Customising and Distributing The Collaborator Survey

Research Instrument Design

Surveys

Behavioural Science

Step-by-step guide to using the create_survey_url() function to correctly customise your survey links for distribution.

Author

Rose Megirian

Introduction

Every experimental design or data analysis output delivered as part of a Service & Support collaboration should be accompanied by an opportunity for collaborators to provide feedback via the AAGI Collaborator Survey. To ensure only relevant questions are shown and that key contextual information is captured alongside responses, each survey link must be customised before sharing - this means selecting details specific to the output and recipient, such as the type of support provided, the AAGI node, and the collaborator’s organisation type, which are then encoded directly into the link.

The create_survey_url() function makes this straightforward. Run it once per output delivery, and it will generate the correct link, print a summary of your selections for you to verify, and copy the URL to your clipboard ready to paste into your delivery email.

Installation and Use

Step 1: Enable the AAGI R-Universe

To install the {AAGISurvey} package containing the create_survey_url() function, you first need to enable the AAGI R Universe in your R session. You only need to do this once. If you have already enabled the AAGI R Universe for another package, you can skip steps 1 & 2 and jump straight to Step 3. Load the package.

Because the AAGI R Universe URL is hosted within a private GitHub group, access to the AAGI Aus GitHub is required to obtain it. If you have the appropriate access, the URL can be found on the AAGI-Aus GitHub page.

Once you have the URL, you can enable the universe using:

# example syntax for enabling the AAGI R-Universe
options(repos = c(   
  aagi_aus = "🦐!!!REPLACE_WITH_AAGI_R_UNIVERSE_URL!!!🦐",   
  CRAN = "https://cloud.r-project.org" 
  ))

Step 2: Install the package

install.packages("AAGISurvey")

Step 3: Load the package

library(AAGISurvey)

Once installed, you only need library(AAGISurvey) in future sessions.

Step 4. Run the `create_survey_url()` function

The create_survey_url() can be run in one of two ways:

Scripted – provide all required arguments directly in your R script.
Interactive – run the function with no arguments and follow the prompts in the console.

Scripted example:

library(AAGISurvey)

# Create a survey URL for design support for small plot 
# trial for a government agency carried out by AAGI Cu

url <- create_survey_url( 
  support_type = "S_D", 
  design_type = "D_SP", 
  analysis_type = ,
  aagi_node = "CU", 
  organisation_type = "O_GRO" 
  )

Interactive example

library(AAGISurvey)

url <- create_survey_url()

Both approaches will:

Generate a Qualtrics survey URL containing the correct metadata
Print a summary of your selections
Copy the URL automatically to your clipboard, ready for easy pasting into an email or other message.

Check the printed summary carefully to confirm all details match the recipient and output.

Accessing Full Function Documentation

For detailed information about all function arguments, including valid coded values for the scripted example, you can view the full documentation in R:

?create_survey_url   

# or  

help("create_survey_url")

Step 5. Include the Customised Survey Link in Your Output Delivery Email

Include the survey link in the same email that delivers your output. Research consistently shows that survey response rates and data quality are meaningfully affected by how and when a survey is distributed (Dillman et al. 2014). Sending the survey alongside the output rather than in a separate follow-up addresses several of these factors directly: the email comes from someone the recipient already has a working relationship with, which research shows meaningfully increases response rates (Ericson 2023; Millar et al. 2021); the request is contextually tied to a specific piece of work rather than arriving as a generic follow-up; and collaborators can respond while the experience is fresh without being asked to recall details later. Reducing the number of separate communications also lowers the friction that is well-documented as a contributor to survey non-response (Dillman et al. 2014).

When adding the link to your email, make sure to:

hyperLink your customised survey URL to the “Go to survey” button.
Include the full survey URL as plain text below the button so that, if the button does not render correctly in the recipient’s email client, it can be copied and pasted directly into a browser.

Template Text to include in your output delivery email

Note

Feedback from our collaborators helps us identify patterns, differences, and emerging insights across the contexts and groups we work with - informing how we prioritise and adapt our work over time to remain relevant and responsive to your needs.

To make providing feedback easy, this email includes a link to a short, anonymous survey about this output and your collaboration experience. Completing the survey takes less than five minutes.

Click HERE to go to survey

(Alternatively, copy and paste this URL into your browser: http://YOUR_CUSTOMISED_SURVEY_URL.com)

Thank you for taking the time to share your thoughts.

Step 6. Save a copy of the email for project records

After sending the email, save a copy in the ‘Correspondence’ folder of the associated collaboration project. This ensures a record is maintained for future reference and accountability.

Further Information

About the Survey

The AAGI Collaborator Survey is designed to help us better understand the impact our Service & Support activities are having within the Australian grains research, development and extension (RD&E) sector. It targets researchers we work with directly through Service & Support collaborations, gathering structured feedback on experimental design and data analysis outputs and the collaboration experience surrounding them. The goal is not only to understand where and how our support is effective, but to identify opportunities to strengthen how we work with collaborators over time.

