The Relative Financial Efficiency of Brazilian Firms and American Firms in the Manufacturing Sector: A Ratio Analysis

Transcription

1 Joural of Applied Fiace & Bakig, vol. 3, o. 5, 2013, 1-18 ISSN: (prit versio), (olie) Sciepress Ltd, 2013 The Relative Fiacial Efficiecy of Brazilia Firms ad America Firms i the Maufacturig Sector: A Ratio Aalysis Parviz Asheghia 1 Abstract This paper evaluates the relative fiacial efficiecy of twety three matched-pairs of U.S. firms ad Brazilia () firms. I this study, efficiecy is measured i terms of profitability, debt maagemet, asset maagemet, ad liquidity maagemet. Paired compariso is employed ad te hypotheses are tested o the basis of the defied ratios. Because matched pairs are used, a appropriate test is the Wilcoxo matched-pairs siged-raked test. All the data for the study were compiled by the author from Merget o Lie. These iclude the most recet five-year time-series data that were available i 2013 for all the te ratios that were tested. The aalysis preseted i this paper idicates the absece of ay statistically sigificat differeces betwee the two sets of firms with regard to most of the ratios examied, suggestig that the U.S ad the Brazilia firms are similar to each other with respect to their fiacial efficiecy. The oly exceptio is that firms have higher retur o equity (ROE) ratios tha the Uited States firms. JEL classificatio umbers: F14, F21, F23, F30, L25, M15, Μ40, N60 Keywords: Fiacial Efficiecy, Brazil, the Uited States, Profitability, Debt, Maagemet, Asset Maagemet, Liquidity Maagemet, ad Fiacial Ratios. 1 Itroductio A ewly idustrialized atio, Brazil is the seveth wealthiest ecoomy i the world. It is also the largest atio by area ad populatio i Lati America ad the Caribbea [1]. As oe of the so called Bric atios, alog with Russia, Idia ad Chia, Brazil has oe of the world's fastest growig major ecoomies, ad its ecoomic reforms have give the atio ew iteratioal recogitio ad ifluece. [2]. 1 Departmet of Ecoomics, Califoria State Uiversity Sa Berardio, A. Article Ifo: Received : May 31, Revised : Jue 29, Published olie : September 1, 2013

2 2 Parviz Asheghia I recet years, Brazil s GDP has grow by a average aual rate of over 5 percet, which exceeds the growth of the U.S. ad Wester Europe. This coditio has cotributed to the ability of this atio to survive the global fiacial crisis with relatively isigificat effects. Brazil was oe of the last atios to plummet ito recessio i 2008 ad amog the first to resume growth i Followig its ecoomic recovery, Brazil experieced GDP real growth rate of 7.5 percet ad 2.1 percet i year 2010 ad 2011, respectively [3]. Although Brazil s rate of ecoomic growth decreased i 2011, ad 2012, strog policy support is predicted to ehace this coutry s rate of ecoomic growth to 4 percet ad 4.1 percet i years 2013 ad 2014 respectively [4]. Give the impressive rate of growth of GDP i Brazil i recet years, may researchers have tried to ivestigate the elemets that fuel the Brazilia firms efficiecy. However, o study has examied the relative efficiecy of the Brazilia firms ad America (U.S.) firms operatig i the maufacturig sector. The purpose of this study is to evaluate the relative fiacial efficiecies of twety-three matched-pair of America ad Brazilia maufacturig firms by employig ratio aalysis. 2 Survey of the Literature There are oly a few studies that cocetrate o evaluatig firms efficiecy i Brazil. However most of these studies cocetrate o measurig productio efficiecy of the firms uder cosideratio. I this sectio we review several of these studies ad preset their results. I their study P. Tecles ad B. M. Tabak [6] aalyze the efficiecy of the Brazilia bakig sector over the post-privatizatio period of Employig a Bayesia stochastic frotier approach, they compare across istitutios ad bak groups. Their results idicate that large baks are the most cost ad profit efficiet, cofirmig the cocetratio process observed i recet years. Additioally, they explai that foreig baks have achieved a good performace by either the establishmet of ew affiliates or the acquisitio of local baks. The remaiig public baks have portrayed improvemets i cost efficiecy; however, they are relatively profit iefficiet. Fially, they fid that the impact of capitalizatio o efficiecy is positive. I his study T. Kida [7] employs a oe-step stochastic frotier model for five developig coutries, icludig Brazil, Morocco, Pakista, South Africa ad Vietam. He cocludes that foreig firms beefit from a better ivestmet climate, which provides sigificat explaatio for the higher efficiecy of foreig firms as compared to domestic firms. The article also examies the importace of vertical spillovers, while cotrollig for the direct impact of the ivestmet climate o efficiecy. The results idicate that firms, especially small local firms that sell more of their output to multiatioals, are more efficiet, idicatig the existece of vertical spillovers via backward likages. I their study N. Nazmi ad J. E. Revilla [8] compare ecoomic efficiecy of Brazil, Idia, ad Chia. They defie ecoomic efficiecy as the gap betwee potetial ad actual output for a give iput combiatio ad techological factor. Usig stochastic productio frotier models, they measure the cotributios of factors of productio ad techology to growth ad estimate o-positive error terms that capture productio iefficiecies i each atio. Their results idicate that Chia ad Idia had relatively

