Media portrayals of regions in the Czech Republic: selected issues

Ma ke ing and T ade 125DOI: 10.15240/ ul/001/2014-4-010 4, XVII, 2014 In oduc ion No su p isingly, 21s cen u y is qui e equen ly e e ed o as he cen u y o in o ma ion. In o ma ional needs cu en ly ep esen such an impo an ca ego y ha hey a e compa able wi h he sa u a ion o physiological needs. Less a en ion is usually de o ed o he e i o ial pa e ns o in o ma ion. E e y e i o y can be cha ac e ized by bo h ma e ial and in angible cha ac e is ics [1], [20]. Spa ial dis ibu ion o popula ion is o u mos impo ance in his espec as human ac i i ies o m ma e ial s uc u es, which su ound us. None heless, all e i o ies a e desc ibable also om in angible pe spec i e. O e all a mosphe e, epu a ion, image as well as posi ion on he men al map can be anked among ypical a ibu es o he in angible dimension o he space [9], [16], [19]. Men al maps ep esen inc easingly popula no ion o ou imes. Men al maps a e o med on he basis o agg ega ion o indi idual images o pa icula spo s and e i o ies. F om his s andpoin , he image o he e i o y should be pe cei ed as o e all pic u e and epu a ion o gi en a ea in he eyes o he wide public. Men al maps e lec he psychological ep esen a ion o space. [5] showed ha men al maps a e no me e p e e ence su aces bu also p edic o s o consequen spa ial beha io . G owing impo ance o men al maps is in compliance wi h ecen upsu ge o so loca ion ac o s. Media – and no ma e whe he p in ed, such as newspape s o jou nals, o elec onic, such as TV, adio o on-line con en – a e undoub edly he p ima y sou ce o in o ma ion. A he same ime, media o med he en i onmen ha di ec s ou e e yday ac i i ies. Media in luence he unc ioning o indi idual subjec s as well as whole e i o ies. In b ie , indi iduals o en e p ises ha do no appea in media a e losing hei exis ence in i ual wo ld and become inconspicuous in he eal wo ld as well. And he same applies o egions. Regions wi hou media publici y a e la gely omi ed by in es o s, ou is s as well as wide public. I he egion appea s in he media i s ill does no mean any ic o y since in es o s, ou is s and he wide public a e p e y sensi i e on he composi ion o news abou indi idual e i o ies. I he egion is depic ed in a posi i e way, i o en lu es he abo e men ioned a ge g oups; ice e sa, nega i e news epel hese a ge g oups and he egion can easily all in o psychological and economic icious ci cle. Pu succinc ly, media in ela ion o egional de elopmen do ma e [4], [8], [10]. Media in luence eade s, lis ene s o TV iewe s in he sho un. In case ha ce ain hemes a e epea ed mo e equen ly, a i udes and beha io o he popula ion can be a ec ed in he longe un. Fu he ac o s shaping men al maps in he longe un a e educa ional ins i u ions, amily o su ounding milieu. Image o he egion should cha ac e ize i s speci ic supply and dis inc i eness. In eali y, e i o ial image e y o en does no co espond o he objec i e eali y, which is caused p ima ily by he p ocesses o agenda-se ing and agenda-cu ing ha a e used in a ious kinds o media a he equen ly. The adi ional mission o media is objec i e, accu a e and unbiased depic ion o eali y. Howe e , con empo a y media can be al eady pe cei ed as co-c ea o s o he milieu we a e li ing in. Rega ding he g ea and s eadily g owing in e connec edness o he wo ld, media can show jus selec i e cu s o complex eali y. People a e hus in o med abou he e en s bu a he same ime media mo e o less conspicuously di ec hei beha io and he pe cep ion o MEDIA PORTRAYALS OF REGIONS IN THE CZECH REPUBLIC: SELECTED ISSUES Jan Sucháãek, Pe Seìa, Václa F ied ich, Ja osla Kou sk˘ EM_04_14_zlom 1.12.2014 11:13 S ánka 125 Ma ke ing a obchod 126 2014, XVII, 4 eali y. Many people e en accep opinions p esen ed by media in a e y sophis ica ed way. Agenda-se ing, i.e. accen ua ion o chosen e en s o agenda-cu