Multiple Linear Regression Analysis of Marketable Supply of Beef Cattle in Kaffa Zone Southern Ethiopia

The share of meat and other slaughter by-products exported from the overall export commodities is not more than 2% and meat consumption is below the average of Sub-Sahara African countries. So, marketable supply of beef cattle have important role in making better income for smallholder farmers. As a result the study was aimed to investigate the impediment that affects beef cattle marketable supply in Gimbo district. Two- stage sampling technique was employed to pick up 196 target sample respondents from the sample frame in the study area. Multiple linear regression models were used for analysis. As the model result indicates: marketable supply of beef cattle were affected positively by:  number of livestock fattened, sex of household head, perception of current price, access to credit , area of land allocated for grazing ,access to market information and educational status. So, improvement of fattening management systems should be taken in to consideration to increase marketable supply of beef cattle in the study area as well as at country level. Keywords : marketable supply, beef cattle, multiple linear regression models, Gimbo, Ethiopia DOI: 10.7176/FSQM/98-02 Publication date: June 30 th 2020

peoples regional state of Ethiopia, part of Keffa zone. The district is located at 440 kilometer from Addis Ababa and 9km from Bonga. The district bordered in south by Decha district, in west by Chena, in north by Gojeb River and in east by Menjiwo. The elevation of this district ranges from 3300 to 4300 meter above the Sea level. The total population of the district is 89892 of whom 44774 are men and 45118 women (CSA, 2014) Ethiopian population census. 81.32% the population were orthodox Christians, 13.67% were catholic and 2.13% were Muslims. In other ways 85% of the populations are rural dwellers' and 15%' are urban dwellers' (GDO, 2019).

Sampling techniques and sample Size
Two-stage sampling technique was employed for this study. In the first stage thirty five smallest administrative units were purposely selected based on livestock production potential among thirty nine smallest administrative units (kebeles) in Gimbo district. In second stage five kebeles were selected by simple random sample methods from 35 rural kebeles. Finally 196 sample sizes were determined by using Cochran's (1963)

Data collection and analysis
Primary and secondary data were used. Descriptive statistics and multiple linear regression models were employed for analysis and STATA 14 was used as a tool of analysis.

Multiple linear regression model specification
Regression of independent variables on one dependent variable is called multiple regression model (Kaya & Guler, 2013). Rather than modeling as a simple regression, better to model dependent variable as a function of several explanatory variables (Gemechu et al., 2020). Multiple linear regression model (MLRM) is selected for its simplicity and practical applicability (Greene, 2003 andDessie et al., 2019). Moreover this model is used to test both economic theories and non-experimental data because it can accommodate many cause variables which may be correlated (Maddala and Lahiri, 1992). Unlike simple regression analysis, multiple linear regression analysis is more responsive to ceteris paribus analysis because it allows us to explicitly control many other factors which simultaneously affect effect variable .So, multiple linear regression models were used to analyze factors affecting beef cattle marketable supply in Gimbo district for this study. The functional relationship of MLRM: = ( 1, 2, 3, … , , ) …………………… (1) Where y =beef cattle marketable supply, xn = cause variables and εm = error terms. MLRM: = + !1 1 + !2 2 + !3 3 + ⋯ + ! + ᴇ …………………….......... (2) Where y = effect variable, Xn = cause variables, = constant, βn = parameters associated with cause variable and εk = error terms.

Diagnosis tests
Before multiple linear regression models were analyzed: multicollinearity tests were conducted to identify the variables that are highly correlated. As (Gujarati, 2004) variance inflation factor (VIF) is used to check multicollinearity among continuous variables before including variables in the model. As a rule of thumb, if the value of VIF is greater than 10, the variables are said to be highly collinear. Likewise, the multicollinearity between discrete variables can be calculated using contingency coefficient (CC). The value ranges between 0 and 1, 0 indicating no association between the variables and value close to 1 indicating a high degree of association between variables. As a rule of thumb, if the value of CC is greater than 0.75, the variables are said to be collinear.
3 Results and discussions 3.1. Descriptive statistics results for continues variables The mean age of household head is about 46 years with standard deviation 11.2 meaning majority of them were productive age Table 2. The average family sizes for livestock producers were 5 persons per household almost similar to the national average which is 5.1 per household (CSA, 2013). So, family size is a peculiar trait in rural societies of many developing countries such as Ethiopia (Shumetie and Mamo, 2019). The average educational status of respondents was 4.58 with standard deviation of 4.29. Thus educations contribute to improve beef cattle marketable supply. The average grazing land of sampled farmers is found to be 4.3 hectare which is greater than the national average i.e 1.37 hectare but it varies from place to place as (CSA, 2013). The average of livestock holding per household was 6.61 with standard deviation of 3.96 in the study area. Among 196 target respondents: 139 were male and the remaining 57 were female headed household Table  3. The marital statuses of sample respondents were 84.6% married, 15.4% single. From 196 sample respondents' 81.5% access to credit and 18.5% not access to credit service. Moreover 90.8% of the respondents' access market information while 9.2% not access market information.   Land allocated for grazing (LAFG): the estimated coefficient of this variable was found to be affect marketable supply positively at 5% significance level. If land allocated for grazing increase by one hectare, number of beef cattle supplied to the market increase by 0.7488. This indicates farmers, who have more farm size most likely to supply beef cattle. This result go with (Dinku et al., 2019) farmers who have more farm size, are most likely to produce sesame, in turn increase marketable supply.
Perception of current price (PCP): this variable affects marketable supply of beef cattle positively at 1% significance level. This implies that, if the price of beef cattle increases by one unit, beef cattle marketable supply increased by 0.6692. This is in line with the results of previous studies by (Dinku et al., 2019) price of paddy received by farmers affects marketable surplus of crops positively.
Access to market information system (AMI): the result of estimated coefficient of this variable affects marketable supply of beef cattle positively at 5% level of significance. This means as the farmer access to market information increase by one unit, marketable supply of beef cattle increased by 0.6985. This result goes with (Gemechu et.al, 2020) factor affecting beef cattle value chain analysis.
Access to credit (ATC): beef cattle marketable supply affected positively at 1% significance level. Being access to credit, increase marketable supply by 1.022. This study in line with (Gemechu et al., 2020) factor affecting beef cattle value chain analysis.

Conclusion and recommendations
As multiple linear regression model result indicate: marketable supply of beef cattle was affected positively by: number of livestock fattened, sex of household head, perception of current price, access to credit, and area of land allocated for grazing, access to market information and educational status. As a result, those factors have to be promoted by concerned body through facilitation of agricultural inputs including credit service, education, and extension service to increase beef cattle marketable supply in the study area. The result recommended that: Food Science and Quality Management www.iiste.org ISSN 2224-6088 (Paper) ISSN 2225-0557 (Online) Vol.98, 2020 development of fattening management systems should be taken in to consideration by development agents, governmental and non-governmental organizations', to create awareness for beef cattle production and marketable supply of beef cattle sectors. Abbreviations: CSA: Central Statics Agency, GDO: Gimbo District Office, MLRM: Multiple Linear Regression Model Availability of data The raw data which confirm the output of this study can be found from the author based on permissible request.

Competing interests
There are no competing interests. Funding Authors have not gotten direct fund for this research work.

Authors' contributions
Plan, analysis and revision of this paper were done by this author.