Sky High Insights: Analyzing Consumer Behavior and Pricing Strategies in the Airline Industry

How to cite

Albarez, Mark; Wong, Jason. 2025. "Sky High Insights: Analyzing Consumer Behavior and Pricing Strategies in the Airline Industry." AEA RCT Registry. AEARCTR-0015661. https://doi.org/10.1257/rct.15661-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Mark Albarez* , Jason Wong

* Primary investigator

Registry ID
AEARCTR-0015661
DOI
https://doi.org/10.1257/rct.15661-1.0
First registered
Last updated
Status
On going
Trial dates
2025-01-21 – 2025-05-09
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15661

Abstract

Current research on airline consumer willingness to pay and preferences focus on existing categories of products and services that are fundamental to a full-service airline. For instance, there is much research on how seat pitch, or the distance between seats, in airline economy and premium economy seats impacts a plethora of consumer behaviors. Other research also focuses on fundamental factors such as seat width, seat recline, and service quality, as well as ancillary additions (i.e., baggage, meals, etc.), all of which have grown to become expected insertions or add-ons to a full-service airline ticket. Innovative products, on the other hand, are different, as they are features that are not expected by consumers and ones that are not critical to an airline seat’s function. Recently, airlines have begun implementations of innovative features across all classes of service. Examples of such innovations include Emirates’s first-class showers aboard Airbus A380 aircraft, Virgin Atlantic’s inclusion of business-class on-board bars, no forward seat recline from Lufthansa’s Allegris premium economy, the skycouch aboard Air New Zealand economy, and well-being zones for all cabins on Qantas project sunrise flights. From this short list of examples it becomes apparent that methods by which to differentiate do not pertain solely to a single flying class, and with this increasing behavior from airlines it is crucial to obtain an understanding regarding how these innovations impact consumer willingness to pay.
Innovative features are especially interesting within economy class. In an environment where many airlines are restricting space and removing enhancements to economy class, a few other airlines are doing just the opposite. The aforementioned examples of Air New Zealand and Qantas are a few of such, but it is especially important to understand how economy class innovative features are perceived by passengers, as these are the ones that will be experienced by the largest number of people. Therefore, this study, seeks to understand this exact gap in current knowledge. To do so, I will use a conjoint analysis to study willingness to pay for both fundamental and innovative features – using fundamental features and their existing knowledge as a baseline. From the conjoint results, I will be able to infer, conditional to fundamental features, how much customers are willing to pay for a set of new innovative features.
In this pre-analysis plan, I outline the methodology, hypothesis, variables, sampling, and data collection procedures.

Experimental design

Participant Recruiting and Sampling
This study focuses on the route between New York City and Los Angeles, as such, our research is restricted to respondents who live in New York State of California, and who have previously travelled on an airplane. Our respondent criteria include 1, participants must be from the states of New York or California, 2, must be over the age of 18, 3, must have flown on a US based carrier during their lifetime, and 4, must have taken a domestic flight within the past 3 years. Each of these restrictions is either enforced through the respondent recruiting platform, or through built-in screen-out questions.
We used Qualtrics to design the conjoint survey and Prolific as the respondent recruiting platform. Prolific has a strong reputation for accurate random samples with participants who are engaged and serious about their given research tasks. Secondly, Prolific offers hundreds of built in filtered that negated the use of excessive screening questions within the survey. Lastly, Prolific’s platform has an intuitive interface that interacted seamlessly with Qualtrics - to ensure data security and quality.
Within the Prolific sample, 13,796 live in New York State and California. After filtering for age (criteria 2) and airline history (criteria 3), we are left with an eligible sample of 8,846. From here we chose a random minimum sample of 320 was also selected – to have reasonable power. Each respondent will be given a fixed compensation averaging $17.00/hour – higher than minimum wage in both states – for their time and effort participating in the survey, the only criteria for non-eligibility will be an indication of a non-human response.

5.2 Survey Methodology
Qualtrics is the chosen software to host the survey. Qualtrics allows us to have flexibility and accuracy within our analysis and data collection process. As one of the largest online survey platforms, Qualtrics has a host of tools that allow the survey to include any customized component, and their new conjoint analysis software permits the research to gain overall efficiency and accuracy with their intuitive user interface and automatically generating statistics (i.e., utility, willingness to pay, cost analysis, etc.). The survey is broken down into the following sections:
1. Participant Consent
2. Screening
3. General Travel Questions
4. Awareness Check
5. Conjoint Questions (1/8)
6. Reflection
7. Demographics
A few notes from each of the sections. Firstly, participant consent is given as to inform each participant that their data will be used and stored for this research and possible subsequent studies, but also that each of their responses are kept strictly anonymous. This was a requirement to obtain Institutional Review Board approval (see section 5.3), and for any human based studies to proceed. The screening section ensures that participants met criteria 4 that did not have a filter on the Prolific pre-study screens. An awareness check is given before the conjoint analysis begins – a simple question asking to complete 5+8 in numeric terms – this ensures that participants are completing the survey with attention in mind. There are a total of 8 conjoint analysis questions, this is a number predetermined by the Qualtrics software based upon the number of attributes and levels included.

Design details

Randomization method
Randomization completed through Prolific systems
Randomization unit
Participants will be recruited from Prolific, where approximately 13,796 individuals reside in New York State or California. After applying screening criteria related to age, airline history, and domestic travel recency, an eligible pool of approximately 8,846 individuals will remain. From this pool, 324 respondents will be randomly selected to complete the conjoint survey hosted on Qualtrics. Within the survey, the attribute levels for each bundle (Seat Type A and Seat Type B) will be randomly assigned using Qualtrics' conjoint module, which generates orthogonal and balanced experimental designs. As a result, each respondent will face randomized combinations of airline product features in each choice task. Because each bundle is randomized independently, the study will not define traditional "treatment" and "control" groups; rather, treatment effects will be inferred by comparing choices between two randomly constructed alternatives across repeated tasks.
Clusters
1 group of Prolific respondents
Observations
324 participants
Arms
8846 Prolific participants who fit the selection criteria
Primary outcomes
Price, willingness to pay for 6 varying attributes

Cited by

No citation data available for this registration yet.

Keywords