PAIR2004

YouGov Report: Feasibility Study of Research into the Arab Spring

Student ID’s:

24067083

24657255

24759031

24635308

24758779

24711977

23397063

24042293

Word Count: 6,006 words.

Contents

Executive Summary…………………………………………………………………………………………………………….3

Rationale…………………………………………………………………………………………………………………………….4

Methodology …………………………………………………………………………………………………………………….7

Results/Findings…………………………………………………………………………………………………………………8

Policy Recommendations…………………………………………………………………………………………………..14

Appendix…………………………………………………………………………………………………………………………16

Bibliography………………………………………………………………………………………………………………………19

Executive Summary

Rationale

  • Empirical data on the Arab Spring is sparsely available and many conclusions being drawn are assumptions based on anecdotal evidence.
  • One of the most significant considerations when conducting research is the reliability of the data already gathered.
  • There are questions about whether existing data is comparable and whether any new data which is collected can be comparable when studying such a large and diverse selection of countries which are currently politically unstable.
  • A key barrier when considering the feasibility of conducting primary research in the Middle East and Northern Africa is the complexity of ensuring research is ethically conducted. A project on this scale will always have extended ethical issues to work around, but given the vulnerability and fragility of countries involved in the Arab Spring such as Syria where the conflict is as of yet unresolved, will make this project more difficult than most.
  • Lack of expertise in collection, interpretation, and presentation of data continues to be a problem in the Middle East and Northern Africa

Methodology

  • The method of research undertaken in this study differs to the methodology of a research paper,as the task given was a study of feasibility, rather than aprimary research project. The decision was made to review previous academic work on the Arab Spring as well ascollecting existing data to determine whether further collection of data in the region would be required.
  • The practicalities and difficulties of carrying out research in the region were then considered by examining previous research work in the Middle East & North Africa.

Results/Findings

  • From a political perspective, censorship over research is still rife in many Arab Spring countries. For a lot of countries, research authorities are still mainly government funded such as in the Maghreb countries and Gulf nations.
  • An understanding of each country’s religious and ethnic make ups will be important when choosing researchers and to ensure good quality data is gathered.
  • Reliable information is exceptionally hard to come by in certain regions, with simple literacy levels of countries such as the UAE being measured irregularly and therefore too outdated to be credible.
  • Thirdly, from a socio-demographic aspect, the customs and traditions present in many communities in the Arab Middle East are exceptionally male dominated. Women are often marginalised which will make achieving representative data difficult.
  • From an ethical perspective ensuring the security of research participants will be vital.
  • There are terrorist organisations operating within these regions which have a history of attacking Western projects within the Middle East.
  • Several countries which researchers will be operating in are governed by Shari'a law, the strictness of which means researchers will have to ensure data is collected in an inoffensive manner.
  • Another key area of ethical concern when conducting the research is ensuring that it is collected in a consistent and fair manner. Technologies available e.g. internet access may be limited so methods which are open to all will need to be used.
  • There will be a wide spread of languages across the 21 countries involved and so ensuring data is accurately translated to so participants are fairly represented will be important.

Policy Recommendations:

  • Using local researchers whose own religions and ethnicities match those of the research participants will be vital to overcome the broad ethnic and religious cleavages within each of the countries targeted and may help reduce the initial tensions that are produced by difficult interview questions and improve the quantity and quality of data that is gathered.
  • Thus, we put forward the policy proposal that time is spent in the recruitment and training of local researchers for your work.
  • It is difficult to offer a uniform recommendation to suggest how ethical difficulties may be overcome but perhaps a pragmatic approach is best using the judgment of researchers on the ground to tackle difficulties as they arise.
  • This report concludes that YouGov should go ahead with the process of gathering data on the impact of the Arab Spring throughout all of the countries involved across the Middle East and North Africa.

Rationale

The purpose of this report is to examine the feasibility of a study into the impact of the Arab Spring in the Middle East and North Africa. In the interest of the aforementioned feasibility we have decided to investigate whether there is a real need for YouGov’s proposed study. With this aim in mind we have decided to review previous academic work on the Arab Spring to analyse what data is currently being used, and also try and collate some secondary demographic and ethnic data of own to see what data is actually available and whether there is a need for further collection of data on the impact of the Arab Spring.

