Abstract:
The paper describes difficulties encountered in the use of cross-sectional annual enterprise records, as a primary source of data, for the estimation of physical response functions for mushroom production in Northern Ireland.
The main difficulties identified were: (1) mathematical inter-dependence between variables arising from the use of aggregated data; (2) observational error related to reliability of the recorded data; (3) determination of the relative importance of increased depth and density of the compost as contributory factors to the response relationship between yield and quantity of compost per unit area cropped.
Additionally, the existence of problems due to heterogeneity in the data arising from differing levels of managerial ability is recognised.
A linear relationship was determined between yield and the quantity of compost per unit area cropped, also between yield and the quantity of spawn per unit area cropped.
In view of the comparability of the basic form of empirical relationships estimated from farm record data to that obtained from controlled experimental work over a similar range of input values, the adequacy of simple crop and livestock response theory as a framework within which to interpret the results is questioned.
The importance of the system of production on yield was established.
It is considered that the difficulties identified do not invalidate the value of the relationships estimated in providing a benchmark for further research aimed at determining multivariate response function(s). The paper recommends that future work in estimating response relationships for mushroom production in Northern Ireland should be of an interdisciplinary nature based upon individual crop data.
|