Annex 4: Post-harvest consumption analysis of sweet potato in Kenya - Survey

Executive summary
Introduction
Purpose and justification
Methodology
Results
Household consumption survey
Summary and conclusion
References

Report by Dr. J. D. Alumira and C. M. Obara

Executive summary

The paper reviews secondary information available on the post harvest analysis of sweet potato in Kenya and subsequent consumption patterns. It also gives the results of a survey by a semi-structured questionnaire, of mainly middle-income Nairobi dwellers on their consumption patterns of sweet potatoes and cassava, relative to the main staple, maize. Factors considered include tribal preferences, intra-household food distribution, influence of rural-urban migration and the preferences of sweet potato in relation to cassava. The findings show that most of the secondary information available is not specific to sweet-potatoes but groups all root-crops (cassava, yams, arrow-roots, sweet-potatoes) together as roots or in other instances refers to all potatoes both sweet and Irish as just potatoes. The supplementary study on sweet potato consumption in Nairobi confirmed that maize is the main staple food followed by beans; sweet potato was highly preferred to cassava as a staple, was consumed by all household members although women and children ate considerably more; was mossy consumed in the boiled form and eaten for breakfast or as a snack. Both the middle and low income respondents consumed sweet potatoes with similar frequency, at similar meals and preferred them in boiled form. That they also faced similar constraints particularly the rising retail prices and changing eating habits due to urban livelihoods. The study was however not able to capture the affluent in society because of limited resources particularly tune.

By Alumira J. D. PhD. and Obara C. M. MSc.

Introduction

Off-farm post-production issues have become critical and the relationship between off-farm and on-farm operations requires to be carefully examined within the wider socio-economic and political context. This, in the wake of liberalisation, is expected to enable the farmers to find new outlets for products they used to sell to the government or other multi-functional organisations. A clear perspective is also expected to be set on the lowering of the transaction costs of food supply and on raising the added value that is gained by the farming population. A prototype framework for the analysis of post-production sector has recently been developed by the Prevention of Food Losses Programme of FAO (Flach, 1995). This frame-work is basic and requires to be fine-tuned to the specific country situation. The first country study for Kenya focused on the post-harvest analysis of root and tuber crops, mainly potato and sweet potato in Meru and Kisii districts; the results of which are yet to be disseminated. This study supplements the earlier one undertaken by a team of consultants both local and international and led by tire University of Hannover, for GTZ; and whose aim was to apply a systems perspective which includes the whole set of operations between harvesting and consumption (storage, processing, handling, transport and marketing), to the analysis of potato and sweet-potato. It is on this background that this study has been born.

Purpose and justification

This study was commissioned by GTZ to address the consumption characteristics of sweet potato in the Kenya, particularly the relationship in demand for sweet potatoes and other food crops. The purpose is to gain an understanding of the different types of consumption patterns of sweet-potato in Kenya. Terms of reference were several:

· Tribal preferences and population of tribes together with their main staples. Breakdown for Kenya and Nairobi

· Urban-rural differences in consumption; also to what extent traditional rural consumption patterns have been modified by urban living.

· Class divisions. How does income status affect consumption patterns of staples especially sweet potato? Breakdown of income status for Nairobi

· Differences within families (male, female and children) in sweet potato consumption

· How sweet-potato is consumed in urban centres. Has there been an expansion in the last years in sweet potato demand?

· Compare the position and preference of sweet-potato with that of cassava

Methodology

A combination of both secondary information and administration of semi-structured interviews was used. A total of 35 mainly middle to low income respondents were interviewed in Nairobi. The sample population consisted of agricultural and Veterinary officers, people in banking profession, housewives, secretaries, clerks and laboratory technologists. A copy of the questionnaire is attached.

Results

Population of Kenyan tribes is as shown in Table 1; whereas Table 2 shows the population of Nairobi in relation to that of Kisii and Meru districts where the other study was conducted. Tribal preferences in terms of staple diets do not vary significantly since the single most important food crop in Kenya is maize (RoK, National Development Plan, 1997-2001). In fact famine in Kenya is associated with an absence of maize, even in the presence of other cereals like wheat, rice, millet/sorghum (UNICEF, 1992). According to the same UNICEF document the second most important staple in the country is beans; others that are depended upon in the absence of these two main staples include starchy roots and tubers (cassava/sweet-potato). These root/tubers are no longer consumed in amounts as high as pre-colonial times because eating habits and preferences have changed in favour of maize and cash crops. Differences do however occur in times of food stress where different options are taken by different tribes so as to cope with the hungry period, especially in incidences of drought and inter-seasonal food deficits. In such times, roots and tubers become handy as hunger crops for some tribes whereas others turn to sorghums and millet. Production figures as shown in Table 3 give a substantial difference between the production levels of the main staple and other relatively significant crops.

