Your Federal Quarterly Tax Payments are due April 15th Get Help Now >>

thermometer by stariya

VIEWS: 16 PAGES: 23

									Michigan Food Thermometer Education Campaign

The Michigan State University National Food Safety and Toxicology Center
(MSUNFSTC), Michigan State University Food Science and Human Nutrition
Department (MSUFSHN) and Michigan State University Extension (MSUE), in
cooperation with the Food Safety Education Staff (FSES) of the Food Safety and
Inspection Service (FSIS), U.S. Department of Agriculture (USDA) and the
Michigan Department of Agriculture (MDA) seek to increase the use of food
thermometers in Michigan. A social marketing approach was adopted to achieve
this objective.

Need

Foodborne illness is a significant public health problem in America, with an
estimated 76 million cases of illness per year, including nearly 325,000
hospitalizations and about 5,000 deaths. The annual cost of medical care and
lost productivity that result from foodborne illness is measured in billions of
dollars. One estimate suggests a cost of $20 billion to $40 billion per year in lost
productivity.

Children, in particular, experience high rates of foodborne illnesses. Rates of
reported Salmonella infection, for example, are more than 10 times higher in
children less than 1 year old than rates among adults. For E. coli 0157:H7 and
Shigella, rates are highest among children aged 1 to 9 years.

Recognizing that foodborne illness is a serious health issue, an objective was
included in Healthy People 2010 to reduce foodborne illnesses by increasing the
percentage of consumers who follow key food safety practices, which include
using food thermometers.

Encouraging thermometer use is also important because new research has
raised questions about past advice to consumers to use color and texture of
meat as a guide to safe cooking. Researchers have learned that color is not a
reliable indicator because meat could turn brown without reaching the internal
temperature necessary to kill dangerous microorganisms that are present. Safety
could only be assured by using a thermometer.

Background

In 1998, a USDA study concluded that the internal temperature and therefore the
safety of cooked hamburgers couldn’t be judged by visual inspection of the color
of the cooked meat. Data showed that nearly 25 percent of the hamburgers
judged to be thoroughly cooked according to color could still be contaminated
with bacteria. As a result, the FSES launched a social marketing campaign to
promote the use of food thermometers in home cooking.

In 2001, the MDA funded a proposal from MSUNFSTC and MSUE for a
comprehensive food safety education program for Michigan consumers. The
proposal included an objective to “Develop and implement a social marketing
campaign emphasizing food safety.” Project staff committed to “develop a broad
social marketing campaign based on extensive input from and collaboration with
consumers and stakeholder groups.”

The shared objectives led to collaboration. The Michigan project staff identified
thermometer use as an appropriate goal for a food safety campaign. The USDA
determined that piloting the food thermometer use campaign in Michigan would
be an appropriate step in the national campaign strategy.

Project History

Initial focus groups conducted by FSIS showed that consumer food safety
knowledge related to thermometer use was lacking. Thermometer use was
generally reserved for special occasions, such as holiday meals. These focus
groups resulted in recommendations to:
      Promote food thermometer use for everyday meals
      Promote thermometer use as a means to improve taste as well as safety
      Target parents of young children who are responsible for the welfare of an
         at-risk group and also appear open to behavioral change.

FSES launched their national food thermometer education campaign in 2000 with
the character “Thermy” delivering the message, “It’s Safe to Bite When the
Temperature is Right.”

In 2001, FSES decided to adopt a social marketing approach in order to increase
the impact of their Food Thermometer Education Program.

Social Marketing

In social marketing efforts, the objective is to create behavior change in a specific
target audience. It is a long-term process. A successful social marketing program
is customer centered. It differentiates itself from the “competition.” Most important
of all, it makes a positive difference in the health and well being of the entire
society by facilitating behavioral change at the individual level.

In his book “Marketing for Social Change”, Andreasen describes a six-stage
process for strategic social marketing. The six steps are:
    1. Listening. Gain a deep understanding of the target audience.
    2. Planning. Translate what was learned in step one to make decisions about
        product, price, place and promotion.
    3. Structuring. Don’t go public without having put the structure in place to
        make the program successful.
    4. Pretesting. Test the plan with the target audience.
    5. Implementing. Work the plan.
    6. Monitoring. Plan routine monitor-and-revise cycles to adjust for what is
        learned during implementation.
Part of listening and planning is determining where the target audience is in the
stages of behavior change and then planning the appropriate actions to motivate
movement through the stages.

Four Stages of Behavior Change
Stage      Name of Stage         Description            Actions to Motivate
  1      Pre-contemplation      Ignorance        Provide awareness
  2      Contemplation          Consideration    Support information Gathering
  3      Action                 Trial            Facilitate “purchase” and
                                                 experimenting
   4     Maintenance            Adoption         Encourage to repeat and
                                                 evangelize

There are several other social marketing considerations:
    In many social marketing campaigns promoting socially desirable behavior,
      there is negative demand – that is, the customer is not asking for or
      demanding the “product” being “sold”.
    Social marketing programs are focused on a long-term commitment rather
      than the short-term profit that drives commercial marketing.
    Competition may take many forms and is undefined until the customer
      defines it. The customer’s competing alternatives must be understood and
      addressed to ensure program success.
    A central tenet of any social marketing effort is to go after the “low-hanging
      fruit – that is, to concentrate on the most responsive markets first. Since a
      campaign must show results in order to survive, it is advantageous to
      begin with the best candidate for demonstrable success.

Initiating the Social Marketing Approach

To build the foundation for gaining an understanding of the target audience a
complete review of past focus group research done for FTEC was undertaken.
These observations were made:
    Most participants are largely unaware of the need for food thermometer
       use; they are initially unconvinced the subject has sufficient priority to
       warrant discussion or action.
    Many participants are aware of the need for hand washing, for having
       clean food preparation surfaces and tools, and have some knowledge of
       cross-contamination.
    Many participants became interested in the subject matter when
       information was presented.
    Many participants became engaged in the discussion when actual
       thermometers were displayed; these devices converted the discussion
       from abstract to tangible.

