AN EVALUATION OF CONSUMER PURCHASE PATTERNS FOR SELECTED FRESH FRUITS AND VEGETABLES Thcsfis for IN» 31.34;:qu 2:! M. A MICHIGAN STATE {EI‘ZWERSIE‘Y Thamas A, Cmagar 39563 MICHIGAN STATE UNNERSITY LIBRARIES llllllllllllllllllUlllllllIIIIHIIIIHIIHH1|I|lllllll||HHll 31293 01025 6554 LIBRARY Michigan State University PLACE IN RETURN BOX to remove this checkout from your mood. TO AVOID FINES return on Of before data duo. DATE DUE DATE DUE DATE DUE usu IcAnAIIIrmath ActIorVEquaI OpponunlIyImtIIwon WWI “Tho road Butt-1m- program at nmgan Stat. Mun-city is under thc spun-ship or the humid. “location of Food Chains” m EVALUATIOE or comm szmsz PATTERNS ma 331mm mm man's um VEDEI‘ABLES Thomas L. Grease: A THESIS Submitted to the college of Business and Public Service of Michigan State Univu'uty and Applied Sciencc in partial fulfillment cf the requiranents for the degree out HASTE? OF ARTS Department of General Mm auricula- in Food Distribution 1956 ACKNOWLEDGWTS The trite: wishes to acknowledge the assistance received fro. a cent manhu- ot people who graciouely deeperated in this etw. First or all, appreciation ie acknowledged to the National Association of Food Chains fa financial help received through the geduete fellowship. Gratitude ie expressed to Dr. David J. Luck, Director of the Bureau of Busineee Research, and Dr. Edward A. Brand, Director of the Curriculum in Food Distribution, for their valuable advice end guidance in the presentation of thia em. The mm- 1. also gratifulLto Dr. Gerald a. mackenbuah and Dr. James D. Shaffer of the Deparhent of Agiculturel Economice who gave willingly or their time in supplying the neoeeeary infor- nation to oomlete thie ItWr’ may . .‘ MWI'DCODDOOOOOO I. IV. V. - TABLE OF COHTEM‘S WCdMWeeeeeeeeeeeeee WflammdSDWeeeaeee ‘M.ofth.MQoeeeeeeeeeee 'Iethedet’Btndy eeeeeeeeeeeeee sounczmnumormmn .... .. .. Objeotim of the manger: State Unveruty WCIMeeeeeeeeeeeee (berationotthel’anel...a....... alarecterietice of the Lancing Papnlatia e mmthOfthCMeeeeeeeeee nmnmhwj-Meeeeeeee ProoeeeingottheData..oo....... fllltetionof,bata............. panama um um emcrms'rrcs a WHMATOMCHASEOOOOOOOOOOO flesh-PotatoPurohaeeIDeu.......... RepackagingmdSiaeflmo...o Packagingcmtainere...-..ooo. e Relationship of Family Characteristics te Mktt‘flmeeeeeeeeea M eeeeeeeeeeeeeeeeee “CHASE PATTERNS AND CHARACTEISTIGS (F ml”mpmmm.....000000 Inabipplel'm'ohaeenata e e e ee e e e e Waginganduaeothrohaee . e e e . Relationship of Penny characta'ietiel te FreehipplePurchaeee e-eeeeeeeeee O‘COOOOUOOOCOOOODD. MSE PAW AND CHARACTWIOS (I WMEMWEQOOOCOOCOOO Freeha'angeru'ehae‘ ODMeeee manhthdeeeeeeeeee hepaokaging and Biae or Purchase 0.... 0.... .0. 0 0 Relationship Between Family maracteriatiee mnmm.muuoegoeeoe m eeeeeeeeeeeeeeeeee CD... 0 0...... g ‘0 OWEN 0“ 3‘5 52553 3 01383518» 62 67 85 93 96 103 CHAPTER VI. FRESH GRAPEFRUIT PURCHASES VII. IDOOOODOQODO PURCHASE PATTHENS AND CHARACTERISTICS OF 0.0... Preeh m-apefruit Purchaae Data Pricing Method Wm COCOOOOOOOCCO Deoiaion Making in Purchaeing Grapefruit Relationship Between Family Gheracteriatiee and Free]: Grapefruit Pm-oheaee OOOOOOODOOOOOOCODO O C C O .0... nm A5 RELATED ‘ro . SPECIFIC mmmxsxm PRACTICBODOODODOO Hetureofthenata Purchasing Pricing e 0 SW 0 0 VIII. SW . e e LPPEIIDIX...-ooo BmummPHOCOIO 1 0.. .9... Niven-tiling and Pronotion Display e e e e e I O O O O O O O O O O O I C O O C '0'... 0.0... .0... 00.... 00.... 00.... .00... 0.0... ee'eeee O 0.9... e e e... 0.0... 0.0.0. THE USE OF HICHIGAH STATE UNIVERSITI calm PANEL Page 165 171 173 LIST 01" TABLE Table Page 1. Quaracteristios of lensing Population, 1950 I. . . . . 15 2. Characteristics of the Average Family in the 11.8.0. Consumer Panel at Different Tine Periods Con- pered to the Average Fonily in the 1950 WCcmeeeeeeeeeeeeeeeeee 1? 3. Ember of Families Roma-ting Continuously for Selected Time Periods, by Year Started Re. porting. Fobmuw 1951 to w 1955. 11.3.0. mwpmdoeeeeeeeeeeeeeeeee 19 h. Tearly Variatim in Expenditure Per Capita and Average Percent of Families Buying Msh Potatoes Each Week by the Thirteen-Pam HeekPa‘iodsoftheIear..oco........ 28 5. Seasonal Variation in Fresh Potatoes Quantity, Eupenditure, hpmditure Rank, and Frequency of Purchases During the Thirteen-Pom- Week Pmm0f19s3eeeeeeeeeeeeeeeeee 30 6. Table of Relative Distribution of Potato Sales by State of Potato Origin Throughout the Fair 3068021501'1953seeeeeseeeeeeeseea 33 7. Distribution of Fresh mchigan Meta Ptn-chases by Size or the unit of Purchases Dtu'ing 1953 . . . . . 35 8. Variation in Sise of Michigan Potato Purchasu and Quantity Pin-dissed by Hethod of Purchase fu- l'hree-Four wed Periods in the First mas-ta d1955eeeeeeeeeee,eeeeeeeeeee 9. Distribution of California Potato Purchases by Bias OfthOMtOfMdIUDWJSSJeeeeeeee 151 10. Variation in Sise of Idaho Potato Purchases and Quantity Purchased by Hethod of Purchase for Three-Four Week Periods in the First Quarter “1955eeeeeeeeeeeeeeeeeeeeee 1&3 11. Distribution of Idaho Potato Purchases by the files efmeUnitofPurchaseDm'ingESJo ee . . . . . M; 12. Distribution of all Other Fresh Potato Purchases by also of the Amount of Purchase During 1953 . . . . h6 Table Page 13. consumer Acceptam of Polyethylene and flesh Window Paper Bags and Corrugated Boxes in Two Bangor Swermarkets for a Six-Heck Period, 1952- l9535eason............o....... ‘19 1h. Genome:- Aeeeptanoe of Washed and Umashed Potatoes in Polyethylene Bags at Two Bangor Mmkets . for a Six-Week Puiod, 1952-1953 Season . . . . . . SO 15. [differences in Factors Related to Frash Potato Pur- chases Am0ng 178 H.S.U. Consumer Panel Fmilies Grouped According to Quantity of Fresh Potato PurchasesPeranilyinBSB ........... 52 16. Differences in Factors Related to Prosh Potato Pur- chases Among 178 14.8.0. Consuna- Panel Fandlias Grouped Aceading to the number of weeks They Purchasedh'eshl’otatoeelnls53 eeeeeeeee 5h 11. Differences in flash Potato Manes Among 178 H.8.U. Gonsmer Panel Families Grouped According to Per “p1uxnm1953OOOCOUOICOCOOODC 55 18. Differences in flash Potato Pm'ohases Among 178 14.8.0. Cmsmer Panel Paulies Grouped Accord- . ingtoSiseofFanilyinl953........... 51 19. Seasonal Variation in Quantity, Empenditure, Maldive-e Rank, and Prequuoy er Pur'oliaseofipples,1953o............ 6" 20. Tearly Variation in Expenditure Per Capita and Average Pacentage of Families Buying Fresh Apples Each Week During the Thirteen-Four week Periodsefl952, 1953, must . . . . . . . 66 21. Distribution of Fresh Apple Purchases by Size of the Unit of Pin-chase, FouroThirteen Week Pm'iods . . . 68 22. The Effect of Merchandising Practices on AppleSales................... 69 23. Variation in Bias of Apple Purchases and Quantiw ”chased by Method of Purchau for Three-Fm UeekPeriodsinthe First Quarterofl955 . . . . 73 21;. Quantity of Apples Bold Per 100 Customers by Specific Hethods in 12 Food Chain Stereo, Pittsburyx,Pennsylvania eeeeee'eeeeeee n Table 25. 26. 27. 29. 30.. 32. 33. 3h. 35. has Infferences in Factors Related to Fresh Apple Par. chases Among 178 11.3.0. Consumer Panel Families Moped According to Quantity of hash Apple PurchasesPerFanilyin1953o........n. 79 Differences in Factors Related to Fresh Apple Pur- mases Anong 178 11.8.8. Consuma- Panel Fullies Grouped According to the Dunba- of Weeks they Wfldmwflnl953...o..o... 81 Diffcences in Factors Related to Fresh Apple Pur- chases Among 178 11.8.0. CW Panel Families Grouped According to the Average Size of Fresh ApplePurohasesinl953............. 82 Differences in Fresh Apple Purchase among 178 $3.0. Consuel- Panel Families Grouped AmrdngtoPerapitaIncome,1953oaoa.o. 83 Differences in Fresh Apple Purchases Among 178 14.5.1.1. consumer Panel Fmilies AesordingtoSiaeofFas-uly,1953 o....... 3h Teeny Variation in Expenditu-e Per Wits and Average Parent of Families Buying flesh Oranges Each Week During the Thirteen-Four We“ PmOdl Of 1952, 1953, 334195“- e e e e e e e 89 Seasonal Variation in Fresh (range Quantity, Expenditure, Expenditm'eaankandF‘ren quencycfPurchase, Dan-ingtheThirteen— meOflkPfl'iOdflfl9s3eeeeeeeeeees 91 Variation in Sise of Orange Purchases and Quantity Purchased by hsthod of Pacheco for Three-Four was: Pmmmthonrstouartcormss . . . . 99 Vciation in Sise of (range Purchases and Quantity Purchased During the Paar-Thirteen Wed: Periods “”53COOOCDDOCIOOOOOGOOCO. 101 Diffcences in Factors Related to Fresh (range Pur- chases Along 168 $3.6. Gmsuner Panel Failies GroupedAoowdingto the QuantityofPresh OrangesMhasedPerFanilyinDSJ.....-. 10“ Differences in Factors Related to Fresh (range Pur- chases Among 168 14.5.8. Consular Panel Fuilies Grouped According to the Number of Weeks They Purchasedi‘resha'angesinl953 eeeeeeeee 10$ Table Page ho. Coupetitive Relationships Between the Twenty-Five Top Fresh Fruits and Vegetables, June ll; toJulyll,1953o................ 138 In. competitive Relationships Between the Twenty-51v. Top Fresh Mite and Vegetables, July 12 toAngust8,1953................ 139 M. Qupotitive Relationships Between the Twenty-Five Top Fresh Fruits and Vegetables, August 9 tog.ptu‘b.r531953eeeeeeeeeeoeere 110 Is. Capetitive Relatimships Between the Twenty-Five Top Fresh Fruits and Vegetables, Novnher 29 “W26,1953..ga........... m CHAPTER I INTRODUCTION Mose of Study Since Pets-m of 1951, a consumer panel hes operated at nohigan State University. The resend: project that supports this panel was originelJJeetablishedsoeeto opceteeontinuouslyforatenyeer period. Thepenelia coupe-dotebout 250£enfllieeinthe Lansing areae. This group of fasiliee regularly subunits a weekly diary, which eouteineadetsiledreeordettheirfoodpm-ohaeesnedednring that M, to the Want of Agricultural Economies, maligan State Unite-iv. (he efthenuwarees invhieh these dateprevidedlv the Judaism State University W Panel m be of usefulness is to the retail food distributor. lath the trend to self-amine mesh fruit and vegetable open- atieneinthereteilfood sta‘es,theretaileriebeooninxna-e isolated tron his customers. In the past he knew hie customers thread! ecvioing their needs; and by talking with thu he not only uncured uhst they wanted and that they bought, but he ob— tained intmtion that enabled him to estinte the future needs of his oatmeal. The present lack of pesonal contact and ehsenee of information es to the total retail market situation for fresh prom haveinsoneeasesreeultedintheretsilerbeingunnsreofhis relative competitive position and the tedndques he light use to increase his sales and profits. It is felt that information which reflects the purchases of the consumer in the narket place will be of value to the food retailer in reaching mhaMising decisions. A comprehensive knosledge of the fresh produce purchase pattu-ns for different types of Mite and vegetables appears to be essential if merchandisers are to pun-dues, remote, display, and price effectively and profitably. Undoubtedly new sorohandisees have gained sufficient knowledge through personal uporience and detailed rents-dc to do an effective Job. To those in thiscategorythie thesisnaybelookeduponasaoo-plenentary source e! information. However, it is believed that there are easy others she have an inadequate knowledge of puehase pattern and therefore metrelymocrule cfthmbsndconjeotureintheirnerchsndise practicese Thus, the purpose of this thesis, “An Evaluation of Gm Purchase Patterns for Selected Fresh Fruits and Vegetables'. is mold. The first purpose is the presentation or processed data collected from food purchase diaries so as to reveal to those intu- estedinrreitandvegetablewohandising thewpes otinfomatiul thatareavailableandctpossibleusofulnoestotha. hesecull moss is to evaluate these data in terms or serviceabdlity for feed retailm by applying these data to cctain specified function daunhsndisingtodeteminenhether theaedata canes-cannotbe adapted to these factions. We Significame of Stock Mating, or the moment of products through the channels of distributim to the ultimate customer, cannot be viewed as having economiciapcrtancetoonlyoneof thenanyinstitutinsl groupsinter- ested in serohandising fruit and vegetables to the collision of all othm involved in urinating of the some products. By the va-y defi. nition of merchandising as defined by the Ana-ices Marketing Associ- atimas 'thflanninginvolvedinnsrketing the rightserohandieeu‘ service at the right place, at the right time, in the right quantities and, at the right price," it would by necessity out across all poups involved in saving the particular product to its finaldestinatial.1 rumor-s, their ssrketing orgaxdsations and associations, wholesalers, and retailers are engaged in nachsndising fresh fruits and vegetables. They are all involved in promotional activities, designing and build- ing attractive packaging natal-isle and dicplm, advertising prom and sense differentiation in an effort to establish costar lqslty and pricing acupetitively so as to attract nee customers. use, an ruearchwhich delves into a study of the’mstousr's habits and rates-noes forapartioularlineoffoedprodnets affectsallin- valved groups. laypeopleuentoregardwketingssssrelyapaesiveactin dlioh the function cftheurketingagmcyisncelytcstmdreaw to supply decent. But in a dynamic econ”, in which cometition is the ecu-dinner, narketing institutions are aims attaapting to ceatenssandlargerdsundsfor their-products. Boseunoertainty my result, but this is one of the facets of economic smith. ‘3. a. Halsh. wind miet Are Its Parts?‘ Yearbook e: Aficulture, MtedStatesDWtcfigi-iculmre,w n, . ., 9 ,p. . Through this stuck, a knowledge of existing and potential new demand as represented by consumer exponenture in dollars and cents and in physical quantities is provided. Retailing is selling to the customer. The retail food store is at the end of the ass-looting chamel. In it the products of the fare in a aultituds of varieties md fares are gathered, it is that part of the (flstributive process that gives purpose to all that which has gone on before it. The basic task of the food retailer is to provide service. He hastodetarminethewants ofhis wetonmandthenacquireand price at competitive levels anywhere from a dosan to a hundred differ- ntkindsoffreshfruits andvegetablesinalloftheirvarietiesand since. The average consumer expects the narketing system to keep the goods flowing continuously into the retail outlets at prices that allow her a rising standard of living. That goal requires Maidens ascents of effort and knots-how by all marketing agencies. Hhatdo theoustonerswantr is msweallthintseknos shat us want. Tet fes questions in marketing are harda' to answer. It is an isportant question, because decisions for all marketers arebasedmthissnswc. inlpsckaginginthequantities desired by oustasers? Is the variety sufficient to attract additional sales and nu custansn‘! Are the right items being promoted as as to attract additional traffic? These are Just a few of the questions that retailers need answers to. Answers to these probleas are provided if the retailer can give a positive reply to the query; Hhat do the customers want? Marketing research has worked out various wm of learning what custoaers want. Techniques such as observation, interviewing, report- ing panels, controlled enpcinents at retail stores, etc. are asking available to the nerchandising executive factual information for decision-akinginplaescfwasteful trial anderroraethedse in efficient marketing system is one that gives the public as nearly as possible what it wants. A Improvmts in effioiacy cane paduallyinsneoonosythatisdoninatedbyfreeentcpriseand mention. The-siniobfcrtheresearohcinbcthpivateand public institution is to note interaction available to all these who can use it. legaflless of the individual problems faced by far-ere and sarketing agencies in the handling and selling of cos-cuties there isonemproblenthatperaeatestheentirenarketingstraetm-e. Thsyneedteunderetandthenarketingproeess better, becauseunder. standingisonsofthehighwaystoinprovenent. Scope of the Stuq Studies of W behavior usueJJy fall into twe classifications: (l) identification of cuteners and (2) their bwing MC. The purpose of such studies is to ascertain who buys, where, what, when, and how. This thesis will have elements of who, what and how woven throughout. ldentifioatim of mtomars sets to determine who the custoaers are. it is not sufficient to study Wing behavior patterns without knowing whose buying is involved. Hence it is necessary to identify the composition of oustwars. Incidentally, the terns 'oustoncs' and 'consuners“ sill be used interchangeably throughout this thesis, although there is a technical distimtim. A customer is the pur- ohasaofgoodsandsenioeswhileaoonaumerietheueucfgocds and services. Ganpcsitien of customers could include the complete gamut of charactu'istice such as can, income, roman, nationality, nan.- cation status, occupation, fondly size, and so on. The haying be- havia' is affected by each of these ohu'acteriaticsy however, the relative simnonnce of each would vary greatly depmding opal the iteneofpurchaees. Inthisetmhritwouldbeinpnacticalsnd meoesesry tostudy all of these eharactu-isties. Buever, fes- certain of the ea-oditiee (apples, potatoes, men-nit, emcee) anattupteillbensdstedeterninetherelationIMpbeteeenfm sine and incms to frequncy of purchase, sins of purchase, and total quantity purchased. ifistinotionshcnldbeudsbeteeenhwinghahitsandhehevic. Mtisatndeneytoeardaparticnlarectionthathssbeonealnoet spontaneous through repetition, while buying behsvier patterns repre- eeetthebwingdesimcfalargenmbcdsnstcners. Gusto-chu- inhabitesndbehvicrarenctfind, dWitnsytakeepericd efunehefore theycanbe changed. iconbinatimofdiffcent fasten-ears sleeve in operation to change food bwing behavior. Severalcftheeefaetcrsmantmes, shoppinscentu's, self- service super-nuke“, frozen foods, prepackaged produce, sewed food: that require little if any preparation, better has refriger- ating facilities, etc. Ascending toiillianipplebaon of StopandShop, noncommu- taminsbehavioreanbezr-onpedint-allittientoa2 l. Plaeecfpnrchase. 2. Ituspnrchssed. 3. Tineendfrequmefpmee. h. monument... S. Bespcnse to sales promotion devices. Date from the Hiohigan State Univu'sity Consular Panel ourtributu infant-ion a: only the second, third, and fourth ital. For purposes of this thesis only fresh fruits and vegetables will be studied from theviewmintofndxatitenyhewnnehofeachiteatvsisecrwdsht, byqamditmdollar,twfm«ncyefpurehase,andhrseasonforthe twenty-fivemoetinportanth‘eehn'uiteenivegeteblesintcrleef upmditurs. incredetailedanalreiswillbeaventcfcnrcfthetwutycfive eo-oditise. These co-edities are potatoes, apples, ranges, and grapefruit. This analysis will att-pt to alanine the relatienship betweuwhothesustacearetnrfuilysiseandinooae,andthe Wendqnanuwoftheiteasthqpurdnsed. ileothesefcar it‘s will be considered for nine of purchase so that acre satie- factmsiaedunitsofpwchuembedisplayedintheMitand vegetable esnntcs. Cmodity merchandising studies that are perti- nent tothe develspnent of the dieeuseionwill alsobepresented. 2 U. ipplehann. nStuchring Gamma BahaVior in Retail Sta-es.“ W Mow 1951. P- 17% Method of Study Customer wants and needs can be studied directly or attanpts m be made to infer than from studies of behavior. Five principal methods can be used in such consumer investigations.) 1. Heasurec of product consumtim. 2. Retail store records of sales, prices, and inventories. 3. Sales experiments in retail stores. 1:. Direct surveys. 50 W38 P311318. anhhas special advantages particular to itselfbathythesna. token each has its own individual drawbacks. Because of the diatoms ofaneperatingreportingpanalandancedforanevaluaticncfthe usefulness of these data for food retailers the fifth technique was ohoeenas baingthesost expedient forthis stew. ilthonghthemmumortingpanelhasbacnchceenbythe author as the technique to be used in investigating consumer behavior, it should not he seemed that this method is necessarily the superic tedmique. The other four techniques have their own peculiar advan- tages and defects in solving the variety of diverse prohlus arising in this field. 3?. Glemnts and 1'. Players. I'11on Thq Tell What we Went.‘I Iearhook of i%oulture, United States Department of iaioulture, ll, De s; p Do We CHAPTER II SOURCE AHD HATURE OF THE Dill Objectives of the Michigan State university Consumer Panel The MichiganmState University Consumer Panel is a.group of‘about 250 families, who~reside in.Lansing, Michigan. These faailiee report their’focd purchases weekly through the sodium.cf a diary; (See Appendix for copy of diary.) Diaries are filled.cnt so that the price per unit, quantity bought and total expenditure for each food.itan purchased are recorded. Completed diaries are then nailed to the Depart-eat of Agricultural Rhona-ice, Michigan State University whererthe data are transferred onto I.B.H; cards. The data free these food.purohase diaries werevthe primary source of inferential for this thesis. The project that suppa'ts this panel was approved in 1958 and was designed‘to»oontinnelfor tenqyears. the first diaries were subsitted to the department in February 1951; however, it was not until.aidesumner of that year'that families were reporting on a regnler'belil. Since that time the nulber ef'penel members has rel-ined.at about 250 families. The objectives of the original project were as follows: "The first is to determine the effect of price changes (both real and money) npon.the quantities of foed.purchased, and the associated timeAIag adjustment. The second objective is to determine the effect of a change in income (both real and:monqy) upon.the quantity purchased and expenditure for 10 various food products, and the associated tine-dag. the third . objective is to measure the effect of price changes and ism flanges upon substitution smug different products. In a sense, therefore, the objectives are to determine price elasticity, inane elasticity, and cross elasticity of demand. The Justification for the establishmm of a panel are as follows: “Ham surveys have established certain relationships seen: different sectors of the population at a point in tine, and new studies of tine series data provide agyegative relatiu- ships over time. But, problems arise in uplaining changes in aggregates without a knowledge of the comment parts. Also, different sectors of the component parts may not respond m ties as they would be expected to based on relationships at a point in time. A second underlying reason for the project in. volvee the time period for which cost consumption data are available, nasal: annual data. Responses to year to year manges aay not represent the responses to which producers and asrketa-smustadjustinmaweftheireeektoveek andmonth to south activities. new reseu'chers in duand analysis have expressed the need for greeenseetionel analysis over tine and for short—period data.“ In addition to providing information related to the studies d elasticities, thepanelaayaleoserveasaneasellentsecreeef infantion related to scam purchase patterns, buying habits, It is this supplementary type of infmation that the some feelssillbeacstneemltotheretailfcodaerchantaeanaidte the successful nanaguent of a fresh knit and vegetable depa't-snt. Operation of the Panel The leadership for the cgsnisation and operation of the Bichigsn State university costumer Panel has been under the direction of 1(3. 6. Web. 'Deeand Analysis free the 11.5.0. censuses Panel.” A paper delivered at a Joint nesting of the American Ste- tistice Association, and the American Fern Economics Association, Hashingtcn, n.0,, Dace-bar 30, 1953. aIbid. p Fe 11;. 11 Dr, Gold 0. Quacksnbush and Dr. Janos n. Shaffer. Dr. Shaffer's doctc'ate dissertation dealt with the methodological problems of «gamma and mung . panel? The initial step in establishing a panel was to obtain a sample . scans of the Losing population; the characta‘istios of the popu- laticnvsere not well know, since the last omens was ten years old. The senses would not on]: give an insight into the characteristics of the lensing pcpulatim, but would provide a pool from which new labors could be drawn in case prospective meshes refused to par-- ticipate «- subsequently dropped out. Iteesdeddedthateeanplecf WWZOOOfaailies, or about seva pccent of the population, would provide an adequate level of reliability, and at the me tin provide a sufficient substitution peel. the 2000 faaily sample nae selected Intention); wtflngevuyfctn'teenthresidentialaddrees frontheetreetand mustnwommtymmmuwn. 1-. Polk andprew. Atotelcf1885 intu'vieIrswere successfullyoonpleted inthespringctl950endverensedae thebeeis for thepanel sample. Frosthe ciginal supleasnb-saspleeasdraennsingfourcm- trolfectcrsehichareasfollosss (1)1ncosecfthefaaily,(2) anchor of individuals in the family, (3) education of the housewife, and (h) age of the house-life. nus sub-sample consists of 300 mu. liesu-sboutcnepercentefthepqmlaticn. Toprovide fee-nu- L. 3 J. D. Shaffc, ”Methodological Basis for the Operation of a console Purchase Panel.” Unpublished Ph. D. thesis, Departnent of Agricultural Economics, Michigan State University, 1952. cooperatore and "drop cuts" a method of whetitntion was provided which involved four control factors in addition to the four mentioned above. Incidentally, the families least liken to cooperate were those in the low or high incone groups, those where the housewife had an eighth pads or lower education, those with woken bones, those where the housewife was elderly, and those ehere'both the housewfle and husband warhead." When panel members drop out, new members are selected from the families in the ample census so as to be as such like the replaced fusily as possible. When panel members move from the city, an attunpt is made to replace them with families who are moving into the city. Provisions are also made for adding a mall anchor of newly created families so that the panel representiveness is maintained over time. iseccmsmplecensuseastakeninl95hasabesis torrevisingthe sample cal to prOVide a new pool of potential member-.5 Honctary payments are made to panel members as an incentive to return their diaries promptly without interruption. Payments range Iran five cents to fifty-five cents per diary, with the largest women being paid to those families who return their diary for the longest periods eithcnt interruption. The sandman that any member can earn is $27.60 a pear.6 hJ. D. Shaffer. “A Plan for Sampling a Changing Population Over Time." Journal of Fem Econadcs, Vol. 37, 1951» pp. 153-163. 53. ll. Riley. "Sane Kcamrcments of Commuter Demand for Heats." Gamblishod Ph. D. thesis, Department of Agricultural Economics, Michigan State University, 1951;. 6Quacloenhush, op. ci . 13 The diary is an itemized listing of all important food products categorized into major food types. In contrast to a journal type of diary the amber needs only to write in the quantities bought, price paid, and the expenditure for each food product or item purchased during the seven days, In addition, where it is appropriate, the member checks certain aspects of the item that are of interest to the department, For example, with fruits and vegetables the method of reservation is checked (fresh, frozen, canned, dried or Jam, jelly, am); starting in 1955, fresh purchases were further classified into prepackaged and bulk. Purchases of about 500 different products can be entered in the diary. Other information that is reported weekly is as follows: (1) all disposable imam received during the week, (2) the mmber of and amount spent for meals away from home, (3) masher of guest meals served at the heme, (b) changes in household memership, and (5) gift food items received, bass-men produce or killed neat items used. As mentioned before, all data are transposed onto 1.3.11. cards. Cards can be sorted into numerous classifications depending upon the information that is desired. These tabulations provide price, quantity, and expenditure infatuation by individual products or product groups for specific time periods according to fanilies or family character- istics. However, the processing burden makes it impossible for data to become available for analysis until after a period of time has elepsed. - muracterietics of the Lansing Papulation So that the information collected from panel members may be useful to retailers in other population areas, a knowledge of the Lansing population is essential. The census of population taken in 1950 pro- vides the most recent complete survey of this area. A emu-y of the pertinent etatistioe of Lansing is contained in Table 1. So that comparisons can be facilitated, etatietice of mchigan and the United States are also included. Population alaracterietice which should be of interest in stuck- ing purchase patterns of selected fresh produce itue are ae follows: 1. 2. 3. In Low percentage of non-unite as compared to both Michigan and the United States. Slightly lower than average number of persons per household. Larger percent of females 1h yearn old and over in the labor force as canpared to Urban mchigan, but a lower percent than the rest of the United Statee. (The number of uplqed fuel» has been accepted by the food industry no one out the reasons for the increased demand for convenience types of foods as opposed to food that requires larger amounts of tile in preparation.) The lever percentage of employed in manufacturing and the lower percentage nnuployed in relation to the rest of Urbenlnehigan. Thieneybepartlyezplainedbytheract that leasing is the etate capital, thus it has a large m of people who are engaged in civil service occu- patient e 15 nan 1 mammrms or LANSING POPUIATION, 1950* fl. fi‘ _. lensing Inchigsn 92,000 6.371 M. Eben mus nohigen Urban Total 1:503 H. 96,1e68 11. 