Social Media and Policy: A Case Study-Ukraine

This case study is the last of five.  They will be incorporated into a full paper exploring the relationship between governments’ presence and activity on social media and policy outcomes.

I. Brief on Ukraine

Ukraine was selected as a case study candidate because it had a large fall in poverty over the analyzed years (over 20% to under 1%) through the creation of their government Twitter accounts, and those accounts were frequent tweeters with high interaction rates.

The country is ranked 84th in HDI- below Georgia and just above Ecuador. Some indicators are moving in healthy directions: literacy rate, secondary school completion, and electricity access have been rising while poverty has been falling.

Ukraine outcomes

According to We Social, 58% of the Ukrainian population are internet users[1], and 29% are active social media users, 72% using those channels every day.  Internet use has been on the rise, but social media use has fallen 24% between January 2017 and January 2018.

As in other case study countries, Twitter is not the most popular social media platform in Ukraine, although it has been competitive, being used by more than a third of the population mid-2017.  31% of Ukrainians preferred Facebook for news in 2017, while 2% preferred Twitter.

Ukraine social.png

II. Statistical Analysis

Comparing the count of Ukraine’s active Twitter accounts and the poverty rate and Gini index using correlation analysis, I found a moderate negative relationship with both indicators, with coefficients of -0.44 and -0.55. In short, as number of accounts rises, poverty and inequality fall.

I used regression analysis to explore the potential causal relationship.  The results, at the 90% confidence level, are compared below:

Statistic Count and Poverty Count and Gini
R2 0.19 0.299
Coefficient -0.95 -0.68
T-statistic -1.89 -2.53
P-value 0.078 0.023

The results support the existence of a relationship between the cumulative number of accounts and both poverty and inequality.  They show a statistically significant, strong inverse relationship.

There are not a large enough sample to run analyses on the count-voter turnout relationship.[2]

III. Background

A major consideration in Ukraine’s foreign and domestic policy is the balance of competing Russian and Western interests.  The timeline below outlines some major events in Ukraine’s foreign relations with both.

Ukraine timeline


Current public perceptions of foreign partners are mixed.  If a referendum were to be held today, 37% would vote to join NATO and 26% would vote against it.  At the same time, positive perceptions of the EU in Ukraine have been growing, reaching 49% in 2018, and 69% link peace and stability with the EU.  66% tend to trust the EU versus 40% NATO, where distrust is at 49%, and 29% the Eurasian Economic Union (founded by Russia, Belarus, and Kazakhstan), where distrust is 48%.

However, there is more to Ukrainian media, politics, and economics than foreign influence, even if reform moves forward with EU membership in mind and support at their back.

The Freedom House rates both Ukraine, excluding Crimea, and its press as partially free (62/100 where 100 is most free and 53/100 where 100 is least free, respectively).  While security and war are major concerns, Ukrainians are more worried about poverty and living standards.  Economic crisis, unemployment, and corruption are other “most pressing problems” facing Ukraine.

Separatist conflict fueled an economic crisis in the early 2010s, prompting the International Monetary Fund to step in with a bailout predicated on reform, continuing to pressure for reform since. The World Bank also began providing technical and financial support to Ukraine following the crisis, “support[ing] high-priority reform measures to address the key structural roots of the current economic crisis in Ukraine and to lay the foundation for inclusive and sustainable growth…” Ukraine’s economy is weakly recovering, with modest GDP growth, rising real wages, and falling poverty.

At the end of 2017, Ukrainians were somewhat more positive, with 30% saying the economy stayed the same or improved in the last 12 months, compared to July 2015 when only 10% said the same.  But 42% still expected the economic situation to worsen in the coming year.

Ukrainians do not have much faith in their system overall.  More than 80% of Ukrainians say the term “good governance” does not apply in Ukraine, and more than 90% say the same of “lack of corruption.”   Trust in government institutions is low: only 16% trust the government and 12% Parliament, and more than 2 in 3 Ukrainians are not satisfied with the way democracy works in their country.

Transparency International scores Ukraine 30/100, or 130th of 180 in the world (alongside Iran, Myanmar, and Sierra Leone), on its Corruption Perceptions Index, although this score has been rising in recent years from a low of 25 in 2013.  National Anti-Corruption Bureau of Ukraine (NABU), also part of post-crisis reform, was founded in 2015 as an independent agency “with the purpose of cleansing government of corruption in order to enable formation and development of successful society and efficient state.”  Since then, however, NABU has faced headwinds from the Ukrainian government, including interference with a US-backed undercover operation and proposed legislation that could hurt the group’s independence, while the creation of an anti-corruption court has been delayed.