ing, i.e. igno ance o selec ed occu ences play inc easingly impo an ole in his con ex [11], [12]. A he same ime, he possibili ies o con ol whe he media ul ill hei mission, i.e. objec i e and unbiased co e age a e p ac ically elimina ed, since he e a e no so many indi iduals ha possess su icien ime and ene gy o check he p ima y sou ces o in o ma ion and o e i y he alidi y o media in o ming. Mo eo e , he quali y o in o ma ion i sel is hea ily dependen upon he way o in e p e a ion and explana ion o e en s o phenomena. Indeed, media bias is one o he mos impo an a ibu es o con empo a y co e age. F om his poin o iew, he concep ion o ga ekeeping is o g ea impo ance. Ga ekeeping should be ea ed as he p ocess h ough which in o ma ion is il e ed o dissemina ion, be i publica ion, b oadcas ing, in e ne o some o he mode o communica ion. Ga ekeeping anges om a epo e deciding which sou ces a e selec ed o be included in as o y o edi o s de e mining which s o ies a e p in ed o co e ed. The p ocess conce ns also media owne s as well as ad e ise s. P ocess o ga ekeeping designs no only which in o ma ion is chosen, bu also wha he con en and na u e o he messages, such as news, will be. The mos impo an a ibu es o ga ekeeping a e as ollows [2], [15]: In accomplishing hei moni o ing unc ions, all media ha e a e y la ge amoun o s o ies b ough e e y day by epo e s. Only a limi ed amoun o ime o space is a ailable in any medium o i s daily p esen a ions o he news o i s audience. The emaining oom is de o ed p ima ily o ad e ising. Wi hin any news o ganiza ion he e exis s anews pe spec i e, a subcul u e sui gene is ha includes a complex se o c i e ia o e alua ing a pa icula news s o y – c i e ia based on economic needs o he medium, o ganiza ional policies and some o he ca ego ies. The abo e c i e ia a e used by edi o s, news di ec o s, and o he pe sonnel who selec a limi ed numbe o news s o ies o p esen a ion o he public. These news a e p esen ed in ways such ha he equi e- men s o he medium and he p e e ences o he audience a e me . Pe sonnel in he media is ac ing as ga ekeepe s, le ing some s o ies pass h ough he sys em bu keeping o he s ou , hus limi ing, con olling, and shaping he public's knowledge o he en i e y o ac ual e en s aking place in eali y. F om e i o ial poin o iew, media cons i u e ce ain in o ma ional ga e be ween inne , i.e. municipal and egional ac o s o e i o ial de elopmen and ou e g oups, such as isi o s, po en ial isi o s, in es o s, non- egional en ep eneu s e c. Local/ egional poli icians, ins i u ions o en ep eneu ial subjec s s i e o he a ac ion o media a en ion and add ess hei o e s, ci izens o employees jus by means o media. And ice e sa, hese ele an playe s o local and egional de elopmen ge he impo an in o ma ion o hei decision- making h ough media again. No su p isingly, media by hemsel es became ele an ac o s o e i o ial de elopmen . Subsequen ly, we a e cu en ly en i led o speak abou media- e i- o ial o di ec ly medi o ial de elopmen , a he han abou me e e i o ial de elopmen [17]. We can dis inguish local, egional and na ional media. Basically, i is scala di e en ia ion o media. Ob iously, na ional media a e he mos in luen ial ones. Fo he pu poses o ou wo k, we will ocus p ima ily upon na ional media, because hey in o m no only abou na ional o o eign issues bu also abou pa icula egions. And jus he news conce ning egions ha a e men ioned a he na ional le el subs an ially co- de e mine he c ea ion o men al maps. In many coun ies, spa ial dis ibu ion o media is a he cen alized and all p incipal media a e concen a ed usually in o he capi al ci y, which is also he case o Czech Republic. A he same ime, one has o conside ha abo e media cen es ypically ac also as adminis a i e, poli ical and socio-cul u al hubs. This is he ac o ha may a ec e i o ial pa e n and composi ion o he