On the basis of a literature review, a questionnaire-based survey was conducted with representatives from 18 agricultural ministries of various countries at the Global Forum for Food and Agriculture 2014 in Berlin, where 65 agricultural ministers participated. Moreover, the final communiqué of the Berlin Agriculture Ministers’ Summit 2014 was analyzed, following the steps of Mayring’s qualitative content analysis [10, 11].
Questionnaire-based survey at the Global Forum for Food and Agriculture
The survey was conducted at the Global Forum for Food and Agriculture (GFFA) in Berlin from 16–18 January 2014. The GFFA is an international conference that is held during International Green Week (IGW) and is organized by the Federal Ministry of Food and Agriculture (BMEL) in cooperation with GFFA Berlin e.V., the Senate of Berlin, and Messe Berlin GmbH each year. In 2014, it took place for the sixth time. Representatives from politics, business, science, and civil society used the forum to meet, share ideas, and enhance political understanding on central issues in the future. The 2014 topic was “Empowering Agriculture: Fostering Resilience—Securing Food and Nutrition” . During the first 2 days, there were expert panel discussions on aspects of the main topic. On the second day, four working sessions were organized by the international organizations FAO, EU-Commission, World Bank, and United Nations Environment Programme (UNEP). The third day was the most important one; here, the International GFFA Panel Discussion was held, and in the afternoon, the Berlin Agriculture Ministers’ Summit took place (Fig. 2).
The GFFA was chosen because it is the worlds largest conference of agricultural ministers, with 65 ministers in 2014. Additionally, the working session of the FAO focused on “food systems for better nutrition”, in which the participants discussed nutrition–agriculture linkages. The results of this and the other three working sessions were included in a final communiqué. The communiqué is the outcome of the summit and represents the common position of the participating ministers. They committed to incorporating it into the ongoing international discussions on agricultural policy such as ICN2 and the post-2015 process .
The target group of the survey were the ministers and representatives of the agricultural ministries participating at the GFFA. A standardized questionnaire was presented directly to them for self-administration. Participants gave informed consent before taking part. Ethical approval was not needed because social research, not medical, was conducted.
Content and design of the questionnaire
Three main items were of primary interest for the survey: first, the prominence of the concept, if countries are already aware of it or have implemented it; second, what their understanding of the concept is; third, the added value of the concept was examined by looking at its aim, its usefulness with regard to nutrition security, and its relationship to the right to food.
The questionnaire (see Additional file 1) was designed based on a literature review where papers published in journals and publications of international organizations were included. Peer-reviewed papers were searched primarily in the databases PubMed, CAB Abstracts, and Web of Science. The webpages of international organizations relevant to the field of agriculture as well as food and nutrition security were searched for nutrition-sensitive agriculture-related publications. Important sources were the webpages of the Food and Agriculture Organization of the United Nations (FAO), International Food Policy Research Institute (IFPRI), the World Bank, and the United Nations Standing Committee on Nutrition (UNSCN). Searches were mainly conducted between November 2013 and February 2014. The keywords used were nutrition-sensitive/nutrition-enhancing/nutrition-smart/nutrition-focused agriculture, promoting/improving nutrition through agriculture, linking agriculture and nutrition/health, nutrition-agriculture linkage, and agriculture for nutrition. To minimize the risk of missing relevant papers, reference lists of key studies were reviewed for additional relevant studies. Only the most recent literature that had been published between 2008 and 2014 was included. Papers were excluded that did not refer to agriculture as a nutrition-sensitive sector or that focused only on a single problem, e.g., aflatoxin and stunting. Only publications in English were used for further analysis.
There was no commonly agreed upon definition for nutrition-sensitive agriculture found in the literature. Moreover, no conceptual framework could be found that explicitly described what nutrition-sensitive agriculture looks like. Therefore, four definitions and six aims found in the literature were cited directly or in a modified fashion [13–17]. Most reviewed publications (90 %) referred to food security, in particular to diversification (82.5 %), less so to health (57.5 %), and rarely to care (37.5 %). There was no agreement in the literature about the most appropriate foods and production systems. Nevertheless, sustainable forms of agriculture were frequently mentioned. The most important cross-cutting issues were gender equity, nutrition education for all actors along the value chain, and collaboration between agricultural, nutritional, and health professionals [2, 7, 8, 13–50].
