Thecurrent paper is based on a positivistic philosophy on deductive approach:hypothesis provided above were deducted from the literature review and havebeen tested. Collecting quantitative data, mono-method has been applied(Saunders, Lewis, & Thornhill, 2007). Due to time limitations, across-sectional time horizon was used in the current paper (Bryman A.
,2012). The required primary data wascollected through the survey with a diverse sample of participants (n=120). A literaturereview allowed to improve the questionnaire’s structure and its content. Inorder to apply a statistical approach to data analysis, the research design wasbased on quantitative questions. The questionnaire includes 9 questions thatare strictly directed to find out whether Facebook influences positivelyelectoral choice.
Questions were separated into 3 dimensions as follows:Dimension (1): politicianparticipation in social networks (Q3)Dimension (2): electorates’trust (Q2)Dimension (3): chances towin (Q6, Q7) In orderto answer the main research question amount of questions is sufficient. Alldata was transferred to the special program for statistical analysis (SPSS) andthe answers were decoded in numbers to facilitate the measurement itself. Thereare three types of data that could be measured in SPSS: scale, ordinal andnominal (Bryman & Cramer, 2011), but considering current research, allquestions from the poll are nominal. For Values of a nominal measure vary incategories. It is not possible to rank the categories, to calculate thedifference or relation between them (“yes/no – answers”) (Bryman & Cramer,2011).
Questions 2, 3, 6, 7 were considered as the most relevant for the papergoals and, hence, for proving hypotheses. All questions are listed in theappendix at the end of the paper. Descriptive Statistics tools such asFrequencies and Crosstabs along with Binomial Test and a qualitativeinterpretation of noticed correlations were applied in Figure 1.
(Figure 1) To prove H1, firstly,it has been figured out the frequencies of the answers on Q3 concerningpeople’s trust towards candidates who use social networks to communicate withcitizens. Then, to Q3 was applied a Binomial test (non-parametric test), whereit was assumed that the proportion of people whose level of trust to thecandidates’ social media use would be more than 55% (0.55). In order to applycrosstab method to questions 2 and 3, Q2 frequencies have been found. Forinterpretation H2 data (how the use of Facebook increases the electorates’trust in public authorities), the frequencies and binomial test were applied toQ6. A test proportion of 85% (0,85) was considered as a percent of people whowould think that it is possible to change citizens’ attitude to a certain candidateif the candidate conducts his pre-election campaign via Facebook. Lastly,crosstab method was applied towards Q3 and Q6.
For the final hypothesis (H3)there have been considered the same steps as for previous ones (H1, H2): thefrequencies of Q7, and crosstab for Q6 and Q7. These methods have been used forgetting statistical data, that contributed to rejection/approval of hypothesis.The current paper aimed to build theory from the datato develop a better understanding of how social media use affectselection results (Bryman A. , 2012). Sample size established thereliability of the study results. Because of time limitation, the random sampletechnique has been chosen.
This technique was chosen in order to escape biasand prejudice of citizens’ attitude toward political leaders use social mediaand to get own independent results. The target audience was established as Germanand Russian universities. Online data collection occurred from25.11.
2017 to 25.12.2017 on the Facebook-pages ‘IUBH Bad Honnef’ and ‘Networking:?????????? ???’ with 4,861 and 1,258 members accordingly. Members from thefirst group are mainly students from International University of AppliedSciences from Bad Honnef, Germany; members from the second group are currentand graduated students and professors from Lomonosov Moscow State University,School of Public Administration, Russia. After closing our survey, the numberof answers received that can be evaluated achieved 120 people.
4.Results and Discussions. In order to answer the main researchquestion, firstly, it is important to find out whether people use socialnetworks as a tool for monitoring world’s political news (Q1). The results ofthe survey showed that there are 89 (74%) of respondents check out the latestnews through the social networks, and 31 (26%) of those who do not do itthrough them. Basically, it means that 26% of all respondents follow the newseither through traditional media or in practice (respondents from School ofPublic Administration are mostly students and professors who have a personalconnection with the public authorities in Russia). As the majority of respondents use socialnetworks as a monitoring tool, it makes sense to continue the current research.The general situation concerning people’s trust towards candidates who usesocial networks to communicate with the public (Q3) looks like:The observed ratio of those who trust such candidatesis 0.47.
