A survey of Smallholder dairy cattle farmers in Tanzania: Farmer Demographic Characteristics and Basic Management Constraints

doi:10.21203/rs.3.rs-4993873/v1

Download PDF

Research Article

A survey of Smallholder dairy cattle farmers in Tanzania: Farmer Demographic Characteristics and Basic Management Constraints

https://doi.org/10.21203/rs.3.rs-4993873/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Since the 1980s, there has been a significant change in the population of smallholder dairy cattle farmers across Tanzania. Therefore, we lack demographic data on that population as well as information on the key constraints reducing productivity. This survey of smallholder dairy farmers across six regions of Tanzania aimed to provide demographic data and to identify those key constraints.

Results

Of the 301 households surveyed, 74% were headed by men, but in Njombe there was an equal number of women and men. Across all regions, the majority of household heads were aged between 41 to 60 years and had ≤ 10 years of experience as a smallholder dairy farmer. Dairying was the major source of income on 84% of households, ranging from 59–100%, across the six regions. Most respondents had primary education but had gone no further (55%); however, in Morogoro 68% of farmers had been in secondary/university education. Across four regions (Njombe, Mbeya, Kilimanjaro and Arusha), herd size of 3–4 animals was most common (32–50%); however, in Morogoro and Tanga most herds had ˃4 animals (66% and 78%, respectively). Zero-grazing was the most common grazing system (75%), but tethering was predominant (68%) in Mbeya. Across all regions natural pastures were the principal feed source. Across all respondents, cash purchase was the most common means of obtaining their first cattle beast (66%), although in Mbeya a gift from a relative/friend (49%) was the most common source., High input costs (93%), unavailability of feed (71%), lack of land (68%) and diseases (62%) were the most commonly identified constraints., while high breeding costs (96%), poor oestrus detection (89%), cows not displaying oestrus (79%) and lack of AI services (51%) were the key constraints to successful breeding.

Conclusion

Although there are significant common pints across all six regions in our study, demographic differences across the regions necessitate a regional approach to development. This approach needs to focus on who the farmers are in that region and what those farmers need rather than trying to implement the same solutions across Tanzania.

Smallholder dairy cattle farmer

household

demographic

farming constraints

Focused initiatives to improve dairy cattle industry in Tanzania started in 1921, when the colonial government of the time introduced the first Holstein and Ayrshire cattle into Temeke, Dar es Salaam (40). This move was followed by establishment of few medium-to-large scale settler dairy farms in the Northern and Southern highland regions of the country. However, most of these farms were nationalised after independence, following the Arusha Declaration in year 1967. This resulted in the dairy sector in the 1970s being dominated by governmental parastatal and state-owned farms, with the formation in 1975 of the Tanzanian Dairy Farming Company. Concurrently, Tanzania Dairies Ltd. was formed, which was responsible for processing and selling the milk to consumers (17, 24). However, subsequent challenges, such as political interference with the milk price and the generally low productivity of the cattle, led to poor economic performance and collapse of these governmental parastatals and state-owned farms in the early 1980s (17, 24).

After this collapse, efforts to bridge the gap between Tanzanian milk production and demand for dairy products within Tanzania began. These efforts were led by private, religious and non-governmental organizations, all of which promoted smallholder dairy cattle farming (16). Smallholder dairy cattle farming refers to a type of dairy farming typically practiced on a small scale with the mean herd size of approximated 4 animals, ranging from 1 to 12 cattle, often involving unimproved genetics and rudimentary management by individual farmers or households (3, 23, 31, 1, 30, 39). These efforts were supported, at the government level, by the Tanzania Livestock Policy, which emphasised the promotion of smallholder dairy cattle farming (24).

Initially, most smallholder dairy farms were owned by high and mid-ranking civil servants, with cows being kept in their owner’s place of residence (“backyard production“), with milk production being a supplementary, rather than primary income source (24, 43). However, beginning in the 1990s, smallholder dairy cattle farming started experiencing rapid growth across Tanzania, with a change in the demographics of farmers, from producers for whom milk was a supplementary income, to producers for whom milk was the primary source of income (19). The smallholder dairy sector thus currently provides significant income and employment across rural, peri-urban and urban areas of Tanzania, as well as being a valuable source of human nutrition in those areas (19, 8, 29).

Of the 33.9 million cattle in Tanzania, 33.8 million (99.6%) are kept by smallholder farmers, with only 142,000 (0.4%) in large-scale farms (i.e. farms that have more than 100 cattle and which have their own input delivery systems (22, 29). Around 70% of the milk produced in Tanzania is from indigenous/local Zebu cattle, with the remaining 30% from improved dairy cattle breeds (2). For smallholders, improved dairy cattle are generally crosses of European dairy breeds (e.g. Friesian, Ayrshire, and Jersey) with local Zebu, especially the Tanzanian Shorthorn Zebu, but also Boran and Sahiwal. These improved dairy cattle are concentrated in the rural areas of Tanga, Arusha, Kilimanjaro and Manyara (Northern Highland regions), Mbeya, Iringa and Njombe (Southern Highland regions) as well as in the Morogoro, Kagera and Dar es Salaam regions (43).

Initiatives to improve the productivity of the Tanzanian dairy industry, particularly smallholder farms, are ongoing (4). These initiatives are a mixture of government-led programmes and external stakeholder-led programmes (with significant Tanzanian government support). They include long-running programmes such as the Dairy Development Program, and Heifer Project International in Tanzania (HPI) Scheme (26). Community Action Research Program (37), AgResults Tanzania Dairy Productivity Challenge Project (2019–2024), and the newer countrywide program for the transformation of the livestock sector titled Livestock Sector Transformation Plan (25).

However, despite the significant change in the population of smallholder dairy cattle farmers across Tanzania since the 1980s, there is limited information regarding the demographics of such farmers. Some localised surveys of smallholder dairy cattle farming have been undertaken, but these have generally been limited to one or two regions: Dar es Salaam (15), Morogoro (7, 9), Tanga (1), and Kilimanjaro and Arusha (42). Larger-scale surveys have been undertaken, but they either did not report demographic data (4) or had a limited analysis of such data (27).

Additionally, most of these studies were conducted in urban and peri-urban areas. A better understanding of the demographics of smallholder dairy farmers (especially at the rural/region level) is needed to help advisers and development agencies to better understand who smallholder dairy farmers are, as well as their goals and challenges. Such understanding could be useful in targeting support for dairy farmers based on individual requirements (especially if structured at the regional level) rather than using a “one-size-fits-all” programme across Tanzania. The aim of the current survey was therefore to elicit information on demographic characteristics of smallholder dairy cattle farmers from thirteen districts across six different regions of Tanzania, and to detail the constraints that the farmers perceive to be affecting their productivity.

Ethical considerations and approval

This research was approved by the Ministry of Livestock and Fisheries through the Ethics Review Board of the Tanzania Livestock Research Institute (TALIRI) (reference number TLRI/RCC.21/007) of the United Republic of Tanzania. Permission letters were firstly provided by the office of Regional Administrative Secretary from the six study regions, and then, from the Executive Directors of the respective District Council (DC), Town Council (TC), Municipal Council (MC) or City Council (CC) of the thirteen study districts of Tanzania mainland. A local veterinarian or livestock officer first introduced the interviewer (all interviews were done by the first author) to the farmer/respondent (usually the family head), then the interviewer explained the reason for the visit. Thereafter, each respondent/farmer was given a written informed consent form for them to sign before participating in the questionnaire interview. If the interviewee was unable to read and write, another family member was called to approve and sign on behalf. Results were anonymized and personal data kept confidential. The specific consent of the relevant participants was obtained for any photographs used to illustrate this study.

Study area and study farmsselection

Information from smallholder dairy cattle farmers was gathered using a cross-sectional study design, from May 2022 to February 2023. Six regions of the Tanzania mainland were purposely selected, principally based on the proportion of improved dairy cattle. Three of these regions were in the Northern Highlands (Arusha, Kilimanjaro and Tanga), two in the Southern Highlands (Mbeya and Njombe) and one (Morogoro) in the Eastern zone. Within each region, district(s) were selected by convenience sampling with the help of local veterinarians/livestock officers (Fig. 1). Within each district, convenience sampling was employed to identify study villages and the first study farm in each village (whose suitability was decided by the local veterinarian/livestock officer and interviewer). Snowball sampling was then employed to select other study farms in a particular village. Subjects could nominate as many further subjects as they wished with non-discriminative sampling being used until > 50 respondents had been identified per region.

(INSERT Fig. 1.)

Field data collection

Structured and pre-tested questionnaire was used for collecting research data. Pre-testing was done in the Morogoro municipality capturing twenty-four smallholder farmers (who were not included in the final questionnaire) and eleven experts (veterinarians/livestock officers and researchers). The questionnaire was formulated so that all the questions were closed, and questions were finally entered into KoboToolbox (Cambridge USA) for subsequent data collation.

Initial information collected from farmers included age, gender, dairy farming experience, involvement (full or part time), education level, dependence on dairy farming (plus other household/farm income generating activities) and decision-making process on farm. Respondents were also asked on farm-related issues, such as source of their first dairy animals, herd size, herd composition as well as breeding practices and preferences. The last section asked the respondents for their opinions regarding the key constraints that affected their farm capturing information related to cattle health and reproduction, availability of veterinary and breeding services, availability of land and feed, and access to markets for their product.

Data management and analysis

Data from the questionnaire were downloaded from KoboToolbox to Excel spreadsheets (Microsoft, Seattle, USA) before analysis using SPSS version 25 (IBM, Seattle, USA). Results are tabulated and presented as overall results and by region. For most responses, multinomial logistic regression was used to evaluate the effect of region on the key outcome. If responses were clearly ordered, ordinal logistic regression was used, provided the proportional odds assumption was met. If this was not met, then a multinomial regression was used. Categories were merged for all analyses where totals were < 10. For all analyses, except where stated, Tanga was used as the reference region, and the category with the highest frequency in the outcome was set as the reference category. Data from Arusha were excluded from the analyses due to small number of respondents; however, they have still been included in the tables and figures for reference. Further, data for individual regions are not reported where differences from Tanga are within the 95% confidence interval (CI) for that region.