news. Howe e , he e a e much mo e sou ces o media bias. In o de o cap u e he ela ion be ween media and egions, au ho s o his pape ocused on TV. TV co e age has one o he highes impac s on he public on he one hand and is ep e- sen a i e enough on he o he [14]. Compa isons show ha TV co e age agenda o a la ge EM_04_14_zlom 1.12.2014 11:13 S ánka 126 Ma ke ing and T ade 1274, XVII, 2014 ex en ep esen s also p ess o adio co e age agenda [14]. Tha is why TV co e age will be aken as a poin o depa u e o his esea ch. The main objec i e o his a icle is o analyze and e alua e selec ed aspec s o NUTS III egions po ayals wi hin na ional TV co e age. Whole esea ch has been accomplished p ima ily om quali a i e poin o iew. Added alue o he pape is subs an ially in luenced by he ela i e newness o he whole opic. 1. Ma e ials and Me hods The empi ical analysis in his pape is based on unique da a pu chased om Media Teno , L d. This company deals wi h a con inuous and sys ema ic analysis o media epo s in he Czech Republic. The Media Teno , L d. o e s a com- ple e se o analy ical se ices based on he con en analysis o media. I accu a ely e alua es bo h quali a i e and quan i a i e aspec s o media con ibu ions. Analy ical and consul ing se ices a e usually used by p i a e companies, poli ical pa ies, PR agencies, local go e nmen s, schola s as well as media edi o s. In his pape , we in es iga ed TV co e age which was ep esen ed by e ening news o 3 p incipal TV companies in he Czech Republic. Ou esea ch he e o e moni o ed Událos i and Událos i, komen áfie o Czech TV, bo h o hem ep esen ing public TV. On he o he hand, Zp á y cons i u es pa o p i a e FTV P ima ai ime and he same applies o Tele izní no iny o TV No a. Ou esea ch co e s he pe iod be ween 2004 and 2011 yea s, which is maximum possible ime in his ealm. Thus one can al eady ge a leas an essen ial o e iew o de e- lopmen al endencies in he esea ched domain. Ou analysis and in e p e a ion is based on con en s uc u e o con ibu ions appea ing in na ional TVs and ela ed o indi idual NUTS III egions in he Czech Republic. Indi idual hemes ela ed o coun y’s sel -go e ning egions include ele an economic, social as well as en i on- men al aspec s o li e in hese e i o ies. Fo he pu poses o his ex , h ee basic esea ch hypo heses ha e been o mula ed. The hypo heses es ed in ou pape a e de ined as ollows: H1: The a e age numbe o con ibu ions in each hema ic pilla is compa able wi hin indi idual egions. H2: The ep esen a ion o con ibu ions om apa icula pilla is compa able among egions. H3: Independence be ween egion and pilla a iables can be expec ed. In his pape , he me hods o co esponden- ce analysis and K uskal-Wallis es p ocedu e will be u ilised. The K uskal-Wallis es o non- pa ame ic analysis o a iance (ANOVA) is usually used when one wan s o es he di e- ence be ween wo o mo e g oups when da a sample doesn’ mee assump ion o no mali y. Co espondence analysis is app op ia e app oach when a emp ing o de e mine he p oximal ela ionships among wo o mo e ca ego ical a iables. When ha ing ca ego ical a iables he me hod o co espondence analysis is simila o he usage o co ela ion analysis o p incipal componen s analysis o con inuous o nea ly con inuous a iables. 1.1 Analysis o Va iance The i s analy ical app oach, which will be applied in his pape , is he analysis o a iance, namely he K uskal-Wallis es . The K uskal-Wallis es is usually used when he e is one nominal a iable and one measu emen a iable, and he measu emen a iable does no mee an assump ion o no mali y. In ha case, a s an- da d one-way ANOVA may lead o inaccu a e es ima es o he p- alue when he da a a e e y a om no mal dis ibu ion. The K uskal-Wallis es belongs o a g oup o nonpa ame ic es s and is equi alen o he s anda d one-way ANOVA. This es is a logical ex ension o he Mann-Whi ney U es and allows us he compa