Survey data on voluntary nature conservation commitments of German businesses and their perceptions towards conservation credits

To preserve biodiversity and ecosystem services, company engagement is crucial. However, available data on manager views and perceptions regarding nature conservation in particular is rare. The presented survey data gives insights into current levels and forms of business commitments for nature conservation. The data contributes to understanding business attitudes towards voluntary conservation action and includes information about factors that influence their engagement. Moreover, the data informs about manager perceptions towards the concept of nature conservation credits and, as such, allows for an evaluation of a certified biodiversity and ecosystem services market. Importantly, the dataset contains essential company characteristics to put responses into greater context. The scope of the survey is limited to German companies from secondary and tertiary sectors. Companies were sampled through proportional stratified random sampling based on size and location. The data was collected through a self-administered online-survey, conducted in 2019. The database comprises responses of 747 companies that logged into the online system. The survey data were in part analysed through structural equation modelling for an investigation of factors that drive voluntary conservation commitments [1]. Related to this analysis, a subset of 618 companies is available that provided sufficiently completed questionnaires. Both datasets, i.e. the raw data as well as the first subset used for analysis, are hosted in the public repository Open Research Data of the Leibniz Centre for Agricultural Landscape Research (ZALF), Germany. The repository also stores all coding information as well as the questionnaire: https://www.doi.org/10.4228/ZALF.DK.149. The dataset can be used, for example, by researchers from the field of environmental business management and strategy.


a b s t r a c t
To preserve biodiversity and ecosystem services, company engagement is crucial. However, available data on manager views and perceptions regarding nature conservation in particular is rare. The presented survey data gives insights into current levels and forms of business commitments for nature conservation. The data contributes to understanding business attitudes towards voluntary conservation action and includes information about factors that influence their engagement. Moreover, the data informs about manager perceptions towards the concept of nature conservation credits and, as such, allows for an evaluation of a certified biodiversity and ecosystem services market. Importantly, the dataset contains essential company characteristics to put responses into greater context. The scope of the survey is limited to German companies from secondary and tertiary sectors. Companies were sampled through proportional stratified random sampling based on size and location. The data was collected through a self-administered online-survey, conducted in 2019. The database comprises responses of 747 companies that logged into the online system. The survey data were in part analysed through structural equation modelling for an investigation of factors that drive voluntary conservation commitments [1]. Related to this analysis, a subset of 618 companies is available that provided sufficiently completed questionnaires. Both datasets, i.e. the raw data as well as the first subset used for analysis, are hosted in the public repository Open Research Data of the Leibniz Centre for Agricultural Landscape Research (ZALF), Germany. The repository also stores all coding information as well as the questionnaire: https://www.doi.org/10.4228/ZALF.DK.149 . The dataset can be used, for example, by researchers from the field of environmental business management and strategy. ©

Value of the Data
• The dataset is, to the knowledge of the authors, the first data source that specifically informs about German business perceptions and actions for nature conservation. The survey focuses on targeted actions for public environmental goods and, as such, makes a valuable contribution to the environmental management literature.
• The data is valuable to those interested in understanding business responses to the loss of biodiversity and ecosystem services. Policy-makers benefit from information about factors that encourage companies to voluntarily commit to nature protection. The data is also relevant for environmental organisations and practitioners with an interest in offering certified biodiversity and ecosystem services, in support of their fundraising effort s f or conservation projects. • The data might be used to compare the results of similar studies, which focus on different countries and/ or certain business sectors or company types. Moreover, the data can be used by researchers that investigate, for instance, organisational decision-making, environmental business strategies, stakeholder management or methodological aspects of company surveys. • The data contains information from diverse company types and, therefore, reflects the multifaceted German business landscape. In addition, the survey captured responses from companies that were already engaged for nature conservation as well as those that were not. • With a sample size of around 700 participants, it is a large dataset suitable for a wide range of statistical analysis methods.  Table 2 -4 of this data paper. The survey data contains information about the voluntary nature conservation commitments of German businesses in secondary and tertiary sectors. It also includes respondents' perceptions about the concept of nature conservation credits and their willingness to pay for biodiversity and ecosystem services.

