End Commercial Tobacco Campaign Public Opinion Survey

This page provides guidance on the Public Opinion Survey for the 2022 End Commercial Tobacco Campaign. 

Quick links:

Clink the links below to jump to that section.  Or click the links on the right to download important information.

Background

The purpose of this survey is to collect data about people’s knowledge, attitudes, and perceptions regarding the tobacco industry’s influence in the community(ies) where your LLA is focusing its End Commercial Tobacco Campaign efforts.  Survey results can inform your project about public sentiments, which you can try to leverage into local policymaker action, stakeholder engagement, community advocacy and education, and much more.

Parts of the survey and pattern of questions

In this document, questions with a hashtag or pound sign (#) are required to be included in the survey for all LLAs.  There are additional required questions depending on the endgame objective in your LLA’s workplan.  Please be sure to review the table of required questions to make sure you have all the necessary CTCP-required questions in your survey. 

There are options for the respondent to opt out of any question in the survey, including ones with the #.  This is not to be confused with the asterisk (*) to show that a response is required for the survey to continue.  When you view the survey in SurveyAnalytics, almost all the questions will have an asterisk and will force a response to the question.  This is done so that a question is not accidentally skipped.  If the respondent does not know how to answer or does not have an opinion about the question, then the data collector should select the response, “I don’t know” or later, “I prefer not to answer.” 

LLAs are able to reorder, edit, or delete questions and answer choices of the non-required questions (questions without #) to better align with project goals.  The survey should start with generic topics.  Then it covers the required and optional questions on relevant policy options.  The last section asks demographic questions about the respondent.  After your LLA makes its edits, this should still be the general pattern of questions. 

Importance of safety

If surveying in-person, the survey should always be conducted in areas that are considered safe for the data collectors.  Immediately leave any situation that feels unsafe.  Stop any survey if it feels as though the respondent is becoming overly irritated or aggressive.  If possible, collect data in groups of two or more per site.  

Follow public health or COVID-19-related protocols and recommendations from local health authorities to ensure health and safety of all while surveying (e.g., maintain six feet of distance between data collector and respondent, wear a mask, properly sanitize devices between uses if giving them to a respondent, etc.) 

Survey administration

For in-person data collection, a trained data collector must administer the surveys on mobile devices loaded with SurveyPocket.  The survey is designed for each data collector to read each question aloud to a respondent and enters the respondent’s answer into SurveyPocket on a mobile device.  In some cases, it may make sense to allow the respondent to enter their own answers onto the devices.   

Another option is to conduct the survey by sharing a link to complete the survey online.  The survey prompts below have some additional considerations for setting online data collection, including at least one location screening question, such as zip code, to avoid bots.  Either way, data collection must be done through SurveyPocket on a mobile device or a link from SurveyAnalytics online.   

Since this is part of the official coordinated data collection efforts, methods and tools should be as uniform as possible.  Here are some important things to note:

  • Questions presented on the screen should not be skipped, so if the respondent does not know how to answer or does not have an opinion about the question, then the data collector should select the response, “I don’t know” or later, “I prefer not to answer.”   

  • If a respondent wants to change a previous answer, select the “Back” button located in the top left corner of the screen.   

  • If someone does not want to complete the entire survey, that is ok.  You can stop in the middle of a survey by selecting the home button located in the top right corner of the screen.  Save it as “Incomplete,” and it will save data up to that point.  Then the device will start a new survey for the next respondent. 

In some situations, it may be beneficial for data collectors to identify themselves with your health department, wear clothes or a badge that associates them with the department, or have department signs nearby.  In other situations, especially in areas that are less likely to respond to a government survey, it might be better for data collectors to identify as a member of the coalition or other organization and wear more casual clothes.  Be strategic when assigning data collectors to neighborhoods.  Someone from the local community organizations may have a trusted reputation to help gather data.   

Who to survey

Focus efforts to survey people that live in the specific jurisdiction listed in your workplan for the End Commercial Tobacco Campaign objective.  It may be strategic to oversample (survey more people) from a specific community or other demographics.  For example, it may be important to survey more people that live in an undecided policy maker’s voting district.  Another example is if you are working on smokefree MUH, it will be beneficial to survey more people that live in MUH.   

