Developing and Implementing Surveys to Assess School ...



Developing and Implementing Surveys to Assess School District Physical Activity and Nutrition Policies:

The Experience and Recommendations from a Pilot Survey Conducted in King County, Washington

Summary and Recommendations

Concerning Survey Development and Implementation

Conducted and Reported by:

Epidemiology, Planning & Evaluation Unit and

Parent and Child Health Section

Public Health-Seattle & King County

The Center for Public Health Nutrition at the University of Washington

October 14, 2004

To obtain a copy of the full report, including the survey findings, please contact Epidemiology, Planning and Evaluation, Public Health-Seattle & King County at 206-296-6818

Background

In collaboration with the University of Washington, Center for Public Health Nutrition, Public Health-Seattle & King County (PHSKC) conducted a mail and web-based survey of school district policies related to physical activity and nutrition in King County. The survey targeted the school district level because the district is where policies are implemented that come from the Office of Superintendent of Public Instruction (OSPI) and local school boards.

This report contains a summary of the process of conducting the survey, and recommendations for local jurisdictions or states wishing to implement similar surveys in their school districts.

Specific Aims

The purpose of the survey was twofold: 1. to generate systematically collected data on district policies; and 2. to test this approach for the purposes of developing a statewide school district policy surveillance system that can be used to measure progress toward the goals and objectives of the Washington State Nutrition and Physical Activity Plan, as well as the plans of other constituency groups. The specific objectives were the following:

1. Using existing measures which have been evaluated for reliability and validity, develop a short survey through a consensus process with potential users of these data, including state and local nutritionists working for public health and school food service and physical activity experts, school district representatives, The Children’s Alliance, Action for Healthy Kids and Seattle School Board members to measure key nutrition and physical activity district policies.

2. To inform the development of a state-wide surveillance system which can be used to track changes over time in school district physical activity and nutrition policies, through the pilot development of survey instruments and methods that can be replicated statewide.

3. To gather baseline data on school policies relating to physical activity and nutrition at the district level in King County.

To provide feedback to the State DOH on process issues related to development and implementation of the survey (surveillance tool), such as identifying appropriate respondents, ease of administration, response rates, item non-response, use of web-based response option, number of contacts with respondents required for response rate of 75% or higher, and respondent burden (e.g., time required to complete the survey).

Methods

Study Design

We conducted a self-administered, mail survey of school district policies relevant to physical activity and nutrition in King County school districts. Mail surveys are appropriate for populations for which there is a reliable address list and when those who are being surveyed are able to respond accurately and completely to the survey in writing. They are often successfully used with professional populations. Additionally, mail surveys provide the respondent with the ability to check their records or to consult with others who may be better able to provide factual information.[i]

We also posted the survey on the World Wide Web and advised the respondents that they could answer using the pen and paper method or the electronic method.

Survey Development

The most comprehensive source of reliable and valid questions related to school nutrition and physical activity is the School Health Policy and Programs Study, An expert panel developed the questions on the SHPPS survey and the majority of questions have been demonstrated to have good reliability and validity.[ii]

To develop the PHSKC school district survey, we first identified the SHPPS questions that matched the objectives established for the Washington State Nutrition and Physical Activity Plan, Action for Healthy Kids Model Policies and relevant Healthy People 2010 objectives. Next, state and local nutrition and physical activity experts were identified and invited to review the draft survey to determine its appropriateness for King County school districts.

Due to the short timeframe for survey development and the busy schedules of local nutrition and physical activity experts, we decided to solicit input on draft survey questions by email and telephone. A consensus process was used, with several opportunities for input from the expert reviewers.

The reviewers provided valuable feedback and guided question selection. Their input is summarized below:

• Focus on district policy questions

• Reduce the number of questions

• Relate questions to Washington State Nutrition and Physical Activity Plan or Action for Healthy Kids Model Policies

• Request copies of nutrition and physical activity district policies

• Be aware that district policies will likely not reflect school-level practice

While the majority of reviewers stated that the data gathered in the pilot survey would be helpful in augmenting their own work, several reviewers expressed the concern that district policies are likely to vary greatly from school food service practice. District policies tend to be general, whereas food service practice may surpass the requirements of district policies. Some food service directors were concerned that only assessing district policies might inaccurately characterize the nutritional quality of food being served to students in schools.