Data Description
The complete dataset ("dataset_complete ") comprises 136 variables with mostly numerical codings, except text entries in 12 columns. In total, the dataset includes observations from 747 businesses that logged into the online system. The logins were tracked by the online system, based on individual participation codes that were assigned to each company address. The corresponding codebook ("dataset_complete_codebook ") provides coding information and related questions in German and English. Generally, binary and dummy variables were coded 1 = "yes" and 0 = "no". Missing values were coded as 999,999 = "no answer" or 977,777 = "not applicable" due to preceding filter functions.
The questionnaire ("Questionnaire_DE ") shows the design of the implemented online survey. It is comprised of four main parts: 1) information about the company, 2) information about its nature conservation commitment as well as aspects influencing (non-)commitment, 3) perceptions about the concept of nature conservation credits, and 4) preferences for nature conservation credits.
A part of the survey data ("subset_1 ) has been analysed through structural equation modelling to investigate factors that influence companies' voluntary nature conservation commitments [1] . The subset consists of 59 variables and 618 observations. For inclusion in the analysis, the respondents needed to answer a question about whether or not their company has been engaged for the protection of biological diversity and natural habitats (cf. variable "engaged "). Respondents that gave no answer to this question or stated "I don't know" were excluded. The subset further differs from the complete dataset in that some variables were renamed and recoded. All information about coding, new and previous variable names as well as related questions in German and English are found in the codebook ("subset_1_codebook ").
The repository also provides the postal letters used to invite the companies to participate in the survey ( "Invitation Letter_A", "Invitation Letter_B" and "Invitation Letter_reminder" ). The files provide the template of the one-page invitation letters as well as the envelopes. The documents are provided in German.
Moreover, a secondary data table is provided, which contains information about the geographical and size distribution of the target population ( "target_population_2017 ). The data was .csv -

Invitation Letter_A
Postal invitation letter template. Invitation letter A was initially sent to 10 0 0 companies. Because it was not as successful as " invitation letter_B " in activating participation, it was not used for the main despatch.

Invitation Letter_B
Postal invitation letter template. Invitation letter B was initially sent to 10 0 0 companies. Because it was more successful than " invitation letter_A " in activating participation, it was used for the main despatch.

Invitation Letter_reminder
Postal reminder letter template, which equalled the design of " invitation letter_B ".
.pdf -target_population_2017 Secondary data retrieved from the database "Destatis" of the German Federal Statistical Office, reflecting the target population of the survey.

target_population_2017_ retrieval
Information about how the target population data was retrieved from the Destatis database.
.txt Table 2 Economic sectors and their inclusion in the target population of the survey.
in data paper Table 3 Distribution of the target population according to size and geographical regions, based on secondary data from Destatis.
in data paper Table 4 Distribution of the sample frame according to size and geographical regions, based on available company addresses from the survey agency meap GmbH.
in data paper retrieved from "Destatis", the open source database of the German Federal Statistical Office. The table includes information based on latest available data from 2017. A detailed step-by-step guide about how the file was retrieved from the Destatis website is included as a text file in the repository ( "target_population_2017_retrieval" ). Finally, Table 1 further includes information that is provided in the subsequent tables of this data paper.

Target population
The target population for the business survey included companies that fulfilled the following requirements: 1) Companies with at least 50 employees: We defined this criterion because we presume that any voluntary engagement requires a certain level of flexibility and possibility in terms of finances and staff capacities, which might not be given for smaller entities. The precise threshold of 50 + employees is commonly used in official EU-or German statistical agencies to define medium-sized companies. Thus, the survey targeted medium-sized (50-249 employees) and large-sized companies (250 + employees). 2) Limitation to company headquarters: Many companies have multiple locations or run individually managed production sites. However, nature conservation engagement is often part of a wider corporate strategy [2] , which tends to be worked out in headquarters rather than minor company locations. To avoid duplicates as well as contacting subordinate company locations without appropriate contact persons, we focused on company headquarters. 3) No companies from the primary sector: The survey targeted companies from diverse business sectors, except those of the agriculture, forestry, fishery or mining sector, cf. Table 2 . We excluded the primary sector because due to their landscape activities nature conservation concerns are inherent to their business. The survey focused on voluntary nature conservation commitments beyond obvious operational good practice. In addition, we excluded three economic sectors from the target population as they were not suitable for our topic of interest as shown in Table 2 . These target population specifications resulted in a total population size of around 79,0 0 0 German companies, according to the German Federal Statistical Office [4] . Based on the latest available data from 2017, around 80% of the target population were medium-sized and around 20% were large-sized companies, cf. Table 3 .

