Scribd
Upload
10 views14 pages

U 1.3 and 1.4 Experimental Method, Defining Experiments

Uploaded by

정우박
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views14 pages

U 1.3 and 1.4 Experimental Method, Defining Experiments

Uploaded by

정우박
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

AP

Research
Methods

PSYCH
Continued
U 1.3
Sampling
Before investigating or testing a hypothesis, researcher must
decide which individuals will be the participants/subjects.

Let's say we want to test the hypothesis:


Coffee helps teachers to be more productive.

What would we need to operationally define?

Then, we need to figure out which teachers will be our test


subjects.

Can we test all teachers? What method of research should we


use to measure? How do we make sure our results are valid?
Sampling
Knowing what our population is (teachers), we should then select
a sampling of the population. The goal is to have a sample that is
representative of a larger population. What might be a
representative sample for us?

Sampling Bias: bad sampling process resulting in


unrepresentative sample. Example: sampling only the 4th floor.

What kind of mindset do you need to have when you're presented with
any data or research findings?
CRITICAL THINKING!

How can you avoid sampling bias???


Sampling
Random selection/random sampling gives everyone an equal
chance to be selected, allows for more generalization of the
research finding.

Is this random sampling?


Selecting random teachers from the school cafeteria on a
Tuesday afternoon
Selecting 20 names from a basket out of 200 total

Stratified sampling: a process that allows researcher to ensure


sample represents the population by some criteria.
For example, dividing group into age 18-25, 26-35, 36-45 then
random sampling those groups.
Sampling
After selecting a representative sample, what do you need to
do to the participants so you can test your hypothesis?

This is where the experiment comes in!

In the case of our coffee experiment, we need to see what


teachers' behaviours are like with and without coffee.
eriments
Exp
3 kinds: laboratory, field, and quasi-experiments (sample not
randomly selected due to ethical reasons; e.g., does cocaine affect
student test performance? do orphans face greater risk of
depression?)

Usually involves an experimental group and a control group.


The groups are also randomly assigned to limit participant-related
confounding variables.

Experimental Control
The independent variable, The group NOT exposed to
group exposed to treatment, used to
treatment. compare results.
eriments
Exp
Cause Effect
The independent variable, The dependent variable,
studied & manipulated. outcome that's measured.

Researchers will try to


randomize the groups as Other Variables
much as possible to reduce The confounding or
the contamination of other extraneous variable,
variables which could affects the certainty of
render a research invalid. research results.
eriments
Exp

Let's practice identifying the variables


Handout 1: Variable Worksheet

Handout 2: Breaking Down an Experiment


eriments
Exp
To ensure the equivalence of grouping, researcher could use:
Group matching by a criteria (e.g., sex, age) then randomly
assign.

To reduce experimenter bias (which is a form of confounding


variable, the experimenter may subconsciously favor action to
confirm their hypothesis) and the placebo effect, one can resort
to a double blind procedure.
Double Blind
Both researcher and
participants don't know
which group is assigned.
eriments
Exp
To reduce participant bias (participant wanting to give "good"
answers, or giving into cues that reveal the purpose of the study -
AKA demand characteristics) and the placebo effect, one can
resort to a single blind procedure.

Interesting note:
Hawthorne effect Single Blind
People perform better The participants don't
solely due to being know which group they're
observed/chosen
assigned.
rrelations
Co
In research, we may ask how two variables are related.
Variables can be anything that’s measurable.

Correlational coefficient (r) is a statistical measure that can


indicate how strong or weak a relationship is. This number will
always be from -1.00 to +1.00
rrelations
Co
What is a positive correlation? (“positive” often means an addition in
psychology, not “good”.)
The presence of one thing predicts the presence of another.

Example: money that you pay for gas…and…amount of gas in the tank
of your car

What is a negative correlation?


The presence of one thing predicts the absence of another.

Example: number of KM your car drove…and…amount of gas left in


the tank
Here you will see data
represented in a scatterplot,
which makes visualizing
trends much easier than data
charts, as you can see from
the example in page 50-51 of
the textbook.

The further away dots are


from the line, the weaker the
correlation.

How do you think this is


represented as a
correlational coefficient?
Let's do some examples.
r = _________
TASK
Review causation and correlation:
read textbook page 52
Check your understanding:
MCQ & FRQ on page 58

You might also like