Criteria for Finalizing Survey Content

Based on reviewer recommendations, the following criteria were developed to guide the wording and content of the surveys:

• Where possible, we tried to keep the original SHPPS wording so that comparisons between local and national data could be made. Wording was changed in only a few cases where reviewer comments made it clear that the SHPPS wording was ambiguous, imprecise or inappropriate in the local context.

• Based on reviewer suggestions, questions were eliminated that did not provide information about policies at the district level.

• Questions were eliminated that did not provide information related to the goals and objectives of the Washington State Nutrition and Physical Activity Plan or Action for Health Kids.

• A few questions were added to capture information about the goals/objectives in the Washington State Nutrition and Physical Activity Plan and Action for Healthy Kids for which there were no corresponding items on the SHPPS survey. These are questions #32 on the Food Service questionnaire, and #29, 30 and 31 on the Physical Activity questionnaire. However, the new questions have not been tested for reliability or validity.

• One reviewer suggested adding questions to the survey to assess physical activity policies specific to asthma. These questions were not added as they were outside the scope of the project goals and objectives and they would have lengthened the survey. Questions not focused on nutrition or physical activity policies were not included to keep the survey length shorter.

The final survey drew heavily from the existing SHPPS survey questions, making state and national comparisons possible for most items.

Identifying the Respondents

All school districts in King County were included in the survey (n=19). District superintendents were informed by a personal letter about the survey and asked for their support, prior to the first survey mailing. A letter was sent to all superintendents in King County, introducing the survey, and signed by Donna Johnson of the Center for Public Health Nutrition. The letter informed superintendents that we wished to survey the individual/s in the district most knowledgeable about physical activity and nutrition.

Follow-up phone calls were made to superintendents’ offices to obtain contact information for these people. In most cases there were two respondents for each school district. The respondents were primarily supervisors or managers of food service programs (or nutrition) for the nutrition survey. Respondents were more varied among those who replied to the physical activity survey. These included an assistant and deputy district superintendent, a director of curriculum, a director of elementary education, and three directors of physical education. The remainder were physical education teachers.

Ensuring Adequate Response Rates

One of the biggest concerns with mail surveys is ensuring an adequate response rate, since low response rates may introduce bias. French,[iii] in a statewide survey of school nutrition policies, reported a 75% response rate after two mailings, and a similar response rate in a smaller study.[iv] The SHPPS achieved a 100% response rate in the state-level survey and an 82% response rate at the district level using mail surveys, with telephone follow-up.[v] Based on this, we anticipated a 75-85% response rate after two to three mailings and telephone follow-up.

The process that we followed to ensure an adequate response rate was the following:

1. All school district superintendents in King County were informed about the survey through a personal letter that asked for their support and aid in identifying the appropriate respondents for each survey, prior to the first survey mailing.

2. Subsequent mailings included the survey instrument and a cover letter, on Health Department letterhead, explaining the purpose of the study and the importance of participation.

3. Respondents were given the option of responding by mail or on-line.

4. Two follow-up mailings were sent to non-respondents.

5. After the third mailing, non-respondents were contacted by phone, and for those who were willing to participate, the survey was administered by phone.

The following are cumulative response rates after each mailing:

Response rates after 1 mailing: 39.5% (15 of 38 received)

Average amount of time between 1st and 2nd mailing: 10 days

Response rates after 2 mailings: 55.3% (21 of 38 received)

Average amount of time between 2nd and 3rd mailing: 7 days

Response rates after 3 mailings: 57.9% (22 of 38 received)

Follow-up phone calls were made one week after postmark of 3rd mailing.

Response rates after telephone follow-up: 78.9% (30 of 38 received)

There were eight non-respondents. Four were refusals, and included three physical activity respondents and one nutrition respondent. The reasons for refusal were different for each individual. One person said there was no one serving in the position that would typically have the expertise to complete the survey. Another refused to participate because he said there were other ways to collect the information requested. Another said he did not have time to complete the survey because it was the end of the year and the survey was not a priority. Finally, the nutrition non-respondent refused to participate because she felt the survey was written from an uninformed perspective, therefore, she was uncomfortable providing answers to certain questions.