ison o mo e han wo independen g oups. I is mos commonly used when we wish o compa e h ee o mo e se s o sco es ha come om di e en g oups [13]. The K uskal-Wallis es does no expec an assump ion ha he da a a e no mally dis ibu ed. Howe e , i does assume ha he obse a ions in each g oup come om popula ions wi h he same shape o dis ibu ion. In o he wo ds, i di e en g oups ha e di e en shapes, o ins ance one is skewed o he le and ano he one is skewed o he igh , o hey ha e di e en a iances, he K uskal-Wallis es may gi e inaccu a e esul s. Fo de ails, see [3]. In ha kind o si ua ion, maybe one could look in o some kind o boo s ap analysis. The null hypo hesis in case o he K uskal- Wallis es is ha he samples come om popula ions wi h ce ain cha ac e is ics. These cha ac e is ics en ail he e is 0.5 p obabili y ha EM_04_14_zlom 1.12.2014 11:13 S ánka 127 Ma ke ing a obchod 128 2014, XVII, 4 a andom obse a ion om one g oup is g ea e han a andom obse a ion om ano he g oup. The K uskal-Wallis es does no es he null hypo hesis ha he popula ions ha e iden ical means, which is he null hypo hesis in case o one-way ANOVA. The K uskal-Wallis es also does no es he null hypo hesis ha he popula ions ha e equal medians, al hough you will see his e o in some s a is ics ex books. Like mos non-pa ame ic es s, he K uskal-Wallis es is also pe o med on anked da a, so ha obse a ions a e con e ed o hei anks in he o e all da a se , and he smalles alue ge s a ank o 1, he nex smalles ge s a ank o 2, e c. Tied obse a ions ge a e age anks; he e o e i he e we e wo iden ical alues occupying he se en h and eigh h smalles places, all would ge a ank o 7.5 e c. The sum o he anks is calcula ed o each g oup, hen he es s a is ic H is calcula ed. His gi en by a a he o midable o mula ha basically ep esen s he a iance o he anks among g oups, wi h an adjus men o he numbe o ies. I he null hypo hesis is ue, he es s a is ic H is app oxima ely chi-squa e dis ibu ed, which means ha he p obabili y o ge ing a pa icula alue o H by chance is he p- alue co esponding o a chi-squa e equal o H. The deg ees o eedom ep esen he numbe o g oups minus 1. Analysis o a iance will be u ilised o he e i ica ion o hypo heses H1 and H2. 1.2 Measu emen o Co espondence in a C oss abula ion Co espondence analysis is a mul i a ia e s a is ical echnique p oposed by [7] and la e de eloped by [6]. The co espondence analysis deno es a nonpa ame ic app oach which does no calcula e any s a is ical signi icance es since i is no based on a p obabili y dis ibu ion assump ion. The co espondence analysis i sel ep esen s an algo i hm which is capable o some measu- emen o co espondence be ween he ows and he columns in a co espondence able. When cen ing he columns and ows o co espon- dence able and using chi-squa e dis ances i co esponds o s anda d co espondence analysis. Ne e heless, using al e na i e cen ing ways combined wi h ano he ype o dis ances allows o an al e na i e ep esen a ion o co - espondence able in a low-dimensional space. As he measu e o co espondence any associa ion, simila i y, con usion o in e ac ion be ween he column and ow a iables can be conside ed. Common ype o co espondence able is a c oss abula ion wi h he cells con aining equency coun s. Howe e , when using a ypical c oss abula ion i does no usually p o ide aclea pic u e o he na u e o co espondence be ween a iables, especially i he a iables o in e es a e nominal and in addi ion con ain nume ous ca ego ies. To summa ize, co espondence analysis is usually pe o med on con ingency ables o size m × nwhe e m ep esen s he numbe o ows and ndeno es he numbe o columns in acon ingency able. F om s a is ical poin o iew, ac oss abula ion is a ype o able in a ma ix o ma ha displays he mul i a ia e equency dis ibu ion o he a iables. A c ucial p oblem o mul i a ia e s a is ics is inding dependence s uc u e unde