Table 1: Population Statistics by Ethnic Group in Kenya, 1989

Ethnic Group	Population
Kikuyu	4,448,302
Luhya	3,083,273
Luo	2,653,936
Kalenjin	2,458,123
Kamba	2,448,302
Kisii	1,318,409
Meru	1,087,778
Mijikenda	1,007,371
Maasai	377,089
Turkana	283,750
Others*	1,844,159
Total	21,443,636

Source: RoK, 1989 - Kenya Population Census (1989) Vol. 1... latest available census results.

"*" Others comprise smaller ethnic groups which include the Mbere, Tharaka, Kuria, Bajun, Boni-Sanje, Pokomo, Taita, Taveta, Swahili Shirazi, Basuba, Dorobo, El-Molo, Njemps, Samburu, Dasnachi-Shangil, Orma, Rendille, Sakuye, Ajuran, Degodia, Goshs, Gurreh, Hawiyah, Ogaden, Somali-so-Stated, Kenya Asian, Kenyan European, Kenyan Arabs, Tanzanians, Ugandans, Indians, Pakistanis, British Other Asians, Other Europeans, and Other Arabs

Table 2: Distribution Of Population By District for Nairobi, Meru and Kisii

District	Population
Nairobi	1,324,570
Meru	1,144,594
Kisii	1,137,054

Source: RoK, Population Census, 1989 Vol. 1 p. 1-1

Table 2 also shows that the population of Nairobi Meru and Kisii is more or less the same. What is different is the composition in terms of ethnic groups represented.

Table 3: Estimated Production of Selected Food Crops in Kenya, 1993/94 - 1995/96 (Million Bags)

	1993/94	1994/95	1995/96*
Maize	23.40	17.73	23.83
Beans	2.39	1.25	2.82
Potatoes	2.26	1.99	2.51
Sorghum	0.95	0.86	1.05
Millet	0.45	0.39	0.47

Source: RoK, National Development Plan, 1997-2001 p. 55, '*' = the 1996 value is provisional

Table 3 shows that potatoes generally hold position three relative to maize and beans as main staples, giving a high potential for potato production and consumption in Kenya Acquisition and consumption of sweet potatoes is constrained by several factors one of which is poverty. According to the 1992 Welfare Monitoring and Evaluation Survey (WMES), the level of absolute poverty in rural areas was 46.4% while in urban areas it was 29.3%; and that 57% of household heads who had no education lived below the absolute poverty line. The 1994 Welfare Monitoring Survey II (WMSII) did not indicate significant shifts in these figures. The rural poor are mainly subsistence farmers, the landless and the pastoralists whose incidence of poverty in 1992 was 52%, 50% and 43% respectively. (RoK, National development Plan, 1997-2001). This has resulted in a consistent increase in rural-urban migration as shown in Table 4. As urban population increases, rural population decreases and is estimated to rise to 23.44% of the projected population by the year 2000. Since a high proportion of rural emigrants to Nairobi are poor, they settle in slum areas.

Table 4: Distribution of Population Between Rural and Urban, 1990-2000 (Million)

	1990	1995	2000
Urban	4.07	5.28	7.44
	(18)	(19.2)	(23.44)
Rural	19.43	22.4	24.36
	(82)	(80.6)	(76.6)
Total	23.7	27.5	31.8

( ) denotes percentage.

