Listen to the radio and decide to have fewer babies

img_3101-1-bb

Other than being an indicator of a household’s welfare status, the ownership of particular durable consumer goods has a much more specific benefit: a radio, a mobile phone, or a television can be a source of information for the household members. So, when data on household ownership of radio and television are collected, they are often taken as a measure of access to mass media, and more specifically to the information through them delivered. In fact exposure to information on television, radio, and in print can increase an individual’s knowledge and awareness of new ideas, social changes, and opportunities, which in turn can affect the individual’s perceptions and behaviors, including those related to health and family choices.

Attending the Development Economics course at Bocconi University – and compulsively surfing Google Scholar – I have found out that numerous studies have analyzed the effect of exposure to mass media, in its different forms, on fertility outcomes. So, when presented with the opportunity to write a paper for my Population Dynamics and Economics class I thought nothing could be more appealing to me than to combine my deep interest for research in development economics and the knowledge I gained living and working in Uganda. So I decided to explore the relationship between media and fertility outcomes a little further.

Uganda has the world’s youngest population, with 57% of its 30.7 million people being below the age of 18, and a further 21.3% between the age of 18 and 30 (UBOS 2010). In addition, with around 94% of Ugandan youth living below US$2 a day, Uganda has the highest poverty incidence among youth in Sub-Saharan Africa (World Bank 2008), making a decrease in fertility rates a priority for the Ugandan government. In fact, the Total Fertility Rate in Uganda between 2009 and 2011 has been on average of 6.2 children per woman.

Given that the situation is so challenging, knowing whether there is an effect of the information available through the media on the number of children women decide to have can be of extreme help when structuring government policies or NGOs programs with the aim of reducing fertility rates. Therefore I tried to determine the effect of mass media access on women’s fertility choices by looking at how their family preferences change when information is accessible through the media. To carry out the regression analysis I used a pooled cross-section dataset of the Ugandan Demographic and Health Surveys (DHS) carried out in 1988, 1995, 2000, 2006, 2009 and 2011.

As the mass media of interest, I chose radio since, in 2011 (latest data available), two-thirds of Ugandan households had at least a radio, whereas only 17.5% had a television. Moreover, an average of 68% of the women from DHS Surveys reported listening to the radio and 66.5% reported hearing through it about family planning at least once; furthermore this percentage increases to 77.5% if only we consider only those who own a radio in their household. It is relevant to notice that no specifics on radio programs contents were available to me, and therefore my research did not examine the effect of a specific policy or message, but the ones of the ownership of a radio and of the exposure to it. It is in fact safe to assume that any information delivered by the media is relevant to the decrease of fertility related outcomes since they bring “modern” ideas forth, and they are commonly associated with the fall of the taboo on family planning methods and with the promotion of the ideal of smaller and more educated families.

As the fertility outcome variable I chose the “Children Ever Born to Women 40/49 Years Old”, which is the mean number of children ever born to women at the end of their childbearing period. The reason behind this choice has to do with the fact that the age-structure in a country with high mortality such as Uganda needs to be taken into account when considering fertility measures.

The results of the regression analysis show that “Radio Ownership” and “Exposure to Radio” have a negative and statistically significant effect on the CEB40/49. I then decided to try to dig deeper, and investigate the channels through which this effect works, whether by a decrease in the demand for children or by an increase in the awareness of women, or if these drivers work basically at the same pace.

To proxy the demand for children, I constructed a variable capturing the preferences on the number of children starting from an explicit question asked in the DHS questionnaire. Taking into account only those women that can have children, I divided them into three categories: those who want no more children, those who are undecided, and those who want more children. As far as awareness is concerned, I defined whether a woman is “Aware and Proactive” based on her knowledge of the ovulatory cycle and of the existence of contraceptive methods, and on her intention to use any such methods. If a woman complies with all the three characteristics, then she is considered “Aware and Proactive”; if she complies only with the knowledge variables she is “Aware but Not Proactive”; if she only complies with the intention to use she is “Not Aware but Proactive”; and at last if she does not comply with either requirement, she is “Not Aware and Not Proactive”.

The only case in which a significance of the interaction variable is present is the heterogeneous effect of “Radio Ownership” on the “Fertility Preferences” across these four groups, but, in general, results show that no definitive answer can be given as far as which channel is driving the change in fertility outcomes.