In order to enhance the project success a more refined level of audience
segmentation was sought. A commercially-available market segmentation
system was selected. Market segmentation systems use statistical models that
combine numerous national surveys of consumer behavior, lifestyle and attitude
with annually updated census data. With this existing data the segmentation
system provides very acute segmentation and audience descriptions. This
approach is often referred to as geodemographic segmentation. The US
MOSIAC system was selected for the FTEC social marketing program. Because
of what had already been learned about the willingness of parents of young
children to adopt new food safety behaviors the US MOSIAC segments with high
indices of households with young children were selected. Four FTEC sub-
segments were selected – “Boomburbs”, “Single Moms”, “Heartlands” and “Rural
Towns”.

Boomburbs                           US Households       13.9%
Upscale, suburban, two-      Predominant Age Range      0-9, 35-44
earner families in newer                  Education     4 or more yrs of college
communities, where the           Median HH Income       $62,913
newest technological                  Housing Type      Single family owner
gadgets are frequently               Ethnic Diversity   White, Asian
sought                                  Family Type     Married with children
Single Moms                         US Households       13.78%
Predominantly African        Predominant Age Range      0-9, 18-24
American and Hispanic                     Education     Some high school
one-parent families in           Median HH Income       $30,626
major metropolitan areas,             Housing Type      Apartment renters
where incomes are low                Ethnic Diversity   African-American,
and the parents are                                     Hispanic
young                                   Family Type     Single parents
Heartlanders                        US Households       8.33%
Middle-class                 Predominant Age Range      5-9, <18, 35-54
Midwesterners with larger                 Education     High school graduate
families and traditional         Median HH Income       $41,458
lifestyles, stressing                 Housing Type      Single family owner
recreation and family                Ethnic Diversity   Whites
activities                              Family Type     Married with children
Rural Towns                         US Households       5.59%
Rural, low-income, less-     Predominant Age Range      0-9, <18, 18-24
educated and                              Education     Some High School
underemployed families,          Median HH Income       $29,089
with limited consumer                 Housing Type      Single family owners
choices                              Ethnic Diversity   White, Native American,
                                                        Hispanic
                                          Family Type   Married with children

Drawing on the input of experts including a nutrition educators and ethnic
marketers the “Boomburb” population with children under ten years of age was
selected as the campaign target. Reasons included their being most likely to
move rapidly through the Stages of Change, their concern for the welfare of their
children, their propensity for acquiring and using new information, their tendency
to be early adopters of technology, and the ability to capitalize on their role as
major influencers of mass culture.
More information on the campaign background can be found at:
http://www.fsis.usda.gov/Frame/FrameRedirect.asp?main=/oa/research/thermom
_edu.htm

Creating the Communication Strategy for the “Boomburbs”

Once the “Boomburbs” Parents were selected as the first target audience the
next step was to determine the best way of communicating with them. Based on
observation, FTEC campaign themes and focal points were developed. Potential
message and style concepts were developed and then tested with a carefully
selected Boomburbs Parents audience.

These tests showed preferences for family scenes, demonstration of
thermometer use techniques, clear, memorable temperature icons, a USDA
endorsement and a prominent web address. The following key areas for
message focus were identified.


Critical Areas for Enhanced Education    Suggest Media Tactics
Education of the correct cooking         Make this a primary campaign visual.
temperatures                             Repeat memorable icons on product
                                         packaging, web site, fact sheets,
                                         magnets, t-shirts, etc., and encourage
                                         press attention in this new approach.
Demonstration of thermometer use on      Create web site content videos and
small meat items, in point-of-purchase   photo, point-of-purchase
demonstrations and in web site videos    demonstrations and family cooking
                                         events. Encourage press attention,
                                         especially food and cooking oriented
                                         TV. Create substantial web site
                                         information and support with fact
                                         sheets.
Display of a variety of thermometer      Create point-of-purchase promotions,
devices and their uses                   extensive web site information and
                                         supporting fact sheets. Provide ideas
                                         and encouragement to wholesalers,
                                         manufacturers, retailers.
Inform clearly on the health risks to    Create substantial web site detail, fact
children related to undercooking of      sheets for parents, children and child
meats                                    health information sources like
                                         pediatricians.

For cost considerations, the Internet was selected as the primary content base.

More information on the communication styles of Boomburbs can be found at:
http://www.fsis.usda.gov/oa/research/rti_thermy.pdf and at:
http://www.fsis.usda.gov/oa/research/boomburbs_style_guide.pdf
Choosing the Slogan

The next step in listening to the target audience was developing the campaign
slogan.

Initial concept testing used a “special event” at selected Williams Sonoma stores,
offering gift certificates or stipends and providing refreshments. The goal was to
design the ideal prototype for promoting thermometer use. A total of 59
participants were interviewed concerning their concept preferences.

Findings included these recommendations:
    Utilize a “catchy” heading that appeals to parents
    Incorporate active images of an individual using a thermometer
    Incorporate the temperature chart in all designs
    Emphasize the notion that “one out of four hamburgers turns brown before
      it has reached a safe internal temperature.” This fact had the greatest
      impact.
    Use images “Boomburbs” would relate to.
    Use a website address that is simple and easy to remember.
    Decrease the amount of information and focus on the main message.

Based on the recommendations new concepts were tested early in 2004 to
finalize the slogan to be used in the Michigan Pilot. The slogan selected is:
                                   “Is it done yet?”
                             You can’t tell by looking.
                        Use a food thermometer to be sure.

