Social Media and Policy: A Case Study- Iran

This is the third of five government social media presence and policy outcomes case studies, which will be incorporated into a full research paper.[1]

I. Brief on Iran

In the original regional review, Iran is unique for large fall in inequality after the establishment of its social media accounts.  Looking at the graph below, you can also see, after the earliest accounts were established in 2009, a small rise in secondary school completion and literacy.  Poverty also fell.

Iran Metrics.png

Additionally, like the other case studies from the mid-Eurasia and Northern Africa list, Iran’s unique political situation makes it an ideal candidate for further study.

Iran is ranked 69th on the Human Development Index, one spot above Georgia and 20 ranks above Ecuador.  According to Statista, it also has the highest number of internet users in the Middle East, ahead of Saudi Arabia by 32.55 million users, reaching 69% of the Iranian population.[2]  We Social finds that 49% are active social media users, up 135% between January 2017 and January 2018.

Like in Ecuador and Georgia, Twitter is not the most popular social media platform in Iran.[3]  Facebook, Instagram,[4] and Twitter are all relatively popular but also banned.  Tweets from Iranian government accounts, then, are more likely not reaching and/or are not meant to reach, most Iranian citizens.  Still, a sizable minority is accessing the platform.

Iran Media.png

II. Statistical Analysis

a. Poverty, Gini, and Accounts

As in the previous case studies (see here for methodology), I use correlation and regression analysis to evaluate the statistical relationship between the cumulative number of government social media accounts and the poverty headcount ratio and Gini coefficient.  The correlation coefficients are -0.512 and -0.855 respectively, indicating inverse relationships: as the number of accounts rise, Iranian poverty and inequality falls.  I am not able to conduct regression analysis due to small sample sizes.[5]

b. Voter Turnout and Accounts

In comparison, correlation analysis shows that the relationship between voter turnout and count of social media accounts is weak, and presidential and parliamentary turnouts move in opposite directions.  Parliamentary turnout has a negative relationship to accounts (-.14) while presidential is positive (.25).

This relationship may be connected to the limited choice Iranian voters are presented with.  Both presidential and parliamentary candidates in Iran must be approved by the Guardian Council.  For the May 2017 presidential election, only 6 of 1,600 candidates were approved.  For the February 2016 parliamentary elections, only 51% of candidates were approved.[6]  However, the perceptions of competitiveness or an undercurrent of dissent can still result in engaged voters and raised turnout.[7]

Again, the sample sizes for regression analysis are extremely small, n=6 and n=5, precluding analysis.[8]

III. Background

The Freedom House firmly rates Iran as a whole “not free,” with a score of 18/100 (where 100 is most free).  Similarly, the 2017 Iranian press score is 90/100 (where 100 is least free).  Control of the country’s institutions rests ultimately in the hands of Supreme Leader Ayatollah Ali Khamenei.  Popular and media ideological dissent is actively suppressed.  The 1979 constitution, which established the current Islamic state, and later laws corset freedom of speech to the bounds of Islamic principles and national/public security/stability, with stiff penalties for violators.[9] Journalists are imprisoned arbitrarily or for acting against national security.[10]

The state-run Islamic Republic of Iran Broadcasting dominates legal TV channels (satellite dishes are used but illegal).[11]  State and hardline media together spin a counternarrative against reformists and activists and the state keeps tight control over foreign journalists’ access.[12]

Additionally, most popular social media platforms are banned since 2009 and sanctions have contributed additional limitations. That has not stopped Iranians from using them: Facebook has been used to protest the mandatory hijab, Instagram to depict daily life, Twitter to dispute the 2009 election, and Telegram, more recently banned and extremely popular, to share protest information.

Iran Insta.png

For the Telegram ban, officials cited the need to “promote homegrown apps that could break Telegram’s virtual monopoly on social media” and how Telegram benefits from Iran’s economy without contributing to it (although the ban may end up hurting the economy), but the app’s use in the December 2017 anti-government, economic protests might have also played a role.

An earlier, well-publicized instance of the use of social media in Iranian protests was after the announcement of Mahmoud Ahmadinejad as the victor of the 2009 presidential election.  It sparked the Green Movement‘s mass protests, with citizens claiming a rigged election.  Traditional and social media outside Iran rallied, spreading the movement and its imagery, including video of the death of a woman at the protests.[13]

Ahmadinejad’s first term resulted in unsustainable spending and inflation, which among other economic issues made him appear incompetent leading up to the 2009 election campaign.[14]  Soon after, he proposed phasing out costly subsidies, sparking a heated debate.  In 2010, Parliament approved some reductions, replaced with direct cash payments to citizens which may have played a role in the fall in inequality.[15] The subsequent and current president, Hassan Rouhani, sought additional cuts in 2013, raising gas prices and ending rationing in 2015.[16]

With these reforms in place, Rouhani secured moderate economic growth, despite political headwinds and unemployment worries.  In addition, the PS+1 Joint Plan of Action (JPOA) (the “Iran Deal”) came into effect in 2015, ending a three decade standoff with the United States and seeming to improve Iran’s economic outlook.  However, these gains have already begun to erode after American President Trump’s decision to pull out of the deal, contributing to the December protests.

