Do Uncontested Elections Have Worse Outcomes?

GraVandius

GraVandius

Retired Troll
Pronouns
His Majesty
1568500806077.png

Do Uncontested Elections Have Worse Outcomes?
​

Following the debacles that lead to Rach's resignation for President the idea came into my mind that the region is more critical of candidates who win in uncontested elections than those with opponents. The easiest position to test this idea with was the most prominent, the presidency. I had already started another study on the Presidency, in which I documented all the Presidential elections I could find in the Oval Room (Stored in the very cool ERI Data Bank). Following that, I contacted HEM about getting access to the raw data from his excellent What Does It Mean To Be A 'Popular' President, And Who Are The Popular Ones? piece in the ENN. HEM was kind enough to grant the access and after about two weeks of work, here are the results.

​

This graph compares the average Approval ratings of the nine totally uncontested (Candidate vs Abstain or ROE) Presidential elections from what I believe was the first one in July 2007 to Rach in the most recent election, to the total average from the aforementioned ENN piece. As you can see the average for totally uncontested elections, is significantly lower than the total average with a difference of 8.29%. This would also notably put the Uncontested average in the unpopular range of HEM's study as it is just slightly below the 58.25% that marked the bottom quartile of his data.

This would seem to suggest that the region is in fact more critical of the winners of uncontested elections. There is another option for this result however. The group contains some of the notable low polling performers Pez201 and Aexnidaral according to HEM's study, ranking last and 5th from last respectively. Since one could assume bad polling means a bad term the other option could be that there is simply some correlation between bad presidential terms and uncontested elections. This correlation could exist perhaps due to a lack of an opposition on ones behind, which could lead potentially to complacency.

Regardless of which of these two theories is true it seems that it might be better for cannidates in the upcoming Presidential Election to have an opponent rather than coast into a coronation.
 
Last edited:
I don't know much about journalism but I think this is a great example of analytical reporting. Great topic.
 
Does the total average include the uncontested elections as well?
 
Izzy said:
I don't know much about journalism but I think this is a great example of analytical reporting. Great topic.
Click to expand...
Thank You Izzy.
Sopo said:
Does the total average include the uncontested elections as well?
Click to expand...
Yes it does, except for Rach's most recent term, as that is HEM's average from his article.

 
GraVandius said:
Izzy said:
I don't know much about journalism but I think this is a great example of analytical reporting. Great topic.
Click to expand...
Thank You Izzy.
Sopo said:
Does the total average include the uncontested elections as well?
Click to expand...
Yes it does, except for Rach's most recent term, as that is HEM's average from his article.
Click to expand...
I'd say it would be more useful to compare uncontested elections to contested elections (rather than total elections) as it would likely help your overall point with a starker contrast.
 
Some thoughts to expand on this:

- As Sopo said, comparing Uncontested vs. Contested.

- Separate Contested elections into bins - for example, "serious challenger" versus "not serious challenger." A distinction can be made by looking at whether opposition had served in any elected office in the past (CA Chair, Senate, President), or had been a Minister.

There is some surprise that uncontested candidates are less popular than contested candidates. I would expect that if you are a 'strong' enough candidate to run without opposition, it's because you'd steamroll anyone - probably because you are perceived as a high quality candidate. Why should we then see lower than average outcomes?
Could this be a polling issue? Uncontested campaigns are highly correlated with low political activity, so you may have discontent from citizens on that dimension which bleeds into the approval rating.

Edit: I missed HEM's " What Does It Mean To Be A 'Popular' President, And Who Are The Popular Ones?", so thank you for linking to that. Amusing that I was the median of all presidents ever. An Average Man. :p
 
PhDre said:
- As Sopo said, comparing Uncontested vs. Contested.
Click to expand...
​
Ask and ye shall receive. The difference is roughly only a percentage point more than my previous calculation, with the Contested average being 66.80% rather than the total average's 65.93%. The fact that this rise is not that large makes sense when you consider that uncontested elections make up only about ~14% of our over 60 elections.

PhDre said:
Separate Contested elections into bins - for example, "serious challenger" versus "not serious challenger." A distinction can be made by looking at whether opposition had served in any elected office in the past (CA Chair, Senate, President), or had been a Minister.
Click to expand...
I had actually attempted to do this with a different study attempting to classify Europeia's presidential elections into different categories. After collecting the data and links for all the presidential elections I basically gave up as the amount of time collecting the rest of that information, Senate, Minister ect., would take a ridiculously long amount of time for the entirety of Europeia's history. The legislative records in particular only go back to the 28th Senate and finding ministers would be a immense amount of time digging through the Conference Hall. The best way would probably to have a group of people who were active in a respective era and eyeball the candidates for this information.
 
That would be an _insane_ amount of information to go through. Maybe a very long-term project if you're motivated.
 
If someone wants to create a data set that anyone with statistical tools (free to download) could run...

Each row would be a unique candidate-election observations.

Columns:
user name
election date
election type
term running for
previous experience field
votes received
election outcome (win or lose)
iterate votes received and outcome for runoffs or as a separate table

You could run all kinds of goodies if you had that.
 
hyanygo said:
Do a test for statistical significance.
Click to expand...
As this might be relevant in the coming election, I've attempted to run a test for statistical significance on excel and I got the following:
t Stat-1.39823679
P(T<=t) one-tail0.091900221
t Critical one-tail1.761310136
P(T<=t) two-tail0.183800442
t Critical two-tail2.144786688

Which according to the following description on the internet, I believe means it is not statistically significant?
Conclusion: We do a two-tail test (inequality). lf t Stat < -t Critical two-tail or t Stat > t Critical two-tail, we reject the null hypothesis. This is not the case, -2.365 < 1.473 < 2.365. Therefore, we do not reject the null hypothesis. The observed difference between the sample means (33 - 24.8) is not convincing enough to say that the average number of study hours between female and male students differ significantly.
Click to expand...
As -1.39 (t stat) is not less than -2.144 (Critical two tail) and -1.39(t stat) is not greater than 1.76(Critical one-tail)
 
You must log in or register to reply here.
Back
Top