,

College of Economics & Management, Nanjing Agricultural University/China Center for Food Security Research, Nanjing 210095, China

1 Introduction

Maize is one of important grain crops in China. In 2014, national maize sowing area was 37.076 million hm2, and total yield reached 0.21567 billion tons, which played an important role in stabilizing total grain output of China. Farmer is main body of agricultural production. As the adopter and beneficiary of new maize variety, farmers behavior finally decides the promotion of new maize variety. Scholars at home and abroad have deeply analyzed the influencing factors of farmer’s adoption behavior for new maize variety. Gafsietal.[1]found that farmer’s acceptance rate to new variety was related to its own risk preference. Hayami[2]thought that farmer’s adoption behavior for rice variety showed negative correlation with its plantation scale. Herath[3]studied the farmers from Sri Lanka region, and thought that farmer’s adoption behavior for new rice variety was affected by the cost of new technology and the predicted yield. Barkleyetal.[4]thought that key factors affecting farmer’s wheat variety selection were variety quality and production characteristics. Hornaetal.[5]analyzed Nigeria farmers’ preferences and willingness to select new rice variety. The research by Wang Xiudongetal.[6]showed that main influencing factors of farmer’s adoption behavior for new wheat variety mainly included procurability of new variety, farmer’s risk awareness and whether or not good variety could increase benefit. Li Dongmeietal.[7]found that the promotion of agricultural technician and purchase behaviors of relatives and friends could generate positive impacts on farmer adopting new rice variety. Via comparative analysis, Tang Bowenetal.[8]thought that cultural level, loan difficulty and information availability had significant impacts on farmer’ adoption behavior for new variety. Based on the investigation on rive growers of Hubei Province, Qi Zhenhongetal.[9]found that farmer’s age, cultural level, health condition, rice planting area and income all affected the famer’s adoption behavior for new variety. The research by Huang Wuetal.[10]showed that household incomes per capita, identity of demonstration household and the proportion of working income to total household income all had important impacts on farmer adopting new peanut variety. Hou Linkeetal.[11]found that farmer risk preference had significant impact on his selection behavior for maize variety. At present, the researches about farmer’s selection behavior for new agricultural variety mainly concentrated in the rice or wheat, especially foreign researches. For this reason, taking maize as research object, introducing the variables: farmer’s risk attitude and awareness, influencing factors of the farmer’s adoption behavior for new maize variety in Huaxian of Henan Province were analyzed by using the binary Logistic model, which aimed to propose feasible channels to improve farmer’s adoption behavior for new maize variety, and provide forceful basis for establishing reasonable agricultural technology extension system.

2 Theoretical analysis and research hypothesis

2.1TheoreticalanalysisSchultz[12]proposed that farmers ad-opted appropriate agricultural technology by combining with their own existing resource condition under the target premise of maximizing profit or effect. Sahaetal.[13]thought that whether or not farmers adopted new agricultural technology depended on the predicted income and the cost of adopting new technology. But in this paper, the authors thought that whether or not farmers adopted new maize variety actually depended on comparative result of production effects between new and old varieties. Supposing other conditions were unchangeable, if the predicted net benefit of adopting new variety was more than that of existing variety, the farmers would adopt new variety. Supposing that newly increased unit cost of adopting new maize variety was the known fixed value, and using cost of old variety was , according to the prior related research thinking[13-14], adopting condition of new variety was set as：λpg(m) - (w+r)m≥p0f(m) -rm. Here,λwas risk preference coefficient,pwas product price after farmers adopted new maize variety,p0was product price of adopting old variety,mwas decision-making scale,g(*) was production function after adopting new variety,f(*) was production function of using old variety. Seen from the above formula, whether or not farmers adopted new variety was affected by the cost of farmers adopted new and old varieties, risk preference coefficient, production scale and farmer’s predicted benefit of adopting new variety, and as well as subjective risk preference of farmer, farmer individual characteristics, family business characteristics and the characteristics of agricultural technology extension.

