Clients are asking how to think about survey research during the COVID-19 global pandemic. Specific questions include:
- Will data be biased? Will self-reported behaviors, preferences and attitudes be biased by the fact that many consumers are living outside of their normal routines right now?
- Should we be running research right now? What types of research should we avoid and what types of research should we continue or even begin?
- What steps should we be taking in survey design and fielding to mitigate bias?
First off, it is undoubtedly the case that consumer attitudes and behaviors have changed in recent weeks due to the outbreak of COVID-19 and its spread across the globe. Travel bans, school closings, the implementation of work from home policies, along with other social distancing measures, have dramatically changed daily life for many Americans and consumers all over the world. The extent to which these changes should be thought of as conditions that are producing non-normative survey data vs.a new normal that consumer-facing organizations will need to understand is an evolving issue, which we will seek to understand in a data-driven way, by broadly tracking consumer behaviors, sentiment, priorities and attitudes in the coming weeks and months.
Some areas where we may see changes in survey data/consumer responses:
Behaviors are the most obvious area in which we expect to see change. Some that will be relevant to the majority of our clients are:
- Purchase frequency – Consumers may stockpile more basic supplies (e.g., food and household needs) and, conversely, forego the purchase of less critical products and services.
- Purchase channel – Many consumers will be moving toward online purchases and home delivery and/or curbside pick up to avoid crowded grocery stores and mass retailers.
- Brand switching – We expect to see much less brand switching than retailer switching, except insofar as a) supply chain shortages require them and b) consumers are looking for more function in a product category in which they had previous valued other characteristics (e.g., we may see some consumers move away from “natural” household cleaners in favor of products with antimicrobial ingredients).
- Behaviors around entertainment and leisure time (e.g., less travel, fewer social activities and perhaps increased digital behaviors such as the use of social media and streaming services, etc.).
- Risk aversion in forward-looking predictions – Consumers seek to reduce risk during times of uncertainty.
Attitudes, Priorities, and Preferences:
We may expect to see that heightened existential and economic anxiety lead to shifts in attitudes, perceptions, and preferences, such as:
- Increased brand and product loyalty, as consumers derive comfort from familiar and trusted sources.
- Increased preference for function over form in product and service design.
- Increased focus on safety and security, avoiding pain/negative outcomes; decreased prioritization of pleasure and aspirations, and approaching positive outcomes.
Do we expect that there will be differences by market?
We should expect that markets where COVID-19 has spread more widely will show more evidence of behavior change, priority shifts, etc. than those with fewer cases. This will likely be true at the country level (e.g., Italy vs. US), and also at the state and city level (e.g., NYC vs. Dallas).
Do we expect any changes that are specific to panels and not a reflection of the population at large?
No, we do not. Whatever changes we see in survey data we can assume to be reflective of changes in the behaviors of the general population. In fact, it’s possible that we may see quicker/increased response rates to survey invitations, given that many consumers are not able to engage in their normal daily or weekend activities, but the great majority remain relatively healthy.
Should we make any changes to our fielding approaches?
In markets and areas where the majority of the population is healthy and/or asymptomatic or only mildly ill, we do not expect to see any changes in fulfillment or fielding times. In regions that have been hit the hardest (e.g., Wuhan Province, China; Lombardy, Italy) we may expect to see longer field times and/or issues with fulfillment with hard to reach audiences; however, such delays will likely be temporary and associated with a 2-4 week period during the virus’s peak. Aytm will monitor the situation closely and advise clients if any fielding delays are expected and/or if a hard to reach group (e.g., the elderly in hard-hit areas) cannot be fulfilled.
Should we make any changes to what kind of research we are doing and/or how we ask survey questions during this time?
This is a difficult question to answer and there is likely no right answer, particularly given that it remains to be seen whether this period is a “blip” or “the new normal.” Generally, we would expect research around new product development (feature prioritization, line optimization, concept testing) to be relatively robust and stable to current conditions. Likewise, we expect brand awareness, preference, and perceptions to generally be stable. Research around shopper insights and purchase behaviors, as well as habits and practices, may see some significant shifts (but again, this can be valuable information). Predictions of future behavior may see an impact as a risk-averse mindset may lead consumers to underestimate spending and leisure behavior and overestimate safety and security-related behaviors. With regard to reports on current or typical behaviors (particularly if a client seeks to get a “pre-COVID-19 read” on consumer behavior), it may help to phrase questions in a longer-term way. Instead of asking “How do you prefer to purchase Product X?”, choose more specific language: Over the past 6 months, how have you most commonly purchased Product X?
Our team of research experts is actively working to produce ongoing data on the impact COVID-19 is having on consumer behaviors and attitudes that we will continue to share in our blog series in the weeks to come.
First in the series, you can read more about how the outbreak is affecting our panelists and their responses to survey invitations here: The Online Sampling Impact of the Coronavirus Outbreak.
Stay safe and we’re here to answer any questions or offer support if you need it.