lying he a iables con ained in high dimensional con ingency ables. When using co espondence analysis app oach, a con ingency able o equencies is usually s anda dized i s ly. Thus, he ela i e equencies ac oss all cells sum o 1.0. In ypical co espondence analysis he en ies in he able ep esen he ela i e equencies in e ms o he dis ances be ween indi idual columns and ows in a low-dimensional space. When applying co espondence analysis p ocedu e, all used da a should be nonnega i e and mo eo e on he same scale. The me hod o co espondence analysis deals wi h ows and columns equi alen ly. The con ingency ables o e se e al measu es o associa ion and es s o associa ion. Fo ins ance, chi-squa e es , Fishe 's exac es , Phi coe icien , con ingence coe icien C o C a- me ’s V can be used. An analysis o con in- gency ables usually includes in es iga ion o ow and column p o iles and es ing o independence ia he chi-squa e es . The co espondence analysis decomposes he χ2 s a is ic associa ed wi h his able in o o hogonal ac o s. A chi-squa ed es , also e e ed o χ2 es , is s a is ical hypo hesis es in which he sampling dis ibu ion o he es s a is ic is a chi-squa ed dis ibu ion when he null hypo hesis is ue. Since co espondence analysis is only desc ip i e echnique, i can be applied o ables whe he he χ2s a is ic is app op ia e o no . When he numbe o EM_04_14_zlom 1.12.2014 11:13 S ánka 128 Ma ke ing and T ade 1294, XVII, 2014 ca ego ies is qui e la ge and he assump ions o chi-squa e es a e no ul illed, he chi- squa e es some imes is no able o de ec he dependence s uc u e clea ly. The co espondence analysis is able o examine he ela ionship be ween wo a iables g aphically in mul idimensional space. One can compu e column and ow sco es and make aplo based on he sco es. Thus, he basic aim o co espondence analysis is o c ea e biplo s o co espondence ables. The co espondence plo allows us o isualize mu ual ela ionships among ca ego ies spa ially on dimensional axes. In o he wo ds, wi h a help o empi ically de i ed dimensions i can be seen which ca ego ies a e close o o he ca ego ies. The e o e, p ojec ing poin s o one a iable on he ec o om he o igin o a ca ego y poin o he o he a iable ep esen s he ela ionship be ween analysed a iables. To sum i up, wi h a help o his app oach one can see which ca ego ies o used a iable a e simila o each o he o /and which ca e- go ies o wo di e en a iables a e ela ed. Mo eo e , co espondence analysis p ocedu e allows us o i supplemen a y poin s in o a space which has been de ined by ac i e poin s. Co espondence analysis is able o wo k wi h nominal a iables, o dinal a iables, and also wi h a io a iables, albei c ea ion o disc e e ca ego ies om a con inuous a iable is gene ally no ecommended. Since in a ypical co espondence able, a iables in he ows and columns a e assumed o deno e uno de ed ca ego ies, he nominal scaling is usually used. I means ha he only conside a ion is he ac ha some objec s a e in he same ca ego y while o he s a e no . No hing can be assumed abou he dis ance o o de be ween ca ego ies o he same a iable. Co espondence analysis will be used o he e alua ion o hypo hesis H3. 2. Resul s and Discussion S a is ical analyzes p esen ed in his pape we e accomplished on he da a ile ep esen ing he indi idual con ibu ions p esen ed in he majo news p og ams om di e en NUTS III egions in he public Czech Tele ision (Událos i and Událos i, komen áfie), comme cial TV No a (Tele izní no iny) and FTV P ima (Zp á y) o he pe iod om Ap il 2004 o Decembe 2011. Con ibu ions a e hema ically di ided in o 30 ca ego ies and 10 pilla s (social, en i onmen al, economic, educa ion and science, public a ai s, Eu opean and in e na ional, secu i y, acciden s, spo and jus ice). Al oge he he amoun o 54,667 con ibu ions was moni o ed by his way. S a is ical da a we e analysed by means o con