In the questionnaire, cross-sectoral factors and elements that should be included in the concept of nutrition-sensitive agriculture for nutrition security, food security, health, and care were surveyed. Furthermore, the representatives should specify which foods in their opinion should be the focus of nutrition-sensitive agriculture and which production system would be most appropriate for it. For each question, the issues were cited that were mentioned in the reviewed publications. The modified UNICEF model served as the basic structure for the arrangement of the questions. The questions had to be answered quickly and easily, since ministry representatives do not have much time to fill out questionnaires. Therefore, it consisted only of 18 closed questions covering five pages.
Students of nutritional science and other academic backgrounds examined the questionnaire for clarity and assessed its duration. Normally, about 15 min were needed to complete the questionnaire. After including the students’ comments, which were mainly on understanding and wording of the questions, the questionnaire was given a second time to employees of Division 622—World Food Affairs of the German Ministry of Food and Agriculture. One important issue that was raised was the distinction between food security and nutrition security found in two questions (questions 10 and 11). These two terms were not considered to be accepted worldwide as different terms, and comprehension of these was not clear. Definitions were given before they appeared in questions, and they were arranged according to the modified UNICEF model, starting with the comprehensive term nutrition security and followed by the three main aspects food security, care, and health. Another aspect was the right to food. To secure a common understanding of the right to food, the definition was given before the question.
Dissemination of the questionnaire
The questionnaire was disseminated at the GFFA to as many representatives of the delegations as possible. The procedure was twofold. First, questionnaires were placed in the business lounge where only the delegations had access and at relevant expert panel discussions and working meetings, namely, the kick-off event, the expert panels on “Opportunities for better nutrition along agricultural value chains: building on learning from rice in Bangladesh” and on “Horticultural contributions to the fight against malnutrition”, and the working session of the FAO about “Food systems for better nutrition”. Second, the study was presented by the author and the study team to the delegations. They asked the delegates to answer the questionnaire on site and send it back via e-mail or to give their contact data in order to receive the questionnaire via e-mail at a later date. At the GFFA, 100 questionnaires were printed and disseminated in hardcopy. This number was chosen based on estimates that about one third would get lost by delegates that took two copies or by not-targeted visitors of the events. On the last day of the summit, the strategy of presenting the study and collecting contact data was more strongly pursued.
As a follow-up, the questionnaire was sent to the collected contact addresses via e-mail. Before sending, it was converted into Word format with text boxes, check boxes, and drop-down lists in order to allow electronic administration. Delegations that had no embassy in Berlin received support at the GFFA through volunteers from the BMEL. These were asked to inform the delegations about the research project and distribute the questionnaire to each representative. After the GFFA, the BMEL staff members who were in charge of the delegations were reminded of the study and asked to send the questionnaire via e-mail again. As a second strategy, other networks were used, e.g., personal contacts to representatives of European Member States were asked to answer the questionnaire. Moreover, the German agricultural attachés in the countries were asked to informally disseminate the questionnaire to their contacts in the agricultural ministries. The first deadline was the 31 January 2014. This was later pushed back to the end of February 2014. Delegates who had pledged to return it but had not up to this point were asked again. By the end of March, data collection was concluded.
The data of the questionnaire were coded, entered into a Microsoft Excel spreadsheet, and converted into a pivot table. A descriptive analysis was made, frequencies were shown, and sub groups analyzed with cross tables . The data were finally transferred into SPSS 11.5 for further cross tabulations, statistical tests, and cluster analysis. Statistical significance tests were only made in some cases to confirm suggestions. The p value was defined as the probability of the observed data, given that the null hypothesis H
0 is true. This contains very limited information. The rejection of a null hypothesis is no basis for estimating the probability that in a replication of the research the null hypothesis will also be rejected . Therefore together with the p value, the effect size phi was reported in order to improve interpretation of the collected data. Effect sizes do not take sample size into account and are therefore independent of the number of participants . Phi is similar to the correlation coefficient; in fact, if both variables are dichotomous, Pearson’s r is equal to phi. Correlations between 0.2 and 0.5 were interpreted as medium sized according to the rules for social science data analysis of Kühnel and Krebs . The cluster analysis was conducted using Ward’s method in order to see whether specific groups could be found. Ward’s method was used because it clusters the results according to the smallest variance within a cluster.