To sum up, my research confirms that there is a significant effect of media exposure on the number of children a woman decides to have, and on how much control she gains over the actual number of children she has. It is imperative for policy makers worldwide to take advantage of this channel when designing policies to control fertility outcomes, especially since the cost-effectiveness of acting through an already existing channel could be groundbreaking.

Other than the research outcome itself, what I truly wanted to share with this post is how I strongly believe that a deep knowledge of the real world situation we are trying to investigate in development studies is fundamental when doing research in this field. It is not enough to sit in classrooms and read papers, the real ideas and breakthroughs of what is needed, even in the research literature, come when travelling to these countries and experiencing life through the eyes of the locals. I encourage you all to take the chance to work and live in developing countries. Just as much, I invite you to take part in poverty labs worldwide, where to share ideas and experiences with people interested in the same issues, and why not, also in our very own LEAP at Bocconi University.

Beatrice Montano

Careers in Development Economics: What does it mean to work in the field as a Research Associate?

Livia Alfonsi, 25 years old, is a former Bocconi student who now works as a Research Associate for BRAC in Uganda. She is currently working on a research project about small firms’ expansion and job creation for the youth. It is a randomized control trial (RCT) that BRAC is carrying out with partners at LSE and UCL. I decided to interview her to know more about what it means to work in the field and how it feels to carry out research in person.

 

Livia, what does it mean to be a Research Associate?

The description of the job really depends on the project you are working on. The tasks are very diverse and require a variety of skills, not only qualities that you learn in school and strictly related to research. In my experience, since day one I have been given a lot of responsibilities. My team is very small, with only four people total: a field manager, a research assistant, an intern and me. That is why I also carry out tasks that would more easily fall under the description of a project manager’s job.

What I do can be broadly divided into two categories: research-related tasks and managerial tasks. Most of the research-related work is carried out after the data collection phase, although even during that initial stage I think it is always a good practice to look at the data coming in, controlling for its quality. It is essential to know if the respondents really comprehend what you are asking them and the enumerators fully understand what they have to ask, and if not, to try to solve the issue before it is too late. After the data collection you have to run consistency and attrition checks, clean and analyze the data. Usually the principal investigator, the Professor who gets the grant and conceives the research project, sends you a to-do list of tasks to perform on Stata. In my experience I have been given a lot of freedom of choice and responsibility about how to carry them out and I have been lucky enough to be involved in all the phases of the projects, from the design of the survey to the writing of the paper.

The managerial tasks are very diverse and are difficult to describe. They are mainly related to the data collection phase, which in my project lasts about 6 months every year. In my case they range from training and assessing the preparation of the enumerators (the people administering the survey; usually locals that master the Uganda language/s) to organizing the purchase of thousands of bars of soap, the reward we give to our respondents. Usually the research associate is not in charge of the daily monitoring of the data collection. I still go on the field sometimes, especially if the interview is important (e.g. we are piloting a new section of the questionnaire or we previously detected issues in one enumerator’s work and we want to double check) and is carried out in English. However, usually it’s the field manager the one who is in constant link with the field, especially due to linguistic barrier.

Managing all the problems that can come up during the data collection phase isn’t easy, anything can happen: the tablet used to collect the data breaks, the enumerators loose the soap bars to reward the respondents (or sell them!) etc. You need to solve all the problems; it can be tiring sometimes but it teaches you a lot!

Another very important task, especially in the first waves, is formulating the survey. Thinking about what to ask and how to phrase the questions is a very delicate chore: you have to think about who is going to answer those questions. In our study, for instance, even if we are interviewing firm owners, the firm is typically very small with no more than 5 employees in most cases, and the owner usually didn’t go to school past 5th grade. You need to phrase the survey questions taking all this into account.

 How did you get there?

 Like many students who are passionate about the field, I first approached the subject during my undergraduate studies, when I attended, during a campus-abroad experience in Vietnam and Thailand, a Development Economics course “growth and macro-oriented”. That has been the first dive into the discipline, from both an academic and a personal point of view. Later on I did a thesis on microfinance and I kept studying development and other applied micro courses during my master program. After internships at the OECD and at UNIDO headquarters, I realized that I needed a field experience. I got in contact with the junior principal investigator of the project I am now working on, who was looking for someone who was able to do data cleaning. At the beginning I worked from Milan, then I went on the field with BRAC, initially as an intern. After 2 months and a half I got the contract as a research associate and now I’ve been here for more than a year.

