Fish Consumption and Accessibility and the Implications for Household Nutrition and Food Security in Ghana and Tanzania

Akua Sakyiabea Akuffo, Purdue University


This dissertation empirically investigates aquaculture adoption on food security, fish demand and fish accessibility in Ghana and Tanzania. There are three separate chapters, each discussing a specific issue relating to food security and fish consumption. Data for chapters two and three are from the round 6 of the Ghana Living Standards Survey (GLSS6) while chapter four uses the Household Budget Survey (HBS) for Tanzania. The second chapter of this dissertation applies the Propensity Score Matching (PSM) approach in logit and probit frameworks to a cross section of 4,011 Ghanaian households, to analyze how fish farming is associated with household food security. The results show that on the margin, the probability of adopting fish farming increases with wealth, location, ecological zone and household size but decreases with household income per capita. The study also observes from the average treatment on the treated (ATT) that, on average, fish farming households have food security scores 15.5 points higher than non-fish farming households. The results also suggest that fish farming households have higher food diversity and frequency of food consumed than the non-fish farming households through direct consumption. Post-estimation analysis indicates that female-headed households have equal probability of being food secure relative to their male counterparts if they adopt fish farming. Chapter Three uses a latent class model of structural heterogeneity in a linear regression framework to examine the importance of location and ethnic group affiliation on fish demand in 2,185 households in Ghana. The results suggest that Ghanaian households fall into two classes concerning seafood demand, which I refer to as ‘Traditional’ and ‘Non-Traditional’ households. For Traditional households, fish and poultry are complementary goods and fish and red meat is substitutes. For Non-Traditional households, fish and poultry are complementary goods and fish, red meat and pork are substitutes. Price is a major concern for consumers in the rural and peri-urban areas, who tend to be more Traditional while taste, diversity: health and nutrition concerns pertain to urban consumers, who tend to be more Non-Traditional. Traditional households are identified as being Akan Christians and located in the forest and savannah areas while Non-Traditional households are identified as a mixture of Akan, Ewe, and Dagomba. Religion does not affect fish consumption by Non-Traditional households, located in mainly the savannah areas. Chapter Four relies on simple linear regression to evaluate the correlation between electricity, communication networks and transportation on access to seafood in 1,730 Tanzanian household. Two measures of seafood accessibility are used, namely: the Fish Access Count (FAC) and the Fish Accessibility Index (FAI). Results showed that access to electricity, communication, and transportation are positively correlated to seafood accessibility by Tanzanian households. Access to transportation is a primary component of access to seafood in urban areas, access to electricity improves access to seafood in rural and peri-urban areas while access to communication is associated with improved access to seafood in rural and urban households.




Shively, Purdue University.

Subject Area

African Studies|Home economics|Agricultural economics

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server