At least 50 smallholder dairy cattle farmers were interviewed per region, except Arusha where only 16 farmers were interviewed. Fewer farmers in Arusha participated following the unavailability of local veterinarians and livestock officers, who were participating in the national livestock identification program, resulting in reluctance among farmers to participate. Overall, across the six regions, 301 farmers were recruited for the survey (Table 1).

Table 1

Number of regions, districts and smallholder dairy cattle farmers visited during a survey on smallholder dairy farming, demographics, and constraints in Tanzania.
S/No	Regions visited	Districts visited	Farmers
S/No	Regions visited	Districts visited	Per District	Per Region
1	Njombe	Njombe TC	54	54
2	Mbeya	Busokelo DC	55	55
3	Tanga	Tanga CC	12	53
3	Tanga	Muheza DC	41	53
4	Kilimanjaro	Moshi MC	8	66
		Moshi DC	18
		Hai DC	40
5	Arusha	Arusha CC	6	16
		Arusha DC	1
		Meru DC	9
6	Morogoro	Morogoro MC	47	57
		Morogoro DC	5
		Mvomero DC	5
Total		13	301	301
Key: DC - District council, TC - Town council, MC - Municipal council and CC -City council

Smallholder dairy farmers’ household, family, and farm demographics

Of the 301 households which participated in this study, 224 (74%) were headed by a father and 69 (23%) by a mother (Table 2). For the analysis of the effect of region, two categories were created (father and mother) with respondents who were recorded as ‘other’ combined with father or mother depending on their gender (male or female). There were differences between regions in who was the head of the household (Fig. 2), with households in Njombe having much higher odds of having a female head of the household than households in Tanga (odds ratio (OR): 5.2, 95%CI: 2.8–13.1).

Table 2

Smallholder dairy farmers household and herd characteristics across the study regions in Tanzania mainland. Data are shown as the number of respondents and percentage of responses within each region n (%).
Character	Category	Region
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n = 54 (18)	n = 55 (18)	n = 53 (18)	n = 66 (22)	n = 16 (5)	n = 57 (19)	n = 301 (100)
Head of the family
Father		26 (48)	48 (87)	45 (85)	52 (79)	12 (75)	41 (72)	224 (74)
Mother		26 (48)	2 (4)	8 (15)	14 (21)	4 (25)	15 (26)	69 (23)
Other		2 (4)	5 (9)				1 (2)	8 (3)
Type of family
monogamy		41 (76)	43 (78)	43 (81)	46 (70)	14 (88)	49 (86)	236 (78)
polygamy		5 (9)	4 (7)	4 (8)	2 (3)		2 (4)	17 (6)
widow		6 (11)		4 (8)	11 (17)	2 (13)	4 (7)	27 (9)
widower				1 (2)	5 (8)			6 (2)
Single parents & unmarried		2 (4)	8 (15)	1 (2)	2 (3)		2 (4)	15 (5)
Participatory decision making in the family
	yes	39 (72)	49 (89)	46 (87)	51 (77)	15 (94)	48 (84)	248 (82)
	no	15 (28)	6 (11)	7 (13)	15 (23)	1 (6)	9 (16)	53 (18)
People taking care of animals
	1, 2	40 (74)	38 (69)	23 (43)	44 (67)	14 (88)	42 (74)	201 (67)
	3, 4	13 (24)	17 (31)	30 (57)	22 (33)	2 (13)	14 (25)	98 (33)
	≥ 5	1 (2)					1 (2)	2 (1)
Source of the first dairy cattle/cow
Cash purchase		37 (69)	25 (45)	27 (51)	48 (73)	13 (81)	50 (88)	200 (66)
Gift (relatives or friends)		3 (6)	27 (49)	11 (21)	17 (26)	3 (19)	6 (11)	67 (22)
NGO^a		14 (26)		14 (26)			1 (2)	29 (10)
Home bred and bank			3 (5)	1 (2)	1 (2)			5 (2)
Total herd size (all cattle)
	1, 2	14 (26)	15 (27)	6 (11)	15 (23)	1 (6)	9 (16)	60 (20)
	3, 4	25 (46)	20 (36)	6 (11)	21 (32)	8 (50)	10 (18)	90 (30)
	5, 6	12 (22)	13 (24)	11 (21)	15 (23)	2 (13)	9 (16)	62 (21)
	7, 8	2 (4)	5 (9)	13 (25)	8 (12)	1 (6)	11 (19)	40 (13)
	9, 10		1 (2)	11 (21)	5 (8)	4 (25)	3 (5)	24 (8)
	≥ 11	1 (2)	1 (2)	6 (11)	2 (3)		15 (26)	25 (8)
Numbers of cows
Respondents with adult cows		n = 52 (96)	n = 48 (87)	n = 53 (100)	n = 60 (91)	n = 15 (94)	n = 54 (95)	n = 282 (94)
	1, 2	32 (62)	26 (54)	18 (34)	33 (55)	10 (67)	18 (33)	137 (49)
	3, 4	18 (34)	20 (42)	22 (42)	22 (37)	5 (33)	19 (35)	106 (38)
	5, 6	1 (2)	1 (2)	9 (17)	3 (5)		5 (9)	19 (7)
	7, 8	1 (2)		1 (2)			6 (11)	8 (3)
	≥ 9		1 (2)	3 (6)	2 (3)		6 (11)	12 (4)
Numbers of heifers
Respondents with heifers		n = 23 (46)	n = 37 (67)	n = 36 (68)	n = 43 (65)	n = 13 (65)	n = 37 (65)	n = 189 (63)
	1, 2	22 (96)	33 (89)	32 (89)	41 (95)	9 (69)	25 (68)	162 (86)
	3, 4	1 (4)	4 (11)	3 (8)	2 (5)	4 (31)	11 (30)	25 (13)
Numbers of calves
Respondents with calves		n = 30 (56)	n = 31 (56)	n = 48 (91)	n = 43 (65)	n = 13 (65)	n = 43 (75)	n = 208 (69)
	1, 2	25 (83)	29 (94)	23 (48)	26 (61)	10 (77)	20 (47)	133 (64)
	3, 4	5 (17)	2 (6)	18 (38)	14 (33)	2 (15)	13 (30)	54 (26)
	≥ 5			7 (15)	3 (6)	1 ()	10 (23)	21 (10)
Numbers of bulls
Respondents with bulls		n = 2 (4)	n = 2 (4)	n = 8 (15)	n = 16 (24)	n = 6 (38)	n = 18 (32)	n = 52 (17)
	1, 2	2 (100)	2 (100)	8 (100)	15 (94)	6 (100)	15 (83)	48 (92)
	≥ 3				1 (6)		3 (17)	4 (8)
For each region, results in bold indicate the category with the highest frequency for that question; ^a non-governmental organisation, e.g. Heifer-in-Trust Scheme

Table 3. Demographics of smallholder dairy farmers and assistants/workers across the study regions in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Region
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Demographics of dairy farmers
Total respondents		n=54 (18)	n=55 (18)	n=53 (18)	n=66 (22)	n=16 (5)	n=57 (19)	n=301 (100)
Age
	≤20	1(2)		1(2)			1(2)	3(1)
	21-40	12(22)	22(40)	9(17)	8(12)	1(6)	8(14)	60(20)
	41-60	25(46)	29(53)	26(49)	36(55)	11(69)	28(49)	155(51)
	≥61	16(30)	4(7)	17(32)	22(33)	4(25)	20(35)	83(28)
Experience in dairy cattle farming (years)
	≤10	27(50)	21(38)	16(30)	20(30)	3(19)	31(54)	118(39)
	11 to 20	16(30)	19(35)	18(34)	14(21)	6(38)	11(19)	84(28)
	21 to 30	7(13)	10(18)	12(23)	20(30)	4(25)	12(21)	65(22)
	31 to 40	4(7)	3(5)	7(13)	10(15)	2(13)	2(4)	28(9)
	≥41		2(4)		2(3)	1(6)	1(2)	6(2)
Involvement in dairy cattle farming
	Full time	41(76)	42(76)	46(87)	50(76)	11(69)	41(58)	231(77)
	Part time	13(24)	13(24)	7(13)	16(24)	5(31)	16(42)	70(23)
Education level
No formal education		2(4)	2(4)	1(2)	2(3)		1(2)	8(3)
	Primary	32(59)	34(62)	35(57)	38(52)	10(63)	17(30)	166(55)
	Secondary	10(19)	9(17)	8(23)	8(12)	1(6)	7(12)	43(14)
	College	5(9)	5(9)	6(11)	15(26)	3(19)	9(16)	43(14)
	University	5(9)	5(9)	3(8)	3(8)	2(13)	23(40)	41(14)

Demographics of farm assistants

Total respondents

n=31(15)

n=25(12)

n=43(21)

n=46(23)

n=6(3)

n=52(26)

n=203(%)

Age

≤20

16(52)

7(28)

7(16)

14(31)

3(50)

24(46)

71(35)

21-30

8(26)

17(68)

17(40)

19(41)

2(33)

23(44)

86(42)

31-40

3(10)

1(4)

14(33)

7(15)

1(17)

4(8)

30(15)

≥41

4(13)

5(12)

6(13)

1(2)

16(8)

Experience in dairy cattle farming (years)

6(19)

2(8)

3(7)

8(17)

17(33)

36(18)

1 to 5

21(68)

16(64)

27(63)

32(70)

6(100)

30(58)

132(65)

6 to 10

2(6)

5(20)

10(23)

5(11)

5(10)

27(13)

11 to 15

1(3)

2(8)

1(2)

4(2)

≥16

1(3)

2(5)

1(2)

4(2)

Involvement in dairy cattle farming

Full time

26(84)

18(72)

35(81)

40(87)

6(100)

51(98)

176(87)

Part time

5(16)

7(28)

8(19)

6(13)

1(2)

27(13)

Education level

No formal education

6(19)

1(4)

4(9)

6(13)

4(8)

21(10)

Primary

16(52)

15(60)

28(65)

30(63)

3(50)

42(81)

134(66)

Secondary

4(13)

4(16)

8(19)

6(13)

2(33)

4(8)

28(14)

College

3(10)

5(20)

1(2)

2(4)

11(5)

University

2(7)

2(5)

2(7)

1(17)

2(4)

9(4)

For each region, results in bold indicate the category with the highest frequency for that question

(INSERT Table 2 and Fig. 2)

Most households (236/301; 78%) were monogamous, with the lowest proportion recorded in Njombe (41/54; 76%), and the highest in Arusha (14/16; 88%) (Fig. 3). Additionally, in most households (248/301; 82%), the entire family was involved in the decision-making process, not just the head of the household.

