You may have read how different studies are entirely based on how much time people spend on social media platforms. The researchers then use the similar data in presenting their theories or conclusions. But what if we tell you that according to a research from Facebook, people have always been pretty bad at telling how much time they spend on any platform across the internet?
Let’s suppose if you were about to conduct a study related to how social media usage potentially affects or get affected by mood, you would take help from the self-reported statistics in both scenarios.
As there is practically no way introduced to measure the mood of any social media user, therefore you are bound to rely on the answers the participants give. You would also naturally assume that people really well know how much time they spend on Facebook, Instagram or Twitter for that matter. But that perception is just like deceiving yourself.
In fact, the researchers always know that the self-reported numbers are bound to have errors and some studies have also proved that. But now this new study from the world’s biggest social media company, Facebook shows that when you compare the self-reporting data with actual server logs, there is a massive disconnection and hence you cannot rely on any numbers provided through such sources for serious scientific work.
When the responses given by particular people were compared with the internal data provided by Facebook, the results showed that majority of people have always overestimated the number of times they have opened the app or used the platform on PC. Take a look at this chart for a more detailed example.
As you can clearly see above, very few people were actually spending more than three hours on the site every day, whereas the majority of them only spent one hour, despite telling different answers. There was also a massive mismatch in the number of logins as many people thought that they open the app 10 times a day or more, whereas the server data showed that it was really very common. Younger people gave more wrong answers which is a sign of big worry because most studies keep their demographics around them.
Nevertheless, all of this doesn’t mean that people don’t use Facebook a lot. Sometimes they even use the app more than their expectations. But this data is entirely focused on giving direction to different studies that rely on self-reported data - which is not a good practice.
As a solution, Facebook suggests that researchers should not incorporate these values in their work directly but instead interpret the self-reported time given by people as merely an estimate in which they can depend on the distribution relative to the respondents rather than absolute measures.
If exact online times are necessary then a tracking app or collaborating with Facebook for the similar purpose can be a great idea!
Read next: Facebook’s Instagram Will Soon Require Suspicious Users to Submit Their Government-Issued ID for Verification
Let’s suppose if you were about to conduct a study related to how social media usage potentially affects or get affected by mood, you would take help from the self-reported statistics in both scenarios.
As there is practically no way introduced to measure the mood of any social media user, therefore you are bound to rely on the answers the participants give. You would also naturally assume that people really well know how much time they spend on Facebook, Instagram or Twitter for that matter. But that perception is just like deceiving yourself.
In fact, the researchers always know that the self-reported numbers are bound to have errors and some studies have also proved that. But now this new study from the world’s biggest social media company, Facebook shows that when you compare the self-reporting data with actual server logs, there is a massive disconnection and hence you cannot rely on any numbers provided through such sources for serious scientific work.
When the responses given by particular people were compared with the internal data provided by Facebook, the results showed that majority of people have always overestimated the number of times they have opened the app or used the platform on PC. Take a look at this chart for a more detailed example.
As you can clearly see above, very few people were actually spending more than three hours on the site every day, whereas the majority of them only spent one hour, despite telling different answers. There was also a massive mismatch in the number of logins as many people thought that they open the app 10 times a day or more, whereas the server data showed that it was really very common. Younger people gave more wrong answers which is a sign of big worry because most studies keep their demographics around them.
Nevertheless, all of this doesn’t mean that people don’t use Facebook a lot. Sometimes they even use the app more than their expectations. But this data is entirely focused on giving direction to different studies that rely on self-reported data - which is not a good practice.
As a solution, Facebook suggests that researchers should not incorporate these values in their work directly but instead interpret the self-reported time given by people as merely an estimate in which they can depend on the distribution relative to the respondents rather than absolute measures.
If exact online times are necessary then a tracking app or collaborating with Facebook for the similar purpose can be a great idea!
Read next: Facebook’s Instagram Will Soon Require Suspicious Users to Submit Their Government-Issued ID for Verification