105,697 8. AGharecteristics Total population Percentage increase 19140-1950 17.0 21.2 18.6 us than age (years) 30.8 29.8 30.2 31.6 30.2 Percentage 65 years old a over 8.0- 7.2 6.6 8.1 8.1 Pccentage White 3.3 7.1 9.5 10.1 10.1; Persons W household 3.16 3.132 303’ 30214 3e38 Percentage of sales 1h years old and ever in laba- force 31.5 80.0 31.9 76A “eh frontage of fenlee 11; years old and ever in labc (we 36.3 27.3 30.2 WAS 35-? We of Mr ' twee enemplqed 13.8 5.1: 5.8 5.6 h.) We upland * in “entering 33.8 110.9 ““3 29.1: 25.9 ‘ Radian income families 81:09? 83519 83815 film $3073 Mont of families with income lea than $2,000 20.7 28.1: 21.4. 32.6 38.6 Meat e! families with income more than fig“ 21e6 15a? 18e6 15.3 1203 Census of Population, "Characteristics of the Papulation.‘ Vol. II, Mean of Census, Hashington, D. 0., 1950. 16 S. Lansing is a city sith a comparatively high level of income. The median family income was $14,097 in 19149. This is seven percent higher than the rest of Urban Michigan, 19 percent higher than the Urban United States, and 33 percent higher than the United States as a total. This higher income level is further brought out by the fact that there is a smaller percentage of Lansing families with incomes of less than $2,000 and a larger percentage of Lansing families with incomes of over $6,000. Reliability of the Panel One question of importance to the researcher who uses data derived fro: a sarmle population is: Is the panel representative of the measured population? It is difficult to obtain an original sanple that is representative of the population as classified in the original sample, but the probl- is even more acute in naintaining sample reliability over a long tine pried. (Innervation of Table 2 reveals that characteristics of the popu- lation have not changed neatly in sons instances, uhile in other instances there appears to be rather wide discrepancies. is pertains to the average age of housuives, the increase in age probably repre- sents the normal age increase of the original samle over the four year period. Additional panel members will be remaited free younger housewives so that the average age of all numbers will more closely reflect the wiginal sample age of 142.? years. 17 TABLE! CHARACTERISTICS or m AVERAGE man In THE 14.3.0. comm mm m 13mm we: PERIcms comma) TO THE Avmca mum IN THE 1950 mm omens" ram: 5195‘; . Tan. T Jan. 1 June 30 Jan. 1 Moteristiu 6mm! 1952 1953 A 1953 1951: (yum) _ 1.2.1 1.5.0 1.5.6 1.6.6 1.6.6 W) 10.9 11.1 11.2 11.3 11.3 Average fasily incone last 0 d . f you- (dollars) ( msa" man. W case. 350-» Average Ian-boo! perm per family 3.28 3.39 3.29 3.13 3.22 “so or failies reporting 1885 20? 21:2 230 2“: *Rilm may 1:. 67. .Prsonel interviu survey of 1885 lensing families. h191.9 inoone after taxes. u1.951 insane after taxes. ‘1952 inoaae after taxes. .1953 insole aftc taxes. £1953 income after taxes. 18 Fluctuations in the average education of the housewife and average masher of persons per family are minor. Deviations in the former have not been over fours-tenths of a year while deviations in the lattu' varied within a range of not more than two-tenths of a person. The characteristics which have shown the greatest amount of change is the level of average fasily income. However, in appraising the reliability of the sample census two conditions should be observed. First, families are classified on the basis of last year's incense, so that on each January 1st average income of panel members is calculated based upon the previous year. Second, since 1950 the avaage level of income has been on the W trend. ”If family incomes in Lansing moved parallel to national disposable income per person, the average level of income for panel members in 1952 (based on 1951 realised income) should have been $14,359. 01 this basis the actual level appeared to be less than desired for an optimum eaxuplej however, this sacred to be corrected as the panel moved into 1953. The panel average of $11,581; for June 30, 1953 was about 22 per- cent above the 191.9 level of $3,738. This compared with an overall increase in national disposable income of 19 percent for the corresponding period. uLocal income data on gross weflly earnings of nanufactur- ing workers in Ingxan County, where Lansing is located, showed an increase of 21 percent tron the last half of 1951 to the first half of 1953. Weekly average income of panel nanbers repa‘ted on a current basis rose 16 percent during the ease pried. The difference in rate of increase could be due to the lag in wage increases received by non-manufacturing workers and to the increase in overtime pay for manufacturing sorter-s Fasiliea with fixed incomes also affected the panel average." Another criterion of sample stability is the continuity of amber of families reporting. Table 3 shows a froquency of distri- bution of families that have participated in the panel for varying 7mm .1... p. 66. 19 TABLE 3 mm or FAMILIES REPORTING oommmusu Fan SELECTED IIME PERIODS, 3! YEAR STARTED REPORTING, FEBRUARY 1951 20 m 1955. 11.3.0. consumm mum." ”‘0 0f ““3 1951 me 1952 me 1953 me 19513 me 1955 39. W1 .. a In is 17 16 3(5) 5 - n 28 12 1 10 9(8) 12 . 26 23 7 5 1700) 190.9) 21 - 51 1h 8 2 23(21) ...-. 52 - 78 28 10 mm) 7(6) ” .... 79 - 10b 10 3 11(11) .... .... 105 «- no 5 10(8) 160.6) .... 131 - 156 7 2!:(20) .... - .... .... 151 - 182 6(3) me?) an .... .... 183 - 208 3805) .... .... .... 2°? '1“ "'3‘ .2122). .34.“..- .asaa... .39... 4...... roe-i 251. 119 72 13 36 .4 a J. D. Shaffer and G. G. Quadranbush. "Cooperation and Sampling in Four 'fears of 3.3.0. consumer Panel Matias." gag Bulletin, inchigan Agricultural hperinent Station, W s P0 . "scum-1r is interpreted as meaning that the fssily sissed no em than 1. percent of the diaries in the time period they were in. () Parentbeses uses that this tuber is still inthe panel. Thq total 251a,.a'86percent of the Kay 1955 panel of 291 tubers. 20 lengths of time. is it can be noted there has been fairly good sta- bility of panel member participation since the inception of the panel in the spring of 1951. Retailing Produce in Lansing There are several sources of supply for fresh fruits and vegetables which are available to the Lansing family. First, and by far the nest i-pwtant, is the combination retail food store and the speciality fruit and vegetable market. In the retail food trade of Lansing, Hichigan there are a total of 208 stores which do a total annual volume of 336,712,000.8 is it might be expected the food retailing structure is not unusual for a city of this ails; Four large corporate chains maintain units in the city. These firms are the Great Atlantic 5. Pacific Tea Camany, Rational Tea Compaq, Kroger Calpany, and Wrigley Sta'es, Int. (lihco March 1955). A local chain operates three super- larkets, and until 1953 another local diam Operated six supermarkets. The second local chain has since been wquired by fictional Tea Company. These firns operate a total of 19 supersarkets in or near Ian-ing, all of mice carry a complete line or fresh produce.9 is indicated by the total number of stores, there are several individually canned superi- Iarkets and superettes and a large number of small neighborhood poseries finish carry promos. 8Anon. "Census of Business, Retail Trade.“ Preliminary Data. Bin-can of Census, Washington, D. 0., 1951.. 9 may’ 22: 01“, p. 63. 21 Competition is quite keen among the supermarket operators for the eonsuner's weekly produce dollar. These stores usually run single or double page ads once a seal: and a less-than-s-page ad one other day or the wed. The primary media is the Lansing State Journal, which is the only local daily newspaper. In addition, one chain uses a radio progras, while two others use spot radio announcements. Undoubtedly this advertising does such to influence the relative quantities of the dittrent varieties that are purchased in Lansing during am one week. The effect is probably such greater than a particular it. is featured by more than one chain. Another source of supply is the producer roadside stand. During the winter soothe these stands are inoperative because of elinate and lack of item to sell. During the sauna. months these stands are of greater impctance in the distribution of produce oomdities through. out the state. However, in the leasing area roadside selling of produce item is quite minportanhm Producers in this area have available to thm a public sarket located within an. corporated mm of Lansing. m. mum esned market is primarily a retail market; however, it does engage in sue sholesale transactions. In addition to farmer representatives there are several other types of sellers in this sarket. First, there 1. the truoker who hm produce from farmers and mm. it in the public market. Second, the dealer who sells outset-state produce an the market, usually citrus and bananas. Information on sales and min estimate by the County Agicultural Agent of Inghan County indicates there are probably no more than half a dozen such stands in the county. 22 use of the market by consumers has been impossible to obtain because records revealing this type of information have not been kept. The last information available was in 1930 which shows that total sales were estimated at $306,815.11 Hwevcr, this figure includes not only fruits and vegetables but items such as most, dairy products, poultry neat, eggs, baked and canned goods, cut nouns and other siscellaneous items. Also to be considered is the fact that it is unknown shat proportion of these sales are made to the ultimate customers as opposed to retailers. It can generally be concluded that the public produce mark at is a relatively unimportant source of supply for fruits and vegetables for the ultimate consumer. hceptions to this conclusion might be nude for certain sumer and fall months when quantity purchases of apples, peaches, tomatoes, and potatoes can be made there. Elna sources of supply for the urban consmser, such as has gown gardens, fruit trees, and gifts from neighbors are excluded from these data since there is no expenditure nsde. m the several different sources of supply available to the lensing census", it can quickly be ascertained that the major source of east produce supply is the retail food store. Processing of the Data is was mentioned earlier, a procedure has been established in the Department of Agricultural Economics fa- the coding of the data fru 110. Uqu. "Public Produce Markets of Michigan,“ Agricultural Went Station, Michigan State College, special Bulletin No. 268, m 1937’ p0 ”e 23 the panel diaries and the punching of them on I.B.M. cards.n From the punched card, tabulations can be made by sorting the cards for the information that is desired. When work began on this study, tabulations which disclosed the seasonal purchase patterns of fresh finite and vegetables was already available for 1952 and 1953, and during the study data on 199; expenditures became available. Intonation on amount of expenditure, percentage of total expenditure for each item, and average percent of {amines buying each week were derived from each thirteen-four week period of 1952, 1953, and 1951:. Processing of the data revealed ninor fluctuation, which will be discussed in the following four chapters. For quantity totals, only the year 1953 was used. In the interpretation of the quantity information a problm arose which involved inconsistencies or reporting units. tor example, oranges are sometimes reported on a unit basis and in other cases on a weight basis, with the reporter failing to specuy the basis of ”Went. In most instances these points of confusion can be easily corrected because the obvious intonation is apparent. than these emissions are not apparent a telephone call to the family involved often clarifies the paint in. question. For some item both gift and home gown mounts are received by panel were. This is particularly true. in bone grown itens during the armor and tall nonths. Although these are scanned 12This procedure for 1.8.8. analysis was developed by Dr. G. G. «3&th an?! me Je De “two 2h they are not reflected as purchases by panel members. For this reason they are not included in total aggregates or other amputations. In the more detailed study of the four comedities, Ganges, mpefruit, apples, and potatoes, tabulations pertaining to indi- vidual fondly purchases were studied. The families were classified in several respects. First, families were sorted into groups based upon the quantity purchased; second, they were grouped according to frequency with which they made purchases; third, families were sorted into three equal income groups; and fourth, families were sorted by amber of persons in the family. These four classifications were then applied to data-nine if they were related to either fanily charactu- istdos or frequency, else, and quantity of purchase. Because of the seasonality considerations involved in the purchase of these ituas only those faudlies who reported their diaries for 50 or more weeks wu'e used. Fran the funny individual purchase tabulation it was possible to record the size of each purchase made by the family during the week. From this information studies of the most Requently purchased siae wit and the sine unit that contribute nest to total quantities can be nude. To facilitate this study additional data bees-e availa- ble in 1955. In: diaries at the start of 1955 made it possible for panel umbers to report whether they purchased fresh fruits and vegetables in prepackaged form or in bulk fu-n. Unfortunately uly the first tilelve weeks of the year are available at this writing. 25 Despite the incompleteness in tine this information will disclose to the food merchant the wide possibilities of this type of data in helping to plan the package sise that offers the greatest potential sales appeal. I Iinitaticn of Data In addition to the limitation of consumer reporting panels mtioned heretofore, there are several limitations which are - Marent in the data itself. First, and perhaps most obvious is the limited geographical area of coverage. The city of lensing is the statistical universe for the panel. To the extent that the Lansing population has characteristics similar to other areas in which the data light be applied, the results new. be mm to be quite similar. is was pointed out in Table l, the Lansing ares wasquitesinilartootherurbanaressinmchiganandinthe United states for certain ohsractu'istics but other characteristics exhibited elements of dissimilarity. A second problem in the use of this data is the imossibility of asking comparisons on the basis of quality characteristics or variety distinctions, There is little that can be done in this area since it is believed that nest housewives could not distinguish varieties unless othesvise marked, and could not identify and report pin-chases by pads. Grades of flesh fruits and vegetables are of wholesale ' rigin and intended to facilitate trading at channels above consumer level. Therefore, it is doubtful, even if these mdes were known by «menus, that they would accurately reflect the characteristics of quality that the customer considers in making her purchase. 26 Moreover, material collected from the panel does not furnish data for the analysis of buying motivations or the effect on bduwicr of information or misinformation about the product. Although the panel is a relatively sensitive indicator of trends, it provides no knowledge of why there are non-users of a product and hence cannot point out how to capture potential markets. Finally, there are undoubtedly some errors in reporting frash - fruits and vegetables. Because fresh fruits and vegetablu consti- tute a fairly large portion of the weekly sales budget it is believed that error of mission are few. Exception to this night be lads where the fruit or vegetable it- purchased is of small quantity and is the only items purchased. more of confusion are probably small but are possible for some of the fruit items like berries and between oranges and tsngelees. Also, there is the possi- bility that acme errors are node in distinguishing between itus that are prepackaged and those that are bulk purchased. CHAPTER III PURCHASE PAWS AND CHLRACTEISTICS OF FREE POTATO PURCHASERS The potato is a basic produce itm which accounts for one-eighth to one-sixth of each dollar that is spent in the produce deparhentsi of lensing, mchigsn. Because there are both early and late varieties of potatoes and because of the keying qualities of'this product, po- tatoes are available throughout the entire year. Ithaslongbeenrecognisedbynsrchantsintheprodocebnsiness that the potato is the 'backbme' item of every fruit and vegetable department. Panel data substantiates this belief: for during most of the months of the year potatoes rank first among all fruits and use-- tables in both dollar upmditnre and quantity consumed. Years in which this is not true are years in which the average price is so low that increased purchases are not sufficient to emanate for the lower price. his feet is vividly portrayed in Table 1;. which shows the e:- penditnre per capita and the average pment of falilies buying sad veal: timing the thirteen-fonr rest periods of 1952, 1953. and 1951.. During 1952 and 1953 petatoes ranked first in expenditure with $.66 and $3.50 per capita being spent respectively for potatoes, while in 1951; fresh potatoes dropped to second position in expenditure rank with 83.02 per espits being spent. The Mating sesame of 1951-1952 and 1952-1953 were ehsracterised as high potato price years, while the mating season of 19534951; was a high production-lee price potato year. 28 new I. rm VARIATION In mammals PER wrm my AVERAGE mom or FAMILIES BUYING FRESH muons EACH um BI ran THIRTEEN-FOUR WEEK PERIODS or THE mm" Ehpenditure Per Capita Average Percent of Finilies Period much week 1952 1953 19Sh 1952 1953 195k I 8 .30 t .31: 8 .18 32% 31.1 321 II .35 .32 .11: 38 32 30 III .37. .29 .18 35 3h 31: Iv .38 .27 .15 to 36 35 v .21 .25 .21. 35 37 to V1 .39 .32 .32 1.6 1:3 1.2 VII .hh .28 .32 1.9 1.2 39 VIII .h6 .25 .30 50 3B 39 II .32 .21. .27 35 37 37 I .37 .23 .23 38 3h 31: II .52 .31 .25 28 31 31 m .35 .20 .21. * 30 21 30 XIII .23 .20 .20 2h 28 28 Total it.“ $3.50 83.02 371 351 351 e lflnmdgan State university Consumer Panel Data. 29 Being a product of inelastic demand the price is subject to large fluctuation when production varies from year to year. Thus the extreme variatim in expenditure per capita during the three years can be accounted for. Table 5 further adds to the clarification of this fact. Viewing the first two-four week periods in 1953 (high potato prices) shows that substantially the same quantity was purchased at considera- bly hifier prices than during the last two-four seek periods in 1953 (later prices). ilthcugh'nct revealed fr0n a study of panel data, there is a long tine downward; consumptim trend that is affecting the potato industry. The year to year fluctuations in expenditure tend to hide the direction of the trend. The most obvious influence that is causing a deer-Ind spiral in potato consumption is the desire fa a better diet. Hr. David 0. Helniccff of Penn Fruit Compaq pointed out at the Rational Associo atien of Food Chaim Produce Clinic of 1951; that in 1912 we ate about 115 pounds of potatoes per person while in 1952 we ate about 100 pounds of fresh and processed potatoeee-a decline of about to percent} Hand in hand with the desire for a better diet and an increased family in. come is the increased dammed for more processed convenience type foods. Perhaps then the increased use of prepackaged potatoes in both unpeeled and other prepared forms will help reverse this trend. Fresh Potato Purchase Data . seen-ding to Table 1;, about 35 percent of the families bought potatoes each week with some variance taking place during the seasons 1'1'). G. Helnicoff. ”Economic Trends in Produce Consumption." A speech presented at the National Association of Food Chains Produce Clinic on March 15, 19511. TABLES SEASONAL VARIATION IN FRESH POTATOES QUANIII'I, mmnmmu, EXPENDITURE RANK, AND mama! 0F PURCHASES DURING THE THIRTEEN-FOUR um PERIODS OF 1953* dit Average Quantity momma-o :13“ m m Pea-cent Tine Period thased in Cents M“ 2:: or 1mm.- Per Person Per Person Vegetables Emilee!) {pounds} (cents) (percent) nae 28, 1952- Jan. 25 " I». 21 Seé e32 1 32 Feb. 22 - 1hr. 21 5.6 .29 1 at Her. 22 . Apr. 18 5.6 .21 2 36 ‘9". 1, ‘I w 16 5.3 .25 z 31 NH 11 'I' an. 13 Se6 e32 1 113 July 12 «- Aug. 8 5.2 .25 2 38 ms 9 ‘I’ We 5 6e: .213 2 31 Byte 6 ’ Me 3 701 e23 1‘ 3“ Me It " Oct. 31 11.5 031 2 31 We 1 9 he 28 5.5 e20 3 27 "We ’9 ' Me 26 Se6 e20 3 23 e Eugen State Univa'sity Consular Panel Data. 31 of the year and between the years. During the first seven months of 1952 there was an unusually large masher of families buying potatoes each week. When conpared with the same time period of 1953 and 19511 this becomes very apparent. The reason for this occurrence seems to be the result of government price control regulations that were in uistenee during 1952. Potatoes were placed under ceiling regulations on January 19, 1952. According to the United Fresh Bruit and Vege— table tssociation the effect of the controls was 'to disrupt potato aux-hating, dvn-ted goat quantities of potatoes to black mm. dxannels, and caused potatoes to disappear from retail stores.“ A chat crop of potatoes which under nm'nsl conditions would hm been spread out over the long aertetdng season, due to price acting as a brake on canonization was need too quickly; so that for a short time inthespringof1952potetoeebecmeelnoetunobtainable. Theerfect at the price control was so disastrous that Congress in Jul: 1952 passed an anth wanting all fruits and vegetables {rm price control. Thiswcurrenceisborneoutby'rebleh, showingthata relatively larger number of families bought fewer potatoes sore fre- qu-nfi: in 1952 than during 1953 and 1952:. Bring 1953 data pertaining to quantity purchased became availa- ble fa us. Table 5 shows the pounds of potatoes that were purchased Want the seasons of 1953. The feature that is most outstanding about this data is the ruerkable stability of quantity purchased w. 2I. P. Beads and R. A. Seelig. 'l'ruit and Vegetable Facts and I’ointers.‘I United Fresh Fruit and Vegetable Association, Washington, De Ge, Novenbu‘ 1952c 32 during most or the year. The exception to this is between aide-August and late October when over 31 percent or the annual supply was pur- chased. This is the time of year when the late producing states such ufldfigsnarehsrvestingandnarketingthedrcrops. Promotionand nos-chandising of potatoes in large sise units of purdnase undoubtedly account for the larger average per capita quantity. fable 5 also indicates that the average pm of raulies buy- ingeaehweekis considerablylargerduringthesmeraonthsthan duringtheetherthreequerterecftheyeer. Hithabcut the sane quantity or potatoee being sold during the sussner as during the rest ettheyearitbeomes apparent thatpuraasesarebeiug-edence frequently and is slaller quantities. Since potatoes are most often nerchandised and displayed by state of origin, the rolledng discussion will sake the distinction of po- tstoeriun. Sothat thestufiofdisplaynethodswill alsobefscilic tated, a table of distributim of potato sales by state of rigin is presented. Table 6 shows the relative inc-tones of each state's potato sales in Lansing throughout the four seasons of 1953. mohigan leads in potato sales during the entire year with the Iain spring and em mention coming frat California and the fall and vinta- ee-petition from Idaho. Maine potatoes are relatively unhportant throughout the year with the higxest prOpcrtion being only 5 percent (hiring the winter months. Idaho's tap sales also occur during the winter months when thq reach 1’; percent of the total sales. Richigen's Woe in potato sales points strongly to the need It Hi¢igea re- tailers to cam and egyessively promote Michigan potatoes. 33 .88 ace-a 8838 5:858 33... anon... 8H 23.3 8” «3:: 8a «$5 8H mmdfi n. one; an 80.“ «u we”; on Rn; .58. seen a mum N new a . «mu; a 8... tea mu 3 as.“ «a omen u 8 .3833 n. 83 a 3m 0 «8 an 98.“ Been N 8... a. a2 H 8“ m 9:. one! 8 Rosa; 8 2m.» 3 93$ 3 no} 533E RS no woman 38 En gamma Home 8.22 .8 amen.” B a 858.— no BHBfiEmE Eng .6 Sm: v Ema. 3h Table 6 provides fm'ther information on the fluctuation of potato sales during the year. The first three quarters of 1953 were note- worthy by their stability in total pounds of potatoes purchased by panel “has. During the last quarter potato tonnage increased by an 15 percent compared uith each other quarter of the year. Lack of seasonality nong total potato sales during the first three quarters of the year is another reason for studying prepackaged sise units fru a state of origin basis. " Repackaging and Size of Pin-chase In studying the prepackam eise requirments for Hidiigan pota- toes tso tables of figures are necessary. Table 7 shows the distri-p hctim of Migan potato purchases by also of the unit or We during 1953. Table 8 shove the variation in sise of Minn potato purchases and quantity purchased by nethed of purchase for three-four seek periods in the first qmep of 1955. Table 6 beosls possible shoe in 1955 consular purchase diaries were changed so as to reflect whether the oonsmer lads her purchases in prepackaged or bulk fern. a look at the first quarter or 1955 (when method of purchase infomutin booms available, Table 8) shoes that about three-fourths of the purchases and three-fourths of we Michigan potato quantity saspurchasedinpropachaged fora. ThelSpoundunit of saleuas the nest upon-tent prepackaged eise, with about 60 percent of the par- chases and 50 pcoent of the quantity of all Michigan potatoes being sold at that package also. Second in importance in terms of quantity purchased was the 50 pound unit. During the first quarter of the year the 50 pound papa bagged potato accounted for between 10 to 15 percent 35 .38 75m #328 9.3.85.5 33m ante... 8a 8a 03 g 8." 8H 8." 8.” a5 a... e.» or" ma m. 3m m4 8” 4:2 o.n «.a n.« a.“ m. a.» m. 3 hm 3.. m4 m. a . «A ».m a; om 1m ed efi are. a. m4 «A o4 an o. o. h. m.. n...” «A «A a; on m.~m 9% m.~m «.mm 3h «Rm ”.3 «:3 ma es es .0 N. a. “0 He Ne 4H as ”H a... 2 33 new 3a 3a .3 3H 3 a, 3 3 3 a. a. a. e.” a .. o 00 N.“ 0.“ H00 00.” . #0.: 00H 5.“ m as JON Me No“ “s 90H As How a - H - - as as . so ............. ELI. .I to al Is ' Ie - so - - as - I! a” - as - as 3353 3392 53590 8.5.5 5555 3.5.5 bean-5 Sancho no no no no no no as 8 .388 aceeeom amoeba aaeohom ans-ohm 0303A nacho.“ 9503a 698.8“ sud ~me .2 8&5er "mm” 8 these.» ammo“ .5 82. RNA .3 not: an n S .E Wafiehm: m 2 .. .~ .96 «n. S .S 3! m S .& thereon ammo" SEE amassed .8 g a: no 33 um $285. 95.2 335“... who» no Sgfififin has Janna asses m .uemexoenoum easel m .38 85a .8538 .3353 33m eager Am 2. E Q. «a .2. am a. mu ms nu E. O O. 0. Ce 0. .0. . 0,. a .0 H .0 a W 00 H co m Ce .0 0e ee Ge 00 ee 8." 0 so be so a e e.e e. .9 so 00 ee 8H s so so se N e .H so n so H ee 8 e n." ee 4 es M ee M . ee Man as 4 cm 00 Q 00 M N fl H H 0‘ N O. .H on H H H H e. H De 0. 00 N 00 H “N .3 Hm 5 pm fl am ma 3 «a R .2 mm m N a m o N o m 3 A o u n." on es es N es H so N O. H so N 00 m - 0 H H M N ee 00 N as so 09 H H m I. H sun—Sm 5 lam ammfi no $958 55 emu. fi 88%.“ has 88.3%." 6a amalgam mo 8mg Hm Emanuele Eocene one among... 832 Bans“: no mfim E 6335 03min 37 of Michigan potato sales, 0f third importance in prepackaged fern is” thelo pound out sise, which accounted for about 5 to 8 percent of Michigan potato quantity. The on): bulk purchases that were of any consequence was the 15 pound sise sales, which amounted to about 15 percent of all Michigan sales. During the winter months Richigan retailers should carry the mority of their potato supplies in Michigan potatoes. These pota- toes should be packaged in 10, 15, and 50 poandsisos. To encourage sales oflargerunite the prieeperpoaadsheuldbeadjustedonthese three else. so that the least per pound price is earried on the 50 pound package, then the 15 pound package, and finally the 10 pound aisa. this should help to encourage larger mu of sale and at the sanetineitrsflects theloscunit oostefhmdliuglergesise packages. Tothosethatnaysish tobuylessthanalOpoundsise theretailersndhiapersonnel shouldberoadyto sorvetbeseous- tuu's needs by opening prepackaged bags and selling loose potatoes froatha. Thissillmtonlysmethesecustmersbatsillhelp build confidence in prepackaged potatoes. It is interesting to note that a potato prefu'ence study oonduted by the United States Departnut of Agriculture in 19“ showed that alaost half of all those interviewed never bought prepackaged potatoes.’ The reasm asst Request): mentioned was: “You can't see what you are getting.’ This feeling was messed by about to percent of all house- wives interviesed. Other frequently mentioned reasons were: “Can't 3"Potato Preferences Among flouemold Customers.“ United States Depart-mt of Agriculture, Miscellaneous Publication Huber 662, 191:8. 38 be bought in small quantities,‘ Wot as good quality,“ and "not available in stores.” These cements indicate the necessity of ask- ing sure that all consent packages of potatoes offered are of satis- tsctc'y quality. During the past sevm years such of this objection to prepackaged potatoes sense to have been overcome. Panel data pointsont thatabout ”percent ofalllancingpurohasesserepre- etched. During the second quarter or the year there was little change in the relative position of each size of unit sale. Of course the actual Hichigan potato sales are redueed because during the spring. the last year's orep has usually been eqlately reacted tron store agestodtsvhilethsneacmpdoes notsousintshsrveetuntillate m. Hevctheless Table 6 indicates that Inchigan'e relative pesitionisathOperoenttonnegeofallpotatoes sales duringthe spring. 