of sociodemographic dif-
ferences in TVV37,38 and eating behav-
iors,39,40 few studies have investigated
potential effect modification by these vari-
ables.6,41,42
Author Affiliations: Eunice
Kennedy Shriver National
Institute of Child Health and
Human Development, Bethesda,
Maryland.
ARCH PEDIATR ADOLESC MED/ VOL 166 (NO. 5), MAY
2012 WWW.ARCHPEDIATRICS.COM
465
©2012 American Medical Association. All rights reserved.
Downloaded From:
http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/peds/
23403/ on 01/11/2017

The objective of this study was to examine associa-
tions between TVV and eating behaviors in a nationally
representative sample of US adolescents. We also sought
to investigate differences across age, sex, and race/
ethnicity and to ascertain the role of snacking while watch-
ing TV. Specific eating behaviors examined included in-
take of fruit, vegetables, sugar-sweetened soda, and sweets;
eating at a fast food restaurant; and skipping breakfast.
METHODS
STUDY DESIGN AND PARTICIPANTS
The Health Behaviour in School-aged Children Study is a sur-
vey of adolescents conducted every 4 years in the United States
beginning in 1995 to monitor and understand youth health be-
haviors, their social context, health, and well-being. Data col-
lection and procedural details have been published else-
where.43 We used data from the 2009-2010 US survey, which
included a nationally representative sample of 12 642 stu-
dents in grades 5 through 10 (mean [SD] age, 13.4 [0.09] years).
The sample was selected using a 3-stage stratified clustered
sam-
pling procedure, with school districts as the primary sampling
unit. The survey was administered in classrooms by indepen-
dent research staff. Black/African American and Hispanic stu-
dents were oversampled to produce reliable estimates for these
groups. There was 86.5% participation among 14 620 eligible
students. Questions were validated in several countries in tar-
get age groups. Youth assent and active or passive parental con-
sent were obtained as required by participating school dis-
tricts. The institutional review board of the Eunice Kennedy
Shriver National Institute of Child Health and Human Devel-
opment approved study procedures.

VARIABLES
Television viewing time was assessed by asking (separately for
weekdays and weekends), “About how many hours a day do
you usually watch television (including videos and DVDs) in
your free time?”. Response categories included “none at all,”
“about half an hour a day,” “about 1 hour a day,” “about 2 hours
a day,” and so on to “about 7 or more hours a day.” Similar
questions and identical responses were used to assess com-
puter use. The survey asked (separately for weekdays and week-
ends), “About how many hours a day do you usually play games
on a computer or games console (PlayStation, Xbox, Game-
Cube, etc) in your free time?” and “About how many hours a
day do you usually use a computer for chatting online, Inter-
net, e-mailing, homework, etc, in your free time?”. Weekday
and weekend responses were combined to obtain average daily
hours of TVV and computer use. The combined variable for
computer use for games and other purposes was used in final
analyses because of the lack of an independent contribution of
separate computer use variables in regression models.
Eating behaviors were assessed by asking, “How many times
a week do you usually eat or drink . . . ” followed by “fruits,”
“vegetables,” “sweets (candy or chocolate),” and “Coke or other
soft drinks that contain sugar” (soda), with response options
for “never,” “less than once a week,” “once a week,” “2 to 4
days a week,” “5 to 6 days a week,” “once a day, every day,”
and “every day, more than once.” Skipping breakfast was as-
sessed by asking (separately for weekdays and weekends), “How
often do you usually have breakfast (more than a glass of milk
or fruit juice)?”. Weekday response options included “I never
have breakfast during weekdays,” “1 day,” “2 days,” “3 days,”
“4 days,” or “5 days.” Weekend response options included “I
never have breakfast during the weekend,” “I usually have
break-

fast on only 1 day of the weekend (Saturday OR Sunday),” and
“I usually have breakfast on both weekend days (Saturday AND
Sunday).” Weekday and weekend responses were combined to
create a single variable indicating whether breakfast was
skipped
at least 1 day per week. Eating at a fast food restaurant was as-
sessed by asking, “How often do you eat in a fast food restau-
rant (for example, McDonald’s, KFC, Pizza Hut, Taco Bell)?,”
with response options for “never,” “rarely (less than once a
month),” “once a month,” “2 to 3 times a month,” “once a
week,”
“2 to 4 days a week,” and “5 or more days a week.”
Snacking during TVV and computer use were assessed in
participants in grades 7 through 10 by asking, “How often do
you eat a snack while you . . . ” followed by “watch TV (in-
cluding videos and DVDs)” and “work or play on a computer
or games console,” with response options for “never,” “less than
once a week,” “1 to 2 days a week,” “3 to 4 days a week,” “5 to
6 days a week,” and “every day.” Leisure-time vigorous physi-
cal activity was assessed by asking, “Outside school hours: how
many hours a week do you usually exercise in your free time
so much that you get out of breath or sweat?,” with responses
for “none,” “about half an hour,” “about 1 hour,” “about 2 to
3 hours,” “about 4 to 6 hours,” and “7 hours or more.”
Students reported age, sex, race (“What do you consider your
race to be?,” with response options for “black or African Ameri-
can” [hereafter referred to as black], “white,” “Asian,” “Ameri-
can Indian or Alaska native,” “Native Hawaiian or other Pa-
cific Islander,” and “other” with a blank space provided), and
ethnicity (“What do you consider your ethnicity to be?,” with
response options for “Hispanic or Latino” and “not Hispanic
or Latino”). We combined responses to create a 4-category race/
ethnicity variable: non-Hispanic white (white), non-Hispanic
black/African American (black/African American), Hispanic,

and
other. Socioeconomic status was assessed by the Family Afflu-
ence Scale, a measure with demonstrated content and external
validity44 that was developed for the Health Behaviour in
School-
aged Children Study on the basis of responses to questions
about
computer and automobile ownership, whether the student shares
a bedroom, and frequency of family vacations.
STATISTICAL ANALYSIS
Descriptive statistics were calculated and compared by sex, age,
and race/ethnicity using regression analysis for continuous out-
comes and the Pearson/Wald test for binary outcomes. Mul-
tiple logistic regressions assessed relationships between TVV
(hours per day) and daily intake of fruit, vegetables, sweets,
and soda, as well as skipping breakfast at least 1 day per week
and eating at a fast food restaurant at least 1 day per week. We
examined interactions of TVV with age, sex, and race/ethnicity
with multiplicative interaction terms and stratified analyses
where
warranted. Relationships between TVV and dietary behaviors
after controlling for frequency of TV and computer snacking
were explored in grades 7 through 10; we further examined the
interaction of TVV and TV snacking in this subpopulation. All
independent variables were continuous except sex and race/
ethnicity. Analyses accounted for the complex survey sam-
pling design using STATA, version 11 (StataCorp).
RESULTS
There were significant sociodemographic differences in
most eating behaviors except fast food, which was not
different by sex but was more frequent for older (aged
�13 years) vs younger (�13 years) participants and for

racial/ethnic groups compared with white youth (P � .001)
466
Downloaded From:
23403/ on 01/11/2017
(Table 1 and Table 2). Odds of daily intake of fruit
and vegetables were higher for younger than older par-
ticipants, for girls compared with boys, and for white and
other groups compared with black and Hispanic youth.
Odds of daily intake of sweets were highest for older vs
younger youth, for girls vs boys, and for black youth com-
pared with other racial/ethnic groups. Odds of drinking
soda at least daily were highest for older vs younger youth,
for boys vs girls, and for black and Hispanic youth com-
pared with other racial/ethnic groups. Skipping break-
fast was more common for older than younger partici-
pants, for girls vs boys, and for black, Hispanic, and
“other” youth vs white participants.
Television viewing did not differ by age group but was
lower for girls than boys (approximately0.1 h/d) and higher
for black participants vs other racial/ethnic groups (ap-
proximately 1.0-1.4 h/d). Computer use was lower for
younger vs older participants, for girls vs boys, and for white
vs other racial/ethnic groups. Television snacking did not
differ by sex or age but was more frequent for black youth
vs other racial/ethnic groups. Computer snacking was less

frequent for girls than boys (approximately 0.25 d/wk) and
more frequent for black participants vs other ethnic/
racial groups (approximately 0.8-1.2 d/wk). Leisure-time
vigorous physical activity was lower for girls vs boys (ap-
proximately 0.4 h/wk) and higher for white participants
vs other racial/ethnic groups (approximately0.2-0.4 h/wk).
Television viewing time was inversely related to fruit
and vegetable intake and positively related to sweets and
soda intake, fast food intake, and skipping breakfast in
models adjusted for computer use, physical activity, age,
sex, family affluence, and race/ethnicity (Table 3). Re-
lationships differed by age and race/ethnicity (Figure);
we did not find differences by sex. Relationships with soda
(Figure, A) and fast food (Figure, B) intake and with skip-
ping breakfast (Figure, C) differed according to race/
ethnicity, whereas relationships with skipping break-
fast differed as well by age (Figure, C). There was a positive
relationship of TVV with soda intake for all race/
ethnicities, although the relationship was weaker for black
(adjusted odds ratio, 1.13; 95% CI, 1.05-1.21) com-
pared with white (1.30; 1.22-1.39) youth (P = .001). Tele-
vision viewing time was positively related to fast food in
white youth (adjusted odds ratio, 1.19; 95% CI, 1.12-
1.27) and in Hispanic youth (1.10; 1.05-1.15); the rela-
tionship with fast food was not significant for black youth.
Skipping breakfast was not related to TVV in youth aged
13 or older (1.03; 0.99-1.07). In participants younger than
13, there was a positive relationship between TVV and
skipping breakfast in white youth (adjusted odds ratio,
1.21; 95% CI, 1.12-1.31) but an inverse relationship in
black youth and no significant relationship in Hispanic
youth (1.01; 0.97-1.35) (Figure, C). Estimates were simi-
lar between white and other participants.

