An analysis of the language in social media messages posted via the Internet determined that, on average, the use of ...

SamA on January 3, 2020

Hello @farnoushsalimian,

The stimulus gives us the average usage of positive words at different points in the day. Positive word usage is common in the morning, decreases gradually in the afternoon, and increases sharply in the evening.

Based on these averages, the conclusion claims to track the mood of a typical person throughout the day. What is the problem with this?

I was able to eliminate A and D pretty quickly, leaving me with B, C, and E. Let's discuss those answer choices.

B. At first glance this seems like a reasonable criticism. However, the fact that "many" people do not use these positive words would not have an effect on the given averages. We still have the same trends throughout the day that the argument is based upon.

C. Are other methods of communication a better indicator of the typical person's mood than social media? We do not know, so we cannot say that C is a problem. There is no reason why word use in social media would have to correlate with word use in other communication forms.

E. This is saying that the people using positive language in the evening are usually not the same people using positive language in the morning. This tells us that, although the averages may be valid, they do not represent the mood shifts of a particular person. We cannot conclude that the typical person wakes up happy, becomes sad in the afternoon, and then feels better in the evening. What if some are morning people and others are night owls? This is the alternative explanation that E suggests.

Abigail-Okereke on March 7, 2023

Hello,
Thank you for your explanation. I now understand why E is correct. However, I interpreted the flaw to be that just b/c someone posts positive things on social media doesn't mean they are happy in real life. So, I was looking for something similar to perception vs. reality. What type of flaw is showcased in answer choice E?

Emil-Kunkin on March 12, 2023

Hi, I think the flaw you noted is valid as well, and as is often the case, this is an argument with more than one flaw in it. The flaw you pointed out is one of data representation, in which the data may not actually reflect the underlying scenario.

The flaw that E pointed out is one of an inconsistent sample. It's possible that the group of morning posters and evening posters has no overlap, but the author assumes that both groups are the same people.