Livestock Research for Rural Development 20 (8) 2008 Guide for preparation of papers LRRD News

Citation of this paper

Impact of improved chicken production on intra-household decision-making among fisherfolks in the Kainji Lake Basin, Nigeria

A O Lawal* and O A Adekunle

* UK Department for International Development (DFID) Nigeria, British High Commission, 10 Bobo Street, Maitama Abuja, Nigeria
a-lawal@dfid.gov.uk

Abstract

Two household decision outcomes were measured namely, the probability that a wife would take decisions on managing her own extra income and income from the sale of chickens, using the Logit model in a study carried out among 120 fisherfolks in four fishing villages on the shorelines of the Kainji Lake, Nigeria.

 

Results show that in about 43% of the households, wives control their own extra income while about 29% said it was jointly controlled.  In about the same proportion (28%) control is solely by the husbands. 

 

Younger and better-educated wives, wives with more marketing experience before marriage, and those with older husbands are more likely to take decisions on managing extra income by themselves.  Although positively related to decision making processes by wives, keeping improved chickens is not significant.  However, the expected probability points to a potential towards increase in sole decision by wives on managing income by her with an increase in the keeping of improved chickens.

Keywords: freshwater fisheries, gender, logit, poultry, socio-economic


Introduction

The Nigerian-German Kainji Lake Fisheries Promotion Project (KLFPP) in 1996 started an “Improved poultry keeping programme” at the National Institute for Freshwater Fisheries Research (NIFFR). The rationale of the Programme was to compensate for possible short-term income loss from fisheries due to the implementation of fisheries management measures restricting the use of the Lake’s resources and to boost alternative income sources especially for the fishermen’s wives (Ayeni and Mdaihli 1997).

 

The strategy chosen was aimed at upgrading in both adult sizes and egg production of local domestic chickens through cross breeding of the slow growing and less productive local hens with the fast growing, improved stock of commercial cockerels of the Black Haco breed (Ayorinde 1997).  Between 1997 and 2001, the KLFPP raised and distributed about 5000 improved cocks to more than 100 communities along the shorelines of the lake. 

 

The objective of socioeconomic impact analysis is to assess the magnitude and distribution of both direct and indirect effects (Grootaert 2002), and measure the impact of projects on both individual and community-level outcomes (Field and Kremer 2005).  These impacts are changes that have occurred for an individual fisherfolk at household or farm level or in the community at large as a result of the adoption of improved chickens. At the household level, important impact indicators include among others, fisherfolks’ income and income distribution, intra-household gender relations, as well as allocation and control of resources.

 

It has been observed that in the African households, men and women and often children, separately control productive resources, take part in independent decisions, manage personal income, assume different responsibilities and favour different investments (Sanginga 1998).

 

The objective of this paper is to examine the status of women in decision making and participation in economic activity as a result of adoption of improved chicken production by fisherfolk households.  Two outcomes were examined, namely the probability that a wife would take decisions on managing her own extra income and income from the sale of chickens.  The questions on decision making are phrased in terms of who makes certain decisions within the household, with the following choices: husband alone, both spouses jointly, and wife alone.

 

Materials and methods 

Data source

 

The study was carried out in four fishing communities; two from each of the eastern and western shorelines of the Kainji Lake where improved chickens were distributed between 1998 and 2001.  A two-stage random sampling technique was used to select a sample of 120 fisherfolks, comprising 60 fisherfolks from each area.  The first stage of the sampling involved the stratification of the lake into western and eastern shorelines to select two communities each, and the second stage of the sampling involved selection of respondents.  A random sample of 30 was drawn from the lists of fisherfolks in benefiting communities. 

 

The Kainji Lake, resulting from the damming of the River Niger is the largest artificial lake in Nigeria and is well known for fisheries products.  Lake Kainji is situated between latitudes 9o 50' - 10o 57' North and longitudes 4o  25' - 4o 45' East. The lake was impounded on 2nd August 1968 and it is 136.8 km in length and 24.1 km maximum width. Its surface area has been variously quoted as approximately 1,300 km2.  At full volume, the water is at the altitude 142 m and at low volume the water is at the 133 m level (Abiodun 2003).

 

The study took place in March 2007 and data collection involved a combination of household schedules, focus group discussions (FGDs), and field observations.  The data was analysed using the Statistical Package for Social Sciences (SPSS 1994)

 

Empirical model specification

 

The approach used is to estimate logistic regression on the whether wife of the household takes some decisions alone, relative to omitted categories.  This approach is similar to that taken by Adato et al (2000) in their analysis of the determinants of intra-household decision making in PROGRESA communities.

