Which candidates are placed on top of lists? Examining characteristics selectors are looking for in candidates to legislative elections

Candidate selection is one of the most important tasks of political parties. It determines who is placed in front of the electorate and thus often the parties’ electoral fortune. Despite this importance, only few parties have implemented formal requirements prospective candidates should meet (Hazan and Rahat, 2010).[1] Due to this lack of formal rules, we know little about the characteristics party gatekeepers are valuing when selecting their candidates. In this blog post, I make a first attempt to answer the question of how individual-level characteristics affect the likelihood of being selected on viable list places.[2]

Incumbency and experience constitute key traits that tend to foster aspirants’ likelihood of being selected as candidates according to previous research (e.g. Gallagher and Marsh, 1988). Beyond these well-researched factors, the literature draws attentions to further characteristics of potential candidates that might determine the chances to be nominated such as age, years of party membership, and sex. The age of candidates is frequently used in election campaigns to promote or undermine candidates; older candidates are alternately portrayed as frail and vulnerable or wise and experienced and younger candidates are described as either beginners or energetic and progressive (Campbell and Cowley, 2014). In fact, how selectors respond to the age of candidates remains empirically understudied. Moreover, when assembling party lists, political gatekeepers might reward membership and prefer aspirants who they have known and whose skills they can easily asses. Hence aspirants who have served as members for many years might be more likely to hold viable list placements. Finally, Gallagher and Marsh (1988) also emphasize that political parties might use lower (electorally hopeless) positions on the list to produce balanced tickets. As a result, female aspirants might have higher chances of being selected in the first place, but it may decrease their likelihood of running on top of the list. Research studying the UK and U.S. show that women indeed tend to be overrepresented among those candidates that have no or little chances to succeed in the election.

To test these expectations, I draw on data from the Comparative Candidate Survey (CCS, 2016) between 2008 and 2010. The CCS offers a unique data set to examine candidates at national levels because it is the only source of information about candidates’ characteristics (e.g. perceived chances to win and socio-demographic details) in a number of European countries. Overall, the sample allows me to study a total of 2066 respondents from the seven countries (Belgium, Italy, Netherlands, Norway, Portugal, Romania, Slovenia) that employ closed ballot system.

A dichotomized variable captures the candidates’ viability based on the respondents’ initial evaluations of their chances to win a mandate as asked by the CSS. It takes the value 0 for those candidates who expected not to win and 1 for those who were certain to win. Besides the variables of main interest (age, experience, incumbency, years of membership and gender), the analysis controls for party-level factors such as quota rules (International Institute for Democracy and Electoral Assistance, 2015) and left-right placement (Comparative Study of Electoral Systems, 2016). Moreover, variables indicating whether the candidate selection was decided centrally (at the national level) and inclusively (by a large selection committee, both variables collected by author) are integrated in the models as selection rules are likely to affect the outcome of candidate selection, too. Finally, district magnitude (Beck et al., 2001) and women’s proportion in the national labor force (World Bank, 2016) control for contextual factors. To account for the un-modelled heterogeneity at the country level, I employ country fixed effects.

Table 1 presents the results of logistic regressions estimating whether a candidate believes to hold a viable placement or not. Model 1 confirms most expectations detailed above. Incumbents are much more likely to be placed on viable positions. Also, older candidates are more prone to be placed on a higher list position and seem to be perceived as more qualified due to their life experienced. Male candidates are around 40 percent more likely to hold a viable list positioning than women. To control whether this significant relationship is driven by men’s higher confidence in their success and to account for gender bias in the evaluation of viability, I conducted a robustness check (not reported here) including a variable capturing whether candidates won in the following election. Incorporating the variable, male candidates’ likelihood for a viable position becomes even stronger and remains statistically significant. The finding thus supports the idea, that party gatekeepers prefer men over women for viable seat positions.

Table 1: Explaining candidates’ likelihood of self-evaluated viable candidacy, logistic regression estimates with country fixed effects.2018_july_blog_tab1Annotations: Exponentiated coefficients * p < 0.10, ** p < 0.05, ***
p < 0.01. All models are estimates with country fixed effect, which are omitted for reasons of better readability.

To analyze how this effect evolves with a candidate’s seniority, Model 2 incorporates an interaction term between gender and age. Figure 1 displays the interaction effect visually and shows that young women are least prone to hold viable list positions, while older women’s chances increase. Hence, party gatekeepers seem to perceive young women as least competent or likely to succeed in elections.

Figure 1: Marginal effects of gender by age on the evaluation of winning chances with 95% confidence intervals.


Two findings are quite surprising. First, involvement in a local parliament does not excerpt a significant effect on the likelihood to hold a viable list placement.[3] Previous engagement at lower political levels does not make a prospective candidate more or less attractive to party officials and/or the electorate. Local political processes might be perceived as very different from national ones and thus not as a pivotal experience for prospective candidates. Second, contrary to my expectation, longer membership does not go hand in hand with a favorable list placement. Motivated party members show their ambition early on and secure the support of party gatekeepers shortly after joining the party. Duteous members of many years, instead, might agree to fill lower list positions without frank interests in entering the legislative.

This short blogpost indicates that party gate keepers have a clear vision, of who should run on the most prominent positions of party lists: incumbents, new party members and men. It would be interesting to know why gatekeepers look for these characteristics in prospective candidates rather than for experience in form of either familiarity with political processes at a different administrative level or with the functioning of political parties gained through a long membership therein. Future research could thus investigate how selections committees evaluate aspirants and what they think voters value most.

Sarah C. Dingler in July 2018


[1] The most comment requirement entails that candidates need be a paying party member for at least one year. Some exceptions include some Bangladesh parties (no membership required), the Finish Social democrats in the early 2000s (four months before the primaries) or Italian and Belgium parties (minimum of five years). Others parties impose a certain deposit (e.g. Canadian Conservative demanded a $1000 deposit in 2009) or have an age limit for prospective candidates (e.g. not older than 70 years). For more details see Hazan and Rahat (2010).

[2] Viability is determined based on the candidate’s list position in relation to the number of seats won by the party during the previous election. As a result, candidates with a list position below the predicted seats can be considered viable.

[3] This effect remains stable when including a lagged variable measuring experience at the local level.



Beck, T., Clarke, G. R., Groff, A., Keefer, P. & Walsh, P. (2001) New Tools and New Tests in Comparative Political Economy. The Database of Political Institutions. World Bank Economic Review, 15, 165-176.

Campbell, R. & Cowley, P. (2014) What Voters Want: Reactions to Candidate Characteristics in a Survey Experiment. Political Studies, 62, 745-765.

CCS (2016) Comparative Candidates Survey Module I – 2005-2013 [Dataset – cumulative file]. Lausanne, FORS.

Comparative Study of Electoral Systems (2016) CSES Module 4: 2011-2016.

Gallagher, M. & Marsh, M. (Eds.) (1988) Candidate Selection in Comparative Perspective: The Secret Garden of Politics, London, Sage.

Hazan, R. Y. & Rahat, G. (2010) Democracy within parties: Candidate selection methods and their political consequences, Oxford, Oxford University Press.

International Institute for Democracy and Electoral Assistance (2015) Global
Database of Quotas for Women. IDEA International.

World Bank (2016) Labor Force Participation Rate, Female (% of Female Population Ages 15+) (Modeled Ilo Estimate).


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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