Rouhani made historic, diplomatic, and economic headway with the deal’s negotiation.  Signed in 2013 by Foreign Minister Javad Zarif, whose personal account is reviewed below, it minimized the threat of sanctions for businesses working with Iranian products, repatriated frozen assets, and lifted secondary American nuclear-related sanctions in addition to UN and European sanctions. Although Iranian foreign policy has not traditionally been in line with economic interests, they were certainly served by these shifts.

Iran’s foreign policy secures the revolutionary priorities but can hurt the country economically, while economic disconnect minimizes regional leadership options. Regardless, Iran does take part in regional affairs, often in opposition to American and Israeli efforts. For example, in Yemen instability became an opportunity for Iran to confront Saudi Arabia, a regional rival supported by the US in the conflict, by providing support to the Houthi movement. Similarly, Iran supports Syria’s Assad, resulting in clashes with Israeli forces.

Iran is also strategically active around the world.  For example, in 2016 Minister Zarif visited Bolivia, Chile, Cuba, Ecuador, Nicaragua, and Venezuela with a business delegation and suggested an opportunity to develop stronger bilateral relations. In the Chavez years, Iran had cultivated a closer relationship with Venezuela, another OPEC member, with multiple high-level visits.

So too is Iran an active participant in global organizations.  For instance, it held the 2012-2015 Non-Aligned Movement presidency and is a member of the G-15, among other international organizations.

The diplomatic see-saw of international diplomacy and conflict mirrors a domestic ideological give-and-take. President Rouhani’s election, and the 2016 parliamentary election, shifted power away from hardliners, although he has not been fulfilled some of his campaign promises or brought reformers into his cabinet. This may be in part because Rouhani has to moderate reform and the domination of the Supreme Leader and the ideals of the Iranian Revolution.

While his victory “is presented to the West as a public mandate for a progressive policy…in fact President Rohani and Iranian Foreign Minister Mohammad Javad Zarif defend and support Iran’s territorial expansion in the region, as well as the exporting of the Islamic Revolution…” and are long-time members of the establishment.[17]  The Rouhani administration’s approach could be seen as pragmatic: the president does not hold the political power in Iran to implement reform as he campaigned on.  Khamenei’s support, or at least acquiescence, drives policy.

The role of the supreme leader may have interesting implications for citizen-government policy communication and social media exchange.  As examined in the Ecuador case study, a top leaders’ tolerance of dissent can play a major role in policy and the tone of social media discussions, even after they have left office.  The Correa-Moreno contrast even bears superficial resemblance to the Khamenei-Rouhani one, where a vocal leader with staying power has granted at least initial approval to a junior reformer who has spoken out against some of the repressive media policies the former favors, accepting some political risk.  In the case of Iran, given the ideological balance that needs to be struck, this apparent “dynamic” may be a strategic one, lending both Rouhani and Khamenei policy-making and/or public relations flexibility.

However, all four Khamenei accounts in the 2017 Twiplomacy data set are unverified, are excluded from the analysis below.[18]

IV. Account Analysis

There are three verified, active Iranian accounts in the 2017 Twiplomacy data set.  Two are personal accounts of President Rouhani, the first using an Arabic description and the second an English, while the third is the foreign minister’s personal account.  You can see some of their basic stats below, showing well-followed, moderately-interactive accounts:

Name Creation Total Tweets Followers Following Tweets/day Since Creation Interaction Rate
Rouhani_ir 22 APR 2013 3,247 420,316 15 2 .45%
HassanRouhani 05 MAY 2013 3,504 630,630 8 2[19] .14%
JZarif 17 JUN 2009 226 695,850[20] 17 0 .78%

Like in previous case studies, I examine 50 tweets, working backward from April 30th.[21] Additionally, MentionMapps show the accounts and hashtags the profile has mentioned, with the central user in orange, connected users in blue, and their connections in gray.  They help illustrate the subjects each account discusses, who they communicate with, and what those accounts are discussing.

a. The Presidential Accounts: @HassanRouhani and @Rouhani_ir

HassanRouhani tweets in English and focuses on community building internationally and among Muslims (38%).

Some tweets, often text-only original commentary, are openly critical of American and Israeli foreign policy.

