The Hypothetico-Deductive Model
• It’s the main framework for inquiry within the scientific method
- Form a theory
- Deduce hypothesis from theory
- Test predictions
- Evaluate theory in light of tests
- The concept of “proof”
--> “If I open my TV and find a colony of hamsters, have I proven my theory that TV’s are
powered by hamsters?”
--> When testing a hypothesis made by a theory, we are actually trying to to disprove the theory,
not confirm it. The scientific method cannot prove that things are true, because it is impossible to
rule out all alternative theories that may better account for details of our observations. The
scientific method is a tool for eliminating theories that sound plausible but are, in fact, wrong.
- Affirming the consequent
--> This is a logical error that is often made when trying to do empirical work
- The Scientific Method
- Scientific Skepticism (How scientists are trained to think about the world, these ways are very
effective in figuring things out) --> It is the practice of not taking the world at face-value but, instead, considering it in close
detail before drawing conclusions.
Scientists are trained to follow a few rules, called critical thinking, when being skeptical:
• Handling extravagant claims
- The more a claim contradicts what we already believe, the more persuasive the evidence
for the claim must be before we accept it.
• Focus on falsifiable ideas or theories
- We only deal with theories that, if they were incorrect, we would have a way to
demonstrate this. We are using a process of elimination. An example would be
creationism, and it is believed that this theory cannot be falsified, because we have no
way of observing it. We cannot prove or disprove this divine being, we can only measure
physical things. It does not mean it does or does not exist, but we don’t have the tools to
approach this question because we have nothing tangible.
• Apply Occam’s Razor
- If two theories explain a phenomenon equally well, we should generally prefer to work
with the one that makes the fewest assumptions. ex: we have two theories: the first says
that clocks work through a series of interlocking gears, and the second says the same
except that they’re pushed by invisible pink unicorns. One makes fewer assumptions; we
would prefer to work with the simpler of the two, which does not mean that the pink
unicorn theory is wrong. Often what happens in scientific research is not that we
immediately falsify a theory but, rather, we devise an alternative theory that is more
succinct. Another wording: do not multiply entities beyond necessity. Positing the
existence of invisible pink unicorns does not add anything extra to our understanding of
how clocks work.
• Replicate Observations
- Findings must be capable of being duplicated by independent researches following the
• Rule out rival theories
- How could we rule out the dog suicide theory? (Scotland bridge) _____________________________________________________________________________
September 19, 2013
A NEW WAY OF UNDERSTANDING
- For most of recorded history, people simply looked to their leaders for understanding.
(ex: egyptian Pharaoh's were considered God-men, and Medieval Kings were The Hands of God)
- Accurate measurement devices are, historically, quite new (ex: Gregorian Calendar (1582) and
Modern Balance Scales (1700s)
- It was difficult to measure the passage of time and seasons in a reliable way
- Once we could weigh things accurately this helped out in commerce and the economy, because
it was hard to compare weights.
- If we want to understand the world by measurement our measurement tools need to be accurate:
Operational Definition. These four:
• It wouldn’t be useful to define aggression as “the distance you can shoot milk out of your nose
while laughing”, we say this definition lacks validity, which is: the characteristic of an
observation that allows one to draw accurate inferences about it.
• Operational definitions are not “valid” or “invalid”. There is a spectrum from “completely
irrelevant” to “exactly what I want to measure”. It’s not one or the other, there are statements or
operations that are more or less valid than others. Ex: SAT scores only account for 16% of the
variability in college students’ performance, but they are still used as an operational definition
of “preparedness for university”. We should be able to take your score on your SAT, and compare it to the marks you receive in University, and they should be similar (share some sort
of relationship). Truthfully, there’s actually only a little relationship.
• Construct Validity: Operational definitions are tools, and sometimes there are multiple tools for
the same job. This term means - the tendency for multiple operational definitions to produce
converging findings. Ex: What are three different ways to operationally define aggression?
• Definition: the tendency for a measurement device to produce the same result when used to
measure the same thing on different occasions
• Suppose we want to operationally define “visual acuity”
• Reliable Definition: The minimum threshold of size before you can no longer read letters is 10
feet away. VS. Unreliable definition: The number of ducks you can shoot down on a foggy day
• Some psychologists study visual perception, and help opticians develop tools for testing visual
activity. You can probably read down to the same place on any day of a given week that you
• The extent to which a measurement consistently differs in a certain direction away from what it
is supposed to be measuring.
• Suppose you wanted to understand the effects of playing video games on aggressive behavior.
To do this, you:
Find people who actually play video games and are familiar with the video games and are
familiar with the v