More information on slogan selection can be found at:
http://www.fsis.usda.gov/PDF/Thermometer_Slogans_and_Concepts.pdf


Setting the Baseline

One advantage to using geodemographic segmenting is the ability to test a
sample population in Michigan and have the results be transferable to the nation.
Taking advantage of that fact, a baseline survey was administered prior to
implementation. This statewide survey identified current thermometer use of
people with children under ten in all four geodemographic groups.

The survey research for the project was conducted by the Travel, Tourism and
Recreation Resource Center at Michigan State University’s Department of
Community, Agriculture, Recreation and Resource Studies. A pre-campaign mail
survey was administered to 2,500 Boomburbs, 1,000 Heartlands, 1,000 Single
Moms, and 1,000 Rural Towns. Participation was encouraged with the offer of a
drawing for $300, $200 and $100 gift certificates. The return rate was 24%. A
phone survey with a target of 100 completed calls was used to measure non-
response bias.
The Michigan Pilot Implementation

After establishing the research baseline, a campaign was implemented focused
on the target audience – Boomburbs with children under 10 - in three counties
using paid and free media and public events promoting the importance of using
food thermometers.

The criteria for the “Boomburbs” was used to select the target counties. Following
is a listing of the twelve counties with the highest percentage of “Boomburbs”
compared to total addresses.

County Name                                           Addresses Targets      %
Oakland                                                  61,437 47,716 77.67%
Macomb                                                   44,146 31,944 72.36%
Livingston                                               10,996   7,615 69.25%
Washtenaw                                                13,053   8,864 67.91%
Kent                                                     28,164 18,691 66.36%
Wayne                                                    80,109 50,800 63.41%
Ottawa                                                   14,031   8,645 61.61%
Kalamazoo                                                10,857   6,413 59.07%
Genesee                                                  21,790 12,761 58.56%
Monroe                                                     8,769  4,971 56.69%
Ingham                                                   11,420   6,437 56.37%
Dickinson                                                  1,254    621 49.52%

Ingham, Washtenaw and Kent Counties were selected for the pilot in order to
provide separation of markets to allow for testing alternate strategies. In the post-
test Genesee County was used for the control.

The Planning Meeting

On April 21, 2004 people from the three selected counties and representatives of
organizations with an interest in food safety and community involvement met on
the MSU campus. The purpose of the meeting was to identify people, places and
things within the three target counties that would lend themselves to the
thermometer campaign intervention in August. They identified such things as:
     Which local media to target with paid advertising
     Where to schedule appearances by USDA Undersecretary for Food
       Safety Dr. Elsa Murano
     Events suitable for participation by the USDA Food Safety Mobile
     Where cooking demonstrations involving local celebrities and chefs might
       be held
     What specialty stores would be good hosts for events
     Grocers, meat markets, farm markets where displays might be set up
     Activities involving youth
The Implementation: Two Active Weeks in Michigan

Following the meeting, the implementation plans were developed.

Purchased media included television and radio ads, magazine ads, and
newspaper ads. Radio remotes were used to heighten awareness of the USDA
Food Safety Mobile locations and the campaign message.

The focus for free media was the visit to the state by Dr. Elsa Murano, USDA
Undersecretary for Food Safety and or Ann M. Veneman, U.S. Secretary of
Agriculture; the visits by the USDA Food Safety Mobile and characters “Thermy”
and “Bac”; and, a radio promotion called “Grill the DJ” which featured on-air
grilling and testing of meat using an indoor grill.

The Implementation took advantage of some existing events and also created
some events. Events will all include “Thermy” and “Bac”, as well as a variety of
additional promotional activities.

More information on the implementation is available in the lessons learned report
“Is it done yet? Use a food thermometer to be sure: A social marketing campaign
encouraging the use of food thermometers”, August 2 – 15, 2004. Contact Trent
Wakenight, Consortium Coordinator, National Food Safety & Toxicology Center,
wakenigh@msu.edu, or Paul McConaughy, Program Leader, MSU Extension,
mcconau1@msu.edu for a copy.

Evaluation

A post test was administered in the three counties where the implementation took
place and in a fourth control county. Administration began on August 16, 2004,
immediately after the completion of the campaign. The post test measured how
successful the various strategies were in changing people’s use of food
thermometers when preparing meat and poultry.

A total of 4,000 Boomburbs, 2,000 Heartlands, 2,000 Single Moms, and 1,871
Rural Towns received the survey, equally distributed between the target counties.
Incentives were used to encourage completion of the survey. The return rate was
30%. A phone survey again measured non-response bias.

Survey results show that among Boomburbs in the target counties there was a
12% increase in people using thermometers (sometimes, most times, always)
and a 33% increase in people thinking about using a food thermometer.

Complete survey results are available in the document “Common Food
Preparation Practices, Survey Report, December 2004” available by contacting
Paul McConaughy, Program Leader, MSU Extension, mcconau1@msu.edu.
     Common Food Preparation Practices

                  Survey Report




By
Donald F. Holecek
Teresa I Herbowicz
Tsao-Fang Y. Bristor
Lori A. Martin
Travel, Tourism and Recreation Resource Center
at the Dept. of Community, Agriculture, Recreation and Resource
Studies

and
Paul McConaughy
Family Consumer Sciences
at Michigan State University Extension




Michigan State University
December 2004
Introduction

Michigan Food Thermometer Education Campaign

The Michigan State University National Food Safety and Toxicology Center
(MSUNFSTC), Michigan State University Food Science and Human Nutrition
Department (MSUFSHN) and Michigan State University Extension (MSUE), in
cooperation with the Food Safety Education Staff (FSES) of the Food Safety and
Inspection Service (FSIS), U.S. Department of Agriculture (USDA) and the Michigan
Department of Agriculture (MDA) seek to increase the use of food thermometers in
Michigan. A social marketing approach was adopted to achieve this objective. The focus
of this report is on evaluation of this education campaign (a.k.a. “program”).