Sample frame and sampling
To gain access to company addresses of our target population we worked with the survey agency meap GmbH. This agency had access to various company databases from its cooperation with address brokers through which company contacts could be rented for the survey. From the available addresses, the survey agency created the sample frame according to the above specified criteria, cf. Table 4 . In total, the agency had around 68,0 0 0 addresses available that matched  the specifications of our target population. The reason that the target population and sample frame differ slightly is that not all companies which were recorded by the Federal Statistical Office were also available in the databases of address brokers. In addition, the information in the Destatis business registry referred to the year 2017, whereas the address databases might have been more or less up-to-date. Based on the sample frame, proportional stratified random sampling was conducted. The strata were defined according to 1) the federal states, as well as 2) the business size, differentiated by the number of employees. In total, 17,0 0 0 company contacts were sampled and invited to participate in the survey.

Questionnaire design
The questionnaire content was based on aspects that were found to be relevant for corporate conservation commitments, according to a preceding qualitative study [2] . Therefore, the questionnaire sought to expand and validate the findings of the qualitative study (ibid.). All in all, the questionnaire consisted of 23 to 30 questions, depending on responses to filter questions, and included mainly closed as well as some open questions. Numerous closed questions were measured on a 5-point Likert scale, which typically included an "I don't know" answer option. The survey required an estimated 15 to 20 min to complete. To ensure high quality, we invited 11 company representatives from various sectors, such as technology consultancy and tourism, to participate in a pre-test. Regarding the questionnaire design we also received consultancy from survey experts of the GESIS Leibniz-Institute for the Social Sciences in Germany.

Data collection
The data was collected through a self-administered online survey between August and December 2019. The online system was optimised for various mobile devices and web browsers. We invited our sampled contacts using postal invitation letters because e-mail invitations often go unread, are quickly deleted or are filtered out to the spam folder [ 5 , 6 ]. One-page invitation letters addressed the respondents personally. In large companies with 250 + employees we primarily contacted the head of marketing and communication. If no such contact person was available in the database we addressed a member of the executive board or first management level. In medium-sized companies with 50-249 employees, we primarily addressed the executive manager or else a member of the first management level. We made this distinction because we presumed that executive managers in medium-sized companies tend to be involved in most decisions themselves, whereas in large or even very large corporations a lot of decision-making powers rest within the specialised departments. The available database from the survey agency meap GmbH did not include contact persons from CSR departments or environmental management, so we could not target employees in those positions.
The invitation letters contained a brief URL link and a company-specific participation code required for logging in to the survey system. They also included information and logos of the affiliated research institute as well as the sponsoring agencies, i.e. two federal ministries and one federal agency. Moreover, we assured confidentiality of the data as well as anonymity of the respondents in both the invitation letter and the login page. The letters as well as the online survey provided the name and e-mail address of the study leader. To encourage participation, we offered respondents to indicate their interest in receiving a survey results report from the finalised study, at the end of the questionnaire.
We initially designed two different versions of the invitation letter to compare the activation success: One design used colour-print, electronic signatures and friendly wording (cf. Invitation Letter_A ), whereas the other design used a more bureaucratic style without colour-printing or signatures (cf. Invitation Letter_B ). A despatch of both versions to 10 0 0 companies showed a higher response success from the bureaucratically-styled letters. Therefore this design was used for contacting the remaining 15,0 0 0 companies, which was done mid-October 2019. At the end of November, a reminder letter (cf. Invitation Letter_reminder ) was sent out to 70 0 0 randomly selected companies that had not yet participated in the survey until then. Even though reminders are an effective way to increase response rates [7] , one or numerous reminders to all non-respondents was not feasible due to budget reasons. Moreover, the identity of the companies and contact persons were not known to the research team and rested undisclosed with the survey agency so that follow-ups through e-mail or telephone were not possible.
From 17,0 0 0 postal invitations, 553 did not receive the letters due to wrong or outdated address information, bankruptcy or unavailable contact persons, e.g. due to retirement or job change. However, it is likely that the number of unavailable contact persons was higher than what has been reported back to the study team. A total of 747 participants (4.54%) logged into the online survey, cf. dataset_complete . A first data analysis with a subset of the survey responses [1] was done with 618 questionnaires (3.76%), cf. subset_1 .

Ethics Statement
We confirm that the participants in the survey have given their informed consent to participate. Participants were ensured that their participation is voluntary and that they can leave or pause the online survey at any point. It was guaranteed that any information provided is treated confidentially and anonymously.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships which have or could be perceived to have influenced the work reported in this article.