genderand social targeting in plantbreeding...conceptual framework: use a marketing approach...
TRANSCRIPT
5 O c t o b e r 0 1 7
Gender and Social Targetingin Plant Breeding
A l a s t a i r O r r , C i n d y C o x , J a c q u i A s h b y
Breeding Cycle – Main Stages
Social targeting and demand analysis
Setting breeding priorities/objectives
Generation/Identification of new variation
Selection in segregating populations
Testing experimental
varieties
Releasing new varieties
Seed production and distribution
Introduction
G.Smith/CIAT
BackgroundPapercommissionedforthisWorkshop,tobeuploadedonWOCwebsite
Objectivesofthispresentation:Ø SharedraftpaperØ UsefulideasfromPenangWorkshopØ Yourfeedbackneededtohelpfinalizethepaper
Aworkinprogress….!
What’s ahead
G.Smith/CIAT
TheproblemAconceptualframeworkApplicationtobreedingprogramsDesigningtargetingstudiesUsinglargedatasetsConclusions
The problem: how to make breeding programs more customer-driven?
G.Smith/CIAT
Breedingprogramsforresource-poorfarmershavelimitedinformationabouttheirclientsandtheirclients’preferences
Why?Ø Nomarketindicators• sales• marketshare• InstantfeedbackfromsalesteamsØ Small-scalestudies,not
representativeatnational,agro-ecosystemlevels
Ø Nodisaggregation(poor/non-poor,gender)
Conceptual framework: use a marketing approach
G.Smith/CIAT
Segmenting:Identifymarketsegments(“ahomogeneousgroupofconsumerswithauniquesetofpreferences”)
“Market”notnecessarilyacommercialmarket,canbeforhomeconsumption
Targeting:selectmarketsegment(s)forspecificproducts
Largeconsumerdatabasestoidentifymarketsegmentsandpreferences:eg.LMS,TargetGroupIndex
Example:TheCIATBeanProgramusedLCMforKenyatotargettheurban/middle-classmarketandparticipation(‘immersion’)tolearnpreferencesforpre-cookedbeans
The Segmenting-Targeting-Positioning (STP) Framework
G.Smith/CIAT
S:Segmentingthemarket
1.Definethemarket
2.Selectbasesforsegmentation
3.Validatethesegments
T:Targetingspecificmarketsegments
4.Constructsegmentprofiles
5.Evaluatemarketattractivenessofsegments
6.Identifywhichandhowmanysegmentstotarget
P:Positioningproductswithtargetmarketsegments
7.Developpositioningstrategy
8.Designappropriatemarketingmixtocommunicatepositioning
Application to breeding programs: Segment the market
G.Smith/CIAT
Steps Description Datarequired
1.Definethemarket Genericmarket:aggregatemarketforaproduct
Relevantmarket:boundarytoguidebreedingprogram
Definedmarket:existingcustomers,potentialcustomers
Target countries,
Agro-ecosystems,
Area planted to crop
Value chains for crop
End uses for crop2.Selectbasesforsegmentation Geographic(where?)
Demographic(who?)
Behavioral(why?)
Region, state
Age,maritalstatus,gender,ethnicity,income,occupation,consumption/sale
Enduses,traitpreferences
3.Validatethesegments Measurable
Substantial
Accessible
Differentiable(responddifferentlytomarketstimuli);
Actionable(programcanbedesignedtoservethemarket);
Size
Purchasing power,
profitability
Growth rate
Location
Distance to market
Application to breeding programs: Target the market
G.Smith/CIAT
Steps Description Datarequired
5.Evaluatemarketattractivenessofsegments
Excludesegmentswithlownumbersofpoorgrowers,sellers,processors
Absolute number of poor and non-poor growers, sellers, processors
Size of segments,
growth rate,
number of resource-poor growers,
location,
distance to market
6.Identifywhichandhowmanysegmentsshouldbetargeted
Comparesegmentswithresource-poorgrowers,sellers,processors
Gender and social targeting
G.Smith/CIAT
Primarytargetsegmentistheresource-poorgrower,sellerorprocessor,notwomen/men
Traitpreferencesreflectgenderrolesinproduction,sale,andvalueaddition
Usegenderastargetingvariablewherewomen/menplaydifferentrolesinproduction,processing,ormarketing
Avoid“genderfundamentalism”Makegender“instrumental”:whodowewanttoinfluencetouseourproduct?Theexampleofpre-cookedbeans:firstidentifyconsumersegments,thenidentifygenderroles
G.Smith/CIAT
At what stage in the breeding cycle is information needed?
STP1 STP2-6macro STP2-6micro
Designing targeting studies within the breeding team
G.Smith/CIAT
Define the question jointlyBreeders’question:Isthereatraitthatcanhelpdevelopanewproductthatwillleadtoabreakthroughinadoptionandresultinimpactonanindustrialscale?Targetingquestion:Isthereatraitthatcanhelpdevelopanewproductthatwillleadtoabreakthroughinadoptionbyresource-poorfarmers/sellers/processors(includingwomen) andresultinimpactonanindustrialscalethatbenefitspoorconsumers?