Television viewing time and TV snacking were inde-
pendently related to eating behaviors in participants in
grades 7 through 10 in adjusted models (Table 4). Tele-
vision snacking was positively related to daily intake of
Table 1. Participant Characteristics by Sex and Age
Characteristic
Total
(N = 12 642)
Sex Age, y
Male
(n = 6502)
Female
(n = 6136) P Value a
�13
(n = 5152)
�13
Family Affluence Scale score, mean (SE) 5.4 (0.06) 5.3 (0.06)
5.4 (0.07) .05 5.4 (0.07) 5.3 (0.06) .25
Food intake, No. (%), times per day
Fruit
�1 5297 (44.6) 2562 (41.8) 2734 (47.5)
�.001
2412 (51.4) 2832 (39.6)

�.001
�1 6724 (54.4) 3577 (58.2) 3145 (52.5) 2392 (48.6) 4299 (60.4)
Vegetables
�1 4382 (38.7) 2067 (35.2) 2314 (42.3)
�.001
1886 (41.7) 2450 (36.2)
�.001
�1 7459 (61.3) 3955 (64.8) 3502 (57.7) 2820 (58.3) 4601 (63.8)
Sweets
�1 2941 (24.7) 1352 (22.4) 1588 (27.1)
.001
1084 (22.5) 1841 (26.3)
�.001
�1 8863 (75.3) 4656 (77.6) 4205 (72.9) 3619 (77.5) 5197 (73.7)
Soda
�1 3637 (30.1) 1881 (30.9) 1754 (29.3)
.01
1277 (26.2) 2347 (33.2)
�.001
�1 8312 (69.9) 4198 (69.1) 4113 (70.7) 3504 (73.8) 4738 (66.8)
Skipping breakfast, No. (%), d/wk
�1 799 (54.8) 3239 (51.8) 3558 (58.0)
�.001
2332 (45.9) 4443 (61.9)

�.001
�1 5236 (45.2) 2879 (48.2) 2355 (42.0) 2529 (54.1) 2644 (38.1)
Fast food, No. (%), d/wk
�1 4256 (34.1) 2215 (35.0) 2038 (33.2)
.17
1574 (31.7) 2666 (36.3)
.02
�1 8218 (65.9) 4184 (65.0) 4034 (66.8) 3465 (68.3) 4682 (63.7)
Television viewing, mean (SE), h/d 2.4 (0.06) 2.5 (0.06) 2.4
(0.07) .005 2.5 (0.07) 2.4 (0.06) .79
Computer use, mean (SE), h/d 2.8 (0.08) 3.1 (0.10) 2.5 (0.08)
�.001 2.5 (0.09) 3.0 (0.09) �.001
TV snacking, mean (SE), d/wk b 3.3 (0.06) 3.3 (0.07) 3.3 (0.08)
.58 NA NA NA
Computer snacking, mean (SE), d/wk b 2.4 (0.07) 2.54 (0.08)
2.24 (0.08) �.001 NA NA NA
Physical activity, mean (SE), h/wk 1.6 (0.03) 1.8 (0.03) 1.4
(0.04) �.001 1.6 (0.04) 1.6 (0.03) .21
Abbreviation: NA, not applicable.
a By t test or Pearson �2 test of overall association.
b Assessed only in respondents in grades 7 through 10.
467
Downloaded From:
23403/ on 01/11/2017

fruit, candy, and soda and to fast food but was unrelated
to vegetable intake and skipping breakfast. Indepen-
dent relationships of TVV with intake of candy and soda
and with fast food were moderately attenuated but re-
mained statistically significant after adjustment for TV
and computer snacking (relationships with eating be-
haviors unadjusted for TV and computer snacking were
similar in participants in grades 7 through 10 as in the
overall sample, so separate estimates are not reported).
Independent relationships of TVV with fruit and veg-
etable intake and skipping breakfast were essentially un-
changed after adjustment for TV and computer snack-
ing. We found no interactions between TVV and TV
snacking for any eating behavior (results not shown).
COMMENT
This study provides estimates of associations of TVV and
eating behaviors in a diverse representative sample of US
adolescents. Television viewing time was associated with
lower odds of consuming fruit or vegetables daily and
Table 2. Participant Characteristics by Race/Ethnicity
Characteristic
White
(n = 5334)
Black
(n = 2302)
Hispanic

(n = 3407)
Other
Family Affluence Scale score, mean (SE) 5.7 (0.06) 5.1 (0.07)
4.9 (0.08) 5.3 (0.09) �.001
Food intake, No. (%), times per day
Fruit
�1 2339 (46.0) 924 (41.3) 1337 (41.9) 634 (46.9)
.02
�1 2821 (54.0) 1222 (58.7) 1868 (58.1) 750 (53.1)
Vegetables
�1 2072 (41.5) 785 (37.0) 930 (30.5) 542 (41.1)
�.001
�1 3032 (58.5) 1327 (63.0) 822 (69.5) 822 (58.9)
Sweets
�1 1068 (21.8) 754 (34.9) 759 (23.7) 333 (24.7)
�.001
�1 4008 (78.2) 1338 (65.1) 2394 (76.3) 1034 (75.3)
Soda
�1 1294 (26.5) 879 (40.7) 1102 (34.5) 337 (25.1)
�.001
�1 3849 (73.5) 1247 (59.3) 2090 (65.6) 1033 (74.9)
Skipping breakfast, No. (%), d/wk
�1 2634 (50.4) 1326 (60.9) 1967 (61.0) 816 (56.1)

�.001
�1 2549 (49.6) 822 (39.1) 1230 (39.0) 569 (43.9)
Fast food, No. (%), d/wk
�1 1528 (29.1) 990 (44.8) 1253 (38.6) 446 (33.7)
�.001
�1 3756 (70.9) 1276 (55.2) 2101 (61.4) 992 (66.3)
Television viewing, mean (SE), h/d 2.0 (0.05) 3.4 (0.07) 2.7
(0.06) 2.4 (0.08) �.001
Computer use, mean (SE), h/d 2.4 (0.08) 3.6 (0.11) 3.1 (0.09)
3.1 (0.13) �.001
Television snacking, mean (SE), d/wk 3.0 (0.06) 4.3 (0.08) 3.4
(0.10) 3.0 (0.10) .02
Computer snacking, mean (SE), d/wk 2.0 (0.06) 3.2 (0.10) 2.6
(0.12) 2.4 (0.14) �.001
Physical activity, mean (SE), h/wk 1.8 (0.03) 1.4 (0.04) 1.4
(0.03) 1.6 (0.06) �.001
a By t test or Pearson �2 test of overall association.
Table 3. Odds Ratios From Multiple Logistic Regressions
Predicting Eating Behaviors a
Characteristic
Eating Behavior, Odds Ratio (95% CI)
Fruit
(n = 9196)
Vegetables
(n = 9069)
Candy

(n = 9047)
Soda
(n = 9155)
Fast Food
(n = 9513)
Skipping Breakfast
(n = 9322)
Television viewing, h/d 0.92 (0.88-0.96) 0.95 (0.91-1.00) 1.18
(1.14-1.23) 1.24 (1.20-1.29) 1.14 (1.09-1.19) 1.06 (1.02-1.10)
Computer/games, h/d 0.99 (0.96-1.02) 0.97 (0.95-1.00) 1.12
(1.09-1.15) 1.13 (1.10-1.16) 1.06 (1.04-1.09) 1.04 (1.01-1.07)
Physical activity, h/wk 1.27 (1.22-1.33) 1.25 (1.19-1.31) 0.96
(0.90-1.02) 0.93 (0.88-0.99) 1.00 (0.94-1.05) 0.95 (0.90-0.99)
Age, y 0.86 (0.82-0.89) 0.93 (0.90-0.97) 1.07 (1.02-1.12) 1.09
(1.04-1.15) 1.06 (1.01-1.12) 1.24 (1.19-1.30
Sex
Male 1.00 [Reference] 1.00 [Reference] 1.00 [Reference] 1.00
[Reference] 1.00 [Reference] 1.00 [Reference]
Female 1.29 (1.14-1.47) 1.43 (1.27-1.61) 1.32 (1.15-1.51) 0.96
(0.84-1.08) 0.95 (0.82-1.10) 1.42 (1.29-1.57)
Family Affluence Scale score 1.13 (1.09-1.18) 1.10 (1.05-1.15)
1.00 (0.94-1.06) 0.90 (0.86-0.94) 1.08 (1.04-1.13) 0.87 (0.84-
0.91)
Race/ethnicity
White 1.00 [Reference] 1.00 [Reference] 1.00 [Reference] 1.00
Black 1.10 (0.93-1.29) 1.08 (0.91-1.28) 1.37 (1.11-1.69) 1.16
(0.94-1.43) 1.62 (1.28-2.06) 1.13 (0.94-1.37)
Hispanic 1.05 (0.87-1.28) 0.78 (0.65-0.94) 0.88 (0.73-1.05) 1.01

(0.80-1.27) 1.43 (1.18-1.75) 1.26 (1.07-1.48)
Other 1.09 (0.87-1.36) 1.00 (0.79-1.27) 0.98 (0.79-1.20) 0.73
(0.61-0.88) 1.16 (0.91-1.46) 1.22 (1.03-1.43)
a Separate models predicting at least daily intake of fruit,
vegetables, candy, and soda and at least weekly eating at a fast
food restaurant and skipping breakfast.
Models are adjusted for all characteristics.
468
Downloaded From:
23403/ on 01/11/2017
higher odds of consuming candy and sugar-sweetened
soda daily, skipping breakfast at least 1 day per week, and
eating at a fast food restaurant at least 1 day per week in
models adjusted for computer use, physical activity, age,
sex, race/ethnicity, and family affluence. The relation-
ship of TVV with this unhealthy combination of eating
behaviors may contribute to the documented relation-
ship of TVV with cardiometabolic risk factors.4,45,46
Many of our prevalence estimates vary somewhat from
previously reported findings, likely because of differ-
ences in assessment methods. Our estimates of TVV (2.4
h/d) and computer use (2.8 h/d) are similar to and more
than an hour greater, respectively, than previous esti-
mates.38,47 Reported vigorous physical activity in our

sample is substantially lower than previous estimates of
moderate to vigorous physical activity.38,47-49 Report of
skipping breakfast in our sample (54.8% �1 d/wk) was
higher than a recent National Health and Nutrition Ex-
amination Survey estimate (20%).50 In addition, our es-
timate of frequency of eating at a fast food restaurant is
lower than findings of a study of California adolescents,
in which 46% reported eating fast food at least 2 times
per week51 and recent estimates from a nationally rep-
resentative sample reporting a mean 2 to 3 fast food meals
per week,52,53 although multiple fast food meals may be
consumed on a single day.
Relationships of TVV with soda intake, fast food, and
breakfast behaviors differed according to race/ethnicity.
The relationship with soda intake was attenuated for black
participants compared with white participants, al-
though relationships were positive for all groups. In ad-
dition, the relationship of TVV with fast food was not sig-
nificant for black youth, whereas this relationship was
positive in other racial/ethnic groups. We found differ-
ential relationships of TVV with skipping breakfast by
both race/ethnicity and age. Among respondents younger
than 13, TVV was positively related to skipping break-
fast in white adolescents but inversely related to skip-
1.5
1.1
1.3
1.2
1.4

