Results: The results of the memory experiment were analyzed using a Randomized Block analysis of variance. For each subject, we computed the mean confidence.

Slides:



Advertisements
Similar presentations
Ed-D 420 Inclusion of Exceptional Learners. CAT time Learner-Centered - Learner-centered techniques focus on strategies and approaches to improve learning.
Advertisements

Cognitive Modelling – An exemplar-based context model Benjamin Moloney Student No:
Reliability for Teachers Kansas State Department of Education ASSESSMENT LITERACY PROJECT1 Reliability = Consistency.
Copyright © Allyn & Bacon (2007) Data and the Nature of Measurement Graziano and Raulin Research Methods: Chapter 4 This multimedia product and its contents.
Sentence Memory: A Constructive Versus Interpretive Approach Bransford, J.D., Barclay, J.R., & Franks, J.J.
Hemispheric Asymmetries In False Recognition May Depend on Associative Strength Cathy S. Robinson & Christine Chiarello University of California, Riverside.
Readings 25 & 26. Reading 25: Classic Memory and the eye-witness Experiment 1 Experiment 2 Conclusion Reading 26: Contemporary Misinformation Effect Memory.
A quick introduction to the analysis of questionnaire data John Richardson.
The Contribution of Perceptual Mechanisms to the Spacing Effect Jason Arndt & Julie Dumas Middlebury College Abstract Recent explanations of the spacing.
Consistency of 9-11 Memories Kristen Sager, Advisor: Dr. Arnold L. Glass What is a Flashbulb Memory? Brown and Kulik argued personal circumstances during.
The Analysis of Variance
CONFIDENCE – ACCURACY RELATIONS IN STUDENT PERFORMANCES We attempted to determine students’ ability to assess comprehension of course material. Students.
 The misinformation effect refers to incorrect recall or source attribution of an item presented after a to-be-remembered event as having been presented.
Reliability of Selection Measures. Reliability Defined The degree of dependability, consistency, or stability of scores on measures used in selection.
Richard M. Jacobs, OSA, Ph.D.
Inferential Statistics
Descriptive and Causal Research Designs
Sampling : Error and bias. Sampling definitions  Sampling universe  Sampling frame  Sampling unit  Basic sampling unit or elementary unit  Sampling.
David Steer Department of Geosciences The University of Akron Learning objectives and assessments Oct 2013.
1 Statistics for the Behavioral Sciences (5 th ed.) Gravetter & Wallnau Chapter 10 The t Test for Two Independent Samples University of Guelph Psychology.
Results Following Signal Detection Theory, Accuracy is calculated as the difference between Real and Foil claim rates, and Bias is the mean of the two.
Frequency Judgments in an Auditing-Related Task By: Jane Butt Presenter: Sara Aliabadi November 20,
1 Focusing on the FCAT Test-Taking Strategies Grades 3-5 Nancy E. Brito, Department of Assessment , PX47521 Information.
TEMPLATE DESIGN © Difference in reaction times between true memories and false memories in a recognition task Marta Forai.
References Arndt, J. & Hirshman, E. (1998). True and false recognition in MINERVA2: Explanation from a global matching perspective. Journal of Memory and.
Individual Preferences for Uncertainty: An Ironically Pleasurable Stimulus Bankert, M., VanNess, K., Hord, E., Pena, S., Keith, V., Urecki, C., & Buchholz,
Music & Studying.
David Steer Department of Geosciences The University of Akron Learning objectives and assessments May 2013.
Chapter 10: Analyzing Experimental Data Inferential statistics are used to determine whether the independent variable had an effect on the dependent variance.
 Are false memories more likely to develop when people are motivated to believe in the false event?  Sharman and Calacouris (2010)
References McDermott, K.B. (1996). The persistence of false memories in list recall. Journal of Memory and Language, 35, Miller, M.B., & Wolford,
1 Psych 5500/6500 The t Test for a Single Group Mean (Part 1): Two-tail Tests & Confidence Intervals Fall, 2008.
Access Into Memory: Does Associative Memory Come First? Erin Buchanan, Ph.D., University of Mississippi Abstract Two experiments measuring the reaction.