'the principal exceptions compared to the sintu' quarter is that there is a relative increase in novcnent of the 10 pound Mt. Table 1 shows um 3t percent. or the purchases and 2h.5 percent of thenohiganquantitysassoldinmpoundmteduringtheseeond quartcoomparedsith 1h.) pare-mo: the purchases and9.lpareut otthequantitwduringthe firstqusrter. AtthessnetinethelS pound unit declined tron 66.7 percent of the quantity to 51.9 percent of the quantity and the large sise units or 50, 60, and 100 pounds also declined. Reasons for the increased customer interest in the smaller 10 pound unite seems to stem from one main factq'. During the spring the last of the late potato storage stocks sove into the marketing ohamls. In the late spring and early mar old potatoes tend 39 to shrivel rapidly and may if kept too long in the home become spongy and start to sprout. is total potato tonnage remains about the suns it becomes evident that the customers are purchasing smaller siaed unite nae frequentlyo Table 5 verifies this fact by indicating that on the average as additional 5 percent of the families made potato pursmases every seek during the months of April, m, June, July, and August then they do during the rest of the year. ThetMrdqusrtchad several significant changes occur-shone.- pared with the second quarter. Table 1 points out that the sain change sastheinereaseinthelargeuniteofpurchase. TheSO, 60, andlm' pmndunitsaooountedfor aboutlTporcentofthetotalquautityin comparison with 12 percent in the second quarter. This suggests that retailers should offer at least one of these sises. Il'he relative quantityothpeundandISpoundsisesdrOppedSporoent each emu-ca eiththeseccndquartc. misdeersassihssallpurdxassssassainly the result of or. increased «1. of large units. no 30 pound sale sise accounted fa- 5.6 percent of the mchigan quantity. Undoubtedly these sales m“ conposed or two 15 pound mu. It. should be re-‘ W that these are than" changes ancng the units thank. and that the actual tonnage of mchigan potatoes increased over the produce quarter. The last quarter was the nest upon-tent potato quarter in terse of total quantity purchased. Hichigsn sold more than to percent, nee-e than an other quarter. The seat important siae was again the 15 pound unit, with more than 50 percent of the mchigeu quantity being sold at that sise. The largest increase in quantity sold seem-radiothelargeunits; theSOpound, 60pound, anleOpound ho sises obtained 5.9 percent, 11.1; percent and 19.8 percent of the Inchigan quantity respectively. lhe sises the retailer should carry are the 15 pound siae and the 50 pound unit. Customers who would buy the 100 pound unit could probably be expected to bw two 50 pound pacing“; likewise the 50 pound unit could serve as a sub- stitute for the 60 pound or bushel size. The California potato does not compete idth the Michigan po- tatoes to the extent that Idaho, Maine, and other late state po- tatoes do. When lflchigan potatoes are starting to disappear tron the narket in the spring, the California Long white potato starts to appear in the retail stores. Table 6 indicates that virtually no California potatoes are on the narket during the first and last quarter of the your and for that reason a discussion of package sise during those quarta's will not be ands. During the spring the California potato was at a seasonal peak insalesvithabout amatofallpetatosalesbeingrealised by this state. table 9 indicat- about 75 parent or the purchases andTSparcentefthequantityisrealisedinthelOpoundpaokage use, 1pmtofthemantityatthe15pmndsise, andépa'oalt otthequantityinthelteSpoquange. misseena toindicate' ' that a 10 pm package would be sufficient for food retailers. The MpractieeofnsrohandisingthelOpoundsiseinepentoptote bags has the advantage of permitting the customers to more easily sbtain a miller MW of potatoes than 10 pounds if they so desire. m, when a tote bag is used, displv signs should clearly indi- cate that the nachandise must be weighed. TABLE 9 DISTRIBM‘IOH OF CALIFORNIA POTATO PURCHASES BY SIZE OF THE UNIT OF PURCHASE DURIM} 1953' 31" web 29, 1953 June 28, 1953 1.. M53 s temb pound, Percent of Percent of Percent of Percent of Purchase Quantity Purchase Quantity 1 - S 11.” M71 8.h% 3.15 6 ‘ 9 1e} 1.0 e8 e5 10 71.6 15.9 7h.o 70.8 11 i. m 100 1a} e8 1’0 15 5.0 7.h 13.0 18.7 20 2.3 h.6 2.9 5.6 30 01 zoo eee eee 1m .3 303 see see 100% - 1001 100% 1001 .Hichigan State University Consumer Panel Data. In the minor California potatoes drapped to 18 percent of the total potato sales; this occurs because of the stream of the new mchigan potatoes onto the retail scene. Table 9 discloses that 71 percent of the California quantity is sold in 10 pound class, 19 per- cent in 15 pound units, and 6 percent in 20 pound units. Thus re- tailers should continue to carry the 1d pound size package. Although the 15 pound unit is responsible for about 19 percent of the California sales when it 1. mas-ed with .11 potato sales this is less than t percent of the total potato quantity. Thus, the 15 pound unit in California potatoes appears to be unnecessary in a successful potato ' nerchandising program. In Lansing the Maine potato is quite mum in sales. Pro- duced at the seas time as the fichigan varieties it must face the 1:2 prospects of a higher transportation charge into Michigan stores than do Michigan potatoes. Essentially the Heine potato is the same type. of potato that Michigan produces. For this reason, as Table 6 shows, less than 2 percent of the total potato quantity was sold during the spring, sumac and fall. Kaine potatoes do slightly better in the sister months when thq attained 5 percent of the total sales. This appears to be the only time in which Lansing retailers would be Justi- fied in handling Maine potatoes. Idaho potatoes are shipped during the entire year and offer coupe- tition to Hichigan potatoes because of their excellent baking quality. rhq are neat inner-tent in the first quarter of the year when they are respmsible for about 11; percent of the total sales. During the amend, third, andfourthquartertheiraaleswereépereent, hperoent, and 7 percent of the total potato sales. Because of thadr Ming shares. teristiss and the high watcher acceptance of their eepericr quality, thce appears to be a need for food retailers to sell Idaho potatoes Ming the entire year. . What should the proposed package size be to merchandise Idaho potatoes? Table 10 shows the variation in size of Idaho potato purchases and the quantity pu'chased by method of purchase for the three-four week periods in the first quarter of 1955. This table gives a detailed breakdown of purchase sises by prepackaged and bulk pwchaeee. Table 11 shows the distribution of purchase sises of Idaho potatoes during each quarter of 1953. However, in this table no distinction is made between prepackaged and bulk purchases. Jasmine-la gaming-fa .38 flood soap-Bo booths. 33m 323... «a 2. E 2. ma no 3 8 an a .a 2. ee J as N so so ee ee es M es H ON 0e n 00 N N OH H 0 so a co M . m.” an 3 ma .3 a 2. a S n." d. 3.. S on n .ee N .ee ee ee H ee as so so ee a as W m n «a m m n c a. e n «a m m .. H tttttttttttttttttt escoaoawtuuttanttctttsanuuu m m MN m m m m m m t m A Maniac 885.3 message 8.28.: E358 .8392 duos co 1oz. co 1%... «o H38 do 33 so 38 do 355 assayed pace Bum poncho.“ succeed £3on aoeofim a." NS new nose. mmfi .85: mmfi «mu HEB.» Sam mmfi .5 god $3 .on you... mmfi J been... Emma 8 go can an.“ a Banana a: 30% 6.. message .8 noise: .8 Eng 55% n5 amass: 848." 9:5 .8 fine nu 855:» 3% .83 ”on: .8538 5.5.85.5 33» amazon. can con OOH 8H 8H 8d 8a 8a New be. do“ 00H ee ee so so R QeH 0e." 0e re es fa Gem at." 8 0.6 «.4 H.m «.n 4.0." «.0 a.» 0.4 mm m.“ 0.. O. .0 .9 O. F. m. a - ”H 0.2. 6.3 m.8 min 4.2. «an. “:8 was 3 Hon H.4— Nod w...” no can he God a I 0 04. W: 0.0 an." A.m ~53 o.n 0.3" m I H tttttttttttttttttttt aceobomautttottttstnutttta ”siege 8.5.5 3355 .83 59.5 8328 #338 3322 do no co co co co no do .38. accused £30qu .Eeohom £803 access." vouched Ed on audio”; 3% W?!“ -Q.. N: no.2. 1. in CI. 4 1...}. .l'." tame." go magi me E: Eafiumgumgecuggefifiagg Md mama. 1.5 During the first quarter, Table 10 shows that from 75 to no per- cent of the purchases were prepackaged purchases while 80 to 85 percent of the Idaho potato quantity was sold in prepackaged form. Most all of the prepackaged sales were made at the 10 pound sias unit. There appears to be little Justification for selling any prepackage siso other than the 10 pound unit. Since most of the bulk sales are also bought in 10 pound increments there should be several of the prepackaged bags spa: for customer inspection so that aw fears of purchasing unde- sirable merchandise will be avoided. As for the rest of the year, the sass practice seems desirable because Table 11 shows that m... 75 to 80 percent of Idaho potatoes are sold in 10 pound sises during the ‘ seeond, third, and fourth quarters. Because of the relatively mall amount of total Idaho potatoes sales aw sises other than the 10 pound package would be unjustified. ‘ Panel numbers also have two other choices in setting the state of potato origin. First they may nark potatoes that cone tron states etha- than the four previously nentiomd. These other states represent an inconsequential anotmt of the total potato supply except during the spring months. Table 6 indicates that during this time the etha- states account for 9 percent of the total quantity. This is the time when some of the early potatoes from the southm states are shipped into Michigan. The last choice the panel members may indicate is ”Don‘t know state." This represents a fairly important amount of the total potato supply. For examle, table 6 shows that 10 percent, 22 percent, 15 percent, and 7 percent of the supply is realized during each one of the quarters of the year. A look at Table 12 indicates the most popular size in the .831 and ed 5&8 no 38 e5 nouns g soon-8n 885." g Henna 80 Egg 33m denuded—s. 8” 8a 8a 8.— 8a 8H 8H 8H ee 00 ee e a.“ no 40H." “OH 8." Me“ OOH “IN “0 00; We ee 0e 8 ma a.“ or“ a.” fin and mi 3.. 8 QM.“ o.mn finn «.8 0.8 5. «.mn 4.8 m" c. m. n4 «. a.” «J in .. 8 8mm 3% «.5 .3... tom mém 99.. «.mm 3 a. a o.m n4. n.n «.n a... o. a; a - c ad fine «.8 83 do." flea 8.2 v.3 m ad ad. o.m me: «A «.8 o... «:3 a - H ee NON ee ee ee ee 00 ee n O. OOH O. O O. O O O. N O OOH O. O. I. I. O. O. H .................... BEEF.------------------ .3358 .958 hedge 8-8.8 3358 cases:— bag 320.2 8 .8 we no no as no mo 5 80.8.. 2.08.. z 80.8. 898. . .H .. 8. won. . . -8.» . 23a . a . no.-. swam m 2 .5 teenage n ma . m 83. sham." go ”395% mo 54 E 8 ans magmgdgggghoaggn «Hg M first and last quarter of the year is the 15 pound size and in the sec- ond and third quarters the 10 pound unit is in predominance. Undoubtedly nest of these potatoes cone tram the aoup of potatoes that have their state of origin in Elohigan, Idaho, Maine and California. Without know- ingtheproduoingsresitwouldbehardtodramarvoonclusionsasto bag sise for these potatoes. However, the larger umber of l to 10 pound sised units of purchase indicates the fact that smaller quantities are more desirable in the spring and some nonths. Both the old and early potatoes at this time of year are more subject to rapid deterio- ration. Packam Containcs Of interest to food retailers in the packaging and merchandising of prepackaged potatoes is the type of container naterial that has been most successful in selling potatoes. Consumer packages for pc- tatoes are usually paper bags which provide little or no oppa-tunity for inspection of the product at the time of purchase. Since visi- bility of contents appears to be a desirable feature of consumer packages for potatoes} a test stow was made in Maine to determine the custucr acceptance of transparent plastic bags for potatoes and. the problems connected with their use.5 1‘The study referred to on page 37 indicated that lack of merchan- dise visibility is a detenent to selling prepackaged potatoes. 5A. Perry. “Plastic Bags for Potato Packages.“ University of Maine, Bulletin 511;, 1953. ha Since both Maine and Michigan market similar varieties of pets- toes it seems that the results which were discovered in the Mae experiments should be adaptable to mchigan potato merchandising. Tests were run in several supermarkets in Bangor, Maine during the late an of 1952 and early winter of 1953. “nu-es containers were chosen for selling the lo pound unit. These were as follows: (1) a regular mesh window paper bag, (2) a printed polyetlvlene bag, and (3) a slatted top corrugated boat. U. S. 11me 1 grade washed potatoes were used in this six week test, During the first two weeks all three containers were sold at the same price; this gave customs a chance to become acquainted with each package without the influence of price affecting the choice, In the following two sets the price of each package was changed so as to reflect the container cost. During the final two weeks the price of the polyetln‘lene package was increased by sore than the container cost to determine what Numif aw, would be paid for potatoes packaged in this ntcisl. the results of this experiment are shown in table 13. Results from these test clearly indicate the demand by customers for product visibility in their selection of potatoes. The desire was strong «laugh that they was willing to pay a tomt precio- ovsr and above the mesh paper window bag which only gives partial visibility. It some if potatoes were priced at an amount which would cover the added cost of packaging in transparent bags that the retailer handling such potatoes would have overcome one of the road- blocks in selling packaged potatoes and would have at the same time added one more cospetitivs merchandising tool to his stock in trade. 119 TABLE 13 comm ACCEPTANCE OF mummm m may 141mm rum BAGS m) CORRIBATED BOXES IN Two mason SUPERMARKEIS FOR 1 SIX-WEEK PERIOD, 1952-1953 season“ W Type of lO-Pound Package Sales Period "Folly. Mesh rm“ ethylene Window Go Bong“ Bass Bass First 2-week Period Dec. 8-20, 1952 Retail price per lO-pound package 59! 59¢ 59‘ Percent of test sales 63.5% 25.7% 10.8% Second 2-week Period Dec. 22 , 1952- January 3. 1953 Retail price per 10-peund package 5?! 55¢ 59! Percent of test sales 62.3% 33.3% m mm 2-woek Period Jan. 5-16, 1953 5 55 5 Retail price per 10-pound package 9d e 9e L. _L g. * Pm, 0 Gite. D. So In this same study another experiment was conducted which should help to overcome resistance to potato sales. Conducted at the same tine as the other experiment,me in different supermarkets, was a test to dstmine acceptability of washed and unwashed potatoes dis- plea-ed next to each other in transparent plastic bags. During the first two weeks both packages we priced at the same amount, so that customers would be permitted to become acquainted with each package. During the nut two weeks the price of the washed potato was advanced two-cents above the price of the unwashed so as to take into account 50 the additional cost of washing.6 In the final two weeks the pm” of washed potatoes were increased four-cults above the 10 pound smashed bag to data-mine how much of a premium the customer would pay for washed potatoes. Results of this six-steak test are shown in Table 1h. TABLE 11; consume ACCMAHCB 01" new AND WASIESD Powwow IN POIIETHILENE BAGS AT Two BANGOR SUPERMABKBIIB FOR 1 3mm PERIOD, 1952-1953 SEASON“ W ‘ r 3: ‘ .. Q ' __£ype of Potatoes Sales Period _ ‘ W ash ed I! h .3 First Z-week period Dec. 8-20, 1952 7 Retail price per 10 pound package 59¢ 59¢ Percent of test sales 85.5% 111.53 Second 2w“ wiod Dec. 22, 1952- ancary 3, 1953 Retail price per 10 pound package 57s 55‘ Percent of test «.1» _ 76.8fi 23.2: Third 2-week period Jan. 546, 1953 Retail price per 10 pound package 59¢ 55‘ Meant of test «1» 62.3% 37.11 * P”, 22' dtgg 1!. 7e Evenatapruiuncftwocents sbovetheaddedadditicnalcostof washing, custcuu's showed their preference for washed potatoes by pure chasing 62.3 percent of the supply compared to 37.7 percent for un- washed potatoes. The results of these sales demonstrate that cleanli- ness has high appeal value to consumers. 63mm“ indicate the cost of washing potatoes are about 20 cents per hundredHeight. 51 Also, of interest to potato merchandisers is the effect of certain family characteristics on the total quantity of potatoes purchased, the else of purchase, and the frequency of pm‘chase. Such information should be of help in determining potential sales possibilities and directing the advertising and promotion efforts of those that are respmibls for produce sales performance. Relationship of mu]: Charactcisties to Fresh Potato Purchases Two methods are employed in studying the relationships that exist betweu family characteristics and fresh potato purchases. The first aethcd is to study families that had different total purchases of potatoes and the second method is to study families that have differ- at faldly characteristics. So that scum-ate yearly results could be obtained, only those faulies which reported 50 or more weeks were used. These 178 families were ranked into five equal groups according to the total fanin purchases and into five equal mups according to frequency of purchase. Then, these sass families were classified into income coupe and sise of the fondly mups. Thus, this provides studies of families by differences in purchase behavior and studies of pwchase behavior by differences in fanny duracteristics. In Table 15 the families wa‘e ranked into five equal groups based spa: the quantity of fresh potatoes they purchased. It is interest- ing to note, even with such a staple item such as potatoes, the wide variance in quantity consumed per capita. The range in average quantity consened per capita was 125.? pounds down to 31.0 pounds per person. Those families that consumed the largest amount of 52 cnmmu ea awescea cc neonoaaa as second “man so seesaw ace-hen ass usuaesve eases AN geese ae_ceoee ease-ice ascend .38 18a 8588 Ruthie 83m. 536s... as.” 22. 8mm» in We m.» an 9mm 3." _ flog as ~.a ~.HH wean howm n.ann 3% ”RA H83 .33 8mm» gm» a.n «.4 u.wfl mom” n.hn swan hobo coda p.02 9qu can: 383 a 3.4 and . 5m» 0.... n.mn 04m. homun n.c~m 83]. Ii .39.: 83a: Sauce. .533 seesaw} esteem} sue-m} new Aeucccnv_hfldseu «on eeoa neeeasue.hamecesv suspend _ aragmmca L, Inga—i... use eeeeaoasu;se_eusa IIIII‘ same." a has: in Engage 0.52 when .8 go 8. 82382 fleeces flags. a: genes was gnome! 2% g gang 9290.— EM on Eng 20:, IH Hannah.” 53 potatoes per capita were the families that also had the highest average sise families, the largest family income, but the lowest average per person income. These same families bought potatoes the most frequently and in the largest average sises. The families that were the lowest More of fresh potatoes were those families that had the smallest sisal families, lowest family income, and media per capita income. Also, the low purchase families were the families that bought potatoes the least often and is the smallest average also of purchase. By grouping the families according to the frequency of purchase, a large range of average amber of purchases was realised. Although the felines that bought most frequently also bought the largest total quantities 5 they didn't hay in particularly large quantities when they did sake pu-chaees. Table 16 points out that the highest one-fifth of the families bought 36 times in a year and in average sises of 13 pounds; while the aediu family bought only 15.5 times in a year, but in an average of only 18.2 pounds. Families that buy most frequent]: tend to have larger families, larger may income, but the lover per capita insane. Next, the families were grouped according to fasdly characteristics to determine to what extent, if any, fanny characteristics are related to purchases of potatoes. By mu grouping the families into three equal groups based upon pereapitainconeitispossihletcdetcsineifswtrendsinpu'chas- in are attributable to differences in income. Il'he results of this classification are wesented in Table 17. High income scans to be related to large quantities of potatoes comumed per capita. Big data one mead-sue cases .8 .28: as cease eases no seesaw .nmua .n nuance». co cornea: ee sauce.“ «man so coo-m." $8 in... 8538 hut->3 3.3m Segue... 8.« 8.“ 8A an... 3.; «~35 co .3. .9982 0mm; .33» 42.3 Hahn» mung 153w coated eon ewes-.5 0 Sam» 43 Rome as.» owns: 3... es. . a floss . «:3 5.3 «.3 «.3 0.9 Agony ensign no one... ewes-rd seam . N. .N N e. a a. nosed 53.5 o .3 8H m m 2. $3 an behaves: . 93.30 0. .o . 0. mm o. essence—:— nenuafi ewshebd m a. a m mu on we on soda-sh sound—in sodas-m sound—ah conga an» as m} on» co 3 en» no m} e5 as m} on» no m} 8.33055 finds-h IIE go 6 uncoom 35ml can 33502 do seem essences...“ , Shotguns: some” an BEER name anaemia an a no fig ems on anemone apnoea a324,,— emaem seemaou 6.3. 2H amaze flange 92.8." emcee 2. flag emcee: E anachron— mug.“ 55 “31.317 DIFFMCES In mass pon'ro PURCHASES mom 178 14.3.0. comm PANEL FAMILIES (mourns accaumn 10 PER cum. Income 1953* ...... —. .—.-_.~—.~._-..-.-—.— -———-.—. n -.....——- _— _ — ”we -n—u—o --.--.-.--~e~-.--.—--.— ..—-. "....’.. .a.......‘... Per cqggp..1nmom.'c;... Family Characteristics Highest 1/3 Middle 1/3 beset 1/3 Family Characteristics Aves-3 per capits inco 8267!: $1597 8891: Average family insane]- 3613h _ $5080 3360? W .13. of family 2.2 3.2 ' 3.3 manti‘ty Average quantity per image quantity per pm (man) 90.? as 86.1 upenditure tors Ava-age aspadi .mchigan sue. University Cmer Panel Data. 1Based a: 1952 issue as reported on January 1, 1953. 2Basedoannherefsealseatenathae, Zlaeals equaling we person. insane families purchased about 5 pccent sore potatoes per capita thandidthelesincue coup, bntzzpccmtncrepcoapitathsn loci. insole faniliee group. The high income group purchased fewer potatoes per family than the other too income groups, howcvc, the eise of the faaily was considerably smaller. Free the espendituree and quantity purchased per capita it can be determined that the 56 nidcne incense group paid more per pound than did the other groups. Since the middle income group purchased the smallest quantity per capita it appears that income is neither directly nor inversely . related to purchases of hash potatoes. This data indicates that potatoes, although considered a low cost food, are not purchased in as great a quantity per capita by low and middle income families as is potentially possible. Promotion efforts might be directed in informing the customer of the high food value received for the small expenditure. chrts might also be spent in carrying potatoes of two or more price lines. By carrying sevu'al lines; higher priced- higher quality potatoes could be used to attract high and middle income mops, while a lower priced-low.- quality line might be used to attract low income fendlies. The last classification is based on families of different sises. In Table 18 form classifications are used: (1) one person, (2) two persons: (3) three and four persons, and (h) five and are persons. the}: group isnot equal ism of families because at either extr-e fadlies are not as prevalent as the middle sised families. The families of largest size were not only the purchasers of the greatest amount of potatoes per family but were the largest purchasers per capita. Second, in purchases per capita was the family of two, third the famly of three and four, and last the family of one n-ber. The melation between family size and purchases per capita scene to be direct at only the two extremes in family size with the middle sised families showing an inverse relationship. Purchases by the large fami- lies occur about every other week compared with the smallest sised 57 TABLE 18 DIFFWCES IN mesa POTATO PURCHASES mom 178 14.3.0. consume PANEL FAMLIIL‘S (mourn) ACCORDING TO SIZE 01" mm IN 1953* Data on Purchases and ~ Si” °f PM; Pauly Characteristics One Two Three a Five 3: . A. Four Over Huber of cases 11' 61 68 32 run: muraoterietics ivu'a siae of f 1.1 2.0 3.3; 5.1: Ava-age family incense! 81529 31:85? $550!: 85960 swag per capita man-3 31:89 82386 81571: 31091 Quantity Aver guantity per 65.9 86.0 76.1: 97.5 capit peunds kpenditm‘e image enditore DQ‘ “Pit 33am $3ew 33e32 $3e56 Bias .1 Average e We Pregame: Average nonhu- cf pom-hoses 8.3 113.7 20.1: 26.8 ”mousse State University Scanner Panel Data. l'Bmsed on the number of meals eaten in the home, 21 meals equal. ingonepersen. 2 Based on the 1952 income as reported on January 1, 1953. 58 families blaring only about ova-y six weeks. Although the family of two munbm ranked second in purchases per capita they spent more in dollars per capita than the high quantity families, which indi- cates that the high ccnsmupticn families paid a lower price per pound for their potatoes. Since the relationships between family characteristics are not clear-cut precise conclusions cannot be drawn. The family of one person sise consumes the smallest quantity of potatoes and has the lowest family income (neat to the lowest per capita income). This suggests a line of potatoes which carries a relatively loo price; also suggested, since this person is probably upland and has little time for steel preparation, is the adoption of a fresh potato that is easily and quickly prepared. This latter suggestion could only beadopted when technology develops a potato that can be pre- peeled and pro—cut so as to withstand quality deterioration. A higher priced and probably a higher quality potato seems as be preferred by the families of two, three, and four. A loll price potato is purchased by the large families. no. is prebehly the result of the lee per capita income received by these families. Judgim from the high frequency of purchase, potatoes would be a good item to feature in store advu-tisaoents which aim to create a price messiah. The high We also of purchase and relative infrequency of purchase chich is a characteristic of small size families points out the necessity for retailers to have loose potatoes readily available for these custom“. In! constmpticn should be coupled uith smaller 59 sised purchases bought more frequently rather than the reverse. This would reduce hues spoilage and arable the aerolnnt to sell other reduce “as due to the increased potato patronage. Summary The potato is the tap item in tonnage movement and usually highest in custaner expenditure dollars. tears in which this is not true are years in which the potato supply is heavy and consequently the price is low. In loosing, the potatoes Ron Michigan are the nest pepular. The most inpa'tant reason for this favorable position of Michigan potatoes is the lower cost involved in marketing them in Michigan. Panel data indicates that the mohigan potatoes were purchased in prepackaged form by over three-foIn-ths cf the customers during the winter months. Prepaokaged size should be a 10 pound, 15 pound, and 50 pound with prices per pound adjusted downlard as the bag siae increases. During the spring nohigan tonnage and relative position with other states Mos-eases. Bag sises should be the 10 and 15 pound sise during the spring. Pbre rapid deterioration of old stmge potatoes and new potatoes is the reason for the customers more Request purchases of mallsr quantities. In the sumac, meldgan tonnage increases over the second period and Hichigan retailm should-again adopt the larger class unit of either 50, 60, or 100 pounds. The last quartc' is the most important Michigan potato quartu' when tonnage was to percent more than cry other quarter. The sises retailers should carry are the 15 pound pack and either the 50 pound bag or bushel. 60 When Michigan potatoes are disappearing from the market, California potatoes are beginning to appear. During the spring this potato accounted for 22 percent of all potato sales. Since the 10 pound bag sold over 70 parent of the quantity, it is the only size retailers should carry dur- ing both the spring and sumer. Idaho potatoes compete with Michigan potatoes during the entire year, however, they are most competitive dur- ing the winter. About 80 pecan of the quantity were prepackaged sales. Most all of prepackaged sales were made at the 10 pound sise unit. There appears to be little Justification for handling any other prepackaged sise. Families that were the lowest purchasers of fresh potatoes were those families with the smallest sised families, lowest family income, and nediun per capita income. These sane families bought potatoes the least often and the smallest average sise purchases. Families that buy most frequently buy on the average of 36 times a year but in slsller average sises than those families that buy only half as often. Families that buy most frequently tend to have larger families, larger family income, but the lowest per capita income. Purchases of potatoes are not positively related to income, although the highest income mop purchased sore potatoes per capita than did the other income groups. The correlation between family size and purchases of potatoes per capita seem to be directly related at the two extremes in family sise with the middle sized families show- ing an inverse relationship. From this data on family characteristics several conclusions can be‘drswn. First, there appears a need for customer education on the 61 low cost, high food value that can be derived from potatoes. Second, he price lines of potatoes should be carried. _‘I'his would appeal to those that desire a higher quality of potato and are willing to pay a premium for quality and at the one time it would permit a low price oupetitive potato for appeal to those fmilies that have 1w incense. Third, the infrequent purchase of potatoes by small families, points out a need for retailers to make loose potatoes readily available to those" customers that purchase only a small total amount of potatoes. CHAPTER IV Panama PATTMS AND CHARACTERISTICS OF FRESH APPLE PURCHASERS “The bulk of the apple crop is purchased by consumers as fresh apples. Over three-fourths of the panel families' expenditure for apples were for fresh apples in 1952 and 195331 ‘ Fresh Apple Put-chase Data During the three years 1952, 1953. and 1951;. apple. ranked third asong fruits and vegetables in constnnsr dollar expenditure, with 82.17 beingspentpcpersosinl953 andwithonly slightvariations tree this accent in 1952 and 1951:. This enemas-es with a per capita dollar upenditure of $3.50 fer potatoes and $3.25 fer bananas, which ranked one and two, respectively. In terms of quantity amassed per person, apple purchases animated to slightly ever 25 pends. Fr 1953. apples stepcchasedbyanavu-ageeffipermtofthefssiliesevcywesk. his means that apples rusted seventh in this atom, with head lettuce, bananas, potatoes, celery, carrots and ranges having a supwiu' frequency of purchases. Thee is, of course, considerable variation in dollar espemiture, quantity expenditure, and frequency of purchase by seasons of the year. 1.1. D. Shaffer and G. G. Quackenhush. ”Consume Purchases of Apples." Michigan Agricultural maeriment Station, East Lansing, Special Bulletin hos, 1955, p. 7. 