Recognizing that food-borne illness is a serious health issue in America, an objective was
included in Healthy People 2010 to reduce food-borne illnesses by increasing the
percentage of consumers who follow key food safety practices, which include using food
thermometer.

Encouraging thermometer use is also important because new research has raised
questions about past advice to consumers to use color and texture of meat as a guide to
safe cooking. Researchers have learned that color is not a reliable indicator because meat
could turn brown without reaching the internal temperature necessary to kill dangerous
microorganisms that are present. Safety could only be assured by using a thermometer.

In 2001, the MDA funded a proposal from Michigan State University Food Science and
Human Nutrition Department, and Michigan State University Extension for a
comprehensive food safety education program for Michigan consumers. The proposal
included an objective to “Develop and implement a social marketing campaign
emphasizing food safety.” Project staff committed to “develop a broad social marketing
campaign based on extensive input from and collaboration with consumers and
stakeholder groups.” The Michigan project staff identified thermometer use as an
appropriate goal for a food safety campaign. The USDA determined that piloting the food
thermometer use campaign in Michigan would be an appropriate step in the national
campaign strategy.

Initial focus groups conducted by Food Safety and Inspection Service showed that
consumer food safety knowledge related to thermometer use was lacking. Thermometer
use was generally reserved for special occasions, such as holiday meals. These focus
groups resulted in recommendations to:
 Promote food thermometer use for everyday meals
 Promote thermometer use as a means to improve taste as well as safety
 Target parents of young children who are responsible for the welfare of an at-risk
    group and also appear open to behavioral change.
Survey Methodology

In order to enhance success of the overall project, a more refined level of audience
segmentation was sought. A commercially available market segmentation system was
selected, which uses statistical model that combine numerous national surveys of
consumer behavior, lifestyle and attitude with annually updated census data. With these
existing data, the segmentation system provides very acute segmentation and audience
descriptions. This approach is often referred to as geo-demographic segmentation. The
US MOSIAC system was selected for the Food Thermometer Education Campaign
(FTEC) social marketing program. Because of what had already been learned about the
willingness of parents of young children to adopt new food safety behaviors, the US
MOSIAC segments with high indices of households with young children were selected:
– Boomburbs - upscale, suburban, two-earner families in newer communities, where
    the newest technological gadgets are frequently sought;
– Heartlands – middle-class Midwesterners with larger families and traditional
    lifestyles, stressing recreation and family activities;
– Rural Towns – rural, low-income, less-educated and underemployed families, with
    limited consumer choices, and
– Single Moms – predominantly African American and Hispanic one-parent families in
    major metropolitan areas, where incomes are low and the parents are young.

Drawing on the input of several experts, including a nutrition educator and an ethnic
marketer, the “Boomburb” parents of children under age 10 were selected as the
campaign target. Reasons included their being most likely to move rapidly through the
stages of behavior change, their propensity for acquiring and using new information, their
tendency to be early adopters of technology, and the ability to capitalize on their role as
major influencers of mass culture.

The survey

One advantage to using geo-demographic segmenting was the ability to test a sample
population in Michigan and have the results be transferable to the nation. Taking
advantage of that fact, a baseline survey was administered prior to implementation of the
campaign. This statewide survey identified where, in relation to the stages of behavior
change, people with children under 10 in all four geo-demographic groups stand.

The survey research for the project was conducted by Travel, Tourism and Recreation
Resource Center at Michigan State University’s Department of Community, Agriculture,
Recreation and Resource Studies. A pre-campaign mail survey was administered to 2,500
Boomburbs, 1,000 Heartlands, 1,000 Rural Towns and 1,000 Single Moms drawn from
the statewide population. Participation was encouraged with the offer of a drawing for
$300, $200 and $100 gift certificates. The return rate was 24%. A phone survey with a
target of 100 completed calls was used to measure non-response bias.

After establishing the research baseline (pre-campaign survey results), a campaign
focused on the target audience – Boomburbs with children under 10 – was implemented
in three counties using paid and free media and public events promoting the importance
of using food thermometers. Ingham, Washtenaw and Kent Counties were selected for
the pilot campaign.
A post-campaign survey was administered in the three counties where the campaign
implementation took place and in a fourth control county (Genesee). The post-campaign
survey was administered immediately after the completion of the campaign to measure
how successful the campaign was in changing people’s views on the use of food
thermometer.

A control county was employed in the study design to assess the influence of possible
non-controllable events on food preparation behavior while the campaign was underway.
Of particular concern was the possibility of an occurrence of a food-related event,
especially one tied to meat, that would generate mass media coverage, which would
confound interpreting pre- and post-campaign results. Fortunately, no such event
occurred during the course of the study.

A total of 4,000 Boomburbs, 2,000 Heartlands, 1,871 Rural Towns, and 2,000 Single
Moms were sampled, proportionally distributed between the target counties. Incentives
were used again to encourage completion of the survey. A response rate to the post-
campaign survey was 30%, far exceeding the expected 20%. A small phone survey was
conducted afterwards to measure non-response bias.

Results of the post-campaign survey indicate the impact of the campaign on a change in
use of food thermometer in meal preparation.

Limitations

In assessing the impact/effectiveness of the FTEC social marketing program, it is
important to consider the following:

1.   The campaign was of relatively short duration, only two weeks.

2.   The investment in the promotion campaign was relatively small, approximately
     $250,000.

3.   The food-thermometer-use promotion campaign constituted only a small portion of a
     plethora of promotional messages on a variety of subjects that residents of the study
     counties were exposed to, thus they might or might not take notice of its message.