Anti-corruption efforts may still be making a difference.  Over 100,000 officials have declared their wealth online due to new legislation, highlighting economic inequality and launching dozens of criminal proceedings.  According to the NABU, it is responsible for 21 convictions and has several hundred “proceedings under investigation.” While corruption is still the 4th most pressing issue facing the country (33%) in 2018, it is down from the #1 issue, with 45% of Ukrainian survey respondents selecting it as a “most pressing problem” in 2017.  At the same time, another survey shows that 46% believe the anti-corruption efforts have been a complete failure.  Only 24% believe the NABU is an active fighter against corruption and Ukrainians are generally reluctant to call state institutions “fighters against corruption,” which could be linked to their overall distrust of their government.

More than half of Ukrainians, on the other hand, “completely” or “rather” trust national TV news, and 48% trust online news.  Ukrainian traditional media is dominated by television77% of Ukrainians used TV for news in 2017, while 54% used news sites and 45% use social networks.  Media bias and ties to business interests are a persistent problem, hurting domestic media’s credibility.  More than 60% say “freedom of the media” does not apply in Ukraine.  While the government has been reforming to minimize state influence in media, it has also restricted access to certain Russian sources.

Russian journalists, broadcasts, and websites have been banned as part of sanctions packages and to guard the public against Russian propaganda, influence, and cyberattacks.  Their influence is a national security concern, as illustrated by the example of the Russian Liberation Movement.  One of the now-banned sites, VK, had been used to distribute fake news propaganda videos featuring the group.  The videos demonstrate connections with anti-Ukrainian fighters in Donetsk, the pro-Russian hacker group CyberBerkut, and a Russian “troll factory.”

But the bans seem to have been at least somewhat effective at limiting Russian sites’ visibility: only 12% of Ukrainian internet users used Russian news sites in the last month, compared to 25% in 2015.  Russian TV watchers are down from 12% to 5%.

At the same time, most Ukrainians may not need government protection from those sources.  Their top requirement of news sources is credibility, and they decide trustworthy sources based firstly on if they know the source and secondly on if different points of view are presented.  Only 6% trust Russian TV and 10% of online news, and those who do live mostly towards the east, with the highest proportions in Sumy and Donetsk. Of those who do use Russian TV and web news, only 32% say the information provided by either source is objective and reliable.

Additionally, Ukrainians against the ban on Russian websites say it has caused a disconnect from friends, family, interests, and followers, in addition to having fundamental flaws like being undemocratic.

The Internews survey, cited multiple times above, also examines the relationship between policy reform awareness and media, one of the relationships my case studies were designed to evaluate as an explanation for a relationship between policy outcomes and governments’ social media use.  The survey finds that Ukrainians have most often heard about pension reform in the last 30 days, 75% from national TV channels and only 3% from social networks.  This is a proportion nearly identical across multiple issue areas: 43% have heard something about healthcare reform and 75% heard about it on TV while 3% said social media. 43% heard about land reform, 75% on TV and 2% on social media.  Across all 3 issues, a majority of respondents said coverage was insufficient.  Those who have heard about the reforms tend to be older and more educated.

Taken together, these statistics show the static dominance of traditional news sources over social media, which is not only a secondary source of information on policy changes but also does not seem to have led to a democratization of government awareness, at least in the Ukraine.

Still, Ukrainian politicians are on social media, successfully and unsuccessfully attempting to influence or gauge public opinion.  And not all issues are equal: another survey shows that 57% of Ukrainians have heard something about the European Union in the last 3 months, 22% from social media.  35% go to social media to find information about the EU.

Finally, as previous case studies have shown, communicating domestic policy to citizens is not the only way governments use social media.

IV. Account Analysis

For details on the methodology used in this section, please see here.

I was only able to view back to August 14th on the @MFA_Ukraine account’s page.  So, as with the Uganda case study, I used August 15th, the first date within that time frame that I could view all the tweets posted during that day, final date for Ukraine data set and worked forwards, with one exception.