1.0
0.9
0.8
White Black Hispanic Other
Race/Ethnicity
Ad
ju
st
ed
O
R
A
1.5
1.1
1.3
1.2
1.4
1.0
0.9
0.8

Race/Ethnicity
Ad
ju
st
ed
O
R
B
1.5
1.1
1.3
1.2
1.4
1.0
0.9
0.8
Race/Ethnicity
Ad
ju

st
ed
O
R
C
Figure. Associations of television viewing and eating behaviors
by race/ethnicity and age. Adjusted odds ratios (ORs) of (A)
drinking soda (�1 instance per day)
among all youth (A), eating fast food (�1 d/wk) among all
youth (B), and skipping breakfast at least 1 day per week among
youth younger than 13 (C) associated
with television viewing time (hours per day) by race/ethnicity.
Error bars indicate 95% CIs. Associations with soda intake were
significantly different for black
compared with white youth (P = .001). Associations with fast
food were significantly different for black (P = .001), Hispanic
(P � .001), and “other” (P � .001) youth
compared with white youth. Associations with skipping
breakfast were significantly different for black (P � .001) and
Hispanic (P = .006) youth compared with
white youth. Odds ratios were adjusted for computer use,
physical activity, age, sex, and family affluence.
Table 4. Logistic Regressions Predicting Eating Behaviors of
Participants in Grades 7 Through 10 a
Characteristic
Eating Behavior, Odds Ratio (95% CI)
Fruit Vegetable Candy Soda Fast Food Skipping Breakfast

Television viewing, h/d 0.91 (0.87-0.95) 0.94 (0.90-0.98) 1.06
(1.01-1.11) 1.13 (1.08-1.19) 1.09 (1.04-1.14) 1.05 (1.01-1.10)
Computer/games, h/d 0.96 (0.93-0.99) 0.97 (0.94-1.00) 1.03
(0.99-1.07) 1.06 (1.02-1.10) 1.01 (0.98-1.04) 1.04 (1.00-1.07)
Physical activity, h/wk 1.24 (1.19-1.30) 1.23 (1.17-1.29) 0.95
(0.88-1.02) 0.94 (0.88-1.01) 1.01 (0.94-1.08) 0.97 (0.91-1.03)
Television snacking, d/wk 1.06 (1.02-1.10) 1.02 (0.99-1.05)
1.20 (1.16-1.24) 1.15 (1.11-1.18) 1.09 (1.06-1.13) 0.99 (0.96-
1.02)
Computer snacking, d/wk 1.02 (0.99-106) 1.02 (0.98-1.05) 1.12
(1.08-1.16) 1.13 (1.09-1.17) 1.09 (1.06-1.12) 0.99 (0.96-1.03)
Age, y 0.87 (0.81-0.93) 0.94 (0.88-1.01) 1.07 (0.99-1.14) 1.09
(1.03-1.16) 1.10 (1.02-1.18) 1.18 (1.12-1.25)
Sex
Male 1.00 [Reference] 1.00 [Reference] 1.00 [Reference] 1.00
Female 1.21 (1.02-1.43) 1.31 (1.14-1.49) 1.26 (1.06-1.49) 0.95
(0.82-1.10) 0.94 (0.78-1.13) 1.61 (1.44-1.80)
Family Affluence Scale score 1.13 (1.07-1.19) 1.08 (1.03-1.13)
0.99 (0.93-1.06) 0.90 (0.86-0.95) 1.13 (1.08-1.19) 0.90 (0.85-
0.94)
Race/ethnicity
White 1.00 [Reference] 1.00 [Reference] 1.00 [Reference] 1.00
Black 1.19 (0.99-1.42) 1.06 (0.87-1.30) 1.48 (1.21-1.80) 1.20
(0.96-1.50) 1.63 (1.26-2.10) 1.29 (1.08-1.55)
Hispanic 1.02 (0.81-1.28) 0.75 (0.61-0.91) 0.96 (0.78-1.20) 1.10
(0.86-1.41) 1.65 (1.28-2.11) 1.27 (1.08-1.51)
Other 1.14 (0.85-1.54) 0.93 (0.69-1.25) 0.90 (0.69-1.17) 0.68
(0.53-0.85) 1.28 (0.96-1.71) 1.38 (1.12-1.71)
a Separate models predicting at least daily intake of fruit,
vegetables, candy, and soda and at least weekly eating at a fast

food restaurant and skipping breakfast.
Models are estimated for participants in grades 7 through 10
only and are adjusted for all characteristics.
469
Downloaded From:
23403/ on 01/11/2017
ping breakfast in black adolescents. There was no rela-
tionship in Hispanic adolescents or in those aged 13 or
older of any race/ethnicity. These findings are poten-
tially explained by a relationship of TVV with sleep du-
ration and morning sleepiness in youth that may vary with
respect to factors associated with child age and race/
ethnicity.54 Our results indicated that relationships of TVV
and intake of fruit and vegetables did not differ accord-
ing to race/ethnicity, contrary to findings of a previous
study using data from the 1999 national Youth Risk Be-
havior Survey.42 In addition, relationships were not dif-
ferent according to sex, as has been suggested in previ-
ous research.6,41 Likely because of the heterogeneous
make-up of the “other” subgroup, we found few differ-
ences from white participants, apart from overall lower
odds of drinking soda daily and higher odds of skipping
breakfast. Differential associations by race/ethnicity sug-
gest the importance of social, cultural, or other contex-
tual factors42,55 that may contribute to variability in as-
sociations of TVV with eating behaviors. However, despite

differential associations by sociodemographic factors, our
findings of associations of TVV and eating behaviors across
all races/ethnicities were nearly universally inconsistent
with healthful eating patterns.
Our results demonstrated relationships of TV snack-
ing with several eating behaviors after adjustment for TVV,
computer use, physical activity, computer snacking, and
sociodemographic characteristics. Although a measure
of overall snacking frequency was not available, our find-
ing of independent relationships of eating behaviors with
both TV and computer snacking suggest that these mea-
sures represent specific modes of snacking rather than
serving as proxies for overall snacking. Results indicat-
ing positive relationships of TV snacking with candy and
soda intake and with eating at a fast food restaurant sup-
port previous research documenting adverse associa-
tions of eating while watching TV and nutritional qual-
ity of foods consumed.20,21,56,57 Previous experimental
research has also demonstrated a positive association of
eating while watching TV on food quantity consumed in
undergraduates29 and children.36 This evidence sug-
gests TV snacking may influence the amount and qual-
ity of food consumed.
Relationships of TVV with eating behaviors were es-
sentially unchanged after adjustment for TV snacking,
and we found no evidence of effect modification be-
tween TVV and TV snacking, suggesting that TV snack-
ing does not fully account for relationships between TVV
and eating behaviors. This evidence, together with our
finding that TVV was strongly and positively related to
the intake of highly advertised foods and either weakly
related or unrelated to rarely advertised foods, supports
a hypothesized influence of TV food advertisement ex-
posure on dietary intake consistent with experimental re-

search.58,59 Although data indicate a recent decrease in
adolescents’ exposure to candy and soda advertise-
ments, exposure to fast food restaurant advertisements
has increased.60 In addition, the nutritional quality of ad-
vertised foods is poor,33,34,61 and fruit and vegetable ad-
vertisements are essentially nonexistent.34,60,62-64 Alter-
natively, these findings may reflect a clustering of
unhealthy behaviors related to unobserved factors, such
as an increased tendency to snack during essentially un-
occupied time or preferential selection of foods condu-
cive to snacking. Associations of TVV and TV snacking
with eating behaviors were evident after adjustment for
several covariates intended to reduce confounding by char-
acteristics such as general preference for health and re-
lated behaviors, although the potential for residual bias
persists in observational research.
Interpretation of these findings must take into ac-
count the relative strengths and weaknesses of this study.
Our findings rely on a brief assessment of dietary intake
rather than a more detailed method (eg, food frequency
questionnaire or dietary recall), which may have hin-
dered our ability to detect significant relationships. An
important limitation is the cross-sectional study design,
which does not allow for determination of causality be-
cause of the inability to establish temporality and the po-
tential for confounding by unobserved factors. Thus, we
cannot rule out, for example, the possibility that youth
with poor dietary behaviors, including a greater ten-
dency toward snacking, are more likely to watch TV. How-
ever, previous interventions to reduce TVV in children
and adolescents have shown evidence of a beneficial in-
fluence on eating while watching TV65,66 as well as on total
energy intake41,67 and fruit and vegetable intake,41,68 sup-
porting a causal role of TVV on dietary intake. In addi-

tion, the internal validity of our findings is strength-
ened by our adjustment for several behavioral and
socioeconomic confounders. Our study is further strength-
ened by the high participation rate and the use of a rep-
resentative sample of US adolescents, including over-
sampling of black and Hispanic youth to enable subgroup
analysis.
These findings show that TVV is related to a cluster
of unhealthy eating behaviors in US adolescents after con-
trolling for TV snacking, computer snacking, socioeco-
nomic variables, computer use, and physical activity. Re-
lationships of TVV with eating behaviors were modified
by age and race/ethnicity, suggesting the importance of
cultural or social factors. Future research should eluci-
date the independent contributions of TVV, food adver-
tising, and TV snacking on dietary intake in this popu-
lation. If these relationships are causal, efforts to reduce
TVV or to modify the nutritional content of advertised
foods may lead to substantial improvements in adoles-
cents’ dietary intake.
Accepted for Publication: October 31, 2011.
Correspondence: Leah M. Lipsky, PhD, MHS, Eunice
Kennedy Shriver National Institute of Child Health and
Human Development, 6100 Executive Blvd, Ste 7B13,
Bethesda, MD 20852 ([email protected]).
Author Affiliations: Dr Lipsky had full access to all the
data in the study and takes responsibility for the integrity
of the data and the accuracy of the data analysis. Study con-
cept and design: Lipsky and Iannotti. Acquisition of data:
Iannotti. Analysis and interpretation of data: Lipsky and Ian-
notti. Drafting of the manuscript: Lipsky and Iannotti. Criti-
cal revision of the manuscript for important intellectual con-
tent: Lipsky and Iannotti. Obtained funding: Iannotti.
Administrative, technical, and material support: Iannotti.

