Is Google biased against conservatives? The short answer is not likely.
Based on our data-driven study, we found that 63% of Google’s search results were non-partisan. While 32% of the results came from left-leaning publications versus 5% of results coming from right-leaning websites.
Now you may see 32% vs 5% and think that Google is biased.
But here’s the truth:
The lopsided results are because of Search Engine Optimization (SEO), not because Google’s algorithm is biased.
Let’s dive into the case study.
Google Liberal Bias: Does It Exist?
63% of Results are Non-Partisan
While 63% is almost a failing score in school, it’s impressive that an algorithm can be so effective.
316 of the results were from non-partisan and unbiased websites. Websites that present facts. Not opinions.
This is super important for politically-driven topics where there tends to be lots of misinformation or “fake news”.
32% Left vs 5% Right
While Google’s algorithm presents unbiased information for the majority of the results, it isn’t perfect. As you can see, left-leaning publications received more coverage and visibility for key political topics.
In a perfect world, there would be equal representation for both parties.
Unfortunately, this isn’t the case.
Now you might be wondering why? Why are there more left-leaning websites showing up for key political topics?
Well, it’s not for the reasons that you probably think.
More on this in second, but it has to do with Search Engine Optimization (SEO).
But before we get into that, let’s dive into some of these topics a little.
Top 3 Left-Leaning Topics
Topics that had the most left-leaning websites present:
- “black lives matter” = 9/10
- “border wall” = 7/10. This one is almost inline with Gallup poll data. 60% of American’s oppose “significantly expanding the construction of walls along the U.S.-Mexico border”.
- “dreamers” = 6/10
Top 3 Right-Leaning Topics
Topics that had the most right-leaning websites present:
- “blue lives matter” = 4/10
- “second amendment” = 3/10
- “pro life” = 3/10
Top 3 Unbiased Topics
The topics below had the highest number of non-partisan results:
- “electoral college” = 10/10
- “gold standard” = 10/10
- “roe vs wade” = 9/10
Google’s algorithm was effective for these topics. It had almost a 100% success rate in showing non-partisan results for these topics.
Top 3 Topics With Fair Representation
“second amendment” = Left: 3, Center: 4, Right: 3
“pro life” = Left: 3, Center: 4, Right: 3. According to Gallup Poll, 46% of Americans are “pro choice” while 49% are “pro life”, so this was pretty accurate.
“gun control” = Left: 2, Center: 7, Right: 1. The first two results were fair, but after that there weren’t any other topics that had “fair” representation. “Gun control” was the closest.
So at this point, it’s clear that right-leaning websites get less visibility and coverage for key political topics.
But the question is: why?
Why Do Google’s Results Seem Biased (from an SEO Perspective)?
First, let me quickly cover how SEO works. Google uses over 200 different ranking factors for determining what pages should show up any given keyword. These factors are NOT created equal.
But in most cases, pages that perform well in Google have the following characteristics:
1. They are targeting a keyword phrase like “black lives matter” or “gun control”.
2. The page provides detailed information about the topic. This is referred to SEO content in my world. More often than not, pages that have bigger word counts tend to perform better.
3. The website and target page have backlinks. A “backlink” is created when another website links to your website.
Google uses backlinks as “votes” for a website. The more high-quality backlinks a website has, the better it will often perform. Out of all of the 200 ranking factors, backlinks are one of the strongest by far.
So now let’s look at the data.
Left-Leaning Websites Are Stronger
On average left-leaning websites in this study were substantially stronger than right-leaning websites. By “stronger”, I’m speaking in terms of website authority driven by the quantity of high-quality backlinks.
Here’s the data pulled from Ahrefs:
Left-leaning websites also have substantially more .gov and .edu backlinks.
So without getting too deep into SEO, the primary reason why there are more left-leaning websites outperforming right-leaning websites is because of the volume of backlinks and overall website authority.
But this raises some other questions:
Why do left-leaning websites get more backlinks?
There are a few possibilities:
1. Left-leaning websites are more likely to engage in active link building. “Active” link building is when a company pays a company like mine to acquire backlinks on other websites.
2. There is a greater pool of left-leaning websites capable of linking compared to right-leaning. According to data in this article, the tech industry (who have very powerful websites) leans left.
3. .edu and .gov websites tend to be more left-leaning and therefore or more likely to link to left-leaning sources of information.
These ideas are difficult to quantify, but there’s worth exploring further.
To Conclude the Google Bias Project
So back to our original question:
Is Google biased? Not likely.
Google’s results were surprisingly non-partisan (61%) for key political topics. It’s not perfect and the results lean left, but it’s likely not because Google’s algorithms are biased.
It’s because those left-leaning websites are simply better at SEO.
Now there is one thing that can’t be quantified, but it could make Google results biased in some cases. And that’s the fact that Google’s algorithm is designed and operated by humans. Humans are inherently biased in one way or another.
You must also consider that Google uses over 10,000 humans for manually reviewing the quality of their search engine results. It’s entirely possible that these humans could suppress pages they don’t agree with.
Bias is inevitable when humans are involved.
If you’d like to see all the data we used for this case study go here (no opt-in required).
How to We Conducted the Google Bias Project
The way we conducted this case study was simple:
1. We found over 300 “controversial” topics using sites like ProCon.org
2. We broke that list down and focused on 50 politically-driven topics (see them here). Why? Because political bias is easier to identify.
3. We scraped Google’s SERPs (Search Engine Results Pages) for each topic using the Data Miner plugin for Chrome
4. We then built a list of websites that were frequently showing up for these key political topics.
6. Then we went through the top 10 for each topic and marked each result as left, center, right, or unknown. This was based on the website’s known political leanings. Not based on the content of the individual page. We made this decision because categorizing each individual result would make the process too subjective on our end.
We analyzed a total of 500 results (50 keywords x 10 first page spots on Google).
Now Your Turn
Were you surprised by this data? Have you seen bias in Google search results? Let me know below.