Open or close the menu

Field experiments and quasi-experiments in marketing and behavioral economics

Time & place: 13.5.-15.5.2019, Hanken, Helsinki

Target group

The course is primarily meant for doctoral students in marketing as well as in behavioral economics, as the course readings reside mostly in these disciplines. However, the methodologies covered on the course are well applicable in research of other disciplines of business administration, economics, and behavioral/social sciences, as well. Thus, students of other disciplines than marketing and behavioral economics are welcome, too.

The course is targeted to doctoral students affiliated with universities in Scandinavian/Nordic countries as well as Baltics. The course is free, but students must pay for travel, accommodation, and meals themselves.

A limited number of participants is admitted to the course (see Quota below).

Description

The course provides methodological understanding of and insights to designing, implementing, and analysing random controlled experiments in real-market/real-life field settings (i.e. outside laboratory and survey/scenario settings) , as well as to analysing field data and secondary register data related to natural-experimental and quasi-experimental settings. The focus of the course is on such aspects of field experiments and quasi-experimental analyses on field data that go beyond standard aspects of experimental methodology (as applied in , e.g., consumer psychology). Thus, this course serves as a complementary, continuation course to Kataja’s introductory doctoral course on experimental methods:  “Experiments as a research method”.

Learning goals

After completing this course, you will

  • be familiar with several data gathering and analysis techniques (i) to overcome limitations of field settings as opposed to laboratory experiment settings (when it comes to inability to e.g., experimentally manipulate all factors of interest and/or directly examining/observing theoretical process mechanisms), as well as (ii) to leverage the advantages of field settings (e.g., assessing heterogeneity).
  • be able to plan and implement data gathering and analysis for various field settings
  • be able to outline overall article setups – including field experiments and/or quasi-experimental analyses combined with additional data/studies – in order to have the articles published in high quality journals of marketing or economics

Course credits and student workload:

The course consists of two parts, that give separate credits: first part 4 ECTS and second part 3 ECTS. When accepted to the course, you can either (a) take only the first part, or (b) take both the first and second part. It is not possible to only take the second part. The first part is an independent learning course with literature reading and exam. The second part is an interactive class/seminar course with paper assignments. Both parts are evaluated on a scale 0-5.

Part I: Literature and exam, 4 ECTS

  • individual/independent work (reading) 102h
  • exam (e-exam, or classroom exam at Hanken on May 13th morning): 3h

Part II: Class and seminar, 3 ECTS

  • individual/independent work (pre-assignment presentation slides; post-seminar paper assignment): 66 h
  • contact teaching/seminar (at Hanken: May 13th-15th): 14 h

Instruction:

For Part I,

  • reading a selection of academic literature, and
  • taking a literature exam.

 

For Part II,

  • pre-assignment based on literature (preparing, individually, as 15-minute presentation on 1 article —  or on 2 articles in pairs)
  • mandatory presence and participation in classroom /seminar discussions (May 13th-15th, 2019, Hanken),
  • short classroom presentation on the pre-assignment, and
  • post-seminar paper assignment (writing a preliminary plan for applying some of the data gathering and analysis techniques introduced in the academic articles on one’s own research topic)

 

Course literature:

Part I

Field Experiments, Basic Models

 

  1. List, John A., & David Lucking-Reiley (2002), “The Effects of Seed Money and Refunds on Charitable Giving: Experimental Evidence from a University Capital Campaign”, Journal of Political Economy, 110(1), 215-232.
  2. Smeets, P., R. Bauer, & U. Gneezy (2015), “Giving Behavior of Millionaires,” Proceedings of the National Academy of Sciences, 34, 10641-10644.
  3. Andrews, Michelle, Xueming Luo, Zheng Fang, & Jaakko Aspara (2014), “Cause Marketing Effectiveness and the Moderating Role of Price Discounts,” Journal of Marketing 78, no. 6 (2014): 120-142.
  4. Tucker, Catherine E. (2015), “The Reach and Persuasiveness of Viral Video Ads,” Marketing Science 34(2), 281-296.