Financial Disclosure: None reported.
470
Downloaded From:
23403/ on 01/11/2017
Funding/Support: This research was funded by grant
HHSN2672008000009C from the Intramural Research
Program of the Eunice Kennedy Shriver National Insti-
tute of Child Health and Human Development and the
Maternal and Child Health Bureau of the Health Re-
sources and Services Administration.
REFERENCES
1. Cohen DA, Sturm R, Scott M, Farley TA, Bluthenthal R. Not
enough fruit and veg-
etables or too many cookies, candies, salty snacks, and soft
drinks? Public Health
Rep. 2010;125(1):88-95.
2. Fungwe T, Guenther PM, Juan W, Hiza H, Lino M. The
quality of children’s diets in
2003-04 as measured by the Healthy Eating Index—2005.
Washington, DC: US
Dept of Agriculture Center for Nutrition Policy and Promotion;
2009. Report 43.

3. Reedy J, Krebs-Smith SM. Dietary sources of energy, solid
fats, and added sug-
ars among children and adolescents in the United States. J Am
Diet Assoc. 2010;
110(10):1477-1484.
4. Aadahl M, Kjaer M, Jørgensen T. Influence of time spent on
TV viewing and vig-
orous intensity physical activity on cardiovascular biomarkers:
the Inter 99 study.
Eur J Cardiovasc Prev Rehabil. 2007;14(5):660-665.
5. Andersen RE, Crespo CJ, Bartlett SJ, Cheskin LJ, Pratt M.
Relationship of physi-
cal activity and television watching with body weight and level
of fatness among
children: results from the third National Health and Nutrition
Examination Survey.
JAMA. 1998;279(12):938-942.
6. Crespo CJ, Smit E, Troiano RP, Bartlett SJ, Macera CA,
Andersen RE. Television
watching, energy intake, and obesity in US children: results
from the third Na-
tional Health and Nutrition Examination Survey, 1988-1994.
Arch Pediatr Ado-
lesc Med. 2001;155(3):360-365.
7. Gortmaker SL, Must A, Sobol AM, Peterson K, Colditz GA,
Dietz WH. Television
viewing as a cause of increasing obesity among children in the
United States,
1986-1990. Arch Pediatr Adolesc Med. 1996;150(4):356-362.
8. Healy GN, Dunstan DW, Salmon J, Shaw JE, Zimmet PZ,
Owen N. Television time

and continuous metabolic risk in physically active adults. Med
Sci Sports Exerc.
2008;40(4):639-645.
9. Hu FB. Sedentary lifestyle and risk of obesity and type 2
diabetes. Lipids. 2003;
38(2):103-108.
10. Jakes RW, Day NE, Khaw KT, et al. Television viewing and
low participation in
vigorous recreation are independently associated with obesity
and markers of
cardiovascular disease risk: EPIC-Norfolk population-based
study. Eur J Clin Nutr.
2003;57(9):1089-1096.
11. Martinez-Gomez D, Rey-López JP, Chillón P, et al; AVENA
Study Group. Exces-
sive TV viewing and cardiovascular disease risk factors in
adolescents: the AVENA
cross-sectional study. BMC Public Health. 2010;10:274.
doi:10.1186
/1471-2458-10-274.
12. Viner RM, Cole TJ. Television viewing in early childhood
predicts adult body mass
index. J Pediatr. 2005;147(4):429-435.
13. Harris JL, Bargh JA. Television viewing and unhealthy diet:
implications for chil-
dren and media interventions. Health Commun. 2009;24(7):660-
673.
14. French SA, Story M, Neumark-Sztainer D, Fulkerson JA,
Hannan P. Fast food res-
taurant use among adolescents: associations with nutrient

intake, food choices
and behavioral and psychosocial variables. Int J Obes Relat
Metab Disord. 2001;
25(12):1823-1833.
15. Larson NI, Neumark-Sztainer DR, Story MT, Wall MM,
Harnack LJ, Eisenberg
ME. Fast food intake: longitudinal trends during the transition
to young adult-
hood and correlates of intake. J Adolesc Health. 2008;43(1):79-
86.
16. Miller SA, Taveras EM, Rifas-Shiman SL, Gillman MW.
Association between tele-
vision viewing and poor diet quality in young children. Int J
Pediatr Obes. 2008;
3(3):168-176.
17. Pearson N, Ball K, Crawford D. Mediators of longitudinal
associations between
television viewing and eating behaviours in adolescents. Int J
Behav Nutr Phys
Act. 2011;8:23. doi:10.1186/1479-5868-8-23.
Medline:21450065.
18. Scully M, Dixon H, Wakefield M. Association between
commercial television ex-
posure and fast-food consumption among adults. Public Health
Nutr. 2009;
12(1):105-110.
19. Barr-Anderson DJ, van den Berg P, Neumark-Sztainer D,
Story M. Characteris-
tics associated with older adolescents who have a television in
their bedrooms.
Pediatrics. 2008;121(4):718-724.

20. Coon KA, Goldberg J, Rogers BL, Tucker KL. Relationships
between use of tele-
vision during meals and children’s food consumption patterns.
Pediatrics. 2001;
107(1):E7. doi:10.1542/peds.107.1.e7.
21. Feldman S, Eisenberg ME, Neumark-Sztainer D, Story M.
Associations between
watching TV during family meals and dietary intake among
adolescents. J Nutr
Educ Behav. 2007;39(5):257-263.
22. Kremers SP, van der Horst K, Brug J. Adolescent screen-
viewing behaviour is
associated with consumption of sugar-sweetened beverages: the
role of habit
strength and perceived parental norms. Appetite.
2007;48(3):345-350.
23. Péneau S, Mekhmoukh A, Chapelot D, et al. Influence of
environmental factors
on food intake and choice of beverage during meals in
teenagers: a laboratory
study. Br J Nutr. 2009;102(12):1854-1859.
24. Ranjit N, Evans MH, Byrd-Williams C, Evans AE,
Hoelscher DM. Dietary and ac-
tivity correlates of sugar-sweetened beverage consumption
among adolescents
[published online September 27, 2010]. Pediatrics.
2010;126(4):e754-e761. doi:
10.1542/peds.2010-1229.
25. Utter J, Neumark-Sztainer D, Jeffery R, Story M. Couch

potatoes or french fries:
are sedentary behaviors associated with body mass index,
physical activity, and
dietary behaviors among adolescents? J Am Diet Assoc.
2003;103(10):1298-
1305.
26. Van den Bulck J, Van Mierlo J. Energy intake associated
with television viewing
in adolescents: a cross sectional study. Appetite.
2004;43(2):181-184.
27. Vereecken CA, Todd J, Roberts C, Mulvihill C, Maes L.
Television viewing be-
haviour and associations with food habits in different countries.
Public Health
Nutr. 2006;9(2):244-250.
28. Ng C, Young TK, Corey PN. Associations of television
viewing, physical activity
and dietary behaviours with obesity in aboriginal and
nonaboriginal Canadian youth.
Public Health Nutr. 2010;13(9):1430-1437.
29. Blass EM, Anderson DR, Kirkorian HL, Pempek TA, Price
I, Koleini MF. On the
road to obesity: television viewing increases intake of high-
density foods. Physiol
Behav. 2006;88(4-5):597-604.
30. Thomson M, Spence JC, Raine K, Laing L. The association
of television viewing
with snacking behavior and body weight of young adults. Am J
Health Promot.
2008;22(5):329-335.

31. Boynton-Jarrett R, Thomas TN, Peterson KE, Wiecha J,
Sobol AM, Gortmaker SL.
Impact of television viewing patterns on fruit and vegetable
consumption among
adolescents. Pediatrics. 2003;112(6, pt 1):1321-1326.
32. Andreyeva T, Kelly IR, Harris JL. Exposure to food
advertising on television: as-
sociations with children’s fast food and soft drink consumption
and obesity. Econ
Hum Biol. 2011;9(3):221-233.
33. Institute of Medicine Food and Nutrition Board. Food
Marketing to Children and
Youth: Threat or Opportunity? Washington, DC: National
Academies Press; 2006.
34. Kunkel D, McKinley C, Wright P. The impact of industry
self-regulation on the
nutritional quality of foods advertised on television to children.
Children Now web-
site.
http://www.childrennow.org/uploads/documents/adstudy_2009.p
df. Ac-
cessed February 8, 2012.
35. Hill JO, Wyatt HR, Reed GW, Peters JC. Obesity and the
environment: where do
we go from here? Science. 2003;299(5608):853-855.
36. Temple JL, Giacomelli AM, Kent KM, Roemmich JN,
Epstein LH. Television watch-
ing increases motivated responding for food and energy intake
in children. Am J
Clin Nutr. 2007;85(2):355-361.