One “active” account has not been since the 2017 Twiplomacy data was collected.  The most recent @Kabmin_UA_e tweet available is from December 19th, 2017. The other account from these case studies that had not been active in some time, namely @HassanRouhani, had not had a tweet for about four months prior to the May data pull but was included in the overall analysis because the dates of the @HassanRouhani tweets overlapped with other accounts’ tweet dates, so the analyzed sets were comparable.  In contrast, @Kabmin_UA_e does not have that overlap, and so, although otherwise meeting the criteria for analysis and so is included in the table below, will be excluded from the rest of the case study.

Additionally, some tweets were excluded from analysis.  These 8 total posts from 5 different accounts  were Ukrainian-language pictures or videos with no accompanying text or translations.  I am not able to read Ukrainian, so these “untranslatables” were excluded from the analyzed data set due to a lack of specific, confirm-able information.

There are 8 active, verified Ukraine accounts is the 2017 Twiplomacy data set.  The 7 analyzed below are @Poroshenko, the personal account of the current president; @APUkraine, the English-language institutional presidential account; @TheBankova, the Ukrainian-language institutional presidential account; @Vgroysman, the personal account of the current prime minister; @Kabmin_UA, the Ukrainian-language account of the Cabinet of Ministers; @PavloKlimkin, personal account of the current Minister of Foreign Affairs; and @MFA_Ukraine, the institutional account of the foreign ministry.

Name Creation Total Tweets Followers Following Tweets/day Since Creation Interaction Rate
Poroshenko 02 APR 2014 9,161 1,133,351[4] 25 8 .02%
APUkraine 09 JUL 2014 9,241 89,904 324 9 .05%
TheBankova 03 JUN 2015 2,374 5,097 286 3 0.69%
VGroysman 20 JAN 2016 3,793 41,336 98 8 .23%
Kabmin_UA 08 JUL 2011 46,823 115,541 189 22 .05%
Kabmin_UA_e 08 JUL 2011 13,674 8,388 5 6 .00%
PavloKlimkin 02 OCT 2011 1,929 284,150 157 1 .06%
MFA_Ukraine 13 MAY 2010 86,184 86,184 1,072 16 .07%
a. Presidential Accounts: @Poroshenko, @APUkraine, and @TheBankova

While all three are technically presidential accounts, @Poroshenko, @APUkraine, and @TheBankova have different tones, purposes, audiences, and even speak different languages, although they do sometimes share the similar content.

President Poroshenko’s personal account tweets keep a generally positive tone and are most often for community-building (40%), celebratory (36%) or reporting (24%), and entirely in Ukrainian.

@APUkraine is the Ukrainian-language institutional account, more focused on informing (64%) and so adopting a reporting tone most often (48%).  @TheBankova is the English-language version, similarly taking a reporting tone (46%), but with an equal focus on community-building, both domestic and international, and informing (44% and 40%).

All three- in fact, all 7 accounts- are focused on communicating to citizens, although @TheBankova is directed at governments/governmental bodies or representatives most often of the presidential accounts (62% versus 48% and 50%).  Both @APUkraine and @TheBankova more often tweet about government functions (60% and 78% to 26%) and community functions (32%- the most among all accounts- to 22%).   Despite being written in English, the lack of context or additional information make @TheBankova tweets appear to be for a domestic, diaspora, or otherwise highly informed audience.  Many of the posts are every brief, with little information outside a link to a government statement.

@APUkraine’s original tweets are similarly robotic.

Despite their brevity, the posts still have enough information to send some mixed messages, including more direct or aggressive presidential quotes out of step with the content shared on the @Poroshenko account.

The accounts’ MentionMapps demonstrate their differences.  Poroshenko’s personal account is connected most closely to international hashtags and accounts: the US Army Airborne’s annual joint exercise; the UN General Assembly; and current and former leaders of the European Council, UN, France, and Turkey, plus the US special representative to Ukraine.

UK Poro

In contrast, @APUkraine is less evidently connected to specific issues.   Top connections are to the Poroshenko personal account, the Freedom House, and a defense company.


@TheBankova, on the other hand, has a wider variety of connections that more closely resemble @Poroshenko’s- international.  Top hashtags are about Russia and the UN while top connected accounts are similarly Poroshenko’s, former French president François Hollande’s, current president of the European Council Donald Tusk’s, and the current chancellor of Austria Sebastian Kurz’s.

Both institutional presidential accounts are also related to the World Boxing Council.  The 56th convention was held late 2018 in Kiev, and this connection reflects tweets in this and other case studies that highlight high level athletic achievements and sports diplomacy for domestic community- and international reputation-building.