It is interesting to see that many academic articles are often only using anecdotal evidence to support their arguments. For example Khondker’s (2011) article “The Role of New Media in the Arab Spring”, uses almost no empirical data. Khondker (2011, p.676) describes the self-immolation of a fruit seller in protest against the Tunisian government and then simply states that new media helped to make this a national event which brought “virtual and real revolutionaries...out in droves to protest”. Khondker makes no attempt to quantify how big a role new media played in spreading news of the event or how many people came out to protest as a result – there is no empirical evidence available on the event. Instead, Khondkerhas made an assumption about the cause and outcome of the protest. Khondker does make attempts to use some empirical evidence in his article - for example noting that a Facebook page called “We Are All Khaled Said”, created to draw attention to the beating and death of an Egyptian blogger, rapidly gained 350,000 followers, showing how new media was being used to disseminate information quickly amongst protesters. However, Khondker’s work shows that empirical data is sparsely available and many conclusions being drawn are assumptions based on anecdotal evidence.

Khondker is not the only academic who has been forced to rely on anecdotal evidence to support his research. Anderson (2011, p.2) describes demonstrations in Tunisia as driven by “neglected rural areas...[and]repressed labour movements”, whereas in Egypt they were driven by “urbane and cosmopolitan young people”. In Libya, Anderson argues protests were led by “ragtag bands of armed rebels” and caused by “tribal and regional cleavages”. Whilst these statements may be true there is no empirical evidence to support them and no demographic data to back up what Anderson describes – they are all anecdotal assumptions. Parchami (2012, p.39) also makes assumptions about the Shia population in Bahrain, describing them as “disenfranchised and disaffected” but makes no attempt to support this. Again Parchami is relying on popular anecdotal opinion and not public opinion survey data. It seems quite common for researchers to make anecdotal statements about motivations for the protests, as seen in Hazran (2012, p.118) and Forstenlecher et al (2012, p.54). That such a large range of academic research is using anecdotal evidence highlights the need for wide-ranging and reliable empirical data in particular, public-opinion surveys to reveal the motivations behind the Arab Spring uprisings.

Wilson & Dunn (2011, p.1,248-9) admit that previous analysis has been reliant on anecdotal evidence because of the absence of empirical data and have attempted to address this withtheir Tahir Data Project – a venture to collect data sets on social media use. Wilson & Dunn examine the use of social media during the Arab Spring, specifically the use of Twitter in the Egyptian Revolution. To gather the data for the Tahir Data Project 1,200 interviews were carried out with protest participants who were selected by ‘snowballing’ (Wilson & Dunn, 2011, p.1,249-1,250). Although Wilson & Dunn’s work is restricted to Egyptian protesters, it does reveal some of the difficulties researchers might face in trying to collect data in Arab Spring regions. Finding protest participants was difficult for Wilson & Dunn – hence the need to use the snowballing method to select participants. Unfortunately the use of this method made it almost impossible to draw a representative sample – 75.5% of Wilson & Dunn’s sample were male and over 77% had a degree. They had little control over the participants selected because of the lack of a sampling frame and the reluctance of some protesters to participate, for fear that their identities might be given away. Wilson & Dunn (2011, p.1,250) argue that although the sample may not be representative of the Egyptian population or the protesters, it is difficult for a researcher to know what the general population demographics actually look like because of the “obfuscation” of Egyptian census data (Wilson & Dunn, 2011, p.1,250). Wilson & Dunn’s work did not involve online surveying, instead interviews were face-to-face and whilst this may have allowed the collection of data from individuals who did not have access to the internet, other problems arose, most notably the security risk for protest participants – hence interviews had to be carried out in semi-public areas like cafes and parks for protection (Wilson & Dunn p. 1,250). They (2011, p.1,251) also compiled a data set using an archive of Tweets using certain hashtags. However, the sole focus was on tweets in the English language such as #jan25 (the date most protests started), automatically excluding Arabic language tweets. This highlights the importance of having local knowledge to collect empirical data to overcome simple difficulties such as language barriers.

Gause (2011) and Parchami (2012) both admit that academics did not expect the Arab uprisings; Parchami (2012, p.35) describing “regional experts...[as]...stunned”. If “the vast majority of academic specialists on the Arab world were as surprised as everyone else” (Gause, 2011, p.81) then this really suggests that there was a lack of understanding amongst academics about the individuals on the ground in the Arab Spring countries, highlighting the need for public opinion research.

Reading academic articles, it is clear that academics often have to go to great lengths to find data on the Arab Spring to support their work. Forrtenslecher et al (2012, p.55, Note 12) use a leaked US intelligence cable from the ‘Wikileaks’ site in their attempts to predict the prospects for democratization in the UAE after the Arab Spring. Hazran (2012, p.119 Note 15) is using data to describe ethnic and religious divisions in Syria which is over 11 years old, pre-dating the Arab Spring. It is unclear whether this data can be considered recent, and therefore relevant enough to draw accurate assumptions from.