37. Giammattei J, Blix G, Marshak HH, Wollitzer AO, Pettitt
DJ. Television watching
and soft drink consumption: associations with obesity in 11- to
13-year-old
schoolchildren. Arch Pediatr Adolesc Med. 2003;157(9):882-
886.
38. Roberts DF, Foehr UG, Rideout V. Generation M: media in
the lives of 8–18-year-olds.
Kaiser Family Foundation website.
http://www.kff.org/entmedia/loader.cfm?url=
/commonspot/security/getfile.cfm&PageID=51809. Accessed
February 8, 2012.
39. Lorson BA, Melgar-Quinonez HR, Taylor CA. Correlates of
fruit and vegetable in-
takes in US children. J Am Diet Assoc. 2009;109(3):474-478.
40. Patrick H, Nicklas TA. A review of family and social
determinants of children’s
eating patterns and diet quality. J Am Coll Nutr. 2005;24(2):83-
92.
41. Gortmaker SL, Peterson K, Wiecha J, et al. Reducing
obesity via a school-based
interdisciplinary intervention among youth: Planet Health. Arch
Pediatr Adolesc
Med. 1999;153(4):409-418.
42. Lowry R, Wechsler H, Galuska DA, Fulton JE, Kann L.
Television viewing and its
associations with overweight, sedentary lifestyle, and
insufficient consumption
of fruits and vegetables among US high school students:
differences by race,
ethnicity, and gender. J Sch Health. 2002;72(10):413-421.

43. Roberts C, Freeman J, Samdal O, et al; International HBSC
Study Group. The
Health Behaviour in School-aged Children (HBSC) study:
methodological devel-
opments and current tensions. Int J Public Health.
2009;54(suppl 2):140-150.
44. Currie C, Molcho M, Boyce W, Holstein B, Torsheim T,
Richter M. Researching
health inequalities in adolescents: the development of the
Health Behaviour in
School-aged Children (HBSC) Family Affluence Scale. Soc Sci
Med. 2008;66
(6):1429-1436.
45. Ford ES, Li C, Zhao G, Pearson WS, Tsai J, Churilla JR.
Sedentary behavior, physi-
cal activity, and concentrations of insulin among US adults.
Metabolism. 2010;
59(9):1268-1275.
471
Downloaded From:
23403/ on 01/11/2017
46. Hardy LL, Denney-Wilson E, Thrift AP, Okely AD, Baur
LA. Screen time and meta-

bolic risk factors among adolescents. Arch Pediatr Adolesc
Med. 2010;164
(7):643-649.
47. Anderson SE, Economos CD, Must A. Active play and
screen time in US children
aged 4 to 11 years in relation to sociodemographic and weight
status charac-
teristics: a nationally representative cross-sectional analysis.
BMC Public Health.
2008;8:366. doi:10.1186/1471-2458-8-366.
48. Nader PR, Bradley RH, Houts RM, McRitchie SL, O’Brien
M. Moderate-to-
vigorous physical activity from ages 9 to 15 years. JAMA.
2008;300(3):295-
305.
49. Nelson MC, Neumark-Stzainer D, Hannan PJ, Sirard JR,
Story M. Longitudinal
and secular trends in physical activity and sedentary behavior
during adolescence.
Pediatrics. 2006;118(6):e1627-e1634. doi:10.1542/peds.2006-
0926.
50. Deshmukh-Taskar PR, Nicklas TA, O’Neil CE, Keast DR,
Radcliffe JD, Cho S.
The relationship of breakfast skipping and type of breakfast
consumption with nu-
trient intake and weight status in children and adolescents: the
National Health and
Nutrition Examination Survey 1999-2006. J Am Diet Assoc.
2010;110(6):869-
878.
51. Babey SH, Wolstein J, Diamant AL. Food environments near

home and school
related to consumption of soda and fast food. Policy Brief
UCLA Cent Health Policy
Res. 2011(PB2011-6):1-8.
52. Gordon-Larsen P, Guilkey DK, Popkin BM. An economic
analysis of community-
level fast food prices and individual-level fast food intake: a
longitudinal study.
Health Place. 2011;17(6):1235-1241.
53. Richardson AS, Boone-Heinonen J, Popkin BM, Gordon-
Larsen P. Neighbor-
hood fast food restaurants and fast food consumption: a national
study. BMC
Public Health. 2011;11:543. doi:10.1186/1471-2458-11-543..
54. Ortega FB, Chillón P, Ruiz JR, et al. Sleep patterns in
Spanish adolescents: as-
sociations with TV watching and leisure-time physical activity.
Eur J Appl Physiol.
2010;110(3):563-573.
55. Furst T, Connors M, Bisogni CA, Sobal J, Falk LW. Food
choice: a conceptual
model of the process. Appetite. 1996;26(3):247-265.
56. Dubois L, Farmer A, Girard M, Peterson K. Social factors
and television use dur-
ing meals and snacks is associated with higher BMI among pre-
school children.
Public Health Nutr. 2008;11(12):1267-1279.
57. Matheson DM, Killen JD, Wang Y, Varady A, Robinson TN.
Children’s food con-
sumption during television viewing. Am J Clin Nutr.

2004;79(6):1088-1094.
58. Halford JC, Gillespie J, Brown V, Pontin EE, Dovey TM.
Effect of television ad-
vertisements for foods on food consumption in children.
Appetite. 2004;42
(2):221-225.
59. Halford JC, Boyland EJ, Hughes G, Oliveira LP, Dovey TM.
Beyond-brand effect
of television (TV) food advertisements/commercials on caloric
intake and food
choice of 5–7-year-old children. Appetite. 2007;49(1):263-267.
60. Powell LM, Szczypka G, Chaloupka FJ. Trends in exposure
to television food ad-
vertisements among children and adolescents in the United
States. Arch Pediatr
Adolesc Med. 2010;164(9):794-802.
61. Powell LM, Schermbeck RM, Szczypka G, Chaloupka FJ,
Braunschweig CL.
Trends in the nutritional content of television food
advertisements seen by chil-
dren in the United States: analyses by age, food categories, and
companies. Arch
Pediatr Adolesc Med. 2011;165(12):1078-1086.
62. Harrison K, Marske AL. Nutritional content of foods
advertised during the televi-
sion programs children watch most. Am J Public Health.
2005;95(9):1568-1574.
63. Palmer EL, Carpenter CF. Food and beverage marketing to
children and youth: trends
and issues. Media Psychol. 2006;8:165-190.

doi:10.1207/s1532785xmep0802_6.
64. Story M, French S. Food advertising and marketing directed
at children and ado-
lescents in the US. Int J Behav Nutr Phys Act. 2004;1(1):3.
doi:10.1186
/1479-5868-1-3.
65. Robinson TN. Reducing children’s television viewing to
prevent obesity: a ran-
domized controlled trial. JAMA. 1999;282(16):1561-1567.
66. Robinson TN, Killen JD, Kraemer HC, et al. Dance and
reducing television view-
ing to prevent weight gain in African-American girls: the
Stanford GEMS pilot
study. Ethn Dis. 2003;13(1)(suppl 1):S65-S77.
67. Epstein LH, Roemmich JN, Robinson JL, et al. A
randomized trial of the effects
of reducing television viewing and computer use on body mass
index in young
children. Arch Pediatr Adolesc Med. 2008;162(3):239-245.
68. Gortmaker SL, Cheung LW, Peterson KE, et al. Impact of a
school-based inter-
disciplinary intervention on diet and physical activity among
urban primary school
children: eat well and keep moving. Arch Pediatr Adolesc Med.
1999;153(9):
975-983.
472

Downloaded From:
23403/ on 01/11/2017
Priming Effects of Television Food Advertising on Eating
Behavior
Jennifer L. Harris, John A. Bargh, and Kelly D. Brownell
Yale University
Abstract
Objective—Health advocates have focused on the prevalence of
advertising for calorie-dense low-
nutrient foods as a significant contributor to the obesity
epidemic. This research tests the hypothesis
that exposure to food advertising during television viewing may
also contribute to obesity by
triggering automatic snacking of available food.
Design—In Experiments 1a and 1b, elementary-school-aged
children watched a cartoon that
contained either food advertising or advertising for other
products and received a snack while
watching. In Experiment 2, adults watched a television program
that included food advertising that
promoted snacking and/or fun product benefits, food advertising
that promoted nutrition benefits or
no food advertising. The adults then tasted and evaluated a
range of healthy to unhealthy snack foods
in an apparently separate experiment.

Main Outcome Measures—Amount of snack foods consumed
during and after advertising
exposure.
Results—Children consumed 45% more when exposed to food
advertising. Adults consumed more
of both healthy and unhealthy snack foods following exposure
to snack food advertising compared
to the other conditions. In both experiments, food advertising
increased consumption of products not
in the presented advertisements, and these effects were not
related to reported hunger or other
conscious influences.
Conclusion—These experiments demonstrate the power of food
advertising to prime automatic
eating behaviors and thus influence far more than brand
preference alone.
Keywords
Food advertising; Priming; Eating behavior; Children; Obesity
According to the U.S. Surgeon General, “Obesity is the fastest
growing cause of disease and
death in America” (Carmona, 2003). And the crisis is not unique
to the U.S.; according to the
World Health Organization (2003), the obesity epidemic is “a
major contributor to the global
burden of chronic disease and disability”. The trend is
especially disturbing among young
people. Over the past 30 years, the percentage of children and
adolescents in the U.S. who are
overweight or at risk of becoming overweight has more than
tripled to 37% and 34%,
respectively (Ogden, et al., 2006).