ingency ables, co espondence analysis and non-pa ame ic K uskal-Wallis es wi h a help o Mic oso Excel and IBM SPSS S a is ics so wa e. 2.1 Desc ip ion o Da a Se As a i s s ep o ou analysis we di ided de aul 30 hema ic ca ego ies o o iginal da a se s ob- Tab. 1: Dis ibu ion o pa icula hema ic pilla s F equency Pe cen SECURITY 13,630 24.9 SOCIAL 12,826 23.5 ACCIDENTS 10,213 18.7 ECONOMIC 5,731 10.5 PUBLIC AFFAIRS 4,030 7.4 ENVIRONMENTAL 3,213 5.9 JUSTICE 2,147 3.9 EUROPEAN & INTERNATIONAL 1,106 2.0 EDUCATION & SCIENCE 983 1.8 SPORT 788 1.4 To al 54,667 100.0 Sou ce: au ho s EM_04_14_zlom 1.12.2014 11:13 S ánka 129 Ma ke ing a obchod 130 2014, XVII, 4 ained by Media Teno L d. in o 10 basic hema ic pilla s in o de o simpli y hei classi ica ion. The desc ip ion o o iginal da a and de aul 30 ca ego ies can be ound in [17] and [18]. Be o e he analyses he sou ce da a we e ans o med om Excel ables in o uni o m da ashee o IBM SPSS S a is ics so wa e and ca ego ized by au ho s. Tab. 1 shows he dis ibu ion o all he news among pa icula pilla s. Almos one hal o he news (49.5%) is o abloid cha ac e , which includes secu i y, acciden s and jus ice. This is in line wi h ga ekeeping concep ion ha has much o do wi h edi o ial s e eo ypes (see also [15]). As i can be seen in Table 1, secu i y and acciden pilla s oge he wi h social pilla occupy he op o he ankings. Eu opean & in e na ional and educa- ion & science pilla s ha e emained a he opposi e end o he ladde oge he wi h spo pilla . In o de o desc ibe mu ual in e dependence among hema ic pilla s and news p og ams con ingency able was c ea ed, see Tab. 2. Table 2 p o ides us wi h syn he ic in o ma ion abou pe cen age dis ibu ion o indi idual hema ic pilla s wi hin esea ched news p og ams. Tab. 2: Con ingency able (% wi hin news p og am) News P og am Tele izní Událos i Událos i, Zp á y To al no iny (CT) komen áfie (FTV P ima) (No a TV) (CT) SECURITY 30.6% 19.3% 17.2% 26.8% 24.9% SOCIAL 21.1% 28.3% 26.7% 20.2% 23.5% ACCIDENTS 20.0% 14.8% 15.0% 22.4% 18.7% ECONOMIC 9.0% 11.3% 14.9% 9.6% 10.5% PUBLIC AFFAIRS 5.2% 10.3% 10.0% 6.1% 7.4% ENVIRONMENTAL 7.1% 4.6% 3.6% 6.5% 5.9% JUSTICE 3.2% 4.5% 5.5% 3.6% 3.9% EUROPEAN & I NTERNATIONAL 1.3% 2.9% 3.2% 1.5% 2.0% EDUCATION & SCIENCE 1.2% 2.1% 2.6% 1.9% 1.8% SPORT 1.3% 1.8% 1.2% 1.4% 1.4% To al 100% 100% 100% 100% 100% (18,558) (14,997) (6,139) (14,973) (54,667) Sou ce: au ho s While he public Czech TV has he la ges sha e o con ibu ions om social pilla , ollowed by secu i y and acciden s pilla s, bo h comme cial ele isions (No a TV, FTV P ima) ha e secu i y pilla on he op, ollowed by ela i ely compa able p opo ion o social and acciden s pilla s. The economic and public a ai s pilla s ha e a g ea e ep esen a ion in news p og ams o Czech TV. All TV s a ions ha e he smalles sha e o Eu opean & in e na ional, spo and educa ion & science pilla s. As a nex s ep o ou analysis we measu ed po en ial dependence in Tab. 2. Resul s a e shown in Tab. 3. Pea son Chi-squa e es o independence shows ha signi ican dependency o pilla s uc u e on TV p og am exis s (Sig. = 0.000). The alue o C ame V coe icien o con in- gency means he weak dependency only. The maximal numbe o con ibu ions is concen a ed a Tele izní no iny o No a TV; sligh ly less con ibu ions is b ough by Událos i EM_04_14_zlom 1.12.2014 11:13 S ánka 130 Ma ke ing and T ade 1314, XVII, 2014 (Czech TV) and Zp á y (FTV P ima). The news p og am Událos i, komen áfie (Czech TV) con ains signi ican ly smalle numbe o con ibu ions, bu his p og am is o ien ed in o impo an and se ious opics wi h mo e de ailed analyses. In o de o p esen desc ibed in e dependence among hema ic pilla s and news p og ams om isual pe spec i e, he co espondence analysis con aining 2 dimensions was used in he o m o sca e plo , see Fig. 1. The ho izon al axis in Fig. 1 (Dimension 1) di ides pa icula