4.   Meal preparation involves behaviors that are passed down from generation to
     generation and result in a firmly ingrained manner in which food is prepared even
     among young adults.

Thus, the program faced two challenges. First, the message had to capture the attention of
the targeted recipients who are inundated daily with competing messages from a
multitude of sources. Second, the message, if it cut through the “noisy” communications
environment and was received, had to be powerful enough to convince recipients to alter
firmly established behavior developed over their lifetime. Under such circumstances,
even small pre- to post-campaign changes in results could be deemed meaningful.
Results

The primary focus of the Food Thermometer Education Campaign social marketing
program was on Boomburb households. However, three other types of households were
also included in the pre- and post-campaign surveys: Heartland, Rural Town, and Single
Mom. These four groups constitute “all respondents” in the each of the stages of the
survey (pre-campaign, post-campaign and control county). “Combined respondents”
constitute all four groups from all three stages of the survey combined (i.e. everybody
who responded to the mail survey).

Results are presented in two sections. In the first section, responses from all respondents
to the pre-campaign survey are presented along with responses from all respondents to
the post-campaign survey obtained from the three counties (Ingham, Kent, and
Washtenaw) where the campaign was implemented. In the second section, the similar
analyses were prepared for responses from Boomburb households.


Section I.             All respondents

A total of 4,327 households responded to the mail survey, 1,343 to the statewide pre-
campaign survey, 2,336 to the post-campaign survey and 648 to the control county1
survey. From studying the data, we conclude that there is a pattern indicating that results
of the control county survey are similar to results of the pre-campaign survey, but not to
the results of the survey of campaign counties. Therefore, results from the control county
survey are not included in the report, because they were not a main focus of the study.

The questionnaire used in the pre- and post-campaign surveys was identical. The first
question asked if any children under 10 years old resided in the household. Results are
presented in Table 1. The percentage of households with children under 10 is almost
identical for the pre- and post-campaign surveys (69.7% vs. 68.7%).

Two conclusions can be drawn from the results in Table 1. First, the study sampling
design was effective in identifying households with children under 10. Second, no bias
was introduced between pre- and post-campaign respondents that is attributable to
differences in percentages of households with children in the two groups of respondents.


Table 1.               Q 1. Are there any children under the age of 10 living in your
household?

                                              Pre-campaign               Post-campaign
    Children under 10 years old living in                                                      Combined respondents2
                                               respondents                respondents
    the household                                                                                      (%)
                                                   (%)                        (%)
    Yes                                            69.7                       68.7                       68.2
    No                                             30.3                       31.3                       31.8




1
    Control county- Genesee County where food thermometer campaign was not conducted, but a post-campaign survey
    was conducted.
2
    Combined respondents- all respondents to the pre- and post-campaign surveys, as well as to the control county survey
    combined.
The second question on both surveys involved weekly frequency of preparing meals that
included meat. Results are presented in Table 2. They clearly show that meat is consumed
at most meals prepared by responding households. Hence, meat preparation safety is an
important issue. The results from the pre- and post-campaign surveys are similar enough
to conclude that differences in propensity to consume meals with meat across pre- and
post-campaign respondents are not a source of bias.

Table 2.      Q 2. About how many times a week do you prepare meals at home
that include meat?

                                                       Pre-campaign     Post-campaign
   Frequency of preparation of meat meals per week      respondents      respondents
                                                            (%)              (%)
   0- times (Please go to question #9)                      0.4              1.6
   1-2 times                                                5.9              5.8
   3-4 times                                               27.2             29.4
   5-6 times                                               34.8             35.7
   7 or more times                                         31.7             27.5




The third question was the most important for assessing the program’s impact. It asked
how often a food thermometer was used while cooking meat. Results are presented in
Table 3, and they are insightful. Nearly 63% of Michigan households that prepare meals
with meat never use a food thermometer when cooking meat. A clear majority of young
children in Michigan are at risk of becoming seriously ill from improperly cooked meat
that they consume. The results do, however, suggest that education can alter meat
preparation behavior. The percentage of respondents who never use a food thermometer
declined from 62.7% in the pre-campaign survey to 57.5% in the post-campaign survey.
It is highly probable that this five-point change is attributable to the program.


Table 3.             Q 3. How often do you use a food thermometer when you cook or grill
meats?

                                                       Pre-campaign     Post-campaign
   Frequency of food thermometer use when cooking or
                                                        respondents      respondents
   grilling meat
                                                            (%)              (%)
   Always (Please go to question #5)                        1.6              2.2
   Most times                                               5.3              5.4
   Sometimes                                               30.4             34.8
   Never                                                   62.7             57.5




The fourth question targeted respondents who don’t always use a food thermometer when
preparing meat. The results presented in Table 4 are discouraging. Over 80% of those
respondents reported that they don’t even consider using a thermometer when preparing
meat. The campaign appears to have had only a small positive impact with respect to this
question. However, in the small telephone survey conducted to assess non-response bias
(see Appendix A), a much greater positive impact (over 10 percentage points) was
attributed to the program. While the sample size drawn for the telephone survey was far
too small to be conclusive, the results do support the larger mail survey result of a
positive impact of the program.


   Table 4.        Q 4. The times when you don’t use a food thermometer when you cook or grill meat, do you think about using
   one?


                                                                        Pre-campaign                 Post-campaign
   Thinking about using food thermometer while cooking or
                                                                         respondents                  respondents
   grilling meat
                                                                             (%)                          (%)
   Yes                                                                       18.0                          18.8
   No                                                                        82.0                          81.2




The fifth question addressed the use of food thermometer when preparing meat for young
children. Results are presented in Table 5. Clearly, these results indicate that young
children are at high risk from meat-borne illnesses associated with the manner food is
prepared in Michigan households. About two-thirds of these households never use a food
thermometer when preparing meat for children. Post-campaign results suggest that the
program did influence some households to adopt safer meat preparation practices. The
percentage of those who never use a food thermometer declined from 66.7 in the pre-
campaign survey to 63.8 in the post-campaign survey. Results of the telephone survey
bias check again indicate an even stronger positive impact of the program.


