For this reason, I narrowed my search to small donor behavior in primary races..Most, to not say all, of these races are more expensive than primaries and are between a Democrat and a Republican, which provides the candidates with access to their political party's resources and networks of big donors.On a primary race, however, challengers rely on all the money they can get and may not underestimate the importance of small donations on their path to victory..Thus, I expect challengers to do better than incumbents in attracting small donations.Instrumental incentives:Instrumental incentives refers to the motivation to donate based on the ability of affecting the result of a race..If we assume the underlying instrumental motivation of donors to be the same in Congressional races as in primary races, there is no reason to think there will be no relationship between competitiveness and small donations.DataUsing data from the Florida Campaign Finance Database of the Florida Division of Elections, I created a database containing whether a candidate is an incumbent or not, a candidate's name, office, political party, vote share, and the aggregate total of donations of $200 or higher s/he received..Small donations by party.Small Donations and Material IncentivesAs mentioned above, in order to test if material incentives were a motivation for small donors to contribute to political campaigns during the 2018 Florida Primary, I compared small donations made to incumbents to those made to non-incumbents..Small donations by incumbent.At first glance, it is easy to notice that the difference on small donations between incumbents and non-incumbents is larger than that between political parties..Thus, material incentives do not seem to be an important motivation for small donors to contribute to political campaigns during the primaries.Small Donations and Instrumental IncentivesDonors with an interest to have a direct influence on a race are more likely to donate to candidates they consider have a real chance at winning the election..The code is the following:#Dividing each vote share by 100Small_donations2 <- mutate(Small_donations2, voteshare = voteshare/100)# Calculating the coefficient by racerace_comp_df = Small_donations2 %>% group_by(officecode) %>% summarise_at(vars(voteshare), funs(1/sum(voteshare*voteshare, na.rm = TRUE)))#Merging with the Small_donations2 data setSmall_donations2 <- merge(x = Small_donations2, y = race_comp_df, by = "officecode", all.x = TRUE)The least competitive race of the 2018 Florida Primary had a coefficient of 1.282 and it was for State Senate District 4 between the Republicans Aaron Bean, who received 87.4% of the votes, and Carlos Slay, who received 12.6% of the votes..The most competitive race had a coefficient of 3.483 and it was for State House District 115 between the Republicans Vance Aloupis, who received 34.1% of the votes, Jose Fernandez, who received 29.1% of the votes, Rhonda Lopez, who received 2.8% of the votes, and Carlos Gobel, who received .8% of the vote.Race Competitiveness by PartyFigure 4 shows the difference on race competitiveness coefficients by party.Figure 4..On average, Democratic and Republican races did not differ on closeness.Race competitiveness by incumbencyFigure 5 shows the difference on race competitiveness coefficients by incumbency.Figure 5..Race competitiveness by incumbencyThe mean competitiveness coefficient for races with non-incumbents is 2.21 (SD = .57), whereas the same figure is 1.80 (SD = .34) for races with incumbents..On average, races with incumbents were neither closer nor had a bigger margin than race with no incumbents.Small Donations and Race CompetitivenessIf people are instrumentally motivated to donate to political campaigns there should be a positive relationship between small donations and closeness..In other words, the more competitive a race is the more small donations it should attract.. More details