Experimental Psychology Notes

CHAPTER 1: EXPERIMENTAL PSYCHOLOGY AND THE SCIENTIFIC METHOD
CHAPTER 2: RESEARCH ETHICS
CHAPTER 3: NONEXPERIMENTAL DESIGNS
CHAPTER 4: SURVEYS AND INTERVIEWS
CHAPTER 5: CORRELATIONAL AND QUASI-EXPERIMENTAL DESIGNS
CHAPTER 6: FORMULATING THE HYPOTHESIS
CHAPTER 7: THE BASIS OF EXPERIMENTATION
CHAPTER 17: ANALYZING RESULTS
I. Which test to use?

a. Levels of Measurement
i. Ratio Scale – has equal intervals between all its values and an absolute zero
point
ii. Interval Scale – measures magnitude and has equal intervals between values. It
has no true zero point
iii. Ordinal Scale – reflects differences only in magnitude: measured in form of
ranks
iv. Nominal Scale – Classifies items into distinct categories
b. Selecting a statistical test
i. How many IVs are there?
ii. How many treatment conditions are there?
iii. Is the experiment between-subjects or within-subjects?
iv. Are the subjects matched?
v. What is the level measurement of the dependent variable?
One Independent Variable Two Independent Variables
Two Treatments More than 2 treatments Factorial Designs
Level of Two Two Multiple Multiple Indepen Matched Independen
measure Independe matched independ matched dent groups t groups
ment of nt Groups groups ent groups groups (w/in and
DV (within groups (w/in subjects) matched
subjects) subjects) groups
(between
subjects
and within
subjects)
Interval T test for T test for One way One way Two Two way Two way
or Ratio independe matched ANOVA ANOVA way ANOVA ANOVA
nt groups groups (repeater ANOVA (repeater (mixed)
(matched measures) measures)
groups)
Nominal Chi Square Chi- Chi-
test square square
test test
II. The Chi-Square Test

a. Nonparametric: doesn’t assume normality
b. Compares frequencies
c. Chi-square: x2
i. Determines whether the frequencies of responses in our sample represent
frequencies expected in the population
ii. 0 when null hypothesis is true (frequencies in sample not different from treated
group)
iii. When larger than critical value, reject null hypothesis
d. You cannot test each subject more than once (all responses must be sampled
independently)
e. Used for nominal data
f. Degrees of Freedom – indicate how many members of a set of data could vary or
change value without changing the value of statistic already known for those data
III. The T Test
a. Parametric test
b. Effects of Sample Size
i. The exact shape of the distribution of t changes depending on the size of the
samples
ii. Small sample = flatter t
iii. Problem: we cannot sample all of the population to find a normal curve
1. Solution: large samples
2. Rarely a problem: Robust
a. The assumptions of the test can be violated without changing
the rates of type 1 and type 2 error
iv. Larger samples make it easier to reject Ho because the critical value of t gets
smaller as sample size (and degrees of freedom) increases
v. It is easier to get significance with a one-tailed test, because the entire 5%
critical region falls in a single tail of the distribution
vi. Confidence interval – represent a range of values above and below our sample
mean that is likely to contain the population mean with the probability level
c. T test for Matched Groups
i. We are evaluating the effect of the independent variable within each subject
ii. Degrees of freedom: N-1 (N is the number of pairs)
iii. The fewer df = the more difficult to reject Ho
1. Takes a more extreme tobs to reach significance in a matched-groups or
within subjects experiment
2. You need a larger t when you have fewer df
iv. Using a within-subjects or matched-groups design lowers the amount of
variability in the data (if we measure the responses of different subjects, we are
likely to get much more variability than if we measure the same subjects or
matched subjects)
v. When using this design: there is a trade off
1. Lower df = lower amount of variability produced by factors aside from
the IV
2. Make the t test for matched groups a more powerful test than the t test
for independent groups
3. Decreases the chance of a type 2 error
IV. Analysis of Variance (ANOVA)
a. Used to evaluate differences between 3 or more treatment means
b. Divides all the variance in the data into component parts and then compares and
evaluates them for statistical significance
i. Within groups variability
1. How much subjects vary from others in the group
ii. Between groups variability
1. How much subjects vary across each different level of the independent
variable
c. Each part presents variability produced by different influences in the experiment
d. Evaluates the likelihood that the proportions we observe could occur by chance

Experimental Psychology Notes

Uploaded by

Copyright:

Available Formats

Experimental Psychology Notes

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Experimental Psychology Notes

Uploaded by

Copyright:

Available Formats

CHAPTER 1: EXPERIMENTAL PSYCHOLOGY AND THE SCIENTIFIC METHOD

CHAPTER 2: RESEARCH ETHICS

CHAPTER 3: NONEXPERIMENTAL DESIGNS

CHAPTER 4: SURVEYS AND INTERVIEWS

CHAPTER 5: CORRELATIONAL AND QUASI-EXPERIMENTAL DESIGNS

CHAPTER 6: FORMULATING THE HYPOTHESIS

CHAPTER 7: THE BASIS OF EXPERIMENTATION

CHAPTER 17: ANALYZING RESULTS

I. Which test to use?

II. The Chi-Square Test

You might also like