Tuesday May 26th

5/25/2020

The Exam #1 debrief is provided below. Feel free to review it, and then go back into the exam in D2L to view the results of your performance in more detail (I turned on the review feature so you should be able to see how you answered each item, and what the correct answers are).

And this is just a reminder that you should be working on your Act 1 project today...it is due on Wednesday (5/27) at midnight. If you have finished this project and would like to dive into Act 2, the material is now posted (except the Exam #2 practice and the Act 2 project...these will be available no later than Thursday morning).

One more thing...a student contacted me and was interested in discussing the Willingham video presented in Act 1 that addressed learning styles, and why they are myth. As an instructional designer, I LOVE this topic and am very happy to host a discussion about Willingham's perspective (which I share) during a Zoom meeting. So I will be sending out an invitation soon for a discussion on Thursday (5/28) at 7:00 p.m. This is completely voluntary, and will run as long as any interested students care to discuss the issue. So look for a Zoom invite soon about this.

Exam Debrief

The first exam is in the record books now, and considering this was the first formal exam many of you have taken in awhile, I would say you did very well! Here are the overall test-level statistics BEFORE any adjustments:

I can live with a raw mean score of nearly 85%, and the standard deviation is acceptable. One student even scored 100%, which is not an easy thing to do on these types of assessments!

Examining the statistics for each item revealed only three potentially problematic questions. These are presented below with a brief explanation of the results.

This was the only item that was missed by more than half the class. The reason why the reinforcement is "intermittent" in this example is because it is assumed that the each student was called on "in random order"...but not every time. To be honest, the phrase "not every time" should have been included in this scenario, so I think that anybody who selected "continuous reinforcement" should not be penalized. So I did adjust the scores of those who selected "Continuous Reinforcement", which brought the overall average exam score to 87%. Nice.

This item was tricky for me too, because I tried to write a negative reinforcement scenario. But there is no denying that from Jeremy's perspective (the behavior of interest in this scenario), being paired with Justin is a response stimulus that is added (not removed), leading to reinforcing Jeremy's behavior of throwing fits. Just because the behavior that is being reinforced is negative behavior (throwing fits) from the teacher's perspective doesn't mean the reinforcement is negative. So this ended up being a challenging scenario, but one that is clearly positive reinforcement.

This was the only other item that was missed by 50% of the class or more. And the reason for this being missed by many is really quite simple. The directions state "Choose all that apply", which means many figured that more than one choice would need to be selected for a correct answer. It WAS pointed out in the directions for the exam that "Choose all that apply" could very well mean only one choice is correct...but old habits die hard, and I am sure this is the only reason why half the class selected another response besides the obvious "Aspects of the theory are not logically consistent". Eight students also thought that "the theory can be falsified" indicated it was a bad theory. Of course, it is just the opposite. All scientific theories MUST be falsifiable, or they are not scientific. It truly is the beauty of science.

2 Comments

Tracy

5/26/2020 04:29:25 pm

So... I guess I don't quite understand the difference between the ideas that "it can be falsified, but hasn't" and "it can be falsified". Could someone provide an example of each??

Greg Sherman

5/27/2020 08:35:12 am

Good question Tracy! When I read your question I immediately thought about an examples of theories in the area of human moral development.

The philosopher Jean Jacques Rousseau developed a theory of moral reasoning that postulates a sequence of five age-related stages through which a person must pass in order to reach moral maturity:

(i) infancy (birth to age 2)
(ii) the age of sensation (3 to 12)
(iii) the age of ideas (13 to puberty)
(iv) the age of sentiment (puberty to age 20)
(v) the age of marriage and social responsibility (age 21 on)

Without going into too much detail, this is not a scientific theory because it is not possible to falsify (prove wrong) these stages. You can't have ideas without sensations, you can't have sentiment without ideas, etc. The logic of these stages make it impossible to disprove.

On the other hand, Kohlberg presents a theory or moral development in which morality develops in approximately six stages. He maintains that no one skips a stage or regresses to an earlier stage.

Level A. Premoral
Stage 1—Punishment and obedience orientation
Stage 2—Naive instrumental hedonism
Level B. Morality of conventional role conformity
Stage 3—Good-boy morality of maintaining good relations, approval by others
Stage 4—Authority-maintaining morality
Level C. Morality of accepted moral principles
Stage 5—Morality of contract, of individual rights and democratically accepted law
Stage 6—Morality of individual principles of conscience

This is a more scientific theory that Rousseau's because it is possible to test whether or not a person displays behaviors based on their stage of morality, and whether or not somebody can skp stages or regress. So Kohlber's theory could be falsified if enough scientific studies reveal that people can skip or regress. I have not done any investigations into whether or not people have empirically tested his theory, but I suspect some have. If nobody has ever tested the theory, then it is a good example of "it can be falsified". If people have tested the theory many times but have yet to find many instances in which it did not hold up, then it would be an example of "it can be be falsified, but hasn't". Other popular examples of "it can be be falsified, but hasn't" include the theory of evolution, plate tectonics, special relativity, and heliocentrism. I know..these are all scientific theories...but hopefully they illustrate the distinction you are looking for.

Tuesday May 26th

Leave a Reply.