pilla s in o 3 basic segmen s: a) Eu opean & in e na ional, public a ai s (public opics), b) Social, spo , jus ice, economic, educa ion & science (socie al opics), c) Secu i y, en i onmen al, acciden s ( abloid opics). On he o he hand, he e ical axis in Fig. 1 (Dimension 2) subdi ides all pilla s in o 2 segmen s: a) Educa ion & science, economics, jus ice, acciden s (e e yday opics), b) Eu opean & in e na ional, public a ai s, social, spo , en i onmen al, secu i y ( emaining opics). Tab. 3: Measu es o dependence in con ingency able Value Asymp o ic Signi icance Pea son Chi-Squa e 2,219.098 0.000 C ame ’s V 0.116 No alid cases 54,667 Sou ce: au ho s Fig. 1: Sca e plo o pilla s Sou ce: au ho s EM_04_14_zlom 1.12.2014 11:13 S ánka 131 Ma ke ing a obchod 132 2014, XVII, 4 In o de o cla i y he p o iles o abo e men ioned TV p og ams in a be e way, we can display bo h pilla s and TV p og ams h ough he co espondence analysis in o one loca ion map, see Fig. 2. Two dimensions o he co espon- dence analysis a e co e ing app oxima ely 98% o he a iabili y o he o iginal c oss able, which de ec s a e y high a e o co espondence. Fig. 2: Sca e plo o bo h pilla s and p og ams Sou ce: au ho s Wi h he help o Fig. 2 one can de e mine which hema ic pilla s a e mo e accen ua ed han he o he s in pa icula news p og ams. The i s dimension (Dimension 1) di ides he public news p og ams (on he le side o he plo ) om he comme cial ones (on he igh ). The second dimension (Dimension 2) acili a es he di e en ia ion o TV news as o hema ic ca ego ies. The con en di e ences be ween he p og ams o public Czech TV a e la ge han he di e ences be ween he news p og ams o bo h comme cial TV s a ions. I can be said ha Událos i, komen áfie (Czech TV) and Zp á y (FTV P ima) a e o ien a ed in o mo e speci ic and p o ound opics. The dimensions in Fig. 2 can be desc ibed as ollows: a) Dimension 1: public – comme cial ( abloid), b) Dimension 2: gene al (socie al) – speci ic ( amilia /indi idual). F om his pe spec i e i is clea ha Událos i o Czech TV p og am concen a es p ima ily on social, public a ai s, Eu opean & in e na ional, spo pilla s, while Událos i, komen áfie a e ocusing on economics, jus ice, educa ion & science pilla s, which is dis inc i e in compa ison wi h o he TV p og ams. I is possible o claim ha Czech TV is able o p oduce wo di e en news p og ams wi h a ious specializa ions and mo e dis inc p o ile. On he EM_04_14_zlom 1.12.2014 11:13 S ánka 132 Ma ke ing and T ade 1334, XVII, 2014 o he hand Tele izní No iny o TV No a is mo e ocused on en i onmen al and secu i y pilla s while o he Zp á y o FTV P ima is closely linked o acciden s pilla . The las esul is consis en wi h he ac ha FTV P ima p oduces a special pa o news called C ime Repo (K imi Zp á y). Resul s we ob ained a e in consonance wi h p e ious analyses [18]. 2.2 Appea ance o News in Va ious Regions and Pilla s In he nex s ep o ou empi ical analysis all he hypo heses as o mula ed in chap e 2 will be e i ied. Fo he pu poses o ou analysis he de aul da a we e ans o med in o he s uc u e wi h he ollowing ac o s:  egion (co e ing NUTS III e i o ies), yea ,  hema ic pilla . The numbe o news pe yea , pe NUTS III egion and pe pa icula pilla se es as quan i a i e indica o s. I should be also men ioned ha wo NUTS III egions – P ague and Cen al Bohemia we e joined in o one e i o y. I was done due o he be e co espondence wi h na u al geog aphical cha ac e is ics on he one hand and supp ession o speci ic u ban cha ac e o he capi al ci y on he o he . Las bu no leas , uni ica ion o P ague and Cen al Bohemia in o one e i o ial uni u ned ou o be co ec and use ul also in p e ious esea ches [17]. a) Hypo hesis H1: The a e age numbe o con ibu ions in each hema ic pilla is compa able wi hin indi idual egions. We will compa e he a e age annual numbe o pos s in each egion and pilla