Analysis of Influenced Factors on Consumer Behaviour: The Demand of Chicken Egg’s in TebingTinggi City

The purpose of this study were to identify factors that affect the demand for chicken eggs and to find out the correlation of factors that affect consumer behavior in purchasing chicken eggs in Tebing Tinggi City. The research used primary data obtained from questionnaires or interviews. The research method was multiple linear regression methods, spearman rank correlation and contingency correlation. The results showed that the price of chicken eggs and the income had real effect to the demand for chicken eggs in the area of research. The result of spearman rank correlation and the contingency correlation showed that there was a real correlation between the income with a consumer behavior in purchasing chicken eggs, while the age, employment, and number of consumer dependents, did not show real correlation with consumer behavior in purchasing chicken eggs.


Introduction
Broiler chicken eggs are one of the popular sources of animal protein in the public. Broiler chicken eggs are relatively cheap and easy to obtain. Due to these factors, almost all people can consume broiler chicken eggs to meet the expected nutritional needs (animal protein). (Lestari, 2009) [1]. The average consumption of broiler chicken eggs per capita per year in Indonesia has increased by 97,398 eggs in 2015 and 99,796 eggs in 2016 (Directorate General of PKH, 2016) [2]. Based on data from the Central Statistics Agency (BPS) of Tebing Tinggi City, in 2016, Tebing Tinggi had a population of 158,902 people. Chicken egg production in the year of 2011 to 2015 sequentially was 343,719 eggs, 360,904 eggs, 362,708 eggs, 371,775 eggs, and 360,105 eggs. The percentage of the population of Tebing Tinggi city with annual production of broiler chicken eggs is 11.3% [3].

Time and Location of Research
The research was conducted in Tebing Tinggi City and began in July 2018.

Sampling Methods
The sample determining method is the trace (Accidental) method, which is done by taking random respondents who happen to buy chicken eggs in several traditional markets in Tebing Tinggi City. Those markets are Gambir market, Inpres market, and Sakti market. The requirements for the sample are as follows: the sample is a consumer who shops for chicken eggs. Determination of the sample in this research are all consumers who bought chicken eggs found in traditional markets in Tebing Tinggi City. According to the Bailey theory, the minimum number of samples for researches using statistical analytics is 30 samples (Hasan, 2002) [4].

Method of collecting
Data in this research is collected using primary and secondary data. Primary data is obtained by direct interview with sample consumers using customized questionnaire. Secondary data is obtained from various literatures, books or journals, related institutions, and also previous researches related to consumers behaviour towards demand for broiler chicken eggs.

Data Analysis Method
Collected data is then analyzed using the hypothesis that is going to be tested: Y = a + b 1 X 1 + b 2 X 2 + b 3 X 3 + b 4

Partial Test (t-test)
t-test is statistical test to see whether each independent variable has effect on control variables. The significance level (α) used is 5%.

Classical Assumption Test
This test aims to test whether the model that is going to be used is free from any classical assumption deviation. OLS (Ordinary Least Square) method is used for this purpose. Therefore, several assumptions must be fulfilled such as:

Multicollinearity Test
Multicollinearity test aims to test whether regression model finds correlation between independent variables. Good regression model should not have correlation between independent variables. If there is correlation between independent variables, then the variables are not orthogonal. (Ghozali, 2005) [7].

Heteroskedasticity Test
Heteroskedasticity test aims to test whether in regression model there is varian inequality from one observation residual to the other. Residual varian that constant from one observation to the other is homoscedasticity, otherwise it is heteroskedasticity.

Assumption of Normality Test
Normality test is used to test assumption on data normality. Good regression model residual which has normal distribution. Normality test aims to test whether in regression model, disturbance variable or residual has normal distribution (Sujarweni, 2015).

Spearman's Rank Correlation Coefficient Test
Statistical data analysis method used in this research is Spearman's Rank correlation method. Jonathan and Elly (2010) stated that Spearman's Rank correlation is used to identify correlation or effect between two variables in ordinal scale, they are independent and dependent variable. The formula of correlation analysis is as follows (Sugiyono, 2013) [9].
Annotation: Rs = Spearman's rank correlation value di = Difference between ranks n = Amount of sample ∑ = Summation notation 2.13 Contingency Correlation Coefficient Test Contingency coefficient is a method used to measure relation (assosiation or correlation) between 2 variables in nominal data type (categorical) (Sugiyono, 2003) [10].

Results and Discussion
Factors Affecting the Demand on Broiler Chicken Egg Source: processed Primary Data Table 11 shows Tolerance's value from all independent variables are larger than 0,1 and VIF's value from all independent variables are smaller than 10. From this result, it can be concluded that there is no multicollinearity violation in the model being used.

