PSYC 305- Tuesday September 11
Basic Concepts:
Populationparameter
Sample:
-entire set of thing of interest
-in most cases its impossible to know the entire population
-descriptive property of pop.
-pop mean () or standard deviation ()
-The part of the po
1.
In experiments the independent variable is manipulated to determine:
a. effects on the individual participants
b. effect on the dependent variable *
c. effects of certain stimuli
d. relation to other variables
2.
If we calculated an effect size and fou
1.
In experiments the independent variable is manipulated to determine: a. effects on the individual participants b. effect on the dependent variable * c. effects of certain stimuli d. relation to other variables
2.
If we calculated an effect size
Assignment 6
Lab number 11
1.
There are some outliers
but the ACT and SAT
scores have relatively
strong positive linear
relationship.
2.
H0: =0. There is no linear association between SAT and
ACT scores.
H1: 0. There is a linear association between SAT an
Two-way independent-groups ANOVA
Labsession3.1
The two-way ANOVA is used to assess the effect of two factors (with 2 or more levels each) on the
dependent variable (DV). When different subjects are used for each combination of levels, it is
considered an
Labsession3
(page1)
Two-way independent-groups ANOVA - APPENDIX
Factor A
A1
A2
B1
X A1B1
X A2B1
X B1
B2
X A1B2
X A2B2
X B2
X A1
X A2
Factor B
simple effect of factor A (within the levels of factor B)
Hypothesesbeingtested
oneforeachofthelevelsinB
A withi
Non-Parametric tests
Labsession5
First,whatareParametrictests?
When a statistical test is performed on parameters (such as means, variances, correlations, etc), it
is called a parametric test.
These include all the ANOVAs, t-tests, and others since means
Kruskal-Wallis H test - Multiple
comparisons
As in ANOVA, if we reject H0, then we
want to perform post hoc multiple
comparison tests to examine which
groups differ significantly.
Kruskal-Wallis H test - Multiple
comparisons
Here is a simple method:
Examp
Labsession4
(page1)
One-way repeated-measures ANOVA
The one-way repeated-measures ANOVA is used to assess the effect of one independent variable
(or factor) with more than 2 levels on the dependent variable.
It is repeated when the same subjects go throug
Lab number 11
1. Hypothesis:
H0: The mean ratings by the 3 marketing managers are equal.
H1: The mean ratings by the 3 marketing managers are not equal.
= .05
Sphericity assumption is violated since p=.009<.05, so compound symmetry
Mauchly's Test of Sphe
Assignment 7
Lab number 11
Hypothesis:
H0: 1=2=3=0
H1: At least regression coefficient is significantly different
from 0.
= .05
Model Summary
R
Adjusted R Std. Error of
Model
R
Square
Square
the Estimate
a
1
.722
.521
.501
1.48253
a. Predictors: (Constan
Lab number 11
1. H0: The mean scores of spatial-temporal reasoning are
the same across all groups.
H1: the mean score of spatial-temporal reasoning are not
the same across all groups.
= .05
ANOVA
scores
Sum of
Squares
Between
Groups
Within Groups
Total
M
Lab number 11
1.
Hypothesis:
Main effect of AGE
H0: under 40 = otherwise
H1: under 40 otherwise
Main effect of MUSIC
H0: Fuzagi = ABBA = Barf Grooks
Interaction effect
H0: The interaction effect is
0
H1: Not all group means are equal
H1: The interaction e
Assignment 8
Lab number 11
Test of interaction effect
Hypothesis:
H0: There is no interaction effect between time spent
stalking before therapy and the therapy group.
H1: There is an interaction effect between the two variables.
= .05
Tests of Between-Su
Linear Regression: Assumptions
Simple & Multiple
Regression (I)
Linear regression assumes that
PSYC 305
Heungsun Hwang
Linearity assumption
Linear
Values of Y are independent and are
sampled at random from the population.
The relationship between X and Y
One-way independent-groups ANOVA (contd)
Labsession2.2
Posthoctests
The ANOVA is used to assess a global (overall) effect of a factor (with 2 or more levels) on the
dependent variable. When it is significant (H0 is rejected), additional tests can be used
STATISTICS FOR EXPERIMENTAL DESIGN PSYC305
ASSIGNMENT #2 (9 points)
Due Wednesday, Oct. 29, 2014, in class
WORK MUST BE INDEPENDENT, ALL QUESTIONS ATTEMPTED and NEATLY
ORGANIZED. SHOW YOUR WORK.
Assignments will be returned Wednesday, November 5th and res
STATISTICS FOR EXPERIMENTAL DESIGN PSYC305
ASSIGNMENT #1 (7 points)
Due Wednesday, Oct. 1, 2014, in class
WORK MUST BE INDEPENDENT, ALL QUESTIONS ATTEMPTED and NEATLY
ORGANIZED. SHOW YOUR WORK.
Assignments will be returned Wednesday, October 8th and resub
STATISTICS FOR EXPERIMENTAL DESIGN PSYC305
ASSIGNMENT #3 (9 points)
Due Wednesday, Nov. 19, 2014, in class
WORK MUST BE INDEPENDENT, ALL QUESTIONS ATTEMPTED and NEATLY
ORGANIZED. SHOW YOUR WORK.
Assignments will be returned Wednesday, November 26th and re
Labsession6
(page1)
Correlation and Regression
CORRELATION
Correlation is defined as the linear relationship between tw o numerical variables.
It is classified by direction (positive or negative) and strength (weak, moderate, or strong).
The Pearson corr
Labsession7
(page1)
Multiple Linear Regression
More than one independent variable (IV) is used to predict the dependent variable in multiple
regression.
The independent variables (denoted as Xj) can be either continuous (numerical) or categorical
(numeri
Labsession8
(page1)
Analysis of co-variance ( ANCOVA )
WhyANCOVA?
The ANCOVA is used to assess the effect of a factor (categorical IV) on the DV, while controlling
for the effect that another extraneous IV could be having on the DV.
When the extraneous va
Repeated measures designs
One-way Repeated
Measures ANOVA (1)
PSYC 305
Heungsun Hwang
One-way repeated measures design
The simplest experiment of this kind may be a
before- and after-treatment design (two
conditions)
Example: One-way repeated
measures des
Regression Analysis
Analysis of
Covariance (I)
In the previous sessions, we have
learned Regression Analysis with
Extraneous variable
Example 1
In experiments, subjects are randomly
assigned to experimental treatments so as
to avoid any systematic bias in
One-Way ANOVA
One-Way ANOVA II (1)
PSYC 305
Heungsun Hwang
One-way ANOVA
SS(B)
df(B)
F ratio
SS(W)
df(W)
VW
(MS(W)
Example: One-Way ANOVA
A random sample of the students in
each row was taken.
The scores for those students on the
mid-term exam were record
One-way independent-groups ANOVA
Labsession2.1
Terminology:
One-way = one factor (categorical independent variable)
Although a factor can have an unlimited number of levels (k),
the name of the design will never mention the levels or their number.
Indepe
Assignment 5
Lab number 11
1.Hypothesis:
H0: The two variables (sex and user groups) are statistically
independent.
H1: The two variables are not independent
= .05
Chi-Square Tests
Value
6.341a
6.545
Pearson Chi-Square
Likelihood Ratio
Linear-by-Linear
A
Basic Statistics:
1. Basic Concepts:
a. Population: The entire sets of things of interest (population parameter)
b. Sample: The part of the population. Typically this provide the data we will look at
(property of the sample)
c. Descriptive Statistics v.s