63 Table 1? depicts this infomatim for the thirteen-four week periods of 1953. " is can be noted in Table 19, apple expenditure during the twelve wed: period from September 6, 1953, until Novmnber 28, 1953, was 37 percent of total annual dollar expenditure, while the quantity bought was almost 50 percent of the annual total. In contrast, during the twelve week period from May 17, 1953 to August 8, 1953, 7 percent of the yscly dollar expenditure was made and 5 percent of the total apple quantity was purchased. The month of October is the nest inns-tent month for apple sales in both dollars and pounds. Purchases averaged about 5-1/2 pounds per person for four weeks or 21 percent of the annual purchases. Results In. a mchigan State University Consumer Panel Repa't reveal that the weighted average price for apples was also the lowest during this Mgh purchase period.)2 For «ample, the average prices per pound in canto-bu, October, and Hovember were 8.066, $.05? and 3.073, respectively. In contrast, during the low purchase period, m- m 17, 1953, until August 8, 1953. ave-ago prices pee poms were 8.15, 8.113 and 8.10 too- each four week interval. However, in using non... State unmoity Consumer Panel price information the follow- ing limitations should be observed: (1) no distinction totes-a grade a' quality is lads, (2) price difference between varieties is not reported, and (3) the sises of purchase units will cause fluctuations in price per pound. Although price and the quantity purchased appear 2123111., p. 11. TABLE 19 SEASONAL VARIATION IN QUE-HIM, mzprtzmrrurm, 3:131:01?er RANK, 13m mzqmzm OF PURCHASE OF APPLES, 1953* Average Ex ndit Time Period , gazinégn fimfingiertgrue‘ $1;ng fifié‘gg Vegetables Week Tpmmds) (percent) Dec. 28, 1952- Jan. 214, 1953 2.00 3 .19 3 29 Jan. 25 - Feb. 21 1.1.9 .17 h 26 Feb. 22 - Her. 21 1.98 .23 3 32 Ear. 22 - Apr. 18 1.07 .12. S 22 Apr. 19 - May 16 .97 .1h 5 21 New 17 - June 13 .53 .08 8 13 June 11; - July 11 .13 .02 22 111 m 12 «- Aug. 8 .55 .05 18 10 Ana. 9 - Sept. 5 1.511 .11 8 20 Sept. 6 - Oct. 3 3.73 .25 2 311 Oct. 14 - Oct. 31 5.1.5 .31 1 35 Nov. 1 - Nov. 28 3.19 .23 2 29 Nov. 29 - Dec. 26 2.72 .23 2 30 1111111131 25.35 $2.15 3 2; A‘__ *Hichigan State University Consumer Panel Data. 65 to be related, there are undoubtedly other merchandising practices that influence purchases. For example, this same report points out that me purchases was made during the four week period following rehruary 22, 1953. than during the four week period preceding that date, and at substantially the same price. Another unique feature of the mchigan State University Bummer Panel data is the report of the We of families bwing a par- ticular modity each week. Table 19 depicts this information for fresh apples during 1953. is can be noted, the variation in average percentage of tandlies buying each week does not fluctuate as lush as one scold expect. However, Table 20 show that differences in seaeml variations of expendittn-e between years seen to be related to the average percentage of families bowing each week. For sample, during the last four wed: period of the years 1952, 1953a and l95h, the seasonal variatia in expenditure Ins 3.18, 8.23, and $.16 per person, respectively; during the same periods the percentage of families buying weekly m 25, 30 and 22 Wt, respectively. It appears that it improved merchandising practices had been followed in Decenher of 1952 and 19514 the amenditm-e for apples could have been increased. One or the problems involved in the merchandising or fresh fruits ad vegetables in prepackaged form is the selection of the most salable constmer unito.3 what would be the most desirable weight for packages 3Prepackaged lsrehandise involves the sale of items in filn, carton, paper bag or in other types of unitised wraps. The dis- tinctia betveen prepackaged and self service should be made. i self-antics produce department is not necessarily one that carries mackaged item. TABLE 20 mum vastly-moo: m mezrmwm m CAT’ITA AND AVERAGE Farmer 02 mums BUYING W33 APPLES anon annex memo THE THIRTEEN-FOUR m: FTRIODS cs 1952, 1953, and 195h* i... l‘scpenditure Per Capita Average Perc ent of Families Period M Buyin Each Week 1952 1953 195h 1952 19 3 1955’ I 3.19 3.13 1.19 31% 293 293 II .18 .17 .22 32 26 35 III .17 .23 .18 29 32 29 Iv .1h .11. .17 23 22 26 v .09 .11: .12 16 21 21 VI .03 .08 .08 6 13 11: VII .02 .02 .011 3 h 7 VIII .07 .05 .06 12 10 13 Ix .11 .11 .11 21 20 18 x .23 .25 .221 36 3h 3h XI .ho - .31 .35 1.2 35 £13 III .23 .23 .28 29 29 32 XIII .18 .23 .16 25 3o 22 a Michigan State University Consmer Panel Data. 67 our applesufcr example, 1, 3 or 8 pounds? Should the amber be standard all year aroiund? These are two of the questions that need answa-ing so that Qerators may package quantities that will appeal to the largest lumber of customers and yet sell the largest possible quantities. A study of the sales distribution of apples should help in making these deoisims. Repackaging and Size of Melissa 1.81. 21 depicts the relative upon-tone. of each 'nnit of pitch-.80 in terms of frequency of purchase and quantity bought. During 1953, panel where were not required to indicate whether the purchases were in bulk or prepackaged form; thus, Table 21 is a combination of both types of purchases. The average siae of purchase for the year was about 6.2/3 pounds and the total W of mohigon State Universiw Omar Panel ambers' pin-chases, on which this table is based, was 2,910. There is of course considerable variation in to. distribution of sins of purchase between the seasons. Ming the last quarter of 1953. when the majority or apples reach nativity, the greatest number cf apple purchases were leads and the largest average purchases were realised. In contrast to the rest of the year, better than 110 percent of the apples were bought in larger than half—bushel sises while a smaller percentage of the apples were bought in the duller nuts. The nest popular size in terms of author of mistomen twine was the 3 pound size, while the h and 5 pound sises were second and third, respectively. However, the bulk of the apples were .o1d between 31 and to pounds, to. bushel being the most impor- tent single eise in that range. .nH a .mmmH .mo: 5835 18am $533 and 3.3.x e355 godsE< asses. ..8H&< no 85 538.. 3.35.35 6 .o .8. than... .n r. 5.6 39.54.. 8.. .3... 32H .3 35$ 82. than no ends .95 32” mam mum 3m 5 Ian ea 6.. Oen Ne eee eee eee eee see. es. 3.". “H Ne." He eee see eee eee see eee ONH‘ b“ O.“ a. see eee eee 000 He: Ne a . MN Jeo De." “eH Ne eee eee eee Dee NF I “a MeNN Mona Ne.“ . @efi see see H.“ We Q: . FM “an Ne.” eee ees eee see FeH Me a as “N «.3 o... .3. .2 5” e. HAH «A .a .. HH n.» to 3H H.0H 3. H... a... .3 «H .. HH ~.H «.H n.~ o.H H... o. a... 4-H S 4e ‘0 HON 00 a.“ De.” OeH We a pH H.“ a.H «.H .3 ~.H e...“ fiH a a. nod 40 in. “O.” a” ”OH 0..” N “em “00 NOOH Med." De: he“ 0e” “.0 Q .3 «.2 o... e.“ .3. a... 4.2 EH m .3. 3: 4.2 3-H £3 a. 3 ”.8 o6... a em 3H H.» EH 3... 3H .12 0.2 H 3 3 3 or: 3H . as... E 3: u M. “ON 00 We“ QOH mom OOH Hem H -I--.Iaeasas--- ....... -Uflg'-----'---.l-'-‘-aeluse Aggy 3358 83.3.5.— .HaHfiga 83th; #353 35.8.. 335.8 835.5 .838 no .3 . he .8 on e. .3 no .5 80.30. §0h0. 80 COO-H0. 5080. "Oswa- . 0.8 . .-. .u . HO .. £50. a @a 850 m .. H. w- 3...: 3.25 mmmH .E 895.95 .mmmH 3 83. £2 .E 5.3. «RH. 3 .3808 emSHfim ems... Eflafimazfioe $5.22 .8 .58 was .5 Hum Hm mmmEoEK 5&4 mam-H .5 zoEmHEmHa “an!“ - 69 During the fall and winter of 1950. the New York State College of Agioulturo conducted a study of merchandising practim in con- hectic: with New for]: But. applem.h The objective of this work was to find thoaox method: at retailing applet: that were most effective in modern W0“. A mubc of merchandising practices were tasted, such an the use of a bulk displq, 1 display of prepackaged npplu, and a display combining bulk and prepnckngod apples; the use of different natal-1:10 for prepackaging apples; ad the pantie. at varying the sin of units in which apple- um priced and puck-.306. The "lulu or this twain weak mt (Soptmba' 25 to Doc-abut 16, 1950) are shown in Table 22. TABLE 22 ms EFFECT OF mcmnmsme PRACTICES on APPLE sum" Mundane Prnotiao Pound- Sold Parloocuatomm Bulimia 2pouzldnm.t..............o n hpNDthooooooooo'ooocoo 13 Prepaokagoddiaplay hpoundodlqahnnobnglooo...... 1-8 Balkandprapndagoddiapm ZpMncmqahanebngmdbulk..... 13 hpoundoollqrhnnobagmdbulk ooo‘co 20 hpound polyothyleno bagandbulk . .. . 23 GWpdyoflvlmobagnndbulk .... 28 1. 13371., E. cit.’ p. 2. h!» 8. Davis. ”plying Inproved Apple Merchandising Punting in Rot-.11 Stern." Canon University Agiculturnl madman. Station, LE. 80?, 1952. 7O Theee teete deer]: show that a bulk display ie not an effective neth of selling epploe in modern self-service etoree. The out eetiefeotory method were to be a combination or bulk end pre- packaged dieplm, with the prepackeged epplee fisplayed in 6 pound unite with e Wlme beg container. Although Michigan Stete “vanity Consumer Panel date in 1953 do not make the dietinctien Wennethodeetdieplnu'typeetuteriel owning, theyde indicate nice of purchase. is hee dread: been mentioned, 37 to 1:8 pound unite account for better then 22 percent of total epple eelee, whileépemdmu eooount for eligluyleuthen6peroent ofthe totel eelee. Thu-e ere eeverel rescue that would tend to eooomt fer thie app-rent W. Fir-t, large eiee mu e! applee were not teeted,beeeneenemet¢eedonoteteoltmitee!mpeendecme for en model! time during thie thirteen week pm‘iod... Snood, ooubie netione of the Inner eieed unite were not displayed. For ample, 2.1:, mépmmtammttuudmmw. nerd, any one miety, the mum, m tested, while nohigen State Univenity commen- Penal dete do not make each aetinctione. Although not e diner-m, the elementetmberetpueheeeemnotteeted. The 6ponndunitmee11mepplee Mattheenetinethenxi-n mt of More new not have been eppeeled to. The unit of eiee mtnetonlybeoepebleoteeningelerge quentityofepplee,~it M be e line that e large mberofehogypenwin pin-chem. Bend upm Richie-n State University Omar Panel date, Beverel tentative conclusion can be drum about eiee of prepackaged 71 unit for displaying during the last quarter of the year. In the enallu' eieee (1 to 6 pounde) e variety of sizes from 3 through 6 pounde die- plved with bulk eeeme to be neat advantageous. These unite could be packaged in transparent fill and nerdlandieed at a eingle price, eneh quomdefcreonamcentecr-Spoundefor eoneotherprice. Since displaye of between 1 to 6 pounde account for nearly 65 percent of the purchelee, ad 25 percent of the quantity, prepackaged unite in theee eieee would appeal to a large unbel- ct cuetcnere. There aleo appeare tcbeaneedfor aninternediate eieepackageofaboutupomde, lime almeet 1 percent of the purchaeee are mede et this unit eiae while 8 percent of the quantity is sold. Two large eiee unite are also needed: e half-huehel and a bushel eiee unit. Although a range of eiaee ie in. cluded in Table 21, an eetinate o: the cannon ealee potential at theee tuceineewculdbcebouthperoentofthepm'chaeee andBSpercentet the quantity. Mingintothe eecondncetimpm'tantquarterinternecfnumberef pcrchaeee, the period from December 28, 1952 to Burch 28, 1953, lppm tccontreetineeveralreepeeteeiththetall quarter. Thehpeundunit predainatee ee thencet imam-tent eingle unit both in terms of per- oectage of pm 0.” mtege of quantity. In both theee Cit“. the caller eiaee becone we laps-tent then during the previcee quarter. WWlthroughépcundunite, itcanbereedily eeenthatthq constitute bettu' than 85 percent of .n purchases and 60 percent of the quenuty. At the eane tine the largu- eieee diminish in input-nee. For example, the 37 to M pound eiee eccounte for 5 percent of the quantity W with 22 percent during the fell quarter. 72 Does this change in distribution pattern mean the retailer should change the package sises he offers for display? to answer this question additional help can be obtained tron 11m Table 23. Data became available in 1955 which distinguished the type of purchase into the two categories of bulk and prepackaged. Tabulations for the first twelve vets of 1955 can be used as supplementary data for this particular period. - Approximately 55 percent of the purchases were made in the pre- packagedfmehilehSpercutverepurehasediathebulktm-a. This fact in itself points strongly toward the same conclusions reached in the New York study, in that apples displayed in both bulk and prepackaged fora increased the pounds sold per 100 answer»; There are several reasmsshyappleaalescanina-easetdxenconsuaerpaekagesareadded. It has long been recognised that additional itus placed on sale in a grocery store increase total sales. Adding consumer packages to bulk displays of apples essentially adds anothu' ital for consumers to con- sidu' in asking their purchases. Then there are always ease shoppce inahurryuhodcnotwmt towait {reproduce clerk teseigh out apples frame a bulk display. They find it more convenient to select a unitieed 13am Also, some customers like to know the exact cost or an it. and a produce clerk seldom reighe out the exact amount at apples Iron a bulk display. Other customers like to select their spples m- a bulk display. With both mm packages and apples displayed in bulk, the retailer is more nearly able to satisfy "We 5113111., p. 2e 73 oxHom ended m .cemexoaaoum eases m .38 Head .8538 3285:: 33m 535:... 3 4m .3 cm mm 9 2 R on a. S mm H38 m 4 H H 0N .. m .. 4H N N .. 3 9 :N m be H O. m 00 N e. m N M H MN I: NH e H n .. m n N N a H N H HH .. N m H a H m N m H h H m H o N a N a n .— m a m m n a m m mm m on w on m an o 4N m 3 a N N m m m m e o o 5 0H. NH n m N .. NH m a H . 0H a N HH o N H O. M O. O. .0 H O. H 1. N D. H .................. “gm ' I. I. - I- .l I. - l. as I as as .l - I. ' . . m m m m m m m MI mi. 5358 838E kahuna Senate 33:55 3295 no no no «on hogan £893 amoeba gush 983.8 vacuum 5“ mmmH 0N nah: wwNH . fiafiw wwwH .NM answer 33 RS .2 been: mmmeN $3 3.. banana ES .6 $348 an... Ea 2H 88%» he: gels moo mméomfi no 8mg E chancel 53,30 52 $325 ”3%: no 33 E .6555 n“ mama. 7h During the first quarter of the year the percentage of prepackaged merchandise sold fluctuated fmm 1414 to 5’4 percent of the total quantity. The last popular prepackaged siss was the h pound unit, with the 3 and 2 pound unite also being important. Very few prepackaged sales Ice lads in large units. Although most retailers do not prepackage in units of ever 5 pounds, it would be possible for consumers to make . purchases of several mailer eise packaged units at one tine. The largest quantities of bulk apples are sold in the large sise units, may, the bushel and half-bushel quantities. From the data contained in Tables 21 and 23 the following recu- aendations can be made as to the sizes of apple displays to use during thefirstquartsrcftheyear. Averietycf siseefrou2throug1$ ponds offers the greatest potential, with the najwity of sises being packaged in the 3 and 1; pound units. in intermediate sized prepackaged untrightbeelininatedu'carriedinonhenallm Thehalf- bashel, 2111mm! eise, sons to ranina'gocd chdee for continuance. The bushel eise unit should be discontinued. Inastudytodetu'nineifmioapples eccldbegivanmem- petitive advantage by cffcim attractive units in two or three siae mite at the swaths, Ralth. ShermanandGlm H. latchell efmio State University deta'mined that the offering of several sises of bags of apples increased apple sales in the stm'oe where they were offered.6 The results of this study, conducted during the fall of 1951 and 1952, were very similar for the two years. In 1951, when 3, 5, and 10 poms! 6R. W. Sherman and Glen H. Mitchell. "Sell Pro-Pack Apples in a Variety of Sines." Pro essive Grocer, January 1951;, p. 92. 75 units was tried the statistical analysis showed that apple sales were increased by about 15 percent compared with when only one size (5 pounds) was used. The importers» hm is that it gives the retailer 15 percent more volume of apples while helping to lower his service cost. During the second year the oombinatiu was changed to include a 3, h, and 8 pound pack. Substantiallyv the same results were obtained in both years, with the exception that in 1952 the 8 pound Mags sold a slightly highc percentage of the total apple sales than had the 10 pound wt the year before. is a side light, a popular theory held snug me oftheretailtredensdispelled. Ithadalwsysbeanbelievedthata large siae unit with a high unit price would not sell. Outrary to this belief, it was found that the 8 pound unit witha price tag of over a dollar had no detrimtal effect on the percentage of sales mounted by the large bag. The second quarter or 1953. in the fichigan sum University consular Panel data, was characta'ised by a low number of purchases and the lowest average use of purchase. During this period panel were made 523 pu‘ohaaee which averaged 3.61 pounds per porcine-e. WiththemeptionefabouthperoentefapplesboughtinthelBte 2!; pound range, there were virtually no apples bought in large mounts. compared with the previous two quartu's, there was an increasingly larger prepa-tion of apples W in 2. 3. h, and 6 pound unite. Pu- prepeaaging purposes a combination of. small sises should be used. a few intc'nsdiate sise packages of 11 w 12 pounds might be successfully tried. It would be Weary to carry the large size bushel or half- bushel at this season. 16 In the third quarter of 1953 the smallest umber of purchases were lads, while avnge purchases amped to 6.6!; pounds per purchase. it this time of the year the early smer apples such as Iellow Trans- parent, Gravenstaln, and Duchess make their first appearance. sises ' these varieties as primarily used the cooking and-baking, the larger ‘ sise of purchase, as indicated in table 21, can partially be elplained. Also the traditional fall termites, Jonathan, McIntosh, and Healthy, begin their was to ma during Septanber. A partial»): strong favorite in terns of quantity was the range of sises betweu 13 ad 21; pounds. This manually means a large Wage or halt-bushel sises were padzased. use important in quantity purchase was the 11 to 12 pound pmchese which accounted for 18 pwcent of the total quantity. This strongly suppa-ts the contention that there is a need for a packagedaboutlotelZpeeads.‘ Again,thenallsiseewerepopn1ar fa frequency of purchase, with the 2, 3, h and 6 pound units account- ing for 65 pucent of the purchases and 35 percent of the quantity. 1. Variety of these small sizes would be ideal for packaging purposes. A recent stucw by the {hated States Department of Agriculture con- rims new of the conclusions drawn from the Michigan: State University cam Panel data as pertains to bag also in the alsaller units] Also Winthie etudyie thecftenaskedqoestion, 'Hhatprioiuunit should be need?“' he study was conducted during the wintc months in Pittsburgh, minim. The findings indicated that more spplss T— 1inon. “Merchandising Studies in Supermarkets - lpples, Lettuce and Tomatoes,“ Agricultural Harketing Smice, United States Depart- sent at Agriculture, Hashingtcn, n. 0., 1955. 77 were sold per 100 customers, using a display of plain polyetkwlene bags lit)! Heights Wing Iron 2 to 6 pounds “catchaveight' in m- nation with bulk Ind based upon a. 3 pound pricing unit.8 Displays ci‘ a printed 5 pound polyetlwlens bag offered in combination with hull: and based on a 5 pound pricing unit resulted in the second largest quantity of sales. Isble 2!; depicts the {our merchandising methods tested. TABLE 21: QUANTITY OF muss son: PER 100 003mm BI spscn‘ro maons II 12 roan can; swam, PITTSBURGH, rmsmmtn lei-mm Hethcd Sales Pr 100 1. Plain polyethylne bags of weights varying fro-Ztoopcunde, in combination with bulk, manth‘Bpmpriflm.‘eeeves-lessee 10.751ble 2. Printed polyetlwlene bags of 5 pounds in eubimuonsithbnlk, andsithsSpm prici unit 0 e e e e e e e e e s e e e e e e e e e e 9e9o lbfl. 3. Plain pouetlvlene bags of weights vary- ingfmztoépcunde,sithmtbulk,uitha 3 pound.prioe unit e e e e e e e e e e e e a e e e e e 9.18 1b.. 1:. Printed polyetlvlene bags of 5 pounds, combined sith bulk, and having a 5 pound prid. unit 6 e e e e e s e 9 e e e e e e a e e e e e e 3.92 Ible 'ucrchccdi-ing Studies in Supermarkets, . 4 cit... p. 3. 8"('."I|t1:h--Ired.ght," s one term in retail trade usage, refers to package seights that are not predetermined, but are arrived at eftu' the packages are filled. 78 Among the four methods tested a significant difference prevailed only between the first and last methods. The data indicated a strong preference fa a choice of 2 to 6 pound IMatch-weight" bags of apples offa'ed at a 3 pound price unit. During these tests about 85 percent of the apples were sold in bags when a combination of bulk and bags was displayed, with the largest proportim of the small sise packages being purchases in the 3 and 1: pound bags. While it is ilpossible to tell from this United States DepartIent of Agriculture study shether the adaption of these recon-tended sise units would result in more apple sales for all retailers as a group, than are good enough cosmetitively to fen-ca thmelves into stores, and these stores adopting than first will benefit first. Relationship of Family Quaracterietics to Fresh Apple Purchases or interest to produce nerchandieers is the relationship that exists between family characteristics and the quantities purchased, the sin of purchase, and the frequenw of WC. One of the sethods by which this information can be obtained is to study the fuilies that had diffu-ent total purchases of fresh apples. Because of the sesame}. nature of apple purchases, only those families shich reported 50 or sore weeks were used. These 178 families were retired into five equal groups according to the quantity of apples they pur- chased. Table 25 sunrises the information on these five groups by factors that are believed to be related to consumptim. hdividual family purchases ranged from no purchases at all to 372 pounds, with the average purchase of the top one-fifth of the families being 200 .nmma .H gonna do nuance?" no 0305 Nmmu no nomemm .839. 333 one.» 38 833a doused eflo gave edges Hm none: as dense eases no venom." .m." .c . S .moo flange 38cm 5593 can induce «85.899 153.534 chance. .63 e co 835.2 8858c. .5383an .e .o o5 833m .n .e nob encode: ad. ed in fin 9n hung on can. cent: .32“ mood. was” 3mg 32“ mesons conned use emshzi «3.3 as. 83» 82a momma neg 5355 onshore ‘ Boo-«taunts .35— em eN eN e o Oi 809 a .n o m n H3 ink“: mmv em.”- e335. m m b OH 4H Amocsoav «any. ha . cannon—fl no ends emcee: exam o 05 non «com 8030.99 no hang 03.3.5 p 4 H. a 0:: w snack in md 93 min 33 . 335% 33 co canto owotma c an on mm mm succeed «on hugfiw 5%» ma 3 8 o2 o8 €3ch has: con nomads Ho 5353". omens: p.358 endgame goofing 333cm Bud—”5..” enough .5 co m} one co m} 23 co m} 23 co m} .25 co m} oofioetoccuono panic whose.“ 530m p.55. vacuum voenwdm one 33:25 no 33 consona— oficma nook cc tees No ocean cognac on?” an a: he nonhuman 3%: name .8 E38 2. Segue awesome undead has genes 5.3. m5 one: 532.; 3a: mg 2. flag exceed 5 Engage mu mama. 80 pounds contrasted with 15 pounds for the lowest one-fifth. There appears to be a direct relationship between the quantity of apples purchased pr family and average sise of purchase per tank, average number of pure chasee, avu'age size of family and, to a lesser extent, average family imme. The families buying the smallest quantity of apples have fewer family members, the highest per capita income, but the lowest average family- ineome. V Comparisons were also made by ranking the families according to number of weekly purchases and average size of purchases per family. This information is presented in Tables 26 and 27. Families varied greatly in the number of times they made week]; purelwmes.9 The range was Tree 0 to h2 times in a yearly period. The flee-age number of purchases for the highest and lowest groups was 25.9 and 3.5, respectively. . Those families buying lost frequently tended to buy smaller quantities of apples per purchase, pay a higher price for apples, buy a total larger quantity, and to have larger families and family incomes, but a laser per capita income. The opposite charac- teristiee prevail among the menu that purchased a run- unbou- d tines. Barking the families by average size of purchases vividly portrays the fact that Quantity purchases are bought at considu'ably lower prices. There seems to be a direct relationship between average size of purchase and total quantity of apples that are purchased. There also appears to 9Each week's purchase is termed as a single purchase as panel data are reported on a weekly basis. This may tend to slightly undu‘estinate the actual number of purchases. .ieeegsveeinwselieegeeegucgce} .nmfi J gefiétgaflflgu .Safiagxofihqslesgn .S .e .mmfi .3: s23?» 1925 mass: and .83 8 3.58.5 H.533 «he gets. fluoaaettasthago. .33 .eaeestflufl .n eggs-3 .44. a; a; Tn e3 aha-nut queers: 02.5 «8.3 39.3 35.3 3.8 melon.“ use! .8.— ease-sq 393 3%. $3. Ram. Rem» «885 bean is... 4 . Savages-.15 Rule oi a.» Q..“ ed 3: 383 13k .285 8v... S «a n. . €2.85 w m «.3495.» as e3- amuse d 83 R 8 2. 8a . ease has: 2» ts arse-new :28 4 .3355 mm 3. 4.3 93 “.3 “ghee:tn§sete . g 8.3....— 83! Ii 83......— Ii 1 - . i 3am» m} .933 .533 cases»; 53m} seas-«haunting...— abes ugh v.35. Yucca in 3 g a g igittessfli-u snmfifiufiafimfififiméofie Ema as 8 flag a: on «5582 8.508 83:5 amuse EH8 6.3. 2..» 384 3285* 3.5 mg 8 Beam e883. 5 ago wags .5an 88 «51.60 3...!- Hm .28: an 933 .12.. no .835: .8 33mm .mmma as £23. .8 v3.39.— na 8.85 «mg no non-mu .829 Hips .83 .38 323” .2“ .n Km? .3: 5.335 datum .335: «an .uofism fig dfiagofi< .323. .3833 go 835.5.“ 853.. 3.38.330 .o .o v5 83.8 .n .n sec ”334.. o.“ in «A . . o.n «.m mud-an «a 8? $6.5 Ed... mmwa omma 2.3» a3» «.88.... 883 ha .335 83$. .53 5.3 3mm» 80m. «8.83 has...“ awash Bagaouflfi Dian m.~.n pend m.m com pom A383 10E 03.853 . 3E R s 8a 3353 m on an... 5585 a!“ .53 $93: 9333 «.3 3: «.3 32 n.» $.35?“ «a his .38: hug-E 4...” m4 «on non 56 A3563 .BA Sign go o «a $9854 n a m o 8 Aggy 8:0 than no 3.? o .Bbd 3mm gags-m madam 333m 03%...“ 33.2; on». H0 m\H 0&0 H0 m) 05 MO m\H 05 H0 S 05 Ho.m\H gavndhmvoflhdso gm pmmHSH 5gb fig 308m vucnwg EC sadnog no 608 Illnflia 8m ofifiha no 38 $822 Na 835m 8? :33 E Ewan?— 392 mg .5 an» 335: an. 8. $988.. fiéoma @523. .32.. EBB .96.: a: 8.5. anéoém .48.: mm?" 8. @933 35.5:— 5 Egg 5 mums" 83 be an absence of a clearly dieoernible pattern between the “rage mnber ei‘ purchases, average family and per capita income, and average size of family. To help in further determining if certain family characteristics are aeeociated uith apple purchases, two additional tablee are pre- eented. Table 28 renke the taniliee by three distinct income groupe based upon per capita income rather than tail: income. Beoaaee per eepite inoone resents a truer picture of a family's financial re- sources than does family ineeue, that ie the eriterion that hae been «looted, Table 29 ranks the families by. average eiae of fanny. TABLE 28 nmmmcxs IN FRESH APPLE PURCHASES mom 178 14.5.0. CONSUMER PANEL mums GROUPED ACCORDING TO mm mm INCOME, 1953* ‘ t: a Data on Purchases and Per Capita Income Clan Family Character! etiee Highest 1/? Paddle 1/3w lamest 173 Family Characteristics Avera per capita inoon ’ 32625 31531 3963 Average family income1 $6016 851118 33877 Ava-age also of men? 2.2 3.2 3.9 Quantity Average capita2 2?pounde)w 29.6 33.9 23.6 mum-e Ava-age expenditure per capita2 $2.69 $2.65 $2.03 *Adapted from J. D. Shaffer and G. G. Quackenbush I'Goneumer Purchaeee of Applee." mohigan Agricultural Ezperimen ant Station, East Lansing, Special 3mm hCS. 1955, p. 22. 1Boned on 1952 income as reported on January 1, 1953. 2Based on number of neale eaten at home, 21 male equaling one person. TABLE 29 Dm‘mmcns 13 MR APPLE PURCHASES new 178 24.5.0. CONSUMER PANEL mamas GROUPED ACCORDING TO SIZE OF FAMILI, 1953* Data on Pun-chance and ~— i Size of Fm? Two ee ve Family Characteristice One A Four Over lumber of eaeee 11 61 68 32 Pencil: ohmoterietice Average nine of tank]- 1.1 2.0 3.1: 5.1; Average family income2 51529 Average per capita income? 31339 Quantiw Ava-age quantity per capital-(pounce) 30.7 Expenditure Average fienditure per capi Size Average e pwchaae par «pit. (pounds) 3.23 Frequenw Average nuber er purchaeee 6.8 32.56 81:85? 