Scribd
Upload
53 views

EAPP Technical Report

This study analyzed differences in Facebook algorithmic recommendations between genders by surveying 30 of the researcher's Facebook friends. Posts were categorized and percentages calculated. Results showed that men are more likely to be recommended neutral, showbiz, political and recreational posts, while women are more likely to see fashion, cosmetic, nature, moral and food posts. Most stereotypes were affirmed, though some differences were seen. The study concluded the algorithm reflects gender tendencies but interests are becoming less defined by gender.

Uploaded by

Semi Sen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
53 views

EAPP Technical Report

This study analyzed differences in Facebook algorithmic recommendations between genders by surveying 30 of the researcher's Facebook friends. Posts were categorized and percentages calculated. Results showed that men are more likely to be recommended neutral, showbiz, political and recreational posts, while women are more likely to see fashion, cosmetic, nature, moral and food posts. Most stereotypes were affirmed, though some differences were seen. The study concluded the algorithm reflects gender tendencies but interests are becoming less defined by gender.

Uploaded by

Semi Sen
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Republic of the Philippines

DEPARTMENT OF EDUCATION
Division of Eastern Samar
DOLORES NATIONAL HIGH SCHOOL
Dolores E. Samar

A Categorical Data Analysis on Facebook’s User Activity and the Algorithmic

Differences between Genders

A Survey Report

In partial fulfilment of the requirements in

English for Academic and Professional Purposes

Mark Lawrence P. Lacdao

12 – Aphrodite

Mr. Igour O. Pomarca

Subject Teacher
I. INTRODUCTION

Facebook has long been dominating the social media sphere when compared to

other services like Twitter, Youtube, Twitch, and others not owned by Meta. But, as the

site’s user base ages it becomes harder for services to retain the same amount of

activity way back when. Meta, on the other hand has a different type of trend as data

taken from statista.com showed the distribution of Facebook users worldwide as of

January 2022, and, age groups at 13-17 only made up 5.6%, 18-24 made up 22.6%,

and with the highest between age groups of 25-34 making up 31%. As the site’s user

base grows older, the trend shows more upwards activity and longer times spent using

and interacting with the site.

The longer you interact with the site, the more likely it is for it use data gathered

from user interaction to compile recommendations in a form of an algorithm that will,

hopefully, have the user allocate more of their time using the app. The study aims to

find the differences between algorithmic recommendations in males and females and

whether or not these recommendations conform to already known stereotypes, such as

men liking sports more than women, or women having more interest in subjects that

feature food, cosmetics, or fashion.

II. METHODOLOGY

Respondents. This study used the purposive random sampling method where 30

respondents were chosen from the researcher’s Facebook friends, both male and

female as illustrated by the table below.


Respondents Amount Percentage
Male 13 43.3%
Female 17 56.7%
Other/NI 0 0
TOTAL 30 100%

Data Gathering Procedure. Data was collected by surveying the existing Facebook

posts of the respondents and tallying the data in the given categories.

Data Analysis. The data gathered was classified into 10 categories: (1) Neutral/Can’t

be determined, (2) Political, (3) Moral/Religious/Philosophical, (4) Showbiz related, (5)

Sports related, (6) Recreational, (7) Nature related, (8) Financial/Business/Livelihood,

(9) Cosmetic/Fashion, (10) Food/Culinary and were further calculated into percentages

in order to determine the statistical configuration and percentage of the number of posts

in relation the respondents’ gender.

III. RESULTS

The researcher conducted a survey with 30 respondents that had a grand total of

241 posts, and in accordance to the pie chart below, results show that 41.90% are

Neutral; 9.13% are Political; 6.22% are Moral; 17.01% are Showbiz related; 6.64% are

Sports related; 2.49% are Recreational; 6.22% are Nature related; 5.80% are Finance

or Business related; 1.24% are Cosmetic; and lastly 3.32% are Food and/or Culinary.
Percentage per Category
Finance; 5.80%
Cosmetic; 1.24% Culinary; 3.32%
Nature; 6.22%
Recreational;
2.49%
Neutral;
Sports; 6.64% 41.90%

Showbiz; 17.01%

Moral; 6.22% Political; 9.13%

Further dividing the 30 respondents by gender and category we get:

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Neutral Political Moral Showbiz Sports Recreational Nature Financial Fashion Food

Male Female Other/NI

IV. CONCLUSION

The algorithm that Facebook uses, when put laymen terms, the user’s likes,

posts, friend’s posts, anything generally related to your account are used to compile
data for the user’s recommendations. Using the results at hand, we can conclude that,

Men in Facebook are more inclined to be recommended posts that are Neutral in

nature, posts that are related to showbiz, and posts that are heavily political and

recreational; Women, on the other hand, are more likely to be recommended posts that

are Fashion and Cosmetics related, post that feature Nature and Moral posts (i.e.,

inspirational quotes, passages, etc.), and posts featuring images or recipes of food,

and, surprisingly the percentage is higher for Women to be recommended posts related

to Sports than to men.

Most like stereotypes were affirmed from the data gathered through the survey,

but, some had been derailed by both men and women. This also supports the possibility

that subjects dominated by either gender can be enjoyed by both, but, only if the trend

shown here continues and it becomes more socially acceptable to like subjects liked by

either gender. In conclusion, the survey report was successful in its aim to differentiate

between algorithmic recommendations between males and females, and proved that it

is possible to for either male-dominated or female-dominated subjects to be socially

acceptable to be liked by the opposite gender.

You might also like