s using one- dimensional analysis o a iance. Since he dis ibu ion o da a se doesn’ exhibi no mal p obabili y dis ibu ion as con i med by Ja que- Be a es he non-pa ame ic analysis o a iance ep esen ed by K uskal-Wallis es will be used. Fi s , i will be compa ed he a e age numbe o con ibu ions pe capi a (pe 100,000 inhabi an s) o a egion among pa icula pilla s. Resul s o K uskal-Wallis es a e shown in Tab. 4. Tab. 4: Non-pa ame ic analysis o a iance (K uskal-Wallis es ) Numbe o news pe capi a K uskal-Wallis H741.189 d 9 Asymp o ic Signi icance 0.000 Sou ce: au ho s The asymp o ic signi icance is less han 0.05, so he hypo hesis H1 can be ejec ed a 5% signi icance le el which means ha he di e ences in he numbe o con ibu ions pe capi a among he pilla s a e s a is ically signi ican . The la ges amoun s o con ibu ions pe capi a h oughou he Czech Republic a e ela ed o secu i y, acciden and social pilla s. Ou esul s con i m he p e ious esea ch indings ha in he news p og ams (especially in comme cial TVs) abloid opics a e clea ly [18]. The same esul can be exp essed ia so ed means plo which is o en used o desc ibe he esul o analysis o a iance es s, see Fig. 3. On he o he hand, Eu opean & in e na ional, educa ion & science and spo pilla s p o ed o ha e he smalles sha e among all pilla s. Mo eo e , he same applies o p ac ically all NUTS III egions. The only excep ions a e Ka lo y Va y and uni ied P ague and Cen al Bohemia egions whe e social pilla is he mos abundan one. b) Hypo hesis H2: The ep esen a ion o con ibu ions om a pa icula pilla is compa able among egions. To e alua e hypo hesis H2 we will again u ilise he analysis o a iance wi h one ac o . When accomplishing he analysis o a iance, he a e age numbe o con ibu ions o a pilla pe capi a (pe 100,000 inhabi an s) will be compa ed among indi idual NUTS III egions as he ac o . EM_04_14_zlom 1.12.2014 11:13 S ánka 133 Ma ke ing a obchod 140 2014, XVII, 4 Abs ac MEDIA PORTRAYALS OF REGIONS IN THE CZECH REPUBLIC: SELECTED ISSUES Jan Sucháãek, Pe Seìa, Václa F ied ich, Ja osla Kou sk˘ Amoun o in o ma ion p o ided by media is g owing i ually e e y day. Su p isingly, a en ion de o ed o he spa ial implica ions o media in o ma ion is a om su icien . Exis ing esea ches b ing ample e idence ha TV co e age embodies media agenda in a sa is ac o y manne as i has a he p o ound impac on he public on he one hand and is ep esen a i e enough on he o he . Ou pape deals wi h TV news epo ing ela ed o NUTS III egions in he Czech Republic. The main pe spec i e adop ed in his a icle is a quali a i e one, which acili a es he e alua ion o TV news con en s. We a e delibe a ely ocusing on na ional TV b oadcas ing, which is he mos in luen ial one. Indi idual con ibu ions we e alua e appea ed in he amewo k o news epo ing in public Czech TV and p i a e No a and P ima. In spi e o se e e limi a ions in media da a, ou pape co e s he pe iod be ween he yea s 2004 and 2011, which helps o unde pin ou conclusions in a su icien way. Con ibu ions, which a e ela ed o indi idual NUTS III egions and esona e a he na ional le el, cons i u e he ocal poin o ou esea ch. TV e lec ion o ma e ial eali y is o in angible cha ac e ; none heless simila ly o ma e ial ca ego ies, one can iden i y spa ial as well as hema ic di e en ia ion and e en pa e ns also in he sphe e o TV news epo ing. The e a e only li le doub s ha media landscape and closely associa ed men al maps ha e been la gely unde es ima ed so a . Thus, hese opics ep esen one o pi o al heo e ical and p ac ical challenges in he ield o spa ial sciences, which holds ue o pos - ansi ion coun y e en mo e. Key Wo ds: NUTS III egions, Czech Republic, media, TV news epo ing, analysis o a iance, co espondence analysis. JEL Classi ica ion: B52, R10, R19, M30, M39, Z10. DOI: 10.15240/ ul/001/2014-4-010 EM_04_14_zlom 1.12.2014 11:13 S ánka 140