The Sentiment of the 2017 Election Programmes
|Date:||06 February 2017|
In the politics of recent years, emotion has played an important role, and this will be no different during the Dutch general elections in March. How do political parties deal with emotions? In a recent research project, Harry Garretsen and Janka Stoker studied political party election programmes for the upcoming elections, comparing the degree to which the emotional wording of the programmes was negative or positive. Results show that the angry voter is generously rewarded by the PVV, the optimistic voter by the VVD, followed closely by the CDA and D66. Whatever the outcome of the elections, the new cabinet taking office after 15 March will have to take these sentiments into account.
This blog is a shortened version of the following publication in Dutch: Harry Garretsen, Janka Stoker, ‘Het sentiment in verkiezingsprogramma’s 2017 op de snijtafel’, Me Judice, 17 January 2017.
As we approach the March elections for the Dutch House of Commons, not only do facts play a role, but voters’ feelings are ever more emphasized. Politicians no longer convince with facts on how well the Dutch economy is performing. The main issue seems to be how to resolve the ‘sense of unease’ of many citizens. This is not new and not restricted to our country, as Trump’s victory and Brexit have shown. Nevertheless, the impression is that political discussions in the Netherlands are more emphatically fuelled by emotions today than in the recent past. How do political parties handle this shift strategically? A good way to find out is to study their election programmes more closely, expressly focusing, not on content, but rather on trying to measure the emotional undercurrent within each programme. We have done this for the 2017 election programmes of 14 Dutch political parties.
How positive are the 2017 election programmes?
Following a recent study by Crabtree et al. (2016), and using the linguistic LIWC program (Pennebaker et al., 2001; Zijlstra et al, 2004), Garretsen and Stoker calculated the ‘positive sentiment’ for each of the programmes. This positive sentiment is calculated by deducting the percentage of words charged with negative emotions from the percentage of words charged with positive emotions in a text. The results show, in the first place, that the governing VVD party, which also provides the Prime Minister, scores highest on positive sentiment . Secondly, when the outcomes for the traditional, mainstream parties VVD/D66/CDA/PvdA/CU are compared to those for the more extremist parties like SP/PVV/DENK/PvdD/VNL/FvD, it is striking that the Dutch mainstream parties clearly score higher on positive sentiment than the more extreme and/or newer parties. The PVV is the only party with a negative (!) score on positive sentiment, but VNL, the PvdD and DENK also exhibit sentiment scores well below the average of 1.43. Finally, it is remarkable that the VVD and the PVV, the two parties leading the polls and likely to be the first ones to be requested or instructed to try and form a coalition after the elections, are furthest apart in terms of their positive sentiment scores.
Does sentiment actually matter?
Of course, the final question is whether the way in which election programmes put sentiment into words actually matters for the election results. We will know on March 15th. Research does show, however, that perceptions and feelings do influence our choice of leaders. This applies to political leaders as well as to CEOs. And if it is true that voters’ feelings carry significant weight at the ballot box these days, the emotional charge of the political message, as worded in these election programmes, might indeed be of great influence.
So, is it better for a party to express a positive or a negative sentiment? Will Prime Minister Rutte’s (VVD) positive sentiment garner more support than Mr Wilders’ (PVV) negative sentiment in March? Politicians and campaigners tended to believe that optimism wins in the end. It is exactly on this point, however, that times seem to be changing, as the election victory of Trump and the position of the PVV in the polls seem to confirm. It remains to be seen who will be the winner on March 15th. Either way, it is clear that feelings do matter and are nowadays easily quantifiable.
- Crabtree, C., Golder, M., Gschwend, T. & Indriðason, I. “Campaign Sentiment in European Party Manifestos.”(under review, 2016).
- Pennebaker, J. W., Francis, M. E., & Booth, R. J. (2001). Linguistic inquiry and word count: LIWC 2001.Mahway: Lawrence Erlbaum Associates, 71, 2001.
- Zijlstra, H., Van Meerveld, T., Van Middendorp, H., Pennebaker, J. W., & Geenen, R. D. (2004). De Nederlandse versie van de ‘linguistic inquiry and word count’(LIWC). Gedrag & Gezondheid, 32, 271-281.