Textbook Notes (270,000)
CA (160,000)
UTSG (10,000)
PSY (3,000)
PSY100H1 (1,000)
Chapter 2

PSY100H1 Chapter 2: Chapter 2 (Reading and Evaluating Scientific Research)


Department
Psychology
Course Code
PSY100H1
Professor
Michael Inzlicht
Chapter
2

This preview shows pages 1-3. to view the full 29 pages of the document.
Chapter 2: Reading and Evaluating Scientific Research
2.1 Principles of Scientific Research
- In 1998, the governor of Georgia put a lot of money in the state budget for
purchasing classical music
- At the time, there was this thing getting around called the Mozart effect
saying that listening to classical music boosts intelligence
o All this hype was based on ONE study and subsequent studies were
not able to come up with the same results
o In the study, 12 adults who listened to classical music did better in a
test of spatial ability than the other adults
o The media inflated this one study and generalized it to mean that
classical music raises intelligence
- This example shows the need for greater scientific literacy
- The single most important aspect of scientific research is that it strives
for objectivity
o Objectivity assumes that certain facts about the world can be
observed and tested independently from the individual who describes
them (e.g. the scientist)
- Objectivity is no easy task because as soon as people observe an event, their
interpretation of it becomes subjective
o Subjective: their knowledge of the event is shaped by prior beliefs,
expectations, experiences, and even their mood
Five Characteristics of Quality Scientific Research
- Quality scientific research meets the following criteria:
o 1) It is based on measurements that are objective, valid, and
reliable
o 2) It can be generalized
o 3) It uses techniques that reduce bias
o 4) It is made public
o 5) It can be replicated
Scientific Measurement: Objectivity, Reliability, and Validity
- The foundation of scientific methodology is in the use of objective
measurements
o Objective measurements: the measure of an entity or behavior that,
within an allowed margin of error, is consistent across instruments
and observers (ex. Kg is used as the unit for weight everywhere in the
world, and scientists don’t get to choose another unit
o There is sometimes a margin of error that scientists need to consider
when measuring something
- A variable: the object, concept, or event being measured
find more resources at oneclass.com
find more resources at oneclass.com

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

o Variables have to be described and measured
Today, high-tech equipment, such as fMRI (functional magnetic
resonance imaging) allows researches to view the brain and
see which areas are activated when stimulated
You can examine certain variable in many different ways
- Whenever a researcher is doing a study, they need to have carefully defined
terms
o Operational definitions: are statements that describe the procedures
(or operations) and specific measures that are used to record
observations
I.e. depression could be operationally defined as a score of x or
higher on the Beck Depression Inventory
o For example, if the Georgia governor looked into the Mozart research,
he would have been able to find some stuff such as:
Researchers have used several different objective measures of
thinking and reasoning in studies of listening to music,
including objective behavioral measures
Based on these measures, the only improvement of classical
music was an increase in spatial reasoning
Spatial reasoning: the ability to look at objects and
mentally manipulate them
Researchers doing this study have seen only a 1.5 increase in
spatial reasoning ability on an )Q style intelligence test and
this increase is perfectly normal. Also, it wasn’t permanent, and
disappeared after 10 mins.
Initially researchers attributed the improvements to classical
music because the same result was not founded in people who
had not listened to the music. But after, in other times,
researchers found the same response after subjects listened to
a horror movie.
o All the above show that we need to research before we make
decisions (the studies have to be RELIABLE)
- Validity: the degree to which an instrument or procedure actually measures
what it claims to measure
o This is quite complex
o Back to depressed example
Someone cannot just arbitrarily say a person is depressed or
not
Instead, for the measure to be valid, a particular score would
have to differentiate depressed and non-depressed people in a
way that accurately maps on how these people actually feel
- In addition to being valid, a measure needs to demonstrate reliability
o Reliability: when it provides consistent and stable answers across
multiple observations and points in time
o There are different kinds of reliability:
find more resources at oneclass.com
find more resources at oneclass.com

Only pages 1-3 are available for preview. Some parts have been intentionally blurred.

Test-test reliability: examines whether scores on a given
measure of behavior are consistent across test sessions
Alternate-forms reliability: this form examines whether
different forms of the same test produce the same results
Why would you need multiple forms of a test?
o Because, let’s say you want to test the memory of
a person with brain damage, so you give them a
test soon after they have acquired that brain
damage. Now, if a couple of months pass, and
you give them the tests again, they might score
better but it might be due to practice and not due
to actual improvement. Thus, having different
forms of a single test would allow us to test for a
certain thing such as memory and eliminate any
improvements that are due to practice and not
actual improvement.
Inter-rater reliability: The raters agree on the measurements
that were taken. If you design an experiment with clear
operational definitions and criteria for the raters, then it is
likely that you will have high inter rater reliability.
For example, psychologists might be interested in the
effects of non-verbal behavior when people interact, so
they might videotape participants solving a problem
and then have trained raters count the number of
touches or the amount of eye contact that occurred
during the experiment (basically, choose a criteria for
the rates to count (such as eye contact) to ensure inter-
rater reliability)
- The experiments that you do, must also be able to be generalizable
- Generalizability of Results:
o Involves examining patterns and trends that will allow us to predict
how most people will respond to different stimuli or situations
o Generalizability: the degree to which one set of results can be
applied to other situations, individuals, or events
o One way to increase the possibility that research results will
generalize is to study a large group of participants
But how large a group is best for studying? Ideally, it would be
best to study the entire population, but this is impossible so
scientists study a sample
Population: the group that the researchers want to generalize
about
Sample: a select group of population members
o How a sample is selected can determine if your results can be
generalized or not
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version