3 The Relative Fiacial Efficiecy of ad U.S. Firms 3 iefficiet productio i the early 1980s but have sice improved productio efficiecy substatially. I the same period, productio efficiecy i Brazil has declied to some extet from relatively high iitial levels. Additioally, the gap betwee productio efficiecy betwee these atios has arrowed sigificatly, implyig more rapid growth i Chia ad Idia as compared to Brazil. O. Yeboah, V. Ofori-Boadu ad T. Li [9], employ Data Evelopmet Aalysis (DEA) to measure ad compare the efficiecy of ethaol productio i Chia ad Brazil. They develop a output orieted model to estimate the extet that output ca be proportioally ehaced without alterig the iput quatities used i each atio. Their results idicate that Chia has bee relatively more efficiet i ethaol productio tha Brazil sice the year E. P. Ribeiro, V. Prochik ad J. DeNegri [10] examie productivity ad growth i the Brazilia iformatics idustry (ISIC 30 office, accoutig ad computig machiery). This idustry which is oe of the fastest growig sectors i maufacturig i the world receives special tax breaks i Brazil. Usig both iterview method ad ecoometric model, they ivestigate this sector after its liberalizatio. The iterviews suggest that firm growth is based either o product differetiatio i the busiess machies subsector (usig close software complemetarities) or retail chais as distributio chaels i computer maufacturig. Their ecoometric aalysis idicates that although the Iformatics Law has a overall positive impact o the firms, its effect o the productivity growth is ot sigificat. O average, less productive firms receive more Iformatics Law beefit. Their fidigs cast doubt o the efficiecy of R&D icetives i its curret form i Brazil. N. Mulder, S. Motout ad L. Lopes [11] examie the relative labor productivity performaces of Brazil ad Mexico i maufacturig as compared to Uited States, durig the Period. Usig the Iteratioal Comparisos of Output ad Productivity (ICOP), they compare Brazil separately with the U.S. i 1985 ad Their results idicate that labor productivity i Brazil was 43 per cet of the U.S. level i 1985, ad labor productivity i Mexico was 27 percet of the U.S. i The extrapolatio to the period idicates that the productivity gaps betwee the Lati coutries ad the U.S wideed, especially i the 1980s. Although Brazil was able to stabilize its productivity differetial, Mexico cotiued to wide its productivity gap relative to the U.S. 3 The Method of Aalysis I this study, fiacial efficiecy is defied i terms of profitability, debt maagemet, asset maagemet, ad liquidity maagemet. Profitability is measured by retur o assets, retur o equity, ad retur o ivestmet. Debt maagemet is measured by log-term debt to equity ad total debt to equity. Asset maagemet is measured by total asset turover, receivable turover, ad ivetory turover. Liquidity maagemet is measured by quick ratio ad curret ratio. The followig te fiacial ratios are used i the aalysis: 1. ROA: Retur o Assets = Net Icome Asset