Data Analysis

Data were analyzed using Epi Info version6. The accuracy of data entry was ensured by double entry.

Evaluation of Survey Process

Because this was a pilot, one of our objectives was to provide feedback to the state on the process of survey implementation. In order to meet this objective, we conducted follow-up telephone interviews with 25 out of the 30 respondents to the survey. The purpose of the follow-up telephone interviews was to hear about their experience completing the survey. The following topics were assessed:

-Professional knowledge and ability to complete survey

-Question difficulty

-Question relevance and importance

-Suggestions for future survey format and content

-Respondent burden

Findings from Evaluation Interviews

Professional Knowledge and Ability to Complete the Survey

Finding the appropriate contact

Determining who in the respective school districts was the most appropriate respondent was an important consideration in ensuring the validity of the findings. In this case the persons most knowledgeable about district policies relating to nutrition and physical activity were sought. Superintendents were notified by mail and asked to identify in a follow-up phone call, the “staff members who have primary responsibility for, or are most knowledgeable about” nutrition and physical activity policies and programs. In the majority of the nineteen school districts contact information was collected for two respondents, one for nutrition and one for physical activity.

Process evaluation interviewees all stated they were the appropriate person in their District to complete the survey. However, almost half of the respondents (12 of 30) were not the original staff members the superintendent identified as “most knowledgeable.” Other staff members completed seven of the physical activity surveys and five of the nutrition surveys. The original individual designated by the superintendent often chose to delegate the survey to a staff member closer to the field. For example, in one district, the Executive Director of Elementary Programs had a physical education teacher fill out the physical activity survey. As in most self-administered mail surveys there is little to no control over who actually fills out the questionnaire.

Providing an answer

Mail surveys allow the respondent the ability to check documents and records or to consult with others who may have a better understanding of the question at hand. Given this, many survey respondents stated that they had conferred with other staff members or looked up policy documents while completing the survey. One physical activity survey respondent, an athletic director, commented, “I had to look stuff up, and I filled it out with the Director of Instruction.”

Half of the nutrition survey respondents commented that they had to confer with other staff members or rely on policy documents to answer questions relating to vending and soda contracts. In those cases it was not the role of the Food Service Director or part of Food Service Management to oversee and operate vending and soda sales. Survey respondents added that it is usually the student body that runs vending and soda sales, and the individual schools that negotiate contracts.

Question Difficulty

Self-administered surveys allow the researcher to ask more complex questions that may have a long list of response categories. The respondent has the luxury of answering the questions at their convenience, and as mentioned above, they can confer with others and their records to provide the appropriate response. However, self-administered surveys allow the researcher no control over how people interpret the questions.

Policy vs. Practice

Policy and practice are two terms that are frequently used together, and sometimes mistakenly used interchangeably. To avoid confusion, we provided a definition of policy in the “Special Instructions” section in every questionnaire. However, respondents reported that they struggled with disentangling their practices from policy. One nutrition survey respondent said, “The questions were vague, kind of hard to say yes or no to because we do things even though it is not written down that we have to” (Food Service Director). A director of nutrition services remarked that the questions were difficult to answer as answering “no” made it seem as though the District was doing things wrong. For example, question 11 asks, “Has your district adopted a policy stating that schools will offer students 3 or more types of milk each day?” Some respondents wrote in the margin that they do offer 3 types of milk, but they don’t have a policy stating they are required to provide 3 types of milk.

Physical activity survey respondents were equally frustrated and had difficulties being asked about policy and not practice. “What is the definition of school policy?” asked one respondent during an interview (Athletic Director). Another respondent, a coordinator of health and physical education, said that the information she provided would have been more accurate if she had been asked about procedures. For example, question 7 asks, “Based on policies adopted by your district, how much physical education are students required to receive while in elementary school?” In the space provided, respondents wrote the time students participated in physical education even though they acknowledged that time was not a requirement derived from district policy.

The comments from physical activity and nutrition survey respondents indicated that both groups had difficulty answering policy questions. The difficulty was not a result of lack of information to answer the questions. The difficulty appeared to be separating policy from practice and then confidently providing a correct answer. Another respondent expressed concern that by limiting the questionnaire to inquiries about district policies that we were missing important information about USDA guidelines – the guidelines that her district and all districts receiving USDA funds follow.

Experience

Characteristics of the respondents influenced their ability to quickly complete the survey. Some respondents who were new to their current position reported that it took them longer to fill out the survey than it might have taken someone with more experience. As one Food Service Director reported, “The questions felt vague and they took me longer to answer because I’m new to the position.” The survey respondent also added that she had been Assistant Director prior to becoming Director, and still there were items on the survey with which she was not immediately familiar.

Aside from difficulties mentioned above, the majority of survey respondents did not find the questions difficult to understand. Most said that the questions were straightforward and easy to understand.

Question Relevance and Importance

One purpose of the physical activity and nutrition surveys was to collect and measure data on key nutrition and physical activity district policies. Here are the respondents’ comments on the questions themselves and their importance.

Physical Activity

The majority (9 of 11) of the physical activity survey respondents stated that the survey questions asked were relevant to physical activity policies in their district. However, many of them pointed out the importance that curriculum and procedures play in their physical activity programs. “It would be more accurate to talk about procedures because we don’t necessarily have policy” (Coordinator of Health and Physical Education).

Nutrition

Unlike the physical activity survey respondents, the nutrition survey respondents were not as confident that the questions asked were relevant to nutrition policies in their district. About one-third (6 of 16) of survey respondents stated that the questions were not relevant. The main source of concern was on the topic of vending and soda contracts. Respondents said that vending and soda contracts had nothing to do with Food Service. A nutrition service specialist who participated in the survey stated the following: “The questions were geared to things that go on in student stores and with vending, and not food service” (Nutrition Service Specialist).

It is worth noting that one respondent reported that her district did not have “district” policies, therefore the questions were not relevant. She said they have food service policies, but they are not district policies coming from the school board or other governing school district body.

Suggestions for Future Survey Format and Content

Format

Only three respondents said that changes to the survey format would make the survey easier for them to complete. One recommended adding an open comments section. Another said the questions were too broad and could be more specific, and in particular mentioned the vending, soda contracts, and advertising questions.

A physical activity respondent was not in favor of the questions that asked him to mark a box if he was attaching documents. He said it was easier for him and less time consuming to just write in the requirements, since the survey provided the space, rather than find the documents and attach them to the survey.

Physical Activity Survey Content

Although there were no questions that were consistently unanswered, half of the physical activity survey respondents had ideas about how to improve the content of the survey. The main recommendation for improving the content was to ask more about curriculum and procedures. Specifically, one respondent suggested wording the questions differently so as not to target just policy. They thought this would help the researcher learn about procedures they practice for which they do not necessarily have a policy.

When asked what other policy issues are important in their particular district, one survey respondent mentioned district insurance policies and the role they play in potential litigation. “There are a lot of things we can’t offer, like roller-blading, climbing walls, and bicycling because liability is an issue.” a respondent noted. One respondent, a Health and Physical Education Curriculum Developer, stated, “You should ask insurance policy questions because you will find a difference in districts as to what activities they can offer because of their insurance situation.”

Nutrition Survey Content

Half of the nutrition survey respondents had ideas about how to improve the content of the survey. A consistent comment was that the survey focus more on food service practice rather than policy. One respondent said, “If the survey is designed to collect information for statewide data collection it needs to be improved to address the issue of current practice as defined in USDA program guidelines” (Associate Superintendent). Another respondent pointed out that because the wording of “policy” elicits a different response than “practice” it would be helpful to first ask about practices to get the participant to understand the difference between practice and policy. He added, “We do things even though the district does not say we have to.” (Director of Food Services).

Similar to the physical activity questions, there were no questions that stood out as particularly problematic, as item non-response showed no particular pattern across respondents.