Use the STP framework as a checklist
G.Smith/CIAT
Segmentingthemarket
Whatarethedifferentendusesfortheproduct?Hastheprogramidentifiedmarketsegmentsintermsofgeography/agro-
ecosystem?Hastheprogramalsoidentifiedmarketsegmentsaccordingtothetypeoffarmer?
Targetingmarketsegments
Doweknowthesize,profitability,andnumberoffarmersineachsegment?Haveweusedtherightindicatorstoidentifyresource-poorfarmers?Dowomenplayamajorroleinproduction/processing/marketing?
Whichmarketsegmentscontainthemajorityofresource-poorfarmers?Whatarethetraitpreferencesofthesefarmers?
Whatproductsareneededtomatchthesetraitpreferences?
Positioningtheproduct
Howwillresource-poorfarmersbecomeawareoftheproduct?Cantheyfinditeasilyiftheywanttotryit?
Howmuchwilltheypaytouseit?
Use the STP framework as a matrix
G.Smith/CIAT
X:Poorsmallholders
Y1:Homeconsumption,valuechain1
Y2:Genderrolesinproduction,sale,processing
Use mixed methods
G.Smith/CIAT
Why mixed methods?
Qualitative methods: trait preferences, gender roles, empowerment Quantitative: trait combinations, segmenting and targeting at national level
Think at scaled and use large datasets
G.Smith/CIAT
OvertoCindy
Using large datasets
G.Smith/CIAT
Guidingquestions:Ø Wherearethepoor(womenand
men)farmersandwhatistheirwelfarestatus?
Ø Onwhatfarmingsystemsdotheyrely?
Ø Whatconstraintsandriskslimittheproductivityofthosefarmingsystems?
Ø Whatcroptraitsmightbestsustainablyraisefarmproductivity,humanwelfareandlivelihoods?
Ø Whatwouldbethebroaderimpactsofsuchchange– whomightwin,whowouldlose(e.g.,gender)
Targeting
First-orderfilter->Environment
Examples:Ø Cropsuitability(GAEZ)Ø SubnationalCropStats(SPAM)
₋ HarvestChoice/IFPRIØ LivelihoodZones(FEWSNET)
Using large datasets
SPAM datasets
Datasetsavailableat:https://www.fews.net/
FEWS NET – Livelihood Zones
Second-orderfilter->Socio-economics
Examples:Ø Population(CIESIN)Ø Poverty(PovcalNet)Ø Marketaccess(Grump,CIESIN)Ø Gender(HHsurveys)
Using large datasets
CELL5M Geo-spatial Database
Category Sub-category(Numberofdatalayers)
Agriculture HarvestedAreaofCrops(134)CropProduction(134)ValueofCropProduction(134)CropYield(134)CropYieldVariability(2)Livestock(16)
Demographics HealthandNutrition(90)IncomeandPoverty(36)Population(12)
Agroecology AgroecologicalZones(4)Climate(7)Elevation(1)FarmingSystems(2)LandCoverandLandUse(21)PestsandDiseases(8)SoilResources(19)
Markets Marketshed(1)Portshed(1)TravelTime(11)
>750datalayersforSSA@10x10kilometerpixels
Example:DevelopmentDomainsinEthiopia
Threelayers:Ø AgriculturalPotentialØ PopulationDensityØ MarketAccess
Chamberlinetal2006
Development Domains
Example:WorldBank’sLivingStandardsMeasurementStudy- IntegratedSurveysonAgriculture(LSMS-ISA)*
Others:WEAI&DHS
*LSMS-ISAdataarenationally-representativeandmulti-topic,withgeo-referencedhouseholdandplotlocations,andinformationonproductionandidentityofmanagers,ownersandlaborersatplotlevel
Sourcing gender-sensitive data
Example: Cassava in Nigeria(FEWS NET + SPAM)
Where are women cassava farmers relative to men? (LSMS)
Where are poor women cassava farmers relative to men?
Households who sell Processed or Unprocessed Cassava - or No Sale
Big Data Platform (beta)…coming soon!
http://ceres.bigdata.cgiar.org/mapview.php
30
- Severalspatially-explicitdemographicandbiophysicaldatasetsavailable
- Micro-levelgender-sensitivedataincreasinglyavailableviaHHsurveys(e.g.,LSMS-ISA,WEAI&DHS)
- WhatabouttheValueChain???
What are the constraints faced by poor farmers and what drives their crop needs and preferences?
What is gender-relevant and what is not?
Conclusions
Conclusions
G.Smith/CIAT
UseSTPasaconceptualframeworkforgenderandsocialtargetingSocio-economictargetingprecedesgendertargetingMakegenderinstrumental
Improvedesignoftargetingstudiesü JointlydefinetheresearchquestionüUsetheSTPframeworkasachecklistormatrixüUsemixedmethodstogeneratedataüThinkatscale