   Table 5.        Q 5. How often do you use a food thermometer when you cook or grill meat for your children under age 10?


                                                                        Pre-campaign                 Post-campaign
   Frequency of food thermometer use when cooking or
                                                                         respondents                  respondents
   grilling meat for children under 10
                                                                             (%)                          (%)
   Always (Please go to question #7)                                           2.0                          2.7
   Most times                                                                  5.8                          5.3
   Sometimes                                                                 25.5                          28.2
   Never                                                                     66.7                          63.8




Question six, like question four, targeted those who don’t always use a food thermometer
but focused on use when preparing meat for consumption by children. Results are
presented in Table 6 and closely follow those reported in Table 4. Even when preparing
meat for highly vulnerable children, the vast majority of households do not use a food
thermometer. The program had only a slight positive impact on this reported behavior.
Results from the small telephone survey, in this case, were equally discouraging.
Table 6.   Q 6. The times when you don’t use a food thermometer when you cook or grill meat for
           your children under age 10, do you think about using one?

                                                            Pre-campaign       Post-campaign
   Thinking about using food thermometer while cooking or
                                                             respondents        respondents
   grilling meat for children under 10
                                                                 (%)                (%)
   Yes                                                          17.2               18.7
   No                                                           82.8               81.3


Question seven and eight followed the same format as questions three and four and
questions five and six. Results are presented in Tables 7 and 8. Even though hamburger is
especially prone to carry pathogens that can cause illness, households are even less likely
to use a food thermometer in preparing hamburgers or to think about using one than they
are in other situations. The program had very small positive impact on the use of food
thermometer judging from pre- and post-campaign results presented in Table 7, and
hardly changed thinking about its use as can be seen in Table 8.

Table 7.    Q 7. How often do you use a food thermometer when you cook or grill
hamburger patties?

                                                            Pre-campaign       Post-campaign
   Frequency of food thermometer use when cooking or
                                                             respondents        respondents
   grilling hamburger patties
                                                                 (%)                (%)
   Always (Please go question #9)                                2.4                2.8
   Most times                                                    2.5                3.0
   Sometimes                                                     6.8                6.4
   Never                                                        88.2               87.8




Table 8.   Q 8. The times when you don’t use a food thermometer when you cook or grill hamburger
           patties, do you think about using one?

                                                            Pre-campaign       Post-campaign
   Thinking about using food thermometer while cooking or
                                                             respondents        respondents
   grilling hamburger patties
                                                                 (%)                (%)
   Yes                                                          13.3               13.8
   No                                                           86.7               86.2




Questions nine and ten were included in the survey to obtain information that could be
used in designing future educational programs on proper food preparation practices.
Results for each of these two questions are presented based on responses from “combined
respondents” (the group that includes pre- and post-campaign survey respondents and
control county survey respondents). This combination is reasonable, because the purpose
of these two questions was to measure a level of trust in and use of different sources of
information about food preparation, levels which would not have been changed by the
campaign.

Results in Table 9 indicate that the two most trusted sources of food preparation
information are “health professionals” and the “USDA”. “Celebrities/popular stars” are
least trusted sources probably because they are often used as spokespersons in
advertisements for all manner of consumer products and many of them may not be
perceived as cooking experts. This is strong evidence for not enlisting popular
personalities in educational campaigns.

Table 9.      Q 9. How much do you trust food preparation information from the
following sources?

                                                   Combined respondents
                                  Very low        Low              High       Very high
 Sources of information
                                    (%)           (%)              (%)          (%)
 Celebrities/popular stars         58.1           34.4              6.8          0.8
 Federal government agencies        4.8           16.6             57.1         21.5
 Food companies                     5.1           25.7             58.5         10.6
 Friends/relatives                  3.1           18.8             56.7         21.3
 Health professionals               1.7            6.6             51.9         39.9
 Local schools                     10.8           38.8             43.7          6.7
 MSU Extension/4-H                  9.7           21.4             52.9         16.0
 News media                        13.1           41.8             39.6          5.6
 Non-governmental organizations    18.3           54.7             24.7          2.2
 State government agencies          7.4           24.4             56.1         12.1
 U.S. Department of Agriculture     4.1           10.9             51.2         33.8


The most commonly used sources of information in preparing foods are cookbooks and
friends and relatives (Table 10). Given that the survey results indicate very low use of
food thermometer when preparing meat, friends and relatives are more likely to be
harmful than helpful information sources of food preparation information. In combination,
results in these two tables suggest that a USDA-endorsed cookbook emphasizing safe
food preparation practices might prove to be effective in changing cooking behaviors; the
USDA is a trusted source and cookbooks are the most often consulted source of food
preparation information.

Another interesting strategy might be to work with publishers of cookbooks to
incorporate a USDA-sanctioned chapter on safe food-handling practices. This would link
the popular appeal of lay press cookbooks with the science related to safe food
preparation and handling. It would combine the best aspects of private publishing with
reliable USDA / government science-based educational materials.

An advantage either of the above strategies would bring to the effort to changing deeply
ingrained food preparation behaviors is that they would in effect communicate the
educational message on a long-term sustainable basis. It would be there each time a
cookbook was consulted in preparing a meal. Media clutter, recall problems, targeting
and proper timing would not be challenges as they were for programs such as the one
evaluated in this study.
Table 10.        Q 10. How often do you use these sources for information about food preparation?



