Statistical significance and p-values

The concept of p-values, used in hypotheses testing, is very difficult to grasp. Often it is misused and misinterpreted, even with the best intents.
To address this issue, the American Statistical Association developed the following ASA Statement on Statistical Significance and P-Values.
Common misunderstanding of p-values are also nicely summarized (an clarified) on Wikipedia.

Teaching and learning strategies

"It is only possible to transfer information, not understanding. The students have to do the difficult part - develop their own understanding of the material - by themselves. Their teachers should help with that and create learning opportunities but nevertheless cannot make the students understand. The students, not the teachers, have to do the hard work - assimilate information and develop understanding."

A beautiful insight into how we learn and how we could learn more effectively is given in a series of short video lectures on "How to get the most out of studying" by prof. Stephen Chew (Samford University):

Concepts one might incorporate in the classes, along with many others:

  • Active Learning - active participation of students
  • Flipped Classroom - switching the usual scheme (getting information in-class, using and assimilating information out-of-class in homeworks, projects, self-study etc.) into a new one (getting information before class, using and assimilating information in-class in discussions, group work etc.)
  • Peer Instruction - a technique that uses polling and discussions in small groups to develop conceptual understanding
  • Just-in-Time-Teaching - frequent feedback both to the students and to the teacher by completing a set of online assignments before each class; also lets the teacher adapt the structure of the class to the needs of the students

Is all this stuff any good? YES! There is scientific evidence to support the use of the concepts mentioned above, such as:

  • Freeman et al. (2014): Active learning increases student performance in science, engineering, and mathematics. PNAS 2014 111 (23) 8410-8415 (full-text available here or here). A metaanalysis of 225 studies in the STEM fields (science, technology, engineering, mathematics) showing that
    • Students in classes with traditional lecturing were 1.5 times more likely to fail than were students in classes with active learning.
    • Average examination scores improved significantly (by about 6 %) in active learning sections.
    • The greatest effects were in small classes (less than 50 students).
  • Other studies showing that flipped classroom model and active learning result in significant increase in learning gains and increase student mastery of both conceptual reasoning and quantitative problem solving:
    • Crouch & Mazur (2001): Peer Instruction: Ten years of experience and results. Am. J. Phys. 69(9), 970-977 (full-text available here).
    • Moravec et al. (2010): Learn before Lecture: A strategy that improves learning outcomes in a large introductory biology class. CBE—Life Sciences Education, Vol. 9, 473–481 (full-text available here).
    • etc.