 

Field Experiments and Heterogeneity Models

 

  1. Lambrecht, Anja, & Catherine Tucker (2013), “When Does Retargeting Work? Information Specificity in Online Advertising.” Journal of Marketing Research, 50(5), 561-576.
  2. Li, C., X. Luo, C. Zhang, Wang (2017), “Weather and Mobile Promotions: A 10-Million-Users Field Study” Marketing Science, 36 (5), September, 762–779.

 

Field experiments Combined with Structural Models

 

  1. Dube, J-P., X. Luo, and Z. Fang (2017), “Self-Signaling and Prosocial Behavior: Cause Marketing Field Experiments,” Marketing Science, 36(6), 36(2), 161–186.
  2. Gneezy, A., U. Gneezy, L.D. Nelson, and A. Brown (2010) “Shared Social Responsibility: A Field Experiment in Pay-What-You-Want Pricing and Charitable Giving,” Science, 329 (5989), 325-327.

 

Quasi-Experiments, Natural Experiments and Diff-in-Diff Models

 

  1. Goldfarb, Avi , & Tucker, Catherine E. (2014), “Conducting Research with Quasi-Experiments: A Guide for Marketers” (March 28, 2014),  Rotman School of Management Working Paper No. 2420920. Available at SSRN: https://ssrn.com/abstract=2420920 or http://dx.doi.org/10.2139/ssrn.2420920
  2. Mayzlin, Dina, Yaniv Dover, & Judith Chevalier (2014), “Promotional Reviews: An Empirical Investigation of Online Review Manipulation”, American Economic Review, 104(8), 2421-55.
  3. Gallino, Santiago, and Antonio Moreno. “Integration of online and offline channels in retail: The impact of sharing reliable inventory availability information.” Management Science 60, no. 6 (2014): 1434-1451.

Field Experiments Combined with Game Theory Models

  1. Fong, Nathan M., Zheng Fang, & Xueming Luo (2015), “Geo-Conquesting: Competitive Locational Targeting of Mobile Promotions”, Journal of Marketing Research, 52(5), 726-735.
  2. Dube, J.-P., Z. Fang, N. Fong, & X. Luo (2017), “Competitive Prices Targeting with Smartphone Coupons,” Marketing Science, 36(6), November, 944-975.

Field Experiments Combined with Social Network Models

  1. Bond, Robert M., Christopher J. Fariss, Jason J. Jones, Adam D. I. Kramer, Cameron Marlow, Jaime E. Settle, James H. Fowler (2012), “A 61-million-person experiment in social influence and political mobilization”, Nature, 489, 295–298.
  2. Ascarza, Eva, Peter Ebbes, Oded Netzer , Matt Danielson (2016), “Beyond the Target Customer: Social Effects of CRM Campaigns”, Journal of Marketing Research, 54 (June), 347-363.

Field Experiments combined with Machine Learning

  1. Cui, Ruomeng, Santiago Gallino, Antonio Moreno, and Dennis J. Zhang (2018) “The operational value of social media information.” Production and Operations Management 27 (10),  1749-1769.
  2. Luo, X, N. Fong, Y. Zhang, & X. Wang (2019). Targeted Promotions on an E-Book Platform: Crowding Out, Heterogeneity, and Opportunity Costs.  Journal of Marketing Research, forthcoming.
  3. Luo, X., X. Lu, & A. Li (forthcoming). When and How to Leverage E-commerce Cart Targeting (ECT): The Relative and Moderated Effects of Scarcity and Price Incentives. Information Systems Research, conditional acceptance.

The literature exam consists of a random sample of 3 questions, which are of the following format:

“Please explain briefly in what way do method(s) X (applied in study by N.N. et al.) allow to

  1. overcome one or several of the following general limitations of field settings (as opposed to laboratory settings)
  • inability to experimentally manipulate a key factor of interest
  • inability to directly observe/test the theoretical process mechanism causing (or mediating between) the effect of the independent variable on the dependent variable/outcome
  • inability to observe the experimental treatment to which an individual got assigned, or the individual-level outcome
  • inability to have controlled, homogeneous participants,
  • inability to fully randomize the treatments
  • other limitations, which?
  1. leverage one or several advantages of the following advantages of field settings
  • ability to efficiently implement more treatment conditions
  • ability to analyse heterogeneity in participants or effects
  • other advantages, which?”

In other words, three questions of this same format are asked, with X/N.N. replaced with three methods/articles included in the reading package.  

You cannot take part in Part II of the course, unless you have taken the literature exam of Part I. The exam will take place in the morning of the first day of Part II of the course (i.e., May 13th, 8am). A quick pass- vs fail check on the classroom exams will be made immediately after the exam (exam @ 8-11am; pass check @11m-12am; Part II seminar starts @12:15am). A full grading of the exam will be made in two weeks after May 13th.

In case you only take Part I of the course, by e-exam (see below), you will receive your grade in two weeks after taking the e-exam.

Part II

The aforementioned articles serve as the literature for the Part II of the course, as well. After the course registration deadline has passed, students will be contacted about who (individuals or pairs) gets to pre-prepare a presentation (15 mins) about which of the articles.

In addition, readings in Part II include lecture handouts and statistical software tips provided by the main instructor, Prof. Xueming Luo.

Assessment:

Part I: grade 0-5

  • Examination 100%

Part II: grade 0-5

  • Pre-assignment presentation slides 50%
  • Post-seminar paper assignment 50%

Recommended Timing:

The course is advised to be taken after the student has attended a basic course on (or gotten familiar with principles of) experimental methods (in e.g.,  consumer psychology).

 

Quota

A max of 25 students are admitted to the course.

 

Additional information

Students get temporary user-IDs to Hanken’s IT systems after signing up to the course.

If you only take Part I of the course, you can also take the exam digitally, as an “e-exam”, at Hanken in Helsinki or Vaasa at any time between April 15st, 2019, and June 14th, 2019. A link to signing up to the e-exam can be found on the course’s Moodle page by April 8th, 2019. The examination room with computers is situated at Hanken (Helsinki and Vaasa). The e-exam can be taken only on Hanken premises. Yet, it can be taken on all days except Sundays, from morning to evening. The examination time is approximately 3 hours.

Upon passing the course, your passing the course will be registered in Hanken’s electronic database. You will get a transcript of records from the study office.

 

To sign up for the course (Part I or PartI&II),please register on the following webpage by March 31st, 2019.

https://www.webropolsurveys.com/S/18C7801D6F7C0F3C.par

 

You will be notified about admission to the course by April 8th, 2019.

 

Examiners 

Professor Xueming Luo, Temple University, Fox School of Business

 

Professor Jaakko Aspara, Hanken School of Economics

 

Prof. Jaakko Aspara will act as coordinator of the course. You can ask further information from him: Jaakko.aspara@hanken.fi

 

Additional instructions on the assignments of Part II

Pre-assignment presentation

After receiving a confirmation of your registration on the course, including information about which of the articles (listed above for Part I) you should prepare a presention for:

  1. Prepare a 15 min slideshow/presentation (max 7 slides) which summarizes
  2. the research questions/hypotheses of the article
  3. the field setting/data addressed in the study
  4. special analyses or techniques applied on the data
  5. how the analyses/techniques in c) help to (i)  overcome limitations of the field setting, and (ii) leverage its advantages , vis-à-vis laboratory setting (see some potential limitations and advantages in the example exam question above, under Part I)
  6. complementary studies/data used in the article (as complementing the field data) – and why
  7. assumption/guess of what aspects of the study the reviewers have considered particularly impressive
  8. Prepare to present your slideshow in class, in the corresponding session outlined in table below

Post-assignment paper

  1. Write a 3-page study plan (Letter size; 1-inch margins all sides; line spacing 1.5 lines, font Arial 11pt)

1)     A 1 page description of the theoretical topic/problem that you intend to investigate with the field experiment (or quasi or natural experiment), and what research question the experiment will answer. Formulate at least one hypothesis to be tested in the study. Include a few relevant references to prior research on the topic/problem, and also briefly indicate whether or not prior studies have adopted field experimental/data approach (and if not, what method/data have they used). You can also replicate an existing study, perhaps adding some new moderating factors.

2)     A 2-page tightly written description of the

  1. field experimental design, outlining the treatment, the dependent and independent variables to be tested, a detailed list of measures (can be put in an appendix) with sources (where are the measures taken from, references), and analysis methods/models,
    or
  2. quasi-experimental field data, as well as analysis methods and models to be applied

3)     A list of references.