Classical Assumption Test
3.2 Heteroskedasticity test is performed using Glejser test. Test result can be seen in Table 12.  Table 12 shows that there is no violation on heteroskedasticity assumption. Using classical assumption that has been performed, it can be concluded that the model being used falls in good model category because it fulfills predetermined assumptions.

Goodness of Fit Test and Hypothetical Test
Factors which affect demand on broiler chicken eggs can be seen in Table 13.  Table 13 shows the value of F hitung is 40,61. This means that variables of price of broiler chicken eggs, income, number of dependants, and price of tempeh is simultaneously and distinctively affecting demand on broiler chicken eggs.

Partial Test (t-test) a. Price of broiler chicken eggs
Analysis result shows regression coefficient of price of broiler chicken eggs in the amount of -0,220 which indicates that price increase in the amount of Rp100/eggs is going to decrease demand of broiler chicken eggs as big as 22 eggs/month. Based on partial test, analysis result on demand of broiler chicken eggs is t hitung in the amount of -10,285 therefore it can be concluded that the price of broiler chicken eggs is distinctively affects demand on broiler chicken eggs in the research area.

c. Number of Consumer Dependent
Based on the analysis, the research obtained a regression coefficient for the number of consumer dependents of 1.335 which means an increase in the number of customer dependents by 1 person, will increase the demand for chicken eggs by 1.3 items / month. Based on the statistical partial test, the demand for chicken eggs obtained t hitung of 1.220 with a significance value of 0.226 at the level of α = 0.05. Therefore, it can be concluded that the number of consumer dependents has no significant effect on the demand for chicken eggs in the research area.

d. Tempe Price
The research obtained a regression coefficient for tempeh price of 0.001 which means that an increase in the price of tempeh of Rp 1,000 / block will increase the demand for chicken eggs by 1 item/month. Based on the statistical partial test, the demand for chicken eggs obtained t hitung of 0.667 with a significance value of 0.506. Therefore, it can be concluded that the price of tempeh had no significant effect on the demand for eggs in the research area. Correlation Between Consumer Characteristics and Behavior in Buying Broiler Chichken Eggs.
Following is the result of Consumer Behavior Expected Score in buying broiler chicken egg  Table 17, it can be seen that parameter 2 is the parameter with the largest percentage of the total score of consumer behavior. Parameter 2 obtained an average score of consumer behavior of 4.32 or 33.64%. Parameter 4 is the parameter with the smallest percentage of the total score of consumer behavior. Parameter 4 scores 1.78 or 13.86%. The consumer behavior high or low scores can be seen from the expected score of Parameter 4, which ranges from 4-20. If the score is 4-9, then the consumer behavior is low. If the score is 10-15, then the consumer behavior is moderate, and if the score is 16-20, then the consumer behavior is high.
3.7 The consumer frequency based on the consumen behavior can be seen from the following  Table 18, it can be seen that majority of consumer's behavior in buying broiler chicken eggs is in the category of medium, 88 persons or 88%.

Correlation Between Consumer Age and Behavior in Buying Broiler Chicken Eggs
Spearman rank correlation analysis was used to determine correlation between Consumer age and behavior in buying broiler chicken eggs. Based on the results of spearman rank correlation analysis, a significance value of 0.174 is obtained, which means that there is no correlation between consumer's age and behavior in buying chicken eggs on the research locus.

Correlation Between the Number of Consumer Dependents and Consumer Behavior in Buying
Broiler Chicken Eggs Spearman rank correlation analysis was used to determine correlation between the number of consumer dependents and consumer behavior in buying broiler chicken eggs. Based on the results, a significance value of 0.468 is obtained, which means that there is no correlation between the number of consumer dependents and behavior in buying chicken eggs on the research locus.
3.10 Correlation Between Consumer Job and Behavior in Buying Broiler Chicken Eggs Contingency correlation analysis was used to determine correlation of consumer job and behavior in buying broiler chicken eggs. Based on the results, a significance value of 0.727 is obtained, which means that there is no correlation between consumer job and behavior in buying broiler chicken eggs on the research locus.

Correlation Between Consumer Income and Behavior in Buying Broiler Chicken Eggs
Spearman rank correlation analysis is used to determine the correlation between the income and comsumers behavior in buying race chicken eggs. The result shows significant value of 0,000 which means that there is a correlation between income and consumers behavior in buying broiler chicken eggs in the research locus.

4.Conclusion
The price of broiler chicken eggs and consumer income are significantly affected the demand of broiler chicken eggs, while the number of consumer dependents and prices of tempeh has no significant effect on the demand of broiler chicken eggs on the research locus. There is a significant correlation between income and consumer behavior in buying broiler chicken eggs, while age and number of consumer dependents and consumer job don't have a real correlation with consumer behavior in buying broiler chicken eggs.