The question about which foods should be in the primary and secondary focus of nutrition-sensitive agriculture was analyzed in two ways: The questionnaires that were filled out as intended were analyzed together. Frequencies were given to all the answer possibilities as a whole and divided by focus. A secondary analysis was made for the questionnaires that were not filled out as intended. In an initial step, we examined food options that were not considered at all. Then, the other foods were classified as primary or secondary according to the most frequently given answer.
Analysis of the Final Communiqué of the GFFA 2014
The Final Communiqué of the GFFA 2014 was analyzed as an indication of the political will and confirmation to implement the concept of nutrition-sensitive agriculture. The first draft of the communiqué was developed by a group of experts from the BMEL in a consolidated process. After completing an initial draft, it was sent to four international organizations, FAO, EU-Commission, UNEP, and World Bank, for further changes and comments. After these were included, the registered Ministries of Agriculture added their comments, which were also included in the final draft and approved by the German Minister of Food and Agriculture. At the Berlin Agriculture Ministers’ Summit, the process was opened again for a 4-h discussion. After reaching a consensus on the communiqué, all 65 participating ministers approved and signed it (own observations). The final communiqué was chosen because the results of the working sessions of the international organizations should be considered in the declaration. The working sessions of FAO concentrated on “food systems for better health”. The session discussed the issues of how to maximize the contribution of agriculture and food systems in order to improve nutrition.
For the analysis, the English version of the Final Communiqué of the GFFA 2014 was used by following the steps of the qualitative content analysis of Mayring (see Fig. 3). The qualitative content analysis according to Mayring is a mixed methods approach that actually contains both qualitative and quantitative analytical steps. The central idea of qualitative content analysis is to start from the methodological basis of quantitative content analysis (analysis of category frequencies) and then conceptualize the process of assigning categories to text passages as a qualitative–interpretive act, following content–analytical rules. Therefore, it follows a step-by-step model for the research process. Because analysis and interpretation is based on content–analytical rules that are defined in advance, reliability is increased [10, 11]. This method is appropriate for analyzing the communiqué because it is a method for investigating statements that are made consciously and explicitly. Moreover, through this technique, the text is interpreted within the communicative context. Both are important aspects for analyzing a political statement with a clear research question.
The focus of the analysis of the communiqué was the subject of the text. The main research question of the content analysis was whether there is political commitment to implement nutrition-sensitive agriculture. Therefore, three sub-questions were defined:
Is the topic nutrition-sensitive agriculture included in the communiqué?
Which elements of the reviewed literature are included?
What importance does nutrition-sensitive agriculture has in the communiqué?
Specific aspects regarding nutrition-sensitive agriculture were filtered out of the communiqué, and their importance was analyzed by looking at the frequency and structure of the communiqué (structuring concerning the contents and the form). Categories were defined deductively and inductively based on the modified UNICEF model and the reviewed literature: food security, health, care, and cross-cutting issues (e.g., gender, sustainability, education, and collaboration, as well as environmental, economic, political, and social contexts). The unit of analysis was defined: one word was the smallest unit (coding unit), and the whole document was the biggest unit (context unit). We first searched for keywords in the document, then, the framework and the three main aspects of the communiqué were analyzed, and finally, the whole document was examined. The analysis was made following the described procedure while reconsidering the categories. A category was broadened if aspects seemed strongly related to those described in the reviewed literature and yet were not reflected accordingly in the existing categories. The categories were mainly reconsidered while holistically analyzing the framework and the three main aspects of the communiqué.