2.2ResearchhypothesisAccording to the above analysis, ch-aracteristic variables of maize planting household were divided into five groups: individual characteristic variable, family characteristic variable, awareness to new variety variable, farmer’s risk attitude variable and agricultural technology extension characteristic variable. Moreover, the below hypotheses for farmer’s adoption behavior for new maize variety and its influencing factors were proposed.

The first group was farmer individual characteristic variables, including farmer’s sex, age and education level. It was supposed that maize grower’s sex, age and education level were related to adoption behavior of new maize variety. Said from the theory, women’s thoughts were more conservative than that of men, and their desires of adopting new maize variety were weaker. As farmer’s age increased, his plantation experience was richer, and he was possible to use new variety. Education years had positive impact on the farmer’s adoption behavior for new maize variety. The second group was family business characteristic variables, including whether or not household had the cadre, whether or not demonstration household, main household income source, maize plantation area and farmland fragmentation. When farmer’s household had the cadre, it indicated they had more chances to touch and understand the related information of new variety and technology, which was conducive to making the selection of using new variety. Under general situation, for the household that main income source was crop farming, farmers tended to use new variety and technology. Maize planting scale had positive effect on new variety selection. The farmers with relative scattered farmland had the condition of declining plantation risk of new variety, and the possibility of adopting new variety was higher than that of other farmers. The third group was farmer’s awareness to new maize variety. In this paper, it was judged by whether or not farmer understood characteristics of new variety. As rational consumer, farmer decided whether or not using new variety was related to his awareness on new variety’s characteristics. Therefore, it was predicted that farmer’s awareness had positive effect on the adoption of new variety. The fourth group was risk attitude variable of farmer. Farmers were the population with stronger risk consciousness, and subjective risk attitude of farmer was related to whether or not actively using new variety. Generally speaking, risk preference type of farmers had higher activity, while risk aversion type of farmers tended to select traditional variety generally planted in surrounding area, and their activity of using new variety was lower. In this paper, the prior two items were summed into risk preference type, while others were summed into risk aversion type. The fifth group was characteristic variable of agricultural technology promotion, including farmer’s joining times in variety training and the times to contact with agricultural technician. Generally, when farmer’s joining times in variety training and the times to contact with agricultural technician were more, farmer was possible to adopt new maize variety. When the times to contact with agricultural technician was more, farmers were possible to touch more information about new variety promoted by government, and obtain more guidance of the related supporting technique, which was all positive excitation for farmer selecting new maize variety.

3 Basic investigation situations

Main production region of maize in Huaxian of Henan Province was taken as typical region. In the investigation, farmers planting maize were selected as the investigation object. According to maize planting scale, four towns (Baidaokou Town, Niutun Town, Liugu Town and Wangzhuang Town) and three villages (Banpodian Village, Zaocun Village and Zhaoying Village) of Huaxian were selected, with 12 natural villages in total. Field investigation was conducted from January 16 to January 27, 2015. Investigators directly entered into farmer’s house for investigation, and 300 questionnaires were obtained, in which there were 279 effective questionnaires, with 93% of effective rate.

3.1BasicsituationsoftheinvestigatedfarmersFirstly, small-scale production was dominant. Maize planting area of the investigated farmers mainly concentrated below 0.67 hm2, accounting for 71.7%, and the investigated farmers with maize planting area surpassing 1.33 hm2only accounted for 14%. Secondly, farmers engaged in maize cultivation were dominated by middle and old age labor force. Average age of the investigated farmers was 47.32 years old, and the farmers of 40-60 years old accounted for 64.2%. Third, education level of farmer was generally lower. Farmers of junior high school and below accounted for 73.1%, while farmers of junior college and above only accounted for 2.9% (Table 1).

