REVIEW SESSION PROBLEMS
1.
2 X 2 ANOVA
You conduct an experiment involving 2 levels of selfconfidence (Low and High) and examine 25 participants
anxiety scores on an interval scale after they speak to one of two groups of differing sizes (Small and Large
Instructions
Assignment 3 consists of three problems covering the topics of ANOV
Each problem is on a separate sheet that can be accessed by clicking
Provide your answer to each problem on the sheet associated with tha
For hypothesis testing questions, us
Good research has
key ingredients:
methodology
theory
measurement
ethics
Educating the public is
tricky
bias we believe
what is consistent with
our fears
mindsets all
vaccines are bad
Table of Contents
1
Correlation
Can
you see
why correlation
can n
Twoway
Between subjects
Analysis of Variance
(ANOVA)
Assumptions of the 2way ANOVA
Perform this ANOVA if it is a 2way design and:
1.
All cells contain independent samples of
participants.
2.
The dependent variable measures interval or ratio
scores that
Oneway
Between subjects
Analysis of Variance
(ANOVA)
Notations
Conditions k = 3
(Levels)
Dependent scores
Factor A: Independent variable level of
sunlight
A1=Low
A2=Medium
A3=High
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X1 = T1
X2 = T2
X3 = T3
M1
M2
M3
n1=5
n2=
In class Problem #1 Report
The F obtained exceeded the F critical
therefore we reject the null hypothesis and
conclude that there was an effect of age on
creativity (F (2, 12) = 14.66, p < .05). Post
hoc comparisons (Tukeys HSD) were used
to determine whe
PSYC 2002
Lecture 14
March 19th, 2012
Post Hoc Test
only when the data is significant, and there are more then 3 groups.
Tukeys HSD honestly significant difference
qk =critical value
HSD= (qk ) vMSwn
N
=(3.77) (v7/5)
=4.46 This value should always be a
PSYC 2002
March 12th, 2012
Lecture 13
Midterm 2 Update:
Multiple choice section is marked and posted
problem solving is currently being marked.
Low
Medium
High
X1
X6
X11
X2
X7
X12
X3
X8
X13
X4
X9
X14
X5
X10
X15
The numbers within the boxes represent ea
PSYC 2002
Lecture 16
March 28th, 2012
Linear Regression
predictor variable  variable that is being predicted X
criterion variable Y
X predicts Y
*On the Final!
pearson correlation, todays lecture will all be on the final.
Linear regression equation: Y
1
CHISQUARE
TESTS
GOODNESSOFFIT
&
TEST OF INDEPENDENCE
ChiSquare
2
1.
Parametric tests:
t and ftests
1.
Test hypotheses about the population.
Assume normal distribution in population.
Uses interval and ratio data only.
Nonparametric tests:
Chi Square
CHOOSING THE CORRECT INFERENTIAL TEST
STRENGTH
RELATIONSHIP
CONTINUOUS
VARIABLES
GOAL
PREDICTION
QUANTITATIVE
PEARSON R
REGRESSION
RESEARCH
QUESTION
ONE
GROUP
DIFFERENCES
NUMBER OF
GROUPS
TWO
ONE SAMPLE TTEST
MORE
THAN TWO
IND. TTEST
DEP. SAMPLES
DATA
T
PSYC2002 March 26th, 2012
Lecture 15.5
Class problem from last Wednesday always interpret interaction effect first
crossover effect use averages in the cells
not significant, might be a type 2 error probably because there was a small sample size.
Male
L17 PSYC 2002
Hypothesis testing By Test of independence
Men
Wome
n
Same
29
17
April 2nd, 2012
Row Total
46
Opposite
4
14
18
Column Total
33
31
n=64
Step 1:
H0: 2 population same men/women do not differ in sex comparing for job title.independent
H1: 2 po
appendix b Statistical Tables
TABLE B.1
THE UNIT NORMAL TABLE*
*Column A lists zscore values. A vertical line drawn through a normal distribution at a zscore location divides the
distribution into two sections.
Column B identifies the proportion in the
528
APPENDIX B
MM
(A)
z
STATISTICAL TABLES
(B)
Proport
ion in
Body
(O
Proportio
n inT
ail
0.50
0.51
.6915
.6950
.3085
.3050
0.52
.6985
.3015
0.53
.7019
.2981
0.54
.7054
.2946
0.55
.7088
.2912
0.56
.7123
.2877
0.57
.7157
.2843
0.58
.7190
.2810
0.59
.7224
FORMULASFINAL EXAM
Where;
Where;
Where;
Where;
Regression:
Where;
and
a = My
 bMx
EQUATIONS FOR BETWEEN GROUPS ONEWAY ANOVA:
SSbetween = SStotal SSwithin
SSwithin = SS for each level
where;

EQUATIONS FOR TWOWAY ANOVA:
SS within = SStotal  SSbetween
Cheat Sheet
2 X 2 Anova Results Interpretations Example: Conclude that there is no significant interaction between selfconfidence and group size (F(1,20)=1.4, ns). There is a significant main effect of selfconfidence on anxiety (F (1,
20)=12.6,p<0.05), w
MIDTERM DECISION CHART
How many samples are there?
SS
Independent
ttest
Is
known?
Yes
2 Same SS
2 Diff
1
Is the assumption of
homogeneity of variance
met? (Fmax)
No
Single sample
ttest
Dependent ttest
Yes
Fmax < F max
critical
No
Fmax > F max
critical
530
APPENDIX B
(A)
STATISTICAL TABLES
(B)
Proport
ion in
Body
(Q
Propor
tion
in
Tail
(D)
Propo
rtion
Betwe
en
Mean
and z
(A)
z
(B)
Proport
ion in
Body
2.50
.9938
.0062
.4938
2.95
.9984
.0016
2.51
.9940
.0060
.4940
2.96
.9985
.0015
2.52
.9941
.0059
.4941
2
Bonus Marks in Psychology 2001 & 2002
All students in Psychology 2001 & 2002 have the opportunity to receive bonus marks to increase their
final grade by electing to participate in psychological research conducted by Psychology faculty, or by
graduate stu
PSYC 2002
Lecture 5
January 18th, 2012
Review:
central tendency mean, median, mode
Mode  when using a nominal scale ( categories)
Mean on interval and ratio scales when you can calculate an average, representative of
all the data in the sample. If the
1.
Hypothesis testing
2.
Decisionmaking
3.
Errors associated with
the decisions
4.
Assumptions of a ztest
5.
Class problems
Population of
psychopaths
n=30
psychopaths
=4; =1
Treatment
Mean = 4
Level of remorse Mean = 6; S=1
Research Hypothesis: What is
Probability
Using the standard normal curve for
determining probabilities
Sampling error
Sampling distributions
Sampling distribution of the means
Inferential statistics: Making inferences about
the population from our sample results
Population
Sample
St
Lecture Outline
Central tendency
Mean, Median, Mode
Shape of distributions
Measures of variability
Range, variance, SD
Standard normal curve
Calculating standard scores (Z scores)
Practice problems
Central Tendency
=
Identifying a single score that descri