Assessing Political Marketing Channel Experiences in the 2015 British General Election
   
Co-Supervisor: Professor Paul Baines
Co-Supervisor: Dr Emma Macdonald
Supported by: Professor Hugh Wilson

  
Broad Areas of the Study

This unique doctoral opportunity is available to a full-time researcher in the interface between political marketing (the use of marketing in political contexts) and marketing communications (specifically multi-channel communication).  It would therefore be of particular interest to those who wish to pursue a future academic career.  The study involves the study of marketing during the next British General Election (expected in 2015).  It involves the use of the innovative real-time experience tracking method, the subject of a recent Cranfield article in Harvard Business Review (Macdonald, Wilson & Konuş 2012), in collaboration with MESH Planning, our market research company partner.  A background including statistics is required.

The Design of the Study

The envisaged study design, to be fine-tuned in collaboration with the student, is as follows.  Our approach uses an online panel of 'floating voters' using a combined online and mobile phone survey collection method.  Data are collected continuously throughout the campaign period of the election and beyond the day of voting.  Data collected include demographics, voting intentions, actual voting, and encounters with a large variety of touchpoints experienced during the campaign including communications generated and/or controlled by the relevant parties, as well as those generated by the media and other sources, such as word-of-mouth through peers.  By touchpoint, we mean all encounters with the brand - in this case, the three main UK political parties - as this term is broader than the notion of media exposure.  Data are to be collected in two forms.  Firstly, an online survey will be used for collecting demographics, voting intentions, and perceptual measures at the start of the campaign; these measures will be repeated at the end of the study with an additional request for actual voting behaviour.  Secondly, mobile phone text messaging will be used to capture contemporaneously, or in 'real-time', the election touchpoints encountered by an individual during the period of the study.  The granularity of touchpoint information provided by the mobile phone methodology, combined with pre-and-post comparisons between intentions and behaviour, enables an analysis of the impact of various touchpoints on voting behaviour.

Our sample of floating voters (defined as respondents who were not 'absolutely certain' about which party they would vote for) will be recruited via an online panel (by market research company Research Now in association with our partner, MESH Planning).  The panel uses quota sampling to recruit a sample of circa 1,100 respondents, representative of the UK population in terms of region, social status, age group and gender.  Respondents will be also screened on the following criteria: 1) access to a mobile phone, 2) ability to send text messages from their phone; and 3) willingness to use their mobile to participate in the research.  Due to the typically four-week commitment required from respondents involved in the study, there is likely to be a significant drop out rate (circa 56%), although drop-outs will be replaced with respondents of a similar profile.  However, for the purpose of this study, to test an innovative new method of panel data collection and to identify the effects of communication experiences on voters, the final sample selected used only those floating voters who participated throughout the whole period of study.

The study design integrates online and mobile phone technologies.  At time t0, respondents complete an online survey which collected their demographic information.  This online survey also measures their likelihood to vote in the upcoming general election (1-10 scale, anchored by 1=absolutely certain not to vote and 10=absolutely certain to vote), which party they would vote for (single response), and their certainty of voting for the selected party (1-10 scale, 1=not at all certain to 10=absolutely certain).  At time t1, four weeks later (or whatever the length of formal political campaign is) and immediately following polling day, the same respondents will again complete an online survey which measures reported voting behaviour including whether they voted (single response) and which party they voted for (single response).  Qualitative data will also be collected at time t1 with regards to the reason for the voting decision.

In between time t0 and time t1, respondents will provide mobile phone text messages each time they experience a touchpoint related to the election campaign.  Each text message will provide information about:

the relevant party;
the touchpoint type;
positivity: the valence of the immediate emotional response to the communication (a single Likert-scale item, "How did it make you feel?", on a 5-point scale anchored by 'very positive' and 'very negative'); and
persuasiveness: the immediate self-assessed impact on consideration (a single Likert-scale item, "How much more likely did it make you to vote for this party in the election?", on a 5-point scale anchored by 'much more likely' and 'much less likely').

What do we Expect to Find?

Uniquely, this approach of collecting touchpoint experiences allows us to assess the relative influence of different direct and indirect marketing communications experiences on voting - an approach that has not previously been researched so comprehensively in the literature with politics or indeed any commercial brand choice.  This method also allows us to assess campaign effects in real-time, so that we can map how different communication techniques are impacting, for example, on voting intention, or brand image attributes.  This makes our study unique and potentially of value beyond political marketing to mainstream marketing, where assessing marketing influence is becoming of increasing importance.

For more information on work published by members of the team using the real-time experience tracking approach, please consult the following articles:

Baines, P., Macdonald, E., Wilson, H and Blades, F. (2011), "The Relative Importance of Channel Experiences and their Impact on Voting Intention", Journal of Marketing Management, Vol 27. No.7/8, pp.691-717
Macdonald, E.K., Wilson, H. and Konuş, U. (2012), "Better Customer Insight - in Real-Time", Harvard Business Review, September,  pp.102-108.
Candidate Profile

Candidates must have a good bachelor's degree in a quantitative discipline (e.g. mathematics, statistics, economics) and a working knowledge of one or more statistical software packages.  Some experience in dynamic modelling (such as panel data analysis) is desirable; an enthusiasm for stretching your statistical skills in this direction is essential.  An interest in politics and/or marketing communciations is clearly helpful though no previous expertise is required.  A Master's degree is desirable but not obligatory.  Candidates with interests in this area are invited to  make contact with either Professor Paul Baines or Dr Emma Macdonald by emailing us on [log in to unmask] or [log in to unmask]