                                                                    All respondents
 Frequency of use of the sources of         Never         Sometimes             Most times            Always
 information about food preparation          (%)             (%)                  (%)                  (%)
 Adult classes                               80.0            16.8                      2.4                  0.7
 K-12 classes                                79.0            17.7                      2.7                  0.7
 Books on cooking                             5.2            41.7                     42.8                 10.2
 Federal government agencies                 42.0            41.7                     13.0                  3.2
 Friends/relatives                            5.2            44.6                     40.6                  9.6
 Internet                                    37.2            46.3                     14.2                  2.3
 Magazines                                   13.7            63.6                     19.9                  2.8
 MSU Extension/4-H                           70.2            22.0                      6.0                  1.8
 Newspapers                                  37.0            53.9                      8.2                  1.0
 Non-governmental organizations              64.8            32.3                      2.5                  0.4
 Radio                                       70.9            25.9                      2.8                  0.4
 State government agencies                   53.1            36.1                      8.8                  2.0
 Television                                  27.5            58.9                     11.5                  2.1


Section II.          Boomburb respondents

A total of 1,838 Boomburbs responded to this study, 590 to the statewide pre-campaign
survey, 986 to the post-campaign survey and 262 to the control county survey.

As can be seen in Table 11, the vast majority of Boomburb households from the pre- and
post-campaign surveys reported having children under 10 as household members. Since
the percentage of respondents with young children is nearly identical in the pre- and post-
campaign surveys, the results obtained do not contain bias due to disproportionate
percentages of households with young children between these two groups of respondents.


Table 11.   Q. 1. Are there any children under the age of 10 living in your
   household?

                                          Pre-campaign          Post-campaign
 Children under 10 years old living in                                                       Combined Boomburbs
                                           Boomburbs             Boomburbs
 the household                                                                                      (%)
                                               (%)                   (%)
 Yes                                          72.2                      72.7                        71.4
 No                                           27.8                      27.3                        28.6




Boomburb households are frequent consumers of meals that include meat, as is the
overall population surveyed for this study. In comparing Tables 2 and 12, it seems that
Boomburb households are slightly less likely to include meat in five or more meals per
week than are all respondents (approximately 62% of Boomburbs vs. approximately 64%
of all respondents).
Table 12.    Q. 2. About how many times a week do you prepare meals at home
   that include meat?

                                                       Pre-campaign      Post-campaign
   Frequency of preparation of meat meals per week      Boomburbs         Boomburbs
                                                            (%)               (%)
   0- times (Please go to question19)                       0.7               1.7
   1-2 times                                                6.8               5.3
   3-4 times                                               30.8              30.8
   5-6 times                                               37.2              36.5
   7 or more times                                         24.5              25.6




Frequency of use of food thermometer in preparing meat is presented in Table 13.
Comparing results for all respondents in Table 3 to those for Boomburbs in Table 13
reveals that Boomburbs use food thermometer somewhat more often than do all
respondents. This table is, again, the most important to assessing the impact of the
program. The pre-campaign survey revealed that 57.0% of Boomburbs never use food
thermometer. As a result of implementing the program, this percentage declined to 51.7%
in the survey of the campaign counties. The percentage-point decline of 5.3 among
Boomburbs was only slightly better than the 5.2-point decline achieved in post-campaign
survey of all respondents. While effective, the program does not appear to have been
more effective in changing Boomburb behavior than it was for all respondent sample.

Table 13.            Q 3. How often do you use a food thermometer when you cook or grill
   meats?

                                                       Pre-campaign      Post-campaign
   Frequency of food thermometer use when cooking or
                                                        Boomburbs         Boomburbs
   grilling meat
                                                            (%)               (%)
   Always (Please go to question #15)                       1.7               2.8
   Most times                                               6.0               5.2
   Sometimes                                               35.3              40.3
   Never                                                   57.0              51.7




Results for question four that targeted non-users of food thermometer obtained for
Boomburbs are also quite similar to the results for all respondents as can be seen in Table
14. Importantly, however, the program was far more effective in making Boomburbs to
think about using a food thermometer than was the case for all respondents. This suggests
Boomburb targeting of the program was more effective than was evidenced in the results
in Table 13.
Table 14. Q. 4. The times when you don’t use a food thermometer when you cook or grill meat, do
          you think about using one?

                                                               Pre-campaign             Post-campaign
   Thinking about using food thermometer while cooking or
                                                                Boomburbs                Boomburbs
   grilling meat
                                                                    (%)                      (%)
   Yes                                                             15.1                     20.2
   No                                                              84.9                     79.8




Boomburbs are slightly more likely to use a food thermometer when preparing meat for consumption by
children than is the case for all respondents, but its use is still very infrequent (Table 15). However, the
program had a positive impact on their use, and the impact was somewhat greater than for all respondents.
This result again supports the effectiveness of targeting Boomburbs in the program.


Table 15. Q 5. How often do you use a food thermometer when you cook or grill meat for your
          children under age 10?

                                                               Pre-campaign             Post-campaign
   Frequency of food thermometer use when cooking or
                                                                Boomburbs                Boomburbs
   grilling meat for children under 10
                                                                    (%)                      (%)
   Always (Please go to question #7)                                2.4                      3.3
   Most times                                                       5.6                      5.6
   Sometimes                                                       29.5                     32.0
   Never                                                           62.6                     59.0




Question six targeted Boomburbs who don’t use food thermometer in preparing meat for
children to consume. Results are presented in Table 16. In the pre-campaign survey,
85.3% of Boomburb respondents indicated that they do not think about using a food
thermometer when preparing meat for children, which is slightly more than the 82.8%
recorded in the pre-campaign survey of all respondents. In the post-campaign survey, the
percentage of Boomburbs who do not think about using a food thermometer declined to
80.1%, a significant 5.2 percentage-point decline. This was a markedly greater decline
than the 1.5 percentage-point decline reported for all respondents in Table 6. This is
additional evidence of the effective targeting of the program to Boomburbs.