Source: RoK National Development Plan, 1997-2001 p. 75

In 1980, Nairobi's slum population was reportedly growing at the rate of 25-30% a year compared to the city's population growth rate of 6-8%. At the end of 1980, poverty levels in Nairobi rose to 30% (UNICEF, 1990 in Mwangi, 1995). Kenya if therefore faced with a challenge of providing enough food at all times for good health for the urban dwellers. Food marketing facilities and channels must hence grow faster than the population especially for the low-income slum dwellers, who spend a very high proportion of their income on food (FAO, 1985 in Mwangi, 1995). Mwangi (1995) whose study was to assess the importance of urban agriculture in household food security for the urban poor also attempted to determine food consumption patterns in terms of food types and sources for low income households in Nairobi. The study covered three population clusters in two slum areas of Nairobi namely Korogocho and in Kariobangi North sub-Location; and the combined villages of Kitui in Pumwani Sub-Location and Kanuku in Eastleigh Sub-Location which were involved in the Undugu Society Urban Agriculture Project (USUAP). Food consumption results are as shown in Table 6 whereas Tables 5A and 5B show monthly income by study group and the mean monthly income for the population aged 15 years and above for selected districts by gender.

Table 5A: Monthly Income By Study Group (Kshs.)

Korogocho Farmers	Korogocho Non-Farmers	USUAP Farmers
Income Kshs/Month
2500	31 (65%)	34 (55%)
> 2500	17 (35%)	28 (10%)
Total	(100%)	67 (100%)

Source: Mwangi A.M, 1995 p. 70

Most of these study population earned less than KShs 2,500 (approx. US$ 50) per month.

Table 5B: Mean Monthly Income (KShs.) For Population Aged 15 Years and Above For Selected Districts

	Male	Female
Nairobi	8,425.9	2,443.5
Meru	2,969.0	564.5
Kisii	1,548.8	1,145.8

Source: GoK, 1994 -Welfare Monitoring Survey II, p. 127

Maize in the form of maize meal was the most consumed cereal while starchy roots, which should include sweet potato, were the second most consumed Table 7 shows the main food ingredients consumed by the study group within a household re-call period of seven days. Again this reiterates the heavy reliance on maize as a staple and dry beans and sweet potato being preferred as the second and third important food crops relied upon, respectively, relative to sweet potato. This study showed that consumption preference of sweet potato among the urban poor was low relative to the other major staples but higher when compared with that of cassava and arrow root.

Table 6: Per Capita Food Consumed (grams per Week) by Food Group and Study Group

	Korogocho Farmers (N=48)	Korogocho Non-Farmers (N=67)	USUAP Farmers (N = 62)
Cereals & Cereal Products	1800	1600	1200
Starchy Roots	600	400	800
legumes & Nuts	400	200	300
Meats, Fish & Eggs	200	200	100
milk	400	200	500
Sugar	200	200	200
Fats & Oils	80	60	90
vegetables	300	400	500
Fruits	200	100	200
Others	6	-	1
Total	4186	3360	3891

Source: Mwangi A.M, 1995 p. 31

Table 7: Main Food Consumed (Grams/Household) By The Study Group

		Korogocho Farmers (N=48)		Korogocho Non-Farmers (N=67)		USUAP Farmers N=62
		% HH	Amount (g)	% HH	Amount (g)	% HH	Amount (g)
Maize:
	Meal	98	5319	98	5904	100	4777
	Dry	58	1503	38	555	61	1423
	Green	67	2742	36	723	37	654
Dry beans		81	2115	56	649	85	1766
Irish Potatoes		58	2148	54	1533	76	3543
Sweet Potatoes		33	933	15	237	35	1176
Arrow root		15	459	14	204	29	562
Cassava		-	-	4	47	3	44
Sorghum/Millet		27	273	21	264	42	660

Source: Mwangi A.M, 1995 p. 66.

Information on the consumption status of sweet potato in the country is not only scanty but also non-specific in most cases. Instead, recent research groups root-crops (sweet potato, cassava, arrow root, yams etc.) together. The WMSII of 1994 valued the production, consumption and sale of selected crops, including root crops in Kenya Shillings. This is as indicated in Tables 8, 9 and 10 respectively.

In Nairobi, root crops were valued at about 14.9% of the value of maize; 28.9% that of beans and 8.6% of the production value of all food crops as shown in Table 8.

