Which single individual is going to influence you in how to vote in the general election more than any other?
Recent research suggests voters will be most swayed by a spouse.
A study entitled 'The Political Values and Choices of Husbands and Wives', investigating 2,846 British citizens' political preferences, finds a 'symmetrical' pattern of influence.
Men and women both give greatest weight to their own political values, but also accord significant impact to their partners' attitudes.
The authors of the study, Man Yee Kan and Anthony Heath from the Department of Sociology, University of Oxford, point out that previous research has found marital and co-habiting partners are a great deal more similar to each other in their voting than are two unrelated individuals with the corresponding characteristics.
Partly this could be because people with similar political attitudes will be more likely to marry (or cohabit) than people with very different preferences. The longer the couples remain together the more similar they become in their voting patterns, suggesting either influence, or that couples who vote differently are more likely to split up.
The findings of the study, published in, 'Journal of Marriage and Family', are described as strikingly out of line with much previous research. This tended to emphasize a pattern of influence where husbands are not so much influenced by their wives, whereas wives are largely swayed by their husbands.
But this new study found a much more symmetrical picture. Both husbands and wives give greatest weight to their own political values, but both also accord some significant weight to their partners'.
The picture is not completely symmetrical, and men show slightly greater independence in their values than women do, but the difference is not substantively very large.
Only in the case of men with a female partner who is the main provider, did the researchers find political independence, with the man according no weight to his partner's values.
The study also found that a politically uninterested man will accord more weight to his partner's values if she is interested in politics, in much the same way that a politically uninterested woman accords more weight to the political values of her interested partner.
These results are in line with the phenomenon that partners' political preferences become more similar over time.
The authors conclude that one possibility is that there has been a recent shift and that there has been a tendency for relationships to become more egalitarian.
The research also found that current occupation is a poorer guide to women's than to men's political values. For women, educational level provides a more powerful predictor.
But there are theories that it is brutal economics that ultimately determines how husbands and wives votes - these theories argue the higher are earnings, the more the individual shifts to the political right.
The impact of an individual's earnings on voting should therefore be larger, the greater is the individual's share in total household earnings.
In a more recent study of British citizens entitled 'How husbands and wives vote', Marte Strøm from the University of Oslo, Norway, found that on average women earn less than their husband, and vote according to their husband's income.
But the new study, published in the journal 'Electoral Studies', also found that if the wife is the maximum earner of the household or works fulltime, she votes more according to her own earnings.
The results show that earnings in general have a large impact on voting behaviour for both men and women.
For every 1% higher household income for men, there is a 0.012 percentage point higher probability of voting Conservative. For women, the probability is 0.016 percentage point higher (the more women earn the greater the likelihood of voting Conservative compared to what happens to men as they earn more). There is a large positive impact of owning your own house on Conservative voting. Those who own their own home are around 10 percent more likely to vote Conservative.
However there is also an asymmetric impact of the spouse income for men and women on average. While men are not influenced by his spouse's income, women are.
The study also found that the impact of income on voting behaviour is sensitive to what age earnings are measured. How much you earn when you are between 33 and 41 increases the impact of income for both men and women over the life-cycle.
Other research finds a modern gender gap in voting where women have become more leftist relative to men and this pattern is observed in all OECD countries over the last decades.
Possibly the growing number of economically vulnerable single women/lone mothers and the higher risk of worse economic outcomes for women accompanying the higher divorce rates, explains this.
Are women not just being influenced in their political ideology and voting patterns by their current situation, but also with half an eye on the possibility of divorce?
Marte Strøm concludes that with women earning an increasing share of household income, her earnings will be increasingly important in impacting voting behaviour in a household, relative to her husband's income.
Given the real but subtle impact of your partner on your vote, the most sophisticated election campaigns in the future might target you - but do it through your partner.
The psychological advantage to the Public Relations machinery in politics is that way you won't know you have been persuaded, and this will make it more difficult to psychologically defend against.
Beware how much your vote is truly your own.