Description: The South African Agricultural Business Innovation Survey describes the patterns of innovation in South African agribusinesses from 2016 to 2018, analysed at an aggregate level for all agribusinesses, and compares trends across three major sub-sectors: agriculture (including crop, animal, and mixed farming), forestry, and fisheries. Abstract: The South African Agricultural Business Innovation Survey 2016-18 examines how commercial agriculture, forestry, and fisheries businesses functioned in the face of enormous and intricate challenges; presenting an in-depth, contextualized profile of innovation patterns in the fisheries and forestry sub-sectors as an example of the type of analysis that policy actors in government, business, universities, and civil society organizations might find useful. Moreover, it suggests how stakeholders, particularly those in government, can use the data to inform agricultural and innovation policy.

Email survey

Web-based self-completion

Businesses are categorised according to the Agriculture Standard Industrial Classification (SIC) codes for economic activities.

Innovation surveys are based on a random stratified sample of business enterprises from the national business register (or equivalent) and results are extrapolated to the original population. The South African Agricultural Business Innovation Survey 2016-18 was based on a random stratified sample obtained from Statistics South Africa (Stats SA) which contained 1 690 businesses. A process of sample cleaning identified 364 businesses as invalid. In particular, these were businesses that were: not identifiable or traceable through several methods, duplicates, or inactive businesses. Invalid businesses were excluded from the original sample, resulting in a final survey sample of 1 326 businesses. In a difficult business climate, despite implementing an extensive advocacy strategy prior to and as part of the fieldwork, 303 businesses responded to the survey. On this basis, the survey achieved an overall response rate of 22%.

Additionally, a simple random sample non-response survey was conducted, as recommended by the Oslo Manual (OECD, 2005) for surveys that achieve response rates of less than 70%. The purpose of the non-response survey was to correct for any bias that might arise due to businesses that did not respond to the survey being less or more innovative than those businesses that did respond. The non-response survey covered 117 or 15% of the businesses that did not respond to the main survey, and 87 firms completed the non-response survey. This gives an overall response rate of 74.3% for the non-response survey. The correction for bias due to non-response was implemented by adjusting the probability weights used to project the sample results to the target population of businesses. This methodology also adjusts the weights for invalid businesses (businesses that were found to have merged or been liquidated).

The results from the survey were then projected to the target population of South African businesses. A great deal of effort was made during fieldwork to ensure that at least one response was received per stratum, to allow for weighting.

Note-: The sampling frame and design is explained in deeper details in the following related document - AgriBIS 2016-18 Agricultural Business Innovation Survey Sampling Report 2017