About the Survey Design

The survey was designed with data quality as the primary consideration, though what quality means for this kind of data differs in important ways from how it is defined for objective or directly measured quantities, and is worth briefly unpacking.

Surveys collect information through participants’ responses, which means they capture what people report, recall, perceive, or judge. This information is inherently shaped by individual experience, interpretation, and context. This is not a weakness. It is the defining characteristic of the data surveys are built to capture, and one of the primary reasons surveys exist: to provide an efficient, scalable way to access information that cannot be measured directly with instruments. Things like experiences, perceptions, judgements, and interpretations. For this kind of data, quality is not defined by statistical precision or replicability in the instrumental sense. It is defined by how well responses reflect participants’ genuine views and experiences given the questions asked and the context in which they were answered.

Understanding this shapes every design decision. Minimising objective burden (the measurable effort imposed by a survey’s length, structure, and demands) is a baseline obligation of good research practice. Asking people to invest time and effort unnecessarily is poor design and, more fundamentally, poor ethics. But research suggests that it is perceived burden, rather than objective burden, that is the more decisive determinant of survey response (data) quality (Kunz and Gummer 2024). Perceived burden reflects each respondent’s subjective experience of the survey, how difficult, effortful, or worthwhile they find it, and consistently predicts response quality across a range of indicators including item completion, attention, and response depth (Kunz and Gummer 2024).

Perceived burden is not determined by effort alone. It reflects the relationship between perceived effort and perceived value, and research consistently identifies motivation, trust, and topic salience as interacting factors that shape whether people participate, complete a survey, and invest in the quality of their responses (Krosnick 1991; Groves et al. 2000). A survey that feels relevant, that comes from an organisation perceived as credible and responsive, and where participation feels like it will amount to something, is experienced as less burdensome regardless of its length.

Critically, perceived burden is shaped largely before a respondent encounters the survey itself, by prior experience, reputation, and whether past participation has felt worthwhile (Bradburn 1978; Krosnick 1991). This means the levers for managing it extend well beyond survey design and into how we operate as an organisation over time. It is why what we do with the data matters as much as how we collect it. Failing to act on responses is not only a missed opportunity, it is an ethical breach of the implicit agreement that research participation represents. And in the communities we work with in the grains sector, where networks are tight and reputations travel, it has consequences that extend beyond a single survey to shape how future participation, and future researchers, are received.

For those interested, these principles and the behavioural and methodological evidence behind them are explored in depth in Behaviourally-Informed Survey Design: A Methodological Guide.

Closing the Loop

Because the value of this survey depends as much on what we do with the data as on how it is collected, it is worth thinking beyond distribution. How findings are shared with collaborators, and whether they can see their input reflected in any changes or communications that follow, directly affects how worthwhile participating feels and by extension the quality and consistency of responses over time.

This is an evolving process and there is room to develop it collectively. If you have ideas about how findings might be acted on, how insights could be communicated back to collaborators in meaningful ways, or suggestions that could improve the structure, wording, or workflow of the survey itself, please share them by emailing CBADA@curtin.edu.au with the subject line AAGI Collaborator Survey.

References

Bradburn, Norman. 1978. “Respondent Burden.” Proceedings of the Survey Research Methods Section of the American Statistical Association, 35–40.

Dillman, Don A., Jolene D. Smyth, and Leah Melani Christian. 2014. Internet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method. 4th ed. John Wiley & Sons.

Ericson, Bonick, A. 2023. “Optimizing Survey Response Rates in Graduate Medical Education Research Studies.” Family Medicine 55 (5): 304–10. https://doi.org/https://doi.org/10.22454/FamMed.2023.750371.

Groves, Robert M., Eleanor Singer, and Amy Corning. 2000. “Leverage-Saliency Theory of Survey Participation: Description and an Illustration.” Public Opinion Quarterly 64 (3): 299–308. https://doi.org/10.1086/317990.

Krosnick, Jon A. 1991. “Response Strategies for Coping with the Cognitive Demands of Attitude Measures in Surveys.” Applied Cognitive Psychology 5 (3): 213–36. https://doi.org/10.1002/acp.2350050305.

Kunz, Tanja, and Tobias Gummer. 2024. “Effects of Objective and Perceived Burden on Response Quality in Web Surveys.” International Journal of Social Research Methodology 27 (3): 385–95. https://doi.org/10.1080/13645579.2024.2393795.

Millar, Morgan M., Hilary A. Hewes, Andrea L. Genovesi, et al. 2021. “The Effect of the Familiarity of a Survey Sender on Response Outcomes in a Large-Scale Survey of Emergency Medical Services Agencies.” Evaluation & the Health Professions 44 (3): 260–67. https://doi.org/10.1177/01632787211030635.