3.2Selectivepurchasesituationofmaizeseedbyfarmer

3.2.1Related situation of seed selection. When farmers answered the problem that "what is the most important factor affecting the selection of maize variety", in 279 farmers, 73.4% of farmers selected "yield", 17.6% of farmers took "stress resistance" as primary factor of seed selection, farmers selecting "quality" accounted for 8.6%, and farmers selecting "growth period" accounted for 0.4%. In the investigation, some farmers thought that when stress resistance of maize variety was strong, especially lodging resistance, maize yield could be guaranteed. Therefore, high stress resistance of seed was the factor considered when farmers selected seed. When farmers answered the problem that "who does have the largest impact on variety selection", 30.8% of farmers selected "their past experience", the farmers selecting "agricultural technician" accounted for 25.8%, the farmers selecting "seed sales staff" accounted for 22.9%, the farmers selecting "demonstration household" accounted for 12.9%, the farmers selecting "neighbors, relatives and friends" accounted for 7.6%. It was clear that agricultural technician and seed sales staff had great impacts on maize variety selection.

Table1Statisticaldescriptionofvariable

VariableHouseholdnumberProportion∥%VariableHouseholdnumberProportion∥%SexCulturaldegreeMale20774.2Primaryschoolandbelow7426.5Female7225.8Juniorhighschool13046.6AgeSeniormiddleschoolorpolytechnicschool6724.0≤30yearsold176.1Juniorcollegeandabove82.931-40yearsold5218.6Whetherornothadvillagecadre41-50yearsold11742.0Yes4516.151-60yearsold6222.2No23483.9≥60yearsold3111.1WhetherornotdemonstrationhouseholdFarmlandarea∥hm2Yes8028.70-0.3310939.1No19971.30.33-0.679132.6Mainincomesource0.67-1.334014.3Farming12645.21.33-3.33176.1Other15354.8Morethan3.33227.9

3.2.2Achievement of seed information and purchase channel. When farmers answered the problem that "what is main way to understand new maize variety", 47.7% of farmers selected "seed company", 21.5% of farmers selected "seed dealer", the farmers selecting "agrotechnical station" accounted for 13.6%, the farmers selecting "neighbors, relatives and friends" accounted for 8.2%, the farmers selecting "television, newspapers and periodicals" accounted for 2.5%, and the farmers selecting "other" accounted for 6.5%. It was clear that main agricultural information obtaining channel of most of farmers was agricultural technology extension service, followed by interpersonal relation, and the farmers selecting media were fewer. For the problem that "where do you buy maize seed", 59.1% of farmers selected "county seed company", 27.6% of farmers bought from "local individual dealer", 9.3% of farmers bought from "the township agrotechnical station", the farmers selecting "scientific research institution", "foreign dealer" and "obtaining through their own social relations" respectively accounted for 1.4%, 0.7% and 0.7%, and the farmers selecting "other" accounted for 1.1% (Table 2).

Table2Purchasesituationofmaizeseedbyfarmer

ClassificationHouseholdnumberProportion%ClassificationHouseholdnumberProportion%ThepersonhadthemaximumimpactonvarietyselectionThemostfactorofselectingmaizeseedAgriculturaltechnician7225.8Yield20573.4Seedsalesstaff6422.9Stressresistance(diseaseresistance,lodgingresistanceanddroughtresistance)4917.6Demonstrationhousehold3612.9Quality248.6Neighbors,relativesandfriends217.6Growthperiod10.4Theirpastexperience8630.8MainsiteofbuyingseedMainwaystounderstandnewmaizevarietyCountyseedcompany16559.1Seedcompany13347.7Localindividualdealer7727.6Seeddealer6021.5Townshipagrotechnicalstation269.3Agrotechnicalstation3813.6Scientificresearchinstitution41.4Neighbors,relativesandfriends238.2Foreigndealer20.7Television,newspapersandperiodicals72.5Obtainingthroughtheirownsocialrelations20.7Other186.5Other31.1

4 Empirical analysis

4.1ModelselectionAdoption behavior of farmer for new maize variety had two kinds of situation: "adopting" and "not adopting". Each farmer could make the best selection based on rationally and comprehensively measuring various kinds of influence factors, and it was a typical two-element decision problem. Therefore, the influencing factors of farmer’s adoption behavior for new maize variety were analyzed by using binary Logistic regression model. The model was set as below:Y=b0+b1x1+b2x2+…+bnxn. Here,Yshowed farmer’s adoption behavior for new maize variety. When farmer adopted new maize variety,Y= 1; on the contrary,Y= 0.xi(i=1,2, …,n) was factor that was possible to affect farmer adopting new variety, andbi(i=1,2,…,n) was regression coefficient of theithexplanatory variable.