Some decision makers may not want to hear from visitors or tourists and only want data from their residents.  If this is the case, the jurisdiction or zip code question can be used as a screening criteria so that only residents of the jurisdiction can take the survey.  In other cases, especially where tourism is important, decision makers may welcome data from visitors.  If this is the case, the jurisdiction or zip code question would not be used as a screening criteria, and anyone can take the survey. 

Only survey respondents 18 or older.  If you want to survey people under 18, please consult your agency’s policies on obtaining parental permission. 

Gaining a representative samples

A public opinion survey is a convenience sample – that means that you survey whoever is around your chosen location(s).  Because data collectors tend to approach people who look like themselves, their sample ends up reflecting who they are (e.g., younger vs, older, non-smokers vs. people who smoke, etc.), instead of the actual population.  There are a few ways to try to make your sample as representative as possible, so that who is surveyed better mirrors the overall population of your community.   

One method is to set quotas for various characteristics such as age or ethnicity, based on the population breakdowns of your county OR you can instruct your team to approach every 5th person who walks by, for example.  Another recommendation is to ensure the present population is naturally diverse.  For example, conducting surveys at a health fair, farmers’ market, or a 5K walk is likely to result in a high proportion of respondents who do not smoke and are interested in healthier lifestyles.  Instead, surveying passersby on a safe corner of a dining district on a busy weekend may result in a more diverse sample.  Whichever route you go, instruct your team during the training, and then observe in the field that this is being implemented properly.    

 Please see your workplan for the number of people that need to be in your sample.  You will want to attempt to survey additional people so that if you need to throw out any data (like incomplete surveys), you will still reach your target sample size.    

 Guidance on sampling can be found at:  

When to conduct the survey

According to each LLA’s workplan, data collection is scheduled to take place July 1 – December 31, 2022.  This is a wide time period to allow for some flexibility among different programs.  However, it is recommended that an LLA completes data collection about a month or so after starting.  There is no specific time frame requirement, but a more limited time period makes your data stronger than a wider time period.   

Data management, analysis, and reporting

Since the survey must be administered through SurveyPocket on a device or online through a link from SurveyAnalytics, here are the instructions to get access to the survey and copy it into your LLA’s account: https://vimeo.com/383413481 

Each LLA must copy the survey into their own account from a folder called, “2022 ECTC Public Opinion Survey.”  Once you copy the survey into your LLA account, you can customize the greeting and edit the questions without an asterisk.  You can also add more questions that would be helpful for your LLA’s efforts.  Keep in mind that you want to include the questions that are relevant to your LLA’s objective.  You might want to include future potential areas of interest for your LLA or efforts of other nearby CTCP-funded partners. 

Please be sure to review the table of required questions to make sure you have all the necessary CTCP-required questions in your survey.

 Instructions for managing your LLA’s survey in SurveyAnalytics (setting up the survey, connecting to devices, syncing devices, analyzing data, and reporting results) are available on the TCEC website https://tobaccoeval.ucdavis.edu/using-mobile-technology.  You may also contact TCEC at tcecTA@phmail.ucdavis.edu.

Great!  Where do I start?

