Agriculture Reference
In-Depth Information
6.7.5 Gender of household head
There is a positive and significant relationship between the probability of cattle sales and
gender of head of household. As the household head change from being male to being
female, the probability of selling cattle increases. This result is not in line with the a priori
expectations of the probability of males selling being more than that of females. Females are
normally involved in many household activities and most of them in the small scale farming
sector are not employed hence they do not have any other source of income. Upon the
death of the spouse, a widow will in most cases be forced to sell cattle in order to meet the
household needs and since women are involved in many household activities adding other
activities related to cattle production and marketing such as cattle herding and taking cattle
to auctions would be difficult. This therefore forces the new household head (the wife:
female) to sell cattle. This probably explains the relationship between gender of household
head and the probability of selling cattle.
6.8 Variation of market opportunities across municipalities
The Pearson Chi-square value for market opportunities across municipalities was less
than 0.05, meaning that opportunities significantly varied across municipalities (Table
6.1). This is in line with a priori expectations that municipalities emerged from different
administrations, with differential support services to farmers, and thus might lead to
differences in market opportunities. According to the Eastern Cape Development
Corporation (ECDC) (2003), the Chris Hani District has a broader base in agriculture
with limited agro-processing industries than Amatole. This therefore results in variations
in market opportunities in these two municipalities.
6.9 Evaluation of model performance
The Hosmer and Lemeshow Goodness of Fit test statistic is 0.794 and it indicates that
the logistic model is a fairly good fit. If the Hosmer and Lemeshow Goodness of Fit test
statistic is 0.05 or less, we reject the null hypothesis that there is no difference between the
observed and predicted values of the dependent variable; if it is greater, we fail to reject the
null hypothesis that there is no difference, implying that the model's estimates fit the data
at an acceptable level. While the model obviously does not explain much of the variations
in the dependent variable, it however does so to a significant degree.
6.10 Conclusion
Education, market distance, body condition and herd size are significantly associated
with municipality while age of head of household, information access, market availability,
transport availability and body condition of cattle significantly affect the probability of
Search WWH ::




Custom Search