We used a Bayesian model to see how winning an election affects longevity. The results suggest that winners live about 8.6 years longer on average compared to losers. The model also showed significant effects from other factors like win_margin and party.
Second Call to brm()
Family: gaussian
Links: mu = identity; sigma = identity
Formula: death_age ~ treatment + win_margin + party
Data: x (Number of observations: 254)
Draws: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
total post-warmup draws = 4000
Regression Coefficients:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
Intercept 73.60 1.79 69.99 77.02 1.00 2594 2716
treatmentwin 8.22 2.81 2.63 13.82 1.00 2592 2712
win_margin -1.38 0.52 -2.38 -0.40 1.00 2594 2500
partyRepublican 3.95 1.45 1.14 6.87 1.00 3685 2693
partyThirdparty -9.49 7.96 -24.89 6.14 1.00 4028 2930
Further Distributional Parameters:
Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sigma 11.28 0.51 10.34 12.36 1.00 4152 2658
Draws were sampled using sampling(NUTS). For each parameter, Bulk_ESS
and Tail_ESS are effective sample size measures, and Rhat is the potential
scale reduction factor on split chains (at convergence, Rhat = 1).
Using 10 posterior draws for ppc type 'dens_overlay' by default.
Using data about US governor candidates from the years 1945 - 2012, we seek to find the relationship between the longevity of Preceptor David Kane and whether or not he wins the Mayoral Election in Newton, MA. Modern medicine has increased the overall lifespan of all candidates regardless of if they won or lost. We modeled age of death as a sum of election result (won/lost), age during election, political party and the win margin. We expect Preceptor to live an extra 8 years (plus/minus 6 years) if he were to win the Mayorial election.
Warning in tidy.brmsfit(x, ..., effects = "fixed"): some parameter names
contain underscores: term naming may be unreliable!
Characteristic
Beta
95% CI1
treatment
treatmentwin
8.2
2.6, 14
win_margin
-1.4
-2.4, -0.40
party
partyRepublican
4.0
1.1, 6.9
partyThirdparty
-9.5
-25, 6.1
1 CI = Credible Interval
The candidates who won for mayor are predicted to live 8.2 years longer than if they did not win. Candidates of the republican party are predicted to live 4 years longer and candidates of the third party are predicted to live 9.5 years less if not elected.
variables: win_margin, treatment, party
values:
Subject: lifespan after mayor election
Preceptor table:
Columns: party: democrat treatment: win or lose win_margin: 0
Outcome: Years lived after won or years lived after lost election (age at death)
Treatment: whether or not the candidate won the election
Covariates: year of election, age of mayor
model type: predictive
Rows:
Units: mayor elections in various counties (candidates for mayor)
EDA: glimpse(governors)
Validity: validity may not hold because the age of the candidates may not be accurate such as the year they were born which can affect the result. Different elections used in the same dataset may also affect the outcome.
Population Table:
Stability: may not hold because the average lifespan changes over time and could be greater than it used to be. Healthcare has also changed over time.
Representativeness: may not hold because there are age restrictions (and possibly health guidelines) for specific elections. Different types of elections from across the country are also needed
Unconfoundedness: may not hold because the candidates could have differing health situations and financial situations to be able to afford proper healthcare