What are the competences and personal qualities you consider necessary to be a good RA?

 First and foremost, it is the ability to adapt. Not only to live in a developing country (of course you might take a bit to adapt to cockroaches in your bedroom or mice in your kitchen!), but also to a new culture, including speaking English in a different way. You need to be eclectic and it is also crucial to have fertile soil to become a “hands-on” person, to solve by yourself the practical everyday issues you are faced with, especially during the data collection.

Then, it is essential to be organized and a fast learner. Nobody is there to follow your work closely, step by step, because your superiors are far away and are often very busy with several research projects. You need to be able to sort things out by yourself. Thankfully there is the Stata list and also some of your colleagues and friends from the university can help you with Stata work! Moreover, in Kampala there is a vibrant community of people working in the development field, it’s easy to find someone ready to help you.

It is also important to have great communication skills: an excellent level of English and the ability to express yourself clearly are essential both during your everyday work and for special occasions. Sometimes, for instance, you are asked to make a speech at meetings with the top management of the project financing institution or at international conferences.

On the technical side, knowledge of what a randomized control trial is, of econometrics and Stata are of course required. Everything else can be more or less learned on the field. For instance, the project I am working on is associated to labor Economics, but during my studies I have never attended a course in this field. My background has been very useful as a signaling device, it taught me rigor, to sort things out by myself, to be organized and work hard, but even if I didn’t know many things I learned them on the way.

What are the main challenges you encounter while doing research?

 When you are at the university, you read beautifully written papers, in which everything seems to work smoothly: randomization works out perfectly, causation poses almost no problem. At the beginning there is a shock if you are used to theory and you are suddenly faced with reality. Ensuring good data quality is not always easy because the survey might be very long and sometimes the respondents don’t understand fully the questions, some other time not even the enumerator understands deeply. Now I know what measurement error is! I really understood the critical importance of the data collection phase, which is often underestimated when you are a student, of a good survey design and of the subsequent check of the data quality. You need to make sure that the information you are asking to provide and the answer you are actually collecting are the same.

Another main challenge is that times and rhythms here are completely different. As a Bocconi student I was used to organize my work with tight deadlines and fast working paces, but here I had to change that: there is a lot of bureaucracy, inefficiencies, and slower working paces. For instance, during the rainy season the data collection had to stop in some areas because of the roads being flooded. Before coming here I didn’t even think about all these practical problems. It is almost impossible to follow the plans and meet the deadlines that you impose to yourself. Related to this issue, physical fatigue in developing countries is an issue and shouldn’t be underestimated when planning the workload. It is quite common that the people you are working with (and yourself) get sick due to parasites, worms or other diseases.

But I need to say that if these can be considered challenges on one hand, on the other they really help you understand what is  feasible in conducting research and what the respondents can reasonably understand. To carry out meaningful research, it’s mandatory to do an experience in the field!

Do you think that something could be done in a different way?

 I think that in general there is room to invest more in the local staff and in capacity building of the NGO. That is essential to have everybody on board, to have everybody really fully understand why they are doing what they are doing and why it is really relevant. For instance, it is important to explain the importance of randomization to avoid that enumerators involuntarily affect the allocation of the treatment! More attention to the quality of data collection would also be desirable. It’s a delicate phase that shouldn’t be outsourced.

What are the main personal challenges you encountered?

 My main difficulty has been finding the key to interact, communicate, and work with Ugandans. It required time to learn how to interact with a new culture, with a different sense of humor, a different way of reasoning, different interests, conversation topics, ways of having fun…

What is most satisfying in your job?

 After all the work you put in the data collection phase, seeing that the data come in and everything goes smoothly gives you a sense of accomplishment. When the first submitted interview comes in the server, we are all very enthusiastic. Also seeing that an enumerator really understood the meaning of a question and benefitted from my explanations and from the training gives me great satisfaction. Moreover, it’s nice to feel that you are having some impact and to feel the gratitude of people here when I tell them that I work for BRAC.

Do you have any advice for a student interested in Development Economics?