The Rouhani_ir presidential account tweets exclusively in Persian.[22]  The account incorporates a wider range of messages than the English language account and focuses more often on internal policy issues.  Written in Persian, this account is potentially communicating more to Iranian audiences, at home or abroad.

Like the English-language account, Rouhani_ir has a community building focus (22%), although the primarily function of the account’s tweets is to inform (52%).

Both accounts primarily tweet original, text-focused content (56% of both accounts’ tweets), and the most mentioned accounts in their tweets are themselves. Only 14% of HassanRouhani’s and 1% of Rouhani_ir’s tweets have links. The main emotion of the Rouhani_ir tweets is commitment (26%),

while 26% of HassanRouhani’s are reporting. Comparing the mention map helps illustrate more of the accounts’ similarities and differences.  The HassanRouhani account mentions primarily international leaders and issues.  Its mention map is made up of foreign leaders’ and organizations’ accounts (an Italian senator, the UK prime minister, the president of Austria, and the International Atomic Agency) and hashtags referring to the “Iran Deal” and the United Nations General Assembly. HR MM.png In contrast, the Rouhani_ir map is more domestically focused.  While the main hashtags are for the UN General Assembly and #WAVE: World Against Violence and Extremism, obviously international, WAVE is an Iranian initiative, and the mentioned accounts are partly made up of Iranian leaders and organizations (the Vice President on Women and Family Affairs, the Iranian Students’ News Agency). HI MM.png Taken together, the accounts address different audiences.  Their analyzed tweets do not seem to network so much as expound, sharing one-direction thoughts and commitments to build community without directly referring to other actors or sources.  They are often addressed to other governments or politicians (78% of HassanRouhani’s tweets and 68% of Rouhani_ir’s).

b. @JZarif

HassanRouhani tweets in English and focuses on community building internationally and among Muslims (38%). Like the presidential accounts, a significant portion of the foreign minister’s tweets expound, without external links or mentions, are original, and are focused on text rather than other media (46%).  His tweets often point at American policy to criticize and/or build community with other nations in disagreement with American actions or leadership, as also demonstrated by the hashtags in his MentionMapp below. 22% of the analyzed tweets from the JZarif account are antagonistic towards or anti-American, more than either presidential account.  The foreign minister more often comments on American’s actions in the region as a whole, versus to focusing American actions against Iran or specific allies.

For 94% of his tweets, the target audience seems to be other governments or politicians, while 82% are aimed at citizens.  “Critical” and “frustrated” are the most common emotions (24% and 24%).

Although he tweets in English and so presumably for an international audience, his MentionMapp indicates that he most often mentions Iranian accounts.

JZ MM.png

c. Overall

Taken in sequence, it appears each account has a role to play, facing increasingly outward and speaking to an ever-wider audience.  Rouhani_ir speaks to Persian readers about domestic issues, informing and making commitments aimed more often at citizens than other accounts.  HassanRouhani speaks to English readers and more often governments/politicians, with a greater focus on foreign affairs and community-building.  Finally, Jzarif is foreign-focused, as you might expect a foreign minister’s account to be, addressing governments and a wider range of policies.  The accounts also share some similarities, from the anti-American, Israel, and Trump messaging to their predominantly expounding format.

Account Average Likes Average retweets Average comments Primary Type Most RT’d account Most mentioned account Most mentioned policy subject Number of days
Rouhani_ir 2808 183 789 O (84%) Azarjahromi (38%) Rouhani_ir (100%) Economic (21%) 253
HassanRouhani 2239 436 363 O (66%) JZarif (29%) HassanRouhani (56%) Sanctions (23%) 452
JZarif 2330 532 369 O (94%) N/A N/A War (33%) 150

Overall, the overlapping messaging, combined with different intended audiences, topics, and tones, suggests a team, rather than an unrelated cluster, of accounts, working to build an international consensus around the Iranian position rather than shape domestic politics or perceptions.  Issues like sanctions,

US/European weapons,

conflict in the Middle East,

and perceptions of Iran,

while hypothetically intended for a primarily foreign audience,[23] also have an impact on the functioning of their domestic economy and government and the stability of the regime overall.

In addition to any influence those tweets may have on domestic policy outcomes, there are also tweets that seem to be directed at Iranians despite the ban, although this functionally admits enforcement failure.

These tweets may be in emulation of other political leaders’ accounts, without the intention of actually being read by Iranian citizens, or only intended for Iranian politicians on Twitter.  They could also be addressed to the global Iranian diaspora rather than domestic communities.