8550}: 35980 $2386 $151!: 31098 33.2 26.5 25.1 32.77 82.28 82.16 3.79 2.h9 1.93 9.61; 13.78 16.72 adapted from J. B. Mar and G. G. Quaokenbueh, 'Coneuner Purohaeee of Apples." Michigan Agricultural Emeriment Station, M mam, 69001.1 1301161113 1405, 1955. pe 220 ‘ 1Baaed on the number of meals eaten at home, 21 meals equaling onepereon. tBaeed on 1952 income as reported on January 1, 1953. 85 On a per up“. an. the middle income group bought more apples per capita but did not spend as much per capita as the top income group, which would suggest that thoee enjoying a higher income pay a higher price for the applee they consmee than do middle income groups. The lowest income group In the lowest group in quantity of apples pur- chased and had the lowest expenditure; however, they tend te m a little more per pound than do the middle inom groups. It nut be member-ed that these are average figures for a particular group and that there is a large amount of variation dthin a pm. When families are grouped according to the lite at family, thee ie a tudeney for the large families to buy more apples per family, but a ennu- number per capita. The larger families buy apples a water umber or tinee dining the year and in large average eieee per purchase, but in maller quantities per person. The families which consist of tee persona are the leaders in average quantiw per capita, average eapenditure per capita, and average the pachaee per capite. This indicatee that probably married couples are the largest coneuaere of applee on a per capita basis. The larger eiee families, where children are numbers, do not consume as new apples on a per capita buie. Thee, childrmmbablydonateataeeewappleeaeadulte. Ihie offers an opportunity for produce merchandise" to increaee apple eelee by praecting apples among child-en. Bullet-J Fresh applee rank third in terms of alpenditure among all freeh fruits and vegetables. ipple sales are highly seasonal, October being the meet important single month for apple purchases. Almost 50 percent 86 of all apples are purchased in over 10 pound units. Since most retailers do not handle such large unit quantities except for limited periods of tine, it suggests the possibility that more retailers night attempt to carry a prepackaged apple unit of between 10 and 12 pounds, a half. bushel unit and a bushel unit during certain selected periods of the year. Unit size of apples should be varied so as to reflect the various seasons of the year. Varying the sizes of prepackaged apples fro: 2 to 6 pounds and displaying along with bulk apples at a selling price of 3 pm offers the greatest opportunity for increasing apple sales. In the mohigan State University consumer Panel data, quantity use related to both frequency of purchase and nice of purchase. The fifth of the families vhich bought the largest amount of apples also bought five times as often as the lowest purchase group and in average sises which were almost three times as great. It would soon that extending the length of time that apples are available, thereby increasing 21's- quenoy of purchase, night be another way of increasing sales. Both the top and middle thirds of the fasilies, based on income per capita, bought a larger amount and spent more per capite than did the lee incoee grmp. Revere, income did not seen to be directly related to large purchases. Large size faunas bought larger quanti- ties, in larger average size, and sore frequently than did the smaller - families: However, on a per capita basis the faulty of two members bought more apples. This would suggest that produce merchandisers could potentially increase apple sales by encouraging children to at ‘0" apples. CHAPTER V PURCHASE PEPPERS AND CHARACTEISTICS OF FRESH “WE PURCHASERS The orange is a fruit which has grown into prominence only within the past 145 years. Fria- to 1910 only about 11 pounds of fresh flanges we conslmedper capite. LpeaksasreachedinmhhuhenhSponnds smccnsumedandsinoethettinetherehasbsenasteachdecline until only about 27 pounds are consumed today} This decline in the fresh market is attributed to the tremendous gains made in the process- ingi-dustxy. Prim-“1915 lessthanlpercent ofthe cropsas proc- essed. Today less than 50 percent of the crop is sold on the fresh nerket, while the trend appears that it will drop still farther. It is difficult to guess when and whee the point of stability between trash and processed sill be reached. Although fresh oranges have enched a wide eoceptasos tr“ over the past us years it in interesting to note that about 5.5 percent of the panel families made no purchssss was 1953. astute of mana- prefu-enoes for citrus products in taxes indicates that about 5 pwsent of the twat-a did not use fresh orange»: Reasons nest frequently 13. La 380113, “Fruit and Vegetable Facts and Pointers," United Fresh Fruit and Vegetable Association, New, 1952. 21. 1. Fugeth, J. i. Bayton and H. W. Hitting, "Citrus Preference Among Customers of Selected Stores," Texas Agicultural and Mechanical College, Bulletin 722, 1950. 88 given for nonuse were: too much trouble to prepare the product, for- hidden by doctor's orders, and dislike of the taste. This same may pointed out why range consumption has enjoyed such a favorable trend.3 Health promoting values that have been stressed by producer advertis- inghas convincsdnaw customers that oranges arebeneficial as a leansofprsventing colds, asvollssanabundant sourceofvitm. (ranges are available every an of the year but are neat abundant fraJanusrythrough W. Ingeneral Flu-idastarts thonssoropyear in October by narketing its early and mid-season varieties until the end of April. Starting in February the Valencia is sold until late July. The other inpwtant producing state, California, starts its season off with the famous Naval mugs in Rovsnber and then the Valencia in Ehrch and continues with the Valencia until Hovembu'. 01' secondary imam-tense are Tens, Arizona, and louisians which market during the fall months until the early spring. Fl'uh (bangs Purchase Dots Table 30 shows the interrelationship that exists bottom «pendi- turespcrcapitaandsverageperoentoffmilies btwingeaohuedtdw- ing the three year period of 1952, 1953, an 1951.. 1953 was a year 1 relative up: menditure per capita compared to 1952 ad 1951» The hiyaest average percent of families bwing each week in 1953 coincides with this higher upmditure. Comparing varied with period it is intmesting to rate that inevu'ypcriod inuhifi the expendituepc oapita exceeds the like period in 1952 and 1951: that the average 31b1de’ Pe 15o TABLE 30 YEARLY VARIATION In EXPENDITURE PER CAPITA AND AVERAGE PERCENI' 0F FAMILIES BUYING FRESH (RANGES EACH WEEK DURING THE THIRTEm-FOUR WEEK PERIODS OF 1952’ 1953: m 19513. Period Expenditm'e Per Capita Avaagfifiho; 02:11:11“ 1952 1953 1951; 1952 1953 19511 I 3 .17 8 .15 8 .17 28% 271 28% n .10 .13 .19 28 31 33 III .19 .19 .20 32 31: 38 IV .13 .18 .15 22 33 :6 V .15 .19 .18 27 3h 28 VI .13 .15 .18 23 26 23 VII .10 .13 .12 18 - 21 19 VIII .08 .11 .10 16 18 11; II .06 .08 .06 11 15 1o 1 .06 .08 .06 11 1h 10 II .08 .12 .09 1h 21 16 III .09 .11. .12 18 25 21 XIII .17 .18 .16 29 30 26 rota $1.59 81.88 $1.71; 211 25% 22% s mmigan State University 00m Panel Data. 90 pment of families buying each week also exceeded the sane Mode of 1952 and. 19511; by the cane tdcen, periods in which the arpenditure was below 1952 and 1951:, the average pu'eent cf families buying each we& was lower. Since the expenditm's per capita and the average percent of families buying each week seen to be related, the increase in expenditure that takesplaeeinoneyear oval- another-year appears to be a function or promotion and merchandising affa't as well as a function of price. For a more detailed look at range purchase date in 1953, Table 33. emu-ates the quantity per capita, expexfiiture per capita, «pendi- turerank, andavu-agepereenterfaliliesbmng eachwesltduringthe thirteen-i’our week periods. About 52 avenge: per «pin are eaten eschyaarataeostotabeat 3.6mtspcrcange. Wee-spar- ebased by an ”cage or about 25 percent of the families each week. or course there is a large seasonal fluctuation in resilies bwing each week and quantity purchased. From the late fall until the ‘late spring the bulk or the camel supply is purchased. I'm sample, from Doom until add-fie; about 60 percent of the total quantity is par. ehaeed. The average percent or families having each week also varies, truahighoffiihpmenttoalosoflhpercent. There appears to be several reasons for this increase in purchasing during the winta‘ months and the sharp decline in purchasing during the mar. First,- both 11m and Oworfl'la are in their peak harvest season during the wintu- and spring months; as the season mess the quality of the fruit beams pooru'. Second, 'seaeonal needs," such as vitamins have become a necessity during the winter to aid in 91 TABLE 31 SEASONAL VARIATION IN FRESH ORANGE QUANTITY, EXPENDITURE, WHIIDITURE RANK AND FREQUENCI OF PURCHASE, DURING THE THIRTEEN-FOUR WEEK PERIOD OF 1953* Quantity Expenditure Menditure 12:32:31: Tine Period Purchased in Gents Rank Anon; of Famili” Per Person Per Person Fruits and Buying End: Vegetables wag (cents) (percent) Dee. 28,1952- an.2h, 1953 h.6 15 h 27 Jan. 25 .. Feb. 21 5.6 18 3 31 Pwe 22 " me 21 SeS 19 k Bk ”at. 22 - Apr. 18 5.1 18 3 33 May 17 a- June 13 3.6 15 S 26 Aug. 9 o Sept. 5 2.3 03 9 15 mt. 6 ‘ Ms 3 2.0 08 9 m M. h - My 31 3.2 12 S 21 We 1 - We 28 3e8 1h 5 25 Rev. 29 - Dec. 26 h.6 18 h 30 M A ._ .___ e Mchigan State University Consumer Panel Data. 92 the prevention and cure of colds. Third, much cf’the seasonal variation is due to shifting to more preferred.produots which becane available dur- ing the stunner months. In a study conducted by the United States anerto’ nest 0!.1griculture on consumer uses and opinions about citrus products the consensus of opinion seemed to be that the citrus and.non-citrus fruits did not compete too greatly.h Competition between citrus fruits and nonpcitrus fruits did not seen.to be entirely dependent upon such matters as price, supply, and :sankstin31asthods. Consumers, in this study, considered citrus fruits as a special food class shich'sas different and not part or the general line of fruits. However, this onfly'sssls to be true during'ths‘lintsr :IonthI. Oranges take a severe drop in;slpenditure'rank during the summer months relative to the nonneitrus fruits. table 31 shoes that during the summer oranges.rank ninth and tenth in.enpenditure while during the sister they'rank third and fourth. This change in sspendi. turn rank position is partly the result of seasonally produced none citrus fruit. One of the controversies in orangeIssrchandising is shether ranges should be priced and sold by the unit or by weight. And of more recent interest is the controversy that exists over'Ehsthsr oranges should be displayed and sold intlccee‘bulk fors.or pftp‘fltlged tors. ind.if in.prepackaged form, what weight or count should be put inthebag soas tosaxiniseboth total sales andyst appeal toa hAnon. 'Consuner's Use of and Opinions About Citrus Products,I United States Departnsnt of Agriculture, Agricultural Information Bulletin nos 50. 1951s 93 large number of cuetmersi Answers to the above can in part be deter- mined from: the discussion that follows in the next two sections. Pricing hotbed Although pens]. date cannot be of help in answering the question so to whether pricing should be by weight or count e review of the history of the problem and the svnilebls pricing studies conducted should be helpful in eppreising the preferences the customer shoes toward these tau methods. The policies end regulations sdnpted by the Office of Price Attainin- trstin during led Wsr II were instrumental in fenilisrising the bwim public end retailers with the method of pricing bulk orengss by weight. Prior to that time, the pricing of orenges by count wee prevalent mg reteilers throughout the Nation with the exception of the West Gout ares where pricing by weight was introduced in the late 1930's. it the emiration of Office of Price Administration controls, produce managers were faced with the problem of deciding whether oranges should be priced by weight or count. After decontrol, many reteilu's resumed selling oranges by count while others continued the weight-pricing method. With both netth of pricing being used, the question arises so to which is the better method for eerketing effi- dm and papuler acceptance. The customer, frequmtly confronted with more than one method of pricing, hes little, it w, basis for detennining which method is bestsuitsdtoherneeds. Gaston andhsbit ottenpleyestrongrole in influencing that buying decision. Undoubtedly in new 1m whether the count or the weight method is used, the customer continues to make her selection on s unit basis, ignoring the pricing method used. 914 The United States Department of Agriculture thought that hy'ccn- ducting research in retail stores on the two methods of pricing bulk oranges-47y count and by weight-n-it might be possible to determine if there uists any difference in potential dnand for oranges resulting tron alternate pricing methods.5 The results of such a study would ' help the retailer detu'nins which method of selling would be nest cupetibls with customer desires. ‘ Tests were conducted in four cities in the Hatheaetern section of the count”. In three of the cities, ranges priced under both ; sethods we placed in edjacsnt displqs. In the fourth, city stores eereequsnydividsdintotwogrcups, eechg'oupsusingenlyene-sthcd of pricing. Cities were selected so that some sf the. was accustomed to pricing by weight, while the others wa'e acclinated to pricing by count. Data tron-lee mordsinthefourwssk pericdindicstsd thet whu’ecustcnersweregiwensnsqualopportuxdtytohuybyocuntrlv pound (3 cities) slightly more than two-thirds of the purchases were nedsonacsuntbssis. Balesinstoreslocatedwhers threwasnet snequsl eppa‘tmity tcbtvbycount erbyweight (1 city) showed that the custmsers offered some resistance to a change in pricing nethods. Sales in stores selling by count were slightly larger than in those selling by weight, the. giving. suppcrt to the conclusion that customers in that area have a preference for purchasing by count. A 53. D. Bosnia and H. R. Trienish, "Comer Buying Practices and Preferences For Mohasing Oranges By Height or Count, In Selected 01%..." United States Department of Agriculture, Washington, D. 0., 19 e 95 In all cities, more oranges were sold by the ccuntopriung method than by the "cat-pricing method. nosevc, this may have been due to force of habit and the in»: of familiarity with edght pricing in sous cities. In interns-s conducted with the customers, we then 70 wt of the total mstonsrs interviewed femd the count-pricing sethod. The study disclosed that principal reasons p.73!) we: (1) more for the m, (2) more undcstnndsble, end (3) h.h1t.6 when customers who purchased their oranges by the 1!th pricing method were questia— ed, the reescns they gets were: (1) more for the nosey, (2) appearance, and (3) better for purchase of mu mm..." Those that gar. Iappearance" es a ressm believed flanges in that display looked better. However, this see not true, because mges in an cues cane fro- the sans source and were of equal quslity, price, end appearance.8 with these facts in nind the reteiler 'sust decide which pricing sethed for ranges is to his advantage. Believer, me conclusions thateanbereachedfrcnthisstudyetronflyindlcetethstpridn; bycemtispreferredbysostenetasers. lsidsfroscesto-errs- sctinthsreappearstebeadefimteedvantsgetowicdngtvcount an the factor of ties taken per sales trsnseetien is comidl'ed. A retailer chum free pricing by weight to pricing by count sill probably find one customer objection during the initial period. have, this must be snowed to familiarise the custaeor with the new nethed end to salute its merchandising merit. 61b1de . p. 9e 71b1de’ p. 10. 81bit}... p. 10. 96 Repackaging and Silo of Purchase The more recent controversy in range merchandising is whether oranges sell best in bulk displm or prepackaged displws or a cabineticn of both. Allied with these questions is the farther query that pertains to the choice of bagged quaratiti es. To gain further insight into this problu a review of the past studies is in ordc. In the 19h9 can. study 76 percent of the nan-s of fresh ranges prams to buy the mm in tall: rather than hag-.9 rho min ree- one for preferring the bulk were: (1) the onrtunity to select fruitofbettu'qcality, and(2) thedesiretcbefreeteselsstthe siaeMntnbez-offrcitnesdsd. AbmtIOperoutcfthesecshad no preform while the running wares-red the prepackaged fora. Among the hmenlmrs who indicated a preform fm' the" prepackaged the reasons given for that chaos m: (l) packaged fruit use less expensive, (2) fruit use easier to handle, while (3) only a small percentage said that the prepeaaged fruit was of bettu' quality. Sass customers reported a preference at the beginning of the sesson for M citrus sold in bags. As the season prey-eased the prefmshiftedtebulkdieplaysasaneansefselectingfruit free from peor keqing qualities, a'yneso, etc. A equity of the hoeswivss who favored prepackaged oranges preferred to purchase thee in 5 pound padrege since.m 9m0th, E. Gite; p. 23. Joni-d0, pg 2". 97 In the United States Departlent of Agriculture stew conducted in Louisville, Kentuclq during 19148, nest of the comparisons between it and the Texas study are quite emu-.11 About 70 pm bought and promo-ed to buy the loose bulk oranges, about 20 percent beught ' and preferred to buy the prepackaged oranges, and 10 percent stated no prefer-mu Among those who preferred bulk oranges, the prefer- ence for that choice seemed to be (1) quality of the fruit was better, (2) chance to see shat thqr were buying, and (3) scenery, because only the exact amount of fruit needed was bought. The prepackaged Osage purchuu‘e favored their nethod of buying because (1) oomenime, (2) more scone-eon, and (3) the quality of fruit no better. It was also inta'esting to note that in this study custasrs bought nest frequently wenges which were purchased by count. Habit and the aistaken idea that purchases made by the pond resulted in poorer quality m1 inability to seleet the mbu- er oranges needed, mooted ferasaJorityprefereneebuyinga‘angesinurdtsefadesencapert w. Inanether stum'eonduetedbytheautedstatssnepertsentd Agriculture in Les ingeles during 1950, essentially the seas results were obtained in this study as in the previous tee.” Again, pricing by count was the most popular of the two methods. Reasons for this ninon. ”Citrus Reference Among Household Consumers in Louisville and Nelson County, Kentuseky,fl United States Department of WCfltW'. Rubi-mag De Os, Informatim millet“ H0. 2, 1950. 12mm, Pa 21s 13'Consmner's Use of and minions About Citrus Products," on ci . 98 preference were: (1) habit, (2) exact nunber of fruit needed can be selected, (3) easier to learn cost, and (h) more convenient. The results of this preference are particularly noteworthy because of the predoninance of pricing by weight on the West Coast. These past studies point strongly toward a need for retailers todisplsyu'angesinaloosearrangementnanna-oneitherbinser displq tables. 'Also pricing by the count seems to sell a y'eatu' quantity of «sages. One of the areas that these studies failed to investigate was the possibility of selling prepackaged ranges by count rather than weight. Since there sea a strong preference fa- prieing by the count, perhaps pester prepackaging acceptance would have resulted if the bag had been sold by units instead of weight. a study of Mchigan State University Consumer Panel data will be helpful in determining if the preferences that were sheen in 19138, 19119, and 1950, are the same as in 1953 and 1955 a whether pre- packaginghadbeoonepopularmoughtooverco-etheebaeetions that fluted toward its use during the earlier period. ‘ um. 32 shoss the variation in number of orange pm'chases and quantity purchased by sise and nethed of purchase for three-four week periods in the first quarter of 1955. Bring the first quarto- of 1955 the peroart of total purc'nases ranged trc- 38 to h? perent for the prepackaged canges. The quantity purchased in prepedcaged fora ranged from M percent to 51 percent of the total quantity. Thus, over the past five years prepackaging scene to have overcome some of the customer objection that was previously incurred. How- ever, about 60 parent of all purchases and 55 percent of the .xdfi enema m .nemsuoeneum Heel w .35 18a also 3:858 33m 835:... am 3 m.~.m m4... 3 am mm 5 an 2 «w on O. O. .9 O. O. .0 O. O. H .0 .H O. 8." o a" .H m. .H .00 O. 0. fl .0 H O. R .0 CI .0 00 H H O. O. . O. .H 00 O. a o n. a a an 2 o n. a m a m an N m a a a .. H .. u a a a 3 es .6 Os 0. es .8 00 H 00 0e. ee 0. “H R R mm mm an an an an an an on on «a De ee .0 me 00 09 H 00 H be H .0 a s- Q m H 2 H m a S n o N «H n o N .0 W.“ m. H to W 00 N .0 W 0. m av H tutu-:01: uuuuuuuuuu grommet nnnnnnnnnnnnnnnnnn m a m m m a m a m a m a 3358 33:02 has 388.3 .3358 836.5 32: 5.” 18a. .3 .38 ea H.305 no 189 no 1%“. co H38. do 8285 accused vceohem peeked aaeohem ocean: vases comm mwfi «3 none: ghee wflwéwm aw _ co 8% $3 .5 538a RS .9” 5&3. mmfi .N 33... ammfi .6 $330 a an." E 385E Em: Edema mom enamels.” mo 8% E flange go a: amalgam means no 53 an 83:53 am an 1% quantity is still made in bulk form. 0n the basis of this evidence it still seems advisable for retailers to avoid a complete suing over to prepackaang. Rather they should feature both bulk and pre- packaged selections together. W to the siss of purchase it can be readily observed from Table 32 that the dcsen sised unit predominated in both prepackaged and bulk pin-chases and quantity taken. Between 60 and 70 meant ofallcangeswescldat thatunitofpm-ohase. ‘rhsnentsise ininpcrtancesasthea dosenunituhich accountedforbstwesnlotc 25 percent od‘ an quantity, with sales about equally amoae batman prepackagedandbulk. most all sales of less than a doseassrs sold in bulk too-a, although these purchases only amounted to about 10 percent of the total quantity they did account for about 20 per- cent of the total transactions. m the basis of these facts it some desirable to feature both bulk and prepackaged merchandise. Two prepackags sisesars recommded; first, the doses pack sise and second, the tee dosen pack siss. Ithe tee dosen sise would be particularly adaptable to merchandising smaller sized oranges and oranges that could be promoted at a seasonally low price- Tahls 33 presuts the variation in size of purchase and quantity purchased during the form-thirteen week periods of 1953. Data in this table are for an entire year, but it does not include method of pumhsss as does Table 32. Il'hs other three quarters of the year present about the sane relative relationships between the sizes. Again the dosen sise unit is the nest popular in both transactions and total quantity 101 .38 .35 aoaoaeoo monotone: 33m cadence... 8H 8H 8." 8." 8a 8H 8H 8.— se so so es es 0e 0 H 8“. 8H m a a u N .. n n 8 .. 2 n .n m A p m o n on I on an «A «N m." we a «N m.— a a .. a a a a n u an um an «m mm 3 on 3 am «a a A : .. u u a a .2 a a fin «a 3 AN m a.“ a 9.. c a a a o a i. a m m .. a Iv .I as - ' as ............ ”flochOQ ................... 33360 canon; 53530 snacks...— hodancno cannons...— hfifiussa sssnchnm 38.5 a.“ «c «c no He no no Ho Ho scenes.“ acsouem essence assess.“ vouch-om age.“ ages anions.— , Joanne.» nos...— R2 .8 pontoon RS emu £533 a... 8.... N3 on.“ 5.3. . no :3 «mod .2. tones?» nmfi .m... .5... . 33 .3 not: «men .8 Snood .l 111' mama." .8 Boga new: Emamdanmpoa m5 223 anemones 5258 e: nonsense amino .8 £3 a .8254» mm ”5de 102 purchased. Second in importance is the two dozen sise unit in terns of total quantity. Third in importance in terms of total quantity butseccndinisqaa'tancsinterns ofnmbercftransaoticnswesthe one-half doses sise unit. Although the same relative relatimhip exists betwaen the sises it would be ruembced that the total quantity fluctuates between the seasons as indicated in Table 31. Thus, a retailer would pedtage a doom, and two dozen sise unit during the entirs year, but only in snsller total mber (lining the em. Units of sale other than the ens-half dozen, one dozen, and two doses sise are negligible. They are mainly in the thru dean sise and ether multiples of 12. m, is the fall there are sons eels nads is crate non half-bushel sise units. The ataonoa of odd med units indicates that very few cangss are sold by the pound such ss 5 pounds or 8 pound basic Regardless of the type of «splay nethcd that is follasd the ilpcrtant taste of decision-aunginthe storeandits effect upon orange sales should be realised by all food retailers. According to the study by the United States anertnent of Agriculture, nest house- wivesusually decide on the kind of dtrus fruit thq want before entering the store.“ Heworer, m of these housewives who had made these decisions changed their plans after entering the store. In nest instances this change of plans is attributed to either one of two factors: (1) poor quality or appearance, and (2) higher price “Ride. p. 260 103 than anticipated for the quality on display. This importance of quality of the fruit on sale was also demonstrated among those who usually waited until in the store, to decide what to bay.” has it appears that retailers need to be alert to the necessity of buy-v ing wanges of good quality and condition, handling than oarenaly while in storage, and policing displays no as to weed out who and decayed fruit. Relationship between Family Characteristics ‘ and Fresh mango Purchases Of interest to produce merchandism is the relationship that exists between different quantities of apples banglt by families and the family characteristics of those fasilies. Also of interefl to the produce merchandise is whether inome or also of family is related to orange purchases. dread with such information the nor- cther is in a better position to sin his advertising message and praeotion effort at those customs-s or potential oustuasrs that are nest likely to canes an increase in orange sales. In cder to show who the families are that are responsible fu' both high and low total quantity purchases; and the families that bw most frequently, two tables of data are presented. In the first table (Table 31;) fanilies who purchased oranges was divided inte five equal ms based upon the quantity of oranges they purchased per family. In the second table (Table 35) families who purchased oranges were divided into five equal groups based upon the frequency with which they pm'chase usages. A—_‘ 15mm. p. 26. .33 .a gas». so vetoes.» as 8605 «mad no bemoan 5033 one Magus easel Hm gone; as cease eases no .335: so usesmu .eownsuo wagon confine.“ 39.3 has no panama .38 Hanna 2588 backspace 8.3 53:35. 5.3 2.3“ cons 3mg «83 mecca: 338 and hate: 35» 3mg» 80% «name name «285 has: has: fin «cm 6." m.n don wand-H mo and. out»: .efiuaggo head o.n pom m.n m.n ml. 3.33 non anaconda we sewn ewsheo< H6 m5 ~.m . 0.3” 4.3 and non oosnounm no swan om§< . A . onaw m s n." ma on assumes—E execs no eon omens: unsung 0 ca 2 S and 338 non 535.6 omens: 5 me an «a... $4 head hon banana omens: 533:0 coda-3h asap send-sh bead-h aside—sh 2: no m} an co m} as «o m} .5 «o m) .5 co m} analyse-oz". has: wearer 550m on enooom vesswfil one sessnohfl do even unsung swaglneeah we gonna: nexus“ confide!“ Home? 5 53m he amalgam $348 when .3 E353 he 8. cfiamoooa anemone wedged nag @828 54...: can 385 mmméoga mfiéo when 2. ESE 382a 2H aegis an fleas .nmm." an Econ no vetoes." es 232: «mad no veeemn donned one gave cases .3 .88: as 333 eases no eons: no nonemu .esmdsuc mg nod—“dash seen» an no no} 105 .33, Hanan goo 53.3»de 33m nsmgodas poo." «.NH New." 423 0.9" has.“ nod season—E on? omega Guam «8% mama» 35m «$3 e33 «383 333 and huh: one; use some 2% some 133 he: can: m.~ «on non o.n Jpn «53.5“ no one... emshe>< _ w nefiafigo has; a a. an a... 93 55.... non 3.33.3 eweeobd .3333 o.~ H6 4:: 5.3 Qun . Eamon? also: no non omenobd ‘ gag academy 833w.” none—awn wean-w." , 832nm." 23.3 efiuom «fluom Sodom oficom $33 538 23a 808m tonal Jemefiuahflfiadmoamuma 883.5. N38: no ~35: 5 gene thmm." a mmczdmo mama 530% HEB. mug «3 E95: may. oh. Qfimaoo”: @3026 age“ awash mgmzoo abomex as 02024 mmméogm H0233 mag OH 95.5% $590.; 2H QED mnflma. 106 According to data presented in Table 31; there was a large range in average purchases among different families. This range on a family basis was from 1469 oranges to l? oranges. There also was a direct relationship between quantity purchased and frequency of purchase and sise of purchase. The highest one-fifth of the families bought oranges most frequently and in the largest average size per purchase. The second one-fifth bought the next most frequently and in the next largest average sise per purchase and so on down the line to the lowest one-fifth. There seem to be no particular patterns of family characteristics that are associated with quantity purchases. The most that night be said is that the families with high quantity purchases were also the families having the highest average family in- cone and the highest average per capita income; however, the reverse was not true. The families who bought the smallest quantities did not have the lowest income. It appears fron this table that there was a slight tendency for larger families to ban the Largest quanti- ties. Table 35 shows the frequency of purchase range between the five groups of families ranked according to the average number of weeks they purchased oranges. There was a considerable drop in frequency between the highest one-fifth who made purchases 32.8 times a year . . and the second one-fifth who made purchases 18.7 tines a year. Fre- quency of purchase was associated with quantity purchased per family; for ample, the highest one-fifth bought M6 oranges while the lowest one-fifth bought only 21 oranges a year. Although there is no definite pattern in regard to average size purchase per family, the nest irequsnt 107 purchasers did have slightly larger purchases than the other families. The highest onecfifth of the families had the largest family incomes and largest per capita income, but again the reverse is not true (the lowest one-fifth did not have the lowest income). In order to detemine if difference in purchases are the result of differentfanily characteristics two additional tables are pre— sented. In Table 36 families were grouped into three equalpes' capita income youps. In Table 37 families were grouped into four groups based upon the average sise of the family. TABLE36 summations IN FRESH ORANGE PURCHASES mom 168 11.5.0. consume PANEL mam cam coanmc to THE Pm CAPITA moms" Families Grouped Into Equal Data on Purchases “d Income Groups Family Characteristics Highest L/B Hiddlfi 1/3 Lguast 1/3 Family characteristics Averagz per capita income $2665 $1593 3 916 Average family inconsa $6070 $5030 $3753; IVUTIg. finily Oil.) 2.2 3oz 309 Quantity Average quantity per Expenditure Average e iture per Olpi $3e16 $2.18 $1e9h ‘Hichigan State University Consumer Panel Data. 1'Based on only those families buying oranges. 