(INSERT Fig. 3)

Total herd size ranged from 1 to 35 cattle, with 3–4 category being the mode herd size (90/301; 30%) (Table 2 and Fig. 4). For the analysis of the effect of region, five categories of herd size were used: 1–2, 3–4, 5–6, 7–8 and ≥ 9. Ordinal logistic regression identified differences across regions in the proportion of farms in one of higher herd size categories. Compared to Tanga, the odds of farms being in the higher herd size category were notably less in Kilimanjaro (OR: 0.3, 95%CI: 0.1–0.5), Mbeya (OR: 0.2, 95%CI: 0.08–0.3) and Njombe (OR: 0.1, 95%CI: 0.07–0.3).

(INSERT Fig. 4)

Of the 301 farms, 19 had no adult cows, 112 had no heifers, 98 had no calves and 249 farms had no breeding bulls. No effect of region on the proportion of farms with milking cows was found but compared to Tanga, farms in Njombe were less likely to have heifers (OR: 0.4, 95%CI: 0.2–0.8), and farms in Morogoro more likely to have bulls (OR: 2.6, 95%CI: 1.02–6.6). The proportion of farms with calves in Tanga was the highest of any region, with farms in Tanga having higher odds of having calves than farms in Njombe (OR: 0.1, 95%CI: 0.1–0.4), Mbeya (OR: 0.1, 95%CI: 0.1–0.4), Kilimanjaro (OR: 0.2, 95%CI: 0.1–0.6) and Morogoro (OR: 0.3, 95%CI: 0.1-1).

Most respondents (201; 67%) reported having fewer than three people who took care of the dairy cattle on their farm (Table 2). As with herd size, there was an effect of region such that, with reference to Tanga, the odds of having more than two people actively participating on farm was lower than in all other regions: Kilimanjaro (OR: 0.4, 95%CI: 0.2–0.8), Mbeya (OR: 0.3, 95%CI: 0.2–0.8), Morogoro (OR: 0.3, 95%CI: 0.1–0.6) and Njombe (OR: 0.3, 95%CI: 0.1–0.6).

Cash purchase was the dominant (200; 66%) source of the first dairy animal (Table 2). Regionally, this was true for all regions except for Mbeya, where a gift from a relative or friend was the most common source (Fig. 5). For analysis of regional differences, data were merged into three groups: cash, gift and other (merging non-governmental organisation (NGO), bank and home-bred). Arusha was excluded from this analysis as there were no farms in the ‘other’ category. Relative to cash purchase, three regions had different odds of a gift being the source of their first cattle beast than respondents in Tanga: the odds were higher in Mbeya (OR: 2.7, 95%CI: 1.1–6.4) and were lower in Morogoro (OR: 0.3, 95%CI: 0.1–0.9) and Njombe (OR: 0.2, 95%CI: 0.05–0.8). For the ‘other’ category, the odds were lower in Kilimanjaro (OR: 0.04, 95%CI: 0.01–0.3), Mbeya (OR: 0.2, 95%CI: 0.1–0.8) and Morogoro (OR: 0.04, 95%CI: 0.01–0.3) relative to cash purchase than in Tanga.

(INSERT Fig. 5)

Farmers/respondents and assistants/workers demographic characteristics

Most respondents were over 40 years of age (238/301; 79%), with the majority (155/301; 52%), being between 41 and 60 years. This was consistent across all regions (Fig. 6 and Table 3). Using ordinal logistic regression with four age categories (i.e. ≤20, 21–40, 41–60 and ≥ 61), the odds of a farmer being in a higher age category were lower in Mbeya (OR: 0.4, 95%CI: 0.2–0.7) compared to Tanga.

Experience in dairy farming was classified into 10-year blocks (≤ 10, 11–21, etc.). The highest proportion of respondents (118/301; 39%) had ≤ 10 years of experience. There was little difference between regions, except that ordinal logistic regression showed that the odds of being in a higher experience category were lower in Morogoro (OR: 0.49, 95%CI: 0.24–0.97) compared with Tanga.

(INSERT Fig. 6)

Most respondents (231/301; 77%) were involved full time in dairy cattle farming (Table 3). The proportion was highest in Tanga (46/53; 87%) and lowest in Morogoro (41/57; 58%), with the odds of being a part-time farmer being higher in Morogoro (OR: 4.8, 95%CI: 1.8–12.4) than in Tanga.

The most common level of education in respondents was primary level (7–14 years) (166/301; 55%) (Table 3 and Fig. 7). For analysis of the effect of region, respondents who had not had a formal education were excluded. Compared to respondents from Tanga, the proportion in each education category was similar across all regions except for Morogoro, where respondents were more likely to have had a higher category of education (OR: 6.5, 95%CI: 3-13.9).

(INSERT Table 3 and Fig. 7)

For the farm workers/assistants, the majority (157/203; 77%) were aged ≤ 30 years and had ≤ 5 years of experience (168/203: 83%), making them generally younger and less experienced than the main respondents (Table 3). Analysis of assistant demographics excluded data from Arusha as there were only 6 responses from that region. Compared to workers in Tanga, assistants from Mbeya (OR: 0.3, 95%CI: 0.12–0.77), Morogoro (OR: 0.2, 95%CI: 0.093–0.45), and Njombe (OR: 0.32, 95%CI: 0.09–0.56) all had lower odds of being in a higher age category. For experience, data were merged into four categories: <1 year, 1 to 5 years, 6 to years and ≥ 11 years. This analysis showed that assistants from Morogoro had lower odds of being in a higher experience category (OR: 0.18, 95%CI: 0.08–0.44) compared to those in Tanga, as did assistants in Kilimanjaro (OR: 0.37, 95%CI: 0.16–0.88). Most assistants were full-time workers (176/203; 87%), and most (182/203; 90%) had had only primary education. The education level of farm workers was similar across the study regions.

Smallholder dairy farmers’ sources of household income

Although 77% of respondents reported full-time involvement with dairy farming, almost all households reported having other sources of income (282/301; 94%) (Table 4). Crop farming was the most common alternative, with 180/282 (64%) gaining at least some income from it. Conversely, only 37% (104/282) and 29% (81/282) of respondents were involved in employment or business, respectively. For analysis of regional differences on other income sources, two categories were created i.e., ‘Yes’ (representing major, moderate and minor) and ‘Not at all’. Excluding Mbeya and Arusha where there are no respondents for the ‘Not at all’ category, the odds for a farmer participating in crop farming were lower in Njombe (OR: 0.1, 95%CI: 0.01–0.9) than in Tanga. Further, odds of a farmer relying on employment as the source of income were highest in Morogoro (OR: 5.2, 95%CI: 2.2–12.4) and lowest in Njombe (OR: 0.1, 95%CI: 0.04–0.4). Lastly, Mbeya had lower odds (OR: 0.2, 95%CI: 0.07–0.6) for its farmers depending on business as an income source compared to Tanga.

Table 4. Income generation activities in the smallholder dairy cattle household across the study regions in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Region
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n=54(18)	n=55(18)	n=53(18)	n=66(22)	n=16(5)	n=57(19)	n=301(100)
Dependency on dairy cattle farming
	yes	8(15)		5(9)	2(3)		4(7)	19(6)
	no	46(85)	55(100)	48(91)	64(97)	16(100)	53(93)	282(94)
Other income sources
Total respondents		n=46(16)	n=55(20)	n=48(17)	n=64(23)	n=16(6)	n=53(19)	n=282(100)
Employment
	Major	12(26)	3(5)	3(6)	7(11)		17(32)	42(15)
	Moderate	4(9)		11(23)	8(13)	1(6)	17(32)	41(15)
	Minor	2(4)	1(2)	5(10)	5(8)	1(6)	7(13)	21(7)
	Not at all	28(61)	51(93)	29(60)	44(69)	14(88)	12(23)	178(63)
Crop farming
	Major	24(52)	20(36)	3(6)	15(23)	3(19)	3(6)	68(24)
	Moderate	17(37)	34(62)	16(33)	31(48)	9(56)	5(9)	112(40)
	Minor	4(9)	1(2)	21(44)	14(22)	4(25)	34(64)	78(28)
	Not at all	1(2)		8(17)	4(6)		11(21)	24(9)
Business
	Major	6(13)		6(13)	7(11)	3(19)	4(8)	26(9)
	Moderate	3(7)	4(7)	7(15)	6(9)	1(6)	4(8)	25(9)
	Minor	2(4)	2(4)	5(10)	12(19)	4(25)	5(9)	30(11)
	Not at all	35(76)	49(89)	30(63)	39(61)	8(50)	40(75)	201(71)
For each region, results in bold indicate the category with the highest frequency for that question

Furthermore, farmers’ responses (major, moderate and minor, excluding ‘not at all’) for their involvement in other income-generation activities (i.e., crop farming, employment and business) apart from dairying, were further evaluated to determine their contribution to the household income. For that, responses were given score values i.e., 3 = major, 2 = moderate and 1 = minor; where the total score was ≥ 6, then dairying was defined as not being the major source of income and defined as being the major source of income when the total score was ≤ 5. Based on this score, dairying was the major income source to 238/282 (84%) households. Across the regions, the highest proportion recorded was 100% (16/16) in Arusha (100% and the lowest was 59% (27/46) from Njombe. Other regions were as follows: 86% (55/64), 88% (42/48), 91% (48/53) and 91% (50/55) from Kilimanjaro, Tanga, Mbeya, Morogoro and Mbeya, consecutively. Logistic regression was used to evaluate the regional differences (excluding Arusha region due to fewer respondents) of household dependence on other income sources. With reference to Tanga, only farmers from the Njombe had higher odds (OR: 4.9, 95%CI: 1.7–13.9) of non-dairying income being their major sources of income.