Oddly, the Supreme Leader Khamenei is rarely mentioned in the analyzed tweets.  The accounts are not used to follow, appease, or highlight Khamenei or his ideas, political or religious.  This may be in part due to disagreement between the president’s and the Supreme Leader’s agenda, but may also expose a strategy to make Iran’s secular leadership appear more independent and so more appealing to Western readers.  The Supreme Leader’s tweets can be inflammatory; other government accounts may play a mitigating role, a strategy that could resemble the foreign audience theory outlined in the Georgian case study, although the Iranian diaspora’s influence could be unique.

V. Discussion and Conclusion

By banning social media platforms, the Iranian government is potentially burning bridges of communication between the government and its citizens.  The bans themselves may not even be effective.  Their existence would seem to limit the probability that there is a causal relationship between the governments’ use of popular foreign, domestically banned social media accounts and policy outcomes.

Statistical support for a causal relationship between social media presence and policy outcomes is weak, but that may be due in part to a unique Iranian social media strategy.  Iranians have been getting around the bans and getting on social media.  Even if it minimizes the potential for a consultative or deliberative relationship between Iranians and their government, it doesn’t mean that they are not listening; that the government is not using the accounts to communicate with them or the Iranian diaspora to sway opinion, build community, or construct a narrative.

Alternatively, that narrative could be meant for foreign ears.  The accounts could be speaking to foreign audiences in order to build a united national social media character or to accomplish specific foreign policy goals.  To prove the latter true, I would need to demonstrate a causal link between social media presence and specific foreign policy objectives, such as the absence of formal wars, falling numbers of internally displaced people, rising trade values, formalization of trade patterns, or other trade, stability, or influence metrics.  Such a relationship could establish that social media presence promotes peace or economic goals, could thereby improve domestic policy outcomes like the poverty rate.

An alternative explanation for the patterns observed in this analysis could be similar to that found in the Ecuador case study: policy outcomes driven more by recent policy changes than by the communicative strategies around them, specifically direct cash payments and the weakening of sanctions in the case of Iran.  This could be especially true when that communication environment is constricted by censorship, as is the case for both Ecuador and Iran.

As this case study stands, the relationship between the presence of political leaders on Twitter and policy outcomes are not causally clear.  Statistical data is weak or nonexistent, while the qualitative review is not conclusive.  It does, however, provide further evidence of at least some government accounts being used to communicate with primarily foreign audiences, to establish common bonds and/or shape a national narrative.  Additionally, it demonstrates the influence of geopolitics and the political structure of a nation on the way it speaks to the world.

[1] Paper will include original analyses (here, here, here, here, here, here, and here), literature review, and five total case studies (Ecuador here, Georgia here).

[2] Although number of user estimates range from 39.15 to 56.7 million.

[3] We Social estimates 5.7 million Facebook users in Iran to 24 million Instagram users.

[4] The country is 7th in the world in number of active Instagram users.

[5] Both samples only had 9 data pairs.  While I did not include this rule in the earlier cases, I will not display key regression statistics unless n=10 for at least one of the samples.

For all three case studies so far, there is not annual data for the outcome metrics.  For the regression I can only use years with data, while I can perform a correlation analysis with “blank” non-zero years.  I use n=10 as the cut-off to include more regressions than I otherwise would have been able to, without resorting to single digit sample sizes.

For the sake of transparency, the policy-count regression shows the (weak, -0.62) relationship between count and poverty to be statistically insignificant at the 90% confidence interval.  In contrast, the relationship between count and inequality is stronger (-1.16) and statistically significant.


[7] See and

[8] The correlation coefficients for cumulative account count and parliamentary and presidential voter turnout are -0.61 and 0.66.


[10] and








[18] Additional research on the roles of unverified accounts versus verified, of the accounts of religious leaders versus political leaders (versus religious leaders who are also political leaders?), and of how individual accounts in a set attributed to a single person or office behaves differently from each other could be illuminating here, and could better flesh out the relationship between government influence in social media, although it is out of the scope of this study.

[19] Interestingly, the English-language presidential account, @HassanRouhani, has not tweeted at all since February 17, 2018, while the Arabic-language account has been more active, with the most recent tweet showing, on May 14th, 2018, is from May 14th.

[20] Equivalent to <1% Iranian population

[21] This may or may not continue to be possible for Ukrainian and Ugandan accounts, as April tweets may no longer be visible by the time I can complete those coding sessions, in which case I will have to shift to a later but uniform start date.

[22] I am not literate in Persian and had to rely on free translation tools to code the content of these tweets.  Due to the well-known shortfalls of these tools, my sense of their tone may be less accurate than for Spanish or English tweets, languages I can read.

[23] Implied to be intended audiences.  I do not claim to be able to read the minds of these leaders, only to make an educated guess at their intended readers based on word choice and syntax.


One thought on “Social Media and Policy: A Case Study- Iran

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s