4.2Variableselection(i)Farmer individual characteristic variables, including sex, age, education years;(ii)Household operation characteristic variables, including whether or not family had the cadre, whether or not demonstration household, main income source, maize planting area and farmland fragmentation degree;(iii)farmer’s awareness to new variety, which was illustrated by whether or not farmer understood the characteristics of new variety;(iv)risk attitude characteristic variable, which mainly indicated subjective risk preference of farmer;(v)characteristic variable of agricultural technology promotion, including farmer’s joining times of variety training and the times to contract with agricultural technician. According to 279 copies of questionnaires, statistical characteristics of each variable were obtained(Table 3).

Table3Explanationandstatisticalcharacteristicsofmodelvariable

VariableclassificationVariablename VariableexplanationStandarddeviationMeanTheexplainedvariableWhetherornotadoptednewmaizevarietyAdoptingnewvariety=10.5010.51Notadoptingnewvariety=0ExplanatoryvariableFarmerindividualcharacteristicvariableSexFemale=0;Man=10.4380.74AgeUnit:yearsold10.09347.32EducationyearsUnit:a3.3238.66FamilyoperationcharacteristicvariableWhetherornotfamilyhadthecadreNo=0;Yes=10.3680.16WhetherornotdemonstrationhouseholdNo=0;Yes=10.5670.31MainincomesourceOther=0;Planting=10.4990.45MaizeplantingareaUnit:mu46.49319.13LandfragmentationdegreeUnit:block1.3042.40FarmersawarenesstonewvarietyWhetherornotknewthecharacteristicsofnewvari-etyNo=0;Yes=10.4670.68RiskattitudecharacteristicvariableFarmerssubjectiveriskpreferenceRiskaversion=0;Riskpreference=10.4990.55Agriculturaltechnologypromotionchar-acteristicvariableTimesofjoininginvarietytrainingUnit:Times0.9161.53TimestocontractwithagriculturaltechnicianUnit:Times1.3962.35

4.3ModelresultsandanalysisUsing SPSS18.0 statistical software, model estimation was conducted by using forced regression(Enter) method, and its regression results were shown as Table 4. Seen from Table 4, the equation had better fitness degree(P>0.1), illustrating that the observed value of dependent variable did not have difference from predicted value of model. Therefore, it could not repel original hypothesis, and thought that data fitting effect was good. (i)In farmer individual characteristic variable, farmer’s education years passed through significance test of the model, and showed positive correlation with farmer’s adoption behavior for new maize variety. It illustrated that the farmers with long education years were more willing to accept new maize variety than that with short education years. By contrasting new variety with old variety, it was decided whether or not planted new variety. For overall traits of new variety promoted on the market were generally better than old variety, the farmers with higher education level were more willing to adopt new variety.(ii)Family operation characteristic variables, whether or not family had the cadre, whether or not demonstration household, main income source and maize planting area all passed through significance test. Cadre family showed negative impact on farmers selecting new maize variety, which was not consistent with prediction. Model result showed that when family had the cadre, the possibility of selecting new maize variety was smaller. The investigation found that demonstration household was generally agricultural pioneer in the village, and was earlier to touch new variety than common farmer. Moreover, demonstration household planting new variety could obtain corresponding subsidy, further increasing the possibility of adopting new variety. Main income source affected farmer’s adoption behavior for new maize variety and showed positive correlation. Maize planting area had positive impact on farmer adopting new maize variety. Whether or not farmer adopted new variety mainly depended on if the increased economic benefit after adopting new variety was significant. When farmer’s planting area was too small, yield increase after new variety input was insignificant, which was difficult to attract farmer’s interest in new variety. For the farmer with larger maize plantation area, the adoption of new variety could obtain larger scale benefit. Therefore, they were willing to plant new variety and hoped to obtain high yield and harvest.(iii)Farmer’s awareness to new variety was an important factor affecting adoption behavior. Model estimation result showed that when farmers understood new variety, they tended to select new variety. In field investigation process, it was found that farmers could understood the characteristics and role of new variety via various channels before adopting new variety, and then judged if adopting new variety.(iv)Farmer’s risk attitude had significant impact on farmer’s adoption behavior for new maize variety. It showed that farmers’ adoption behaviors for new maize variety were affected by themselves risk attitude, and risk preference type of farmers were more possible to select new maize variety than risk aversion type of farmers. Government should enhance supporting strength after farmers adopted new variety, and declined income loss brought by new variety for farmers, thereby promoting that they adopted new variety at ease.(v)In agricultural technology promotion characteristic variables, contract times between farmers and agricultural technicians had significantly positive impact on farmers adopting new variety. As the bridge between new variety and farmer, agricultural technician transmitted the information related to new variety to the farmers in the guidance process, thereby promoting more farmers to adopt new variety.