Address for correspondence: Jennifer L. Harris, Department of
Psychology, Yale University, P.O. Box 208205, New Haven, CT
06520,
[email protected]
Publisher's Disclaimer: The following manuscript is the final
accepted manuscript. It has not been subjected to the final
copyediting,
fact-checking, and proofreading required for formal publication.
It is not the definitive, publisher-authenticated version. The
American
Psychological Association and its Council of Editors disclaim
any responsibility or liabilities for errors or omissions of this
manuscript
version, any version derived from this manuscript by NIH, or
other third parties. The published version is available at
http://www.apa.org/journals/hea/
NIH Public Access
Author Manuscript
Health Psychol. Author manuscript; available in PMC 2010 July
1.
Published in final edited form as:
Health Psychol. 2009 July ; 28(4): 404–413.
doi:10.1037/a0014399.
N
IH
-P
A
A
uthor M
anuscript

N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
http://www.apa.org/journals/hea/
This obesity crisis has been fueled by reductions in physical
activity, as well as
overconsumption of foods high in fat and sugar (Institute of
Medicine (IOM), 2006). Health
authorities believe that the accumulation of unhealthy messages
communicated to children
through food advertising is a leading cause of unhealthy
consumption (Brownell & Horgen,
2004; IOM, 2006). Every day, children view, on average, 15
television food advertisements
(Federal Trade Commission, 2007), and an overwhelming 98%

of these ads promote products
high in fat, sugar, and/or sodium (Powell, Szczpka, Chaloupka,
& Braunschweig, 2007).
Moreover, food advertising to children portrays unhealthy
eating behaviors with positive
outcomes. Snacking at non-meal times occurred in 58% of food
ads during children’s
programming (Harrison & Marske, 2005). In addition to good
taste, the most common product
benefits communicated include fun, happiness and being “cool”
(Folta, Goldberg, Economos,
Bell, & Meltzer, 2006; Harrison & Marske, 2005).
A number of reviews have examined the research on advertising
to children and conclude that
food advertising leads to greater preferences and purchase of
the products advertised (Hastings
et al., 2003; IOM, 2006; Story & French, 2004). In addition, as
assessed through correlational
and quasi-experimental studies, heavier media viewing often
predicts more unhealthy diets and
higher body weight among children (see IOM, 2006). A few
studies have also examined effects
of food advertising on actual eating behaviors, usually assessed
by food choices following
exposure to advertising (see Hastings et al., 2003; IOM, 2006).
One study with high ecological
validity exposed children at an overnight camp to a daily
cartoon with candy or fruit advertising,
PSAs, or no ads (Gorn & Goldberg, 1982). Over a 2-week
period, children who saw the candy
ads selected fruit and orange juice as a snack less often than the
other children.
The literature reviews also highlight, however, the need for
further research -- specifically,

more studies that establish a direct causal link between food
advertising and unhealthy diets.
To begin to address this need, Halford and colleagues recently
demonstrated that groups of
children eat more immediately after viewing a series of 8–10
children’s food commercials than
after watching commercials for other products (Halford,
Boyland, Hughes, Oliveira, & Dovey,
2007; Halford et al., 2008; Halford, Gillespie, Brown, Pontin, &
Dovey, 2004). Additionally,
these effects occurred at the category level, (i.e., increased
consumption transferred to foods
not included in the presented advertisements). However, the
authors did not obtain support for
their proposed mechanism: specifically, that overweight
children have greater recognition
memory for food advertisements, which in turn leads to greater
consumption.
The literature reviews also emphasize the need to extend food
advertising research beyond
children; to-date, very little is known about such effects on
adolescents and adults. Finally,
most research has examined advertising for calorie-dense, low-
nutrient foods. As a result, we
know very little about how advertising for more nutritious food
affects eating behaviors. The
present research addresses these gaps in our knowledge and
utilizes a new approach to study
food advertising effects using contemporary social-cognitive
theories.
Advertising as a “real-world” prime
Social-cognitive theories suggest a subtle and potentially far-
reaching effect of food advertising
on eating behaviors that may occur outside of participants’

intention or awareness (i.e.,
unconsciously; see Bargh & Morsella, 2008). Priming methods
provide a means to test for
these automatic causal effects. In priming studies, relevant
mental representations are activated
in a subtle, unobtrusive manner in one phase of an experiment,
and then, the unconscious,
unintended effects of this activation are assessed in a
subsequent phase (see Bargh & Chartrand,
2000). Priming research has already demonstrated that a variety
of complex social and physical
behaviors – such as aggression, loyalty, rudeness, and walking
speed – can be activated by
relevant external stimuli (i.e., the primes) without the person’s
intent to behave that way or
Harris et al. Page 2
1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P

A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
awareness of the influence (see Dijksterhuis, Chartrand, &
Aarts, 2007). The mechanism
through which behavior priming operates appears to be an
overlap or strong association
between representations activated by the perception of a given
type of behavior, and those used
to enact that type of behavior oneself (Dijksterhuis & Bargh,
2001) – the same mechanism that
creates tendencies toward imitation and mimicry in adults
(Bargh, 2005; Chartrand & Bargh,
1999) and which serves as a vital support for vicarious learning
in young children (Tomasello,
Call, Behne, & Moll, 2005).
An important real-life source of priming influences is the
media, including television programs
and advertisements. Exposure to aggressive or alcohol-

consuming models in media can prime
aggressive behaviors and alcohol consumption in the viewer
(see Anderson & Bushman,
2002; Roehrich & Goldman, 1995). Studies that have focused
specifically on advertising
effects have shown that ads can prime positive expectancies of
the effects of alcohol
consumption (Dunn & Yniguez, 1999) and positive attitudes
towards smoking (Pechman &
Knight, 2002).
External cues and consumption behaviors
Research among adults confirms that external cues have a
significant influence on food
consumption behaviors. Exposure to the sensory properties of
palatable food increased
subjective desire and consumption, even though participants
were already fully sated (Cornell,
Rodin & Weingarten, 1989). Subsequent studies confirmed and
extended this finding, showing
that exposure to sensory-related food cues increases
consumption (Federoff, Polivy & Herman,
1997; Jansen & van den Hout, 1991; Rogers & Hill, 1989).
Moreover, food advertising
typically focuses on the immediate sensory gratifications of
consumption (i.e., the ‘hot’,
appetitive features), making resistance to these messages even
more difficult (i.e., the ‘cold’,
rational process of self-restraint; Loewenstein, 1996; Metcalfe
& Mischel, 1999). In light of
these findings, Lowe and Butryn (2007) proposed that palatable
food stimuli can trigger
hedonic hunger, or “thoughts, feelings and urges about food in
the absence of energy deficits”.
Consumption behaviors can also be activated through automatic

processes. External cues, not
related to the sensory qualities of food, (e.g., container size and
shape, food variety, and portion
size) affect amount consumed without the consumer’s
knowledge (Wansink, 2006). The
behavior of other people is another important external
behavioral cue, and people automatically
mimic others’ eating behaviors, including food choice and
amount of food consumed, without
realizing they are doing so (Johnston, 2002; Tanner, Ferraro,
Chartrand, Bettman & van Baaren,
in press). The unconscious nature of these influences is further
established by studies in which
primes of thirst-related words or smiling faces, presented
subliminally, outside of the
participant’s conscious awareness, increased beverage
consumption among thirsty individuals
(Strahan, Spencer & Zanna, 2002; Winkielman, Berridge, &
Wilbarger, 2005).
Food advertising
Advertising for food and beverages communicates potentially
powerful food consumption
cues, including images of attractive models eating, snacking at
non-meal times, and positive
emotions linked to food consumption (Folta et al., 2006;
Harrison & Marske, 2005). We
propose that the messages presented in television food
advertising similarly have the power to
act as real-world primes and lead to corresponding eating
behaviors. Given the types of foods
and consumption benefits typically promoted in food
advertising, what is primed is usually
snacking on unhealthy foods and beverages (Harrison &
Marske, 2005; Powell, Szczpka,
Chaloupka & Braunschweig, 2007).

In the following studies, we experimentally test whether
television food advertising, embedded
as it would naturally occur within a television program, will
prime, or directly activate, an
1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P

A
A
uthor M
anuscript
automatic increase in snack food consumption. Because these
effects are hypothesized to occur
outside of conscious awareness, the intention or ability to
regulate impulsive tendencies should
not affect the outcome. Therefore, we predict that food
advertising that conveys snacking and
fun (i.e., those typically shown during children’s programming)
will automatically cue eating
behavior among adults as well as children. In addition, in line
with the Halford et al. (2004,
2007, 2008) findings, we predict that the advertising will affect
consumption of any available
foods, not only those that were advertised.
We designed the studies to replicate conditions in which
individuals are typically exposed to
food advertising on television, as well as to minimize
participant awareness that the
experiments involved advertising (versus television viewing, in
general). All advertisements
were embedded within a television program during naturally-
occurring commercial breaks,
and the total number of food advertisements was consistent with
the number typically presented
during a similar amount of programming time. Experiments 1a
and 1b utilized common types
of children’s food advertisements as stimuli and measured

effects on snack food consumed by
children while watching television. Experiment 2 investigated
the effects of both snack- and
nutrition-focused food advertising on adult consumption of a
range of healthy to unhealthy
snack foods. To further minimize awareness of the true purpose
of the experiments, the
advertisements were not related to the brands or types of foods
to be consumed by participants.
Experiments 1a and 1b
In Experiment 1a, we tested our primary hypothesis that
elementary-school-aged children
would consume significantly more snack food while watching a
cartoon that included food
advertising. In Experiment 1b, we recruited children from a
more ethnically and
socioeconomically diverse school district and added a
participant incentive ($20 gift card).
Except where noted, recruiting and experimental procedures
were identical in Experiments 1a
and 1b.
Method
In both experiments, children were randomly assigned to watch
a cartoon that included either
food advertising or other types of advertising and were given a
snack while watching. Children
watched alone to eliminate potential imitation, social
facilitation or self-presentation effects.
Parents also completed a short questionnaire with information
about their child.
Participants—In total, 118 children participated: 55 in
Experiment 1a and 63 in Experiment
1b; 56 girls and 62 boys; and 59 children each in the food and

non-food advertising conditions.
The two conditions did not differ significantly on any of the
child characteristics measured,
including age, weight status and ethnicity (all ps ≥.16). We
received complete data for 108
participants; 92% of parents returned the questionnaire.
Children’s ages ranged from 7 to 11
years (M = 8.8 years).
To determine children’s weight status, we utilized height and
weight information provided by
parents and compared children’s body mass index (BMI) to age-
and sex-normed percentiles
published by the Centers for Disease Control and Prevention
(CDC, 2007). As recommended
by the CDC, children with BMI’s below the 5th percentile were
classified as “underweight”,
those in the 85th to <95th percentiles were classified as “at risk
of overweight”, and those in
the 95th or higher percentiles were classified as “overweight”.
Under these criteria, 3% of our
participants were underweight (n = 3), 62% were normal weight
(n = 66), 21% were at risk of
overweight (n = 23), and 14% were overweight (n = 15). There
was no significant difference
in children’s weight status between Experiments 1a and 1b, χ2
(3, N = 107) = 4.52, p =.21; and
the combined rate of at-risk and overweight children (35%) was
comparable to the 37%
incidence for children in the U.S. (Ogden, et al., 2006).
1.

