Facebook Mood Experiment Sebastian Caldas Slides based on those by - - PowerPoint PPT Presentation

facebook mood experiment
SMART_READER_LITE
LIVE PREVIEW

Facebook Mood Experiment Sebastian Caldas Slides based on those by - - PowerPoint PPT Presentation

Facebook Mood Experiment Sebastian Caldas Slides based on those by Liam Li and Nicholay Topin (Fall 2018) Related Article: Everything We Know About Facebook's Secret Mood Manipulation Experiment | It was probably legal. But was it


slide-1
SLIDE 1

Sebastian Caldas

Slides based on those by Liam Li and Nicholay Topin (Fall 2018) Related Article: “Everything We Know About Facebook's Secret Mood Manipulation Experiment | It was probably legal. But was it ethical?” by Robinson Meyer

Facebook Mood Experiment

slide-2
SLIDE 2

Background

slide-3
SLIDE 3

Facebook modified users’ newsfeeds to see how it impacted their moods.

3

slide-4
SLIDE 4

Facebook modified users’ newsfeeds to see how it impacted their moods.

4

slide-5
SLIDE 5

Facebook modified users’ newsfeeds to see how it impacted their moods.

5

slide-6
SLIDE 6

Timeline

slide-7
SLIDE 7

2012

  • Facebook gathers data
  • Cornell IRB approves analysis

2013

  • Paper written by FB / Cornell researchers
  • Submitted to PNAS

2014

  • Published in PNAS
  • Media reaction and users’ outrage (June 2014)

7

slide-8
SLIDE 8

2012

  • Facebook gathers data
  • Cornell IRB approves analysis

2013

  • Paper written by FB / Cornell researchers
  • Submitted to PNAS

2014

  • Published in PNAS
  • Media reaction and users’ outrage (June 2014)

8

slide-9
SLIDE 9

2012

  • Facebook gathers data
  • Cornell IRB approves analysis

2013

  • Paper written by FB / Cornell researchers
  • Submitted to PNAS

2014

  • Paper published in PNAS
  • Media reaction and users’ outrage (June 2014)

9

slide-10
SLIDE 10

Detractors vs. Proponents

slide-11
SLIDE 11

Facebook manipulated its users’ moods

Detractors present the following concerns:

Users were unable to provide informed consent The IBR approval was conditioned

  • n Facebook’s

approval

11

slide-12
SLIDE 12

Users were unable to provide informed consent

Detractors present the following concerns:

Facebook manipulated its users’ mood Facebook manipulated its users’ mood Previous emotional contagion studies had been limited to crawling news feeds, without modifying them. The IBR approval was conditioned

  • n Facebook’s

approval

12

slide-13
SLIDE 13

The IBR approval was conditioned

  • n Facebook’s

approval

Detractors present the following concerns:

Facebook manipulated its users’ mood The IBR approval was conditioned

  • n Facebook’s

Users were unable to provide informed consent

  • According to the APA,

informed consent should be

  • btained using reasonably

understandable language.

  • Participants should be able

to withdraw their data.

13

slide-14
SLIDE 14

Users were unable to provide informed consent

Detractors present the following concerns:

Facebook manipulated its users’ mood Users were unable to provide informed consent The IBR approval was conditioned

  • n Facebook’s

approval the researchers’ “revision letter said they had Cornell IRB approval as a ‘pre-existing dataset’ presumably from FB, who seems to have reviewed it as well in some unspecified way...” - Dr. Fiske

14

slide-15
SLIDE 15

Meanwhile, proponents present the following arguments:

The use of the data is legal under Facebook’s terms of service The experiment helps Facebook improve its service The impact on users was small and went unnoticed

slide-16
SLIDE 16

Meanwhile, proponents present the following arguments:

The use of the data is legal under Facebook’s terms of service The experiment helps Facebook improve its service The impact on users was small and went unnoticed The use of the data is legal under Facebook’s terms of service “Facebook users relinquish the use of their data for data analysis, testing, and research.” - Facebook’s TOS

slide-17
SLIDE 17

Meanwhile, proponents present the following arguments:

The use of the data is legal under Facebook’s terms of service The experiment helps Facebook improve its service The impact on users was small and went unnoticed The experiment helps Facebook improve its service “We do research to … make the content people see on Facebook as relevant and engaging as possible. A big part

  • f this is understanding how

people respond to different types of content.” - Facebook

slide-18
SLIDE 18

Meanwhile, proponents present the following arguments:

The use of the data is legal under Facebook’s terms of service The experiment helps Facebook improve its service The impact on users was small and went unnoticed The impact on users was small and went unnoticed

slide-19
SLIDE 19

Discussion

slide-20
SLIDE 20

Do you believe something unethical was done?

20

slide-21
SLIDE 21

Was the data collection justified?

21

slide-22
SLIDE 22

Should researchers use data collected in similar manners?

22

slide-23
SLIDE 23

Why is informed consent necessary in this situation and not for for A/B testing on ads?

23

slide-24
SLIDE 24

My opinion

slide-25
SLIDE 25

Do you believe something unethical was done?

  • If ethics are a social construct, then the public outcry

may mean something unethical was indeed done.

  • The root of the problem seems to be the perceived

potential Facebook has to modify our behavior.

  • Opting out is a must in this type of experiments.

25

slide-26
SLIDE 26

Was the data collection justified?

  • The motivation for the study is vague.
  • Facebook may be underplaying the results.
  • Although emotional contagion studies are common,

they had never been performed at this scale.

26

slide-27
SLIDE 27

Should researchers use data collected in similar manners?

  • I don’t know.
  • On one hand, the data may already be there. But do

we want to encourage similar endeavours?

27

slide-28
SLIDE 28

Why is informed consent necessary in this situation and not for for A/B testing on ads?

  • Reactions to ads and emotional responses are

different.

28