Smallholder farmers’ dairy cattle management system and feed sources

Across the 301 farms, 225 (75%) of respondents kept their dairy cattle under a zero-grazing /intensive system, in which forages were harvested daily and brought to the cattle in their shelter (Figs. 8, 9 and 10 and Table 5). This system dominated across all regions except Mbeya, where tethering at pasture was predominant (37/55; 68%). Across the different regions, the odds of semi-intensive farming (compared to intensive/zero grazing) were higher in Mbeya (OR: 12, 95%CI: 3.03–47.5) and Morogoro (OR: 6, 95%CI: 1.9–19.1) than in Tanga.

Table 5. Dairy cattle management systems and feed sources utilized by the smallholder dairy farms across the study regions of mainland Tanzania. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Region
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n=54 (18)	n=55 (18)	n=53 (18)	n=66 (22)	n=16 (5)	n=57 (19)	n=301 (100)
Cattle rearing system
intensive/zero grazing		52(96)	9(16)	48(91)	65(99)	13(81)	38(67)	225(75)
semi-intensive			9(16)	4(8)	1(2)	3(19)	19(33)	36(12)
tethering		2(4)	37(68)					39(13)
extensive				1(2)				1(0.3)
Feed sources
Fodder plot at farm
	major	10(19)	19(35)				3(5)	32(11)
	moderate	16(30)	11(20)	10(19)	8(12)		2(4)	47(16)
	minor	15(28)	18(18)	14(26)	31(47)	3(19)	7(12)	88(29)
	not at all	13(24)	7(13)	29(55)	27(41)	13(81)	45(79)	134(45)
Crop residuals
	major	41(76)	5(9)	1(2)	12(18)	6(38)		65(22)
	moderate	12(22)	28(51)	2(4)	46(70)	10(63)	9(16)	107(36)
	minor		22(40)	48(91)	8(12)		47(83)	125(42)
	not at all	1(2)		2(4)			1(2)	4(1)
Natural pastures (roadside and swamp areas)
	major	49(91)	37(67)	49(93)	51(77)	11(69)	44(77)	241(80)
	moderate	5(9)	16(29)	1(2)	10(15)	5(31)	6(11)	43(14)
	minor		1(2)	2(4)	3(5)		4(7)	10(3)
	not at all		1(2)	1(2)	2(3)		3(5)	7(2)

Conserved feeds (hay + silage)
	major	8(15)						8(6)
	moderate	12(22)		3(6)		1(6)	3(5)	19(6)
	minor	4(7)	3(6)	11(21)	10(15)	6(38)	5(9)	39(13)
	not at all	30(56)	52(95)	39(74)	56(85)	9(56)	49(86)	235(78)
Bought fodder
	major	2(4)	2(4)					4(1)
	moderate	3(6)	1(2)					4(1)
	minor	5(9)	21(38)		2(3)	1(6)	3(5)	32(11)
	not at all	44(82)	31(56)	53(100)	64(97)	15(94)	54(95)	261(87)
Grazing
	major	1(2)		4(8)			16(28)	21(7)
	moderate		2(4)	1(2)			1(2)	4(1)
	minor		33(60)					33(11)
	not at all	53(98)	20(36)	48(91)	66(100)	16(100)	40(70)	243(81)
Cut fodder from outside
	major	51(94)						51(17)
	moderate	3(6)	9(16)					12(4)
	minor		20(36)	3(6)	1(2)		10(18)	34(11)
	not at all		26(47)	50(94)	65(98)	16(100)	47(83)	204(68)
Supplementary feeding (concentrates)
	Yes	54(100)	55(100)	52(98)	61(92)	15(94)	54(95)	291(97)
	No			1(2)	5(8)	1(6)	3(5)	10(3)
For each region, results in bold indicate the category with the highest frequency for that question

(INSERT Figs. 8,9 and 10)

Sources of feed are shown in Table 5 and Fig. 11. Natural pastures (i.e. naturally occurring grasses, legumes, and other species) ranked as the most used source of feed, with 294/301 (98%) respondents using at least some natural pasture, and 241 (80%) using it as the major source of feed. Most respondents gave at least some supplementary concentrate feeds to their cattle; only 10/301(3%) stated that they did not do so at all.

(INSERT Tables 4 and 5)

Some feeds were used to a very limited extent including conserved forage, bought fodder and actual grazing; all of which were used by < 25% of respondents. Across all regions, not at all was the most common response for those feeds (Table 5). Other more commonly used feed sources had regional variations. Cut forage was used by 32% of respondents. Of those, 54/97 (55%) were in Njombe, with most of those (51) reporting that it was a major source of feed (a category only used in Njombe). Fodder plots were used by 55% of respondents, but this varied appreciably by region, with respondents in Njombe and Mbeya having higher odds of reporting that they used fodder plots (at any level) than respondents in Tanga (OR: 3.8, 95%CI: 1.7–8.7, and OR: 8.3, 95%CI: 3.2–21.7, respectively), while respondents in Morogoro were less likely to report that they used fodder plots than those in Tanga (OR: 0.3, 95%CI: 0.1–0.7).

Most respondents used crop residuals (only 4% did not) but the level of use varied markedly by region. Compared to Tanga, the odds of a respondent reporting moderate/major use of crop residuals (as opposed to minor/not at all) were much higher in Njombe and Mbeya (OR: 883, 95%CI: 88.9–8770 and OR: 25, 95%CI: 6.9–90.3, respectively).

(INSERT Fig. 11)

Smallholder dairy farming constraints

Farmers were asked to rank the important farming constraints (Table 6 and Fig. 12), with high costs of inputs ranked as very highly significant by 229/301 (76%) respondents and by most farmers in all the six study regions. Lack of enough land (150/301; 50%) was the second most important constraint, while unavailability of feed (24/301; 41%) was the third.

Table 6. General constraints in smallholder dairy cattle farming across the study regions in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Regions						Total
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n=54 (18)	n=55 (18)	n=53 (18)	n=66 (22)	n=16 (5)	n=57 (19)	n=301 (100)
Lack of enough land
Very highly significant		31(57)	30(55)	11(21)	32(48)	15(94)	31(54)	150(50)
Important		4(7)	3(5)	15(28)	18(27)	1(6)	12(21)	53(18)
Moderate		9(17)	8(15)	6(11)	3(5)		3(5)	29(10)
Minor		4(7)	9(16)	20(38)	11(17)		9(16)	53(18)
Of little importance		6(11)	5(9)	1(2)	2(3)		2(4)	16(5)
Unavailability of feeds
Very highly significant		26(48)	25(46)	5(9)	13(20)	7(44)	14(25)	90(30)
Important		10(17)	14(26)	17(32)	46(70)	9(56)	28(49)	124(41)
Moderate		7(13)	7(13)	9(17)	4(6)		3(5)	30(10)
Minor		5(9)	7(13)	19(36)	3(5)		11(19)	45(15)
Of little importance		6(11)	2(4)	3(6)			1(2)	12(4)
High costs of inputs
Very highly significant		42(78)	52(95)	35(66)	50(76)	13(81)	37(65)	229(76)
Important		4(7)	1(2)	13(25)	15(23)	3(19)	15(26)	51(17)
Moderate		5(9)		5(9)			2(4)	12(4)
Minor		1(2)	2(4)		1(2)		3(5)	7(2)
Of little importance		2(4)						2(1)
Lack of money to buy inputs
Very highly significant		17(32)	23(42)	4(8)	4(6)	1(6)	1(2)	50(17)
Important		13(24)	20(36)	12(23)	20(30)	11(69)	12(21)	88(29)
Moderate		11(20)	3(6)	9(17)	9(14)		3(5)	35(12)
Minor		8(15)	3(6)	25(47)	30(46)	4(25)	38(67)	108(36)
Of little importance		5(9)	6(11)	3(6)	3(5)		3(5)	20(7)
Unpredictable milk market
Very highly significant		33(61)	6(11)	4(8)	20(30)	11(69)	4(7)	78(26)
Important		8(15)	16(29)	10(19)	18(27)	4(25)	7(12)	63(21)
Moderate		4(7)	4(7)		5(8)		5(9)	18(6)
Minor		3(6)	17(31)	23(43)	18(27)	1(6)	19(33)	81(27)
Of little importance		6()	12(22)	16(30)	5(8)		22(39)	61(20)
Unavailability of breeding service
Very highly significant		17(32)	3(6)		2(3)		1(2)	23(8)
Important		14(26)	10(18)	8(15)	3(5)	1(6)	9(16)	45(15)
Moderate		10(19)	6(11)	3(6)	5(8)		1(2)	25(8)
Minor		9(17)	27(49)	26(49)	37(56)	9(56)	20(35)	128(43)
Of little importance		4(8)	9(16)	16(30)	19(29)	6(38)	26(46)	80(27)
Diseases
Very highly significant		20(37)	8(15)	2(4)				30(10)
Important		13(24)	19(35)	24(45)	47(71)	15(94)	39(68)	157(52)
Moderate		10(19)	21(38)	15(28)	5(8)		5(9)	56(19)
Minor		9(17)	5(9)	10(19)	11(17)	1(6)	8(14)	44(15)
Of little importance		2(4)	2(4)	2(4)	3(5)		5(9)	14(5)
For each region, results in bold indicate the category with the highest frequency for that question

For the assessment of regional variations, three ordinal categories were created i.e., High (encompassing very highly significant and important), Moderate and Low (encompassing little importance and minor) and not at all (data from Arusha were excluded from this analysis because of the absence of data in one or more categories). Insufficient land was reported to be an important constraint by farmers in Kilimanjaro (OR: 3.2, 95%CI: 1.5–6.8), Morogoro (3.1, 95%CI: 1.4–7) and Njombe (2.1; 95%CI 1-4.5) than in Tanga. Availability of feed was regarded as of greater importance by farmers from Kilimanjaro (OR: 11.7, 95%CI: 4.6–29.8), Mbeya (OR: 3.4, 95%CI: 1.6–7.3), Morogoro (OR: 3.6, 95%CI: 1.6–7.8) and Njombe (OR: 2.7, 95%CI: 5.9–1.3) than those from Tanga.

