Tahrir Data Project

Preliminary Descriptive Analysis

Discursive Context

• Pessimism and Utopianismsince Iran and Moldova

• 2011>Deen Freelon’s 4-part taxonomy

• The Hype and the Blowback

• Still an anecdotal debate

• Still: «Was it a Facebook Revolution?»

The Tahrir Data Sets

• TDS-a: protester media-use– 1200 interviews between Feb 24 & March 1– In-depth, “superficial” questionnaire

• TDS-b: coordinator media-use– 35 individuals, semi-structured, purposive sample

• TDS-c: Twitter and transnational networks– #jan25 tweets between Jan 21 & Feb 11– 675,715 tweets, 106,000 users, 27 languages

TDS-a: Protester Sample

• “Hidden population”• Predominantly:

–young, –well educated, –non-politcally active, –wired–men

Frequencies:media use

Blogs (123) Twitter (138) Radio (229) Email (285) Facebook (447)

SMS (492) Print (612) Phone (869) TV (978) Live Comm'n (989)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Used in Protests

How much did protesters use media?

How much did protesters use media?

Blogs (123) Twitter (138) Radio (229) Email (285) Facebook (447)

SMS (492) Print (612) Phone (869) TV (978) Live Comm'n (989)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Use in General Used in Protests

Email (285)

SMS (492)

Phone (869)

Radio (229)

Print (612)

Blogs (123)

Twitter (138)

TV (978)

Facebook (447)

-10% 0% 10% 20% 30% 40% 50% 60%

57.5%

16.1%

10.8%

8.1%

5.4%

3.8%

3.0%

1.7%

0.7%

Drops in Medias' General Use to Protest Use

Med

ia (#

med

ia u

sers

)

Email (285)

SMS (492)

Phone (869)

Radio (229)

Print (612)

Blogs (123)

Twitter (138)

TV (978)

Facebook (447)

-10% 0% 10% 20% 30% 40% 50% 60%

57.5%

16.1%

10.8%

8.1%

5.4%

3.8%

3.0%

1.7%

0.7%

Med

ia (#

med

ia u

sers

)

Email (285)

SMS (492)

Phone (869)

Radio (229)

Print (612)

Blogs (123)

Twitter (138)

TV (978)

Facebook (447)

-10% 0% 10% 20% 30% 40% 50% 60%

57.5%

16.1%

10.8%

8.1%

5.4%

3.8%

3.0%

1.7%

0.7%

Med

ia (#

med

ia u

sers

)

Real Time Media?Synchronous Media?Social Media?Suitability for Protest Activity

Logical Regressions:Out on first day &

first time protesting

Protested on Day 1Exp (b) P

Age 1.009 0.27Male 1.295 0.11Education 0.991 0.87Net at Home 1.297 0.21Net on Phone 1.091 0.54Uses Text Gen 1.092 0.56Uses FB Gen 1.443* 0.01Uses TW Gen 1.429* 0.05Uses Blogs Gen 1.197 0.4Uses Email Gen 0.818 0.31Uses Phone Gen 1.22 0.46Uses TV Gen 0.483** 0.01Uses Radio 1.089 0.56Uses Print 1.366* 0.04_cons 0.341* 0.04N 1029ll -654.933* p<0.05, ** p<0.01, *** p<0.001

Not 1st ProtestExp (b) P

Age 1.034*** 0Male 1.034 0.84Education 1.03 0.61Net at Home 1.589* 0.03Net on Phone 1.118 0.44Uses Text Gen 1.227 0.19Uses FB Gen 1.349* 0.04Uses TW Gen 1.467* 0.04Uses Blogs Gen 0.961 0.86Uses Email Gen 1.11 0.61Uses Phone Gen 0.673 0.13Uses TV Gen 0.763 0.34Uses Radio 0.723* 0.04Uses Print 1.781*** 0_cons 0.099*** 0N 1029ll -631.21* p<0.05, ** p<0.01, *** p<0.001

Frequencies:Media rankings and protester behaviour

Phone Face To Face Facebook Satellite TV Email Twitter Documentation SMS Radio Print0%

10%

20%

30%

40%

50%

60%

70%

80%Im

port

ant

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Impo

rtan

t

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Info

rmati

ve

Used

Used

Used

Used

Used

Used

Used

Used

Used

Used

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Moti

vatin

g

Important

Informative

Used

Motivating

Media Arranged by Overall Scores

Perc

ent o

f Tot

al P

ossi

ble

Scor

e

Media Rankings (1)

Phone (869)

Face

To Fa

ce (989)

Faceb

ook (447)

Satel

lite TV

(978)

Email

(285)

Twitter

(138)

Documen

tation (n

o defined

users)

SMS (

492)

Radio (2

29)

Print (6

12)0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

75%

53%

65%

25%

52%

42%

10%

17%

5%

81%

46%

63%

21%

55%

42%

9%16%

3%

36%

51%

84%

43%

29%33%

7%

37%

8%

58%

43%

81%

20%24%

76%

6%

17%

4%

ImportantInformativeUsedMotivating

Media (# media users)

% o

f top

sco

re p

ossi

ble

from

med

ia u

sers

Media Rankings (2)

Phone (869)

Face

To Fa

ce (989)

Faceb

ook (447)

Satel

lite TV

(978)

Email

(285)

Twitter

(138)

Documen

tation (n

o defined

users)

SMS (

492)

Radio (2

29)

Print (6

12)0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

75%

53%

65%

25%

52%

42%

10%

17%

5%

81%

46%

63%

21%

55%

42%

9%16%

3%

36%

51%

84%

43%

29%

33%

7%

37%

8%

58%

43%

81%

20%

24%

76%

6%

17%

4%

ImportantInformativeUsedMotivating

% o

f top

sco

re p

ossi

ble

from

med

ia u

sers

0%

10%

20%

30%

40%

50%

60%

70%

80%

52%

46%

32%

26%

11%

6%5% 4% 5%

3%

Media Ranking Aggregatessignificance by full sample of respondents

Media (# of media users)

% o

f pos

sibl

e sc

ore

0%

10%

20%

30%

40%

50%

60%

70%

80%

52%

46%

32%

26%

11%

6%5% 4% 5%

3%

62%

48%

73%

27%

40%

48%

8%

22%

5%

significance by full sample of respondents

significance by media user respondents

% o

f pos

sibl

e sc

ore

Key Protest Metrics

Reliability

• Blogs (>tw, live, fb, em)

• Exaggeration for Blogs and TW in content types and motivations

Documentation• Sources:

TV, Live, Phone, FB, Print • Users

FB, live TW, phone, sat

.00

1.00

2.00

3.00

4.00

Media (total responses/total media users)

Relay

• Facebook and TV strongest showers

• Both as info types and activities

• The strange showing of live communication

Implications of TDS-a

Digital distinguish themselves through:• Degreee and character of media use • Behavior of media users• User relationships to media

TDS-c: Tweeters

1a. 1b.

Power law distributions for tweeters and retweeters.

TDS-c: Transnational communication flows

TDS b: Coordinator Strategies

• Traditional media & Hybridity• Strong ties and weak ties• Transnational Information Flows as

–Security–Motivation–Principle

Preliminary Conclusions

• Digital media are distinct• Digital media vs hybrid media• Context and contingency• We need more sophisticated models

– Utopianism and pessimism do not suffice– Information ecologies– Functional network mappings

A Facebook Revolution?

Silly Question

Moving Ahead

• Implications for further research– Grounded research– Local research– Open research – Comparative research

• Objects for further research– Mapping – Relay – Duplication and comparison

http://tahrirdata.info