Influence of institutional factors on smallholder farmers’ marketing channel choices - Institutional Constraints to Small Farmer Development in Southern Africa - page 90

from production areas to market places. The emerging and smallholder farmers in the Kat
River Valley lack proper marketing infrastructure (Magni, 1999). For instance, vegetable
farmers opt to sell from their homes because they do not have marketing sheds and proper
storage facilities.

5.5 The methodology

The multinomial logistic regression model was used to test the institutional and technical
factors that influence households from using greater depth marketing methods, which have
the potential of increasing their incomes. According to Matungul et al. (2002), the greater
the depth in marketing methods used by households, the greater the expected income.

Multinomial logistic regression can be used to predict a dependent variable, based on
continuous and/or categorical independent variables, where the dependent variable takes
more than two forms (Hill et al. , 2001). Furthermore, it is used to determine the percent
of variance in the dependent variable explained by the independent variables and to rank
the relative importance of independent variables. Logistic regression does not assume linear
relationship between the dependent variable and independent variables, but requires that
the independent variables be linearly related to the logit of the dependent variable (Gujarati,
1992). Pundo and Fraser (2006) explained that the model allows for the interpretation of
the logit weights for the variables in the same way as in linear regression.

The model has been chosen because it allows one to analyse data where participants are
faced with more than two choices. In this study, smallholder farmers are faced with three
choices, which are; formal market participation, informal market participation and non-
market participation. Firstly, the farming households are assumed to decide whether
to market their products or not. When they choose to market, they then decide on the
marketing channel to be used (either formal markets or informal markets). These decisions
are made based on the option, which maximizes their utility, subject to institutional and
technical constraints.

As such, the utility maximizing function can be given as:

Max U = U (C k , R k , R ik ; H u ) (1)

Where:
Max U denotes the maximum utility that can be attained from agricultural production;
C k represents the consumption of produced goods by the household;
R k represents revenue gained from formal market participation;
R ik represents revenue gained from informal market participation;
H u represents a set of institutional factors shifting the utility function.