4 4 Parviz Asheghia 2. ROE: Retur o Equity = Net Icome Commo Equity 3. ROI : Retur o Ivestmet = 4. LTDE: LT Debt to Equity = 5. TDTE: Total Debt to Equity = Net Icome Ivestmet Log Term Debt L T Debt +Preferred Stock +Commo Stock Total Liabilities Shareholder sequity 6. TATO: Total Asset Turover = Reveue Assets 7. RTO: Receivable Turover = Net Credit Sales Average Accout Receivable 8. ITO: Ivetory Turover = Sales Ivetory 9. QR: Quick Ratio = Cash +Accout Recievable +S T or Marketable Securities Curret Liabilities 10. CR: Curret Ratio = Curret Assets Curret Liabilities 4 The Limitatios of the Study ad the Sample Size The problems of measuremet i comparative aalysis have bee discussed by a umber of researchers (see [8] & [9]). Measuremet problems, as related to firm compariso, ceter aroud two mai issues. First, the choice of idustries, ad secod the kid of data. As for the first issue, ideally the two groups of firms should be similar with regard to product heterogeeity ad size. They should also operate i a similar eviromet ad market structure. However, the limitatios of the sample size i most of the empirical studies call for restrictio i choosig firms for compariso. The secod questio has to do with accoutig differeces that exist betwee coutries. These differeces might lead to biases i the measuremet of the ratios employed i the aalysis. For example, measuremet s problem could arise because the two coutries may use differet procedures for the valuatio of icome producig assets which affects icome statemets. I oe coutry, marketable securities may be treated at the lower of cost or market value (LCM), whereas cost method could be utilized i aother coutry. These differeces might affect comparability of the two coutries with regard to asset turover, ad profitability ratios. Today, oe of the challeges faced by regulatory accoutig agecies globally is how to dimiish or eradicate the asymmetry of iformatio appearig i reports that are prepared accordig to the stadards of various atios. Numerous iteratioal agecies such as the Iteratioal Accoutig Stadards Board (IASB), ad the Iteratioal Federatio of Accoutats (IFAC), have tried to develop highly reliable iteratioal accoutig stadards. I achievig this task, they attempt to foster the mergig of differet local

5 The Relative Fiacial Efficiecy of ad U.S. Firms 5 accoutig stadards ito a iteratioal accoutig stadard. As a result, the iteratioal fiacial reportig stadards (IFRS) are icreasigly acquirig global acceptace. This is evidet by the umber of firms that have adopted iteratioal accoutig stadards i their fiacial statemets. I Jauary 2005, roughly 7,000 compaies listed with the Europea Uio coverted their statemets to the IFRS requiremets [12]. Today, about 113 atios worldwide require or permit IFRS reportig for domestic listed compaies [13]. I the case of our study, Brazil (), o July 13, 2007, the Securities ad Exchage Commissio of Brazil - Comissão de Valores Mobiliários, or CVM required listed compaies to publish their cosolidated fiacial statemets accordig to IFRSs, startig with reportig periods edig i Although the Uited States still adheres to GAAP fiacial reportig, the Securities ad Exchage Commissio voted o August 27th, 2008 to publish for public commet a proposed roadmap that could lead to the use of IFRS by U.S. issuers begiig i Give the aforemetioed limitatios, the fidigs should be iterpreted with cautio. Followig the most acceptable criteria set by empirical studies, each pair of firms chose i this study cosists of oe U.S. firm ad oe firm producig similar products ad havig approximately the same size. All the data for the study were compiled by the author from Merget o Lie. These iclude the most recet five-year time-series data for all the te ratios that were available i Table 1 shows Idustry SIC code, descriptio ad umber of matched pairs that were selected from the maufacturig sector i each coutry. Table 1: The Idustrial Sectors ad the Number of Matched Pairs of Firms Idustry SIC code Descriptio No. of Pairs 2812 Alkalies ad chlorie Idustrial iorgaic chemicals Medicials ad botaicals Pharmaceutical preparatios Paits ad allied products Cyclic crudes ad itermediates Idustrial orgaic chemicals Fertilizers, mixig oly Petroleum refiig Blast furaces ad steel mills Electrometallurgical products Steel pipe ad tubes Gray ad ductile iro foudries Noferrous wiredrawig & isulatig Metal cas Fabricated plate work (boiler shops) Costructio machiery Motorcycles, bicycles, ad parts Telephoe ad telegraph apparatus Motor vehicles ad car bodies Motor vehicle parts ad accessories 2 Total Number of Pairs 23