Table 8: Value of Crop Production By Province (Kshs. Million)

Province	Maize	Beans	Sorghum	Millet	Root Crops	Total Food crops
Nairobi	24.9	12.8	-	-	3.7	43.0
Eastern	3,624.1	3,933.5	151.1	319.4	768.4	9.714.3
Nyanza	8,195.7	1,274.3	7,196.0	249.5	1,122.8	19,929.8
Central	4,230.5	1,377.7	13.1	-	3,919.6	16,760.3

Source: Computed from the GoK (1994), Welfare Monitoring Survey II p. 189

However, consumption values as shown in Table 9 attribute the value of root crops in Nairobi to 3.5% of the total value of consumption of all food crops; 6.9% of that of maize and 10.8% of the consumption value of beans. In Eastern, Nyanza and Central Provinces, both production and consumption values are much higher hence pointing to a need for post-harvest analysis of this crop.

Table 9: Value of Consumption by Crop By Province (Kshs. Million)

Province	Maize	Beans	Sorghum	Millet	Root Crops	Total Food crops
Nairobi	14.4	9.3	-	-	1.0	24.8
Eastern	2,748.8	2,957.5	107.5	182.2	524.2	6,872.0
Nyanza	6,069.3	592.8	1,919.2	136.4	599.7	9,793.0

Source: Computed from the GoK (1994), Welfare Monitoring Survey II p. 191

According to Tables 8 - 10, the lowest production and sale values for root crops are found in Nairobi. This is expected because Nairobi is a major urban area in the country. Interestingly are the very low levels of root-crop consumption values in this city compared to Eastern and Nyanza provinces which are predominantly rural; and where household incomes are relatively much lower. This suggests a negative correlation between income status and root crop production, consumption and sale. As income levels rise, root crop consumption decreases. There is need to investigate this further. Again Tables 8 to 10 show the importance of maize, beans, sorghum and millet in comparison with that of root crops.

Table 10: Value of Sale of Crop By Province (Kshs. Million)

Province	Maize	Beans	Sorghum	Millet	Root Crops	Total Food crops
Nairobi	6.3	3.5.	-	-	2.7	14.1
Eastern	766.8	893.2	41.2	107.3	241.4	2,614.4
Nyanza	1,874.9	619.0	155.2	104.8	348.1	4,504.8

Source: Computed from the GoK (1994), Welfare Monitoring Survey II p. 190

Tables 11-13 (see Appendix) give a profile of various crop and their contribution to household welfare. Table 11 shows the contribution of various crops to household cash income through sales. Again this is higher in the rural districts of Meru and Kisii than in Nairobi and in the rural areas than on the urban Nationally, cash income from sale of maize is highest in comparison to other crops. Table 12 shows the mean monthly expenditure on major staples per province. The districts included in the Table and falling under their respective provinces are those with significant production of roots and tubers. The major staple considered here is maize relative to root crops. The mean monthly household expenditure on food is highest in the urban centres compared to rural areas and in Nairobi compared to Kisii and Meru district. This is as shown in Table 13. This Table also shows the household expenditure on purchase of root crops and that this is slightly higher in urban than in rural areas.

Table 14: Hectarage and Production Statistics for Sweet Potatoes and Cassava by Province and Year

Province	Crop	Year
		1991		1992		1993		1994		1995
		Ha.	Prod.	Ha.	Prod.	Ha.	Prod.	Ha.	Prod.	Ha.	Prod.
Central	S. Potato	1824	12055	2021	17030	3573	29920	2195	16863	1573	11312
	Cassava	946	6099	963	7006	954	14778	805	51100	373	286
Eastern	S. Potato	-	-	-	-	3453	25624	4023	37937	6864	46691
	Cassava	-	-	-	-	2557	18355	3554	31067	5168	40136
Rift Valley	S. Potato	826	5962	629	5856	966	8610	1660	18788	1598	16120
	Cassava	922	10321	1338	16174	1040	10100	1136	11708	974	9943
Nyanza	S. Potato	9733	75950	10546	93404	12296	148871	15380	185065	16330	175725
	Cassava	18466	152007	21375	178458	27764	293886	25804	250743	25132	250311
Western	S. Potato	7887	83848	7324	77158	7054	69054	7853	68667	-	-
	Cassava	20085	182974	19866	169516	14771	127710	19624	154217	-	-
Coast	S. Potato	-	-	-	-	-	-	-	-	-	-
	Cassava	-	-	-	-	-	-	-	-	-	-
N/Eastern	S. Potato	-	-	-	-	-	-	-	-	-	-
	Cassava	-	-	-	-	-	-	-	-	-	-
National	S. Potato	37070	405444	16148	249863	32709	315141	27300	285000	31313	315461
	Cassava	100474	984108	83716	565295	116232	631300	86190	780000	46703	435244
	Maize	1380000	26000000	1407000	27000000	134485	19500000	1500000	34000000	1438740	29870368

