Why this study that confronts Android and iPhone is biased

Why this study that confronts Android and iPhone is biased

PC vs. Mac. Nintendo vs. Sega. PlayStation versus Xbox. Android vs. iPhone. Tech showdowns exist and they always attract huge audiences. If you’ve come across this card, don’t be fooled.

The comparison between Android and iPhone has been the subject of debate for a long time, about which is better, which is more widespread, etc. Thus, we can compare the market shares of the operating systems thanks to various studies and observations, statistics extracted from the Internet or even the figures of the mobile operators.

Recently, a card was widely shared on social media, following a Demotivateur post.

These maps (…) are the first to show with great precision which Android or Apple is the most widespread in each country of the world, continent by continent.

This is a study by Electronics Hub, details of which can be found here. It is not about the most popular operating system, but about which operating system is the most popular in the world.

To carry out this study, they used SentiStrength, an academic linguistic analysis tool. The study was conducted on Twitter, where 347,856 brand-related tweets from around the world about positive and negative sentiments were analyzed. SentiStrength assigns short texts a score from 1 to 5.

74 of the 142 countries surveyed prefer Android and 65 prefer Apple

The results obtained allowed them to create several maps of the world where they classify Android and iPhone according to people’s preferences.

Why you should be careful with this study

Twitter data has many qualities that attract researchers and other marketers. They are extraordinarily easy to collect. Furthermore, they are available in gigantic quantities. And, with a simple text limit, they’re easy to scan.

Because of these attractive qualities, several hundred academic studies have been published using Twitter data, including multiple attempts to predict disease outbreaks, election results, movie box office and stock market movements solely from the content of tweets.

We can use social media data for research, that doesn’t mean we should. Each social media platform presents its own bias among the users who choose to use it.

For example, Twitter users do not represent people who do not have access to the Internet. In France, Twitter has 16.1 million monthly users and 5.1 million daily users (Médiamétrie, 2022). 38.5% of these users are between 25 and 34 years old and 56.4% are men. In reality, Twitter does not represent the entire world at all. Especially since this study is subject to samples that depend on the vocabulary used, but also on the language in which you are tweeting.

Under these circumstances, any collection of tweets will be biased and inferences based on the analysis of these tweets will not match the characteristics of the population. In summary, if you want to study the elites and the channels of influence of the elites, Twitter is definitely a good candidate. This also includes a high proportion of technology products such as smartphones, especially premium phones.

A biased sample cannot be corrected by collecting more data, and these biases have important implications for research based on Twitter data. This suggests that Twitter data is not suitable for studies where representativeness is key, such as the Electronics Hub study topic.


To follow us, we invite you to download our application for Android and iOS. You can read our articles, archives and watch our latest YouTube videos.

#study #confronts #Android #iPhone #biased

Leave a Comment

Your email address will not be published.