N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
We also obtained children’s combined race/ethnicity and prior-
week television viewing from

parents. Participants in Experiment 1a were primarily white,
non-Hispanic (95%), whereas our
sample in Experiment 1b was ethnically diverse: 61% were
white, non-Hispanic (n = 39), 20%
black, non-Hispanic (n = 13), 10% Hispanic (n = 6), 6% Asian
(n = 4) and 2% other or mixed
ethnicity (n = 1). According to their parents’ report, children in
Experiment 1a watched very
little television (M = 1.1 hours per day). Parents in Experiment
1b reported significantly higher
child television viewing (M = 2.0 hours-per-day), t(107) = 4.77,
p <.01; and that their children
were more likely to have a television in their bedrooms (48%
vs. 4% for Experiment 1a
participants), χ2 (1, N = 107) = 25.95, p <.001. In Experiment
1b, we also collected child reports
of their own television viewing: children indicated that they
watched significantly more
television (M = 3.2 hours-per-day) than their parents reported
that they watched, t(56) = 4.35,
p <.001. This level of child-reported television viewing is
comparable to the 3.2 hours-per-day
reported by 8- to 10-year-olds in a large U.S. study that utilized
a similar methodology (Roberts
& Foehr, 2004).
Procedure and Materials—Parents with children in participating
schools received a letter
inviting them to volunteer with their children for a study to
understand television influences.
In Experiment 1b, we also recruited 6 children from a summer
camp in the same school district.
Parents received a description of the experimental procedure.
Parents who requested more
information were informed that we were measuring how food
advertising affects eating

behaviors, but asked not to share that information with their
children before the study. All
parents provided written informed consent, and all procedures
and materials were approved
by the university’s Human Subjects Committee. Participants in
Experiment 1a did not receive
compensation, and Experiment 1b participants received a $20
bookstore gift card.
The children met with the experimenter individually at their
school or camp for approximately
30 min. in an unoccupied classroom or conference room. For
school participants, sessions were
held after school. If the child asked about the purpose of the
study, the experimenter informed
her or him that we were interested in finding out about the kinds
of things that children like,
including television shows and foods.
Following a get-acquainted activity, the children watched a 14-
minute episode of “Disney’s
Recess”, a cartoon typically viewed by 7- to 11-year-olds. In
this episode, the class goes on a
field trip to a science museum. One-half of the children were
randomly assigned to watch a
version that included 4 30-sec. food commercials inserted
during 2 designated advertising
breaks. These commercials promoted snack and breakfast foods
of poor nutritional quality
using a fun and happiness message (a high-sugar cereal, waffle
sticks with syrup, fruit roll-
ups, and potato chips), and were chosen to represent the types
of food commercials that are
most commonly shown on children’s television (Powell et al.,
2007). The other half watched
the same cartoon with 4 non-food commercials (games and

entertainment products). All
commercials had aired during actual children’s television
cartoon programming.
Children also received a large bowl of cheddar cheese
“goldfish” crackers (150 gr.) and a glass
of water, and were told that they could have a snack while
watching. (Advertising for goldfish
crackers was not presented during the cartoon.) The
experimenter then left the room, returned
after the cartoon was finished, and asked the children when they
had last eaten prior to the
experiment. Participants in Experiment 1b also highlighted the
programs they had watched on
the previous weekday and Saturday on a television programming
grid. After the children left,
the experimenter weighed the remaining goldfish and recorded
the amount consumed.
Separately, parents completed a short questionnaire that asked
for the number of hours and
minutes their child had watched television on each of the past 7
days, whether the child has a
television in his or her bedroom, how often the child ate a snack
or meal while watching
1.
N
IH
-P
A

A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
television in the past 7 days, how much their child likes
goldfish crackers, and their child’s
height, weight, and demographic information.
One debriefing was held for all children following completion
of the sessions at their school
or camp to minimize the possibility that children would share

information about the purpose
of the study with future participants. Interested parents also
attended, and all parents received
a debriefing in the mail.
Results
Identical procedures were followed during the cartoon-viewing
portions of Experiments 1a
and 1b, and the amount of goldfish crackers consumed did not
differ between the two studies
(p =.68) (see Table 1). Therefore, to increase the power of the
statistical analyses, we combined
results for the two experiments in the following analysis of
eating behaviors.
As predicted, children who saw the cartoon with food
advertising ate considerably more (45%)
goldfish crackers while watching (M = 28.5 gr.) than did
children who saw non-food advertising
(M = 19.7 gr.), t(116) = 3.19, p =.01, d =.60.
Importantly, most child characteristics did not predict or
moderate consumption (see Table 1).
ANOVAs were conducted with advertising condition and child
categories, including weight
status, gender, television in the child’s bedroom, and white,
non-Hispanic versus ethnic
minority, as between-participants factors. All models showed a
main effect of advertising
condition (all F(1,105) ≥ 7.03, p <.01). In addition, there were
no significant main effects for
any of the child characteristics (all Fs ≤.75, ps ≥.39) and no
significant interactions with
advertising (all Fs ≤ 1.13, ps ≥.29).
Additionally, we found similar results when we conducted

separate regression analyses to
predict snack consumption using a standardized version of each
continuous variable, a dummy
variable for condition, and the interaction term. The amount of
goldfish crackers consumed
was not significantly correlated with amount of time since the
child last ate, child’s age, parents’
assessment of their children’s appetite, snacking while watching
TV in the past week, parents’
reports of their child’s weekly TV viewing, or children’s
reported TV viewing (collected in
Experiment 1b only), (all ps ≥.29) or with any of the interaction
terms (all ps ≥.42). Only
parents’ assessment of how much their children liked goldfish
crackers, β =.20, t(3,104) = 2.13,
p =.04, predicted amount consumed. Therefore, regardless of
the child characteristics
examined, children consumed more after viewing the food
advertising,
Discussion
These results provide strong support for our hypothesis.
Children who saw food advertising
ate 8.8 grams more during the 14 min. they watched TV in this
experiment. At this rate,
snacking while watching commercial television with food
advertisements for only 30 min. per
day would lead to 94 additional kcal. consumed and a weight
gain of almost 10 pounds per
year, if not compensated by reduced consumption of other foods
or increased physical activity.
Unexpectedly, of the child characteristics measured, only liking
of goldfish crackers (as
reported by parents) predicted amount consumed. We caution
against making definitive

conclusions about differences in eating behaviors between
different groups of children, as some
parent and child reports, including child’s weight and television
viewing may be biased.
However, the lack of significant moderating effects for any of
the child characteristics
measured suggests the considerable power of food advertising to
consistently influence
consumption across a highly diverse sample of children. In
general, then, the effect of food
advertising was consistent with an automatic link between
perception and behavior, and in line
1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A

uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
with most other recent demonstrations of behavioral priming
effects (Dijksterhuis & Bargh,
2001; Dijksterhuis et al., 2007).
Experiment 2
In Experiment 2, we expand on the above findings to predict
that food advertising will also
prime eating behavior among an adult sample. In addition, we
examine whether effects on
eating behavior are simply due to exposure to images and
thoughts of palatable foods or whether
the product benefits presented in the advertising differentially
affect consumption. Specifically,
we hypothesize that exposure to food advertising with that
promotes snacking, fun and
excitement will prime greater consumption of snack foods than
advertising that conveys
nutrition benefits. Although we did not specifically test the
effects of advertising for different
types of foods, these messages are commonly used to promote

calorie-dense, low-nutrient food
products in both adult and children’s food advertising (Harrison
& Marske, 2005), whereas the
nutrition message tends to be used in advertising for somewhat
healthier products. Finally, we
examine individual differences in food advertising effects. Prior
research has demonstrated
that women who habitually diet and monitor their weight (i.e.,
restrained eaters) may be
especially prone to increased eating when exposed to external
food cues (Federoff, Polivy, &
Herman, 1997; Jansen & van den Hout, 1991). As a result, we
hypothesize a general effect of
snack advertising on increased eating, but a more pronounced
effect on restrained eaters.
Method
As in the first experiments, we attempted to replicate viewing
conditions in which participants
would be naturally exposed to food advertising. In Experiment
2, however, participants were
not provided with a snack while watching. Instead, they were
asked to participate in an
ostensible ‘second experiment’ to test consumer products. In
this second study, they tasted and
rated snack foods that varied in perceived nutritional value.
Participants—Participants were 98 university students between
18 and 24 years old.
Restrained eaters (i.e., those with scores ≥ 15 on the Eating
Restraint Scale; Herman, Polivy,
Pliner, Threlkeld & Munic, 1978) included 31 women and 8
men; unrestrained eaters included
29 women and 24 men. Participants were racially and ethnically
diverse: 61% were of white,
European-American descent only (n = 55), 7% were black only

