Can social media tell who will win Monday?

Predicting the election on Monday has been very difficult. People are mostly separated by strong emotion and are using this as a way of telling what to expect. Previous polls that put the opposition in the backseat where rubbished and trashed particularly on social media. More recent polls that have put the opposition in good standing have been accepted. Likewise poll results that predicted landslide victory for the ruling party have been rubbished.
Whether the opposition or ruling party will win this election has therefore depended on the wishes of the person being asked or reading a report. Social media has proven to be very personal space where individuals will entertain things they agree with and block off what they disapprove of.

If you are using social media sentiment you might conclude that the opposition MDCAlliance led by Nelson Chamisa will win. If you are using opinion polls the answer is either side might win or the country is headed for a GNU. This is a reasonable expectation considering that those most active on social media are youths who have been captured to some extent by opposition politics of generational consensus.

On Twitter, the sentiment is clearly anti- Mnangagwa and the establishment but in Zimbabwe Twitter only has about 90 000 users which makes it unsuitable to have a more general prediction of Zimbabweans’ behavior particularly those in rural areas. On Twitter those posting positive things on the ruling party were mostly recipients of vitriol and intolerant language. Those praising the opposing were generally applauded and had their material shared. Sharing of material in these personalized space is a sign of approval and appreciation.

Where accusations of political violence are traded, opposition supporters are unlikely to share or retweet the negative publicity. Ruling party supporters have however been quick to share and retweet any content portraying the opposition as violent. Where individuals feel strongly about an issue they would share and aprove of it regardless of whether it is true or not.
sentiment 5

On Whatsapp the numbers are slightly different. There are reported 5, 2 million Whatsapp users in Zimbabwe. Not all these Whatsapp users are registered voters. Some of the most avid users of Whatsapp are children.

On whatsapp we targeted neutral groups such as residents’ associations; groups of old school mates, journalists and other neutral groups. Political party groups or groups set up by party activists were avoided but this is where more political debate is occurring. Political activists whatsapp groups tended to be biased along party lines. They are formed specifically to mobilise people of one party within a certain area. So a Zanu PF cell in Stoneridge Harare South for example is exclusively for the party supporters in that area to coordinate their political activities.

On Whatsapp the sentiment is negatively biased against Emmerson Mnangagwa in the 8 neutral groups that were analysed which had a sample size of 300 people. People discuss four main topics, elections, corruption, cash crisis and service delivery. The people in these groups are mainly between the ages of 26 to 40 years old. The study was conducted 10 days before the election where more or less people are set about their electoral choices.

sentiment 6

On corruption the sentiment was clearly against president Mnangagwa. Whatsapp users doubted his commitment to fighting corruption giving examples of perceived corrupt ministers in his cabinet. The debate became more intense when former energy minister Samuel Undenge was jailed for corruption involving $12000. Most users saw it as politicking and expressed themselves through memes.Instead of celebrating that a corrupt official had been jailed they questioned why he was targeted. This suggests a lack of confidence in the fight against corruption by Mnangagwa.

sentiment 4

Most of the participants on these groups discussed the cash problem as their biggest grievance. They had no confidence that Zanu PF has solutions for the cash problems. They are also delusional about the mega deals signed by Mnangagwa. They mocked Mnangagwa’s efforts to revive the economy arguing that he was part of the team that had destroyed it in the first place.

On elections the main content shared was of Nelson Chamisa’s oversubscribed rallies and those depicting Mnangagwa as unpopular such as videos of people purporting to be walking out of his rallies. Another big topic within the groups was the ‘rigging’ by Zanu PF through ZEC. There were a lot of forwarded messages about how ZEC intended to rig the elections as well as warning messages educating recipients on what to do to counter the rigging efforts.This showed that the participants realy believed Zanu Pf would rig the elections

But there is a catch. It is only the minority who participate in most of the groups. In one group with 118 members for example only 6 people participated in political debate on a regular basis. The rest were quiet leaving us to speculate whether they consume the political content being shared on the platforms. Only about 7% of the people groups under review participated actively in commenting on the posted content. There was no way of telling if they shared the content without commenting. Only about 10% of the people in these whatsapp groups we considered as neutral engaged actively with political content.

We also analysed the type of content being shared in these groups. Most of it was fake news in form of doctored pictures or deliberate falsehoods.Whether one believed this news or not depended on their preconceived political preferences. If one supports Zanu PF for example they are likely to believe any news disparaging MDC Alliance whether such news is true or not. A good example is the views of users on what Kofi Annan said when he visited the country.Twitter and Whatsapp users under review approved or disapproved of what the former UN secretary general said based on their political preferences.

One therefore cannot conclude with certainty how the election will go based on social media analysis.We have to wait and look to other indicators for more accurate predictions. Some of those who are active on Whatsapp in some groups professed not to be legitimate voters. There was no way of measuring if the dormant whatsapp group members are consuming political content in their groups and if they are convinced by this content or if it will shape their political choices. A bigger sample size of Whatsapp groups may have given a more accurate reflection of things. Study of political or activists whatsapp groups may be more useful to see if there is a connection between online and offline actions. In the meantime we must look beyond social media to find that illusive answer.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s