Testing Behavioral Incentives
AbstractWhat motivates you to clean your room/mow the lawn/wash the dishes (substitute your own responsibility at home)? What motivates you to do something you really like to do? Or think about this: you'd like to get your younger brother or sister to do you a favor. What strategy works best? Here is a project designed to test which incentive strategy works best for encouraging small children to complete a task.
Andrew Olson, Ph.D., Science Buddies
This project is based on:
- Carson, E.M., 2004. Praise vs. Prize, California State Science Fair Abstract. Retrieved September 5, 2006.
The objective of this experiment is to find out what form of positive reinforcement (praise or a reward) is most effective in influencing four-year-old children to complete a task.
Research has shown that positive reinforcement is an effective means of influencing behavior. This experiment looks further to determine which type of reinforcement—praise or reward—is more effective with young children.
To do this project, you will test a random sample of four-year-old preschoolers to see how quickly they can complete a simple shape-matching task. One group will receive stickers as a reward, the other group will receive verbal praise.
You will need to get permission and assistance from one or more pre-school teachers to do this project. You will also need to have written permission from the children's parents before conducting the study. The Experimental Procedure section explains how to construct the survey. The Science Buddies resource, How Many Survey Participants Do I Need?, will show you how to figure out how many respondents you need to recruit in order to achieve your desired level of confidence that your results are representative of the total population.
Terms and Concepts
To do this project, you should do research that enables you to understand the following terms and concepts:
- Positive reinforcement
- Behavior modification
- Reinforcement theory
- Praise and behavioral outcomes
- Random sample
- How many participants do you need in your study in order to be 90% confident that the results from your study are representative of four-year-old pre-school students in general? How many participants for a 95% confidence level?
- What type of positive reinforcement will work best for influencing the behavior of four-year-old children?
- The following references are good sources of information on reward and reinforcement in educational settings:
- Cotton, K., 1998. Instructional Reinforcement. School Improvement Research Series. Retrieved September 5, 2006.
- Cotton, K. (2000). The Schooling Practices That Matter Most, a pamphlet which can also be ordered online from Amazon.com.
- This website has descriptions and calculators for several statistical tests, including the Student's t-test that you can use in this project:
Kirkman, T. (n.d.). "Student's t-Tests," Department of Physics, College of St. Benedict & St. John's University. Retrieved February 23, 2006.
Materials and Equipment
To do this experiment you will need the following materials and equipment:
- Two groups of four-year-old pre-schoolers:
- Witten parental consent must be obtained
- You will also need to obtain consent from the preschool teacher(s) involved
- The more subjects you can test, the more reliable your results will be (see the Science Buddies resource, How Many Participants Do I Need?, to see why.
- Children's blocks in four different shapes (square, circle, rectangle, triangle)
- Board with outline of square, circle, rectangle, and triangle
- Stop watch or timer
Note: There are special considerations when designing an experiment involving human subjects. For an experiment such as this, which involves children, parental consent must be obtained for all participants prior to beginning the experiment. You will also need to obtain advance permission from the teacher(s) whose classes are involved in this project. In addition, the experimental design must be approved by the fair's scientific review committee (SRC) prior to the commencement of experiments or surveys. Please refer to the ISEF rules for additional important requirements for studies involving human subjects.
- You will need to plan well ahead for this experiment in order to obtain SRC approval and parental consent forms for your survey participants. The SRC will need a detailed description of your proposed experimental procedure. They will also need to approve the parental consent forms before you begin.
- The task for the test is straightforward: the children are supposed to place the shapes inside their matching outlines on the board. Their goal is to do the task accurately, and as quickly as possible. You will time how long it takes each child to complete the task.
- The children should be randomly divided into two groups:
- each child receives instructions at the beginning of the task,
- one group is given instructions only and receives praise as a reward during the task,
- the other group is given instructions and told that they will receive a prize (stickers) as a reward for completing the task;
- note that the praise group is not informed about the prize group, nor is a prize discussed with them at any point (the children receiving the stickers should leave them outside the classroom until the entire test is completed).
- Both groups should be tested in the same location, outside the classroom door at a quiet table with minimal distraction. Also, this way no child will watch any other children during the testing.
- Record the test results and birthday of each participant.
- After the tests are complete, calculate the average time for each group. Is there a difference?
- More advanced students should calculate whether any difference between the average scores is statistically significant.
Ask an Expert
- Use birthdays to calculate the age of each participant, and make a scatterplot of test score (y-axis) vs. age (x-axis). This graph will show you if age is a major determinant of performance. Do you see a correlation between age and test score?
- Extend the study to children of different ages. Do different age groups respond in the same way or differently?
- Design an experiment to test the effectiveness of alternative forms of reward such as toys, candy, or special activities.
- Design an experiment to test the effectiveness of praise from different individuals, such as a stranger, a teacher, a parent, or a peer.
If you like this project, you might enjoy exploring these related careers: