Header image

Behavioural Science Lab


Gallup Pakistan has setup Pakistan’s first Behavioural Science Research Lab, working with a Behavioural Science Ph.D. Research Fellow at Warwick Business School. Similarly, Gallup Pakistan is working in the field of advance data analytics and using concepts such as ‘Predictive Modeling” for providing better future insights to our clients. In a landmark study, Gallup’s researchers assisted a Yale University study in application of Conjoint Analysis on electoral outcomes.

The focus of the research being undertaken is to conduct lab experiments to develop behavior change interventions. This approach is becoming increasingly popular in the enhancement of evidence based policy making.

Gallup Pakistan’s Behavioural Science Studies:

1. ‘Measuring Social Integration and Community Cohesion in Pakistan’

Gallup Pakistan working with Centre for Decision Research and Experimental Economics (CeDEx) at the University of Nottingham and UNDP Pakistan

 Academic researchers involved: Professor Chris Starmer, Dr Emily Wyman and Dr Umar Taj

Subjective perceptions of closeness to others can be easily measured, and have been shown to reliably index aspects of social relationships that affect decision-making processes and so determine concrete outcomes for individuals and groups. Pertinent to measuring social integration, one branch of ongoing research at CeDEx has investigated ‘Oneness’, a measure of how subjectively close people feel to other individuals. CeDEx research has shown, firstly, that this simple tool can be used to reliably index an impressive range of relationship attributes (e.g. how often individuals interact, how close they feel) and, secondly, that Oneness scores strongly predict levels of group cooperation.

Oneness might be adapted in different ways for different purposes. For instance, they might be used to assess basic variations in vulnerability between populations (e.g. to assess whether youths in certain locales perceive themselves as particularly distant from family, or key figures in the local community). They might also be used to address hypotheses about the efficacy of specific interventions (e.g. how community centre or skills training interventions increase a sense of youth integration and group cohesion, or reduce identification with high-risk groups).

Due to socioeconomic conditions, youth populations in some parts of Karachi have been identified as at special risk of vulnerability. As part of a strategy for tackling this issue, UNDP Pakistan has been providing skills training to youth from these vulnerable areas of Karachi and helping them find employment in the textile and garment industry in Karachi. This project comprises a field experiment that will, in collaboration with UNDP Pakistan use Oneness to empirically and quantitatively monitor changes in youth integration and social cohesion.

2. Can media advertisements change public attitude?    

Gallup Pakistan is collaborating with Avri Bilovich and Umar Taj (Research Fellows in the Behavioural Science Group at Warwick Business School) to understand the role that media advertisements play in changing public attitude. This is the first of its kind study in Pakistan and in response to finding an answer to the question that we get asked repeatedly by our clients: do media advertisements change public attitude.

The research involves using direct and indirect preference elicitation methods with the aim of exploring the effects of different types of media advertisements (informationally rich and informationally poor) on modifying political attitudes. Additionally, this project will inform us on the malleability of direct and indirect elicited attitudes which is methodologically important for understanding the drivers of people’s behaviour.

The methods being used are: Discrete Choice Experiments, Rating Scales, Likert Scales, and Free response.

3. Microfinance and Consumer Decision Making: Applying Behavioural Economics to Improve Product Design in Pakistan

Pakistan has one of the lowest financial inclusion figures in South Asia with only 16% of the population formally banked and 23% using formal financial services. These figures contrast significantly with the regional average of 46%. There are many reasons to believe there is a strong unmet demand for these financial services: more than half of the Pakistani population saves (but only 8% in formal financial institutions); one-third of the population borrows (but only 3% using formal credit products). The lack of engagement in formal financial institutions presents an exciting opportunity to understand what are the psychological and behavioural factors that currently obstruct consumers from using these services. These insights can also help inform us to design financial products that best fit the populations they are designed to serve.

There has been a growing body of research highlighting the importance of psychological and behavioural biases that play a role in consumer’s financial decision making. People frequently engage in financial behaviours that run counter to their long-term well-being: they save too little, spend too much, and fail to avoid unnecessary fees.

Gallup Pakistan is collaborating with Joe Gladstone (Assistant Professor of Consumer Behaviour at UCL School of Management) and Umar Taj (Research Fellow in the Behavioural Science Group at Warwick Business School) to measure behavioural biases through a series of choice experiments and observe how certain biases lead people to make sub-optimal decisions. In particular, we will be interested in:

  • Pakistani people’s ability to make optimal investment decisions in very simplified investment tasks
  • How framing of information and complexity in the characteristics of the products affect consumer’s decision making in the Pakistani context?
  • Investigating whether current products on the market would be more attractive if elements of the costs were re-formatted

4. News Sharing and Voting on Social Networks: An Experimental Study

Gallup Pakistan collaborated with Professor Matthew Shum (J. Stanley Johnson Professor of Economics in the Division of Humanities and Social Sciences at California Institute of Technology), Kirill Pogorelskiy (Assistant Professor of Economics at The University of Warwick) and Umar Taj (Research Fellow in the Behavioural Science Group at Warwick Business School) to study the relationship between news sharing and voting on social networks.

Abstract of the forthcoming paper:

We study the relationship between social media and information aggregation by voting. Our experimental treatments mimic the features of social networks in the presence of media bias in order to address concerns that voters obtaining their political news via social media may become more polarized in their voting behaviour. Our results suggest substantial effects of polarization at the expense of efficient information aggregation: subjects engage in “liking” behaviour by selectively sharing the news that is favourable to their party and down weighting unfavourable news in their voting decisions. At the same time, social networks raise collective decision making efficiency, if there is no media bias. All in all, subjects behave as if information sharing and voting is expressive of their party preferences even when (by design) their preferences have a common value component. We supplement lab results with survey evidence from Pakistan, and find that it is broadly consistent with these patterns.

You can download the paper from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2972231

5. Can media advertisements change public attitude?   

The field of behavioural science has gained a lot of traction over the past few years. However, most if not all of the research studies that led to discovery of the latest behavioural insights have involved W.E.I.R.D participants: Western, Educated, Industrialised, Rich and Democratic.

We want to be sure that these latest behavioural insights apply within the Pakistani context as well. In this regard Gallup Pakistan has started a behavioural science replication series where every week we are replicating the classic studies in the field of behavioural science with the Pakistani population as the participants. We will be sharing the results soon with you through our Monthly Behavioural Economics Newsletter