(n = 7), 14% Asian only (n =
13), 7% Hispanic only (n = 6), and 9% mixed race or ethnicity
(n = 9). Participants received
Introduction to Psychology course credit or $10.
Materials—A 16-minute, abbreviated version of an
improvisational comedy television
program (“Whose Line is it Anyway?”) was used as the
television-viewing stimuli. The
program included 11 commercials (4 min. total), inserted during
2 commercial breaks. Three
versions were created; each version included 7 of the same non-
food commercials. In addition,
one version included 4 commercials for food and beverages with
a snacking message that
emphasized fun and excitement (2 fast-food products, candy
bar, and cola soft drink); another
included 4 food and beverage commercials with a nutrition
message (granola bar, orange juice,
oatmeal and an “instant breakfast” beverage); and the control
included 4 additional non-food
commercials. These commercials were inserted into non-
prominent positions during the
commercial break (i.e., not the first or last commercial) to
reduce the likelihood that participants
would pay more than their usual amount of attention to the food
commercials.
Pre-testing with a sample of college students confirmed that the
food advertisements
communicated the intended product benefits (see Table 2). The
commercials were also matched
on other persuasion-related characteristics. Pre-test participants
reported similar moderate
levels of enjoyment for all commercials (M = 5.59 out of 10 for
the snack ads, 5.53 for the

nutrition ads, and 5.05 for the control ads), F(2, 158) = 1.20, ns.
In addition, past consumption
1.
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A

uthor M
anuscript
of the foods in the snack and nutrition ads did not differ
significantly (M = 1.78 out of 6 for
the snack ads and 2.11 for the nutrition ads), t(102) = 1.37, ns;
nor did future intent to purchase
the foods (M = 4.78 out of 10 for the snack ads; M = 5.20 for
the nutrition ads), t(102) = 1.37,
ns. The only significant difference found was that participants
were less familiar with the
nutrition commercials (M = 1.13 out of 6) than the snack (M =
1.47) or control (M = 1.68)
commercials, F(2, 158) = 6.91, p <.01. Familiarity was low,
however, for all commercials
tested.
Procedures—All participants were tested between 3 and 6 p.m.
to minimize initial
differences in hunger. On average, participants had last eaten
2.8 hours earlier (SD = 2.5). They
were informed that the first study examined effects of television
on mood, and were randomly
assigned to watch one of the three versions of the television
program. To increase the
believability of the cover story, participants were informed that
they were in the “comedy
condition”, and that the experimenter had kept the commercials
to make the viewing experience
as realistic as possible. Before and after watching television,
participants completed a PANAS
current mood assessment (Watson, Clark, & Tellegen, 1988). To
assess hunger without alerting

participants that the study involved food, hunger and thirst
ratings were embedded within the
PANAS assessment. As with the mood measures, participants
responded on a scale from 1
(very slightly/not at all) to 5 (extremely) in response to “How
hungry/thirsty do you feel right
now, at this present moment?” All participants watched in a
small, comfortable room, by
themselves.
In line with the cover story, participants were then asked to
move to another room, with a
different experimenter. They were seated at a table with 5 pre-
measured snack foods including
very healthy (carrots and celery with dip), calorie-dense,
nutrient-poor items (mini chocolate
chip cookies and cheesy snack mix), and items perceived to be
moderately healthy (trail mix
and multi-grain tortilla chips). They also received a bottle of
water. Until this point, participants
were not aware that the study involved food. As in the prior
experiments, none of the snack
foods tested had been advertised during the television segment.
Participants were instructed to
take at least one bite of each and rate it on a variety of
dimensions, but also told they could eat
as much as they liked. The experimenter then left the room.
After the participants finished the tasting, they informed the
experimenter, who removed the
food items and asked them to complete questionnaires to assess
perceived healthiness of the
foods tasted, restrained eating, and demographics. These items
were assessed at the end of the
session to avoid affecting eating behaviors with reminders of
health or dieting (other than those

presented in the advertisements). The weight of each food
consumed was recorded, as well as
the total amount of time spent eating. Finally, the first
experimenter conducted a funnel
debriefing (Bargh & Chartrand, 2000) to probe for awareness of
the experimental hypotheses
and effect of the advertisements on subsequent eating behavior.
Unaided recall of specific
advertisements was also obtained during the debriefing.
Results and Discussion
During the funnel debriefing, most participants indicated that
they had noticed the advertising,
but believed our cover story that the study involved television
and mood. To ensure that the
following analyses demonstrate effects of food advertising that
occurred outside of
participants’ awareness, however, we eliminated the data for the
few participants (4 each in
the snack and nutrition advertising conditions) who correctly
guessed that the study concerned
effects of food commercials on eating behaviors or who
believed that the food commercials
might have influenced what or how much they ate.
As intended, participants rated the cookies and snack mix as
very unhealthy (M = 2.71 out of
10 and 2.31, respectively), the vegetables as very healthy (M =
7.71), and the trail mix (M =
1.
N

IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
4.92) and multi-grain chips (M = 4.92) in between. In addition,
participants reported fairly high
taste ratings for all the foods, with the lowest ratings for the

multi-grain chips (M = 6.46 out
of 10), and the highest ratings for the vegetables (M = 7.64) and
cookies (M = 7.70).
Advertising effects on consumption—Participants ate the most
vegetables (M = 34.3
gr.), (vegetables also weighed the most), followed by cookies
(M = 17.9 gr.), trail mix (M =
12.3 gr.), snack mix (M = 9.4 gr.) and multi-grain chips (M =
7.2 gr.). To adjust for weight
differences in the foods, we computed z-scores for amount of
each food consumed and averaged
the standardized scores to obtain a single food-consumption
score for each participant.
According to this measure, a positive score indicates a total
consumed of “X” standard
deviations above the sample mean, and a negative score
indicates a lower-than-average amount
consumed.
To control for potential individual differences in our dependent
variables, we conducted all
analyses using ANOVAs with advertising condition, gender and
restrained eating as between-
participants factors. As predicted, the main effect of advertising
condition was significant, such
that participants who saw snack ads ate more (M =.51) than did
control participants (M =.07)
or those who saw nutrition ads (M = −.13), F(2,78) = 3.72, p
=.03, η2 =.09. An ANOVA to
predict eating time also showed a main effect of advertising,
F(2,78) = 5.05, p <.01, η2 =.12.
Again, participants who saw snack ads ate for the longest
amount of time (M = 13.1 min.)
compared to the other participants (M = 9.8 min. for the control
and M = 8.7 min. for nutrition

ads).
Planned comparisons of the two types of food ads to each other
and the control confirmed that
participants who viewed the snack ads consumed significantly
more than those who viewed
the nutrition ads, F(1,49) = 8.57, p <.01, η2 =.15, and the
difference in consumption between
snack ads and the control approached conventional significance,
F(1,51) = 3.24, p =.08, η2 =.
06. The difference between nutrition ads and the control was not
significant (p =.30).
As predicted, there was a trend for restrained eaters to eat more
overall than unrestrained eaters
(M =.31 vs. −.01), F(1,78) = 3.34, p =.07, η2 =.04. Men also ate
considerably more than women
(M =.50 vs. −.20), F(1,78) = 15.05, p <.001, η2 =.16. The
Advertising x Restrained Eating
interaction approached significance, F(2,78) = 2.75, p =.07, η2
=.07, and the Advertising x
Gender interaction was reliable, F(2,78) = 3.25, p =.04, η2 =.08
(see Figure 1). The snack
advertising had powerful effects on men and restrained eaters;
with both groups consuming
approximately 1 SD more after exposure to snack ads versus
nutrition ads or no food ads.
Female unrestrained eaters, however, ate similar amounts across
all conditions.
Potential mediators and moderators of the effects—We then
examined whether the
effects of advertising on consumption behavior were mediated
by hunger or mood. ANOVAs
to predict change in hunger and mood (before and after viewing)
showed no main effects of

advertising (ps ≥.58), or interaction effects on change in mood
(ps ≥.50). The 2-way interactions
between advertising and both gender and restrained eating on
change in hunger were significant
(F(2,78) = 3.68, p =.03, η2 =.09; F(2,78) = 2.86, p =.06, η2
=.06), but these effects were opposite
those found for consumption behaviors. Restrained eaters and
men reported feeling less hungry
after viewing snack advertising (M = −.41 and −.44) and more
hungry after viewing nutrition
advertising (M =.44 and .54), indicating a complete dissociation
between reported hunger and
eating behaviors.
We also examined potential predictors and moderators of total
consumption, including hunger
and mood at the time participants arrived at the experiment
(time 1) and after they had watched
the television program (time 2), as well as the number of
commercials recalled (awareness).
Again, ANCOVAs to predict total consumption using hunger,
mood and awareness variables
1.
N
IH
-P
A
A
uthor M

anuscript
N
IH
-P
A
A
uthor M
anuscript
N
IH
-P
A
A
uthor M
anuscript
as covariates showed no significant relationship to amount
consumed (all ps ≥.20). Only one
interaction between these potential moderator variables and
advertising condition approached
significance: advertising and hunger at time 2, F(2,78) = 2.61, p
=.08, η2 =.06, (all other ps ≥.
16). Further analyses revealed that hunger immediately prior to
eating, was related to amount
consumed only for participants who had viewed nutrition
advertising (r =.57, p <.01). Hunger

was not, however, significantly related to amount consumed for
participants in the snack ads
and control conditions (rs <.10, ps ≥.59). These findings further
support the direct influence
of the snack advertising on consumption, as effects were
unmediated by subjective internal
states such as hunger.
Finally, we examined the relationship between taste and
healthiness ratings and actual
consumption for individual foods. Taste ratings were positively
correlated with amount
consumed for all foods (ranging from r =.23, p <.05 for
vegetables to r =.45, p <.01 for snack
mix), but perceived healthiness was related only to the amount
of vegetables consumed, r =.
21, p <.05 (all other rs ≤ ±.10, ps ≥.34). ANCOVAs to predict
amount consumed of individual
foods, using rated taste of that food as a covariate,
demonstrated significant main effects of
advertising on cookie, F(2,76) = 4.01, p =.02, η2 =.10, and
multi-grain chip consumption, F
(2,76) = 11.46, p <.001, η2 =.23. In all cases, however, the
direction of influence was the same.
Participants who saw snack commercials ate the most of every
food, regardless of healthiness,
and those who saw nutrition commercials ate the least (see
Figure 2).
Discussion—Experiment 2 demonstrates that adults are also
susceptible to the automatic
effects of food advertising on consumption behavior.1 These
effects were extremely powerful
for men and restrained eaters. We also demonstrated that the
influence of the snack ads
continued after exposure (such that they carried over to the

DHHS Director Intro Memo

Recomendados

Recomendados

Más contenido relacionado

Más de harrisonhoward80223

Más de harrisonhoward80223 (20)

Último

Último (20)

DHHS Director Intro Memo