This preview shows pages 1–2. Sign up to view the full content.
Psychology Methods and Designs Overview
General Information
:
I)
Types of Variables
:
a.
True IV
– the IV levels can be doled out to people randomly
b.
False IV
– the IV levels are predetermined for each individual and outside of the experimenter’s control
c.
Discrete Variable
– each level is a different category
i.
Level often described by words, not numbers (ex – male/female)
ii.
No intermediate values
iii.
Types of Discrete Variables
:
1.
Ordinal Variable
– categories have a logical order (ex – short, medium, tall)
2.
Nominal Variable
– categories do not have a logical order (ex – sad, mad, happy)
d.
Continuous Variable
– each level is a number
i.
There are intermediate values
ii.
Study often has many if not infinite levels (e.g. – height, IQ, reaction time)
iii.
Types of Continuous Variables
:
1.
Ratio Variable
– if zero really represents no value, the numbers are literal and 2 times a
value is really twice the value of the variable (ex – height, age)
2.
Interval Variable
– otherwise (ex – temperature, IQ)
e.
Note
:
i.
Good designs don’t turn continuous IV’s into discrete IV’s (loses information)
ii.
It’s always better to have the DV’s (what you measure) be continuous variables (better for
getting significant results)
iii.
If IV’s are discrete, it’s best to have a balanced number of participants in each level
iv.
In describing IV and DV, make sure you use operational definitions (very specific and
empirically measurable)
II)
Inferential Statistics
:
a.
Inferential statistics test two competing hypotheses
i.
H
0
(Null Hypothesis)
– nothing is going on (the IV has no causal effect)
ii.
H
1
(Alternative Hypothesis)
– something is going on (the IV has a causal effect)
1.
Only one can be true and they describe all the possibilities
iii.
Process
:
1.
Begin by accepting H
0
and reject H
0
only if the data contradicts it, which automatically
means that you accept H
1
iv.
p
– probability that H
0
is correct (if p
< or = .05, reject H
0
and accept H
1
)
b.
Types of Errors
:
i.
Type I Error
– wrongly rejecting H
0
(probability of type I error = .05)
1.
Make sure that the assumptions of any statistical tests are not violated and one is using
“robust” tests, ones that are pretty correct even if some of their assumptions are violated
ii.
Type II Error
– wrongly rejecting H
1
1.
Helps to avoid type II errors if there is a lot of participants and the DV is interval/ratio
III)
Reliability and Validity
:
a.
Reliability
– how repeatable and consistent the test is
i.
Test/Retest Reliability
(measures repeatability) – test
1
time
1
= test
1
time
2
1.
Only pertains to constructs that are traits, not states
a.
Traits (stable), states (easily changeable) (e.g. – traits:states::depression:sadness)
2.
If the time interval is too short they’ll remember how they answered the test the last time,
but if the time interval is too long then the constructs might have changed
ii.
Internal Consistency
(measures consistency with the test) – compares all the items to each other
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
This note was uploaded on 04/17/2008 for the course PSC 3000 taught by Professor Sherman during the Spring '04 term at UC Davis.
 Spring '04
 SHERMAN

Click to edit the document details