Source: MOALD&M, Crops Division, 1997.
Ha. = Hectares
Prod. = Production

Table 14 gives the hectarage and production of sweet potato, cassava and potato since the year 1991 to 1995. In Central province, hectarage under sweet potatoes increased progressively in the period 1991-1993 and then began a gradual descent in the 1994-95 year period as did its overall production. Cassava production in this province consistently increased, rising to a maximum in the year 1994 and then decreasing sharply in 1995. In Eastern Province, cassava production showed a marked increase between the years 1993-1995 compared to sweet potato. In Nyanza province both sweet potato and cassava increased in production in the period 1991 - 1994 and then dropped in 1995. Nationally, sweet potato production remained more-or less the same throughout the period 1992-95 with a maximum in the year 1991. Cassava production nationally fluctuated in the period under review. These statistics have an implication for the demand in sweet potato and cassava over the tune period. The gradual increase in production may be a result of a correspondingly gradual increase in demand.

Further statistics from the Ministry of Agriculture, Livestock Development and Marketing give the yearly average prices (Kshs.) per bag of sweet potatoes in the period 1993-1996 as shown in Table 15 below.

Table: 15 Sweet Potato Yearly Average Prices Per Bag For Selected Districts, 1992-1996

District	Year
	1992	1993	1994	1995	1996
Nairobi	700	920	1,021	830	887
Meru	285	517	1,764	506	474
Kisii	245	342	679	545	589
Mombasa	1,001	919	1,359	1,277	1,661

Source: MOALD&M, Marketing Division

This secondary information has revealed a potential for expansion in the root crop sub-sector. However, this information is not specific to particular root crops (arrow root, cassava, sweet potato, yam etc.) and hence limiting in its use. To supplement this and with special reference to sweet potato consumption patterns in Nairobi, a study was conducted whose results are in the section that follows.

Household consumption survey

As earlier stated, the study sample comprised of middle to low income earners, drawn from several different ethnic groups. The ethnic groups represented included the Kikuyu, Luhya, Luo, Kisii, Kalenjin, Kamba, Taita, Meru and Embu.

Table 16: Percent Distribution of Respondents by Tribe, (N=35)

Kikuyu	Luhya	Luo	Kisii	Kalenjin	Others*
32.2	30.1	10.3	7.6	7.6	12.2

Others* included the Kamba, Meru, Taita and Embu.

All the respondents had lived in Nairobi for more than five years. The study sought to find out the three most important staples among this urban population. Results are as shown in Table 17a below.

Table 17a: % Distribution of the Three Most Important Staple Foods, (N = 35)

	Most Important	2nd Important	3rd Important
Ugali	67.5	11.4	2.9
Rice	2.9	28.6	8.6
Githeri/Beans	18.1	14.3	22.9
Bananas	0	11.4	20
S. Potatoes	5.7	2.9	2.9
Others*	2.9	5.7	14.3

* Others comprise of cassava, fish, meat, vegetables, chapati and Irish potatoes.

Maize was found to be the most important staple as indicated by 67.5% of the respondents. Another 18.1% considered beans consumed in venous forms, but commonly mixed with maize and boiled (githeri/muthokoi/mushenye) to be the main staple. Rice was the second most important staple among these urban respondents as shown by 28.6% of the people interviewed; another 14.3% considered beans as the second most important staple. 22.9% of the interviewees said that beans (githeri/mzokoi/mushenye/mahenjere) were the third most important staple whereas another 20 % rated bananas as the third most important staple. These ratings were considerably influenced by the ethnic backgrounds of the respondents.

The respondents were disaggregated on tribal basis and their preferences between maize, cassava and sweet potato analysed. Results show that almost all of them preferred maize as the main staple Table 17b shows tribal preferences between cassava and sweet potato

Table 17b: % Distribution of Study population by tribal, Preferences between Cassava and S. Potato (N=35)

	Kikuyu,	Luhya	Lou	Others
	n = 11	n = 11	n = 7	n = 6
S. Potato	90.9	81.8	100	66.6
Cassava	0	0	0	16.7
Both	9.1	9.1	0	16.7
none	0	9.1	0	0
Total	100	100	100	0

Others include, Kamba, Taita, Embu, Kalenjin and Meru.

