Tanzania Journal of Agricultural Sciences 2015, Vol. 13(1) : 26-40
Njombe is one of the districts in Tanzania that suffers from the problem of poor adoption of
recommended maize varieties. The study was conducted to investigate the influence of independent
and intervening variables in determining the adoption behaviour of recommended maize varieties
among maize growers in Njombe District of Tanzania. The study employed a cross sectional research
design where data were collected from 113 respondents randomly selected to represent other maize
growers from four villages namely, Kibena, Ulembwe, Uwemba and Igagala. The findings show
that each investigated intervening variable has influence on adoption of recommended maize
varieties as expected. They explain 86.6 percent of the variation in adoption (R2 =0.866, p=0.000).
In particular, the NT (Beta = 0.659, p=0.000) and the efficiency misperception (Beta = -0.232,
p=0.008) that make the biggest contribution. On the other hand, most of the independent variables
investigated have no influence except for the level of education and area used to grow maize. The
total contribution of independent variable towards explaining the variance in adoption is only 18.7
percent. This is reflected in the significant R2 of 0.187, which implies that the total influence of
intervening variables highly explains the influence compared to that of independent variables. This
indicates that the intervening variables are the best predictors of the adoption behaviour and the
influence of independent variables is manifested in the adoption behaviour through the intervening
variables as postulated in Duvels (1991) model of behaviour determinants. This calls for further
testing of the model in different social cultural settings and crops to see its relevance in determining
the adoption behaviour.