 I definitely would like to stress the importance of a significant field experience in a developing country for at least one year. In a field experience you usually have many more responsibilities than in an internship in the headquarters of an international organization, and you learn many more things in the same time span. Of course studying and getting good grades is important, also as signaling. And if the grades aren’t so good, try to compensate with work experience and collaborations of any kind. Initially you might want to arm yourself with patience and you should be prepared to work hard and often for free. Also keep in mind that networking plays, as always, an important role since the formal channel isn’t always the only way to go. Don’t be afraid to be cheeky; take the initiative, write emails to people working in the field you wish to work in, look for opportunities!

 

 

Viola Corradini

 

 

Ethnic Fractionalization and Segregation in Developed Societies

18102405_4ebbc40a70

Two years ago I spent four months in Chicago, studying at Northwestern University. The campus is very close to downtown and I used to take advantage of my spare time to explore the “Windy City”. I was very impressed by the lights of the skyscrapers and the crowded beaches along lake Michigan, but when I asked for advice on new places to see, everyone would warn me: “Don’t go south, it is dangerous”.

The south of the metropolitan area of Chicago is known for being the place where gangs organized their illegal activities; still today the crime rate is very high and shootings are shockingly frequent. Driven by curiosity, I decided to go and personally have a look. Everything seemed regular, until I turned a corner and the scenario in front of me changed completely, as I was entering the historic black neighborhood: Bronzeville. Black children were playing on the streets, and shops were crowded with African-Americans. In front of a pub on Martin Luther King Drive, some black men stared at me while drinking their beer. It was probably quite unusual for them to see white people around there. At the same time, it was very unusual for me to see so many African-Americans all together: I could rarely see them in the richer parts of Chicago.

Later on I discovered that many black people settled in this area between 1910 and 1920 to escape from oppression in the southern States. At the beginning the area was flourishing, but it soon became overcrowded, as black people were impeded to buy houses outside of the “Black Belt”. The neighborhood turned into a very disreputable area, with high crime rate and poor living conditions.

After this occurrence I started thinking skeptically about the urban structure of Chicago. Little Italy, the Irish district, the Greek area, the Black neighborhood are just some examples of the many parts in which the metropolitan area of the city is fragmented. I was asking myself: do societies benefit from ethnic fractionalization? What are the economic and social consequences of urban segregation?

During the Development Economics course I attended at Bocconi University I discovered that many economists have been trying to create theoretical models and to analyze data to understand more in detail the causes and the implications of heterogeneous societies. I was also given the basic instruments to understand the large literature about this topic and I became familiar with measures of poverty, ethnic fractionalization and segregation.

There are many theories and predictions about this topic. Some researchers value ethnic fractionalization and claim that it generates growth, as various abilities and inputs can be combined together in the production function of the economy. Other authors suggest that the beneficial effect of ethnic heterogeneity is particularly strong in more advanced societies, where these various resources are efficiently combined together. Conversely, less developed societies could have difficulties in coordinating inputs of different ethnic groups. In many heterogeneous societies the amount of desired public good is lower compared to more homogeneous communities. Conflicting preferences over type and amount of public goods could explain this phenomenon. Alternatively, public good production could be easier in homogeneous societies whose members share common language and knowledge. At the same time in homogeneous communities it is easier to find and punish free riders. Furthermore, studies report a negative effect of ethnic segregation on the quality of government (measured considering accountability, stability, rule of law, control of corruption). In fact, segregation reduces contacts among different racial groups and consequently reduces trust across them.

To sum up, it seems that ethnic heterogeneity is beneficial when the resources brought by the different ethnic groups are efficiently shared and combined together, but detrimental to the quality of government when ethnic groups are strongly segregated and isolated.

It could be that cities like Chicago, with an advanced economy and fractionalized population, are not benefiting from the positive effects of ethnic heterogeneity due to its urban segregation. At the same time, racial segregation might have negative economic and social consequences on people living in segregated areas. In the United States, for instance, elementary and secondary education is largely financed by property tax. This means that poor neighborhoods can afford lower quality schools and teachers. Poor education translates into bad performance in the labor market and even more persistent economic and urban segregation. Could this vicious cycle be interrupted through a different financial support of schools? Could poverty decrease by designing urban plans that help removing the barriers to buy houses in areas with high quality public services? These and many other questions are still to be further analyzed.