Table 17b shows that all the tribes interviewed preferred sweet potato to cassava.

Thirty four out of 35 people (97.1%) interviewed ate Sweet potatoes in a boiled form. The respondents were asked to give the number of times they had eaten sweet potatoes in the previous fortnight, the previous month and every fortnight while they were stiff in their rural homes and had not moved to live in Nairobi. The frequency of consumption is as shown in Table 18.

Table 18: Frequency of Consumption of Sweet Potatoes (%) (N = 35)

	Last 14 days	Last month	In rural home
Once	34.3	34.3	25.7
Two times	8.6	14.3	8.6
Three times	2.9	11.4	22.9
> Four tunes	0	5.7	17.1
Not at all	42.9	34.3	8.6

Table 18 shows that a substantial number of respondents had not eaten sweet potatoes in their households in the previous fortnight (42.9%) and in the previous month (34.3%). About 34% had consumed sweet potatoes only once in the previous fourteen days and month. These results also show that sweet potato consumption was more frequent in the rural areas five years ago than it is being consumed in Nairobi today as shown by 17.1% of the respondents who ate sweet potato more than four times in a fortnight in their rural homes before they migrated to Nairobi, but now only 5.7% of the respondents do that in Nairobi

This study also wanted to find out the intra-household consumption patterns of sweet potatoes. Results show that sweet potatoes are eaten by all the members of the family. When asked who ate most of the sweet potatoes, (9.1%) said it was the husband, (12.1%) said the wife, (27.3%) the children and (30.3%) said others as indicated in Table 19. 17.9% could not tell who ate most of the potatoes consumed in the family.

Table 19: % Distribution of Member of Family That Eats Most of the Sweet Potatoes (N = 33)

Family Member	% Distribution
Husband	9.1
Wife	12.1
Children	27.3
Visitors/Workers	3.3
Other*	30.3
Do not know	17.9
Total	100

* Other here refers cases where all members of the family eating equally.

Sweet potatoes were mostly eaten at breakfast time (51.4%) or as a snack (31.0%) as shown in Table 20A. Non of the respondents indicated eating Sweet potatoes in ceremonial places like weddings/funerals.

Table 20A: % Distribution of Meals at Which Sweet Potatoes are Consumed (NT = 35)

Meal	% Distribution
Breakfast	51.4
Lunch	5.7
Supper	2.9
Snack	31.0
Supper	6.1
Do not eat at all	2.9
Total	100

Table 20B is a comparison of the nutritive values of the main staple maize the second most important staple beans and sweet potato, as computed from the National Food Composition Tables for Kenya (Sehmi, 1993).

Table 20B: Nutritive Values of Selected Staples in Kenya

Food Type	Calories (KCal/100 g)	Protein/100 g	Ash (minerals)/100 g
Maize	358	10.0	2.3
Cassava	82.0	0.74	0.72
Sweet-Potato	69	0.98	0.96
Beans	336	18.0	4.08
Potato (English)	81	2.01	1.08

Source: Computed from Sehmi, 1993 - National Food Composition Tables for Kenya... pages 51-60.

Most of the respondents ate Sweet potatoes either when there was a lot in the market (32.3) or when the main staple was not available (19.4%). A substantial number of respondents said they mostly ate potatoes when they received some as a gift (16.1%) or when they harvested from their gardens (16.1). See Table 21. Another 16.1% did not eat potatoes at all in their households.

Table 21: % Distribution of the Study Population By Reasons For Most Consumption of Sweet Potatoes (N=31)

Reason	% Distribution
Main staple not available	19.4
When there is a lot in market	32.3
When received as a gift	16.1
Other*	16.1
Did not eat at all	16.1
Total	100

*Other = grow them in own garden and eat after harvesting.

A total of 35 people were asked whether they ate cassava 65.7% said they did whine 34.3% (%) did not. Reasons for not eating cassava were varied: medical due to suffering stomaches after eating cassava, cassava is not readily available in the local markets while others admitted that they had not thought of cassava as food!

The study population was also asked to state their preference between cassava and sweet potato. Results are given in Table 22. 88.6% preferred sweet potatoes to cassava, 8.6 % had no preference and regarded each of the two crops to be of equal importance whine 2.9% preferred cassava to sweet potatoes.

Table 22: Preference Between Sweet Potato and Cassava (N = 35)

Type of Food	% Preference
Cassava	2.9
Sweet Potato	88.5
No Preference	8.6
Total	100

Respondents were also asked to rank, maize, sweet potatoes and Cassava in terms of the most expensive, second most expensive and then the least expensive. Table 23 shows the results.

Table 23: % Distribution of Respondents By Cost-Ranking of Maize, Cassava and Sweet potatoes (N = 35)

Food Type	Most Expensive	2nd most Expensive	3rd Most Expensive
Maize	66.4	25.1	8.5
Sweet Potato	31.4	48.8	20.9
Cassava	3.8	25.7	70.5

Maize which is the main staple as earlier shown was found to be the most expensive (66.4%). The second most expensive was sweet potato (48.8%) while the least costly of the three was Cassava (70.5%)

The survey also sought to find out if more people were eating sweet potatoes and cassava then than in the previous five years. Results are as in Table 24 below. Less people are now eating sweet potatoes and Cassava than they did five years ago.

Table 24: % Distribution of Respondents by Shift in Consumption of Sweet Potato and Cassava Over the Past Five Years (N = 35)

	More people eating	Less People eating	No change	Don't Know
Sweet Potato	37.1	45.7	5.7	11.5
Cassava	31.4	45.7	8.6	14.3

Reasons for the decreased consumption levels of sweet potato were cited as: people have become more modernised and prefer modern exotic foods and grow less of the potatoes; sweet-potato is not as readily available today as it was five years ago; Kenya's rainfall pattern is erratic and unreliable leading to decreasingly unreliable sweet-potato harvests; poor crop husbandry practices especially pest control lower production; increasing population resulting in increased land fragmentation and subsequent decreased landholding sizes, hence difficult to cultivate more potatoes since priority is given to the main staple; availability of sweet potato is not uniform throughout the country because not all areas in the country produce the crop; hence sweet potato is increasingly becoming expensive. Other respondents (37.1%) felt that more people are eating sweet potatoes than five years ago for a number of reasons: eating habits have changed in favour of sweet potatoes; sweet potatoes are cheaper and easier to cook; there are times when there is not enough maize in the market and households do not harvest enough to meet all their food requirements hence a resulting situation of less of the staple food in the market at high prices; sweet potato is drought tolerant and makes up for shortages of the main staple especially when rainfall fails; cost of living has become generally high; sweet potatoes are preferred to bread at breakfast time because they are cheaper and sweet and as such one does not require sugar in their tea.

Reasons for decreasing consumption of cassava were several variety of alternatives available has increased; cassava is not readily available; production levels are very low; it is expensive. Reasons for increased cassava consumption were that it is drought tolerant and therefore it can be harvested throughout the year; it is cheaper than bread and used as an alternative for breakfast.

Summary and conclusion

Secondary information shows that maize is the predominant staple in Kenya, among all the tribes. The survey has revealed that,

· the second most important staple among the ethnic groups interviewed is rice whereas sweet potato is highly preferred to cassava

· the frequency of sweet potato consumption per fortnight is lower in the Nairobi today than it was in the rural areas five years ago whereas all the family members consume sweet potatoes, children eat the most, followed by the wives and then the husbands. There was no indication of high sweet potato consumption during ceremonies and special occasions

· sweet potato in Nairobi was mostly consumed in a boiled form and either at breakfast time or as a snack. There were few instances where it was consumed for lunch or dinner.

· Consumption of sweet potato and cassava had decreased over the last five years; and this was attributed to low production levels, high retail prices and a change in eating habits due to rural-urban migration.

· both the middle and low income respondents consumed sweet potatoes with similar frequency, at similar meals and preferred them at similar meals and in boiled form. That they also faced similar constraints particularly the rising retail prices

· the study was not able to capture the high income class because of limited time.

