Testing Testing’s Threshold
Research has indicated that testing enhances memory (Roediger & Karpicke, 2006; Carrier & Pashler, 1992). Research has also indicated that some testing enhancing memory whereas too much testing can deteriorate memory (Bangert-Drowns, Kulik, & Kulic, 1991). This research seeks to investigate where the threshold in the learning curve is. If some testing enhances memory and too much testing deteriorates memory then there must be a threshold at which testing enhances memory optimally. Roediger and Karpicke (2006) found that repeated testing, as apposed to studying, enhanced long-term memory for prose. Subsequently this research will focus on long-term memory, as it is after all, long term memory that one generally seeks to use tests to identify. Carrier and Pashler (1992) found that testing acted as a memory enhancer absent of the idea that testing was merely another form of study.
This research will be conducted in a between groups single test design. Independent measures will include a 0-test, 3test and 6-test condition. Dependent variables will include the use of a history lesson and 7 tests through which memory can be evaluated, six tests for the experimental conditions and one for the final analysis of results. Controlled variables will include the use of the same classroom and same instructor for all three test conditions. Each testing session should likewise be held during the same time of day. In this way it will take three days of testing to test all three groups. Children will not be tested for more than 45 minute, the time of one class period. This will lessen the extraneous variables of fatigue and boredom. Also the time of day is important to lessen the impact of individual differences concerning their last meal, sleep schedules, and fatigue from the school day in general. The same classroom and instructor is used to maintain a sense of continuity, should the instructor’s style influence learning or more windows in one classroom as apposed to another. By using the same classroom and instructor these potential extraneous variables are lessoned.
Participants will be chosen at random from a group of eligible high school students. These students will then be randomly assigned to one of the three groups. A total of 20 students will be assigned to each group. All groups will attend their regular classes until called for this experiment. A classroom, cleared of distractions has been procured for this study. A veteran will be used as instructor for the class.
On the first day the 0-test group will be called to the classroom. This group will be given the chapter from the history lesson. The class will then be instructed that they have 30 minutes to study the text and will be given a test at the end of the thirty minutes. On the second day the 3-test group will be called to the classroom. They will also be given the text but with instruction that they will be tested every ten minutes with a final test at the end. On the third day the 6-test group will be called. This group of students will be given the text and instruction that they will be tested every five minutes with a final test at the end.
Tests will be scored by percentage, one to one hundred. Only the final test will be counted as the dependent variable. These tests will be averaged across the classroom groups. In this way one number is discovered that will represent the overall standing of each individual group. The pre-tests will not be counted.
Maintaining an SPSS file that will be updated at the end of every day by the veteran will preserve data quality. Data will be entered directly after scoring the tests. All test scores will be maintained in a private laptop owned by the University. This along with all hard copies of the tests will be maintained in a locked office when not in use.
Data will be analyzed by one-way ANOVA for independent groups in SPSS. It is hypothesized that the 3-test group will have significantly higher (p < .05) test scores than either the 0-test or the 6-test groups.
This research seeks to answer the question; Is there a threshold to testing’s memory enhancing power. The 0-test group having no testing and being left primarily to their own devices, within reason, seems too lax an environment for students to learn in. The 6-test group, being tested every five minutes seems rather excessive. In this way, should the results come back conclusive evidence will be given to support more advanced and more thorough research in this area.
The above stated hypothesis was arrived at after reviewing similar research in the field concerning learning thresholds and testing (Roediger & Karpicke, 2006; Carrier & Pashler, 1992; Bangert-Drowns, Kulik, & Kulic, 1991). It is important to the field of memory as it relates to the acquisition of long-term memory. It is also important to the field of education to understand these thresholds and how best to use testing, not only as a method by which to asses students, but as a method by which to enhance the learning power of students.
Bangert-Drowns,R.L., Kulik, J.A. & Kulic, C.L.C. (1991). Effects of frequent classroom testing. Journal of Educational Research, 85, 89-99.
Carrier, M., & Pashler, H. (1992). The influence of retrieval on retention. Memory & Cognition, 20, 633-642.
Roediger, H.L., & Karpicke, J.D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Association for Psychological Science, 17(3), 249-255.