SSRN Author: Dustin WhiteDustin White SSRN Content
https://www.ssrn.com/author=2506296
https://www.ssrn.com/rss/en-usFri, 15 Mar 2019 01:07:59 GMTeditor@ssrn.com (Editor)Fri, 15 Mar 2019 01:07:59 GMTwebmaster@ssrn.com (WebMaster)SSRN RSS Generator 1.0REVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disag- gregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. This work provides guidance to practitioners and researchers facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students’ ability to guess correctly is misspecified. This measure converges to Hake’s (1998) under certain conditions. We then use simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques. Using recursive partitioning trees fitted to our simulation results, we provide the practitioner concrete guidance based ...
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1770743.htmlWed, 13 Mar 2019 23:55:33 GMTREVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disag- gregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. This work provides guidance to practitioners and researchers facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students’ ability to guess correctly is misspecified. This measure converges to Hake’s (1998) under certain conditions. We then use simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques. Using recursive partitioning trees fitted to our simulation results, we provide the practitioner concrete guidance based ...
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1769765.htmlSun, 10 Mar 2019 16:21:19 GMTREVISION: Improving Student Performance through Loss AversionAs shown by Tversky and Kahneman (1991), framing an outcome as a loss causes individuals to expend extra effort to avoid that outcome. Since classroom performance is a function of student effort in search of a higher grade, we seek to use loss aversion to encourage student effort. This field experiment endows students with all of the points in the course upfront, then deducts points for every error throughout the semester. Students perform three to four percentage points better when controlling for student ability and domain knowledge. This result is significant at the 1% level in our most robust specification.
https://www.ssrn.com/abstract=3048028
https://www.ssrn.com/1736742.htmlTue, 06 Nov 2018 20:41:48 GMTREVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disaggregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. This work provides guidance to practitioners and researchers facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students' ability to guess correctly is misspecified. This measure converges to Hake's (1998) under certain conditions. We then use simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques.
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1732453.htmlSun, 21 Oct 2018 17:13:03 GMTREVISION: Improving Student Performance through Loss AversionAs shown by Tversky and Kahneman (1991), framing an outcome as a loss causes individuals to expend extra effort to avoid that outcome. Since classroom performance is a function of student effort in search of a higher grade, we seek to use loss aversion to encourage student effort. This field experiment endows students with all of the points in the course upfront, then deducts points for every error throughout the semester. Students perform three to four percentage points better when controlling for student ability and domain knowledge. This result is significant at the 1% level in our most robust specification.
https://www.ssrn.com/abstract=3048028
https://www.ssrn.com/1710394.htmlSat, 28 Jul 2018 08:41:19 GMTREVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disaggregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. This work provides guidance to practitioners and researchers facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students' ability to guess correctly is misspecified. This measure converges to Hake's (1998) under certain conditions. We then use simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques.
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1702317.htmlSat, 23 Jun 2018 13:57:40 GMTREVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disaggregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. Unfortunately, empirically estimating this probability isn’t feasible without a large number of observations (students). This work provides guidance to practitioners facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students’ ability to guess the correct answer is misspecified. We then use Monte Carlo simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques.
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1693879.htmlSun, 20 May 2018 15:15:36 GMTREVISION: On Guessing: An Alternative Adjusted Positive Learning Estimator and Comparing Probability Misspecification with Monte Carlo SimulationsInstructors and researchers have used the ‘flow’ of knowledge (post-test score minus pre-test score) to measure learning in the classroom for the past fifty years. Walstad and Wagner (2016) and Smith and Wagner (2018) move this practice forward by disaggregating the flow of knowledge and accounting for student guessing. These estimates are sensitive to misspecification of the probability of guessing correct. Unfortunately, empirically estimating this probability isn’t feasible without a large number of observations (students). This work provides guidance to practitioners facing this problem. We introduce a transformed measure of true positive learning that under some knowable conditions performs better when students’ ability to guess the correct answer is misspecified. We then use Monte Carlo simulations to compare the accuracy of two estimation techniques under various violations of the assumptions of those techniques.
https://www.ssrn.com/abstract=3168663
https://www.ssrn.com/1688028.htmlTue, 01 May 2018 05:12:29 GMTREVISION: Improving Student Performance through Loss AversionAs shown by Tversky and Kahneman (1991), framing an outcome as a loss causes individuals to expend extra effort to avoid that outcome. Since classroom performance is a function of student effort in search of a higher grade, we seek to use loss aversion to encourage student effort. This field experiment endows students with all of the points in the course upfront, then deducts points for every error throughout the semester. Students perform three to four percentage points better when controlling for student ability and domain knowledge. This result is significant at the 1% level in our most robust specification.
https://www.ssrn.com/abstract=3048028
https://www.ssrn.com/1684590.htmlThu, 19 Apr 2018 08:24:18 GMT