These findings show that there is potential for post-harvest analysis of sweet potato and subsequent implementation of relevant operations in Kenya as a whole. The finding that some people have never thought of cassava as a food needs to be further explored and also the low production levels of the same. Kenya's population is swelling, particularly in urban areas whereas in the rural areas migration to the arid and semi-arid farming lands (ASALS) continues. The implication of this is that the kind of crops produced in these ASALS are those relevant to the agro-ecological zones; these now include dryland farming crops like sweet-potato and cassava. Consequently, eating habits are changing. There is therefore a need for increased accessibility of the populations to these foods. One important aspect that was not explored by this study is the competitiveness of the marketing operations for sweet potato in Kenya. This is crucial because the degree of competitiveness has a bearing on the magnitude of sales and indirectly influences production levels. These in turn depend on consumption levels of the product. It is necessary therefore that these parameters be investigated further so that the streamlining of post-harvest operations of storage, processing and transportation augurs well with wholesale and retail marketing, hence increased production by the farmers..

References

Central bureau of Statistics, 1996 - Welfare Monitoring Survey II, basic Report, 1994. pp. 341

Mwangi M.A, 1995 - The role of Urban agriculture for food security in low income areas in Nairobi Food and Nutrition Studies Programme, Report No. 54/1995 pp. 84.

Republic of Kenya, 1997 - National Development Plan, 1997-2001. pp. 254.

Republic of Kenya 1994 - Kenya population census, 1989, Vol. 1.

Sehmi K J. 1993 - National Food Composition Tables and the Planning of Healthy Diets in Kenya Government Press, Nairobi pp. 200.

UNICEF/RoK, 1992 - Children and Women in Kenya, A situation Analysis. Regal Press Ltd., Nairobi pp. 162.

Appendix 1: Table 11: Mean Monthly Household Income From Sale of Crops By District (KShs.)

District	Maize	Cash	Wheat	Cereals	Vegetables	Beans	Roots	Sugar cane	Fruits	Other Crops income	Total Crops income
Nairobi	1.1	0.0	0.0	0.0	0.0	0.6	0.5	0.0	0.3	0.0	2.4
Meru	102.9	687.3	42.8	16.2	84.6	212.6	97.9	89.8	31.8	6.8	1,372.6
Kisii	538.5	216.3	0.0	19.7	12.1	181.2	15.9	586.6	65.1	22.5	1,658.1
Rural	440.0	206.6	201.6	18.0	208.4	59.7	83.7	77.2	29.2	93.3	1,417.7
Urban	66.0	10.3	0.2	0.1	16.9	7.6	6.6	0.2	1.4	0.6	110.5
National	363.4	166.3	1603	14.3	169.1	49.0	67.9	61.3	23.5	74.3	1,149.3

Source: GoK (1994),Welfare Monitoring Survey II pages

Appendix 2: Table 12: Mean Monthly expenditure on Major Staples Per Province (Ethnic Group)

Province (District)	Cereals (%)	Root Crops (%)
Nairobi	25.6	4.2
Central	40.1	10.1
Nyandarua (District)	27.1	23.3
Coast	43.1	3.6
Nyanza	51.8	2.8
Kisii (District)	57.5	1.7
Homa Bay (District)	54.3	4.0
Rift Valley	53.0	3.1
Western	43.2	5.2
Eastern	44.2	3.7
Meru (District)	35.2	8.1
Machakos (District)	44.7	2.0
North Eastern	28.6	0.7

Source: GoK (1994),Welfare Monitoring Survey II p. 142

Appendix 3: Table 13: Mean Monthly Household Food Expenditure (inclusive of Own Consumption) By Broad Expenditure Categories By District (KShs.)*

District	Cereals & Cereal Products	Meat & Meat products	Oils	Vegetables/Fruits	Roots	Sugar	Tea	Other Food Products	Purchases Food Expenses
Nairobi	1,707.0	1,823.0	550.1	1,324.9	278.9	628.1	276.6	79.7	6,591.5
Meru	1,554.9	1,051.5	186.8	905.8	356.8	244.2	108.7	8.9	2,890.9
Kisii	3,395.2	950.6	394.2	464.4	99.4	461.3	110.8	31.0	4,033.8
Rural	2,316.2	886.6	207.2	561.6	204.9	378.0	103.2	86.7	3,266.2
Urban	1,635.2	1,543.4	434.8	1,028.3	250.9	493.1	197.5	78.2	5,458.8
Total	2,176.4	1,021.4	253.9	657.4	214.4	401.7	122.5	85.0	3,716.3

Source: GoK (1994),Welfare Monitoring Survey II pages 142-143

* All food categories include own house consumption

** Total food expenditure includes purchased food expenditure, own crop consumption & livestock products consumption.