One survey respondent, a Food Service Director, gave very specific feedback about some particular questions she did not feel were appropriate. She stated that asking about the fat percentage of only meats and poultry products was not realistic: “The fat percentage is calculated for the whole meal, not just the meat product.” She also said that asking about policies regarding how many milk types and the number of foods containing whole grains offered is not something with which a school district would be concerned. She thought that was too specific for a district to have a policy dictating what school food service provides.

Respondent Burden

Though the self-administered mail survey design used in this investigation is associated with lower respondent burden, it is still an important issue to consider. While some respondents may welcome the opportunity to participate and recognize the importance of providing data on their district to help assess physical activity and nutrition policies, others perceive surveys as an imposition to their time.

Importance of Participation

Some survey respondents readily comply when asked to complete a survey. And according to evaluation responses, the majority of respondents had no concerns about participating. However, a few wanted to know why they were being asked to participate and how the information would be used.

It is worth noting that one physical activity respondent and one nutrition respondent were very concerned about participating in the survey. They did not feel they were thoroughly informed as to why they were being asked to provide information on district policies and what specifically was going to result from their participation. It is possible that in these cases the survey was passed on by the original recipient, perhaps without the cover letter describing the purpose of the survey.

Time

Survey participants reported that completing the survey took them anywhere from 5 to 60 minutes. On average the nutrition survey took nine minutes fewer to complete than the physical activity survey (15 and 24 minutes respectively). One survey participant, a Chief Academic Officer, stated emphatically that completing the survey was an imposition on his time. “I was the appropriate person to complete the survey but it was not a good use of my time. It was an imposition that the survey took so long. You need to understand the time commitment you are asking people to make.”

Use of Web-based Response Option

The web-based format was used by 14% (2/14) of those answering the physical education questionnaire. Twenty-five percent of nutrition questionnaire respondents opted to use the web. One respondent reported that she returned the survey on-line; however it was never received. This may be due to user or technical error.

Use of U.S. Postal Service for Returning the Survey

The majority of respondents opted to use the postal service to return their questionnaire. Regrettably, one survey was lost in the mail. Since envelopes with return addresses and postage stamps accompanied the surveys, there is no explanation for this survey loss.

Recommendations for Statewide Implementation

Survey Content

Suggestions that Apply to BOTH Surveys

1. Recommendation: Require respondents to provide their name on the questionnaire (it was optional on the pilot survey).

Rationale: Having a name will make it much easier to follow-up with a respondent in case there are any questions concerning a returned, incomplete questionnaire. Although we had the names of each intended respondent, in some districts, intended respondents sometimes delegated the survey to another employee. In these cases, when names and titles weren’t provided, it was more difficult to re-contact respondents for whom data was missing. This said, there are drawbacks to requiring the provision of names – the major one being that non-confidential surveys can result in the respondent being more likely to provide socially desirable answers.

2. Recommendation: Include the following question:

Are your District’s policies on nutrition and physical activity included on the District’s website? (yes/no)

If yes, please provide the web address where the policies are located.

Rationale:

Respondents who did not send us their district’s policies with their survey indicated that their policies are readily available on the web. A staff member, who is well versed in locating information on the web, searched for the policies of each District on the web. She found that 18 of the 19 districts in King County have web sites. However, she also reported that it was either “very hard” or “not very easy” to locate the district policies on the majority of these web-sites. Having the web site address would allow the state to locate policies more easily and then, if desired, validate the responses that were provided. The state could also conduct a content analysis of the policies and identify strengths across the District policies and any gaps that should be addressed.

3. Recommendation

If the State is interested in which policies exist and how they are implemented, questions will need to be developed that inquire about the Districts’ practices. These practice questions should be followed with questions that ask if there is a district policy for the issue at hand. If the State is interested in determining what the District schools’ practices are, then it is recommended that the State conduct a survey at the school level to determine how policies are implemented. The SHPSS survey has a school-level practice module.

Rationale: Several respondents hand-wrote next to the policy questions that their practices are not mandated by policies. Schools are the settings in which policy is implemented in practice – or where practices are performed because District policies are “silent” or not specific about practices required to implement policies. A school-level practice survey would complement the district-level policy survey.