 

The dependent variables specified in the Logit equations is the fisherman’s wife decision making power in the household, measured as dummies.  In the first equation, it takes a value of 1, if wife spends own extra income and 0, otherwise. In the second equation, it takes the value 1, if wife controls money derived from sale of chickens and 0, otherwise.  The explanatory variables are defined as follows:

 

AGEH, WIFEAGE – age of husband and wife in years

LABOUR – time devoted to caring for chickens, 1 if husband spends more time, 2 if husband and wife equally, 3 if wife spends more time, 4 if husband spends no time

EDUH, WIFEEDU – number of years of formal schooling by fisherfolk

BUYINPUT – purchase of inputs (feeds, vaccines, drugs) in chicken production, 1 if husband, 2 if husband more than wife, 3 if husband and wife equally, 4 if wife more than husband and 5 if wife

WIFEOCCU – wife’s occupation before marriage

IMPROVE – dummy defined for improved chicken production and takes a value of 1 if household is currently keeping improved chickens and 0, otherwise

 

Empirical results and discussion

 

Generally, the likelihood that a woman would spend her own extra income was good while the likelihood that a woman would control income from the sale of chickens was fair; an average woman had about 60% predicted probability of controlling her own income while an average woman had about 33% predicted probability of controlling income from the sale of chickens in the household.

 

Table 1 shows that three variables were significant in determining the probability that a woman would control own extra income.  These are age of the husband (+), age of wife (-) and educational qualification of the wife (+). 


Table 1.  Parameter estimates for the Logit regression 1 (woman controls own extra income)

Variables

B

S.E.

df

Sig.

Exp(B)

Age of husband

.185

.080

1

.021*

1.203

Age of wife

-.189

.085

1

.027*

.828

Educational qualification of husband

1.210

.798

1

.129

3.354

Educational qualification of wife

1.564

.821

1

.057**

4.778

Wife’s occupation before marriage

-.132

.188

1

.484

.877

Time devoted to caring of chickens

-.012

.396

1

.976

.988

Purchase of inputs

.435

.377

1

.249

1.545

Improved chicken production

.344

.478

1

.471

1.411

Constant

-7.688

2.708

1

.005

.000

*, ** indicates that the variable is significant at 5% and 10% levels respectively


Table 2 however shows that only wife’s occupation before marriage (+) was significant in determining the probability that a woman would control money from the sale of chickens in the household. 


Table 2.  Parameter estimates for the Logit regression 2 (woman controls money from the sale of chickens

Variables

B

S.E.

df

Sig.

Exp(B)

Age of husband

.108

.081

1

.182

1.114

Age of wife

-.075

.085

1

.379

.928

Educational qualification of husband

1.031

.684

1

.132

2.803

Educational qualification of wife

-.741

.703

1

.292

.476

Wife’s occupation before marriage

-.594

.207

1

.004*

.552

Time devoted to caring of chickens

-.064

.415

1

.877

.938

Purchase of inputs

-.090

.373

1

.810

.914

Improved chicken production

.289

.511

1

.571

1.336

Constant

-2.101

2.204

1

.341

.122

*, indicates that the variable is significant at 1% level


Overall, in about 43% of the households, women decide on how to spend control their own extra income while about 29% said it was jointly controlled.  In about the same proportion (28%) control is solely by the husbands.  Mean age for husbands was 44.78 years while the mean for the women was 31.73 years.  The signs on the estimate indicates that an increase in the age of the husband would significantly increase the probability that a woman would control own extra income.  However the negative sign attached to the estimate for women indicates that an increase in the age of a woman would significantly decrease the probability that she would control her own extra income.  This scenario suggests that older men would tend to support their wives to control their own extra income whereas younger women would tend to control their own extra income. 

 

The frequency distribution of women’s occupation before marriage showed that farming was about 19%, fishing 28%, hawking and petty trading 51%, while only 2% were full-time students.  All these occupations apart from being a full-time student, involves monetary transactions and experience with marketing.  FGDs with women revealed that young girls involved in fishing and farming also market products for their parents.  This would thus explain the relationship between the wife’s occupation before marriage and probability of controlling money from the sale of chickens.  More importantly, the FGDs also revealed that chicken production is much more a women’s activity, than men’s.  The household survey results also showed that in 59% of the households, women spend more time attending chickens than men, compared with 33% of the households, where men and women spend about the same time.  In 8% of the households, men do not spend any time at all attending chickens.

 

The variable defined as keeping of improved chickens although positive, was not significantly related to women’s decision making processes.  However, the expected probability that keeping improved chicken would impact on decision making is 41.1% (calculated as 100 x (Exp (B) – 1).  This means that with a unit increase in the number of improved chicken, the odds that a woman would control own extra income is 41.1%.  In the same vein, the odds that a woman would control income from sale of chickens, is 33.6%.