  1. Review this page and the survey (there's a link to download the survey on the right or bottom of your screen).  
  2. Login to your LLA's SurveyAnalytics account to edit the survey.  Guides for how to navigate SurveyAnalytics are on our website and there are videos at the bottom of the page.  
  3. Copy the Public Opinion Survey from TCEC's Survey Analytics account into your LLA's Survey Analytics account.
  4. Edit your survey in SurveyAnalytics so that it is customized to your intended audience and purpose.  In particular, edit the introduction text so that it lists the organization your data collectors represent.
    1. If you plan on conducting the survey in Spanish, please note that the Spanish translation has already been included in your SurveyAnalytics account.  However, any changes made to the English version should also be made in the Spanish version.  
    2. Make any changes in the English version, i.e. adding, deleting, or editing any of the informational text, question text, or answer options. 
    3. Make any changes in the main, "Edit" window in your SurveyAnalytics account.  Then edit to the Spanish translation by go to: Languages > Manual Translation > Question & Answer Translation.  The new questions you've created should have a new, blank corresponding box to enter the Spanish translation.  To be clear, you are not creating a whole new survey in Spanish.  The SurveyAnalytics program allows you to collect data in both English and Spanish in the same survey.  Here's a video on how to edit the Spanish text.
    4. Note that the data are all put together into one data set.  If you want the Spanish and English data to be separate, we suggest creating two separate surveys.  You may want to take a few test surveys in both English and Spanish so that you can see what it looks like in terms of collecting the data and then later seeing how the data are stored in SurveyAnalytics.
  5. Test and distribute your survey.
    1. If you plan on conducting the survey on your mobile devices, the devices need to be connected to the survey.  It is NOT the same as what you used for the observations.  Each LLA needs to create unique device keys so that the devices are linked to the LLA’s public opinion survey.  Be sure to synch all devices that will be used to collect data.  You'll also need to download the SurveyPocket app onto each device.  Click here for Apple devices. Click here for Android devices.  For more information on collecting data on mobile devices, check out the video resources on our website.
    2. If you plan on sharing a link or QR code for respondents to use their own devices to collect data, the link is available and the top right of your screen in SurveyAnalytics.  To get to the QR code go to: Distribute > Share > and the QR code is there for you to download and share.  Please note, there are ways to prevent (not completely eliminate) bots.  Check out our newsletter on this topic.
    3. Please be sure to test every feature you plan on using and have data collectors practice using it.  E.g. if you plan on collecting data in Spanish, please test and practice the survey in Spanish.  If you plan on sending the survey link electronically or sharing a QR code, please test the link or QR code.  If you plan on collecting data on a mobile device, please test and practice that way.
  6. Monitor data collection
    1. Check in with your team.  Regroup and retrain as necessary.  
    2. Inspect the data as it comes in.  Ensure data are being uploaded properly.
    3. Aim for a little more than your minimum sample size.
  7. After data collection is completed, you can use SurveyAnalytics to analyze your data.  The data can also be exported and analyzed in Excel or your statistical analysis software of your choice. 
    1. Resources on analyzing quantitative data 
    2. Resources on reporting results
    3. Resources on data visualization

FAQ

  • How do we share our Survey Analytics (SA) account with other members of our team?

  • Each LLA only has one SA account. You’ll need to share the SA username and password with your team members. Keep in mind that only one person can be logged in at a time. We at TCEC have a group chat where we ask each other if anyone is in SA so that we don’t cause any problems with simultaneous logins.

    While only one person at a time can be in the SA account, multiple people/devices can still be used to collect data. This step requires either sharing a link or QR code to take the survey or connecting devices to the appropriate survey using LLA-issued device keys.

  • Should we administer the public opinion survey in both our primary and secondary jurisdictions?

  • It depends! LLAs that chose to complete their observations in both primary and secondary jurisdictions this year should also complete the public opinion survey in both primary and secondary jurisdictions. Furthermore, all the other accompanying activities (policy record review, key informant interview, Midwest Academy Strategy Chart, and other related activities) should also be done on the same timeline.

    If an LLA chose to wait until 2023 to begin work in their secondary communities, then they can also wait until 2023 to administer the public opinion survey for those secondary communities.

  • Do we use the same device keys/passwords from our observation surveys to administer the public opinion survey?

  • No. The TCEC-issued device keys from the observations will NOT work. Each LLA needs to create unique device keys so that the devices are linked to the LLA’s survey. It is best to make sure your LLA has completed all ECTC observations before disconnecting from the TCEC-issued device key that connected the device to the observation. Alternatively, you can use a separate device than you used for the observations. Devices can only be connected to one account at a time. For the public opinion survey, you want to be connected to your LLA’s account.

  • Can we survey non-residents of our County or jurisdiction?  Do we have to purposively select respondents that are impacted by our objective, or can we survey the general public?

  • It depends! Some decision makers only want to hear about what their constituents think.  In other situations, especially where tourism is a big part of the community) it might be helpful to hear from both visitors and residents alike. There is no specific requirement around this, but each LLA should take a moment to decide the intended purpose and audience for this activity, then select respondents based on that. For example, LLAs with a smokefree MUH objective will want to oversample (get more people) that live in MUH. For a smoke-free park objective, it makes more sense to cast a wider net and survey the general public.

  • Are we allowed to conduct the survey both in-person with a mobile device and online through a link?

  • Yes, just be sure to state in your report that different modes were used.

  • Are we allowed to collect responses on paper and then manually enter those into the Survey Pocket?

  • It is possible, but not recommended. This adds an extra step to the work and introduces the potential for data entry error.  It is much more efficient to conduct the survey electronically.  If you choose to collect data on paper, be sure to still use SurveyAnalytics to enter the data.  This way, each entry is as uniform as possible.  