Self-assessment Accuracy: the influence of gender and year in medical school self assessment Elhadi H. Aburawi, Sami Shaban, Margaret El Zubeir, Khalifa.
PORTFOLIO ASSESSMENT OVERVIEW Introduction  Alternative and performance-based assessment  Characteristics of performance-based assessment  Portfolio.
From Bad to Worse: Variations in Judgments of Associative Memory Erin Buchanan, Ph.D., Missouri State University Abstract Four groups were tested in variations.
Learning Objectives Copyright © 2002 South-Western/Thomson Learning Using Measurement Scales to Build Marketing Effectiveness CHAPTER ten.
Assessment and Testing
JAM-boree: A Meta-Analysis of Judgments of Associative Memory Kathrene D. Valentine, Erin M. Buchanan, Missouri State University Abstract Judgments of.
Introduction Relationship between Extraversion and Delay Discounting of Social Interactions Sara L. Daugherty and Daniel D. Holt University of Wisconsin-Eau.
Chapter 6: Analyzing and Interpreting Quantitative Data
Research Topics in Memory
Reliability performance on language tests is also affected by factors other than communicative language ability. (1) test method facets They are systematic.
Chapter 6 - Standardized Measurement and Assessment
Conclusions  Results replicate prior reports of effects of font matching on accurate recognition of study items (Reder, et al., 2002)  Higher hits when.
1 Focusing on the FCAT Test-Taking Strategies Grades 6-8 Nancy E. Brito, Department of Assessment , PX47521
1 Focusing on the FCAT Test-Taking Strategies Grades 9-11 Nancy E. Brito, Department of Assessment , PX47521
1 Topic 14 – Experimental Design Crossover Nested Factors Repeated Measures.
Project VIABLE - Direct Behavior Rating: Evaluating Behaviors with Positive and Negative Definitions Rose Jaffery 1, Albee T. Ongusco 3, Amy M. Briesch.
AP Exam Development and Grading The AP Physics exams are developed by a committee of high school and college physics faculty After the exams are administered,
Effects of Word Concreteness and Spacing on EFL Vocabulary Acquisition 吴翼飞 (南京工业大学,外国语言文学学院,江苏 南京211816) Introduction Vocabulary acquisition is of great.
Assessing Musical Behavior
Data and the Nature of Measurement
Difference in Mls poured between the subject and the researcher
Emilie Zamarripa & Joseph Latimer| Faculty Mentor: Jarrod Hines
Experiment 2 – Discussion Experiment 1 – Discussion
Mental Rotation of Naturalistic Human Faces
Effects of Targeted Troubleshooting Activities on
RELIABILITY OF QUANTITATIVE & QUALITATIVE RESEARCH TOOLS
Alison Burros, Nathan Herdener, & Mei-Ching Lien
Which of these is “a boy”?
Course name: Weekly Planning
Meredith A. Henry, M.S. Department of Psychology
Post event discussion (PED) and EWT
Tasks & Grades for MET1.
Tasks & Grades for MET3.
Validity and Reliability II: The Basics
Psych 231: Research Methods in Psychology
COMPARING VARIABLES OF ORDINAL OR DICHOTOMOUS SCALES: SPEARMAN RANK- ORDER, POINT-BISERIAL, AND BISERIAL CORRELATIONS.
Chapter 8 VALIDITY AND RELIABILITY
Presentation transcript:

Results: The results of the memory experiment were analyzed using a Randomized Block analysis of variance. For each subject, we computed the mean confidence rating on Day 1 for the targets correctly identified in the first test, as well as the mean confidence rating for each type of distracter. The means are presented graphically in Figure 1. We found a significant effect of word type as reported below. We repeated this procedure for each confidence / test pair, and found significant effects of word pair in each test for: Confidence 1 / Test 1, (F = 22.90, df = 3, 111 p <.001) Confidence 1 / Test 2, (F = 2.78, df = 3, 111, p <.05) Confidence 2 / Test 1, (F = 20.45, df = 3, 111, p <.001) Confidence 2 / Test 2, (F = 4.73, df = 3, 111, p <.01). We then performed a posteriori tests to determine what confidence ratings were different from one another, using Tukey’s q to control the alpha level for each collection of tests. The results are shown in Table 1. The headings CiTj refer to confidence ratings on the ith study list for the jth test. More findings show that students were more accurate on their first recognition test (Day 2) than on the second (Day 3) (t = 6.227, df=37, p <.001), Figure 2. The partial correlation between accuracy and retrospective predictions was significant (r = 0.443, df = 35, p <.05) on the first test when confidence was used as a control factor, but not on the second. Accuracy was better for the first test than for the second (t = 7.106, df = 37, p <.001), Figure 3. The mean number of Target, High Associate, Medium Associate, and Low Associate responses for each test day is shown in Figure 4. A high response rate to targets suggests that participants were able to distinguish between the targets and the associates. Confidence and Accuracy Relations in Student Performance Allen A. Newton Advisor: Dr. John M. Ackroff We assessed the accuracy of students’ predictions of their ability to recognize words from a study list on a subsequent recognition test. Their confidence was measured on a forced rating scale. Participants accurately predicted their performance and their confidence reflected such results. Participants made accurate estimates of their overall scores after the tests was completed, although the accuracy of their confidence ratings varied across measurement (subjective estimate) and assessment (performance on the recognition task). Method: Research Participants: Students from two sections of an undergraduate course in Cognition at Rutgers University participated in these experiments for extra credit. Procedure: A set of 49 words from the University of South Florida norms (Nelson et al., 2004), was chosen as a study list for students. The first day, students were asked to remember a 4 to 6 letter word presented for 5 sec and then asked how memorable each word was from a scale of 1 through 5. On day two (two days later), students again saw the word list and re-rated each word. After the word list was studied, students were given a distracter task rating the similarity of various random shapes. A forced choice test consisting of each target along with a high, medium, or low associate was then given. (For example, for the target (T) DRAFT, the high, medium, and low associates were (HA) BEER, (MA) ARMY, and (LA) WIND, respectively.) Students returned 5 days later for a second recognition test. At the end of each recognition test, the students were asked “How many words do you think you recognized correctly?”, with choices of 90%, 80%, 70%, 60% and less than 60%. They were also asked to indicate how accurate they thought that prediction was. People are frequently asked to make decisions about how confident they are in judgments they have made. Studies in eyewitness identification have shown that confidence is not always related to accuracy. Most people have had the experience of being sure they were right about something, only to be proven wrong. Students are called upon to make decisions about the degree to which they know course material on a regular basis. Decisions about how and what to study, and for how long, are based on a perception of how well one understands the material for upcoming exams. Students who feel they are well-prepared sometimes receive feedback to the contrary – they perform poorly on exams. There has been a reasonable amount of research on students’ ability to predict their academic performance. Prohaska (1994) suggests that students’ confidence is a good prediction of their accuracy. A high confidence rating showed a high accurate response; similarly a student who gave low confidence rating gave an inaccurate response. The interesting thing about confidence and accuracy is that most of the present research focuses on the organized structure of real life situations such as academic material. In this study, we presented an unorganized list of words for participants to learn, and asked them to asses their confidence in their ability to remember the words in a future memory test. Mandler suggests that studying a list essentially means organizing it, because recall fundamentally depends on organization. Our stimuli did not fall into semantic categories, in which an overall theme is apparent in a group of words, nor were any of the words related to each other in any other obvious way. Any