The conclusion we have drawn from this literature review is that current academic literature on the Arab Spring is based on generalisations, assumptions and anecdotal evidence or data which is out-of-date or of unknown origin and reliability. This clearly demonstrates the need for a YouGov survey to provide accurate and up-to-date public opinion survey data which academics can use, as currently no comprehensive data set exists.

After concluding our literature review we have decided that it would be useful to attempt to gather some secondary data ourselves to examine what data is currently available to researchers and then to determine whether this data is reliable and useable. Examining the limitations of existing data might provide further validation that YouGov should go ahead with its research project. The data we collected is divided into 3 tables – Table 1: ‘Population Data’, Table 2: ‘Ethnic & Religious Population Breakdown’ and Table 3: ‘Arab Spring Events’ all of which are presented in our Appendix.

Table 1 demonstrates that there is a wide range of basic population data – population size, density, urban population and GDP available for each of the 21 countries YouGov has proposed surveying and it is also available from a reputable source – the UN Statistics Division. However, this data does not greatly assist with work relating to the Arab Spring, at least not directly. We were also able to collect data on the religious and ethnic cleavages within the states, with the exception of Palestine and Yemen. This data, however, is from the CIA Factbook which although this might be considered a reputable source, the CIA Factbook will not divulge how it’s data was collected or its sources which severely restricts its usefulness as a source of reliable and credible data. This ethnic and religious data is useful though to demonstrate that there are broad social and cultural cleavages within these countries which will need to be considered when conducting research which we will discuss further in the ‘Results/Findings’ section of our report. Finally Table 3 displays the types of protest that have occurred in each of the countries and whether any instability is ongoing, which is useful when considering the risks and practicalities on further research in each state. The secondary data we have gathered has proved that there is basic demographic information currently available from the UN, however data on the impact of the Arab Spring is not existent – there is a need for YouGov to collect this data.

Methodology

Choosing the right method of research is crucial when writing a research paper. For instance, Shuttleworth(2009) states that “Selecting the correct type from the different research methods can be a little daunting, at first. There are so many factors to take into account and evaluate.” The method of research undertaken in this study can be considered unique compared to the general methodology used in a research paper, as the task setis a study of feasibility, rather than the actual research project itself. Therefore, a different approach needs to be undertaken. We have already made the decision in our Rationale to review previous academic work on the Arab Spring, and then exploreexisting secondary data to determine whether further collection of data in the region is required. Secondary data analysis was chosen as it is a very flexible method that can be applied to a variety of different media. It is an approach to the analysis of documents and texts, rather than gathering data, in a sense it is not a research method as such (Bryman, 2008, p.1). We concluded in our Rationale that there was a need for YouGov to undertake further research.

Our Rationale began with a literature review. This is what Jupp (2006) has defined as “the further analysis of an existing dataset with the aim of addressing a research question distinct from that for which the dataset was originally collected.” This involved a critical study of a variety of articles written on the Arab spring, to which we would try and find a justification for YouGov’s research. A critique was made of the approach that other researchers had used and what findings had been gathered. We felt that a literature review would give us a good indication on how previous research had been done on the topic of the Arab spring and any mistakes made by previous researchers could be identified. Previous studies identified to us that often, researchers have to go to great lengths to find data on the Arab Spring to support their work. Khondker’s (2011) work shows that empirical data is sparsely available and many conclusions being drawn are assumptions based on subjective evidence. Furthermore, research has also presented confirmation that a large range of academic research is using subjective evidence and therefore indicates the need for comprehensive and reliable empirical data.

As our Rationale has concluded that there is a need for YouGov to undertake research on the impact of the Arab Spring we will move on in the ‘Results/Findings’ section of our report to look at the whether it possible to attemptthe research in the regions identified – North Africa and the Middle East, and, if so, what problems might inhibit the gathering of data in these countries.Primary research, in particular public-opinion surveys,will be required to reveal the true inspirations that led to the Arab Spring uprisings. Therefore, we will look at the success of past attempts to conduct primary research in the region.

Our next step will be to look more closely at the countries involved in the Arab spring and to assess what potential problems could hinder the collection of data in these countries. Factors such as communication willbe considered. We will examine whether, in Gulf Countries such as Iran, Bahrain, and UAE, it will be difficult for social science research to be conducted by a Western research organisation. Religion will also be examined as a potential obstacle in gathering sufficient data in the Middle East. Obviously data needs to be reliable in order to make useful comparisons between the numerous countries proposed by YouGov, so the report will also explore if it is possible to collect reliable data by examining academic articles on the importance of data reliability, and more specifically the difficulties faced bypast research in the Middle East and North Africa when trying to collect reliable data.