2Based on 1952 income as reported on January 1, 1953. 3Based on nuaber of meals eaten at home, 21 meals equaling 03. person. 11.1.)! n . 0‘ TABLE 37 DDTERERICES IN FRESH ORANGE PURCHASES mom 168 31.5.11. consume PANEL FAMILIES smug ACCORDING TO AVIBAGE s12}: 01" mam" Data on Purchases and -L 312° °tlpam13f Family Characteristics One A Two Three 3. F1" 8: Four Over Family characteristics Average size of fami 1.1 2.1 3.6 5.5 Averag family income; $1380 am 35336 $6101 Average per capita income3 $1250 $2385 $11479 8110!: Quantity Average quantity per 60.13113 71e6 8900 h9e1 h9e6 Menditure Average expenditure per capita $3.18 $3.23 $1.80 $1.81 Sise Average sise purchases per person 6.2 6.1; 3.8 2.6 Frequency Average amber of weeks purchased 11.6 13.8 13.0 18.9 *Hichigan sue. University Consumer Panel Data. 1"Based on only those families buying oranges. 2Based on number of meals eaten at home, 21 meals equaling one person. 3Based on 1952 income as reported on January 1, 1953. 109 W grouping the families according to per capita income the range is from a high #2665 to a low of $916. The family size was largest fa' the low income group families and smallest for the high incone group families. Thus, family income ranked in the some newer as per capita‘. home. The quantity of oranges that are purchased per capita is directly related to per capita income. This range was free: 88 was per year per capita to h? oranges per year per capita. However, on a family basis, the middle income group purchased slightly more ranges than high income groups. Relationship between income and auditors per capita is also direct. Even though pa capita income is directly related to per capita quantity, families of all insane groups buy about the sane number of oranges per family. Thus sise of family is a sore imartant determinmt in arriving at difference between family purchases than is incense. And last by grouping the families according to family 31:. (Table 37); it is interesting to note that as the family size becomes larger the family income beeoaes larger, however, the per capita income is lowest for large families and highest for families of two mesbers. Although larger sise families some more oranges per family than small families they consume only about five-Maths as new oranges per capita as the family of two members. ' Since most one and two amber faulies are composed of adults and most times and more number families are composed of adults and children, the reduced ptn'chase per capita of large families must be the result of low consumption on the part of childrm. The large families also buy more frequently but in smaller quantities per person. Li A 110 The data on family characteristics as related to fresh orange purchase behavior has limited direct application to food retailing except for the part that advertising and promotion could 13183. To increase orange consumption children must be encouraged to use oranges in their diet as well as in the parent's diet. The theme of a produce advertisement might take the fern of encouraging the parent to place an orange in the school lunch box or has oranges help to combat winter colds, etc. Sumner-y The fresh range is a fruit which has some into general use only during this century. The peak has been reached in consumption and a decline has already started to take place. This decline has been due to the phenoninel grosth .which the processed orange has emperienoed. Although fresh oranges have and still do enjoy a wide sceepme of use, it is sin-prising to learn that about 5 percent of the lensing families bought no fresh oranges during 1953. In comparing orange purchases from year to year it was noted that increases in expenditu-e appear to be related to promotion and nerchandising effort. (h: the average 25 percent of the families bought ranges each week. Thus, oranges are excellent feature items for advctising. A review of the studies of pricing methods revealed that when given a choice to select from a display priced by count and a display priced by weight the customer prefers to asks her selection from displays priced by count. Aside from favorable customer reaction there appears to be a definite operating advantage to ‘reteilu's when the factor of tile taken per transaction is considered. Studies conducted several years 3.:7' tt‘l.‘. 7. I I t. 111 ago definitely show that customers preferred selecting oranges from loose bulk displays. However, panel data in the winter quarter of 1955 revealed that about he percent of the purchases and 15 percent of the quantity is sold in prepackaged fora. Retailers could best maximise their orange sales by displaying both prepackage and bulk oranges. Fa- prepeckagng purposes the dozen sise unit seems to be the favorite quantity that the customer purchased during the year. Due to the fairly large quantity sold in two dozen sise units (about 22 percent) it some advisable for retailers to also «my this sised units. There is cmsiderable difference between families in the mount of flanges they buy and the frequency with which they buy. Failies an bought the greatest quantity had larger incomes and tended to have large families. The ease family duracteristics are also true fa fdlies that bought the most frequently. When families were grouped according to per capita income and average size of family, it was discovered that there was a direct relationship between per capits income and total quantity purchased per capita. There was also a tendency for large families to consuls a much smaller quantity. of oranges per capita than smaller sised fsnilies. Thus it was concluded that to increase total orange cone adaption through food retailer efforts it would be necessary for practice and advertising to stress the need for oranges in the child's diet. .J‘V 1.". [can I” w. .‘. CHAPTER VI PURCHASE PATTERNS AND CHARACTERISTICS OF FRESH WEFRUIT PURCHASES The yepefruit is a basic produce itm on the market throughout the entire year; however it is nest plentiful from October through Kay. Fla-ids is the principal source of supply with Texas, Lriscna, and California also shipping relatively large quantities of the fruit. Table 38 shows that virtually no grapefruit were sold in lensing dur- ing the months of July, August, and septuber. Due to the U11 quantities shipped and the generally poor quality shipped during the smear months this item should be drapped from the food retailers stock list at this season. Fit-ssh Grapefruit Mchase Data The expenditure per capita during the three year period (1952, 1953 and 1951s) ranged from a low of 87 to 95 cents. Shown in Table 38 is the avu'sge parent of felilies having each week. Ch a yearly basis the percent of fsnlies having each week was quite stable for the three year period (approximately 11; percent). But due to the seasonal purchase pattern exhibited by grapefruit pm-chasers the 1!; percent of the faIilies buying each week becomes urn-ealistic and taxis to hide the impact of the mopping behavior fa' the shorta- tine period. For maple, in 1953 the range was Ra almost zero in the summer months to 29 pccent of the families bw‘ing each wed: during February. 113 TABLE 38 YEMEN VARIATION IN EXEIDITURE HER CAPITA MID AVERAGE REFCEI-IT OF FAMILIES BUYING FRESH GRAEFRQIT EACH WEEK BY FOUR EECK PERIODS OF THE YEAR Expenditure Per Capita Average Percent of Families Period Buying Each Week 19§2 1953 1991: 1952 1953 19% I 11¢ 11¢ 11¢ 21% 20% 25% II 10 16 13 17 29 25 111 13 15 13 25 26 25 Iv 12 13 11 21; 2h 22 v 10 10 8 20 18 17 v1 8 S 6 1h 10 13 m h 2 3 6 6 7 VIII a a a ' .. ee - se IX 2 a a 3 ee .. X l a a 3 es 3 XI 5 S h 10 12 10 XII 9 9 8 18 19 17 XIII 10 10 9 19 21 18 Total 95¢ 96! 87¢ 111% 15% not *Hichigan State University Consumer Panel Data. “loss than 1¢. ll); Part of the difference in expenditure per capits between 1953 and 19511 my be W by differences in pa‘csnt of families buying each week. Comparing like periods of 1953 and 19515 with each other reveals that there were more periods in 1953 than in 1951: when the percentage of families bwing each seek was greater. It is iqossible without price and quantity information to determine the exact function price played in causing both the increased expenditure per capits and the increased mmber of families buying grapefruit each week. Homer, much of this increase nest be due to differences in sex-chandising effort at both prodwer and retailer levels. Table 39 gives a more detailed picture of the grapen'uit purchasing behavior of the lensing population during 1953. The average quantity scammed per capita use 12 yapefroit. The fluctuation in range of quantity purchased during the year was fro. a high of 2.6 grapefruit pacqoitaduringreln'uarytcaloeefpractieallynonednringthe am. consumption starts again in the early fall when tsnpu'atu'ee beams more cool and the new harvest season begins. Grapefruit consumption per capita is directly related to the nabcoffamiliesbwingeachwsd. nonunion“ éppoarl that increased pwchases of gapefruit are nest likely to occur by selling more families grapefruit over a langu- poriod rather than by selling Immpefruitper faulyintheveeklyperiod. Panel detaindicates that y'apefruit acceptance is far fro. universal. More than 16 per- mt of the families bought no grapefruit at all during 1953. In the Texas stow of citrus prsfuences, health promoting values and pleasing tests were the outstanding reasons given by 115 Tm 39 SEASONAL VARIATION IN FPESH grammar (mm-31w, exmmmns, mmmmm 1mm, AND warm or" mnczmsz DURING THE THIRTEEN-Pam mm 11131033 or 1953* Average Quantity Expenditure hperditure Percent Time Period Purchased in Cents Rank‘Among of Families Per Person Per Person Fruits and Buying Each Vegetables Week ‘ ' Tcentsf ' (percent) Jan. 2h, 1953 1.11 11 6 20 Jan. 25 - Feb. 21 2.6 16 5 29' Her. 22 ‘F Ame 18 1.5 13 6 21} Apr. 19 - May 16 1.0 11 7 18 w 17 - June 13' .s s 12 10 M 12 " Aug. 8 I ee «as so Aug. 9 " 30915. 5 8 ee ee ee 3817b. 6 e. Oct. 3 8 es ee ee Oct. h .- Oct. 31 .6 S 9 12 NOV. 1 ‘l‘ NOV. 28 101 7 7 19 Nov. 29 - Dec. 26 1.3 10 6 21 12.0 96 8 15 at Michigan State University Consumer Panel Data. 81533 than 1¢e 116 honenakers for the general acceptance of fresh grapefruit} Omen thought that fresh grapefruit was particuln'ly good for children a a means of preventing colds. Taste, as sell as health practice values is one of the important attributes of citrus products. In this sale study fresh aspen-nit ranked second in taste preference among all terms of citrus products (fresh oranges ranked first). Hith the fail-h large percentage of non-grapefruit uses, as . exists in the Lansing population, it should be helpful to More to understand the reasons for names. Reta-ring again to the fans stedyofpreterencee itsaepointedout that dislikeeftasteeaethe Iain hinch'ance to purchase of the promo? n» unsatisfactory taste ‘ use usually expressed in terns of "bitter“, “acid", and "our". Another reason that appeared as the trouble that was necessary in preparing the item for use. Thoua little new be done by the re- tailer to correct the latter complaint, thee is an oppwtnnity to help overcome the ebJeotios registered against taste. By pushes- inggrapefruit ofgoodquslityandmllnturityendthenbyhandlinc andrctating Wt moteflytheretailersillbe aiding the- situatim by displaying a grapen'uit that is not mulsive to the taste because of quality factors. Similar to the controversy that exists over pricing and displq-n in; methods sith mes is the controversy over these same unchan- dieing techniques nu: aspen-nit. Do customers prefer to bid fresh 1:. i. aneth, J. A. Bayton and H. w. Hitting. "Citrus Prefer- ences Am Customers of Selected Stores.“ Texas saicultural and Beehanical College, Bulletin 722, 1950. 21bit!” Pa 17s ll? grapefruit priced by unit or priced by count? Do customu-s prefer to buy frash @1th in prepackaged or bulk form; and if in prepack- aged fm what weigxt or count should be placed in the bag so as to maximise sales and customer acceptance? Answers to the above can in part be determined from the following sections. Pricing Method Although panel data cannot be of help in answering the question as to whether pricing should be by weight or count, a review of the history of the problem and the available prim studies conducted should be helpful in appraising the preferences the custwer es- hibits toward these two methods. Faniliaritywithbuyingbythepoundm gainedhythe customer during the tine that regulations enforcing this method was adapted by the affioe of Price Administration. is with oranges the mly prior ups-isms that the customerhedsithpricingby countsas intheiest Coast area, share this method of merchandising was introduced in the late 1930's. Eda-ring again to the Tease study on consum- preference on eitrus pmhaees it use reputed that 60 percent of the hamkers preferred pricing by count, 10 percent preferred pricing hy weight, ammwmmurrmntuumomhod.’ homage:- msms for prefu‘ring the count method were: (1) a desire to select the exact nunbt of fruit needed, and (2) it was more con- venient. It was interesting to note that when most consumers were offu-ed grapefruit priced by the pound that they still made their selection by the number desired. 3mm, 9. 2n. 118 In soothe citrus preference study using a cross sample of the nut“ States it was dstm'zsined that the direction of preference was for fresh aspen-nit to be priced by count.h However, about one- third of the users indicated that thq bed no preference in this matter. Reasons given in this study for their preferences were slightly different than, in Texas. Reasons most frequently given ace: (1) that this was the method thq we most accustomed to, and (2) pricing w count made the fruit less expensive because the exact number neededceuldbeselectedendatthesasetieebiggertruitccnldbe bougt. 1t 1- most probable that habit in the nest new reason fwaprei‘ermefu'bwinggrapefruitpricedbycmt. Theocstm of planning the purchases or adequate quantities in t“ of a specific number appears to be fairly deep-seated. the homemaker, kneeinghosmewshsis providingfer, sakes her sstieateotqoantity in tan. or masher of aspen-nit. But there appears to be little lofisinpreferringanyonemberefmzits solcngsstheproduot can be selected by that“ «iterion the shopper wishes to use, encept that it m be more trouble te have it weighed. Also, the Mr say sent to know at the time of selection Just has such the fruit will cost, which my not be possible when pricing by weight. than WW habit and the mistaken idea that if they bought in the pound they scold get peorc quality and could not select the mber needed, accounted for the eaJeritw preference for bwing himn. “Commer's Use or afl Opinions About Citrus Products,” United States Department of Agriculture, washingtcn, D. 0.. Agricultural Information Bulletin No. '50, 1951. 119 grapefruit in units of count. Bonner, all of this is not to say that customers would necessarily buy a greater quantity of gapefruit if cm were sold by count. In vein of the present day practice of increased shipments of grapefruit marketed in nesh begs priced by weight, a word of caution isnecessary. mcughinaanyinstances thesebagshavehadgoed acceptance in sue stores, it still behooves the retailer not to- nerchandiee these to the exclusion of grapefruit priced by count. is long as the custanere' stated prefcenee is for this pricing by count and there actions bear this set it scene probable that this should yield a hidxar veins of grapefruit sales for swans store. res-hops the llichigan State University Oonmer Panel data, which is of sores-sent cigin than meweviceslyeeetionedstudiee, sill shedsmndatcnthisbehavicr. Tmingnutethesectiuu W, the previous question along with questiuls posed in the first section will be discussed. Michell-118 The no recent controversy in grapefruit nercl'xsndising is whether grapefruit sell best in bulk displays or prepackaged displm «- a cmbinetion of both. And the second part of the seas question is that quantities should be packaged at different seasons of the year if prepackaging seems desirable. is with oranges, a revise of the histwy of the situation as it existed four or five years age should be valuable. Then by attaching mchigan State thivmity Consumer Panel data, which is of the more recent aigin, a coepari- scneenbenade. 120 In a study by the United States Dopartnent of ig-iculture in lcuisville, Kentucky there'uas an starchelning desire for grape- mm. that 1. displvedlcoems To the reply, mm the aspen-m ycu.beught already in a bag or lccse, out of a bin?', 95 percent beughtrloees displayed grepetruit. ‘Hhen asked as to theiryprefen~ secs about 90 percent stated they sould prefer to buy loose dis- played grapefruit and about 5 percent had ncrpreference as to either seabed. The first series od'reasons fer this preference centered around the rector of quality. therewsss a feeling that it;yeu lbcught‘flfoplnklged grapefruit.yec.gst a.fes’pocrer quality fruit andh-uit ofunsrensise. mesecondsa-ies otressmhodtcdc Idth economy. The eccnnny lay in.ths fact that this nethsdnees lees easteful because they could buy only the2nnnber they needed at any 81V‘I&tillb Substantially the seas results‘sssercbtained in mmummmmwtuammzmmmm Tessa studyQ‘ theeejpest studies pcdnm strongly’tcserd the need.£br'reteilsss to carry a supply ct’lcose displayed grapefruit and tcupriee these grspstruit by'counterather‘thannseight. abs-ver, it should be sectioned that during this period c: tieereest at the prepackaged arspetruit was seld hy~ssighte Whether'this would.hese a'beerinc engthejpces'acceptanceAof’prepeckeged grapefruit or not see ispcesi- his to data-sins. Blush. "Citrus Erefercnces Among Household Consumers in.Lonisville and Eileen County, Kentucky3' United.8tates Department or Agriculture, Hhshington, D. 0., Agricultural InformationhBulletin Nb. 2, 1950. 6'Oonsuner's Use or and Opinions thout Citrus Products," and "Citrus Preferences Among Customers of selected Stores,I 22. cit. . . 12.11 11141.1...»1‘ .Ilp. ,3.:.Hnr“ujhu1‘4iflhrwd “sue IF... ”4 o V. 121 A study of the mchigan State University Consumer Panel data sill be helpful in determining if the preferences that were shown in l9h8, 19b9, and 1950 are the some as in 1953 and 1955 or whether pre- packaginghasbeconepopular enoughintheinterinperiodtocvercae the objections that prevailed in the earlia' period. rm. to shoes the variation in umber of fresh grapefruit purchases and quantity purchased at different mute of sales for three-four week periods in the first quartc of 1955. The range in prepackaged sale of pape- fruitsasfrenalcscthporcontteh9porcentoflegz‘ape1ruit quantity. Thus iteppearsthat ovu'thepastfiveyesrslewcfthe objection formal: raised against prepackaged grapefruit have boa dispelled. lava-thelcss about 60 percent of the purchases and 55 percent of quantity 1- am purchased in bulk. m the basis of this evidence it still some desirable to avoid a couplets change over to prepackaged grapefruit. Rather it seems more apprOpriate to use both methods of display. The question of mba' of mat.- to put in a package must still be answered. Table bl shows that the lost popular sise mmt is the half dceen sise. m four, five, eight mu in bag also send quite well. These sane mu were also popular in the bulk purchases with the further addition of tee and three unit also being node in large quantities. Undoubtedly the variance encountced in prepackaged unit-a reflect to acne extent the sale of grapefruit be- ing realised at the pricing unit suggested on the price card. For «male. kid .4, 6 for «d, or 5 for -¢, etc. It appears that there is no clear-cut umber of units in a bag which sell decidedly . a... ..wfi1.« b..- ‘1. 122 .xasm queen m .vemnxeeneum enema m .35 15a .3538 P1235 81m Eugen... am 3 mm 3 mm «a 3 mm mm 3 3 an M H H ee ee N ee 00 ee n ee ee 99 d av 4H N . e e e e e e H M e e H N 0 H n “H ae H e m n n h. a n a m o a n on H H O I H e e e e e e e e e 0 fl e e H a n 3 a 3 a p n a u a n a o .. .I. .. .. a A H A H a A H p on mu m Ana a." a." m." «a 5 an e d... o m m m o m a. m s a m c a m m a m m p a m m «a c an p a mu m mm d 0H a an n m a mg n n n .. o H a .. AA A m n dd u u e e e e N e. 0 e e O I H . O O O O 0 e N e e H lllllllllllllllllll uncenem I I I I I I I I I I I I I I I I I I I , m m m m m m m m m m m m 933 .8526 E3530 3383 3358 .8385 32: fi H30." do .38. no 38. no .38. mo 33.. do .33 no mmfiwéa aneonom veecuem aceoaem accouem aaeenem aceeaem mmnomma ES .8 :83. mm? .8 bean; mmfi .3 55.. mmma .5 has; $3 .on 55» mmfin .u been: km? .8 $548 “.95 mma 2H mQOHmHm mmmkvmbomlmmmmh moa madmumbm ho nomam:_um qmmdmumbm MBHBZ4 48.3 2.2;~ no.3» Roma «$3 1-85 e35 hot: on on .2 «.n fin «he! no .3. .mp5: oofioflsoflau his. 04 «4 HA mod com 333 «on eesnokon no ends ewsbebd a.“ a..." in as no hid to eeenohfi no 038 swan-54 scan .3 o... .3. man 3.“ 2.32.5 execs «o bean omens: headset— 0 e e e e 3 “00 ma 0: on hmH m5 hen—33:55 emewebq a. a. . . and m and AN N cm r 9: hen 5.353". swans: badge audio-m codified evade-m gonna-m 33.3..." 05 no m} 05 we m} on» we m} 05 mo m} 95 no m} uoaaugcshano tough auozoa £¢H50h Ghana nnooom 900nm“: nae nondnohsm no sfisa mango dang e>fim 35 was sandwich .32 a HE: and $383 Summoning was. .6 “mango an 2. Banach? E80 admin can: 3:328 5.3. 3H 28.: $532 humanize can 2. 933% $08,: an noggin «A and." .33 on gnaw no seated?» as second «mg no weaemn donned one wand—ace cases .8 .033 as noose cases no 9355 no panama Jan—hog wflbfi 33239 2.93 bane cc venom." .88 15m «538 beta.» 83» Sound... mom; owns 033 mans «$3 «335 Jeane to emote: Hmmm» 33a Bmi 3N3 comm. «.52: 535 that: o.« fin ”A «J. «.n would do one. $824 . Shuffle-none has on «.4 no nm em 53 to enema—E soda encased Guam m6 m5 don 5.3 Wmfi hood . hen 5.3.3.6 owehsbd 535.5 j E to 3A «.3 332% uses: we gonna: emekebd gong Ended 822: 33?: .233 Sane-a .5 co m} .5 «o m} as co m} .5 «o m} on» co m) advantage has: €38 god each. 388 wanna on. 83% no 38 .398 38 £2 35 osné sided g on»? .n Suwanee mean engage Ema muons no woman an on coqfiooo.‘ awesome mfidafi no?" ~35sz 56:: 2: Eco: mmaaoéa «Hagoéu maxi 2. 593mm 305% on Eozga 9 mama . 130 According to data presented in stlo 1:3 than is a rungs from a high sversgs purchase of 114? grapefruit. per family (hiring s not to a. low stage of 5 grapefruit per family dm'ing s year period. Thco m s diroct relationship batsman quantity purchased and frequency of purohsss. lbs largest quantity oonsmm of grape- fruit also bought gator smonnts of fruit par purchase than tho locust quantity consumer. There seems to be no disoeu-niblo pstm Wtboqnsntityotn‘uitpwohassdsndthstsnflyinomr per «pm income. the. is s slight tendency for tho largc pun-.- ohsssrs to ban large mun»; hams:- tho diff-rm or. probably too slight. to 13. significant. ram. ];3 shows the frequency of push”. rang. hm fin groupsoffadliursrkodscoordngtotho “mmofwsoks they purchassd matron. no mo in Hoqusncy m from s high affidtinosinsyw toslwsvmgoofl.htinosinsnsml priod. Thsdropfrom thobim you}: totho secondhigbost group no the lax-gut. In that instance, the damp was so “mg. orlZpurolusossyosr. Although thsmtlrsqmntpm-ohusswm also the suns twins. vho also bought tbs largest quantities-— they won faniliss that bought in only slightly largo:- averse. mepm‘ohsso. hsmtoprankingfamflissbought inns]; slightly largo:- W per pin-onus than tbs thus lousst ranking funnies. 111m is no dioou'aiblo pattsrn that shows I rolstion- ohip between the family dursctoristios and frequency of purchase. In order to dotsmino if differences in purchases are the result of differ-mt family characteristics two sdditionsl tables 131 are presented. In Table hb families were grouped into four groups based upon the average size or the family. The families that did not buy grapefruit during 1953 were excluded from the tabulation. TABLE M; nmmmcns IN morons 33mm: TO FRESH ORAPEFRUIT PURCHASES mom the 14.3.0. consume PANEL Fm GROUPED ACCORDING TO ms PER CAPITA INCOME" Families Grouped Into Equal Data on Purchases and Income Groups Family Characteristics Highest 1,} Hiddle 1/3 an”, 1/3 —w——— Family characteristics Average per capita income2 $2726 $1613 $910 Average family nun-.2 $6196 $51.02 83635 Average family sise3 2.2 3.3 3.8 Quantiw Average quantity per Expenditure Average nditure .Hichigan State University Consumer Panel Data. 1based on only those families buying grapefruit. zBeaed on 1952 income as reported on January 1, 1953. 3Based on number of meals eaten at home, 21 meals equaling one person. By groqlng the families according to per capits income the average per canita income was $2726, $1613, and $910, respectively, 132 fa- eech amp. The highest income families had the mallest family msnbu's. The highest mouse families bought about twice as such grapefruit per capite as the middle and lowest incoae grows. Thus, high per capita quantities is related to high per capita Met-e. loam, this large diffu'ence between groups is somewhat diminished Ihm family purchases are considered. (Thu-e are larger average size families in the middle and la! income gulps.) Nevertheless, high quantity purchases sees: to be associated with high income. ' lhe last gasping is by average sise of the fauly and the results of this grouping are shown in Table 15. Family income is largest for the largest sized family but is lowest on a per capite basis. The family cmposed of tee Imus enjoys the largest per capita income. The family of two purchases twice as much grapefruit per capita as the next ranking youp and schemes about 28 psront more on a family basis than the families composed of more than five. Undoubtedly the fact that the family of tile ujoys the highest per capita income is an influencing factor in this high consumption. Since one and two amber families are most likely to be entirely coupesed of adults and nest three and more nonhu- fanilies are osmoeed of adults and childrn, the reduced purchases pa capita of large families lust be the result of low consumption on the part of children. ihis section on relationships between family characteristics and fresh grapefruit purchases probably be limited application to food retailing except for the information that is provided on who is reepmible for high and low quantity purchases of fresh grapefruit. “ivnlt'flnv- $.11 XII I ‘ mth 133 DIFFERENCES IN FACTORS RELATED TO FRESH GRAPEFRUIT PURCHASES AMONG 1143 14.3.11. COBSWLER PANEL FAMILIES GROUPED ACCORDING To AVERAGE SIZE OF FAMILI IN 1953* 1 ~ g: ,3==== Data on Purchases and 312° of Family Family Characteristics One Tue Three &. Fivesa Four Over Family characteristics ‘ Averag size of f . 1.1 2.0 3.8 h.8 Average finily imone3 $1281 $5151. $5705 359% iverag; per capita income 811h7 32h31 31550 31103 Quantity Average quantity per capita 15.5 33.0 11.2 10.7 Average quantity per family 17.0 66.0 1.2.6 51.h Expenditure Average expenditure per Clpiu 31.13 32.53 8 e93 3 e82 Size Average size purchases per person 2.7 2.8 1.h 1.2 Frequency Average anchor of weeks purchased 5.8 11.7 8.1 8.9 re Michigan.8tats University Consumer Panel Data. 1Based cn.only those families buying grapefruit. 2Basedonnunber of‘neals eaten at home, 21 meals equaling one person. 3Based on 1952 income as reported on January 1, 1953. 13h This information could provide a basis for directing advertising usages. For example, it has been pointed out in these past tables that grapefruit consumption 1- lowest among families having children and families of lower per capite income. Thus, advertising night teke the tern of the low cost per serving during the peel: hmeet eeeson or the nutrition evsilsble per serving, etc. Bunny The grapefruit is s belie produce ital: on the what about eight months of the year. Virtually no yepetruit were sold in the Leasing area during the owner swaths of 1953. The percent of families bwing each week dun-inc 1953 ranged free e low of sore in themertimetoahighoprercentinthepeeknonthef February. Grapefruit conemnption per capite is related to the number d' families buying, It appears that oonsmption is most likely to increase by selling more families yepefrnit over s longer t1. pried rethu' than selling we grapefruit per family in the shut tine pa'iod. More than 16 percent of the penal felines bought no yepetrut st e11 during 1953. Principe]. reasons for mu were unsatisfactory taste and excess mount of ties in preparation. By careful selection, correct handling methods, end good merchandising techniques the retailer may partially overccme the 1m objection. A review of previous studies chewed that pricing by count was the pricing method preferred by customers. It is most probable that 135 ‘ habit is the reason for this preference. In View of the trend to do mare merchandising by weight it still behooves the merchant not to exclude pricing by count. live years ago abmt 95 percent of the watchers preferred to buy their mpefruit in the loose bulk form. However in 1955 about 55 percent of the customers preferred the bulk purchase method. Thus over the past five years new or the objections raised against prepackaging have been dispelled. It appears that there is no clear- cut number of gapefruit to place in a bag which sell decidedly better than aw other sise. Thus, a retailer should change the number of unitsinsbagtofitaparticulsr nerchandising progrsa. ’ The decisions that the hamster nukes in the store is the inpca'tant decision of whether to buy or not. Value or the inter. action of quality and price together is the final deternents in making that decision. Planned purchases are frequently altex'ed after the visual inspection is node. Families that buy the largest quantities of papefruit are realise that are of small size and have large per capita income. Advertising to increase consumption should point out the low cost per securing and the health promoting values that it has for children. [1145' 11.91.? «4.11 4th.