4. Recommendation:

Add an “additional comments” section to the end of the survey to provide respondents with the option of clarifying any answers or making any additional comments about the survey.

Rationale:

It was common for respondents to write comments in the margins next to questions. Providing a section for “additional comments” will provide respondents with more room for explaining their district's policies or clarifying their responses.

Suggestions for the Physical Education District Survey

5. Recommendation: Include the following question in the physical education district questionnaire: “Does your district have a person who has the primary responsibility for the physical education and activity curricula and programs?

Rationale:

Adding this question is recommended because in the pilot survey, few districts had one person responsible for the physical education program.

6. Recommendation: Re-write questions concerning time in the Physical Education District Questionnaire (questions 7, 17, and 20) so that the respondent must provide answers in common units of time. These units would be specific to the school age (e.g., for elementary school, the unit would be hours per week; for middle school and jr. high and high school it would be hours per week x semesters per year. This could be done by providing response options that respondent checks off. After establishing whether or not a credit is equal to the same number of hours across districts, an equivalency chart could be provided that shows how many hours are included in a credit so that all answers are provided comparable.

Rationale:

In the pilot survey, respondents were allowed to write in their responses to these questions and these tended to be in different time units – especially for the middle/jr. high and high school questions. Establishing common units will result in more interpretable findings.

7. Recommendation: Re-write physical education requirement questions to reflect Washington State laws concerning physical education. For example, the question could be re-written as the following:

State law requires 4 semesters (2 years) of physical education credits in order to graduate from high school. How many PE credits does your District require students complete in order to graduate from High School?

Rationale: Asking this question will allow the State to determine if most Districts are adopting State standards for graduation, or if any have more additional requirements.

Suggestions that apply to the Food Services District Questionnaire

8. Recommendation: Change question 4 in the food service district questionnaire to: “Does your district require that schools offer meals that have a total fat content that is limited to 30% or less of calories from saturated fat?”

Rationale: We suggest this change because several respondents indicated that fat content is provided for entire meals, not just the meat itself that is contained in it.

9. Recommendation: Include the following question with yes/no response categories:

“Is your District food services department responsible for

1) breakfast

2) lunch

3) after-school or extended day program snacks

3) staff meetings

4) meetings attended by students’ family members

5) school stores, canteens, or snack bars

6) vending machines

7) concession stands

If no for any of the above questions, please indicate which departments are responsible for these food services. If you aren’t sure about which District departments are responsible for these different settings that provide food, can you please investigate and provide the answer?

Rationale: We suggest adding this question because half of the respondents in the Food Services District process survey responded that they are not responsible for the provision of soft drinks or junk foods. Additionally, Action for Healthy Kids recommends that all food items sold in schools should be under the management of the Food Services Program, this question would help measure progress toward this objective.

Survey Administration

Recommendations that Apply to BOTH Surveys

10. Recommendation: The School District Superintendents should be aware of the survey prior to receiving the letter of introduction to the survey. The results of the pilot survey could be presented to school administrators at one of the Washington Association of School Administrators meetings as a way to introduce a possible statewide survey. In addition, a description of the process that will be used at the State level should be described to them at this meeting. Announcements could also be made in any newsletters that the District Superintendents receive and other communication channels.

Rationale: It is important that superintendents be aware of the survey and to have their full support since they are gatekeepers for participation for their District. It is also possible that they could provide important feedback about the survey and its administration if they are more aware of its content and the process of implementation.

11. Recommendation: The cover letter introducing the survey to the potential respondents could be improved so that it is more concise and emphasizes the following: 1) the role of policy and its importance; 2) a description of how policy may differ from practice; 3) most knowledgeable staff selected to complete survey and 4) the clarification that the survey focuses on determining District policies and is not about evaluating the of the District or the staff. It is suggested that, as was done in the pilot survey, the selected respondents be informed of how the data will be used and that the results of the survey will be available . Since everyone being surveyed is aware of the epidemic of overweight and obesity in students, this data can be deleted from the letters of introduction.

Rationale: The cover letter could be improved upon based on feedback that we received in the process evaluation. A strong response rate from the mailings was obtained in the pilot, however, the cover letter introducing the survey could be more concise and focus on issues that were identified in the process evaluation discussions.

12. Recommendation: The cover letter that introduces the survey to respondents should be stapled to the questionnaire so that whoever completes the survey will have the opportunity to understand why the survey is being conducted.

Rationale: This is recommended because during our process interviews we spoke with a few people who did not know the purpose of the survey because it had been passed onto them to complete by the original recipient without the introductory letter.

13. Recommendation: Conduct two mailings instead of three.

Rationale: The third mailing did not significantly increase response rates. However, this may have been related to the fact that the end of school was imminent and the survey was competing for staff’s very limited time.

14. Recommendation: Follow-up the first mailing with a post-card reminder with the internet address for the web-based survey. The reminder should be mailed so that the respondent receives it 3-4 days after receiving the survey.

Rationale: This may help increase response rates following the first mailing and possibly increase web-based survey usage.

15. Recommendation: Follow-up the second mailing with a personal phone call reminder to the recipient.

Rationale: As indicated by the pilot data, this most likely will help to increase response rates.

16. Recommendation: The State should consider having a person following-up with respondents to capture missing data on surveys.

Rationale: Most of the returned surveys had missing data that was obtained from follow-up phone calls. This improves the quality of the data significantly.

17. Recommendation: Offer both pen and paper and web-based administration methods with e-mail reminders.

Rationale: Although web-based responses were low for the pilot survey, email reminders with links to the web questionnaire may increase response rates.

18. Recommendation: The State should consider the academic calendar of Districts prior to implementing the survey and carefully select a time when other surveys or standardized testing are not being conducted and avoid the beginning and end of the academic school year.

Rationale: Out of necessity, the pilot survey was implemented at the end of the school year. This is a time when District staff are busy wrapping up their responsibilities prior to summer vacation.

Dissemination of findings

1. Recommendation: It is important that the results of the survey be given to the respondents and the superintendents. They should be allowed to see their survey responses as well as the aggregated data. This can be done in the form of a follow-up report that accompanies a note thanking them for their participation. It is also recommended that the results be disseminated at the Joint Conference on Health, the School Food Service Directors Annual Meeting, Washington Association of School Administrators, PSTA, and to the media to inform the public.

Rationale: Access to and awareness of information on District policies that relate to nutrition and physical activities in the schools may result in improved policies to combat the growing overweight epidemic.

Coordination of State, District, and School-level Surveys

1. Recommendation: Given the increasing number of research studies that focus on youth nutrition and physical activity, it is critical that the State, or some institution/agency, create a system to coordinate the content and implementation of school-based surveys at the State, District, and School levels.

Rationale: Coordination of surveys will reduce replication of efforts and reduce respondent burden. It also provides an opportunity for researchers to complement each other’s efforts and build on each other’s findings.

Feasibility of This Approach for Statewide Policy Surveillance

A mail and web-based survey is a feasible means of collecting data on School District policies related to physical activity and nutrition. We achieved a 79% response rate after three mailings and telephone follow-up. However, the third mailing yielded few additional responses. Therefore, we think that it would be possible to achieve an adequate response rate from a mail and web-based survey with two mailings and telephone follow-up.

-----------------------

[i] Salant, P., & Dillman, D. A. (1994) How to conduct your own survey. New York, NY: John Wiley & Sons.

[ii] Smith TK, Brener ND, Kann L, Kinchen SA, McManus T, Thorne J. Mehodology for the School Health Policies and Programs Study 2000. J Scho Health. 2001;71(7):260-265.

[iii] French SA, Story M, Fulkerson JA. School food policies and practices: A state-wide survey of secondary school principals. Journal of the American Dietetic Association 2002;102(12):1785-1789.

[iv] French SA, Story M, Fulkerson AJ, Gerlach AF. Food environment in secondary schools: a la carte, vending machines and food policies and practices. AJPH 2003;93(7):1161-67.

[v] US Department of Health and Human Services, Centers for Disease Control and Prevention. School Health Policies and Programs Study.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download