What I learned from this visit to Bronzeville and even more after a field experience in Africa is that an attentive and curious observation of the real world is often at the basis of economic research. LEAP, the new poverty lab at Bocconi University, by promoting students’ research activity and collaborating with NGOs in Italy and abroad, tries to investigate on how to concretely alleviate poverty and reduce inequality. This way, research becomes compelling, effective and socially relevant.

Sara Spaziani

A Day in the Field

img_1957

I remember it was a cold March in Milan. I was sitting in class as Professor La Ferrara was teaching the Marshallian Hypothesis, a theory that seeks to explain the efficiency gains from a variety of contractual forms. The purpose of that lecture was to demonstrate why land markets in many developing countries show coexistence of fixed rent and sharecropping tenancy contracts, although the latter is less efficient. If you attended that class, you would know that the risk aversion can partially explain why we observe this pattern. However, I remember getting lost in thought and wondering “What does it really mean? What does it feel like being a farmer in Africa who has to face that decision?”.

Few months later I found myself in a field in Kamuli with a beautiful view of the lush Ugandan landscape. I was interning at BRAC, a development organisation dedicated to alleviating poverty by empowering the poor, for a project aimed at testing the Marshallian Hypothesis experimentally. I recall I was really excited because I was going to have the opportunity to meet a real farmer and obtain some answers to my question. But reality is always different from expectations. I realised we couldn’t speak the same language, and that she looked really distracted. All of a sudden, I got the feeling that my presence, a “muzungu” (white person) landed from Italy in the middle of Uganda, seemed a little odd.

That day I realised that I went to Uganda with this simple question and I found out that the world is much more complex than a model, that I had so many more questions in my mind and that the woman herself, farming in the sun with a baby strapped on her back, must have been wondering “Why is this muzungu here?”

So, here I am, one year after that experience still wondering why those days in Uganda have changed my life so much and why I would recommend every student who is passionate about development economics, and wants to put his interest and skills to action, to pack his bags and head towards a developing country for an experience in the field. I guess the first questions that come to mind are “Am I ready for this experience? What can I expect?” Since I had the same doubts almost a year ago, let me share what the elements of a typical day in the field looks like.

Be ready, because you will see things that are not considered normal by most of your peers; not all of them will be pleasant, but they will be challenging, fascinating and sometimes exciting. You can find yourself stuck in chaotic traffic in a minivan for hours, listening to a street preacher that reminds you that God is forgiving, while a local seated next to you tries to sell you the chicken he holds in his bare hands. You could meet a cheerful group of children all perfectly dressed up for school walking in the street between an open-air garbage dump and stray cows. Or you could even be stuck in the mud with a GPS tracker in your hand trying to measure the coordinates of a field.

According to my experience, there is no “typical” day, as I have been involved in many aspects of the project. My work largely depended on the stage of the Randomized Control Trial; it was between the midline and the last follow-up surveys. The internship entailed field visits to the project districts in addition to analytical work from the country office in Kampala. This meant that my days were basically divided between getting intimate with Stata, partaking in numerous meetings in the country office, and traveling to the field to ensure the regularity of the data collection process.

As far as the analytical and managerial work is concerned, my duty was to coordinate the team members in the field (20 enumerators and 6 field supervisors), to write daily reports on the data collection progress (trends, challenges and additional findings from the field) and to spend most of the time on the data cleaning and analysis.

Aside from this daily work at the office, I had the opportunity to go in the field and meet the enumerators and program administrators to solve the problems directly on site. Sometimes there were technical problems with the tablets, sometimes it was just a communication issue, and other times it was a problem of incentives. You can understand that it was truly a “hands on” experience that brought me into the real world, unlike a conventional desk internship. This is why I think studying poverty is intriguing but direct exposure to the field is crucial, as it allows us to participate substantively in many issues that we normally discuss only passingly in class.

Perhaps, the most unexpected aspect of working for an NGO in a developing country is the fact that you’ll find yourself constantly inspired by the sincerity, determination and resourcefulness of the people with whom you are working. My experience in Uganda was honestly monumental in sharpening my abilities and in developing a holistic interest in poverty alleviation programs. I still feel lucky I was side by side with truly inspiring people, working to overcome difficult situations despite unfavourable odds.

With this post I hope I have inspired you to seek your own field experiences. If so, I invite you to get involved in the activities of  LEAP, the new poverty lab at Bocconi University.

Emilio Dal Re