What comes to your mind when you think of forest science? Maybe scientists monitoring biodiversity in the forest? Or researchers projecting future scenarios? But what about social science research around forests? In our Martelkom project, we are exploring societal perceptions of forest management and the interactions between foresters and the public. To do this, we decided to conduct a nationwide survey in Germany. This cannot be too complex – right? Or can it?
You might think that creating a survey is quick and easy: just draft some questions on the topics you want to study. Yet, in our studies, we have been told to never underestimate surveys and their complexity as a data collection tool. So, when we were tasked to develop a survey on societal perceptions of forest management, we were eager to dive in. Now that the survey is sent out, we’re still surprised by how much time it took – from crafting the right questions to identifying our target group, refining wording, and selecting the best response types. But let’s break it down step by step.
Advantages of a literature review
We began with a semi-systematic literature review to investigate the current state of our two main interests: societal perceptions and preferences for forests, and communication and interaction between forest stakeholders and the public. A literature review is key to identifying knowledge and research gaps and thus sharpening the focus of a survey study. A semi-systematic review means following clear guidelines to ensure transparency and replicability of the process and results (Snyder et al., 2019). We developed a specific search string of keywords to filter relevant databases like “Web of Science” for relevant scientific papers – and ended up with 1.307 hits!
The keywords we used were: (Forest) AND (Perception OR preference* OR value OR view* OR opinion OR choice* OR attitude* OR expect* OR belief*) AND (aesthetic* OR “scenic beauty” OR structur*) AND (societ* OR citizen* OR civic OR public* OR visit* OR recreation)
Reading all these papers was out of the question, so we applied additional selection criteria such as publication date, language, and empirical study design. After filtering, we were left with about 50 papers for in-depth study. These papers revealed a wide variety of methodologies, diverse definitions of “the public”, and even more diverse insights into societal perceptions and preferences for forests. Equipped with this theoretical foundation, we felt ready for some hands-on research.
Practical marteloscope exercises
We started with practical exercises in the forest to see how certain groups from the public interact with foresters. Using Marteloscopes (special silvicultural training sites), we explored their potential as a communication tool between foresters and the public. What makes these sites unique is the abundance of forest data that participants can access on a tablet, allowing them to navigate their experience in the forest independently. The Marteloscope exercises were throughout a positive, interactive experience for both foresters and participants – not to mention the researchers who got to spend time in the forest. For forest visitors, taking on the perspective of a forester uncovered many previously (unknown) details, like the complexity of forest management decisions, the nuances of wood quality, and the diversity of microhabitats (small biodiversity in the forest like fungai, woodpecker holes etc). On the other hand, the numbers and data in the tablet often contrasted with the rather recreational and emotional perspective of forest visitors – the non-quantifiable “magic of the forest”. But what about the underlying assumptions and expectations the public holds towards the forestry sector?
The process of finding survey questions
Getting back to our goal of exploring societal perceptions and preferences for forests, it was crucial to narrow down our research questions. To align the survey with our experiences from the Marteloscope exercises, we decided to focus on perceptions of multifunctional forest management including communication and interaction between foresters and the public. To avoid reinventing the wheel, we took our initial inspiration from the WaMos survey (WaldMonitoring sozio-kulturell), a socio-cultural forest monitoring project in Switzerland that runs every 10 years. Moreover, we reviewed other forest perception surveys across Europe. We adapted some questions from the WaMos survey and developed new ones tailored to our specific research goals, adding a set of standardised socio-demographic questions to ensure representativeness of our results. We also had to select underlying theoretical concepts to inform the questions and guide the analysis of our findings. In our final version of the questionnaire, each question must be clearly linked to our research goals, concepts and the existing knowledge.
Selecting the final set of survey questions turned out to be quite a process. Should we use single choice, multiple choice, ranking, or matrix with Likert-scales? Each type has its advantages and difficulties, offering different sorts of data and require various methods of analysis. For example, multiple choice gives equal weight to all options, while ranking questions highlight priorities. Matrixes on the other hand can provide detailed insights on preference scales. We adapted and refined our focus step by step – by discussing ideas with colleagues, getting feedback, revisiting the literature, and testing the survey multiple times. This back-and-forth process of refining questions and adapting our focus of the survey design took several months.
But that’s not all – let us have a look at the logistics of the survey.
Why working with a polling company?
Defining our target group was a critical part of the survey work. We aimed to capture a broad range of forest-related perceptions from the public in Germany, rather than focusing on specific demographic groups. Since it was not feasible to survey every person in Germany, we needed a sample that is (statistically) representative of the German population. A representative sample would allow us to generalize the results to a broader context – in this case, the perceptions of the general German public. We decided to gather 2.000 responses, ensuring representativeness across age, gender, education level, and location (diverse German states). Of course, this method cannot perfectly capture the full diversity of the population, but we aimed to get as close as possible.
To achieve this, we contracted a professional polling company. These companies use diverse online panels of voluntary respondents, most often for market research purposes, to ensure the desired sample is reached.
Next steps
A pretest, conducted by the polling company, will provide us with a small sample of responses, allowing us to refine our methods before collecting the full database. Once the data is in, we will check it for quality and representativeness before diving into analysis. We are planning to publish a research article based on the survey results, and we hope that our findings will contribute to a communication strategy that fosters ongoing dialogue between foresters and the public – even after the Martelkom project.
Curious to learn more? Stay updated here: Martelkom – WaldWerte diskutieren
Literature: Snyder, H. (2019). Literature review as a research methodology: An overview and guidelines. Journal of Business Research, 104, 333–339.