UK Bank

b. Ministers’ Accounts: @VGroysman and @Kabmin_UA

@VGroysman, the current prime minister’s account, is more often directed at businesses than other accounts (32% compared to @Poroshenko’s 12%).  50% of its tweets are informing, but has a range of tones: 30% reporting, 18% commitment, and 14% celebratory.

Unlike the presidency, the prime minister’s office is not using multiple accounts for different messages or styles, and so reflects range of both.  However, the @Kabmin_UA account does share some characteristics: both accounts are most focused on economic policy (48%) and together Groysman’s account and Cabinet of Ministers’ most often tweet about specific bills, 42% and 44% respectively, and often give advice or information to citizens about policies’ implications and opportunities to benefit from them.

The Cabinet of Ministers account shares @VGroysman’s policy focus, though most often reporting (50%) and informing (48%), the tone and purpose by now most associated with institutional accounts in most countries. It also retweets, the most often- 86% of the sample is retweets- and 34% of all the account’s tweets are @VGroysman content, further supporting the view of the account as a complement of the Prime Minister’s.

At the same time, @Kabmin_UA has stronger connections to multiple hashtags and other accounts, including a government procurement system, the country brand, the Ministry of Economic Development and Trade, the Ministry of Infrastructure, and the acting Minister of Health, as well as, unlikely enough, the American event Burning Man.

UK Kabmin

Unlike other MentionMapp’s, @VGroysman shows few and weak connections to conversations or accounts, although the account has the highest interaction rate of the set.  Those on the map include an academic, a businessman, and government accounts.


c. Foreign Ministry Accounts: @PavloKlimkin and @MFA_Ukraine

Finally, the foreign ministry accounts, the personal and the institutional, are similarly complementary.  @MFA_Ukraine reports (58%) while @PavloKlimkin acknowledges or celebrates (38% and 32%).

The institutional account is for informing (36%) as much as community-building (36%) and inspiring action (28%), while the minister’s account is mostly for community-building (60%).  When policy is mentioned, its often passingly.   Both focus on communicating with citizens (96% and 90%) but @MFA_Ukraine’s tweets are more often also directed at other governments/governmental representatives (92% to 66%)- more often than any other account.  This might be why, in part, @MFA_Ukraine tweets often in English, (52%), second only to @TheBankova and compared to 2% of @PavloKlimkin’s posts.  However, both also use more languages than other accounts, including Italian, French, and German for some tweets, sometimes posting the same message in Ukrainian and another language in the same tweet.

Klimkin’s account is more often anti-Russian than other accounts.  Although there are few posts in this dataset that are aggressively anti-Russian, 20% of the whole set, at least 6% of each accounts’ posts, are at least somewhat critical of the nation’s policies, while @PavloKlimkin is in 24% of analyzed tweets.

Both accounts are connected to #Russia, though otherwise have very different connections.  Klimkin is also connected to #Sentsov, a hunger-striking Crimean prisoner in Russia who was arrested for plotting terrorist acts.  @PavloKilmkin and @MFA_Ukraine’s mentions of Sentsov’s imprisonment make up 92% of the mentions in the dataset.

@PavloKlimkin’s other connections include foreign politicians from Latvia, Canada, and the US, while @MFA_Ukraine is connected to domestic people and institutions like Poroshenko and the Ukrainian Parliament accounts.


d. Summary
Account Average Likes Average retweets Average comments Primary Type Most RT’d account Most mentioned account Most mentioned policy subject Number of days
Poroshenko 549 190 46 O (100%) N/A N/A Russia (57%) 9
APUkraine 267 92 17 O (68%) Poroshenko (100%) Poroshenko (86%) Russia (37%) 9
TheBankova 66 36 3 O (96%) N/A Poroshenko (100%) Russia (41%) 13
VGroysman 124 50 14 O (100%) N/A N/A Economic (62%) 15
Kambin_UA 218 89 17 RT (86%) VGroysman (63%) N/A Economic (60%) 15
PavloKlimkin 298 118 18 O (88%) MFA_Ukraine (50%) PavloKlimkin (27%) Russia (63%) 31
MFA_Ukraine 73 57 5 RT (62%) N/A PavloKlimkin (25%) Russia (72%) 6

In the Ukrainian set of tweets, we see institutional accounts complementing personal accounts, often retweeting them and promoting their message with a stronger focus on informing, with a more neutral tone.  Each set has an institutional account that tends to retweet a portion of the personal account’s tweets from the same period (APUkraine and Poroshenko, Kabmin_UA and Vgroysman), though the Pavloklimin-MFA_Ukraine relationship is less straightforward.  MFA_Ukraine retweets a wide variety of accounts and PavloKlimkin retweets primarily MFA_Ukraine, though this relationship too supports the concept of complementary government account sets.

