| Abstract | Rad govori o većini čimbenika koji utječu na ponašanje birača u prvome dijelu, dok je u drugom dijelu riječ o mogućnosti utjecaja samih predizbornih anketa na birače. U radu se daje povijesni pregled istraživanja javnoga mnijenja, od „straw poll“ novinskih anketa u 20. stoljeću do današnjih naprednih tehnika i metoda znanstvenog ispitivanja javnoga mnijenja. Vezano uz metode i tehnnike, anketno je ispitivanje najvažnija i najučestalija metoda ispitivanja javnoga mnijenja. Upravo se u predizbornim istraživanjima najčešće koristi metoda ankete. Da bi anketa bila valjana i pouzdana, jedan od važnijih čimebenika je reprezentativnost uzorka. Reprezentativnim se uzorkom smatra onaj koji pruža rezultate sukladne onima što bismo ih prikupili ispitivanjem cijele populacije, međutim, smatra se da ne postoji neki objektivni kriterij za utvrđivanje reprezentativnosti uzorka, te da reprezentativnost uzorka ovisi o puno čimbenika, od kojih je neke teško kontrolirati. Odbijanje ankete i neizjašnjavanje ispitanika jedni su od ključnih metodoloških problema društvenih, a tako i predizbornih istraživanja. Upravo navedeno utječe na reprezentativnost uzorka, jer zbog visoke proporcije odbijanja anketa i uskraćivanja odgovora, čak i najkorektnije planiran anketni uzorak može postati pristran. Kako od mišljenja dolazi do djelovanja? Pojedinac prvo razmišlja o nekom pitanja, interpretira ga i konstruira mišljenje, zatim to svoje mišljenje izražava javno, a sve to dovodi do djelovanja i samog čina ponašanja, u slučaju izbora, izbornog ponašanja. Na izborno ponašanje utječe mnogo čimbenika, od osobina, stavova i socijalnih oblježja birača, te pripadanja nekoj socijalnoj grupi, preko programa stranke/kandidata, povjerenja u stranku/kandidata, do mogućnosti ispunjavanja biračevog cilja činom izbora. Da bi se provjerio mogući utjecaj predizbornih anketa na birače, potrebno je da ih oni zapažaju i smatraju vjerodostojnim. U Hrvatskoj se sve više takve ankete opažaju i sve im se više vjeruje. Međutim, čvrstih dokaza za postojanje utjecaja preizbornih anketa na ponašanje i preferencije birača, zapravo nema. Naime, svako društveno istrašivanje, pa tako i predizborna, ovise o situaciji i vremenu, stoga uvijek traže stalno provjeravanje, i pitanje utjecaja predizbornih anketa na birače zapravo ostaju otvoreno. Postoji nekoliko teorija potencijalnog utjecaja preizbornih anketa na birače, od kojih nijedna nije potpuno dokazana. |
| Abstract (english) | The paper discusses most of the factors that influence the behavior of the voters in the first part, while in the second part they are about the influence of the electoral polls themselves on the voters. The paper gives a historical overview of public opinion polls, from the "straw poll" of newspaper polls in the 20th century to today's advanced techniques and methods of public opinion polling. With regard to methods and techniques, the survey is the most important and most common method of public opinion polling. It is precisely in the preelection research that the survey method is most often used. For a survey to be valid and reliable, one of the most important factors is sample representativity. A representative sample is considered to be the one that provides the results consistent with those that we would collect them by examining the entire population, however, there is no objective criterion for determining the sample representativeness, and that the representativeness of the sample depends on many factors, some of which are difficult to control. Rejection of the survey and non-disclosure of the respondents are one of the key methodological problems of social as well as pre-election research. The aforementioned is influenced by the representativeness of the sample because of the high proportion of rejection of the survey and the denial of the response, even the most accurate planned survey pattern may become biased. How does thinking come from acting? The individual first ponders a question, interprets it and constructs the opinion, then expresses his/her opinion in public, all of which leads to action and the act of acting itself, in the case of election, electoral behavior. Electoral behavior is influenced by many factors, traits, attitudes and social voter choices, membership of a social group, party/candidate programs, party/candidate trust, ability to fulfill the voter's goal by election act. In order to check the potential impact of pre-election polls on voters, it is necessary that they observe and consider them credible. In Croatia, more and more such surveys are perceived and believe them all. However, solid evidence for the existence of pre-election polls on the behavior and preference of voters does not really exist. Namely, any social research, even preelection, depends on the situation and the time, so they always require constant scrutiny, and the question of the influence of pre-election polls on voters remains open. There are several theories of potential influence of pre-election polls on voters, none of which has been fully demonstrated. |