6 6 Parviz Asheghia 5 The Testig of Hypotheses Paired compariso is used to compare the relative fiacial maagerial efficiecy of U.S. ad firms. Te hypotheses are tested o the basis of ROA, ROE, ROI, LTDE, TDTE, TAT, RTO, ITO, QR, ad CR. I all of these cases the ull hypothesis states that there is o differece betwee U.S. ad firms with regard to the ratio that is beig compared. The alterative hypothesis explais that these ratios are differet. Matched pairs are used, a appropriate test is the Wilcoxo Matched-pairs Siged-raked test. This test is ideal because it is a oparametric test, ot requirig a large sample size. This test gives more weight to pairs that show a large differece tha pairs idicatig small oes. I this maer the Wilcoxo test is similar to the t-test but it deals with ordial data. This test is oe of the most powerful o-parametric tests. Eve for small samples its power is about 95 percet of that of the t-tests (see [14], [15], & [16]). To coduct the Wilcoxo test, first the differeces betwee each pair, with regard to the ratios that are beig compared, are computed. The these differeces are raked o the basis of their absolute values. Next, the sums of the raks of the positive ad egative differeces are used as the test statistics T-, ad T+, respectively. Fially, the appropriate absolute T value i each table is chose as the test statistic ad is compared to the critical value give i the table of critical values for the Wilcoxo test, for the umber of observatios (N). 5.1 Profitability The results of the tests for profitability, measured by ROA, ROE, ad ROI, are show i Tables 2 through 4. The values of the test statistic (T) i these tables idicate that all of the ull hypotheses of similarities betwee ROA (Table 2) ad ROI (Table 4) caot be rejected at the 5% level of sigificace. ROA similarities mea that U.S. firms ad firms are similar i terms of efficiecy by which they use assets to geerate profit. The similarities of ROI meas that the U.S. firms ad Brazilia firms are similar with regard to the efficiecy by which they maage the ivested capital to geerate profit. The ull hypothesis that ROE of U.S. firms ad Brazilia firms does ot differ is rejected at 5% level of sigificace, idicatig that firms have higher ROE ratios as compared to U.S. firms. This meas Brazilia firms are more efficiet tha U.S. firms i terms of geeratig icome for their shareholders. 5.2 Debt Maagemet The results of the tests for debt maagemet, measured by LTDE ad TDTE are show i Tables five ad six, respectively. The values of the test statistic (T) i these tables idicate that all of the ull hypotheses of similarities of betwee the U.S. firms ad Brazilia firms with regard LTDE (Table 5) ad TDTE (Table 6) caot be rejected at the 5% level of sigificace. This implies that U.S. firms ad firms are similar with regard to the efficiecy by which they maage their total debt ad log term debts. 5.3 Asset Maagemet The results of the tests for asset maagemet, measured by TATO, RTO, ITO are show i Tables seve through ie, respectively. The values of the test statistic (T) i these tables idicate that all the ull hypotheses of similarities betwee the U.S. firms ad

7 The Relative Fiacial Efficiecy of ad U.S. Firms 7 Brazilia firms with regard to TATO (Table 7), RTO (Table 8), ad ITO (Table 9) caot be rejected at the 5% level of sigificace. This implies that U.S. firms ad Brazilia firms are similar with regard to the efficiecy by which they maage their assets. 5.4 Liquidity Maagemet The results of the tests for liquidity maagemet, measured by QR ad CR are show i Tables te ad eleve, respectively. The values of the test statistic (T) i these tables idicate that all the ull hypotheses of similarities betwee the U.S. firms ad Brazilia firms with regard to QR (Table 10), ad CR (Table 11) caot be rejected at the 5% level of sigificace. This implies that U.S. firms ad Brazilia firms are similar with regard to the efficiecy by which they maage their liquidities. 6 Cocludig Remarks This study evaluates the relative fiacial efficiecy of U.S. firms, ad Brazilia firms, usig a sample of five-year time-series data for a set of 23 matched firms that are chose from seve idustries. To the extet that the data are ot biased i the cotext of the limitatios set i this study, the foregoig aalysis suggests although Brazilia firms are more efficiet tha the U.S. firms i terms of ROE, they are similar i terms of the other two profitability ratios of ROA ad ROI. The results of the aalysis of this study also idicate that there are o sigificat differeces betwee the U.S. firms ad Brazilia firms with regard to their efficiecy by which they maage their debts, assets, ad liqidity. I a earlier study Liu ad O Farrell examie the potetial differeces i fiacial ratios betwee Brazilia ad U.S. compaies with 60 matched-pairs of compaies from three major maufacturig idustries that are most attractive to U.S. ivestors for the year They coclude that Brazilia firms have lower log term debt to total capital, ad asset turover ratios, but higher duratio of payables. This study differs from their study because each matched pair examied i their study is based o a oe-year data i 2006, but this study employs all the five-year data that were available i This study examied the comparative efficiecies of the ad U.S. firms i terms of profitability, a debt maagemet asset maagemet, ad liquidity, portrayig the relative effectiveess of the executives of these firms i maagig their fiaces. It would be helpful to compare productio efficiecies of the firms with the U.S. firms to see if firms are as efficiet as their couterparts i the U.S. i maagig their productio process. This would require the measuremet of total factor productivity, capital productivity ad labor productivity ad could be the subject of further research i this area.