With the help of SPSS (ExactSig. (1-tailed)) p-value = 0.041. So if the proportion of those whose level of trustshould grow in the entire population is 55% then there is only 4.
1% chance tofind them in a sample of 120 people. The hypothesis is usually rejected whenthis chance is smaller than 5% (p < 0.05) (Lehmann, 1986), so the ratio of those whosetrust would grow is smaller than 55%. Figure 2 reveals that the respondents indicate a highresponse toward the social networks use by public authorities. The Frequenciesof the answers on Q2 as follows: (Figure 2) Applying theBinomial test in this case and assuming that the ratio of those who thinkpublic authorities should use social networks in order to be closer to citizensequals 85%. In the present research, the observed proportion is 87% and thep-value=0.
267 which prove hypothesis of 85%. Finally, to prove H1 answers Q3and Q2 have been compared, in order to see whether people who expect candidatesto communicate to the public via social media, in reality, trust them more thancandidates who do not: Do you trust more to those public authorities who communicate with citizens via social networks? Yes No Should public authorities use social networks in order to be closer to citizens? Yes 51% (54) 49% (52) No 14% (12) 86% (2) (Table 3) The analysis in the Table 3illustrates that the majority of the respondents approve the idea of use ofsocial networks use by politicians. However, at the same time among those whosupport this point of view, almost half of them (49%) asserted that social mediause does not raise their level of trust towards such political leaders. Inother words, the use of social media tools became a “must”, one of theconventional tools, rather than a winning advantage. A politician nowadayssimply cannot go without them. For proving H2, frequencies of Q6 hasbeen figured out and the Binomial test was applied. The observed ratio is 87%, the p-value is26.
7% as in the previous case for the Q3. Then to prove H2 Crosstab method forQ3 and Q6 has been made: Do you trust more to those public authorities who communicate with citizens via social networks? Yes No Does it possible to change citizens’ attitude to certain candidate if this candidate conducts his pre-election campaign via Facebook? Yes 45% (54) 42,5% (51) No 2,5% (3) 10% (12) (Table 4) The results indicate the same situation as with the previous hypothesis:the majority of respondents believe that the use of Facebook by candidatesaffects the positive electorate attitude towards them. But among those whothink so, almost half of the respondents would not trust the candidate moredespite of his (candidate’s) activity in this social network.
Thus, the use ofFacebook has a positive influence on public opinion but using this network asthe only driver of manipulation of public opinion is not enough to establishtrust between candidates and citizens. Referring to the theoreticalfoundations, some authors slso pointed out that social networks are vitallyimportant tools for creating the needed level of trust with the electorate(Susila, Dean, & Harness, 2015; Schneider Hahn, Scherer, Basso, dos Santos, 2016). However, according to current results, Facebook useas an independent instrument is not effective enough in building trustfulrelationships with voters. For the final hypothesis (H3) there havebeen considered frequencies of the Q7: (Figure 5) And below crosstab method for Q6 and Q7 was recieved(Table 6): Is it possible to win elections using FB? Yes No I do not know Facebook should serve as an auxiliary, but not the only tool Is it possible to change citizens’ attitude to certain candidate if this candidate conducts his pre-election campaign via Facebook? Yes 42,5%(51) 14% (17) 29% (35) 3% (4) No 0 6,5% (8) 4% (5) 0 (Table 6)Theresults shown in Table 6 provide that the majority of respondents considerFacebook use as an influencing tool on the public, among that majority,respondents also see the opportunity to win elections with the help of this socialnetwork. Also a significant share of respondents answered “I do notknow”, which indicates either the incorrectness of the question (one ofthe respondents expressed such opinion) or its ambiguity. Indeed, those whoanswered on Q7 in own words, mentioned that Facebook can become a factor inwinning elections but just in conjoint with other means and channels. Thus, itcan be asserted that Facebook affects chances to win the elections, but onlythe use of a proper mix of tool and channels, including social media, canactually provide the victory.
Takinginto consideration all mentioned above and answering the main research questionit can be concluded that: Facebook is an instrument that is able to buildtrustful and consistent relationships with the electorate and hence caninfluence the electoral choice in favor of a certain politician. However, it isimportant to note that using solely Facebook might not be effective in winningelections, so it should be used as auxiliary (additional) mean, not the onlyone.