Table4Logisticanalysisoffactorsaffectingfarmer’sadoptionbehaviorfornewmaizevariety

VariableBS.E.WalsdfSig.Exp(B)Sex-0.3560.4380.66010.4160.701Age0.0020.0190.01610.8981.002Educationyears0.228***0.0739.67610.0021.256Whetherornothadvillagecadre-1.299*0.6054.60610.0320.273Whetherornotdemonstrationhousehold0.934*0.4943.57410.0592.546Mainincomesource0.652*0.3912.78110.0951.920Maizeplantationarea0.039*0.0242.72310.0991.040Farmlandfragmentationdegree0.0980.1550.39710.5291.103Whetherornotunderstoodthecharacteristicsofnewvariety1.456***0.40113.13610.0004.287Subjectiveriskpreference1.343***0.37812.63710.0003.831Joiningtimesofvarietytraining0.4050.2772.12810.1451.499Timestocontractwithagriculturaltechnician0.411***0.1537.19710.0071.509Constant-5.8861.25022.17110.0000.003-2Loglikelihood204.418CoxandSnellRsquare0.480NagelkerkeRsquare0.640HosmerandLemeshowfittinggoodnesstest0.742

Note:*,**and***respectively showed that regression coefficients were significant at the levels of 10%, 5% and 1%.

5 Conclusions and suggestions

5.1ConclusionsBased on investigation data of maize production by 279 households in three villages and four towns of Huaxian, Henan Province on January of 2015, the influences of farmer individual characteristic, family operation characteristic, farmer’s awareness to variety, farmer’s subjective risk preference characteristic and agricultural technology promotion characteristic variable on farmer’s adoption behavior for new maize variety were tested. The research showed that in farmer individual characteristic, farmer’s education years had significantly positive effect on the adoption for new maize variety; in family characteristics, whether or not family had the cadre, whether or not demonstration household, main income source and maize planting area affected farmer’s adoption behavior for new maize variety; farmer’s awareness to new maize variety and farmer’s subjective risk preference significantly affected the adoption of new variety. Additionally, in agricultural technology promotion characteristics, contract times between farmer and agricultural technician affected the adoption behavior for new maize variety.