8 8 Parviz Asheghia Table 2: The Wilcoxo Test for the Compariso of ROA Ratios of U.S. ad Firms ROA ki ROA ki TOTAL T+=143 T-=93 ROA ki = ROA of the kth U.S. firm i the ith yea; i = 1...5; k =...23 ROA ki = ROA of the kth firm i the ith year; i = 1...5; k =...23 U.S. RD = Rak of DK; DK = ROA ki - ROA ki ; PRS = Positive rak sum; NRS = Negative rak sum.

9 The Relative Fiacial Efficiecy of ad U.S. Firms 9 Table 3: The Wilcoxo Test for the Compariso of ROE Ratios of U.S. ad Firms ROE ki ROE ki 1 * * * * * * TOTAL T+=125 T- =106 *Data ot available ROE ki = ROE of the kth U.S. firm i the ith year; i = 1...5; k = ===== ROE ki = ROE of the kth firm i the ith year; i = 1...5; k = U.S. RD = Rak of DK; DK = ROE ki - ROE ki ; PRS = Positive rak sum; NRS = Negative rak sum

10 10 Parviz Asheghia Table 4: The Wilcoxo Test for the Compariso of ROI Ratios of U.S. ad Firms ROI ki ROI ki 1 * -6.86* * * * * * TOTAL T+=209 T-=0 *Data ot available ROI U.S. ki = ROI of the kth U.S. firm i the ith year; i = 1...5; k = ROI ki = ROI of the kth firm i the ith year; i = 1...5; k = U.S. RD = Rak of DK; DK = ROI ki - ROI ki ; PRS = Positive rak sum; NRS = Negative rak sum.

11 The Relative Fiacial Efficiecy of ad U.S. Firms 11 Table 5: The Wilcoxo Test for the Compariso of LTDTE Ratios of U.S. ad Firms LTDTE ki LTDTE ki 1 * * 6 * * * * * * TOTAL T+=30 T-=90 *Data ot available LTDTE U.S. ki = LTDTE of the kth U.S. firm i the ith year; i = 1...5; k = LTDTE ki = ROE of the kth firm i the ith year; i = 1...5; k = U.S. RD = Rak of DK; DK = LTDTE ki - LTDTE ki ; PRS = Positive rak sum; NRS = Negative rak sum.

12 12 Parviz Asheghia Table 6: The Wilcoxo Test for the Compariso of TDTE Ratios of U.S. ad Firms TDTE ki TDTE ki 1 * * 6 * * * * * * TOTAL T+=26 T-=71 *Data ot available LTDTE U.S. ki = LTDTE of the kth U.S. firm i the ith year; i = 1...5; k = LTDTE ki = ROE of the kth firm i the ith year; i = 1...5; k = U.S. RD = Rak of DK; DK = LTDTE ki - LTDTE ki ; PRS = Positive rak sum; NRS = Negative rak sum.

13 The Relative Fiacial Efficiecy of ad U.S. Firms 13 Table 7: The Wilcoxo Test for the Compariso of TATO Ratios of U.S. ad Firms TATO ki TATO ki * TOTAL T+=101 T-=133 *Data ot available TATO U.S. ki = TATO of the kth U.S. firm i the ith year; i = 1...5; k = TATO ki = ROE of the kth firm i the ith year; i = 1...5; k = U.S. RD = Rak of DK; DK = TATO ki - TATO ki ; PRS = Positive rak sum; NRS = Negative rak sum.

14 14 Parviz Asheghia Table 8: The Wilcoxo Test for the Compariso of RTO Ratios of U.S. ad Firms RTO ki RTO ki * * TOTAL T+=93 T-=161 *Data ot available RTO U.S. ki = RTO of the kth U.S. firm i the ith year; i = 1...5; k = RTO ki = ROE of the kth firm i the ith year; i = 1...5; k = RD = Rak of DK; DK = RTO - RTO ki ; PRS = Positive rak sum; NRS = Negative rak sum.