Each account in the set tends to address a different audience with a different tone, as outlined in the analysis above.  As highlighted in the Georgia case study and mentioned here, language is the chief hint of the intended audience, international or domestic, of government tweets.  Like in the Georgian case, both English and the national language were used by Ukrainian accounts to establish a consistent narrative and draw attention to conflicts with Russia.  However, English was used less frequently in the Ukrainian data set.  This can be attributed to, in part, the inactivity of @Kabmin_UA_e, which would have been part of the complementary ministers’ accounts set as the non-Ukrainian audience arm, as @TheBankova likely is for the presidential set.

Uniquely, the Ukrainian accounts also use mixed language tweets, with the foreign-language portion directed at particular countries or statesmen and the Ukrainian simply translating the message for a domestic audience.

In some ways, the accounts work together to create a well-rounded national narrative against Russia. Five in 7 mention policy of or towards Russia more often than any other policy issue.  Celebrations of national heritage, military successes and manufacture, diplomatic cooperation, and even natural landmarks are all opportunities to leverage Russian policies to build community among Ukrainians, to seek acknowledgement and action from foreign powers and citizens, and to define the Ukrainian nation as one simultaneously ancient and still coming together.

V. Discussion and Conclusion

While Ukrainians may not trust their government, this study has shown that it is providing them valuable information about their government and its policies.  Tweets have been used to try to connect citizens with valuable social programs, and that may have played a role in citizens’ knowledge of and ability to benefit from those programs.  This could have had an impact on policy outcomes as the statistical analysis suggests, even if social media is not the main way citizens’ get informed about issues.

At the same time, this kind of interaction does not seem to be the primary purpose of the government accounts.  Their informing function also reports on the actions of public officials, a move towards transparency that, according to surveys, does not seem to be helping gain public trust, and building a national narrative by sharing those officials’ quotes and diplomacy to that effect.

Building a national story may be as important, though less direct, a contributor to falling poverty and inequality compared to legislation and programs.  National narratives set agendas, build community, and can strengthen the international relationships needed to ensure, or restore, peace.  Other data shows that citizens are more often getting information about the EU on social media, compared to other issues, hinting that how these accounts are currently being used may make them an important source of foreign affairs information for domestic citizens.

That does not explain, however, the relative lack of channels for external communication, if narrative-building is the intention.  One of the questions that naturally arises from the groups of accounts examined here is why do the minister’s and the foreign ministry not have a foreign-language, external-facing account as the presidential group seems to have?  The now inactive English-language Kabmin account would have filled that space; why has it been left inactive?  Why, in this case study and in others, does not each group of complementary accounts not only adopt different tones and styles, but also have matching, verified, active accounts that are focused on either a domestic or an international audience?

Obviously, different countries have different priorities when it comes to their national narrative, and the primacy of international opinion among those priorities will be variable.  What is the decision-making process, then, that goes into the formation of government social media accounts?  The English version of the Kabmin account was created the same day as the Ukrainian version, but what about the foreign ministry or government bodies in other case studies?  If social media accounts do have effects on policy outcomes, elections, diplomacy, and the formation of a national narrative, the decisions around their creation are just as important as the decisions about how to use them.

So, as in other case studies, I propose this as a potential area for further research.

This is the final case study.  It has supported some of the theses and conclusions of other case studies, while also finding support for the idea of different account types playing different, and sometimes complementary roles.  It, as well as the other case studies, also supports a connection between policy outcomes and governments’ social media presence, even if they have not conclusively demonstrated a consistent strength or direction in that connection.  Together they have used several unique strategies to examine that connection, and together have helped me think of new ways to examine it going forward.  As always, I welcome any and all interested researchers to pick up any banners or threads presented here to come to a better understanding of the social media-policy outcomes connection.

[1] Estimates range from 19.68 million to 23.30.

[2] Sample sizes are n=6 and n=5, parliamentary and presidential voter turnout respectively.  Their correlation coefficients are both negative (-0.7 and -0.93), indicating that as government social media presence rose, voter turnout fell.

[3] Sources include:

[4] Equivalent of 2.7% of Ukraine population


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