Project Activities
The researchers investigated whether an educational intervention, aimed at decreasing whole number bias errors, helped postsecondary education students and adults, ages 18-24, to more accurately solve health-related math problems pertaining to COVID-19 statistics. The participants in the study completed 10 days of daily diaries and were re-contacted to complete a follow-up survey a year later.
Structured Abstract
Setting: All data were collected online with college students or with adult Qualtrics panelists.
Sample: The majority of the college sample (n = 270) self-reported being female. Participants in the Qualtrics panel (n > 1,000) studies were recruited to be representative of the U.S. population by sex, race, ethnicity, and educational attainment.
Intervention: The educational intervention involved analogies, worked examples, and visual displays, such as number lines, which helped adult participants calculate COVID-19 case-fatality rates as they solved health-related math problems.
Research Design and Methods: Researchers conducted two online randomized control trials (RCTs) with adults recruited through online panels. Participants were randomly assigned to a business-as-usual or educational intervention condition. For 10 consecutive days, participants completed evening daily diaries to track the impact of the intervention on risk perceptions and preventive health behaviors (e.g., social distancing, wearing a mask). For both RCTs, participants were re-contacted one year later to complete a follow-up survey. In addition, researchers conducted an online RCT of college students assigned to one of four risk visualization conditions: no visual display, number line visual display, risk ladder, or icon array.
Control Condition: The control condition was a business-as-usual control. That is, participants in the control condition completed the same health-related math problems as those randomly assigned to the educational intervention condition, but they did not engage with the educational intervention, which included a worked example and analogies to a more familiar context.
Key Measures: Key measures included rational number estimation/comparison, objective/subjective numeracy, math anxiety/attitudes, health-related math problem-solving accuracy and strategy reports, risk perceptions, preventive health behaviors, and trait anxiety.
Data Analytic Strategy: The researchers assessed whether health-related math problem-solving accuracy and strategy reports differed across the educational intervention and business-as-usual control conditions with a logistic regression and mixed-effects models. Additionally, linear mixed-effects models were used to assess whether the educational intervention had an impact on risk perceptions and preventive behaviors across the 10-day daily diary period.
Key outcomes
The main findings of this project are as follows:
- The adults who were randomly assigned to the educational intervention were more accurate at answering health-related math problems about COVID-19 than were those who were randomly assigned to the business-as-usual control (Thompson et al., 2021). Those in the educational intervention were less likely to describe whole number bias strategies as the way they solved these health-related math problems.
- The efficacy of the educational intervention was replicated in a second panel of adults (Fitzsimmons et al., 2023). Additionally, for those who learned from the educational intervention, the benefits persisted after a one-day delay.
- Number lines are used less frequently than risk ladders and icon arrays to convey health statistics. However, in a sample of college-aged students, results revealed higher health-related math problem solving accuracy for those who solved problems accompanied by number lines. In fact, performance did not differ on the health-related math problems for those participants who solved the problems with no visual display at all, an icon array, or a risk ladder (Mielicki et al., 2022).
- Changes in approach and avoidance behaviors over the 10 days of daily diaries indicated large individual differences. Discrete emotions, including fear, guilt/shame, and happiness were associated with more recommended preventive behaviors (e.g., hand washing, social distancing). Fear and COVID-19 worry indirectly influenced each other to facilitate more behavioral engagement. While emotions, such as being worried, strongly predicted individual differences in behavior across the 10 days of the daily diary, they did not predict why behaviors occurred on one day versus another (Coifman et al., 2021).
- Cultural worldviews, such as individualism (e.g., valuing individual freedom over the group) and hierarchy (e.g., valuing social stratification over equality) played a significant role in shaping people’s risk believes about COVID-19 (Cheng et al., 2023).
People and institutions involved
IES program contact(s)
Project contributors
Products and publications
Study registration:
Publications:
Find available citations in ERIC for this award here.
Journal articles
Cheng, I., Taber, J., Simonovic, N., Coifman, K. G., Sidney, P. G., & Thompson, C. A. (2023). The associations of cultural worldviews, political orientation, and trust with COVID-19 risk beliefs in the U.S. Social and Personality Psychology Compass, 17(11).
Coifman, K.G., Disabato, D.J., Aurora, P. Seah, T.H.S., Mitchell, B., Simonovic, N., Foust, J.L. Sidney, P., Thompson, C.A., & Taber, J.M. (2021). What drives preventive behaviors during a pandemic? Emotion and Worry. Annals of Behavioral Medicine, 55(8), 791-804.
Fitzsimmons, C. J., Sidney, P. G., Mielicki, M. K., Schiller, L. K., Scheibe, D. A., Taber, J. M., Matthews, P. G., Waters, E. A., Coifman, K. G., & Thompson, C. A. (2023). Worked examples and number lines improve U.S. adults’ understanding of health risks as ratios. Journal of Applied Research in Memory and Cognition. Advance online publication.
Mielicki, M., Schiller, L., Fitzsimmons, C. J., Scheibe, D., Taber, J. M., Sidney, P. G., Matthews, P., Waters, E. A., Coifman, K., & Thompson, C. A. (2023). Number lines can be more effective at facilitating adults' performance on health-related ratio problems than risk ladders and icon arrays. Journal of Experimental Psychology: Applied, 29(3), 529–543.
Thompson, C. A., Taber, J. Sidney, P. G., Fitzsimmons, C. J., Mielicki, M., Matthews, P., Schemmel, E., Simonovic, N., Foust, J., Aurora, P., Seah, T. H. S., Disabato, D., Coifman, K. (2021). Math Matters: A Novel, Brief Educational Intervention Decreases Whole Number Bias When Reasoning about COVID-19. Journal of Experimental Psychology: Applied (Special Issue on Risk Perceptions), 7(4), 632-656.
Available data:
Questions about this project?
To answer additional questions about this project or provide feedback, please contact the program officer.