15 The Relative Fiacial Efficiecy of ad U.S. Firms 15 Table 9: The Wilcoxo Test for the Compariso of ITO Ratios of U.S. ad Firms ITO ki ITO ki * * * TOTAL T+=53 T-=157 *Data ot available ITO U.S. ki = ITO of the kth U.S. firm i the ith year; i= 1...5; k = ITO ki = ROE of the kth firm i the ith year; i= 1...5; k = RD = Rak of DK; DK = ITO - ITO ki ; PRS = Positive rak sum; NRS = Negative rak sum.

16 16 Parviz Asheghia Table 10: The Wilcoxo Test for the Compariso of QR Ratios of U.S. ad Firms QR ki QR ki * * * * * * 22 * * 2.21 TOTAL T+=61 T-=75 *Data ot available QR U.S. ki = QR of the kth U.S. firm i the ith year ; i = 1...5; k = QR ki = ROE of the kth firm i the ith year; i = 1...5; k = RD = Rak of DK; DK = QR - QR ki ; PRS = Positive rak sum; NRS = Negative rak sum.

17 The Relative Fiacial Efficiecy of ad U.S. Firms 17 Table 11: The Wilcoxo Test for the Compariso of CR Ratios of U.S. ad Firms CR ki CR ki TOTAL T+=1331 T-=143 *Data ot available 0 CR U.S. ki = CR of the kth U.S. firm i the ith year; i = 1...5; k = CR ki = ROE of the kth firm i the ith year; i = 1...5; k = RD = Rak of DK; DK = CR - CR ki ; PRS = Positive rak sum; NRS = Negative rak sum.

18 18 Parviz Asheghia Refereces [1] World Bak, Brazil: Overview, The World Bak, Retrieved o 04/01/2013. [2] Wikipedia, Brazil, Retrieved o 04/20/2013 [3] World Bak, Ibid. [4] OECD: Brazil-Ecoomic Forecast Summary (November 2012), OECD, Retrieved o 09/05/2012. [5] P. Tecles ad B. M.Tabak Determiats of Bak Efficiecy: the Case of Brazil, No 210, Workig Papers, Series, Cetral Bak of Brazil, Research Departmet. [6] T. Kida, Foreig owership, sales to multiatioals ad firm efficiecy: the case of Brazil, Morocco, Pakista, South Africa ad Vietam, Applied Ecoomics Letters, 2012, vol. 19, issue 6, pages [7] N. Nazmi ad J. E. Revilla, Ecoomic Efficiecy ad Growth: Evidece From Brazil, Chia, ad Idia, No RP2008/86, Workig Papers from World Istitute for Developmet, Ecoomic Research (UNU-WIDER). [8] O. Yeboah, V. Ofori-Boadu ad T. Li, Chia-U.S. Potetial No-food Ethaol Exportatio No 56469, 2010 Aual Meetig, February 6-9, 2010, Orlado, Florida from Souther Agricultural Ecoomics Associatio. [9] E. P. Ribeiro, V. Prochik ad J. DeNegri, Productivity ad Growth i the Brazilia ad Iformatics Idustry , Workig Papers from Uiversidade de São Paulo, Faculdade de Ecoomia, Admiistração e Cotabilidade de Ribeirão Preto. [10] N. Mulder, S. Motout ad L. Lopes, Brazil ad Mexico's Maufacturig Performace i Iteratioal Perspective, Workig Papers from CEPII research ceter. [11] Ilse Maria Beure ad Roberto Carlos Kla, Divergeces betwee the GAAP ad GAAP, Joural of Accoutig ad Taxatio Vol. 2(2), pp , August [12] Ibid. [13] P. Asheghia ad W. Foote, The Productivities of U.S. Multiatioals i the Idustrial Sector of the Caadia Ecoomy, Easter Ecoomic Joural, (April-Jue, 1985), [14] W. Medehall, J.T. McClave ad M. Ramey, Statistics for Psychology, 2 d ed., Massachusetts, Dubury Press, pp , [15] S. Sigel, S. Sigel, Noparametric Statistics for the Behavioral Scieces, New York, McGraw Hill, 1956.