Table 16. Q 6. The times when you don’t use a food thermometer when you cook or grill meat for
          your children under age 10, do you think about using one?

                                                               Pre-campaign             Post-campaign
   Thinking about using food thermometer while cooking or
                                                                Boomburbs                Boomburbs
   grilling meat for children under 10
                                                                    (%)                      (%)
   Yes                                                             14.7                     19.9
   No                                                              85.3                     80.1


Question seven addressed the use of food thermometer in preparing hamburgers. The
results for Boomburbs, as reported in Table 17 are as shocking as for all respondents
reported in Table 7. Responding Boomburbs do not appear to realize the high risks
associated with improper preparation of hamburger meat. As was the case for all
respondents, the program appears to have had only a marginal positive impact on the use
of food thermometer in cooking hamburgers. The program appears to have had a slightly
greater impact on encouraging Boomburbs to think about using a food thermometer when
preparing hamburgers. Results presented in Table 18 indicate the program resulted in a
1.7 percentage-point increase (12.2% to 13.9%) in the number of Boomburbs who think
about using a food thermometer when cooking or grilling hamburgers.


Table 17.          Q 7. How often do you use a food thermometer when you cook or grill hamburger
     patties?

                                                            Pre-campaign        Post-campaign
   Frequency of food thermometer use when cooking or
                                                             Boomburbs           Boomburbs
   grilling hamburger patties
                                                                 (%)                 (%)
   Always (Please go to question #9)                             3.1                 3.4
   Most times                                                    1.9                 2.9
   Sometimes                                                     8.3                 7.4
   Never                                                        86.7                86.3




Table 18. Q 8. The times when you don’t use a food thermometer when you cook or grill hamburger
          patties, do you think about using one?

                                                            Pre-campaign        Post-campaign
   Thinking about using food thermometer while cooking or
                                                             Boomburbs           Boomburbs
   grilling hamburger patties
                                                                 (%)                 (%)
   Yes                                                          12.2                13.9
   No                                                           87.8                86.1


Responses by Boomburbs to questions # 9 and #10 are provided in Tables 19 and 20, and
as it was in the case of all respondents, results presented for each of these two questions
constitute combined responses from Boomburbs to the pre- and post-campaign surveys as
well as to the control county survey.

Boomburb responses to these two questions closely follow those for all respondents
presented in Tables 9 and 10. The two most trusted information sources of Boomburbs
are “health professionals” and “USDA.” The most commonly used sources of
information are “cookbooks” and “friends and relatives.” Thus, the strategies implied by
these results are similar to those discussed in conjunction with Tables 9 and 10.
Table 19.        Q 9. How much do you trust food preparation information from the following sources?

                                                                   All Boomburbs
                                         Very low           Low                    High   Very high
 Sources of information
                                           (%)              (%)                    (%)      (%)
 Celebrities/popular stars                 58.1            35.3                    6.2       0.4
 Federal government agencies                3.9            14.8                58.4         22.9
 Food companies                             5.0            24.9                60.0         10.0
 Friends/relatives                          3.0            19.9                56.0         21.1
 Health professionals                       1.2             5.8                52.3         40.8
 Local schools                             10.1            37.6                45.8          6.5
 MSU Extension/4-H                          9.1            22.0                53.5         15.4
 News media                                11.9            41.2                41.8          5.1
 Non-governmental organizations            16.3            55.5                25.9          2.2
 State government agencies                  6.4            22.3                59.1         12.2
 U.S. Department of Agriculture             3.4            10.2                53.4         32.9




Table 20.        Q 10. How often do you use these sources for information about food preparation?

                                                                   All Boomburbs
 Frequency of use of the sources of        Never         Sometimes           Most times    Always
 information about food preparation         (%)             (%)                (%)          (%)
 Adult classes                              79.3            17.8                    2.3       0.6
 K-12 classes                               80.3            16.6                    2.3       0.8
 Books on cooking                            4.6            40.8                   44.6      10.0
 Federal government agencies                41.2            43.2                   12.7       2.9
 Friends/relatives                           4.4            47.2                   39.7       8.7
 Internet                                   31.7            49.3                   16.7       2.2
 Magazines                                  12.2            64.4                   21.1       2.3
 MSU Extension/4-H                          74.5            19.1                    5.3       1.1
 Newspapers                                 33.7            57.4                    8.2       0.7
 Non-governmental organizations             62.6            34.6                    2.5       0.2
 Radio                                      71.4            25.7                    2.8       0.1
 State government agencies                  53.1            36.8                    8.5       1.6
 Television                                 26.5            59.7                   12.1       1.7
Conclusions

Despite the media clutter and ingrained behavior challenges facing this program, it was
clearly effective in modifying meat preparation behavior. Results also confirm that the
percentage of the population that regularly use food thermometer in preparing meat, even
for children, is uncomfortably low. Its use was found to be the lowest in preparing
hamburgers which are the most common sources of meat-borne illnesses. The program
design was effective in targeting households with children. The program itself was
notably more effective in modifying the behavior of Boomburbs. The results from the
non-evaluation questions (#9 and #10) on the survey that were included in search of
guidance for future educational campaigns proved to be especially interesting.

The quality of data obtained and analyzed is quite good. Response rates to the mailed
surveys exceeded our expectations. There appeared to be no bias in the data based upon
internal data comparisons or on a small sample of non-respondents queried by telephone.
Finally, nothing occurred during the campaign that was likely to have influenced meat
preparation behavior of the population in the study area, hence the changes that were
observed are most likely to be attributable to the program that was implemented during
the campaign.

								
To top