 

The strategy chosen by the KLFPP was aimed at upgrading in both adult sizes and egg production of local domestic chickens through cross breeding of the slow growing and less productive local hens with the fast growing, improved stock of commercial cockerels (Ayorinde 1997).  Between 1997 and 2001, the KLFPP raised and distributed about 5000 improved cocks to more than 100 communities along the shorelines of the lake.  Although not all fisherfolks in all the communities benefited from this distribution, it was however expected that non beneficiaries would see the benefits of keeping improved chickens and eventually adopt its production.

 

Survey results show that in addition to beneficiaries of the project, a number of fisherfolks bought their own improved cockerels and introduced these into their flocks.  Overall, 54% of the households are currently keeping flocks that contain at least one improved chicken species, while 46% were keeping flocks all of which were of the local chicken species.  Table 3 below shows the mean flock size, income from sale of chickens, number of chickens consumed/ given out as gifts and number that died in the last one year, by households keeping or not keeping improved chickens.


Table 3.  Means of variables in household chicken production

Variable

Households not keeping improved chickens

Households keeping improved chickens

Mean flock size

7.13a

20.17b

Mean annual income

N 1,598.18a

N 6,245.38b

Mean number consumed or given out as gifts in the past year

2.6a

5.77b

Mean number that died in the past year

11.53a

8.62a

a, b row means with different alphabets are significantly different


It can be observed that mean flock size, income and number consumed or given out as gifts are significantly higher in households keeping, than in households not keeping, improved chickens.  Similarly, the mean number of chickens that have died in households keeping improved chickens is less than for households not keeping, although not significant.

 

Conclusion 

The above regression analysis for two decision making outcomes has identified a number of variables that appear to be significant determinants of decision making within households — and, by extension, are correlated with women’s bargaining power.   

 

Wife’s education and age of husband are positively related, while her own age is negatively related, to making decisions solely on controlling her extra income.  Similarly, wife’s marketing experience prior to marriage is positively associated with her making decisions solely on controlling income from the sale of chickens.  Thus a woman is more likely to decide on her own how to spend her extra income if she is younger, if her husband is older, and if she is better educated.  She is also more likely to decide on her own how to spend income from the sale of chickens if she has more marketing experience before marriage.

 

The keeping improved chickens, by itself, though positive, have an insignificant effect on decision making patterns.  What is more interesting is the expected probability which points to an increase in sole decision-making by wives with regard to control of extra income and income from the sale of chickens.  In conclusion therefore, the existing decision making patterns and potential for change, is consistent with KLFPP’s focus on boosting alternative income sources especially for women, and by extension, the potential of empowering women to participate more fully in household decision making.

 

References 

Abiodun J A 2003 Evaluation of Fisheries Catch Trend on Lake Kainji, in Nigeria 1995-2001.  Journal of Applied Sciences and Environmental Management 7 (2): 9 – 13

 

Adato M, de la Brière B, Mindek D and Quisumbing A 2000 The impact of PROGRESA on women’s status and intrahousehold relations.  Final Report submitted to PROGRESA. Washington, DC: International Food Policy Research Institute. Mimeo

 

Ayeni J S O and Mdaihli M 1997 Identification of non-fishing income opportunities around Kainji Lake.  A Consultancy report commissioned by the Nigerian-German (GTZ) Kainji Lake Fisheries Promotion, New Bussa (68pp). 

 

Ayorinde K L 1997 Preliminary impact assessment study of cockerel exchange programme in fishing villages in the Kainji basin.  A report submitted to the Nigerian-German Kainji Lake fisheries promotion project.  Department of Animal Production, University of Ilorin, Nigeria

 

Field E and Kremer M 2005 Impact Evaluation for Slum Upgrading Interventions. Retrieved July 25, 2006 from http://siteresources.worldbank.org/INTURBANPOVERTY/Resources/kremer_field.pdf

 

Grootaert C 2002 Socioeconomic impact assessment of rural roads: methodologies and questionnaires.  Retrieved July 25, 2006 from http://siteresources.worldbank.org/INTISPMA/Resources/383704-1153333441931/11274_Cgrootaert-Impact-Rural_Roads].pdf

 

Sanginga P C 1998  Adoption and Social Impact Assessment of improved agricultural technologies: the case of Soybean in Benue State, Nigeria.  PhD. Thesis, University of Ibadan, Ibadan, Nigeria

 

SPSS 1994 Statistical Package for Social Sciences, SPSS Base 6.1 for Windows User’s Guide. SPSS Inc., Chicago, IL.



Received 7 May 2008; Accepted 2 June 2008; Published 5 August 2008

Go to top