associative errors made in the test would imply something about the memory process, since they could not be attributed to interference from other words in the list. These associative errors in the context of the unorganized structure should allow us to investigate the basic elements of accuracy and confidence. References: Ackroff, J. M., and Rouse, R. O. Jr. TSD and coding in STM. (1970) Psychonomic Science, 21, Busey, T. A., Tunnicliff, J., Loftus, G. R. & Loftus, E. F. (2000) Accounts of the confidence-accuracy relation in recognition memory. Psychonomic Bulletin and Review, 7, Mandler, G. Organization and memory In K.W. Spence & J.T. Spence (Eds.), The psychology of learning and motivation (Vol.1). New York: Academic Press. Morton, J. (1969). Interaction of information in word recognition. Psychological Review, 76, Nelson, D. L., McEvoy, C. L. & Schreiber, T. A. (2004) The University of South Florida word association, rhyme, and word fragment norms. Behavior and Research Methods, Instruments, & Computers, 36, Prohaska, V. (1994) "I know I'll get an A": Confident overestimation of final course grades. Teaching of Psychology. 21(3) Oct 1994, Rawson, K. & Dunlosky, J. (2000) The rereading effect: Metacomprehension accuracy improves across reading trials. Memory and Cognition, 28(6), Figure 2: Accuracy and Prediction on the lab recognition tests Figure 3: Accuracy and confidence on the lab recognition tests Discussion: When studying for exams students can sometimes find it hard to assess how well they know the material being covered. However, research has shown that students are able to recognize the difference between their accuracy and confidence in test taking. Once students can make accurate assessments of their confidence and accuracy they are able to determine which course materials to study. We have found that students are able to distinguish material they do not know and the material they do know based on their confidence rankings. Using the “retrospective confidence ratings” of Busey et al. (2000), (confidence rankings collected at the end of the recognition test), we found accurate assessments for each test. When confidence in the assessment was controlled for, however, the predictions were accurate for the first test only. Also, a rereading effect occurred from the first day to second day. Given that students were able to “re-read” the study list and re-rate their confidence, they predicted and performed better on the first lab exam than on the second lab exam. A 10% decrease in their prediction rate also suggests that the re-reading improves accuracy, suggested by Rawson and Dunlosky (2000). We also found that organizing material into themes is not necessary. What is necessary for accurate performance is confidence of the material to be recognized. Mandler’s results imply that the ability to organize study material is one way for students to increase their confidence in their ability to recall the material in subsequent test situations. But we know that rote memorization works – it’s not as effective as learning material that has some internal cohesion, but it still works. In our task, students had to rely on rote memorization, since there was no was they could organize the materials. We expected confidence to be lower for HAs that for Targets, lower for MAs than for HAs, and still lower for LAs that MAs. The general pattern of results is consistent with this expectation, although as Table 1 shows, most of these differences failed to reach significance. The LAs had higher False Alarm rates in some conditions than MAs; this finding may be an artifact of the low number of False Alarms in general. Alternatively, one could argue that on the recognition test, the HA and MA choices were similar enough to the target to raise the activation level of the target's logogen (Morton, 1969) so that they were rarely confused, but that the LAs were dissimilar enough that they served as true foils in the recognition test, and were selected more frequently than MAs in Test 2, where overall performance was worse. Understanding the factors underlying the these results is a topic for further study. PAIRC1T1C1T2C2T1C2T2 T / HA T / MA4.19* T/ LA10.36** **5.18** HA / MA3.71* HA / LA9.88** **3.66* MA/ LA6.16** **2.83 Table 1. Tukey’s q for difference in mean confidence for response type *p <.05 ** p <.01 Figure 4: Mean number of responses for each word type for each day of testing Students are able to change their performance response depending on their confidence which can be seen in Figure 3. Students can clearly see if they lack the confidence in their performance they can readily predict their performance which could be helpful during study periods. As Figure 4 shows, students were able to learn the list items through rote memorization quite effectively. Given the opportunity to make false alarm responses that were a priori likely to occur given that they were semantically related (Ackroff & Rouse, 1970), their error rate was extremely low. Exposure to the words for a relative long duration (5 sec) may be responsible for this. Our major finding is the relationship between students’ confidence in their ability to recall study materials and their performance. When they’re sure, they do well; when they’re uncertain, their performance is mediocre. Figure 1: Mean Confidence of Targets and False Alarms