“ a. . CHAPTER VII THE USE OF MICHIGAN STATE UNIVERSITY CONSUMis‘R PANEL DATA AS THEMED TO SPECIFIC ”CHANDISIM PRACTICES In an econow of plenty, uhm'e wiring and producing facilities are adequate to provide goods far in excess of those required to meet basic hm food needs, aonmer uante m highly upon-tent. Under these circtmstances, customers can choose whether to buy or not to buy. It is here that information on purchase behavior becomes a vital element in providing merchandisers with facts on which to base merchandising decisions. Because customers need not buy itelss produced to satisfy rants, rather than needs, they can transfer their revere from one product to soothe or from one retailer to another. The interned merchandieer can hasten the acceptance. fr his produce and influence the stability of his returns by being ocgrdsant of the shopping behavior of customs-s. In the foregoing diapters the Iriter has discussed in some detail the usefulness of the mchigan State University consumer Paneldataasitpertains toaparticclarprodnst. Also discussed in these comedity chapters was attain merchandising techniques and methods uhich are of current marten” in marketing these products. It was discovered that mchigan State mini-alt: W Penal data had its greatest applicability in the area of package sises for an individual modity. But the question still remains: Are there other areas of usefulness for these data as applied to 137 the ‘ overall functions of retailing all produce (as opposed to indi- vidual modifies), and if so what are its limitations? Before discussing the individual functions it might be well to first look at the nature of these data and the form in which it would appear in after initial processing. Hature of the Date First, data may be presented for an individual comedity over an ext“ time period or for a brief time period. This is the type of data that were presented in the previous four chapters. Second, the data may be presented for a large lumba- of modities inagventineperiod. Thisisthetypecfdatathatuillbeused in this chapter. his second technique permits an examination of all commodities handled in a produce dwartnent at one time ratha' than a detailed examination of the single item. Relationships between ituns become more decipherable, and since my retailing functions must be analyzed from the complete department viewpoint rather than the individual «meets, these types of data are necessary. Table 106 shows the competitive relationships between the top twenty-five fresh fruits and vegetables from: June 11;, 1953 to Jul: 11. 1953 , and Table In shows the same relationship for the period July 12, 1953 to August 8, 1953. Table he, August 9, 1953 to September 5, 1953. and Table 19, sovember 29, 1953 to December 26, 1953. Pa- purposee of this chapter only these four four-Heel: periods of the you need be sheen. An examination of the tables shoes that twenty-five items in the produce department soc cunt “'3‘"- TABLE h6 138 COMPETITIVE RELATIONSHIPS BETWEEN THE TWENTY-FIVE TOP FRESH mums AND mmmms, JUNE 11. to Jun 11. 1953* nditure Quantity agxgzrcent of Average . Expenditure Product Purchased Total Fresh Percent Rsnk For 100 Fruit and of Families Persons Vegetable Buying Each Expenditure w°°k 1 Strawberries 151 quarts 19.h 36 2 Potatoes 538 pounds 9.8 ha 3 Bananas 1142 pounds 8.? h} h Husknelons 71 each 6.6 El 5 Tonetoes hS pounds 6.h 28 6 Head lettuce 97 heads 6.h Sh 7 'waternelons 250 pounds 5.2 1h 8 Oranges 320 each h.7 21 9 Celery 37 bunches 3.8 26 10 Raspberries 2:! quarts 3.1 7 TOP 10 7b.). 11 Lemons 130 each 2.7 18 12 Cucumbers 55 each 2.2 22 13 Carrots 1.3 bunches 2.1 at 1h nature onions h8 pounds 1.8 18 15 Sweet.cherries 16 pounds 1&6 h l? Radishes 68 hunches 1.5 2h 18 Green onions )9 bunches 1.2 15 19 Peaches 11 pounds 1.2 6 2O Sour cherries 13 pounds 1.1 2 21 Grapefruit 22 each .9 6 22 Apples 13 pounds .8 h 23 Peppers 23 each .7 ll 25 Asparagus 8 bunches .6 h TOP 25 9he8 All others 5.2 “Michigan sum University Constmer Panel Data. eeeeeeeeeeeeeeeeeeeeeeeeee III-Iii!!! .4 . | 1 u .h. , .9 . .1- , . . E ..1 pi. or w I“? N. . _ E Q TABLE h? 139 COMPETITIVE RELATIONSHIPS BETWEEN THE TWENTY-FIVE TOP FRESH FRUITS AND WETABLES, JUL! 12 m AUGUST 8, 1953* Expenditure Quantity as Percent of ‘ Average Expenditure Product Purchased Total Fresh P9” ”W Rank Per 100 Fruit and 01 3331111“ Persons Vgggublg Buying Mb Expenditure "9" 1 Raspberries 133 quarts 17.3 214 2 Potatoes 517 pounds 8.1; 38 3 Banana! 137 pounds 7.9 113 h Tomatoes 81: pounds 6.8 35 5 Muslmelon- cantaloupe 59 each 5.5 29 6 Sweet corn 362 ears 5.2 28 7 Head lettuce 70 heads 5.1 ha 8 Peaches 9h pounds 33.6 26 9 Watemlons 205 pounds 3.8 11 10 Oranges 30? each 3.5 18 Top 10 68.1 11 Blueberries 2O quarts . 2.9 10 12 Celery ho bunches 2.8 214 13 ,Sueet cherries 22 pounds 2.8 6 )5 Cucumbers 89 each 2.1 21 16 Carrots h? bunches 2.0 23 17 Sour cherries 31 pounds 1.9 3 - 18 Apples 55 pounds 1.9 10 19 Mature onions 3.6 pounds 1.6 18 2O Snap beans M; pounds 1.5 10 21 Gabba 1:8 pounds 1.2 18 22 ledishee 1:8 bunches 1.1 17 13 Green onions 36 bunches 1.0 12 2).; Plums 10 pounds .8 h as Peppers 2:. each .7 10 top 25 9h.3 111 others 5.7 4'» 1 Michigan State University Consumer Pam]. Date. TABLE ha comrrITIvs RELATIONSHIPS 13an THE TunmJIVE TOP FRESH FRUITS AND VEGETABLE-‘5, AUGUST 9 To smnnm 5, 1953* Quantity as Percent of Percent Expenditure Product Purchased Total Fresh of Families Persons Vegetable Week Expenditure 1 Peaches 535 pounds 1h.h h2 2 Potatoes 615 pounds 9.7 37 a, Husknelone 98 each 9.6 37 Bananas 103 pounds 7.3 33 S Tomatoes 273 Pounds 7.3 29 6 Head.1ettuce 63 eadh 5.8 . to 7 Sweet corn. hh? each 5.6 27 8 Apples 15h pounds h.5 20 9 Oranges 230 cad! 3.5 15 10 Blueberries l6 quarts 3.2 11 Top 10 7009 11 'Uaternelons 1h3 pounds 3.0 9 12 Celery' 32 bunches 2.8 22 13 cucumbers 166 each 2.h 13 1h Peers 89 pounds 2.2 6 15 Mature onions 6? pounds 2.1 18 16 Lemons 89 each 2.0 . 13 1? Grapes 20 pounds 1.8 9 18 Carrots 3h pounds 1.8 18 19 Peppers 6h each 1.3 12 20 came. 17 pounds 1.2 16 21 Snap beans 18 pounds 1.1 7 22. Plums 13 pounds .8 h 23 Radishee 22 bunches .1 9 28 Green onions 18 bunches . 7 2S Squash 1? pounds .6 6 IQ 25 9503 All others hell “hichigan State University Consumer Panel Data. TABLE 19 commune RELAIIDIEHIPS BETWEEN THE WHY-FIVE TOP r9353 FRUITS AND VEGETABLES, Move-em 29 TO Daemon-I 26,1953" Eacpenditure Quantity as Pacent of Average Expenditure Product Purchased Total Fresh Fem“ Bank For 1CD Fruit and of FamilEi: “1'90” Vegetable m Ekpenditure 1 Bananas 1116 pounds 111.0 M: 2 AppIe. 272 pounds 13.3 30 3 Potatoes 563 pounds 10.7 28 1; Changes 1:65 each 10.1: ’30 5 Head lettuce 80 each 8.2 9 6 Grapefruit 131 each 5.8 21 7 001m 30 bundae- 5.5 30 8 Tangerinee 270 each 11.9 18 9 Carrots 52 bunches 11.1 25 10 Grapes 23 pounds 2.5 10 Top 10 . 79.8 11 Cranberries 15 pounds 2.1 11 12 Cabbage 1.5 pounds 1.8 15 13 Tomatoes 19 pounds 1.8 6 11; Mature onions Ma pounds 1.7 10 15 Sweet potatoes 2 pounds 1.6 7 17 Squash 25 pounds .9 5 18 Radishes 13 bunches .9 7 19 Cucumbers 11. each .8 7 20 Lamas 2h each .8 6 21. Spinach h pounds .7 3 22 Turnips 114 pounds .6 S 23 Brussels Sprouts 3 pounds ' .5 2 2h Snap beans 3 pounds .h 2 25 Green onions 6 bunches . 2 All others 11.7 .Mchigan state Univu'sity Bummer Panel Data. lhz for the major share of the dollar value. Also shown for the four- week period is the quantity consumer per 100 persons and second, the average percentage of customers buying the item in any on- ueek period. Anaed with this type of data the discussion can nos proceed to an evaluation of its merit in tens of applicability to attain merchandising functions. Purchasing (he of the uses that might be made of this information is its use as a basic stock list for produce during each of thirteen four- ueek periods. It is realised that new stores carry more than these twenty-rive itus, but on the other hand there are probably new stores that do carry these twenty-dive cannodities let alone the different varieties and sises of each itu. Table 1&9 shows that the tap ten itass accounted for 79.1; percent of the sales and the nest fifteen items accounted {w an additional 15.9 percent of the sales. The question sight well be asked: if ten itans contribute almost 80 percent of the value, thy carry am more items? In other nerds, is there a case for Variety? is has been sheen in these tables there are a few "old stand-by” fruits and vegetables that make up the bulk of the average retailer's sales, both dollar and volume wise. These staples-opotatoes, head lettuce, tomatoes, salty, and carrots in the vegetable line; and bananas, apples, oranges, grapefruit, and strawberries in the fruit line asks up roughly 65 percent of the dollar value handled. These figures appear to build up a strong case for specialization. 1143 But do these figures tell the whole story? Host of the better retail operators are convinced that they do not; and from the evi- dence that is available on common- cpinions, consumers appear to be glad that nan: retailers feel that way. there is sound evidence that consumers will make extra efforts to shop in stores that stock basic items and those seasonal specialties that add “spice“ to the produce department. Each comedity has its‘season, but the develcment of new varieties, the use of better cultural practices, and the more flexible marketing systole have greatly broadened the base of availability. The sodas-n fully stocked produce department should carry the basic twenty-five commodities for each period of the year and then add to that list depending spa spaee, cutoasr wafer. ”.m. imperisonori‘ableh7sith‘i'ableh9shsesthatisthe smartherewu‘enineitensinthelistthatudnetappearonthe vista. stock list. lime, a retailer can emance his profits by pruoting these seasonal items and at the ease tine encourage in— creased constellation of the standard itus. Retailers who have made produce displm a during card in their stores have fond that sales have increased in other depart- nests. This is a natural consequence of the increased trend toward "one-stop” shopping. (I: the other hand, those retailers who have relied on a few staples to carry the load have discovered the fairly direct relationship between the number of items stocked and the mbc of customers. lhb, then specialty itans are in and out of season so quickly that certain retailers have discounted their importance. But while the higxly seasonal items like raspberries may constitute less than a three percent of annud sales, it may account for as such as 17 percent stall flesh fruitandvegetable salesincnenoath. Goat and hamning problems maybe greatn- on such itus, but so are the profit margins. Panel data shows that customers do buy these products resaly and in significant volume. Pm'chasing‘cr ordering is the first step in the produce oper- ation. Because of the perishable nature at the predict every order is a risk, but it can be a carefully calculated risk. Instead of depending entirely won his Judmt in buying: aerchandise, the retailer shsuld have an effective nsrohandise control mtes so that heuillknouuhich arehis fastestanduhich srehis slowest selling items. Bush a syste- vould help his to keep sufficient unbersottastnovingitassinetookandaveidevu‘stodtingslss saving items. Panel dsta in the torn presented in the tables can beusedssa guideto estinatinguhat it“ shuldbe stocked“ in that quantities. Mar, panel data would be a pee substitute for accurately maintained records or a particular store. For the day-team ptn'ehasing, sales must be estimated for the period covered by the m. that good operators would use sales records for the previous week and adjust this to thdr espectations for the fellas- ing weds. Nevertheless, panel data indicates when a particular commodity is going to be in peak demand. These seasonal items represent short-tern opportunities that must be anticipated and planned for in advance. 1145 There are three limitations to using panel data as an aid in purchasing and they are centered around the general nature of the data. First, there are several factors that influence produce sales iron seek to week. The weather will not only affect the sales tron sad: tcseekbutsillinnanyeases determinehoulcngthe itucan be kept without spoiling. Peak supply and danced conditions do not alums coincide from one year to the next. Promotion items that {satire a low price will often change the earnest of an it. for a trial period. Competing prices and nerdxandising actions can change the sales picture for a store. In other wade, panel data are not an accurate guide for day-to-dq M38138. Hanover, share advance twingcfcarletitens is medtheueefulnessofthedata arein- craased. Uhenana-deraust baplaoedtendeya pricrtoaxpected delivery, panel data provides a more accurate estimate of the expected movement, but its usefulness hue is still limited. The swond limitation is one involving the limited applicao bility to am we store. Panel data are taken from a statistical universe mosed of an entire metropolitan area. To the extent that any one store serves a neighborhood that is materially differ. ent in its food buying behavior M the statistical universe, the usefulness could be expected to be dissipated to that deg-ea. Undoubtedly, certain nati onality groups have buying behaviors that are unique to themselves. is pointed out under the four comedity chapters, income has sue effect on what is or is not bought. And certainly class of store such as “supermarket", "superettes", and '51! and pqe' sta'es could be expected to have different patterns of 1h6 produce sales. Nevertheless panel data shows what the customers are buying and what the ovu'all potential sales could be. If these two limitations are kept in‘nind this data can sure as a useful guide for determining basic stock lists and as a guide for alert- ing‘the merchandiser to the forthcoming seasonal itm. The third limitation applies only to certain of the cmodity items. To know that a octain anchor of apples are sold per capita and that they have a certain frequency of distribution is not enough in bwing apples. It is necessary to knee shat varieties and sises of apples are involved. Panel data does not provide this infatuation because of the difficulty of reporting other than Just the cancdity involved. This liniation does not apply to lettuce, carrots, celery, ate., whee the inflame of variety and sise are less inert-at. idvu'tising and ”suction This section is not intended to present a detailed account of the principles or mechanics of retail advertising. that will be presented is a discussion and examples of has panel data can be used in selecting itans to be advu'tised. The form that item selection advertisement would take is the direct action response as opposed to indirect action or institutional type. The purpose of the former type is to em goods out of the store by bringing peOple into the market in response to an advertisement. me forms that this adver~ tising takes are several in nature but most are naely variations of price proaction. master, with fruits and vegetables the use of educational type adva’tising along with produce price advertising 1h? is a fruitful area to break down prejudice against a product, promote intelligent bwing, and to suggest new uses for a product. Only by way of a thorough man-standing of the customer, his nature, his motives, his appetites, and his desires, can an advsro tisasent be sritten toward the custonu'. ‘mc customer is looking fa' nealtine ideas, for good things at reasonable prices, for household suggestions, for weekly specials, and for new individual wants and needs. In short, the operator must be capable of pre- jecting his thinking so as to realize shat people desire.1 In selecting items for advertising the operator should offer customers the itens that they want at the time that they want then. This aeans choosing items that are tinsel: because of season, holi- days, local events, and national advertising of producers; item that have side local appeal 1 item that are fimqucntl: purchased; and it's that help to sell other itu. mother factor that the water should consider is variety. Inclusion of varied it“ in his advertisements broadens there appeal. This is the area in which Inchigan State University Comer Panel Data can be helpful. A low at the tap ten itans in either Table 116, Table 1:7, Table Ill w l'sbls 349 shows which its!- have side appeal and are timely. The average percent of families hw- ing each week in all tables reflects the frequency of pin-dues. To cpl-in has this information might be used, a produce mclmndieing 1L. Javitch, J. Silver-busy, and L. Stienberg. "A Comprehensive Study of Food Chain Newspaper Advertising.“ Umublished Seminar, )fldxigan State University, 1951;. 1&8 prey-am for the answer months will be illustrated, The summer months were purposely chosen because they can be the met interesting and profitable ncnths to the produce department. The department under- goes a radical change during the pried from May well into September, sndthewcducensnhas tobequioktotake advantageofallthese changes. Perishsbility becomes more important, products are difbe ant, colors are acre varied, harvest and season deals are are prominent. According to s. H. Bolstad of Yon's Grocery Om, the four udcr em promotions involve salad vegetables, berries, salons, and soft fruit.2 ‘ Coming into mau- supply thoughout. the country in my and continuing until early fall are the sold vegetables which offer the best opportunity for sustained promotions throughout the smer. Profits on these items are usually better than average and there colors lend themselves to salad-boul-type of displays asking it possible to build related item displays of dressings, messes, spice, etc. The salad-bowl vegetables can serve to perk up ales periods tln‘cugh their cool eye-appealing attractive- sees. Tables 1:6,”, and M shoe that head lettuce always sakes an attractive droning card because of its frequnt purchase (range from to percent to 5’: percent of the families btwing each week). lettuce featured with the other volume sellers such as tomatoes, celery, cucmbers, carrots, cabbage, green onions, and peppers 2ll. mm 83 “Freshness Keys Smnm' Produce," Chain Store £2, July 1955, Po 11,9 are excellent for in-the-store promotions. As the home grown season arrives those items that are locally produced offer high volume possi-: bility for the store and high value produce for I the customers, Tona- toes are another good feature item ranking high in quantity mount and frequency of purchase throughout the entire am. In the Lansing area they hit there peak demand during August when hone-gown supplies are on the market. According to the data celery and cucumbers would make good secondary feature itus in the produce advertisement. The second major suamer pronoun involves berries with July being a peak aonth for all berries except strawberries which reach a peak during June. According to Table ’46 Movies were first in eunpemiiture rank with about 20 percent of the fresh fruit and vegetable dollar going toward their purchase. About 36 percent or” the families bought strawbcries each week and over a four-week period about one and a half quarts were purchased per person. During July more money was spent on fresh raspberries than my other fruit. Table I"! shows that 133 quarts were purchased per 100 capita in a four-week period and that about one-fourth of the families bought this fruit each well. mueberries were the eleventh neat Wt its: in July. With baries new of the operetws try to add novelty to their department by buying on the early season markets. This leans that the berries nust be top quality for in most cases the price will be higher and the appeals nest be in quality, freshness, and uniqueness. In fichigsn it would be desirable to feature buries when they first arrive on the market and then follow than through the season clinaxing the season with a peak of the harvest special. 150 This a department can earn a reputation for offering the unusual special and yet it will stress the best We. Tables to. M, and he show that melons are good feature itena hon laid-June to early September. Watermelons are at their peak in mchigan from add-June to aid-July. Although they are not a fre- quently bought itch, during their peak they do account for a large tonnage and are especially timely for Independence Day promotions. 0n the other hand cantaloupea seen to hit two peaks. During the period covered in Table h6 cantaloupes um purchased by about 1.1 percent of the families each week and during the sme four week period 71 osntaloupes sere purchased per 100 persons. These early saltaloupes are from California. is the season continues California declines in inpa'tence while the first of inchigon's crop begins; this accounts for the slight decline in constellation as reported in Table M. Ii‘he second peak hits during August when consumption climbed to 98 oantaleupee per 100 persons. Thus cantaloupes offer good promotion possibilities as 'first of the season specials' and “hone—m features.“ The fourth group of feature itus involve the morons soft fruits which the consumer anticipates as the smear months arrive. All of the soft fruits are highly perishable so they not be aoved fast. Cherries start off the soft fruit season in Michigan. They start about the first of July and continue for about one month. is a pro-atlas its they could be featured strongly for one week and then out back. According to Table M, sweet cherries accounted for 151 about 22 pounds per 100 persons in that four seek period with only 6 percent of the families buying. Sour cherries accounted for 31 pounds per 100 persons and only 3 percent of the families buying each week. In spite of the small number buying, the high consumption and high margins make both of these good in-the-store promotions. A split table display might be incorporated which would show met cherries for eating on one side and sour cherries for coming and baking on the other. Following the cherries is the peach deal parade. West coast and South Atlantis states start the parade during July, with llioldgan hitting the peak froe mid-August to nid-soptonber. table ha, which covers the period iron August 9, 1953 to Sept“ 5, 1953, please peadzes as the number one frnit and vegetable in both dollar expendiu ture and frequent or purchase. During this period 535 pounds sore purohased per 100 capita, 114.1; pee-sent or the fresh mm. and vegetable dollarsasspoudonpeaohss andh2poroeutorthe resiliesboughtth. timing the seat. is favorable as these results are, new retails-s fail to take advantage of this promotion qapa'tunity. Oupstition free produoer sources and high spoilage are the most frequently cited reasons for this apathetie attitude. Aggressive prmotion of bushel sises for earning will contribute to orvsrall. produce sales value; while featuring the pound size for eating will take advantage of the traffic building nature of this item The last of the najor soft fruit item is W88. Although the peak season for grapes does not come until early fall, the Thompson Seedless grape is in prominence dm'ing August. Table hB shows that 152 about 20 pounds were consumed per person during this period while about 9 percent of the families made week]: purchases, There is one vegetable item that comes into the produce depart- ment for major attention during the mar. Table 1;? shows that met corn started to come into the market in large quantities during the middle of July. Gm ranked sixth in expenditure with 362 ears purchased per 100 capita during the four weeks and 28 parent of the families bought corn ova-y week; Table 118 indicates that corn dropped to seventh in eXpenditure rank but the quantity purchased per capita increased to Us? cars per 100 capita and the frequency of purchase ‘ running allost‘identioal to Table LT. During this latter period the home-yous season is on in mmigan and as indicated in Table 1:8 larger quantities are purchased per shopping trip. The pranotion' of seasonal itana such as tormd during the summer 1:; meugm offers the eperatcr . reel Opportunity to add inta-est- ing variety to the produce departaent. Panel data helps to point out to the retailer the items that are nest important in terms or volme and frequency of purchase. These two factors of volume and frequency of purchase are two'important criteria in selecting product items that Hill have maximal: pranotion appeal. Panel date provide a guide in «letting item by use of these two criteria. (bee the item has been selected advance planning as to type of display, location of display, and related items tie-in are possible. The first three limitations to using this data for advertising and promotion apply as they did for purchasing. First, fluctuations in the mated-dew narket situation provide a limiting factor in 153 suggesting a definite date for a particular itm promotion. However, promotions should bsplanned far enough in advance so that preper ordering can take place, displays planned and built, and personnel informed. When such advance planning takes place, panel data be- comes more effective.- The limitation of applicability to any one store is less of a problm in advertising and promotion than in hiring. The items as shown in the panel data are the items that customers are most interc- estsd in. Information that shows the relative interest in the items, such as panel data does, is all that is'nesdsd to provide a guide as to promotion potential. The ordu‘ing a- buying of specific quantities for arc one stas takes a sore exacting type iota-nation that panel data can provide. The third limitation of lack at knwladgs as to variety, sise, quality, etc. is less of a limitation to use than are the tuner two. All that panel data should do is provide a guide as to the relative opportunity each item provides as a pronotion possibility. Although further information about a product would be helpful it should not be a detriment to panel data use. For example, in Table 1:8, tomatoes appear to be a such better promotion iten than apples; howevu', during the period award in Table h? the opposite situation prevails; apples are a better promotion feature than tomatoes. Again this limitation will only apply on ituas that are heroinandised in wide ranges of sises and varietissu 'Il'mfltnifllaw u‘ [Kurtaw \é o... 151,. . 84 1w ‘2» «NW! 15h Display How to make the best use of space is a ”Q problm in the food retail store. To a self-swvice operator this is particularly inpu- taut. His store is set up on the theory that goods are sold by dis- plqring then. The sis in allocating space within the selling area is to maximise the productivity of the entire display area by causing each square foot to asks its preper contribution. is desirable as this information is in allocating store space to the produce department, panel data would be of no' help in supply-.- ing figures on which to make an appraisal. Several measures of productivity of space could he considered in asking allocations. Among those that can be expressed on a square-foot basis are net profit, gross profit, sales in terns of dollars, and sales in terms of quantity sold. Net profit represents the most desired aeasure. When 0‘08! profit instead of net profit is used as a measure of space productivity, some adjustnents are necessary to allow for those cosnzcdities for which operating costs are relatively higher. Dollar sales per square foot of floor space is one step farther reamed from net profits than is gross profit. This is because dollar sales nuet be adjusted for variaticn in both camodity lax-gins and mating costs. Data on quantity sold takes no account of return to the store. Sin” panel data does not provide infmation on any of these four neasurss of performance its usefulness as a guide to produce department space allocation is negligible. In the meanest of space within the produce department panel data is again of little usefulness. For panel data to be effective, .J. I. .IIIIJJ'LIi’sharniF repay .. a y n... u . , — a!“ e-Ie _ . 19""! 155 fresh fruits and vegetables would have to be sold in proportion to the mount of space devoted to it. For example, using Table 148, peaches would receive 114 percent of the space because it accounts for 2.1: percent of the total fresh mat and vegetable expenditure. likewise bluebl'ries and watermlons 3 percent of the space, cantaloupe 10 pacent and squash less than 1 pacsnt. The primary fault of displaying in proportion to dollar sales value becomes obvious from considering the before-sentimed items. namely, bulky itms such as squash, watermelon, peaches, m cantaloupes require a certain minim: space regardless of their sales potential. Less than 1 percent of the space would hardly be adequate for squash) while blueberries sill probably require less space than watermelon although they both account for. 3 percent of the sales dollar. In a study by H. Wayne Hitting on produce €10th space utilization some interesting obserVations were made on the mage- nent of spec-.3 It can found that potatoes and onions have low rates of sale per square feet. In contrast to those two iitms, dollar sales per square foot of floor space for tomatoes and lettuce are larger than the percentage of total floor space theywoocupy. However, there are several factors that acct be considered before it can be concluded um up.” should be shifted from potatoes and onions to toeatoes and lettuce. First, high-value non-bulb itus nay require less space 33. Wayne Bitting, "Produce Department Space Utilisation, Gross Mgins and Operating Costs in Selecting Retail Stores, Charlotte, N. 0." United States Department of Agriculture, Market Research Report no. 36, Hashington, D. c., 1953. 156 then bulky lee-value items. Second, increased sales may not result in prepcrtion to the increased display space. Third, it might be advantageous from the standpoint of efficient use of labor and min- taining adequate selectivity for autumn, to allocate space beyond that indicated on the basis of dollar sales per square foot of floor spm. Fourth, the margin of gross profit is an influencing factor . in allocating space. ‘ The second factor of sales not increasing in proportion to additional spans given then is further elaborated upon in a study by Walter B. Hinkle, Jr... entitled ”Merchandising Fresh Fmitsand Vegetables in Retail Stereo."h In the eta-es studied, large displays usually sold more of individual fruits or vegetables than stores having mall displays, but the dirfarmee in sales were usually far from proportionate to the increase in size of display arses. FC' mle, it tool: a 223 percent increase in “rage sise of cauliflower displaytobringaboutellflpercnit incresseinsalesperstwe. It required a 296 percent difference in chaplay to bring about a 66 per- cent increase in the sale of tomatoes. Greater results were usually obtained by increasing use a: display on Friday md sum-m than during the fore part of the week. Since total new display in a produce departnent is limited, an increase in size of displv of one commodity means that the area allotted to some other commodity must be correspondingly reduced. l‘Hulter B. Hinkle, Jr. 'Practices Affecting Sales and Spoilege.” Merchandising Fresh Fruits and Vegetables in Retail Stores Part II. varsity Igimfltuarfixpainentgutfon, LE. “3, I935. R :I-PLIEL .WA.» Ar: . . , , .sv . qu . . 2‘ IE. a = so , ~. «I . 1!-HLLV. Etna; .IJI . , .. w 157 The allocation of space to produce items is probably tied up with the way the retailer buys, rather than the salon requirementa. Honuperishablee are bought in fairly large quantities. They are ' then allowed to take up a large part of the display space to mini- nine movement into and out of storage space. fibre frequent deliver- iea of the perishable. ere required and the onstm has been to pile up limited stock on bond as with the lees pcishable products. Although panel data is of limited usefulness as {Ir aa indicat- ing the correct amount of spec. that is necessary'to maximize salsa and net profits, it can be of use in a general my. One of the principles of produce rack display is to space the large sellers so that lower volume "impulse" items can be intermingled among the more treqnently bought items. Phnel data.indicetee which items ere sell- ing beet and'ihet there frequenqy of purchase ia.- in the season: change certain produce itqea ehmld be given we open, eone less space, and some eliminated. Panel data indicates this in a general way. For example,'hy comparing Table h? with Table hB, indicates that the amount of apaoe peaches were given in luguat should.be greater than July, and apple space should likwioe be getter in Auguat thsn'in My.» An examination of other items in thee. tablea and other item between tables will reveal the need for ohanging the relative size of display of each commodity between seasons. These findings indicate the need for keeping records on individual itema so that the return per square foot of floor space is known. Such intonation will enable the store operator i. . . . .lI-K. z .. .e. . u, Iliu‘ulfiflfiHUéV: .u.. no. u I . ' a —u‘-. .-.. L1“ l-,b - Wm; 158 to allocate space more efficiently. For the reasons elaborated above, panel data does not provide this information. Pricing The problem of pricing merchandise is one that is a vital problem to all merchants whether they are large or small. Prices so low that they do not cover costs or expenses, or so high that they lessen sales volume will have an adverse effect on profits; and if not corrected by use of more realistic prices, will soon put a stars out of business. Also, since new men give high prices as a reason for trading at more than we store, the grocer who is interested in increasing his store voltme, holding his regular customers, and making his: store a one-stop food center must be cmpetitive in price. In general, well defined limits govern the'anount of margin that a store operator can add to the purchase price of his sea-cham- dise, the lower limit being one that will yield sufficient dollars of gross profit to cover operating costs. The upper limit is the one that will give the largest net profit above his operating costs that his competition will permit.- Mover, most retailers follow a aidfle-ofcothe-e-oad practice in establishing an overall produce department gross margin. This enables the retailer to be canpeti- tive and yet expend futm'e volmer Likewise, prices are not set so that operating expenses are merely covered, unless a large share of food retail business is being temporarily sought for that new be a more profitable field latero' 159 The point of departure for fresh fruit and vegetable pricing is the price necessary to cover, merchandise cost, merchandise losses, operating expense, and leave the residual net profit. This would give the operator a certain ova-ell gross profit percentage figure from which to base calculations. For example, if an opu'atcr has thefollowing costs; operating expenses 11mm, merchandise loss 5 pore-at, and not profit (desired) 3 percent, the overall pricing objective would be 25 percent. However, this does not mean that an operator would necessarily markup each produce ’ item 25 percent. Host progressive food retailers use flexible sax-gins between item and on the sansitens over ties. In the study by H. Wayne Bitting, the author points out that individual stores follwed no consistent pricing poliq.5 Margins on single itess within a particular store fluctuated over a three week period and margins on single itune between stores at a par- ticular time period fluctuated widely. This seems to point out that individual retail“ do not fellas an inflexible setup fa- all produce item but rather vary the markup according to ethn- criteris. And second, this stw points out overall pricing objectives in terse of gross profit vary between stores. mess I'othes' criteria" that are used in pricing fruit sake it difficult to subject pricing practices to am scientific formula. The criteria or factors that cause price variance between stores are as follows: Sutting’ 20 01%, p. 12. he. cw l. 2. 3. 160 Different pricing objectives (includes operating costs, merchandising losses and net profits). Different costs of 'nerohandise (based on differences in quantity and quality bought). Competition. The criteria and factors that causes margin differences between itm within a store are! within items over a time period are as follows 3 1. 2. 3. h. 5. Advertised "less leader“ features (low margins are taken on these itans in the hope that high volus sales of the particular items will compensate for the la margin end/or items carrying a high- margin will be purchased; and second, a low price impression for the store will result). Different costs of merchandise (based on difference in quantity and quality purchased and difference due to Inpphr and damn! factors). Volume of sales (items bought in larger amounts and sue them“: usually «217'me mu main). Operating costs (items requiring more handling usual): any was! main). Spoilsge (items in whid: spoilage, markdowns, and othc losses no: high usually require high.- margins). The objective of the sales plan is to arrive at specific margins for groups of produce items while maintaining a sales balance between high and low margin items in relation to their sales volume. The margins are combined so as to arrive at an overall department margin .‘ 161 that is adequate to cover upensee and leave a net profit. The plan also permits a flexible pricing systnn, so that feature items m be prmoted without sacrificing profits. Thus, the plan permits morass to be taken on individual items, so that consideration is allwed for cospetitien, volume potential, operating costs, losses, and promotional featm‘es. A study of past sales records enables the merchant to «flute quite achn'ately his produce requirements for a given tine period. Then, by placing realistic flexible nargim on groups of produce item of a km coat, future dollar sales can be estimated. By totaling the estimated sales for each group of produce it“ a department sales potential will result. Itaus are gathered into the following maps for. convenience purpose; citrus, bananas, other frnit, vegetables, potatoes, and feature item. The margins assigned are arbitrary but yet they take into account the critcia of volume sales, operat- ing costs, and spoilage. A most accurate plan results when costs are known, but barring the knowledge of known cost a sales plan can be derived in advance by using panel data. In place of a known cost the percentage of total fresh fruit and vegetable expenditure can be used. By using the data in Table 1:8 and assuming that during one of the weeks of the months the retailer would like an approximate 2b percent gross margin, a sales plan as follows could result. fl; Keith .3!“ I.» ea Eel-I'd}. ~'!'*"%; 162 Product Grog Produce Sales Gross Profit pencil-ed Distribution Citrus . 6% 25% (25% s 6%) - 1.5 W 7 20% (20% x 7%) - 1.h Other fruit 18 30% (30% x 18%) - SJ; Vegetables 28 35% (35% x 28%) - 9.8 Potatoes 110 20% 20% x 10%) - 2.0 Main feature)- lh 5% ( 5% a: 1M) - .1 Secondary feature: 11 15% (15% a 17%) - 2,6 100% Groes Margin 234.16 1Main feature items - peaches. 23mm." feature items - mum. and tomatoes. This sample sales plan shows that planned goes nargin of 2h.h percent can be realised if all of the merchandise is sold and the selected margins realised. It must be recognised, however, that these are planned sales and in retail practice only a portion of it is realised because of m, shrinkage, mm, pilferage, cash register m, etc. AM a study of department records over a period of tine, a figure representative of these unrealised sales can be made. Depmding upu the store, season, equipment, etc., this figure ranges from about 2 percent to 6 percent. Thus, sub- treating a shrinkage figure from the planned margin, an adjusted aargin figure is arrived at. We, fluh percent - h percent - 20.h preset. ’ The enchant nust now decide shether this margin is too low, too up, or shut right. His past «parishes and sales record are the best guide for determining this. To raise the margin, the nargin on em or all groups should be raised. To lower, the process should be reversed. All produce in any one group does not necessarily ....... Elle-III .! 15""! a. Hit; ’zeix 06‘ w 163 have to be priced at the same profit margin. Instead, the merchant nay use average margins of profits methods. For example, the ”met move" type of produce are priced "on the easy side" margin of profit in «def to speed sales and turnover. mus, panel data are quite useful as a guide to produce price planning. However, where known merchandise costs and relative merchandise :1st are known for a particular store, the usefulness of panel data is diminished. But, it should be emphasised that prist- ing tables should only be used as a guide and should not supplant the individuality and experience of a food retailer. Because of the uniqueness of a particular operator and his store the art of merchan- dising will never be reduced to a scientifically precise formula. Mary In the four commodity chapters, individual items were studied intensively over an extended time period. Little thought was given to the relationships that exists between products for a short tine intn'val. In this chapter the aphasia is changed away free intensive study of one item to a study of the usefulness of lichigan State University Omar Panel data as an aid in performing certain speci- fied functions of retailing. ‘lhe approach taken was to sundae the relatillhipshetweu itees‘ during a four-week period to see how this infmaticn was applicable to the functions of purchasing, advertis- ing and promotion, displaying, and pricing. is an aid to purchasing, panel data was a useful guide for advance planning because it indicated a basic stock list based upon the most important dollar-wise items. However, as a dw-to-day 16h purchasing tool the usefulness was limited because of fluctuations in daily supply and demand, the difference in product deoand betwew stores, and the range in quality and variety characteristics within a single item. These some limitations apply to data used in adver- tising and practice, but again the panel data is a good guide in selecting advertising features based upon the oriterip of its. sales volume and frequency of purchase. Panel data was not too useful as a guide to allocating item display space within a produce department because of the necessity of devoting more space to bulky items to receive a unima- of space than there relative sales value would indicate. In pricing, panel data is an excellent guide in sales planning. By knowing the relative movement of itens certain flexible margins can be assigned to these itms so as to arrive at an overall satisfactory you margin. However, pricing tables will never supplant the individu- ality and experience of a food retailer. CHAPTER VIII swam ' In 1951 the Michigan State University Consumer Panel was estab- lished as a means for receiving information on the food purchase behavior of Lansing, Inchigan families. Diaries are filled out each week so as to contain the food :1th purchased, quantity bought, price paid per unit, eapmditure, and, for some products, method of purchase and type of product preservation. One of the areas of usefulness of these data is to the retail food trade. By providing information on she the customers are, what they bw, when thsybuy, andhovnuohthey buy,panel datahelpstcansverthe question» What do the unto-acre nut?" 'l'hepurpcseof the studorsastoshcw thetypesofdetethstare available fro: the panel for use by retaila-s she handle fresh fruits and vegetables and second, to evaluate its potential as an aid in retailing. To accomplish these purposes an intmsive study of four cmdities was presented. In order to develop the subject nest cowletely studies that pertain to the comedity were discussed along Iith pertinent panel data. Secondly, panel data were presented by relating it to cctain functions of retailing through the use of all the major fresh fruits and vegetables in the short time interval. The first commodity studied was potatoes, which are the top its: in tomage moment and usually highest in consumer emenditure dollm. The Michigan potato is the most popular of all potatoes 166 and should be carried by retailers in Michigan during nest of the year. During the fall prspaok sises should be 15 pounds, and 50 pounds. In the winter the 10 pound size should be added to this list. In the spring the sizes nest in demand were the lo and 15 pound units. When the early crop comes to market during the late sumer the large size unit of either 50 pounds or 100 pounds should be added to the 10 and 15 pound been. The Oalifornin potato cones onto Met in the spring at about the same time that Michigan stocks are depleted and are off the market by the time Hichigsn's crap starts in the late m. The 10 poundsiae unit scans to be the only Justifiable sise. The same sise package should also be mployed with Idaho potatoes, which are in canpetition with Michigan potatoes. Heine potatoes should only be carried in the winter season in the Lansing area. In spite of their pulsation in lensing, the Maine potato accounts for only 5 per- cent of the potato expenditure during their peak winter «soon. In spite of the predominance of prepackaged potatoes, the bulk sale of potatoes still accounts fa- about 25 percent of all sales. It would be a good practice to offn' a snail dispr of loose potatoes from either a display bin or table. Fa those who do not choose to use cull open displays, several prepackaged bags should be opened fa- selection purposes. ‘lhis will not only serve those customers desiring bulk potatoes but will build confidence in pre. packaged displays. In studies to determine the acceptability of various packaging containers the results clearly indicate the damn! by customers for 167 product visibility in their selection d potatoes. The desire was strong enough that they were willing to pay a four cents premium for a polyethylene package over and above the mesh paper window bag which gives only partial visibility. In conjunction with this study it was ascertained that customers were willing to par a two cents premium fa' washed potatoes over and above the added cost of washing. From the data on family characteristics, several conclusions can be drawn. First, there appears to be a need fu- eustcner edu- cation on the low cost-mg food value that can be derived fru potatoes, Second, two price lines of potatoes should be carried; one a quality potato at a higher price and second, a low price competitive potato. Third, the infrequent purchase of potatoes by snell families, points out the need to lake loose bulk potatoes available to those families that purchase only a small total amount of pets“. The second connodity studied was apples, which ram-d third in amnal fresh produce apenditure. Octoba‘ was the nest inpa'tant month for sales of this seasonal commodity. Since ever 50 percent stall saleswu-einunits oflOpmds are?» andsincenw retailers fail to carry large units of pmehase for an extended tine period it suggests that new opportunities for increased apple sales are being missed. To solve this problem the bushel and half- bushel sise could be successfully sold during certain select periods of the year. In the smaller sises the package weights should be Varied so as to take advantage of the seasonal nature of apples. A retailer should not package one weiQt alone, and: as in pounds, but should vary the 168 weights between 2 to 6 pounds with the majority of the bags packed at the nest pepular seasonal sise. Panel data and merchandising studies have shown that bulk displays alone and prepackaged displays alone are relatively ineffective when compared with combinations of bulk and prepackaged apples. The objective is to appeal to the desires of the maxim number of customers without sacrificing labor efficiency. Data on the relationship between family characteristics and fresh apple purchases points to the need of having (1) the period when apples are available for sale extended, and (2) education that would encourage children to eat more apples. Although the fresh orange is still a large volume item, its sales in the past few years have fallen at the profit of the frozen orange Juice concentrates. In comparim fresh orange purchases free year to year it appears that increases in expenditure are related to practice and neohandising effort as well as price. A review of the studies on pricing methods revealed that customers preferred selecting their oranges when priced by count rather than by weight. In view of this preference it would still behoove the retailer to avoid a caaplete change over to pricing by weight. The prepackaged orange is far from being universally accepted. In the winter of 1955 about to percent of the purchases and 16 percent of the quantity was sold in prepackaged form. The only sises that seen logical to package are the one and two dozen units. Families that buy the largest quantities of oranges to families that have luger family incomes and larger numbers of persons. There was a direct relationship between per capita income and per capita 169 quantity, Although larger families buy more «ranges per family, they buy muer amotmts on a per capita basis, which would suggest that orange consumption is low among children. The last commodity studied was grapefruit. This item is on the narket in Lansing in all but the comer months. From the data it appears that consunption per capita increases as the percentage of families having increases. More than 16 percent of the lensing families bought no grapefruit at all in 1953. is with oranges most customers still prefer to buy their grape- fruit priced by count rather than by weight. The change in prepackag- ing from alaost couplets absmoe in 1919 to about 1:5 percent of quantity being sold in prepackaging form in 1955 indicates that many of the former objections to this fort: of merchandising have been shunted. Since there appears to be no definite size of package that outsells my other sise a retailer should «change the quantity to fit a particular enchan- dieing situatiu. Although new customs have node the decision to buy a particular type of citrus product before entering the food store, that decision to boy was often changed after an inspection of the archandise. The retailer can help to stimulate the desire to purchase by selecting fruit of good quality and full naturity and then follow the desirable handling and merchandising practices. Since most of the grapefruit are purchased by families that are of small sise and large per capita income, advertisunents to increase consumption should point out the low cost per serving for the health value derived. 1.55.45“ , , m 1. I (H. A L. M..I§u3m¢r.p...unmlr_ n. . . "‘ t' 1“ U“ “ ‘ .331“; 170 In the four comedity chapters emphasis was placed on the indi- vidual item while in Chapter VII the emhasis was changed to a study of all fruit and vegetables in the short tine interval. Relating this information to specific retailing functions revealed the following conclusions: 1. As a guide to purchasing, panel data were useful as a tool in planning a basic stock list, but as a tool in day-to-dsy ordering the usefulness as limited because of fluctuations in daily suple and duand, differences in demand hm stores, and differences in the range of quality and variety. 2. The same limitation applied in advertising and'prcnctim, but the data are useful in selecting promotion item based on the criteria of sales 1volume and bequency of purchase. 3. Because of the variety of factors other than sales Velma and sales distribution that affect display allocation, panel data are of little help in reaching merchandising decisions. 1;. Pricing sales plans can be derived based upon panel data; however, their usefulness decreases in relation to difference between sales distribution as reflected at store level and as reflected in panel data. Panel data serve as an excellent guide in helping the indiVidual reach merchandising decisions but its use should be tomes-ed with the experience and ability of each individual retail operator. APPENDIX 172 _ :‘y “fl . "“1 ‘v u? - cw smuomurm Bots . moments, I". and I. livers. ‘flos They Tell What We Want.“ Iearboot of AEcultux-e. United States Department of Ayioulture, n. e as. 1951‘, pp. 207.211. I Parfanberger, A. I. %cholog in Advertising. (lineage: A. w. Shay p. e Cm, 1926, The hogessive (h'ocer. Hoe to Make Money S Fresh Fruits and Ye stables. Ha I“: Buttes-ink , corporated, 9' PP- Halsh, R. H. “And What Are Its Parts?" Iearbodc of A culture. Outed States Department of Agriculture, Heshiugtgn, 5. 5., 195“, pp. 8-11. Periodicals Anon. uProduce lineal." Chain Store Age. lurch, 1955, pp. 91-126. Applebau, W. "Studying Consular Behavior in Retail Stores." Journal of Marketing. October, 1951, p. 172. Bolstad, H. H. 'h’eshness lqs Smer Produce." Chain Store Age: July, 1955, pp. 63.62. McCoha, H. "Stock More and Sell More." The Grocers Digest. July, 1952. Shaffer, J. D. and C. C. Quedtenbush. "Cooperation and Sampling in Four Iears of M.S.U. Consumer Panel Operation." g%erlz Bulletin. Agricultural Experiment Station, Michigan State versi y. ugust, 1955s PP. 85-103. Shaffer, J. D. 'A Plan For Sampling a Changing Pepulation Over Time." Journal of Far- Eoonoues, Volume 37, 1951:, pp. 153-163. Sherun, R. W. and G. H. Mitchel. "Sell Prepaotaged Apples in a Variety of Sins." Massive Grocer. January, 1951;, p. 92. Bulletins Anon. "Citrus Preferences Aaong Household Consumers in Louisville and Ielson County,Kenu1dcy.' United States Department of Agriculture, Washington, D. 0., Information Bulletin No. 2, January, 1950. 17h Anon. "Consuner's Use of and Opinions About Citrus Products." United States Department of Agriculture, Washington, D. Cu Agricultural Information Bulletin No. 50, 1951, pp.‘ lql6. Anon. "Effective Displv." Produce Handbook. Part II. Proceeding of the SKI Regional Hanagers Heating. Supermarket Institute. Anon. "Merchandising Studies in Supermarkets-u-Apples, Lettuce, and Tomatoes." United States Department of Agiculture, Washington, De 0e, ‘MSGIB. 1955' 9 pp. Anon. "Q'dering ma Inventory Control." Produce Handbook. Part I. Preceedings of the SE]: Regional Hangers nesting. Supermarket InatitUtCe Pp. 6‘9e Anon. "Potato Preferences Anong Household Consumers." United States Department of Ag-iculture, Washington, D. C. Iiisoellaneous PUDlicttlan “be 622, 19h3. Buis, I. P. and R. A. Seelig. "Fruit and Vegetable Facts and Pointers." (united Fresh Print and Vegetable Association. Washington, D. C., 19 2. . hitting, H. H. "Produce Department Space Utilisation, Cress Hal-gins, nnd Operating Costs in Selected Retail Stores, Charlotte, I. 0." United States Departnent of Agriculture, Washington, D. C. Market Research Report Ho. 36, 1953. Davis, L. 3. "Applying Inproved Apple Icohandising Practices in Retail Stores." Cornell University, artment of Agricultural Economics, Ith.c., He 1., Bu11Ctin|870, 19 g 35 pp. Dosnie, E. D. and 1.3. Trienish. 'Consmer Buying Practices and Preferences for Purchasing Ganges by Height or Count in Selected Citiec." United States Depart-ant of Agriculture, Washington, D. 0., Juan, 19S°e FUgoth, ‘et‘e, 8e ‘e quton, Ind H. H. Hittinge .61tru. PreferIDOOl Along Customers of Selected Stores." Texas Agriculture and Mechanical. College, Bulletin 722, 1950. Cousin, H. R. "Consular Acceptance and Retailing losses. Consumer Packaging as a Method of Retailing Fruits and Vegetables in the lortheast." Part I. Cornell University Agricultural meriment Station, Ithaca, lies York, Bulletin 870, 1951, 35 pp. Hinkel, W. B. "Practices Affecting Sales and Spoilage. Kerchandising Fresh Fruits and Vegetables in Retail Stores." Part II. Cornell University Agricultural Encperiment Station, Ithaca, New Iork, LE. 61?, 1952, 52 PP- 175 table, B. 1.. and R. Mauby. "Retail Produce Departments." Agriwlturel mansion Service and Agricultural kperiment Station. Purdue University, Station Circular 382, 1952, 11 pp. Perry, A. "Plastic Bags for Potato Packages." University of Maine. Bulletin 511‘, m, 1953e Pillar, R. "A Guide To Better Handling and More Efficient Merchandising of Fresh Fruits and Vegetables." fictional League of Wholesale Fresh Fruit and Vegetable Distributors. Washington, D. 0., Harsh, 19158. Rasmussen, H. P. "Comer harshness of Fresh Fruits at Retail." Cornell g‘niversigy Agricultural Experiment Station, Ithaca, N. I. Bulletin 9, 19 . __________. "Consumer Purchases of Fresh Fruits at Retail." Cornell Uni- ;Seisity Agricultural bzperiaent Station, Ithaca, New York. Bulletin 191:9. . D Shaffer, J. D. and G. C. Quaokenbush. "Consumer Purchase of Apples." Ap'icultural kperinent Station, Elohigan State University. Uqu, 0. "Public Produce Harkets in Inchigan." Agricultural heri- mt Station, luchigan State College. Special Bulletin No. 268, 193?, Fe 13c Unpublished late-1.1 Bacall, B. R. "Promce Hsrchandising in a Retail Cooperative." Unpublished ILA. Thesis. Graduate School of mohigan State College, 1951;, 99 pp. Javitoh, I... J. Silva-burg, and L. Steinberg. "A Cosprehensive Study of Food Chain luspaper Advertising." Unpublished Suinar. Inchigan State University. 1951:, 68 13). Riley, B. 1!. "Sue Hcasuresents of Comer Denand fa- Heats." Unpublished Ph.D. nmu. out-mt. School of Michigan out. 0011620. 1951:. 17!: pp. haffer, J. D. "hethodlcgcal Bases for the (pa-anon of a Consumer Purchase Panel." Unpublished Ph.D. Thesis. Graduate School . of Iichigan State College, 1952, 751 pp. Speeches Helnicoff, D. G. "We trends in Produce Connunption." A speech presented at the National Association of Food Chains Clinic on march 155 195,4. 176 Quackenbush, G. G. "Demand Analysis From the 11.3.0. Consumer Panel. A paper delivered at a Joint meeting of the American Statistical Association and the American Farm Economics Association. Wash- i-nztm. D. 0., December 30, 1953. mscellaneous Anal. "Census of Business, Retail Trade." Preliminary Data. Bureau of Census, Washington, D. 0., 1951;. Anon. "Census of Population. Characteristics of the Population." Volume II. Bm'eau of Census, Washington, D. 0., 1950. WW"I”WI""“llllWWll””“lllHlW'll 293010256554