5.2Suggestions(i) Completing farmer’s technical training mechanism. On the one hand, government should enhance the investment on rural education business, and improve nine-year compulsory education’s quality, thereby improving farmer’s cultural quality of science and technology. On the other hand, promotion department of agricultural technology should sufficiently use various forms to propagate new variety for farmers, thereby promoting farmers to adopt new maize variety.(ii)Accelerating land circulation and encouraging scale plantation. Empirical analysis showed that enlarging plantation scale could promote the adoption of new maize variety. Local government should vigorously encourage scale plantation, give certain support for the required hardware and software facilities in scale plantation process by large growing households, and provide fund subsidy or broad loan permit for large growing households, further promoting farmers initiatively to adopt new variety.(iii)Enhancing and perfecting subsidy policy and increasing comparative benefit of maize plantation. On the one hand, government could increase finance subsidy to increase farmer’s income, such as grain direct subsidy and purchase subsidy of agricultural appliance, further improving farmer’s activity of planting grain. On the other hand, government could indirectly decline maize plantation cost, such as actively implementing tax relief policy, thereby improving farmer’s activity to plant new variety.(iv)Providing rural commercial insurance and declining farmer’s income risk. When promoting new maize variety, government and extension departments of scientific research could provide the insurance for farmers adoptingnew variety via the form of commercial insurance. It was conducive to decreasing farmer’s risk in adoption process of new variety, and increasing farmer’s activity of adopting new variety.

[1] GAFSI S, ROE T. Adoption of unlike high-yielding wheat varieties in Tunisia[J].Economic Development and Cultural Change, 1979(28):119-133.

[2] YUJIRO H. Induced innovation, green revolution and income distribution[J]. Comment Economic Development and Cultural Change,1981(30):169-176.

[3] HERATH HMG, HARDAKER JB, ANDERSON JR. Choice of varieties by Sri Lanka rice farmers: Comparing alternative decision models[J].American Journal of Agricultural Economics,1982(64):87-93.

[4] BARKLEY AP, PORTER LL. The determinants of wheat variety selection in Kansas, 1974-1993[J]. American Journal of Agricultural Economics, 1996(78): 202-211.

[5] HORNA JD, SMALE M, VON-OPPEN M. Farmer willingness to pay for seed-related information rice varieties in Nigeria and Benin[J]. Environment and Development Economics, 2007(12): 799-826.

[6] WANG XD, WANG YC. Analysis on the choice behavior of new wheat varieties of peasant households based on high-class seeds subsidy policy[J]. Chinese Rural Economy, 2008(7): 24-31. (in Chinese).

[7] LI DM, LIU Z, TANG S,etal. Analysis on the intentions of farmers’ selection on new rice varieties and the influencing factors——Based on the investigation of 402 peasant households in primary rice production regions of Sichuan Province[J].Problems of Agricultural Economy, 2009(11): 44-50. (in Chinese).

[8] TANG BW, LUO XF, QIN J. Analysis on the influencing factors of different attribute technologies adopted by farmers[J]. Chinese Rural Economy, 2010(6): 49-57. (in Chinese).

[9] QI ZH, LIANG FL, ZHOU H,etal. Empirical research on influencing factors of the rice farmers choosing new varieties. Based on the statistics of Hubei Province[J].Journal of China Agricultural University,2012(2):164-170.(in Chinese).

[10] HUANG W, HAN XQ, ZHU GM. Analysis on the influencing factors of new varieties’ acceptance of peanut farmer households[J]. Journal of Agrotechnical Economics, 2012(12): 12-21. (in Chinese).

[11] HOU LK, QIU HG, BAI JF,etal. Effect of farmers’ risk appetite on the input of agricultural production factors[J]. Journal of Agrotechnical Economics, 2014(5): 21-29. (in Chinese).

[12] THEDORE W. SCHULTZ. Reforming traditional agriculture[M].(LIANG XM) Beijing: The Commercial Press, 1987. (in Chinese).

[13] ATANU S, ALAN LH, ROBEIT S. Adoption of emerging technologies under output uncertainty[J]. American Journal of Agricultural Economics, 1994, 76(11): 836-846.

[14] KONG XZ, FANG SH, PANG XP,etal. Analysis of the effect of household endowments on the agricultural technology adoption decision in west China[J]. Economic Research Journal, 2014(12): 85-95, 122. (in Chinese).

AboutKIT

The Royal Tropical Institute (KIT) in Amsterdam is an independent centre of knowledge and expertise in the areas of international and intercultural cooperation, operating at the interface between theory and practice and between policy and implementation. The Institute contributes to sustainable development, poverty alleviation and cultural preservation and exchange.