Lack of money to buy inputs was regarded as of greater importance in both Mbeya and Njombe (OR 7.32, 95%CI: 3.2–16.6 and 3, 95%CI: 1.4–6.2, respectively) than in Tanga, while the unpredictability of the milk market was regarded as of greater importance in Njombe (OR: 11.4, 95%CI: 0.2–0.7), Kilimanjaro (OR: 4.8, 95%CI: 2.2–10.4) and Mbeya (OR: 2.3, 95%CI: 1–5.2) than in Tanga. In respect to constraints for successful breeding, farmers in Njombe regarded the lack of a breeding service as more important than farmers in Tanga (OR: 9.5, 95%CI: 4.1–22.2).

The importance of disease as a constraint was analysed using a multinomial regression, as the proportional odds assumption was not met (P < 0.001). The odds of being in the Low rather than the High category was similar across regions. For the Moderate category, farmers in Kilimanjaro and Morogoro had lower odds of being in the Moderate rather than the High category when compared to farmers in Tanga (OR: 0.2, 95%CI: 0.1–0.1 and OR: 0.2, 95%CI: 0.1–0.7, respectively).

(INSERT Table 6 and Fig. 12)

Dairy breed selection criteria and breeding practice

Preference for one or more breeds (Fig. 13) was expressed by 160 (53%) respondents, whereas 59 (20%) had no specific preference (Table 7). Farmers in Njombe were more likely to report having a breed preference than farmers in Tanga (OR: 21.5, 95%CI: 4.7–97.6). Of those who expressed a breed preference, the focus was milk production (146/160: 91%), followed by easy handling (108/160: 68%), large body size (100/160: 63%) and ease of getting pregnant (37/160: 60%).

Table 7. Breed preference and dairy cattle selection criteria used, among smallholder dairy cattle farmers across the study regions in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character

Category

Regions

Njombe

Mbeya

Tanga

Kilimanjaro

Arusha

Morogoro

Total

Total respondents

n=54 (18)

n=55 (18)

n=53 (18)

n=66 (22)

n=16 (5)

n=57 (19)

n=301 (100)

Do you have breed preference?

yes

52(96)

34(62)

29(55)

20(30)

4(25)

21(37)

160(53)

1(2)

6(11)

9(17)

18(27)

8(50)

17(30)

59(20)

I do not know

1(2)

15(27)

15(28)

28(42)

4(25)

19(33)

82(27)

Total respondents

n=52(33)

n=34(21)

n=29(18)

n=20(13)

n=4(3)

n=21(13)

n=160(100)

Breed selection criteria

Body size

strongly agree

33(63)

18(53)

9(31)

17(85)

4(100)

19(90)

100(63)

agree

8(15)

13(38)

19(66)

2(10)

44(28)

neutral

4(8)

4(3)

disagree

7(13)

3(9)

1(3)

1(5)

12(8)

Disease resistance

strongly agree

31(60)

5(15)

4(14)

18(86)

58(36)

agree

19(37)

22(65)

20(69)

16(80)

4(100)

2(10)

83(52)

neutral

1(2)

6(18)

2(7)

3(15)

1(5)

13(8)

disagree

1(2)

1(3)

3(10)

1(5)

6(4)

Easy handling

strongly agree

41(79)

27(79)

20(69)

11(55)

2(50)

7(33)

108(68)

agree

8(15)

3(9)

6(21)

9(45)

2(50)

12(57)

40(25)

neutral

2(4)

3(9)

3(10)

2(10)

10(6)

disagree

1(2)

1(3)

2(1)

Milk production

strongly agree

47(90)

30(88)

26(90)

19(95)

4(100)

20(95)

146(91)

agree

4(8)

3(9)

3(10)

1(5)

12(8)

neutral

1(2)

1(3)

2(1)

Easy to conceive

strongly agree

44(85)

10(29)

2(7)

3(15)

1(25)

60(38)

agree

7(13)

18(53)

24(83)

15(75)

3(75)

19(90)

86(54)

neutral

1(2)

6(18)

3(10)

2(10)

14(9)

For each region, results in bold indicate the category with the highest frequency for that question

Table 8. Breeding methods and practice of artificial insemination by the smallholder dairy cattle farmers in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Regions						Total
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n=54 (18)	n=55 (18)	n=53 (18)	n=66 (22)	n=16 (5)	n=57 (19)	n=301 (100)
Breeding method
artificial insemination		17(31)		29(55)	34(52)	10(63)	12(21)	102(34)
	both	33(61)	2(4)	24(45)	32(49)	5(31)	42(74)	138(46)
	natural service	4(7)	53(96)			1(6)	3(5)	61(20)
Total respondents		n=18(10)	n=55(31)	n=24(13)	n=32(18)	n=6(3)	n=45(25)	n=180(100)
Bull sources
Own Bull			2(4)	8(33)	4(13)	1(17)	6(13)	17(9)
Community bulls only			4(7)					4(2)
Own and neighbours’ bulls		3(17)	3(5)			2(33)	28(62)	46(26)
Neighbours’ bulls		15(83)	46(84)	16(67)	28(88)	3(50)	11(24)	113(63)
Total respondents		n=31(14)	n=2(1)	n=53(24)	n=66(30)	n=15(7)	n=54(24)	n=221(100)
How is AI service obtained
*Immediately after calling the AI technician		3(10)		8(15)	20(30)	5(33)	3(6)	39(18)
	**Regularly	16(52)		35(66)	31(47)	10(67)	39(72)	131(59)
Irregular with some interruptions on weekend		12(39)	2(100)	10(19)	15(23)		12(22)	51(23)
AI service Waiting time
	1_to_3_hours	12(39)						12(5)
	4_to_6_hours	3(10)			1(2)	1(7)	1(2)	6(3)
	7_to_9_hours	9(29)		52(98)	53(79)	14(93)	49(91)	177(80)
	10_to_12_hours	7(23)	2(100)	1(2)	12(18)		4(7)	26(12)
NB: For each region, results in bold indicate the category with the highest frequency for that question. Farmer decides insemination time and call the technician at the time of insemination *Farmer calls the technician, explains the signs and duration, the technician decides the insemination time

Table 9. Constraints for successful breeding in smallholder dairy cattle farming across the study regions in Tanzania mainland. Data are shown as number of respondents and percentage of responses within each region n (%).

Character	Category	Regions						Total
Character	Category	Njombe	Mbeya	Tanga	Kilimanjaro	Arusha	Morogoro	Total
Total respondents		n=54 (18)	n=55 (18)	n=53 (18)	n=66 (22)	n=16 (5)	n=57 (19)	n=301 (100)
Unavailability of AI service
	major	19(35)	37(67)	1(2)	4(6)		2(4)	63(21)
	moderate	20(37)	3(6)	2(4)	7(15)		2(4)	34(11)
	minor	10(19)		20(38)	11(12)	1(6)	15(26)	57(19)
	not at all	4(7)		28(53)	44(67)	13(81)	34(60)	123(41)
	never used	1(2)	15(27)	2(4)		2(13)	4(7)	24(8)
Unavailability of breeding bull
	major	11(20)	1(2)	5(9)	2(3)		2(4)	21(7)
	moderate	18(33)	6(11)	6(11)			3(5)	33(11)
	minor	7(13)	27(49)	6(11)	10(15)		7(12)	57(19)
	not at all	12(22)	21(38)	6(11)	24(36)	6(38)	33(58)	102(34)
	never used	6(11)		30(57)	30(46)	10(63)	12(21)	88(29)
High breeding cost
	major	34(63)	20(36)	28(53)	25(38)	6(38)	29(51)	142(47)
	moderate	15(28)	25(46)	22(42)	35(53)	8(50)	18(32)	123(41)
	minor	4(7)	7(13)	1(2)	5(8)	1(6)	6(11)	24(8)
	not at all	1(2)	3(6)	1(2)		1(6)	3(5)	9(3)
	never used			1(2)	1(2)		1(2)	3(1)
Bull located at far distance
	major	19(35)	4(7)	7(13)	2(3)		4(7)	36(12)
	moderate	13(24)	6(11)	4(8)	3(5)	1(6)	5(9)	32(11)
	minor	7(13)	28(51)	6(11)	22(33)	1(6)	10(18)	74(25)
	not at all	9(17)	17(31)	6(11)	8(12)	4(25)	26(46)	70(23)
	never used	6(11)		30(57)	31(47)	10(63)	12(21)	89(30)


Poor heat detection
	major	13(24)	3(6)		7(11)	1(6)	3(5)	27(9)
	moderate	21(39)	21(38)	11(21)	16(24)	2(13)	23(40)	94(31)
	minor	13(24)	21(38)	37(70)	41(62)	11(69)	25(44)	148(49)
	not at all	7(13)	10(18)	5(9)	2(3)	2(13)	6(11)	32(11)
Cows not showing heat
	major	4(7)	1(2)	1(2)	1(2)		1(2)	8(3)
	moderate	27(50)	21(38)	4(8)	8(12)		8(14)	68(23)
	minor	7(13)	18(33)	36(68)	47(71)	12(75)	38(67)	158(53)
	not at all	16(30)	15(27)	12(23)	10(15)	4(25)	10(18)	67(22)
For each region, results in bold indicate the category with the highest frequency for that question

(INSERT Fig. 13)

Breeding methods are summarised in Table 8 and Fig. 14. In general, the most frequently applied breeding method involved a mix of natural service and artificial insemination (AI) (138/301; 46%). At the regional level, AI alone was the most common method in Arusha (10/16; 63%), Kilimanjaro (34/66; 52%) and Tanga (29/53; 55%), while most farmers used a mix of AI and natural service in Morogoro (42/57; 74%) and Njombe (33/64; 61%), and natural service only predominated in Mbeya (53/55; 96%). Relative to Tanga, farmers were less likely to report using AI alone in Morogoro (OR: 0.2, 95%CI: 0.1–0.5) and Njombe (OR: 0.4, 95%CI: 0.2–0.9) (this analysis excluded Mbeya as no farmers in that region reported using AI alone).

