Hard work beats talent when talent won't work hard

Had a conversation this weekend that touched on this point that many of our unconscious or semiconscious modern practices and behaviors are the products, not of informed design (H/T Adam Ferguson "the result of human action, but not the execution of any human design") but of effective evolutionary selection.

We did not have the data to empirically prove a superior design, we fell into practices that worked.

Three basic things that teachers have been doing for a long time (but which annoy students) have been recently validated by science. First, you can eliminate cheating by sitting students randomly. (Prior to that, 10% of students at a top university science course cheated!) 1/3 https://t.co/gxKqkP4qEN

— Ethan Mollick (@emollick) January 19, 2020

Click for the thread.

Key points

You can eliminate cheating by sitting students randomly. Without random seating, baseline cheating was 10% at an elite university.

Giving lots of small quizzes and tests turns out to be one of the most effective ways to boost memory. It also increases future learning and helps you access related topics more effectively than spending the same time studying

Emphasizing attendance is smart: “class attendance a better predictor of college grades than any other known predictor of academic performance, including scores on standardized admissions tests such as the SAT, high school GPA, study habits, and study skills.”

The three studies are:

Catching Cheating Students by Ming‐Jen Lin and Steven D. Levitt. From the Abstract:

We develop a simple algorithm for detecting exam cheating between students who copy off one another's exams. When this algorithm is applied to exams in a general science course at a top university, we find strong evidence of cheating by at least 10% of the students. Students studying together cannot explain our findings. Matching incorrect answers proves to be a stronger indicator of cheating than matching correct answers. When seating locations are randomly assigned, and monitoring is increased, cheating virtually disappears.

Ten Benefits of Testing and Their Applications to Educational Practice by Henry Roediger, Adam L. Putnam, and Megan A. Sumeracki. From the Abstract.

Testing in school is usually done for purposes of assessment, to assign students grades (from tests in classrooms) or rank them in terms of abilities (in standardized tests). Yet tests can serve other purposes in educational settings that greatly improve performance; this chapter reviews 10 other benefits of testing. Retrieval practice occurring during tests can greatly enhance retention of the retrieved information (relative to no testing or even to restudying). Furthermore, besides its durability, such repeated retrieval produces knowledge that can be retrieved flexibly and transferred to other situations. On open-ended assessments (such as essay tests), retrieval practice required by tests can help students organize information and form a coherent knowledge base. Retrieval of some information on a test can also lead to easier retrieval of related information, at least on delayed tests. Besides these direct effects of testing, there are also indirect effects that are quite positive. If students are quizzed frequently, they tend to study more and with more regularity. Quizzes also permit students to discover gaps in their knowledge and focus study efforts on difficult material; furthermore, when students study after taking a test, they learn more from the study episode than if they had not taken the test. Quizzing also enables better metacognitive monitoring for both students and teachers because it provides feedback as to how well learning is progressing. Greater learning would occur in educational settings if students used self-testing as a study strategy and were quizzed more frequently in class.

Class Attendance in College: A Meta-Analytic Review of the Relationship of Class Attendance With Grades and Student Characteristics by Marcus Credé, Sylvia G. Roch, and Urszula M. Kieszczynka. From the Abstract:

A meta-analysis of the relationship between class attendance in college and college grades reveals that attendance has strong relationships with both class grades (k = 69, N = 21,195, ρ = .44) and GPA (k = 33, N = 9,243, ρ = .41). These relationships make class attendance a better predictor of college grades than any other known predictor of academic performance, including scores on standardized admissions tests such as the SAT, high school GPA, study habits, and study skills. Results also show that class attendance explains large amounts of unique variance in college grades because of its relative independence from SAT scores and high school GPA and weak relationship with student characteristics such as conscientiousness and motivation. Mandatory attendance policies appear to have a small positive impact on average grades (k = 3, N = 1,421, d = .21). Implications for theoretical frameworks of student academic performance and educational policy are discussed.

The last two studies reinforce the other cultural trope - diligence and conscientiousness are the great equalizers over raw IQ. Or, as my son's wrestling coach you used to tell his wrestlers "Hard work beats talent when talent won't work hard."