  • How do I enter or edit the Spanish Translation?
  • In collaboration with the Hispanic Coordinating Center, the Spanish translation is already included in your SurveyAnalytics account.  However, any changes made to the English version should also be made in the Spanish version.  In particular, you'll need to fill in the name of the organization your data collectors should represent.  This is stated in the introduction of the survey.  You will also need to update the zip code, city, and/or community in the first few questions.

    First, make any changes in the English version, i.e. adding, deleting, or editing any of the informational text, question text, or answer options.  Make any changes in the main, "Edit" window in your SurveyAnalytics account.  Then edit to the Spanish translation by go to: Languages > Manual Translation > Question & Answer Translation.  The new questions you've created should have a new, blank corresponding box to enter the Spanish translation.  To be clear, you are not creating a whole new survey in Spanish.  The SurveyAnalytics program allows you to collect data in both English and Spanish in the same survey.  

    Here's a video on how to edit the Spanish text.
  • Instead of asking respondents to enter their age, can we modify the question and ask them to select their age range?

  • Yes. We could not come to a consensus about what age categories would be most useful, so the question was left broad. 

  • Are we allowed to remove any sexual identity and gender identity (SOGI) questions if we deem them unnecessary?

  • No. Many of the demographic questions have a hashtag (#) which means the questions need to be included in the survey.  The respondent may choose not to answer the question and should select, “Prefer not to answer.”

  • If we hand the device to the respondent, how do we know if they need the quit information if we can’t go back in to see if they intend to quit tobacco in the next 30 days?

  • It was suggested to offer quit resources to everyone regardless of what the respondent selects in that question.

  • If we say the survey out loud, do we need to read all the following ethnicity questions if they have already identified with one?

  • Another way to collect this data is to ask the question and let the respondent answer.  If their answer fits into one of the existing categories, select that answer, otherwise type in the answer the respondent provided.

  • I don't know if we can complete the surveys by Dec. 31st.  What can we do?

  • Just like with any activity, if you anticipate not being able to finish by the deadline, please communicate that to your team, which also includes your PC.  This way, there are no surprises during the progress report process, and they can try to help brainstorm options to address any challenges.  

    One LLA suggested to, "Just start."  Sometimes we can get bogged down by trying to strategize an make the best plan, but for some communities, you just have to get out there and start.  

    Another LLA was able to send the survey through their HR so that all county employees could be sent the survey.  This was helpful for their county-wide policy work, so it might need to be adjusted for your specific needs.

    Others suggested that with the holidays approaching, it could be a good time to administer surveys at malls and shopping centers where people are likely to be.

    Another option is to use a polling firm that can collect your surveys for you via phone.  Here is a reference from a fellow LLA: KGS Research - https://www.kgsresearch.com, Zeshawn Javed zeshawnj@kgsresearch.com or Rashad Khan rashadk@kgsresearch.com.  This is especially helpful for those with some salary savings due to being short staffed.  Essentially, you can work with your PC and PM to use those salary savings to pay a firm like KGS to collect the data for you.

Additional resources

References

Chung-Hall, J., Fong, G. T., Driezen, P., & Craig, L. (2018). Smokers’ support for tobacco endgame measures in Canada: findings from the 2016 International Tobacco Control Smoking and Vaping Survey. CMAJ open, 6(3), E412.

Middleton, A.; Sanford, A.; Botkin, J.; Coxe, N.; Tran, D.; Aguilar, J.; Villalobos, J.2021. Reducing Availability of Tobacco in Santa Clara County. California: Santa Clara County Public Health Department Tobacco Free Communities Program.

California Department of Public Health, California Tobacco Control Program. Data User Query System (DUQS): Education Attained, 25+ (from U.S. Census Bureau 2015-2019 American Community Survey). Available at https://www.tcspartners.org/Campaigns/DataToolsAndResources/DataQuerySystem.cfm

Wang, M. P., Wang, X., Lam, T. H., Viswanath, K., & Chan, S. S. (2015). The tobacco endgame in Hong Kong: public support for a total ban on tobacco sales. Tobacco control, 24(2), 162-167.

Lykke, M., Pisinger, C., & Glümer, C. (2016). Ready for a goodbye to tobacco?—Assessment of support for endgame strategies on smoking among adults in a Danish regional health survey. Preventive Medicine, 83, 5-10

Documents