(INSERT Fig. 14)

Of the 200 farmers who used natural service on some occasions, 180 answered the question about where they sourced bulls from. Overall, they largely (113/180; 63%) opted to hire bulls from nearby farms, with most of the rest using a combination of their own and their neighbours’ bulls (46/180, 26%). Only 17 (9%) reported exclusively using their own bulls.

Of the 240 farmers who used AI for at least some inseminations, 221 answered questions around AI use. Across all regions, most farmers reported receiving the service regularly (131/221; 59%) and most (179/221; 80%) reported that waiting time from informing the AI technician to the actual insemination ranged between seven and nine hours.

(INSERT Tables 7 and 8)

In response to questions about the constraints around successful breeding (Fig. 15 and Table 9), almost half of the respondents, 142/301(47%), indicated that the high cost of breeding was a major constraint, with 96% (289/301) identifying high breeding costs as being at least a minor constraint. This is a higher proportion than the effects of poor oestrus detection (89%, 269/301), cows not displaying oestrus (79%, 234/301,) and unavailability of AI services (51%, 154/301). Constraints around breeding varied markedly across region. Except for high breeding cost, compared to farmers in Tanga, farmers in Njombe were more likely to report that all the constraints listed in Table 9 were major/moderate constraints on their farm. The relevant OR were 43.3 (95%CI: 11.7–160) for unavailability of AI service, 4.4 (95%CI: 1.9–10.4) for unavailability of breeding bull, 5.6 (95%CI: 2.4–13.1) for bull being located at a distance, 6.5 (95%CI: 2.7–15.4) for poor oestrus detection, and 12.9 (95%CI: 4.5–16.2) for cows not showing heat. For Mbeya, farmers had higher odds (compared to Tanga) of reporting the unavailability of an AI service (OR: 44.4, 95%CI: 12–82.6), poor oestrus detection (OR 3; 95%CI: 1.3–6.9) and cows not showing heat (OR 6.4; 95%CI: 2.2–18.6) as major/moderate constraints. Despite having similar percentages of farmers using mixed breeding (see Table 7), farmers in Kilimanjaro were less likely to report bull availability and bull distance as major/moderate constraints than farmers in Tanga (OR 0.12; 95%CI: 0.03–0.6, and 0.3; 95%CI: 0.1–1, respectively), while farmers in Morogoro were more likely than farmers in Tanga to report that heat detection was a problem (OR 3.2; 95%CI: 1.4–7.5).

(INSERT Fig. 15 and Table 9)

Smallholder dairy cattle household, family, and farm demographics

In order to achieve sustainable development of smallholder dairy cattle farming in Tanzania, a proper understanding of the key participants involved is essential. This includes a thorough knowledge of the features of the farmers and their farms, locality, and farming practices.

Most farms/households (74%) were headed by men. This is consistent with other recent surveys, such as that of Kashoma and Ngou (12) who reported, based on a survey of farms across 17 districts, that 68% of households were headed by men. However, both of these percentages are much lower than previous reports, e.g. the 90% reported by Kivaria et al. (15), in the Dar es Salaam region and the 93% reported by Swai et al. (42) in Kilimanjaro and Arusha. These findings indicate that there has been an increase in the number of women involved in smallholder dairy farming. One positive reason for this increase is the focus of NGOs on empowerment of women-led households. This was most apparent in Njombe, where almost equal numbers of men and women were household heads, consistent with the report by Msangya et al. (26), that 53% of the households in Njombe supported by the Heifer in Project International Trust were led by women. The other, less positive, reason for increased involvement of women in smallholder dairying is likely to be rural-urban migration of men searching for better wages, therefore, by default, leaving women to manage the household and its livestock (35).

Smallholder dairy farming in Tanzania is generally collaborative, with household heads including other household members (spouse and/or children) in decisions related to farming. In most regions, a herd size of 3 to 4 animals was the most common (42). However, in both Tanga and Morogoro, the most common herd size was > 4 animals. The somewhat larger herd size reported in Tanga and Morogoro may reflect, at least in part, the proximity of these regions to Dar es Salaam, which is the biggest milk market in the country. However, the rapid increase in the population in the Dar es Salaam region in the recent past (20) has not been reflected in recent increases in the size of farms in either Morogoro (32) or Tanga (46), so market size is probably not the sole factor. The similarity of farm size across most regions is also reflected in the number of people/farms actively involved in farming, with 1–2 people actively involved on the majority (67%) of farms. Tanga had both larger herds and more active people/farm, with 3–4 people per farm being the most common category, probably reflecting increased involvement of family members as previously reported (41).

Across the six study regions, cash purchase was the principal (66%) source of a household’s first dairy cattle beast, consistent with the 72% reported by Swai et al. (42) in Kilimanjaro/Arusha. The only region where cash purchase was not the principal source was Mbeya, where a gift from a family member or friend was as common as cash purchase (49 vs 45% of respondents, respectively). This process, known as “kufufya” by Nyakyusa speakers, involves an individual giving a heifer or cow (which could be pregnant or non-pregnant) to a relative or friend. When that animal calves, the original owner takes back the milking cow if the newborn calf is female, or if it is a bull calf, the new owner keeps the cow (until it produces a female calf) but shares the milk it produces with the original owner. This process has been popularised using the Kiswahili slogan “Kopa Ng’ombe lipa Ng’ombe” (meaning: borrow a cow, pay a cow). Similar techniques have been used by NGOs (14, 26), with the recipient of a cow raising female offspring before passing on the younger animals to a new family/owner (De Vries (6); “Passing on the Gift”). Likewise, Kopa Ng’ombe lipa Ng’ombe, Passing on the Gift has been a regionally specific process with only Tanga and Mbeya reporting that a significant proportion of respondents got their first cow from an NGO.

Farmers/respondents and attendants/workers demographic characteristics

Most respondents (51%) in this survey were aged between 41 and 60 years, with 39% having < 10 years of farming experience. According to Swai et al. (42), only 11% of their respondents had < 10 years of experience. The higher figures in this survey, even in the regions studied by Swai et al. (42), strongly suggest that there have been a significant number of new entrants into dairy farming over the last 10 years. This appears to be particularly the case in Njombe and Morogoro as both regions had ≥ 50% of respondents with < 10 years’ experience.

As expected, farm assistants were younger (77% were < 30 years of age) and less experienced (83% had < 5 years of experience) than farm owners. For both groups, primary education was the highest education level (55% for farm owners, 66% for assistants) and dairy farming was a full-time occupation (77 vs 87%, respectively). Having a majority of smallholder farmers and assistants with at least primary education signifies that they are more likely to adopt and apply modern farming techniques for improving their productivity.

Smallholder dairy cattle households’ sources of income

Across all respondents, very few (6%) relied solely on dairy cattle for their income. However, even for those who relied on other household income sources (i.e., employment, crop farming and business), the majority (84%) of them still depended on smallholder dairying as their major income source (with exception of the Njombe region). According to Swai et al. (42) dairying was the major source of income for only 32% of respondents in Kilimanjaro and Arusha regions, while Kashoma and Ngou (12) reported 56.4%. This shows an increase in the number of households relying on smallholder dairying year-by-year. This trend can be expected to result in a stable household economy, providing sustainable income and improved nutrition. Furthermore, it indicates a significant growth in the sector, leading to an improved food security, economic development and increased employment opportunities for many Tanzanians. Therefore, this study suggests for the government and other stakeholders to support smallholder dairy cattle farmers in achieving increased and sustainable productivity.

Smallholder dairy cattle grazing system and feed sources

There were marked differences across regions in the source of non-dairy income. In Morogoro, 64% of respondents had moderate/major income from employment. This may be related to the proximity of institutions like Sokoine University of Agriculture and access to government-related jobs. Gillah et al. (7) reported 19% of farmers in Morogoro were government employees and 15% were retired officers. For the remaining regions, crop farming was the principal alternative source of income. In most regions it was listed as a moderate/minor source of income by respondents (64–84% depending on region); however, in Njombe it was a major source of income for 52% of respondents. This indicates that smallholder dairy cattle farming is expanding beyond urban and peri-urban areas, traditionally managed by government officers, to rural areas, where crop farmers are increasingly adopting these practices. Involvement of crop farmers in smallholder dairy cattle farming was also recorded in the previous studies (42, 12).

Zero grazing was the predominant system in all study regions (67 to 99% of respondents) except for Mbeya (only 16% of respondents). Zero grazing or “cut-and carry” uses natural pastures, with fresh forages being obtained from roadsides, highways, and non-cultivated areas such as communal lands, uncultivated fields, communal swamp areas and communal grazing areas (44, 43). Tethering was the predominant system in Mbeya (68% of farmers) which may reflect the longer periods in the Mbeya region (Busokelo district in particular) when grazing is available, inasmuch as the area is a cooler highland area (770 to 2865 metres above sea level) which receives 1500 to 2700 mm rainfall per year (33, 21). The current survey evaluated the use of eight on-farm feed sources. The survey identified five types of feed source: 1) commonly used feed sources with limited variability across regions (i.e. concentrates, and natural pastures); 2) commonly used feed sources with large differences in level of use by region (crop residuals); 3) rarely used feed sources with > 75% farms reporting no use and > 90% of farms no more than minor use (i.e. grazing, bought fodder and conserved feeds); 4) feed sources that are rarely used in most regions, but commonly used in one or more regions (e.g. cut fodder from outside); and 5) feed sources that vary markedly across and within regions (e.g. fodder plots in farm). These regional differences in feeding practices need to be considered when developing support programs. Furthermore, these results also show that it would be too simplistic to assume that all farms within a region use the same feed sources at the same time and in the same way, such that, even at the regional level, feed supply will require multifactorial solutions to achieve success.

The use of crop residuals as livestock feed could have many benefits: cost savings (crop residuals are more affordable), availability (mostly are locally available) and greater utilization of agricultural by-products. Such practice can help to guarantee sustainable agricultural practices and more efficient resource management within the community of smallholder dairy cattle farming in countries like Tanzania.

Dairy breed selection criteria and breeding practice

Only 53% of farmers across the six regions had a breed preference. Surprisingly, the proportion of farmers from Njombe who had a preference was much higher than this average (at 96%). The reason for this difference is unclear. Despite the low proportion of respondents who had a specific breed preference, most respondents (91%) expressed their opinions on the cattle attributes they desire. Unsurprisingly, high milk production being the primary reason for their choices, as highlighted in the previous surveys done in Tanzania (43, 9, 4). This preference (and the preference for large body size especially in Morogoro, Tanga and Mbeya) probably reflects the marketing of “high yielding” cows to smallholders rather than direct experience of the respondents of such cows playing an important role in improving farm incomes (5).

Breeding method was very dependent on the region, reflecting the generally poor AI infrastructure across much of Tanzania and the centralisation of the AI service at the National Artificial Insemination Centre (NAIC) in Arusha. Regions which predominantly used AI were generally close to the NAIC, with regions further away using a combination of natural insemination and AI. Nevertheless, across the six regions, 80% of farmers used at least some AI. For five of the six regions, > 90% of farmers reported using some AI, with only Mbeya having majority of farmers using bulls only (96%). Centralization of AI centre in Tanzania could be the reason for irregular availability of semen and other consumables like liquid nitrogen to farmers located away from NAIC. Similarly, as reported previously by Mwanga et al. (28) in their study on factors influencing breeding decisions by smallholder dairy farmers in Sub- Saharan African countries. Also, irregularities in supply of IA consumables by Kashoma and Ngou (12). Other report (10), over 60% of farmers in Morogoro and Tanga reported to experience challenges related to AI delivery system. Though there are some imported doses of semen by the private organizations (13, 34), but plans are underway to improve the available semen production by involving the public-private partnership and development partners (34). Further, Kusiluka et al. (18) reported lack of breeding services to smallholder dairy cattle farmers in the eastern zone of Tanzania.

The cost of AI varies with distance from the NAIC. The proximity to the AI centre means that AI services can be offered to farmers at a relatively low cost (i.e. between 15,000 and 25,000 TZs per insemination (~ US$6–10) for first insemination). Nevertheless, irrespective of distance, AI was seen by many respondents as being expensive. The lack of a good AI infrastructure in Tanzania probably limits productivity on dairy farms, inasmuch as AI confers significant advantages over natural mating in terms of genetic gain and disease control (36). Increased use of AI could be greatly beneficial for smallholder productivity, reducing the risk of disease from the use of untested local bulls and improving production efficiency and longevity, especially if the focus is not just on increasing milk production. Furthermore, to sustain productivity, emphasis must be placed on improving management and nutrition of the animals. This is because inadequate nutrition and management of animals may result in reduced milk yields, increased health disorders and most importantly, impaired reproductive ability (45). The high cost and unreliability of AI service were among the constraints highlighted in the previous studies from Tanzania (10) and in sub-Saharan African countries (28).

Smallholder dairy cattle farming constraints

Two areas of constraints were examined: general constraints and specific breeding-related constraints. High costs of inputs were identified as the leading constraint in the general constraints section of the survey, with 76% of all farmers reporting it as a highly significant constraint (range: 65–95% by region). In general, high input costs seemed to be perceived as a more important constraint than high breeding costs alone. In all regions, the proportion of farmers who reported high input costs as a highly significant constraint was higher than the proportion who reported high breeding costs as a major constraint. High cost of inputs was also reported previously: concentrates and veterinary drugs (15) in Dar es Salaam and in northern Tanzania (42). Furthermore, in all regions, the most common response in relation to high input costs was that it was a highly significant constraint, whereas, for breeding costs, the most common response varied between major and moderate, depending on region. Insufficient land and feed were also seen as highly significant/important constraints by most respondents, except in Tanga, where most respondents (51% and 59%, respectively) thought that they were of minor to moderate importance. This regional difference may reflect the high number of swamp areas and abandoned sisal plantations in the Tanga region areas from which farmers can reliably obtain forage.

The cost of breeding was seen as a major constraint by a high proportion of respondents in all regions (> 80%), with no evidence that being near the NAIC made that cost less of a constraint. This was also highlighted previously that farmers are unable to pay for the AI service due to high cost and irregularities in AI delivery system (12). Irregularities in the AI delivery system was also reported to be a challenge for smallholder dairying improvement in Rwanda (38). In contrast, distance from the NAIC did affect whether the lack of an AI service was an important constraint, with the regions close to Arusha reporting no or minor concerns and the regions far from Arusha reporting major/moderate concerns. Indeed, concerns about reproductive constraints were particularly prominent in Njombe, with all constraints (other than cost) being ranked as more important in Njombe than in Tanga. The pattern for Mbeya was similar but the differences were less marked. This probably reflects the different pattern of breeding in the two regions – natural service only predominates in Mbeya, while mixed and AI only predominate in Njombe. More reliance on AI leads to more concerns – again highlighting the importance of improving the AI infrastructure in dairying regions remote from Arusha.

The regional differences in the relative importance of constraints again highlights the importance of developing regional solutions to improve smallholder dairy farm productivity, while the variability in the relative importance of constraints across farms within a region highlights the importance of understanding what is driving those differences at the farm level. Previous studies on smallholder dairying constraints in Tanzania include inadequate, seasonal and irregular forage availability in Arusha, Kilimanjaro (42) and Dar es Salaam (15). Also, the high cost and poor conception/pregnancy rates achieved by AI (10) was considered by farmers from Tanga and Morogoro as being a significant disincentive to breed by AI.

This is the first large-scale survey of demographics, management system and constraints of smallholder dairy farms across the whole of Tanzania. It has identified that smallholder dairy cattle farmers differ in various aspects from each other and most importantly from one region to the other. The typical unique characteristics of smallholder dairy cattle farmers are based on their locality. Primary differences include gender, age, and experience of the farmers among the study regions, which, in turn, reflect the structure of the dairy industry and the proximity of alternative income streams between regions. Understanding these differences will help the implementers to recognise what is really preferred by the farmers and not thinking that whatever is planned “ought to” suit the farmers, as this may not be the case. For example, another study (4) on Tanzanian smallholder farmers’ preferred traits of animals, was primarily determined by agro-ecological zones and production systems that had been instigated as a result of differences in climatic conditions and feed sources. Another study in Kenya (11) on breeding objectives for sustainable smallholder dairy farming highlighted the importance of considering differences in production systems, as failure to do so may result in failure of the program.

Nonetheless, the present study highlights key areas that require attention within the specific agro-economic climate of regions in Tanzania, which should at least provide a basis for developing well-founded schemes for the development of the dairy industry. Thus, sustainably developing a smallholder dairy farming industry in Tanzania needs to take account of these regional differences and provide targeted, flexible region-specific solutions and not the ‘one size fits all’ solutions. Therefore, there is a need to identify the nature of the farmers first before implementation of any ‘herd-improvement’ programs, especially those based upon one particular set of assumptions. This could be the major reason for the success or failure of such initiatives established by the government, NGOs or other stakeholders. Further qualitative studies should be undertaken to gather further information on the reasons for these differences, which was not featured in the present study.

Ltd – Limited
NBS - National Bureau of Statistics
OCGS - Office of the Chief Government Statistician
MoLF – Ministry of Livestock and Fisheries
TALIRI - Tanzania Livestock Research Institute
DC - District Council
TC - Town Council
MC - Municipal Council
CC - City Council
USA – United States of America
IMB - International Business Machines
CI - Confidence Interval
OR - Odds Ratio

Ethics approval and consent to participate

Consent for publication

Authors consent to the publication of this article and participants permission for publication was achieved where appropriate.

Availability of data and materials

The generated and analysed dataset used in this study are available from the corresponding author upon acceptable request.

Competing interests

The authors declare that they have no competing interests.

Funding

This work was supported by the Massey University Foundation Scholarship

Authors' contributions

All authors contributed to the study design and conception. Data collection was done by Athans Ngou. Data analysis and interpretation were done by Athanas Ngou, Richard Laven and Timothy Parkinson. The first draft of this manuscript was written by Athanas Ngou, and all authors commented on the versions of the manuscript. All authors read and approved the final document.

Acknowledgements

The authors would like to thank Massey University, New Zealand for their invaluable assistance and support during the research process. Also, we acknowledge the support obtained during data collection from the Sokoine University, Tanzania; the regional, district and village veterinary/livestock officers and smallholder dairy cattle farmers from the visited study sites.

Footnotes

Ngou et al.

Other Important Information

This manuscript is submitted as free article i.e., no Article Processing Charge (no APC) in response to email I received.

Alonso S, Toye PG, Msalya G, Grace D, Unger F. Smallholder dairy farming in Tanzania: Farming practices, animal health and public health challenges and opportunities. 2014.
Brett C. A new way to boost smallholder dairy productivity in Tanzania. worldbank org/voices. 2019. https://blogs.worldbank.org/en/voices/new-way-boost-smallholder-dairy-productivity-tanzania. Accessed 16 August 2024
Chagunda MG, Bruns E, King JM, Wollny C. Evaluation of the breeding strategy for milk yield of Holstein Friesian cows on large-scale dairy farms in Malawi. The Journal of Agricultural Science. 2004;142(5):595-601.
Chawala AR, Banos G, Peters A, Chagunda MGG. Farmer-preferred traits in smallholder dairy farming systems in Tanzania. Tropical Animal Health and Production. 2019;51(6):1337-44.
Chung YB. Historicising sustainable livestock intensification and animal genetic improvement in Africa: towards a decolonial multispecies climate justice. The Journal of Peasant Studies. 2024;51(3):674-94.
De Vries J. Passing on the gift as an approach to sustainable development programmes. Development in Practice. 2012;22(3):373-84.
Gillah K, Kifaro G, Madsen J. Management and production levels of cross-bred dairy cattle in Dar es Salaam and Morogoro urban and peri urban areas. Livestock Research for Rural Development. 2013;25(9).
Gillah K, Kifaro G, Madsen J. Urban and peri urban dairy farming in East Africa: A review on production levels, constraints and opportunities. Livestock Research for Rural Development. 2012;24(11):198.
Gillah KA, Kifaro GC, Madsen J. Effects of management practices on yield and quality of milk from smallholder dairy units in urban and peri-urban Morogoro, Tanzania. Tropical animal health and production. 2014;46:1177-83.
Kanuya N, Mlangwa J, Shoo J. Constraints associated with the use of Artificial Insemination (AI) and Natural Service (NS) in smallholder dairy herds in Tanga and Morogoro, Tanzania. Tanzania Veterinary Journal. 2014;29(1):9-20.
Kariuki C, van Arendonk J, Kahi A, Komen H. Deterministic simulations to determine the impacts of economic and non-economic breeding objectives on sustainable intensification of developing smallholder dairy farms. Livestock Science. 2019;226:7-12.
Kashoma I, Ngou A. Insight into the adoption and success of artificial insemination services in smallholder dairy farming systems: A cross-sectional study. Tanzania Veterinary Journal. 2023;38(1).
Katjiuongua HN, S. Tanzania smallholder dairy value chain development: situation analysis and trends. Noairobi, Kenya: International Livestock Research Institute (ILRI); 2014. https://cgspace.cgiar.org/server/api/core/bitstreams/3b2a4eaf-8c8e-4030-8df5-a7c2364b68a9/content. Accessed 15 Jan 2024.
Kayunze K, Urassa J, Mwakalobo A, editors. Food status and income levels among dairy and non-dairy farmers in Chunya district, Tanzania. Proc Of the 28th Scientific Conference of Tanzania Society of Animal Production (TSAP), 28, pp11-17; 2001.
Kivaria F, Noordhuizen J, Kapaga A. Prospects and constraints of smallholder dairy husbandry in the Dar es Salaam region, Tanzania. Outlook on Agriculture. 2006b;35(3):209-15.
Kurwijila L, Boki K. A review of the small-scale dairy sector-Tanzania. Milk and dairy products, post-harvest losses and food safety in Sub-Saharan Africa and the near East. FAO prevention of Food Losses Programme. 2003.
Kurwijila L, Henriksen J, Aboud A, Kifaro G. Strategies for market orientation of small-scale milk producers and their organisations: proceedings of a workshop held in Morogoro, Tanzania, 20-24 March, 1995. 1995.
Kusiluka L, Karimuribo E, Mdegela R, Kampaga A, Bundala A, Kivaria F, et al. Constraints associated with production in smallholder dairy farms in the eastern coastal zone of Tanzania. Bulletin of animal health and production in Africa. 2006;54(4):286-96.
Limbu F. Agricultural technology economic viability and poverty alleviation in Tanzania. Presented at the structural Transformation Policy Workshop held at Serena Hotel, Nairobi Kenya. 1999:27-30.
Lupala J. Exploring unbalanced urban spatial expansion in sprawling cities. Case study of Kimara Matangini, Kibululu and Dovya settlements in Dar es Salaam City, Tanzania. Central European Journal of Geography and Sustainable Development. 2021;3(2):62-84.
Makala A. Mbeya region investment opportunities. In: Presidents' Office RAaLG, editor. Mbeya Region: Regional Commissioner, Mbeya; 2017. https://mbeya.go.tz/storage/app/media/uploaded-files/mbeya%20region.pdf. Accessed 15 Jan 2024.
Mbwambo N, Nigussie K, Stapleton J. Dairy development in the Tanzanian livestock master plan. Gates Open Res. 2019;3(628):628.
McDermott JJ, Staal SJ, Freeman HA, Herrero M, Van de Steeg J. Sustaining intensification of smallholder livestock systems in the tropics. Livestock science. 2010;130(1-3):95-109.
Mdoe N, Wiggins S. Returns to smallholder dairying in the Kilimanjaro region, Tanzania. Agricultural Economics. 1997;17(1):75-87.
MoLF. Livestock Sector Transformation Plan (LSTP) 2022/23-2026/27. In: Ministry of Livestock and Fisheries URoT, editor. 2022.
Msangya M, Urassa J, Mahonge C. Contribution of Heifer In-Trust schemes to smallholder households’ well-being in Highland area of Njombe Region, Tanzania. Livestock and rural research and development. 2015;26(12).
Mwambene P, Chawala A, Illatsia E, Das S, Tungu B, Loina R. Selecting indigenous cattle populations for improving dairy production in the Southern Highlands and Eastern Tanzania. Livestock Research for Rural Development. 2014;26(3).
Mwanga G, Mujibi F, Yonah Z, Chagunda M. Multi-country investigation of factors influencing breeding decisions by smallholder dairy farmers in sub-Saharan Africa. Tropical animal health and production. 2019;51:395-409.
NBS, OCGS. National Sample Census of Agriculture 2019/20. Tanzania: National Bureau of Statistics (NBS) and Office of the Chief Government Statistician (OCGS); 2021. https://www.nbs.go.tz/nbs/takwimu/Agriculture/2019-20_Agri_Census_%20Main_Report.pdf. Accessed 15 Jan 2024
Nell AJ, Schiere H, Bol S. Quick scan dairy sector Tanzania: Dutch Ministry of Economic Affairs; 2014.
Njombe AP, Msanga Y, Mbwambo N, Makembe N. The Tanzania dairy industry: Status, opportunities and prospects. In Ministry of Livestock and Fisheries Development. African Dairy Conference and Exhibition, Dar es Salaam, Tanzania 2011 May 25.
Nkya R, Aboud A, Kessy B. Improving the productivity of smallholder dairy cattle in peri-urban Morogoro, United Republic of Tanzania. IAEA-TECDOC--1102; 1999.
Nyunza G, Mwakaje AG. Analysis of Round Potato Marketing in Tanzania. International Journal of Business and Social Science. 2012;3(23).
Ogutu C, Kurwijila LR, Omore AO. Review of successes and failures of dairy value chain development interventions in Tanzania. Nairobi, Kenya: International Livestock Research Institute (ILRI); 2014. https://cgspace.cgiar.org/server/api/core/bitstreams/e23e3576-8cc9-4adf-a6a4-3c4740094b8c/content. Accessed July 15, 2024.
Ojango JM, Mrode R, Okeyo A, Rege J, Emerge-Africa K, Chagunda M. Improving smallholder dairy farming in Africa. Achieving sustainable production of milk. 2. 1st ed: Burleigh Dodds Science Publishing; 2017. p. 371-96.
Parkinson TJ, Morrell JM. Advantages and disadvantages of artificial insemination. In: Noakes DE, Parkinson, T. J., & England, G. C., editor. Veterinary Reproduction and Obstetrics. 10th ed. Philadelphia, PA: WB Saunders; 2019. p. 746-77.
Pasape L. A Review of Success Factors Behind Community Action Research Program (CARP): A Case of Experiences From Smallholder Dairy Farmers of Lushoto in Tanzania. Journal of Participatory Research Methods. 2022;3(2).
Rugwiro P, Pascal N, Manirahaba E, Aphrodis T, Abijuru J, Nizeyimana B, et al. Assessment of challenges associated with artificial insemination service delivery in dairy cattle in Rwanda. International Journal of Veterinary Sciences and Animal Husbandry. 2021;6:45-57.
Suleiman T, Mdegela R, Karimuribo E. Characteristics of dairy farming and its effect on milk production: a case study of Unguja island of Zanzibar, Tanzania. Livestock Research for Rural Development. 2016;28(10):174.
Sumberg J. Policy, milk and the Dar es Salaam peri-urban zone: a new future for an old development theme? Land Use Policy. 1997;14(4):277-93.
Swai E, Karimuribo E, Schoonman L, French N, Fitzpatrick J, Kambarage D, Bryant M. Description, socio-economic characteristics, disease managements and mortality dynamics in smallholder's dairy production system in coastal humid region of Tanga, Tanzania. Livestock Research for Rural Development. 2005a;17(4).
Swai E, Mollel P, Malima A. Some factors associated with poor reproductive performance in smallholder dairy cows: the case of Hai and Meru districts, Northern Tanzania. Livestock Research for Rural Development. 2014;26(6):1-2.
Swai ES, Karimuribo ED. Smallholder Dairy Farming in Tanzania. Outlook on Agriculture. 2011;40(1):21-7.
Urio NA. On-farm research on utilization of crop residues by smallholder dairy farmers in Hai district, Tanzania. Towards Optimal Feeding of Agricultural Byproducts to Livestock in Africa. 1986:17.
Van Saun RJ. Dry Cow Nutrition: The Key to Improving Fresh Cow Performance. Veterinary Clinics of North America: Food Animal Practice. 1991;7(2):599-620.
Zylstra L, Lyimo C, Rutamu I, editors. Milk production and Marketing in Tanga Region: Efficiency of farmer co-operatives versus private sector. Conference Proceedings presented at the Workshop on Strategies for Market Orientation of Small-Scale Milk Producers and their Organizations; 1995; Morogoro (Tanzania), p20-24, March 1995. https://www.fao.org/4/X5661E/x5661e0f.htm. Accessed July 15, 2024.

No competing interests reported.

Download PDF

Editor assigned by journal
29 Aug, 2024
Submission checks completed at journal
29 Aug, 2024
First submitted to journal
28 Aug, 2024

You are reading this latest preprint version

A survey of Smallholder dairy cattle farmers in Tanzania: Farmer Demographic Characteristics and Basic Management Constraints

Status:

Version 1

Abstract

Background

Results

Conclusion

Figures

Introduction/background

Methods

Ethical considerations and approval

Field data collection

Data management and analysis

Results

Smallholder dairy farmers’ household, family, and farm demographics

Farmers/respondents and assistants/workers demographic characteristics

Smallholder dairy farmers’ sources of household income

Smallholder farmers’ dairy cattle management system and feed sources

Smallholder dairy farming constraints

Dairy breed selection criteria and breeding practice

Discussion

Smallholder dairy cattle household, family, and farm demographics

Farmers/respondents and attendants/workers demographic characteristics

Smallholder dairy cattle households’ sources of income

Smallholder dairy cattle grazing system and feed sources

Dairy breed selection criteria and breeding practice

Smallholder dairy cattle farming constraints

Conclusions

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1